Lopez et al v. Griswold et al

1
IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLORADO
Civil Action No. 22-cv-00247-JLK
GREG LOPEZ,
RODNEY PELTON, and
STEVEN HOUSE,
Plaintiffs,
v.
JENA GRISWOLD, Colorado Secretary of State, in her official capacity, and
JUDD CHOATE, Director of Elections, Colorado Department of State, in his official capacity,
Defendants.
FINDINGS OF FACT, CONCLUSIONS OF LAW, AND ORDER FOR JUDGMENT
Kane, J.
This case presents constitutional challenges to Colorado’s campaign finance laws.
Plaintiffs seek to have the individual campaign contribution limits and the voluntary spending
limits scheme in §§ 3 and 4 Article XXVIII of the Colorado Constitution declared violative of
the First Amendment to the U.S. Constitution and ordered unenforceable. I conclude that
Colorado’s contribution limits are closely drawn to the sufficiently important state interest of
preventing quid pro quo corruption or its appearance. Therefore, they do not violate the political
expression and association rights of candidates or contributors that are guaranteed by the First
Amendment to the U.S. Constitution. I also conclude that Colorado’s voluntary spending limits
scheme merely offers candidates an alternative campaign financing option that does not infringe
on their First Amendment rights and is not coercive. Consequently, I hold that the challenged
provisions of the Colorado Constitution do not violate the U.S. Constitution and deny Plaintiffs’
claims.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 1
of 142

2
I. FINDINGS OF FACT
Plaintiffs Greg Lopez and Rodney Pelton are current and former candidates for state
office in Colorado. Plaintiff Steven House is a frequent contributor to Colorado candidates.
Defendant Jena Griswold is the Colorado Secretary of State. Defendant Judd Choate is the
Director of Elections at the Colorado Department of State.
Plaintiffs claim Colorado’s limits on the amount individuals can contribute to candidates
for state office are so low that they place a substantial burden on the speech and association
rights of those candidates and contributors. This burden, according to Plaintiffs, cannot be
justified by Colorado’s stated interest in preventing quid pro quo corruption or the appearance of
such corruption. Quid pro quo corruption is the exchange of official acts for money, including
campaign contributions. Preventing such corruption or its appearance is the only government
interest the U.S. Supreme Court views as potentially important enough to justify limiting
campaign contributions. Plaintiffs also claim that Colorado’s voluntary spending limits scheme
impermissibly penalizes candidates for state office who do not accept the expenditure caps and
thus burdens their First Amendment rights without serving a compelling state interest.
Plaintiffs previously sought a preliminary injunction on their claims, and I denied that
Motion, see generally Mem. Op. & Order on Mot. for Prelim. Inj., ECF No. 26. Plaintiffs
appealed, and the Tenth Circuit dismissed the appeal as moot or prudentially moot. See Lopez v.
Griswold, No. 22-1082, 2023 WL 1960802 (10th Cir. Feb. 13, 2023) (unpublished). The matter
then proceeded to a bench trial for me to render a final determination on Plaintiffs’ claims.
A. Article XXVIII
In 1974, the Colorado legislature enacted its first campaign finance reform, limiting
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 2
of 142

3
contributions to $100 per candidate. See Trial Ex. 352 at 7, Trial Ex. 374. More than twenty
years later, the Colorado legislature and the people of Colorado “began a back and forth” where
the legislature would try to raise campaign contribution limits and the people would try to lower
those limits. Trial Tr. at 24:7-9; see Trial Ex. 374. In 1996, the Colorado legislature passed a law
that increased individual campaign contribution limits from $100 to $5,000 for gubernatorial
candidates; $2,500 for Secretary of State, Attorney General, and Treasurer candidates;1 $1,000
for state Senate candidates; and $500 for state House of Representatives and other candidates.2
See Trial Ex. 353 at 2-3, Trial Ex. 374. In 1996, the Colorado electorate responded to those
increases by passing Amendment 15, a citizen’s initiative that amended Colorado law to cut the
contribution limits that had been passed by the legislature to $500 for gubernatorial, Secretary of
State, Attorney General, and Treasurer candidates and $100 for state Senate, state House, and
other offices. See Trial Ex. 356 at 3, Trial Ex. 374. In 2000, the Colorado legislature again
increased the contribution limits to $5,000 for gubernatorial candidates; $2,500 for Secretary of
State, Attorney General, and Treasurer candidates; $1,500 for state Senate candidates; and
$1,000 for state House and other candidates. See Trial Ex. 354 at 1, Trial Ex. 374.
In 2002, Article XXVIII of the Colorado Constitution was proposed by a citizen’s
initiative as Amendment 27 and enacted by popular vote. See Colo. Ethics Watch v. Senate
Majority Fund, LLC, 269 P.3d 1248, 1253 (Colo. 2012). Amendment 27 lowered or prohibited
campaign contribution limits. See Trial Ex. 11 at 2, Trial Ex. 357. The following table, which
was provided to the Colorado electorate as part of the 2002 Ballot Information Booklet, clearly
1 I refer to the contribution limits for Governor, Secretary of State, State Treasurer, or Attorney
General as the limits for “statewide” or “Tier 1” candidates.
2 I refer to the contribution levels for Colorado Senate, House of Representatives, Board of
Education, Board of Regents for the University of Colorado, or District Attorney as the limits for
“legislative” or “Tier 2” candidates.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 3
of 142

4
identified the limits that would be imposed if the measure was passed.
Trial Ex. 11 at 3.
Of the 1,338,989 votes cast in the 2002 election on Amendment 27, 890,390—or
66.5%—voted to adopt it. See Trial Ex. 360 at 144-45. Article XXVIII is now Colorado’s
“primary campaign finance law.” Colo. Ethics Watch, 269 P.3d at 1253. In addition to lowering
or prohibiting various categories of Colorado’s campaign contribution limits, Article XXVIII
addresses other aspects of campaign finance and includes the option for candidates for Colorado
state office to accept voluntary spending limits. In adopting Amendment 27, the people of
Colorado endorsed the following statement regarding the purposes of the reforms:
The people of the state of Colorado hereby find and declare that large campaign
contributions to political candidates create the potential for corruption and the
appearance of corruption; that large campaign contributions made to influence
election outcomes allow wealthy individuals, corporations, and special interest
groups to exercise a disproportionate level of influence over the political process;
. . . and that the interests of the public are best served by limiting campaign
contributions . . . .
Colo. Const. art. XXVIII, § 1.
Article XXVIII § 3 and its accompanying regulations establish the limits on the funds
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 4
of 142

5
donors may contribute to candidates for public office during an election cycle.3 Effectively, a
candidate may accept a single contribution from an individual that is double the contribution
limit during either applicable election cycle. Trial Tr. 872:21-873:6; see also Trial Ex. 351 at 20
(asterisk stating: “A candidate may accept the contribution limit for both the primary election
and the general election.”). Individual contribution limits under § 3 are based on the office the
candidate seeks. Currently, an individual is limited to a contribution of $725 per election cycle—
per primary and per general election—to a candidate for Governor, Secretary of State, State
Treasurer, or Attorney General, and $225 per election cycle to a candidate for state Senate, state
House of Representatives, state Board of Education, Board of Regents for the University of
Colorado, or District Attorney. See 8 Colo. Code Regs. § 1505-6, Rule 10.17.1(b).4
Beginning in the first quarter of 2007 and then every four years thereafter, all limits are
“adjusted by an amount based upon the percentage change over a four[-]year period in the
United States bureau of labor statistics [C]onsumer [P]rice [I]ndex for Denver-Boulder-
Greeley,”5 and then “rounded to the nearest lowest twenty-five dollars.” Colo. Const. art.
XXVIII § 3(13). When Article XXVIII was first adopted, the individual contribution limit for
3 The Colorado Constitution provides three definitions for “election cycle,” which are,
(a) The period of time beginning thirty-one days following a general election for
the particular office and ending thirty days following the next general election for
that office; (b) The period of time beginning thirty-one days following a general
election for the particular office and ending thirty days following the special
legislative election for that office; or (c) The period of time beginning thirty-one
days following the special legislative election for the particular office and ending
thirty days following the next general election for that office.
Article XXVIII § 2(6). Stated another way, the limits described in Article XXVIII and the
Colorado Code of Regulations apply to the primary election and the general election.
4 These limits are not the same as initially authorized in 2002. They were adjusted upward in
2023. See 8 Colo. Code Regs. § 1505-6, Rule 10.17.1.
5 The Denver-Boulder-Greeley Consumer Price Index no longer exists. Trial Tr. at 571:23-25,
871:6-7. Now, it is the Lakewood-Denver-Aurora index. Id. at 572:1-2, 871:5-6.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 5
of 142

6
Tier 1 candidates was $500 per election cycle. See Trial Ex. 11 at 3. Based on the inflation
mechanism, that limit has increased to $725 per election cycle, for a total of $1,450. See 8 Colo.
Code Regs. § 1505-6, Rule 10.17.1(b)(1). The individual contribution limit for Tier 2 candidates
has similarly increased due to the inflation mechanism, rising from $200 to $225 per election
cycle, allowing an individual to contribute a total of $450. See id. at 10.17.1(b)(2). Still,
Colorado’s individual contribution limits are among the lowest, if not the lowest, limits in the
country. See, e.g., Trial Ex. 1; Trial Tr. at 149:14-150:7.
Section 3 also sets limits for political parties’ contributions to candidates for state office.
These limits per election cycle are $789,060 for gubernatorial candidates; $157,805 for Secretary
of State, state Treasurer, and Attorney General candidates; $28,395 for state Senate candidates;
$20,500 for state House, state Board of Education, Regent of the University of Colorado, and
District Attorney candidates. See Colo. Const. Art. XXVIII § 3(3)(d); 8 Colo. Code Regs.
§ 1505-6, Rule 10.17.1(i); Trial Ex. 351 at 20.
Article XXVIII § 4 establishes the State’s voluntary spending limits scheme. Under this
section, a candidate may choose to certify to the Secretary of State that he or she will limit
campaign spending for the election to specified amounts. See Colo. Const. Art. XXVIII § 4(1).
When a candidate for a state office announces his or her candidacy, the candidate must declare in
the candidate affidavit whether he or she accepts the applicable voluntary spending limit. See
Colo. Const. Art. XXVIII § 4(3); Colo. Rev. Stat. § 1-45-110(1). Candidates who accept these
limits also agree that any personal contributions count as political contributions, which are
subject to the contribution limits set by § 3. See Colo. Const. Art. XXVIII § 4(2). The agreement
to accept the spending limit is irrevocable unless another candidate for the same office refuses to
accept the spending limit, in which case a candidate has ten days to withdraw his or her previous
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 6
of 142

7
acceptance. See id. § 4(3), (4).
Currently, Colorado’s voluntary spending limits per election cycle are $3,945,300 for
gubernatorial candidates; $789,025 for Secretary of State, state Treasurer, and Attorney General
candidates; $141,975 for state Senate candidates; $102,500 for state House, state Board of
Education, Regent of the University of Colorado, and District Attorney candidates. See Colo.
Const. Art. XXVIII § 4(1); 8 Colo. Code Regs. § 1505-6, Rule 10.17.1(j). Candidates may
advertise their compliance with the voluntary spending limit. See Colo. Const. Art. XXVIII §
4(6). A candidate who accepts the spending cap enjoys a doubled individual contribution limit
during the election if another candidate in the race has not accepted the spending limit and that
candidate has raised more than ten percent of the cap. See id. § 4(5). Section 4 includes an
inflation mechanism to adjust each spending limit, which is identical to the inflation mechanism
in § 3. See id. § 4(7).
B. Parties and Claims
Plaintiff Greg Lopez is a former Colorado gubernatorial candidate who ran in 2018 and
2022. Trial Tr. at 43:24-44:3. He was defeated both times in the Republican primary. Id. at
48:10-49:2. He has followed Colorado politics for thirty-six years. Id. at 88:14-15. He was not
aware of any occurrence of quid pro quo corruption involving campaign contributions. Id. at
88:16-18. In his campaigns, Mr. Lopez relied mostly on individual contributions, see, e.g., id. at
75:15-16, and did not receive any contributions from his political party, id. at 99:2-5. During his
2018 gubernatorial primary race, Mr. Lopez raised a little more than $60,000. Id. at 89:2-4. He
did not receive any maximum individual contributions to his campaign. Id. at 90:17-92:5. During
his 2022 primary race, he raised approximately $150,000, id. at 76:20-23, including 26
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 7
of 142

8
maximum individual contributions, id. at 92:6-18. His opponent raised a little more than $1.4
million, including approximately $600,000 in self-funded contributions. Id. at 76:24-77:15.
Based on how much money the incumbent candidate raised, had he advanced to the 2022 general
election, Mr. Lopez testified he would have needed millions of dollars to compete with the
incumbent candidate. See, e.g., id. at 83:17-84:8. He did not run against an incumbent candidate
in either the 2018 or 2022 primary elections. Id. at 93:16-21. He did not accept the § 4 voluntary
spending limits in either of his campaigns. Id. at 101:4-8.
Mr. Lopez intends to run for statewide office again in 2026, id. at 50:17-18, 102:2-4, and
plans to fund his campaign through individual contributions, id. at 53:15-20. During his 2022
gubernatorial primary, Mr. Lopez visited all sixty-four counties in the State of Colorado. Id. at
57:22-24. He explained that an effective campaign requires resources such as money, campaign
signs, advertisements, means of transportation (i.e., flights and rental cars), and staff. Id. at
52:20-53:11. Anticipating his 2026 campaign, he estimated he would need approximately $3 or
$4 million to run a competitive campaign, which again would include visiting every Colorado
county. Id. at 52:18-53:2, 88:9-13. To that end, Mr. Lopez maintains an active Colorado
candidate political committee for statewide office, called “Greg Lopez for Governor,” which has
up-to-date contribution and expenditure reports. Id. at 50:19-51:9, 102:13-14; see also Trial Ex.
19. As of his trial testimony, his committee, while considered active, had not accepted any
contributions since 2022. Trial Tr. at 103:2-7.
Plaintiff Rodney Pelton is a Colorado State Senator serving his first term in the Senate
after four years in the Colorado House of Representatives. Id. at 105:24-25. He has followed
Colorado politics for twenty-five years. Id. at 125:24-25. He was not aware of any occurrence of
quid pro quo corruption involving campaign contributions. Id. at 126:1-3. In his first Colorado
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 8
of 142

9
House campaign, in 2018, he did not run against an incumbent; he won the primary after his
challenger dropped out of the race, and he defeated his opponent in the general election. Id. at
106:20-25. Senator Pelton raised about $18,500 during that race, and he agreed that it was a
sufficient amount of money for that election. Id. at 126:9-16, 127:11-13. He chose to accept the
voluntary spending limits for his 2018 campaign. Id. at 133:10-13.
Senator Pelton ran unopposed during his 2020 campaign for reelection to the Colorado
House of Representatives. Id. at 107:11-12. He raised $20,000 during that race and again agreed
that the amount he raised was sufficient. Id. at 127:14-128:19. He also accepted the voluntary
spending limits during his 2020 campaign. Id. at 133:22-24.
In 2021, Senator Pelton ran for state Senate after redistricting in Colorado created a new
state Senate district. Id. at 107:13-16. He ran unopposed during the primary and defeated his
opponent in the general election. Id. at 107:16-20. During his Senate campaign, Senator Pelton
raised a little more than $40,000, which he testified was sufficient to run an effective campaign.
Id. at 128:20-129:22. As in his other campaigns, he again chose to accept the voluntary spending
limits. Id. at 134:7-9. His campaigns were funded through contributions from individuals, small
donor committees, and county political parties. Id. at 110:7-9, 121:9-13.
Senator Pelton intends to run for reelection in 2026, id. at 108:19-20, and maintains an
active Colorado candidate political committee with up-to-date contribution and expenditure
reports, id. at 108:21-109:5; see also Trial Ex. 20. He does not intend to accept the voluntary
spending limits during his 2026 campaign because he made “a calculation” about his strategic
best interests. Trial Tr. at 135:2-9. He testified that the main resources required to run for the
legislature are time and money for expenses, including travel, food, and advertising. Id. at
109:10-121:8. As of the date of trial, Senator Pelton had raised a little more than $800 for his
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg 9
of 142

10
2026 campaign. Id. at 109:8-9. In his previous campaigns, he had “at least one” but possibly
“several” donors attempt to contribute more than Colorado’s contribution limit. Id. at 124:19-23.
Plaintiff Steven House is a former Chairman of the Colorado Republican Party. Id. at
138:13-14. Mr. House also ran for Governor and U.S. House of Representatives; he was
unsuccessful in both campaigns. Id. at 138:16-19. His testimony centered on his contributions to
candidates running for Colorado statewide, legislative, and other offices. Id. at 138:24-139:12.
He claims he has contributed over $280,000 to Colorado state candidates since 2010. Id. at
138:24-139:4; see also Trial Ex. 108 (showing contributions over $10,000 to various candidates,
including the maximum amount to statewide and legislative candidates, during the 2014 four-
year election cycle); Trial Ex. 109 (showing contributions over $14,000 to various candidates,
including the maximum amount to statewide and legislative candidates, during the 2018 four-
year election cycle); Trial Ex. 110 (showing contributions over $4,700 to various candidates,
including the maximum amount to statewide and legislative candidates, during the 2022 four-
year election cycle).6 Mr. House intends to financially support various Colorado candidates for
as long as he can, Trial Tr. at 152:24-153:1, and, if permitted, would contribute more than the
maximum amounts presently allowed by law, see, e.g., id. at 158:19-23.
Secretary of State Griswold and Director of Elections Choate administer, investigate, and
enforce Colorado’s campaign finance laws, including Article XXVIII §§ 3 and 4 and related
statutes and rules. See Colo. Const. Art. XXVIII § 9; Colo. Rev. Stat. § 1-45-111.5.
Each Plaintiff asserts facial challenges to provisions of §§ 3 and 4 of Article XXVIII.
Plaintiffs claim these provisions infringe upon their rights guaranteed under the First
6 Mr. House testified that he has donated more than $280,000 to Colorado candidates across 202
donations since 2010. However, the evidence provided to me shows does not demonstrate that.
See, e.g., Trial Exs. 108-110.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
10 of 142

11
Amendment of the U.S. Constitution, as extended through the Fourteenth Amendment. Again,
Plaintiffs’ first claim (“Claim One”) alleges that the individual campaign contribution limits
under § 3 deprive them of their First Amendment freedoms of speech, association, and assembly
and that the State’s interest in adopting those limits—the prevention of corruption and the
appearance of corruption—cannot justify them. Amended Compl. ¶ 55, ECF No. 46. Plaintiffs’
second claim (“Claim Two”) asserts that, “[b]y doubling an opponent’s contribution limits when
a candidate refuses to abide by expenditure limits,” § 4 “punishes candidates [who] choose to
exercise their First Amendment rights fully, and the donors who would support them,” thus
depriving Plaintiffs “of the First Amendment’s protection for free speech and association.” Id. ¶¶
57, 62. Plaintiffs bring these claims against Defendants in their official capacities pursuant to 42
U.S.C. § 1983.
C. Fact Witness Testimony
At the bench trial, in addition to their own testimony, Plaintiffs presented testimony from
Richard Wadhams and Benjamin Engen. Defendants presented testimony from Stephen Bouey,
Bernard Buescher, and Kerry Donovan.
1. Stephen Bouey
I begin with the testimony of Stephen Bouey because he provided valuable background
information. Since 2009, Mr. Bouey has worked for the Colorado Department of State and
managed Colorado’s campaign finance and lobbyist disclosure programs. Id. at 836:24-837:3. He
testified about Colorado’s campaign finance disclosure database, which is called Transparency in
Contribution and Expenditure Reporting or TRACER. Id. at 841:2-4. TRACER contains reports
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
11 of 142

12
submitted by committees and candidates detailing required disclosures, contributions, and
expenditures. Id. at 839:14-18, 841:20-22. TRACER began operating in 2010 but contains legacy
campaign finance data. Id. at 839:23-840:3. Mr. Bouey explained that it was not possible to
import all legacy data into TRACER because of conversion errors or hard-copy disclosure filings
from the county level. Id. at 840:6-841:1.
Although TRACER is maintained by the Colorado Secretary of State, the office does not
verify the accuracy of the information submitted. Id. at 842:4-6. Candidates and committees are
responsible for complying with the law and accurately reporting contributions and expenditures.
Id. at 925:19-22, 926:2-4. Audit warnings are built into the system, but the warnings do not
prevent a committee or candidate from submitting the flagged information. Id. at 842:7-13, 21-
25. These warnings provide guidance to TRACER filers based on the Colorado Constitution,
state statutes, and regulations. Id. at 865:5-9. TRACER submissions are available to the public,
which allows the media and watchdog groups to review the data. Id. at 843:6-9.
A member of the public can file a complaint with Mr. Bouey’s office if he or she
identifies a potential campaign finance violation. Id. at 843:10-16. A team in the office would
then review the complaint and determine whether a violation occurred. Id. at 845:18-22. If the
violation can be cured, the team would settle or dismiss the complaint. Id. at 846:10-13. An
incurable complaint, however, would lead to an investigation during which the team would
request and review additional information from the complaint’s respondent. Id. at 846:14-19.
The outcome of the investigation would result in a dismissal of the complaint or, if that would be
inappropriate, it would proceed to adjudication. Id. at 846:21-22. Mr. Bouey aims to read every
complaint that is filed in his department and was not aware of any complaints related to quid pro
quo corruption and campaign contributions. Id. at 926:5-7, 926:14-16. Similarly, he was not
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
12 of 142

13
aware of any enforcement actions for quid pro quo corruption involving campaign contributions.
Id. at 926:17-20.
Mr. Bouey also described specific reports available in TRACER. Campaign contributions
and expenditures are reported, as are each level of political parties’ contributions and
expenditures. See id. 607:7-12, 861:16-23.
Part of Mr. Bouey’s job is adjusting contribution limits and voluntary spending limits for
inflation. This involves analyzing inflation over a four-year period based on the now-named
Denver-Aurora-Lakewood CPI and adjusting the limits as needed. Id. at 870:24-871:6. If an
adjustment to either limit is necessary, the new limit is published in the Code of Colorado
Regulations. Id. at 872:8-15; see also Trial Ex. 351 at 20. Mr. Bouey agrees with Plaintiffs that
Colorado’s contribution limits have not perfectly matched inflation. Trial Tr. at 874:8-10. He
explained, however, that as the contribution limits increase, the required four-year inflation rate
to increase the limit will be lower, despite the rounding requirement. Id. at 875:3-10. In other
words, when the contribution limits reach a new and higher threshold, a smaller percentage
change will raise the limit a greater number of dollars and thus is less likely to be neutralized by
the rounding requirement. See id. at 874:22-875:1. For example, to overcome the rounding
requirement for $200, there needed to be a 12.5% increase in inflation over the four-year period,
but for $225 that increase need only be 11.1%. See id. at 875:3-10.
Mr. Bouey also testified about the growth in Colorado’s electorate. From 2004 to 2020,
there was a 54% increase in the number of ballots cast in Colorado in presidential elections. Id.
at 876:19-877:4; Trial Ex. 371. During that same period there was an overall increase of 78% in
the number of ballots cast in Colorado statewide elections. Trial Tr. at 878:4-8; Trial Ex. 372.
This means the pool of potential donors has increased while the total number of House and
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
13 of 142

14
Senate districts has remained the same. Trial Tr. at 878:21-22, 879:4-9.
Mr. Bouey also addressed campaign volunteer services and expenses. An exemption for
time-based volunteer services is carved out of the definition of campaign contributions in the
Colorado Constitution. See Colo. Const. art. XXVIII, § 2(5)(b) (defining “contribution” as “not
includ[ing] services provided without compensation by individuals volunteering their time on
behalf of a candidate, candidate committee, political committee, small donor committee, issue
committee, or political party . . .”). Thus, time spent performing volunteer services for a
campaign does not count as a contribution. On the other hand, volunteer expenses that “produce
a thing of value” are considered contributions. Id. at 885:14-18. Mr. Bouey explained that
infrequently candidates report volunteers’ mileage, but they are not required to do so. Id. at
887:13-20, 889:13-23. This testimony was based on his understanding of Keim v. Douglas
County School District, 399 P.3d 722 (Colo. App. 2015). See Trial Ex. 359. If a candidate
chooses to overreport, that would not trigger a TRACER audit flag. Trial Tr. at 890:14-16.
Mr. Bouey discussed Colorado’s voluntary spending limits. As previously explained, § 4
allows a candidate to choose to limit the amount of money he or she will spend during an
election cycle. See supra Section I.A. In the election cycles between 2014 through 2022, the
percentage of eligible candidates who opted into the voluntary spending limits ranged from 30%
to 37%. Trial Tr. at 895:14-17; see also Trial Ex. 380. He was not aware of any election cycle
involving 40% or more of eligible candidates opting into the voluntary spending limits. Trial Tr.
at 895:21-23. Further, in 2022, 88% of the candidates who opted into the spending limits and
made expenditures spent less than 50% of the limit, while only 3% spent up to the limit. Trial
Ex. 379. Of the candidates in 2022 who declined the voluntary spending limits and made
expenditures, 62% spent less than 50% of the limit, while 21% spent more than the limit. Id.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
14 of 142

15
2. Richard Wadhams
Richard Wadhams is a former Colorado Republican State Chairman who has managed
state and national political campaigns and worked for multiple elected officials. Trial Tr. at
633:4-11. Mr. Wadhams now works as a political consultant and pundit. Id. at 634:2-14. He
testified that the state Republican and Democratic parties follow “[b]ylaws and tradition” to not
donate to primary candidates during the primary election. Id. at 638:10-21, 675:14-18. But a
party could choose to donate to intra-party primary candidates if it wanted to. Id. at 675:21-23.
While the state level Republican Party may donate money to candidates, it prioritizes getting out
the vote rather than donating money to individual candidates. Id. at 643:6-10. To aid in its get out
the vote efforts, the state Party receives “a lot” of money from the national party. Id. at 671:4-7.
The Party’s get out the vote initiative benefits multiple candidates at once, id. at 672:2-4, and
involves hundreds of volunteers making phone calls, knocking on doors, and sending mailers,
which supports candidates without giving direct financial support, id. 671:11-672:1.
Much of Mr. Wadhams’s testimony centered on state political party contributions to
candidates. He asserted that contributing to a candidate in a safe district or one that votes heavily
for the other party is a waste of money. Id. at 648:1-4. Instead, he explained, the state Republican
Party looks to contribute to competitive races, meaning those with candidates who have a chance
of winning. Id. at 648:4-6. The Party, according to Mr. Wadhams, would not want to donate
money to a candidate in a heavily Democratic district or even a candidate in a heavily
Republican district. Id. at 672:18-24.
Mr. Wadhams walked through many years’ worth of contributions from both the
Republican and Democratic state parties and noted that most candidates did not receive the
maximum contribution from the party. See, e.g., id. at 649:14-651:13; Trial Exs. 56-59. He
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
15 of 142

16
testified that a political party “technically could give [to] all of [the candidates], but the reality is
they can’t because [the parties] can’t raise that much money.” Id. at 653:22-23. His testimony
included facts that showed Colorado’s state Democratic Party outspent Colorado’s state
Republican Party. Id. at 655:13-17. He explained that the Republican Party does not donate the
maximum amount to its candidates because it is difficult to raise the money, which requires the
Party to prioritize. Id. at 656:13-15. It is difficult to raise money because, among other reasons,
donors prefer to give money directly to candidates rather than a party. Id. at 656:23. Independent
expenditure committees, or “IECs,” spent more on Colorado campaigns than political parties,
which was possible because there are no limits on IEC contributions. Id. at 658:7-16; Trial Exs.
83, 84.
Finally, Mr. Wadhams testified that in his experience following Colorado politics, which
spans decades, he was not aware of any appearance of quid pro quo corruption involving
campaign contributions. Trial Tr. at 668:6-11.
3. Benjamin Engen7
Benjamin Engen has worked for Colorado candidates’ campaigns in a consulting capacity
during every election cycle since 2006. Id. at 565:15-17. He now works as an economist focusing
on short-term U.S. economic forecasting, which requires him to monitor the Consumer Price
Index and Federal Reserve actions, and as a staff analytics engineer, which requires him to
maintain databases and other analytical aspects of his business. Id. at 564:17-565:9, 590:1-12.
Mr. Engen’s testimony focused on the inflation mechanism in Amendment XXVIII. In
7 Mr. Engen also testified as an expert witness. This section summarizes his testimony as a fact
witness. His expert testimony is addressed later in these Findings of Fact.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
16 of 142

17
2007, the first year the campaign contribution limits could be adjusted, the individual limit to
statewide candidates was raised from $500 to $525. Id. at 572:14-16. The individual limit to
legislative candidates was not adjusted based on inflation and remained at $200. Id. at 572:17-20.
In 2011, the next time the limits could be adjusted, the individual limit to statewide candidates
was raised from $525 to $550. Id. at 572:23-573:3. Again, the individual limit to legislative
candidates remained at $200. Id. at 573:4-7. In 2015, the individual limit to statewide candidates
was raised from $550 to $575. Id. at 573:15-17. Once again, the individual limit to legislative
candidates remained at $200. Id. at 573:18-20. In 2019, the individual limit to statewide
candidates was raised from $575 to $625. Id. at 575:17-19. The individual limit to legislative
candidates remained at $200. Id. at 575:20-23. Finally, in 2023, the individual limit to statewide
candidates was raised from $625 to $725. Id. at 576:12-14. And, for the first time, the individual
limit to legislative candidates was raised from $200 to $225. Id. at 576:15-23.
Mr. Engen defined the Consumer Price Index (“CPI”) for Denver-Aurora-Lakewood
from 2002 until 2022. See generally Trial Ex. 40. Using the CPI, he calculated the dollar value
for the individual contribution limit to statewide candidates if the inflation mechanism did not
require the amount to be rounded down to the nearest $25. Trial Tr. at 578:9-579:4. Without
rounding, he explained, the current contribution limit for statewide candidates would be $824
rather than $725; a 14% difference. Id. at 580:6-10; see also Trial Ex. 40 at 3. For legislative
candidates, Mr. Engen concluded that the current individual limit would be $330 rather than
$225; a 47% difference. Trial Tr. at 581:20-23; see also Trial Ex. 40 at 4.
4. Bernard Buescher
Bernard Buescher is a former Colorado elected official who has had successful and
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
17 of 142

18
unsuccessful campaigns. In 1998, he won the nomination for lieutenant governor, but he and his
running mate lost the election. Trial Tr. at 943:10-17. Mr. Buescher’s first successful campaign
was for a seat in the Colorado House of Representatives in 2004. Id. at 943:18-22. He
successfully ran for reelection in 2006 but lost his bid for reelection in 2008. Id. at 943:23-24. He
was then appointed Colorado Secretary of State, but he lost the race that would have allowed him
to continue in that position. Id. at 943:25-944:4.
Mr. Buescher testified about his experience raising funds for his campaigns. In 2004, he
raised $120,000 from contributions, and he loaned his campaign $20,000. Id. at 947:2-5. He
believed this was enough money to run an effective campaign at the time. Id. 947:13-15. In his
next two reelection campaigns, he raised more money than he raised in his first campaign. Id. at
949:10-11, 23. This included donations from small donor committees, which enjoy a higher
contribution limit than individual donors. See, e.g., id. at 964:23-25; 967:23-24. Although he was
the incumbent and raised about four times more than his opponent in 2008, he lost the race. Id. at
950:22-951:4; see also Trial Ex. 394. Similarly, when he lost his seat as Secretary of State in the
2010 election, he raised more money than his opponent. Trial Tr. at 954:10-19.
Mr. Buescher explained that the contribution limits required him, simply, to get more
donors. Id. at 956:1-6. As he testified, “the more donors [he] got, the more advocates [he] got.”
Id. at 956:6. Mr. Buescher did not accept the voluntary spending limits in his state legislature or
secretary of state races. Id. at 959:5-7. He testified that he never had a volunteer be unable to
contribute money to his campaign because the volunteer had “maxed out the individual
contribution limit by driving around the state and volunteering” on his behalf. Id. at 960:15-19.
And he never experienced a donor or potential donor telling him that he or she was contributing
money to his campaign and thus expected him to vote or act in a certain way. Id. at 961:1-3.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
18 of 142

19
However, Mr. Buescher gave an example of a lobbyist approaching a legislator and informing
the legislator that the interest he or she represented had half of a million dollars to spend on the
legislator’s race, but the lobbyist was not yet sure if the money would be spent to support the
legislator or his opponent. Id. at 971:2-5. The lobbyist informed the legislator which issues were
of most concern and made it clear if he voted the way the donor wanted, he would have $500,000
for his race. Id. at 971:1-8. Mr. Buescher testified that there were many examples of similar
interactions, and that he considered those to be quid pro quo corruption. Id. at 971:10-20.
5. Kerry Donovan
Kerry Donovan is a former Colorado elected official. She first ran for an open seat in a
newly created Colorado State Senate district in 2014. Id. at 794:11-12, 796:2-6. Ms. Donovan
explained that, during her campaign, she felt comfortable asking for maximum contributions
from donors because $400 is “a pretty low dollar amount for most people.” Id. at 798:21-24.
During her 2014 campaign for state Senate, Ms. Donovan raised a little more than $160,000 and
spent a little more than $150,000. Id. at 799:9-24; see also Trial Ex. 343. She testified that she
felt she raised enough money to run an effective campaign. Trial Tr. at 799:25-800:4.
Ms. Donovan explained that her campaigns did not aim to contact every voter in her
district. Id. at 802:9-12. Instead, her campaigns focused on targeting so-called persuasion voters,
or those voters who fall in the middle of the political spectrum. Id. at 803:11-25. Her 2014
campaign, in which she was successful, included mailers, advertisements, and travel throughout
her district. Id. at 800:17-802:8. Her 2018 campaign, which she also won, was similar. In that
campaign she raised a little more than $237,000 and spent a little more than $175,000. Id. at
806:22-807:1; see also Trial Ex. 343.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
19 of 142

20
Over the course of her two Colorado Senate campaigns, she received approximately
2,357 separate contributions. Trial Tr. at 809:22-25. Ms. Donovan also received approximately
$40,000 in contributions from the local and the state Democratic Party during both of her
campaigns. Id. at 810:6-9; Trial Ex. 403.
Ms. Donovan did not accept voluntary spending limits during either of her Senate
campaigns, and she did not feel pressured to accept them. Trial Tr. at 812:21-813:6. She also
testified that she never experienced a volunteer being unable to financially contribute to her
campaign because that person had maxed out their contributions by driving around the state on
her campaign’s behalf. Id. at 813:7-11.
Finally, Ms. Donovan testified about an experience she had in which a donor attempted to
pressure her to vote in a certain way in exchange for a monetary donation. During a candidate
interview with a group focused on medical issues in Colorado, Ms. Donovan was asked how she
would vote on a hypothetical bill that would likely be raised in the next session. Id. at 814:7-11.
During the interview, the group “directly implied” that her answer would impact whether she
received the group’s endorsement and maximum financial contribution, which was
approximately $4,000. Id. at 814:11-13, 815:2-6. She could not remember the specific group or
the individuals involved, could not recall if she encountered the group or individuals again while
she was in office, and did not consider reporting the incident to the Secretary of State. Id. at
831:14-22, 832:10-833:6. Ms. Donovan did not have any similar experiences with individual
contributors, id. at 815:7-9, nor did she recall any news stories or specific stories from her former
colleagues in the Colorado state Senate with similar experiences, id. at 829:6-14. I find Ms.
Donovan’s testimony to be credible, even though she could not remember every detail regarding
her candidate interview with the group and did not report the incident.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
20 of 142

21
D. Expert Witness Testimony
The parties presented testimony from several experts. Dr. David Primo, Benjamin Engen,
Dr. Christopher Bonneau, and Dr. Damon Cann testified on behalf of Plaintiffs. Dr. Abby Wood
and Dr. Douglas Spencer testified on behalf of Defendants.
1. Dr. David Primo
Dr. David Primo, a professor of political science and business administration at the
University of Rochester. He focuses his research on the effect of rules, institutions, and strategy
on democratic outcomes, including the effect of campaign finance rules. Trial Tr. at 219:14-21.
His research includes peer-reviewed and non-peer-reviewed publications. Id. at 220:4-221:6.
The expert opinion Dr. Primo formed on Colorado’s campaign contribution limits and
public perceptions of government relates to trust in government, how people feel about their
government, and whether they believe their government is corrupt. Id. at 224:22-225:7. He
testified that “Americans think campaign contributions . . . [are] the source of corruption.” Id. at
317:21-23. Nevertheless, Dr. Primo concluded that Colorado’s contribution limits exerted no
meaningful effect on quid pro quo corruption or the appearance thereof. Id. at 225:11-15. While
Coloradans were included in his dataset, Dr. Primo did not perform a separate Colorado-specific
analysis in forming his opinion. Id. at 309:1-21. He did not analyze Colorado voter attitudes
about quid pro quo corruption before or after Amendment 27’s enactment, id. at 313:18-19, or
whether the existence of quid pro quo corruption would be different in Colorado if the
contribution limit amount was changed, id. at 313:20-314:9.
Dr. Primo quoted the statement in Buckley v. Valeo, 424 U.S. 1 (1976) (per curiam),8 that
8 Buckley and its importance to this case is discussed below in my Conclusions of Law.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
21 of 142

22
“Congress could legitimately conclude that the avoidance of the appearance of improper
influence is also critical if confidence in the system of representative government is not to be
eroded to a disastrous extent.” Id. at 232:18-22. From this, he gleaned that the Supreme Court
“care[s] about the appearance of corruption because of its effects on confidence and trust in
government.” Id. at 232:23-233:1. Thus, he focused his opinions on the public’s trust and
confidence in government as a proxy for corruption and the appearance of corruption, but not
specifically quid pro quo corruption or its appearance. Id. at 232:24-233:1, 319:5-320:1. When
confronted with his deposition testimony, he conceded that a lack of corruption is not a necessary
component of trust in government and a lack of quid pro quo corruption is not sufficient to create
trust in state government. Id. at 322:21-323:2.
Contribution Limits and the Appearance of Corruption
Dr. Primo’s first analyzed the effect of state level contribution limits on the appearance of
quid pro quo corruption. He created five variables to classify state level campaign finance laws,
which he called “campaign finance regimes.” Id. at 244:5-7. He defined “Regime 0” as no
restrictions, meaning the state has no contribution limits or public funding. Id. at 244:25-245:1.
“Regime 1” was defined as states with only corporate contribution limits. Id. at 24:8-9. “Regime
2” was defined as states with both corporate and individual contribution limits. Id. at 247:14-18.
During the thirty-year period Dr. Primo studied, to the best of his knowledge, there was never a
state that had individual contribution limits but no corporate contribution limits, which is why he
did not separate those limits in his study. Id. at 248:19-22. He did not account for whether the
contribution limit amount impacted the perception of trust in government because he decided
that the amount was not as important as whether or not a contribution limit existed. Id. at 274:16-
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
22 of 142

23
22, 275:8-9, 276:20-22. “Regime 3” was defined as states with corporate and individual
contribution limits and gubernatorial public funding. Id. at 245:15. “Regime 4” was defined as
states with corporate and individual contribution limits and gubernatorial and legislative public
funding. Id. at 245:15-16. Based on the state’s campaign finance laws at a given time, a state
could only be in one regime at a time. Id. at 246:3-4.
Dr. Primo created a dataset for individual level data from a nationwide survey he and a
colleague conducted in 2015-2016, id. at 236:2-4, and nationwide surveys conducted by other
organizations between 1987-2017, id. at 231:1-7, 236:4-7. His peer-reviewed findings were
published as part of a book in 2020. Id. at 236:8-14. Dr. Primo’s 2015-2016 survey contained
responses from people across the United States, including Coloradans, to questions, such as
“[D]o you believe that the following routine political activities are likely to be corrupt if a
politician were to take a particular political position for the following reasons?” Id. at 229:13-16;
see also Trial Ex. 305. His survey showed, for example, that 63% of Americans thought that a
politician taking a particular position because of a political party leader’s pressure was a corrupt
action. Trial Tr. at 229:19-22. Based on the responses, Dr. Primo opined that courts’ definition of
corruption is different from the American public’s definition, or image, of corruption. Id. at
237:17-24. He concluded that Americans think campaign contributions and political pressure are
sources of corruption; Americans think that “everything in politics is basically corrupt.” Id. at
317:21-25.
The approximately 50 or so nationwide surveys conducted by other organizations
between 1987-2017 included nearly 60,000 responses. Id. at 236:4-7, 243:11-12. These surveys
involved questions like, “How much trust and confidence do you have in [your] state
government when it comes to handling state problems?” Id. at 250:21-23. A person could
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
23 of 142

24
respond with either “not at all,” “not very much,” “a fair amount,” or “a great deal.” Id. at
250:23-24. Dr. Primo applied a scale from 0 to 100 to the different survey answers. Id. at 251:8-
9. The scale was divided into equal parts depending on whether there were four or five possible
responses. Id. at 251:18-20. Dr. Primo utilized the midpoint of the divided part to score the
response.9 Id. at 252:3. He then combined the state level and individual level data so that each of
the individual respondents’ data attached to the features of the state in which they lived. Id. at
240:16-20, 241:13-15.
Dr. Primo ran a regression analysis on this data. Id. at 253:23-24; Trial Exs. 119, 120. A
regression analysis aims to answer the question of what is X’s, the independent variable’s, effect
on Y, the dependent variable. Here, the dependent variable was not quid pro quo corruption or its
appearance. Trial Tr. at 320:19-23. Instead, the dependent variable Dr. Primo chose was the 0 to
100 scale about individual’s trust and confidence in state government. Id. at 261:23-25, 319:25-
320:1. The independent variables were the Regime categories he created. Id. at 262:2-5. Dr.
Primo also included fixed effects, or controls. One fixed effect gave a state its own variable to
identify responses from that state, id. at 265:19-20, and another gave the year of response its own
variable, id. at 266:9-12.
Dr. Primo’s regression analysis showed that the level of trust people have in their
government does not statistically differ between Regime 1 and Regime 2. Id. at 272:14-19. He
clarified that, compared to Regime 0, a person living in a Regime 1 state lands 1.88 points lower
on the trust scale and a person living in a Regime 2 state lands 1.74 points lower. Id. at 271:5-11,
271:22-2; Trial Exs. 119, 120. In other words, people living in a state with no contribution limits
9 Dr. Primo admitted that he had not previously seen other researchers use this type of measuring
approach. Trial Tr. at 252:1-15.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
24 of 142

25
or public financing are statistically indistinguishable in terms of trust in government from those
living in states with corporate contribution limits or corporate and individual contribution limits.
Id. at 273:25-274:4.
Dr. Primo also analyzed his results using a robustness check. See Trial Exs. 121, 122.
Controlling for all other factors, he examined his identified trends in state trust before and after a
reform went into effect in the respondent’s home state. Id. at 279:17-21. The data did not show
any “strange or weird or important” changes in trust before or after corporate contribution or
individual contribution limit reforms were enacted. Id. at 279:22-280:1; Trial Exs. 121, 122. This
ensured the observed results of his regression analysis were not based on changes in contribution
limit laws that happened at points where trust in government was trending one way or the other.
Trial Tr. at 278:21-279:1.
Based on his regression analysis, Dr. Primo testified that “[i]f a state were to eliminate all
contribution limits and public funding systems today, it should expect nothing much to happen
with regard to trust and confidence in state government.” Id. at 274:7-10. He did not study the
impact of the amount of a contribution limit; he analyzed only whether the limit existed in a
state. Id. at 302:25-303:8. However, other “experimental studies” have found that the public sees
“every amount of campaign contributions as corrupting, so it doesn’t really matter the amount.”
Id. at 277:2, 277:7-11. Dr. Primo acknowledged that his research did not directly test the
appearance of corruption. Id. at 323:17-24. His research suggests, however, that quid pro quo
corruption is considered the most or nearly the most corrupt activity. Id. at 339:13-23 (explaining
that out of all the activities respondents labeled as corrupt “the general concept of quid pro quo
corruption, however we wrote it up for the questions, was, I’m pretty certain, at the top”). This is
consistent with research done by Dr. Douglas Spencer and Dr. Alexander Theodoridis, as
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
25 of 142

26
discussed below. Id. at 339:24-340:1.
Contribution Limits and Corruption
Dr. Primo’s second analysis focused on the effect of contribution limits on actual quid
pro quo corruption. He first performed a literature review. He did not identify any published,
peer-reviewed research studying the effects of campaign finance laws or contribution limits on
quid pro quo corruption in the United States. Id. at 289:5-8. This is likely because it is difficult to
find and study these effects. Id. at 289:14-18. Second, Dr. Primo reviewed the materials
produced in this case. Similarly, he did not identify any produced evidence that would show the
effect of contribution limits on quid pro quo corruption. Id. at 289:25-290:3, 312:15-19. He
hypothesized, but did not test, that if a politician was going to act corruptly, he or she would find
a different or more creative way to do so than a $500 contribution. Id. at 291:18-21, 306:8-14.
His ultimate opinion was that there is no evidence to show that quid pro quo corruption is related
to the strength of a state’s campaign finance laws. Id. at 288:17-20.
Rebuttal to Dr. Wood
Dr. Primo rebutted Dr. Abby Wood’s expert opinions. His review of the results in her
report identified a statistical error.10 Id. at 284:7-10. I will discuss the substance of Dr. Wood’s
testimony in the next section of my order, but Dr. Primo critiqued her approach to placing states
into “strengthening” or “weakening” categories. He explained that a state remains a
strengthening state for the entire study period if it strengthens its campaign finance laws at one
10 This expert opinion was not disclosed to Defendants before trial. Trial Tr. at 284:11-12. Dr.
Primo did not file a supplemental expert report in response to Dr. Wood’s report or data, which
he received at least one year before trial. Id. at 297:3-5, 298:3-9.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
26 of 142

27
point and never weakens them. Id. at 285:14-19. The states Dr. Wood coded as weakening
similarly stay in the weakening category. Id. at 285:19-21. Plugging this time-based variable into
a regression analysis with state fixed effects, double counts the data. Id. at 286:10-15. This
causes certain state fixed effects to drop out of the analysis to avoid an error. Id. at 287:7-12.
Findings on Dr. Primo’s Testimony
I do not doubt the credibility or reliability of Dr. Primo’s testimony. His research has
been peer reviewed and is based on a broad, robust dataset. However, I find that Dr. Primo’s
research design makes his opinions unreliable and minimally relevant in this case. His opinions
are offered as evidence on a key consideration: Colorado’s interest in preventing quid pro quo
corruption or its appearance, which was the stated purpose for adopting its campaign
contribution limits. Yet Dr. Primo did not directly analyze Colorado voter attitudes about quid
pro quo corruption or the appearance thereof before or after Amendment 27’s enactment.
Amendment 27 was added to the Colorado ballot on a citizen’s initiative, making Colorado
voters’ opinions on corruption even more pertinent to this constitutional challenge.
Dr. Primo’s research focuses on trust and confidence in government rather than
corruption or its appearance. He collected data on whether Americans believed that certain
“routine political activities are likely to be corrupt if a politician” takes a position, for example,
to gain favorable media coverage or due to pressure from party leaders. This relates to the
electorate’s view of the government rather than the impact of contribution limits on corruption or
its appearance. While some Coloradans were included in his overall database, it does not address
Coloradans’ specific views. The surveys performed by other organizations that produced the
remainder of Dr. Primo’s broad dataset did not appear to ask respondents about corruption or a
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
27 of 142

28
similar concept. Instead, those surveys asked about respondents’ trust in their government to
perform certain actions or solve problems. Again, this does not relate to the electorate’s view on
corruption or its appearance. Further still, Dr. Primo’s conversion of every survey response to a
100-point scale is imprecise. Although this approach attempts to standardize responses across
surveys with a varied number of answer choices, it assigns unequal weight to responses
depending on the survey format and removes distinction between responses. See Trial Ex. 305;
see, e.g., Trial Tr. 1016:18-1017:7.
Finally, based on my review of the evidence, Dr. Primo’s emphasis on Americans’ view
that all is corrupt in politics does not lead to the inference that Colorado voters adopted the
campaign contribution limits without quid pro quo corruption in mind. His survey asked
respondents whether certain activities were corrupt, and his testimony did not persuade me that
this provides insight into the Colorado electorate’s view of limiting contribution limits to avoid
the appearance of corruption in politics. I explore these deficiencies in Dr. Primo’s research
further in my Conclusions of Law.
2. Dr. Abby Wood
Dr. Abby Wood is a professor of law and political science at the University of Southern
California. She focuses her research on governance bureaucracy and campaign finance, which
includes disclosures, limits, and bans. Trial Tr. at 979:8-20. Her research includes peer-reviewed
and non-peer-reviewed publications. Id. at 980:1-6. Dr. Wood was called to testify as a rebuttal
expert to Dr. Primo and Dr. Bonneau11 and opined on the Colorado electorate’s concern with
11 I do not find the portion of Dr. Bonneau’s opinion that Dr. Wood rebuts to be relevant to my
analysis. As such, I do not address it in my Findings of Fact.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
28 of 142

29
quid pro quo corruption. Id. at 988:18-25. Her testimony centered on her finding that “the
Colorado electorate that enacted Amendment 27 and still the Colorado electorate . . . of today is
concerned with quid pro quo corruption and not only [undue] influence.” Id. at 988:22-25. None
of her own statistical research or analysis that she presented during trial was peer reviewed. Id. at
1044:17-19.
To determine whether the phrase “corruption or appearance of corruption” in § 1 of
Article XXVIII is limited to undue influence, Dr. Wood analyzed the text of Amendment 27,
national surveys, and newspaper data. Id. at 990:10-21. She noted that the people of Colorado
intended corruption and undue influence to be separate measures because the text of Amendment
27 included those two concepts as two separate bullet points.12 Id. at 991:9-14. Those two
concepts are described in the purposes and findings of Article XXVIII:
The people of the state of Colorado hereby find and declare that large campaign
contributions to political candidates create the potential for corruption and the
appearance of corruption; that large campaign contributions made to influence
election outcomes allow wealthy individuals, corporations, and special interest
groups to exercise a disproportionate level of influence over the political
process. . . .
Trial Ex. 357 at 2. Thus, she opined that, in adopting Amendment 27, the people of Colorado
were concerned with quid pro quo corruption specifically.
Americans’ and Coloradans’ Views on Corruption
Using data from Dr. Douglas Spencer and Dr. Alex Theodoridis, Dr. Wood ranked
respondents’ views of corruption based on seven political actions: self-funding one’s campaign,
12 Based on the admitted trial exhibits, I do not see that bullet points were provided to the
electorate. Instead, I see that corruption and undue influence were included in the purpose of
Amendment 27 as two separate clauses. See Trial Ex. 357. This leads me to agree with Dr.
Wood’s findings related to the Colorado electorate’s intentions.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
29 of 142

30
holding more meetings with donors than non-donors, accepting money from organizations that
do not disclose its donors, accepting a high-paying job in an industry helped while the candidate
was in office, meeting more with lobbyists than the general public, promoting donors’ interests
over the general public’s interests, and exchanging votes for money. See Trial Tr. at 992:4-5;
Trial Ex. 318. Dr. Wood analyzed the results of nationwide responses from voting-age adults and
then isolated twenty responses from Coloradans. Trial Tr. at 992:21-25, 993:11-13, 1052:1-3,
1053:13-14. From there, she estimated Coloradans’ views on corrupt activities. Id. at 992:21-25,
993:11-13, 1052:1-3, 1053:13-14. She did not resample the Colorado data in this portion of her
analysis. Id. at 993:2-16.
The data showed that Americans and Coloradans ranked self-funded campaigns as the
least corrupt action of the seven identified. Trial Ex. 318. Americans and Coloradans ranked
exchanging money for votes as the most corrupt. Id. Unlike most Americans, however,
Coloradans are less concerned with self-funded campaigns and politicians meeting more with
donors than non-donors. Trial Tr. at 994:18-20. Dr. Wood admitted that by estimating the data,
the confidence intervals were wider for the Colorado-specific data compared to the national data,
id. at 993:14-16, and she was able to create statistically significant results, id. at 1055:3-7. Wider
confidence intervals mean the measurement of that result is less precise than a result that showed
a narrower confidence interval. Id. at 1063:2-5.
Again using Dr. Spencer and Dr. Theodoridis’s data, which contained twenty responses
from Coloradans, Dr. Wood estimated the percentage of Americans and Coloradans who believe
quid pro quo corruption is more problematic than three political actions: favoring donor interests
over the public, meeting more with lobbyists than the public, and meeting more with donors than
non-donors. Id. at 995:6-9, 1052:1-3; Trial Ex. 319. She found that Coloradans are more likely to
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
30 of 142

31
believe that bribery is more corrupt than each of those three items; specifically, they were much
more likely to believe that bribery is more corrupt than meeting more with donors than non-
donors. Trial Tr. at 995:16-20; Trial Ex. 319. She concluded that Coloradans were most
concerned about quid pro quo corruption. Trial Tr. at 998:10-11.
In this aspect of her study, Dr. Wood resampled, or bootstrapped, the Colorado-specific
data from Dr. Spencer and Dr. Theodoridis’s data 10,000 times to generate a confidence level.
Id. at 995:23-996:1, 996:22-23, 997:14-17. She noted that by resampling the Colorado data, the
differences between the national data and the Colorado data became statistically significant. Id.
997:6-7. Her resampled data showed that Coloradans are more concerned about quid pro quo
corruption compared to the United States population. Id. at 998:13-16. Based on her review of
the data, Dr. Wood was confident that Coloradans reflected the same opinions as shown in the
national dataset. Id. at 998:10-12. She explained that if she had based her opinion on the twenty
responses from Coloradans, meaning if she chose not to resample, that would remove the
statistical significance of her conclusions, but would not change her conclusions. Id. at 998:4-12.
Dr. Wood also collected data from other sources to analyze Coloradans’ concern with
quid pro quo corruption. She used data from the American National Election Study (“ANES”)
for 2000 and 2002 and the Cooperative Congressional Election Study (“CCES”) for 2006, 2007,
and 2008. Id. at 999:7-8, 12; Trial Ex. 320. The ANES survey data, which had thirty-two
responses from Coloradans, was based on respondents’ answers to whether they think that “quite
a few of the people running the government are crooked,13 not very many, . . . or . . . hardly any
of them are crooked.” Trial Tr. at 999:20-23, 1056:6-9. The CCES survey data was gathered
13 Merriam-Webster includes “crooked” as a synonym for “corrupt.” Corrupt, Merriam-
Webster.com, https://www.merriam-webster.com/thesaurus/corrupt (last visited March 26,
2026).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
31 of 142

32
from respondents’ answers to what they believe is the most important problem facing the country
today, and one possible response was corruption. Id. at 999:24-1000:4. Dr. Wood’s analysis
showed that in 2000, compared to the average American, Coloradans were 8% more likely to
find that “quite a few” people running the government were crooked and 4% less likely to find
that “hardly any” were crooked. Id. 1000:7-12; Trial Ex. 320. In 2002, compared to the average
American, Coloradans were 3% more likely to find that “quite a few” people running the
government were crooked and 12% less likely to find that “hardly any” were crooked. Id.
1000:13-17; Trial Ex. 320. In 2006, 2007, and 2008, about 10% of Coloradans and Americans
chose corruption as the biggest problem facing America. Trial Tr. at 1001:23-25; Trial Ex. 320.
As the last part of her research, Dr. Wood analyzed sixty-two Colorado-based newspaper
articles written in 1998 through 2002 that mentioned government or political candidate
corruption. Trial Tr. at 1004:1-2, 6; Trial Ex. 321. A “nontrivial” number of these articles were
opinion pieces. Trial Tr. at 1060:13-17. She divided the articles into four categories: quid pro
quo only, influence only, influence or quid pro quo, and other. Id. at 1004:10-13. Of the sixty-
two articles she reviewed, eleven discussed only quid pro quo corruption, four discussed only
improper influence, twenty-nine discussed quid pro quo corruption or improper influence, and
eighteen did not fall into one of the other categories. See Trial Ex. 321. Thus, she found that
17.7% of the articles discussed only quid pro quo corruption whereas 6.5% of articles discussed
only improper influence. Trial Tr. at 1005:7-10. She ran a proportional t-test, meaning she
calculated the difference between the proportion of articles that discussed only quid pro quo and
those that discussed only improper influence with the sample size. Id. at 1005:11. She found that
the difference was statistically significant, i.e., the observed outcome in her proportional t-test is
unlikely to have occurred by chance. Id. at 1105:14-15.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
32 of 142

33
Rebuttal to Dr. Primo
Dr. Wood was also called to rebut Dr. Primo’s testimony. She identified three issues with
Dr. Primo’s opinions. She first took issue with his failure to include contribution limit amounts
in his study. Id. at 1007:9-10. Instead, as previously explained, Dr. Primo created five campaign
finance “regimes” for his study, these were whether the state had (Regime 0) no contribution
limits, (Regime 1) corporate limits, (Regime 2) corporate and individual limits, (Regime 3)
corporate and individual limits plus gubernatorial public financing, or (Regime 4) corporate and
individual limits plus gubernatorial and legislative public financing. Id. at 1023:2-20. Trial Ex.
119. Dr. Wood opined that by having all states with individual contribution limits set as Regime
Two, Dr. Primo simply looked at the effect of adopting any contribution limit. Trial Tr. at
1013:19-22. In fact, she called Dr. Primo’s regime categorization “coarse” due to the significant
variation within each regime. Id. at 1018:18-1019:4.
Second, Dr. Wood opined that Dr. Primo’s judgment calls in his measurement and
modeling undermined his findings. Id. at 1007:13-18. Trust or lack thereof does not equal
corruption, and his analysis ignores other issues that affect trust in government. Id. at 1014:3-4,
17-19. She also opined that this portion of his analysis pooled together trust and confidence,
which creates noise and, based on the questions asked, may not trigger an ethics-based response
from respondents. Id. at 1016:16-18. She also believed that Dr. Primo’s measuring of
responses—a sliding scale providing none, very little, some, quite a lot, and a great deal as
choices—could lead to measurement error because of individuals’ differing conceptions of those
choices. Id. at 1016:23-2, 1017:5-7. Dr. Wood’s analysis of Dr. Primo’s data only looked at the
percentage of respondents who gave the top answer, which removed any assumptions between
the other sliding-scale choices. Id. at 1018:10-14.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
33 of 142

34
Third, Dr. Wood opined that Dr. Primo omitted what she considered a key variable:
whether a state is strengthening into a higher regime or weakening into a lower regime. Id. at
1007:22-24. Including this variable, she found, would alter the results of Dr. Primo’s regime-
based analysis. Id. This variable is important because she would expect to see a difference and a
change in trust levels based on whether the state is moving between the regimes Dr. Primo
described. Id. at 1021:23-1022:3. Because he did not account for trust levels in states that moved
into different regimes, the changes in trust would average out close to zero in his findings. Id. at
1022:10-16.
In this so-called regime-trajectory study, Dr. Wood used the five regimes created by Dr.
Primo and further classified the states based on whether they are “strengthening,” or moving up
regime levels, “weakening,” or moving down regime levels, “mixed,” or moving up and down
regime levels, or the regime level does not change. Id. at 1025:5-8; Trial Ex. 315. If a state
changed its campaign finance laws in either direction it would always be included in the
“strengthening” or “weakening” category, regardless of the year that change was instituted. Trial
Tr. at 1047:3-11, 14-20. Dr. Wood did not initially conduct a regression analysis but instead
calculated the difference between states based on raw data. Id. at 1028:8-15. Her analysis found
that when a state moved from Regime 0 to Regime 2, the raw data showed that respondents were
1.7% more likely to select the top rating for trust. Trial Tr. at 1025:15-20; Trial Ex. 315. When a
state moved from Regime 1 to Regime 2, respondents were 4.5% more likely to select the top
rating for trust. Trial Ex. 315. When she looked at the weakening states, when a state moved
from Regime 3 to Regime 2, respondents were 3.1% less likely to select the top rating for trust.
Id. This led Dr. Wood to conclude, using Dr. Primo’s measures and adding in the regime-
trajectory variable, that if a state strengthens its campaign finance laws from having no limits to
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
34 of 142

35
having corporate and individual limits, the people may have a higher trust in their government.
Trial Tr. at 1027:15-19. Based on her findings, if she had designed Dr. Primo’s study, she would
have controlled for the regime changes. Id. at 1027:24-1028:1.
Dr. Wood then ran a regression analysis using Dr. Primo’s model and included her
regime-trajectory variable. Id. at 1030:18-24; see Trial Exs. 315, 316. She ran this additional
analysis to see if the effect observed by Dr. Primo would change. Trial Tr. at 1031:2-6. As
previously discussed, Dr. Primo criticized Dr. Wood’s regression analysis. She addressed this in
her testimony and admitted that by dividing Dr. Primo’s regimes for her regime-trajectory
variable, she inadvertently divided the data such that the “state” fixed effects, or a control for
state in the analysis, for seven of the states were unable to compute or, as she described it, those
fixed effects dropped out. Id. at 1031:16-18, 1032:4-6. Dr. Wood opined that the state fixed
effects dropped out for seven of the states in her regression analysis because, after dividing the
responses based on regime trajectory, there may have only been one state in a division, which
would cause the state fixed effect to drop out. Id. at 1033:8-13. Dr. Wood testified it would not
be possible to study the impact of the regime-trajectory variable on Dr. Primo’s model without
the state fixed effects dropping. Id. at 1037:3-6. In response to Dr. Primo’s criticism, she reran
her analysis and excluded the state fixed effects. Id. at 1036:17. Without the fixed effects, her
findings remained generally constant, but some statistical significance disappeared and the size
of some estimates shrank. Id. at 1036:17-23. Thus, she testified that Dr. Primo’s criticism did not
impact the results or conclusion of her regime-trajectory study. Id. at 1037:10-13.
Dr. Wood criticized Dr. Primo’s model for including a large number of fixed effects.
First, she explained that Dr. Primo’s model included many controls and fixed effects, but he
failed to show any build-up of those controls, which would show how he arrived at his final
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
35 of 142

36
conclusion. Id. at 1034:19-1035:7. When numerous controls and fixed effects are included in a
model, this tends to help a researcher who is “trying to show the relationship exists.” Id. at
1036:6-8. Because Dr. Primo did not show the build-up to his conclusion, it is not possible to
observe any pressure applied to the relationships in his model or any progress made to establish
the final results. Id. at 1034:22-1035:14. Second, in explaining why state fixed effects dropping
from her analysis was not necessarily a reason to discount her analysis, Dr. Wood noted that Dr.
Primo also experienced “party” fixed effects dropping out during his analysis. Id. at 1032:1.
Findings on Dr. Wood’s Testimony
To start, Dr. Wood admitted she made a mistake in her analysis of Dr. Primo’s research:
Before reaching her conclusions, she did not notice that certain state fixed effects had dropped
out of her model. Although this may happen in her line of work, it is troubling that Dr. Wood
missed this occurrence in her final report. It is also troubling, however, that Dr. Primo did not
submit a supplemental report identifying this error and instead chose to inform Dr. Wood of her
error for the first time during his trial testimony. Dr. Wood later testified that she removed the
state fixed effects from her model and found that her findings remained generally consistent with
the findings in her report, minus some statistical significance. This lends confidence that her
testimony is reliable, although I would not say the same thing if her written expert report was
admitted as a trial exhibit.
Plaintiffs also argue that because Dr. Wood’s research did not undergo a peer-review
process, I should discount it. I disagree. Presenting peer-reviewed data is not a prerequisite to
expert testimony. Dr. Wood categorized Dr. Primo’s regime-based variables into more specific
categories for her analysis. A lack of peer review does not make that data unreliable. Thus I will
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
36 of 142

37
consider and give it weight in my analysis.
As for the remainder of her testimony, Dr. Wood’s review of Dr. Spencer and Dr.
Theodoridis’s survey and the other sets of national surveys focused on corruption or
“crookedness,” which is relevant to the questions presented here, or at least more relevant than
trust in government as a corollary to quid pro quo corruption. I have no reason to doubt the data.
I understand Dr. Wood resampled the responses from Coloradans in those surveys to obtain
statistically significant findings. Although the number of Colorado-specific responses was low,
resampling is a generally accepted practice in statistics. I will accept Dr. Wood’s testimony as
part of my analysis here.
3. Benjamin Engen
As previously stated, Mr. Engen works as a political consultant, economic consultant, and
staff analytics engineer. Id. at 590:1-12. He earned his undergraduate degree in political science
and economics from the University of Colorado. Id. at 595:7-17. He also completed the
equivalent coursework in mathematics to qualify for a mathematics major. Id. at 595:17-22. Mr.
Engen holds a master’s degree in business administration, which included classes on quantitative
methods for operations and marketing, but he has not earned a Ph.D. Id. at 595:25-596:3, 596:24-
597:9. He has not published any papers in a peer-reviewed journal. Id. at 598:13-15.
As an expert for Plaintiffs, Mr. Engen presented five conclusions based on his research
regarding the impact Colorado’s campaign contribution limits has had on Colorado’s elections.
First, he opined that after 2002, there was a noticeable decrease in the number of first-time state
Senate candidates winning their general election races compared to incumbent candidates. Id. at
604:15-19, 605:8-17. Second, he opined that legislative campaigns are experiencing so-called
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
37 of 142

38
income disparity because the campaigns that were already raising the most money are raising
more and more money compared to those campaigns that were raising less money. Id. at 605:18-
606:1. Third, he opined that because the cost of running a campaign has increased, it is more
difficult for candidates to reach all the voters in their district. Id. at 606:2-8. Fourth, he opined
that unaccountable actors, such as independent expenditure committees, are holding more
influence in elections. Id. at 606:9-20. Finally, he opined that contribution limits have created the
opportunity for wealthy candidates to have an advantage in their campaigns. Id. at 606:21-607:3.
I have no reason to doubt or question the data on which Mr. Engen based his opinions;
the TRACER database and election results as maintained by the Colorado Secretary of State are
trustworthy sources. I do, however, find a flaw in the foundation of his analysis for his first
opinion regarding post-2002 state Senate general elections and second opinion regarding so-
called income disparity between campaigns. Mr. Engen chose to define “incumbent” to mean a
candidate “who has previously held a Colorado[-]level office, regardless of the” office or district
represented. Id. at 608:24-609:1. He admits that this is different from the definition used by most
academics and political scientists. Id. at 609:2-6. All other experts in this case who rely on the
definition of incumbent for their opinions use a different definition than the one chosen by Mr.
Engen. Mr. Engen did not convince me that his outlier definition is more accurate than the
traditional, widely accepted definition of “incumbent.”14 This purposely chosen definition skews
14 Mr. Engen explained that he “narrowed the data set that [he] used to only legislative races
specifically to control for confounding factors” such as the fact that his definition of incumbent
would include, for example, the Denver mayor running for governor. Trial Tr. at 787:13-788:18.
And, with that narrowed data set, there were “only two instances in which a candidate
transitioned from first-time to incumbent in a way that would be at odds with the definition [of
incumbent] proposed by the defense.” Id. at 788:19-22. “Those two scenarios were, one, the
scenario of a state representative running for the state senate, and, two, a state representative or
state senator’s district number changing as a result of reapportionment . . . .” Id. at 788:12-789:1.
This explanation does not cure my concerns. Crafting a data set to fit an outlier definition raises
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
38 of 142

39
Mr. Engen’s results as more candidates are considered incumbent candidates. The fact that Mr.
Engen’s conclusions are based on a definition that is starkly different from the standard one used
by political scientists and academics creates not only a lack of confidence in his related opinions,
but also doubt about his expert testimony as a whole. His use of an outlier definition strikes me
as self-serving to meet what he believes ought to be the end conclusion. Mr. Engen’s
questionable methodology and credibility combined with his lesser qualifications and lack of
peer-reviewed publications cause me to find that his opinions based on his analysis of
incumbents are not reliable or persuasive, especially compared to those of the other well-
qualified experts in this case. Thus, I disregard Mr. Engen’s first opinion and do not detail it as
part of my Findings of Fact.
Further, as I explain in my Conclusions of Law, three of Mr. Engen’s opinions—his
second opinion that legislative campaigns are experiencing so-called income disparity and his
last two opinions that unaccountable actors are holding more influence in elections and that
contribution limits have given wealthy candidates an advantage in their campaigns—are not
helpful in evaluating the considerations prescribed by the U.S. Supreme Court. Thus, I do not
include those opinions or their bases in my Findings of Fact.
I do accept Mr. Engen’s third opinion—that it is more difficult for candidates to reach all
voters because campaign costs have increased. The cost of every voter contact from a campaign,
which includes direct mail campaigns, printing yard signs, and paid advertising, has increased.
Id. at 684:2-13. The increased costs have made it impossible for campaigns to reach every voter
in a candidate’s district. Id. at 686:4-6. Mr. Engen used the price of stamps as an example. In
more red flags, and I am not persuaded that a state representative running for state senate is an
incumbent, even if that candidate might experience some of the same benefits an incumbent
candidate does. See infra note 45.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
39 of 142

40
2002, when Amendment 27 was enacted, the price of a stamp was $0.37; now it is $0.72. Id. at
688:9-13. Postage alone would be a large campaign expense, more than the average state Senate
campaign could spend to reach every voter in a district. Id. at 685:23-686:3. He explained that
inflation has increased “a cumulative of 64.7 percent over the course of [2002 through the trial
date], while the ability of [legislative] campaigns to raise contributions . . . has only increased by
12.5 percent.” Id. at 688:14-19; see also Trial Ex. 40. These facts and Mr. Engen’s opinion on
this matter are relevant and useful for my analysis.
4. Dr. Christopher Bonneau
Dr. Christopher Bonneau is a professor of political science at the University of
Pittsburgh. His empirical research focuses on state supreme court elections, judicial elections,
election competitiveness, electoral contestation, and campaign spending. Id. at 370:20-371:7. He
has published peer-reviewed articles on campaign finance, including with Dr. Damon Cann, who
also served as one of Plaintiffs’ experts here. Id. at 371:22-24, 372:7-16. Dr. Bonneau has not,
however, published any articles or books about campaign finance laws outside of the judicial
elections or judicial politics context. Id. at 373:20-374:12, 374:13-17. Colorado does not have
judicial elections; the state only has retention elections. Id. at 374:18-20. Dr. Bonneau argued
that his research translates beyond judicial elections because the same factors influence different
types of elections. See id. at 394:1-15. Here, Dr. Bonneau opined on the impact of campaign
finance law on elections, election competitiveness, and democratic representation.
Contribution Limits and Competitive Elections
Dr. Bonneau testified about his survey of mostly peer-reviewed publications on whether
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
40 of 142

41
Colorado’s campaign finance laws adversely affected the ability of non-incumbent candidates to
successfully compete in elections. Id. at 376:2-8. He found that Colorado elections are not
particularly competitive in that regard, and he concluded that the likely reason is because the
limits on campaign contributions lead to low campaign expenditures. Id. at 377:12-15. Without
limiting to Colorado, he opined that individual campaign contribution limits are a cause of the
disadvantage challenger candidates have when competing against an incumbent candidate. Id. at
380:6-11, 396:21-24. Dr. Bonneau noted that Colorado’s contribution limits are lower than all of
Colorado’s neighboring states. Id. at 388:16-19.
Dr. Bonneau defined a “competitive election” as “an election in which two candidates
have a decent chance of winning, or more than one candidate has a decent chance of winning.”
Id. at 378:1-3. This definition takes into account indicators such as percentage of votes received
and incumbent losses. Id. at 387:11-12. Dr. Bonneau explained that in elections where the
incumbent15 is running, the playing field is not level. The incumbent enjoys what is known as the
“incumbency advantage,” which is an advantage gained from, for example, name recognition,
having a well-established funding base, and the ability to benefit and communicate with voters in
the district. Id. at 378:18-379:4. Fundraising is one major aspect of the incumbency advantage;
an incumbent generally has a larger established fundraising network than a challenger. Id. at
379:5-13. Dr. Bonneau agreed that contribution limits are passed to reduce corruption, but he
opined that these laws have led to an unintended consequence of giving an advantage to
incumbents. Id. at 384:6-13. In forming this opinion, however, he did not study whether there
15 Dr. Bonneau uses the traditional definition of the word “incumbent,” meaning the candidate
holding the office at the time of the campaign. See Trial Tr. at 426:8-11. He confirmed that this
is the standard definition of “incumbent.” Id. at 426:17-19. This reinforces my determination that
Mr. Engen’s definition and related opinions should be discounted.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
41 of 142

42
was a direct link between contribution limits and the likelihood of a challenger defeating an
incumbent. Id. at 398:9-12.
One study in Dr. Bonneau’s survey concluded that states with no contribution limits see
an advantage given to incumbents. Id. at 402:7-10. He attempted to refute a separate study, by
Dr. Thomas Stratmann, that found a $500 cap on individual contributions increases the
likelihood of a close race and of an incumbent’s defeat by explaining the study included
uncontested races, which cannot be competitive. Id. at 389:19-390:8. Dr. Bonneau’s criticisms
do not convince me to ignore this study. Even with the inclusion of uncontested races, which
presumably could not involve an incumbent losing, the researcher found evidence that a $500
cap on individual contributions increased the likelihood of an incumbent’s defeat. If anything,
including uncontested races may have watered down the study’s results.
Overall, Dr. Bonneau’s testimony related to whether Colorado’s campaign finance laws
adversely affected the ability of non-incumbent candidates to successfully compete in elections
does not weigh heavily in my analysis. He did not study any direct link between contribution
limits and the likelihood of a challenger defeating an incumbent, which is a necessary link to
determine the effectiveness of a challenger’s campaign. Further, he did not dispute that if one
looked at just the likelihood of an incumbent candidate losing, contribution limits do not further
entrench the so-called incumbency advantage. Id. at 398:23-399:1. I accept Dr. Bonneau’s
testimony, but I find it either irrelevant to the questions of this case or lacking in robustness to an
extent that I am not swayed by it.
Campaign Spending by Incumbents and Challengers
Much of Dr. Bonneau’s testimony focused on campaign spending. When it comes to
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
42 of 142

43
money spent in a judicial election, he opined, a dollar spent by an incumbent is not equal to a
dollar spent by a challenger. Id. at 382:14-18, 407:5-10. According to Dr. Bonneau, a challenger
benefits more from campaign spending than an incumbent. Id. at 383:7-13. Thus, he concluded
that incumbents benefit from contribution limits. Id. at 383:22-23. However, he admitted that the
study he relied on to reach his conclusion found that while challenger spending provides
additional information about the challenger, it also works to decrease the amount of information
about the incumbent; it does not increase overall voter information insomuch as it levels out the
available information. Id. at 404:1-5, 15-18. I accept Dr. Bonneau’s explanation that judicial
elections and legislative elections are impacted by the same factors. Id. at 394:12-13. Thus, I
consider this aspect of Dr. Bonneau’s testimony and analysis, including the negative impact on
information related to the incumbent.
Contribution Limits and Corruption
Finally, Dr. Bonneau also discussed research related to the effect of campaign
contribution limits on corruption. He concluded that contribution limits do not decrease
corruption. Id. at 387:17-24. He reviewed a 2013 study that analyzed corruption prosecutions and
convictions of all state government workers, including both elected and non-elected officials. Id.
at 418:25-419:4, 423:3-9. This study did not consider any cases in which the prosecution chose
not to charge an individual nor did it consider any data related to the appearance of corruption.
Id. at 419:24-420:9. Dr. Bonneau stated that because the appearance of corruption “is not a
crime,” he was not concerned about the limits of the data in this 2013 study. Id. at 422:15-21.
Relying on this study to opine that there is no evidence that state campaign finance reform,
which includes contribution limits, does not decrease corruption, is too broad a statement based
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
43 of 142

44
on the testimony from Dr. Bonneau. Further, the dataset includes non-elected officials, which
would not show any connection between corruption and contribution limits. I find Dr. Bonneau’s
testimony about this study to be irrelevant and not sufficiently reliable. Therefore, I do not give it
any weight.
5. Dr. Damon Cann
Dr. Damon Cann is a professor of political science at Utah State University. He teaches
research methods and focuses his research on elections and representation. Id. at 439:10-21. He
has published peer-reviewed articles on campaign finance, with a focus on judicial elections,
including with Dr. Bonneau. Id. at 441:13-23, 506:10-11. His testimony centered on the
empirical portion of his and Dr. Bonneau’s study. Id. at 445:5-13.
Dr. Cann opined that Colorado’s campaign contribution limits negatively affect the
competitiveness of the elections in a state. Id. at 444:24-445:1, 446:25-447:3. To reach this
conclusion, he analyzed data from 2010-2016 related to campaign finance, funds raised, election
outcomes, and trends in electoral competitiveness. Id. at 447:5-15. Dr. Cann defined “electoral
competitiveness” as rates of contestation in primary and general elections, amounts of money
raised by incumbents16 and challengers, and the share of votes received by incumbents and
challengers. Id. at 445:15-446:1. He focused on more than just incumbents’ win rates to define
competitiveness because of factors other than the number of votes received impacting that
metric, such as an incumbent choosing not to run for reelection. Id. at 446:5-15; see also id. at
16 Dr. Cann explained that an open seat election in which the current elected official is term-
limited and cannot run means that “it’s not truly an incumbent who’s facing a challenge[r] if
there’s no incumbent running in the race.” Trial Tr. at 458:9-16. This, again, conflicts with Mr.
Engen’s definition of “incumbent,” and further reinforces my decision not to consider his related
opinions.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
44 of 142

45
384:14-18 (incumbent success rates are “one measure of electoral competitiveness, but not
necessarily the best measure”). Unlike Dr. Spencer, Dr. Cann excluded from his primary
contestation analysis those incumbents who do not run either by choice or due to term limits. Id.
at 515:13-15, 1113:6-8.
Primary Contestation Rates
First, Dr. Cann discussed his non-regression analysis of rates of contestation in state
lower-chamber17 legislative primary elections, or whether an incumbent candidate faced a
challenger in the primary. Id. at 449:6-11. He chose not to run a regression analysis in
developing this first opinion because he views primary elections as more homogeneous and thus
require fewer controls. Id. at 555:20-25. While incumbent candidates often do not lose their
primary elections, id. at 509:23-510:1, this analysis is important to show whether voters
exercised a choice in the primary election, especially in a state where districts tend to be won by
one party or the other, id. at 450:5-8, 451:3-12. In those single-party districts, primary elections
rather than general elections are the elections where the incumbent is held most accountable. Id.
at 451:13-16. Although he testified about the importance of single-party districts to his
conclusions, he did not control for single-party districts in his analysis. Id. at 519:9-11.
Using data from Ballotpedia, Dr. Cann analyzed rates of primary contestation in
Colorado and in other states during typical, non-special state legislative election years for state
house elections. Id. at 453:3-17, 454:2-3. He did not include open seat races, or those elections
without an incumbent candidate, in his calculation, even if the reason for that open seat was due
17 Dr. Cann explained his study encompasses state lower-chamber legislative elections. Trial Tr.
at 453:18-20. I will interchangeably use state house and state lower chamber in my Findings of
Fact regarding Dr. Cann’s testimony.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
45 of 142

46
to a term-limited incumbent. Id. at 513:1-3, 20-22. Dr. Cann admitted that term limits create
open seats and tend to attract more challengers than when an incumbent runs for reelection. Id. at
514:5-7,12-14. His decision not to consider races without an incumbent excluded a substantial
portion of the data. In 2010, for example, more than 20% of state lower-chamber legislative
elections were for open seats. Id. at 516:12-14; see Trial Ex. 340. In 2012, nearly 37% of state
lower-chamber legislative elections were for open seats. Trial Tr. at 516:22-24; Trial Ex. 340. In
2014, a little more than 26% of state lower-chamber legislative elections were for open seats.
Trial Ex. 340. And in 2016, a little more than 23% of state lower-chamber elections were for
open seats. Id.
Dr. Cann’s review of the contested primaries data found that Colorado was less
competitive in primary elections than the majority of states: In 2010 it was the least competitive,
in 2012 it was the sixth least competitive, in 2014 it was the third least competitive, and in 2016
it was the eighth least competitive. See Trial Ex. 123. In the 2010, 2012, 2014, and 2016
elections, Colorado saw, at most, 10% of its state representative incumbents face primary
challengers. See id.; Trial Tr. at 456:8-10. However, if Dr. Cann included open seat elections in
his analysis, Colorado would be a more electorally competitive state. See Trial Tr. at 521:20-24);
Trial Ex. 392 (showing Colorado was the eleventh-most competitive state in 2012 for state
legislative elections when ranking the state’s primary contests, open seats, and major party
competition).
Dr. Cann’s study also analyzed the relationship between a state’s contribution limits and
the state’s primary contestation rates. He chose to divide the states into two categories: “very
strict,” which included those states with individual contribution limits less than $1,000 per
candidate in an election cycle and $2,000 limits for political action committees per election
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
46 of 142

47
cycle, and “not strict,” which included all other states, including those without any individual
contribution limits. Id. at 459:19-23, 518:4-6. Dr. Cann created these categories for his analysis.
Id. at 518:11-13. He did not perform any stress testing to determine if other contribution limits
would change his conclusions, and he did not control for any variables. Id. at 518:14-22.
Dr. Cann testified that 20% of U.S. states fell into his “very strict” category. Id. at
459:25. Eight states have state house contribution limits that fall at or below $1,000. See Trial
Ex. 1. The remaining states fell into his “not strict” category. Trial Tr. at 460:1. He found that the
states that individual contribution limits at or below $1,000, i.e., the “very strict” states, averaged
about a 14% primary contestation rate, while the states with higher contribution limits, i.e., the
“not strict” states, averaged between a 23% or 24% primary contestation rate. Id. at 460:2-7;
Trial Ex. 129. Dr. Cann opined that this difference was due to the contribution limits in the states
he categorized as “very strict.” Trial Tr. at 460:10-13. However, in his testimony, he did not
identify the data points that specifically led to this conclusion. Instead, he explained that any
unknown variable causing the difference in primary contestation rates would “have to be
something that was correlated both with being in the strict/not-strict category and the primary
contestation in order for an admitted factor to cause bias.” Id. at 519:2-4. In forming his opinion,
he did not attempt to determine any other potential cause of bias. Id. at 519:5-6.
General Election Contestation Rates
Second, Dr. Cann discussed contestation in state house general elections. To analyze this,
he performed a probit regression. Id. at 462:19. A probit regression analysis is one in which the
dependent variable is binary, meaning Dr. Cann coded his general election variable as either
contested or not contested. Id. at 462:21-463:2. He then identified a set of independent
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
47 of 142

48
variables—including contribution limits, incumbent party, election competitiveness in the state,18
presence of an independent redistricting commission, and whether the election was the first after
redistricting—as control variables to explain why the election was or was not contested. Id. at
463:3-16. He selected these variables based on his review of relevant literature and his own
experience. Id. at 464:9-12.
Dr. Cann used a principal component analysis to create what he called his contribution
limit restrictiveness variable. Id. at 465:2-8. This allowed him to create a single index of what
multiple variables represent to create one variable. Id. at 466:2-8. In this case, he used four
categories of contribution limits as the variables: individual limits, political action committee
limits, labor union limits, and corporate limits. Id. at 465:14-18. These four variables, he found,
correlated to a degree that allowed the single-index variable to be used to summarize a state’s
campaign finance limit regime. Id. at 466:9-17. Only one of these four variables, however, is
challenged in this lawsuit: the individual contribution limit. In fact, Colorado bans corporate and
union campaign contributions, a fact of which he was aware when he selected the variables for
his single-index variable. Id. at 525:7-9. Dr. Cann found that the individual contribution limit
component showed “a very robust correlation” with the single-index variable itself. Id. at
468:16-23. When asked why he chose to use a principal component analysis rather than
accounting for each of the four individual indicators, Dr. Cann explained that his choice provided
a more holistic view of a state’s campaign finance system. Id. at 468:24-469:5. However, Dr.
Cann’s campaign-limit variable not only included three variables that are not being challenged
here—two of which are banned in Colorado—but also excluded political party contribution
18 Dr. Cann did not control for partisanship within districts for the election-competitiveness
variable, Trial Tr. at 527:11-24, but instead looked at a state’s deviation from the 2020
presidential election, id. at 556:15-20.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
48 of 142

49
limits, of which Colorado has among the highest in the nation. Id. at 526:16-20, 527:1-4; Trial
Ex. 375.
Looking at general elections from 2010 through 2016, without detailing causation, Dr.
Cann found that the combined effect of the four factors—contribution limits, incumbent party,
election competitiveness in the state, presence of an independent redistricting commission, and
whether the election was the first after redistricting—is associated with whether a general
election is contested. Trial Tr. at 469:24-470:3. He found that states with more strict campaign
limit regimes, like Colorado, resulted in 3% fewer contested general elections compared to those
states with a less strict regime. Id. at 470:4-14; Trial Ex. 130.
Candidate Fundraising and Campaign Contribution Limits
Third, Dr. Cann analyzed the effect of campaign contribution limits on the amount of
money candidates raise. Using fundraising data from contested state house elections from 2010
through 2016, he performed a “seemingly unrelated regression analysis” on the relationship
between the amount of money candidates can raise and the stringency of the contribution limits.
Trial Tr. at 474:24-475:5, 475:25-476:3, 477:3-4, 477:13-14. A seemingly unrelated regression
analysis estimates two regression models at once. Id. at 477:15-16. In his analysis, he controlled
for partisanship for both the incumbent and the challenger. Id. at 477:19-23. He chose this
control because of debate on whether there is a partisan fundraising advantage. Id. at 477:5-10.
Using the contribution limit, single-index variable he created, described above, Dr. Cann found
that the strictness of contribution limits shows a “substantial effect” on the amount of money a
candidate raises. Id. at 477:24-478:1. The stricter the limits, the less money incumbents and
challengers raise. Id. at 478:1-4. He opined that, holding partisanship constant, if a state with no
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
49 of 142

50
contribution limits were to impose the same limits Colorado has, for example, a challenger
would raise $40,000 less per election cycle and an incumbent would raise $70,000 less per
election cycle. Id. at 478:10-20; Trial Ex. 131.
2010-2016 Colorado House Elections and Contribution Limits
Fourth, Dr. Cann discussed the effect of contribution limits on the 2010-2016 Colorado
House of Representatives election results. He analyzed whether the fundraising impacts
predicted in his third analysis would affect how candidates performed in Colorado elections.
Trial Tr. at 481:21-24. This analysis controlled for the presence of third-party candidates and
competitiveness of the previous election. Id. at 482:4-12. He expected to find diminishing
marginal returns in campaign spending: As the amount of money a campaign spends increases,
the percentage of the vote received per dollars spent decreases. Id. at 483:10-13. For example,
the first $20,000 spent by a state legislative campaign will go farther, “all else being equal,” than
the next $20,000 spent. Id. at 483:14-21.
To start, Dr. Cann looked at the effect of a challenger’s fundraising on the share of votes
earned by an incumbent Colorado state legislator. See Trial Ex. 127. His analysis showed that the
first $10,000 or $20,000 raised by a challenger’s campaign is associated with a statistically
significant drop in the share of votes earned by an incumbent. Trial Tr. at 485:6-10, 489:12-13;
Trial Ex. 127. He focused his analysis on the effect of campaign fundraising but noted that
fundraising and spending “are going to be closely correlated with one another.” Trial Tr. at
486:20-22. He did not analyze the relationship between funds raised and funds spent. Id. at
539:11-13. The rate of return shown by Dr. Cann’s analysis began to level out, i.e., the difference
to the share of votes earned by the incumbent became smaller, once a challenger raised more
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
50 of 142

51
than $50,000 or more. Id. at 485:11-16; Trial Ex. 127. However, the average fundraising for a
Colorado lower-chamber challenger across the 2010-2016 period was $24,680. Trial Tr. at
494:14-18. In his analysis, Dr. Cann assumed a constant average level of incumbent spending.
Id. at 488:5-9. An incumbent who spent more or less than average would not change the shape of
the curve; instead the curve would shift up or down. Id. at 488:10-16. The curve would remain
the same regardless of whether the incumbent raised $0 of additional funds or $70,000 of
additional funds. Id. at 544:12-16; Trial Ex. 127. Thus, under Dr. Cann’s model, even with
higher or lower incumbent fundraising, the value of each dollar spent would remain the same.
Dr. Cann also looked at the effect of an incumbent’s fundraising on the share of votes
earned by the incumbent candidate. See Trial Ex. 128. He found that the effect of incumbent
fundraising was not statistically significant to the share of votes earned by the incumbent; based
on the data, he could not be 95% confident that the observed effect was because of incumbent
fundraising or spending rather than a sampling error. Trial Tr. at 489:10-11, 491:19-22. Dr. Cann
concluded that even though incumbent candidates are able to raise more money under less strict
contribution limits, it is the challengers who are advantaged by the less strict contribution limits
given the larger effect challenger dollars spent on the election outcome. Id. at 489:25-490:1,
493:15-19.
Applying his opinions to Colorado, Dr. Cann opined that if Colorado had no contribution
limits in place for lower-chamber legislative elections, an average challenger would raise
$40,000 more per election cycle. Id. at 494:19-21. An additional $40,000 in fundraising, based
on his statistical modeling, would lead to an average decrease of 3.2% in the share of votes
earned by the incumbent candidate in each election. Id. at 495:4-6. Dr. Cann opined the outcome
of one or two elections a year may change if the incumbent lost 3.2% of his or her share of the
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
51 of 142

52
vote. Id. at 495:12-15. Similarly, without any contribution limits, Dr. Cann concluded that an
average incumbent would raise $70,000 more but this would not lead to a change in the share of
votes earned by the incumbent. Id. at 495:19-496:1. He did not, however, analyze whether
Colorado challengers win elections at lower rates than in states with higher individual
contribution limits. Id. at 546:21-25.
Rebuttal to Dr. Spencer
Finally, Dr. Cann rebutted the testimony of Dr. Douglas Spencer. Dr. Cann stated he was
not persuaded by Dr. Spencer’s analysis related to whether contribution limits impact electoral
competitiveness. Id. at 498:3-4. He opined that Dr. Spencer’s focus on incumbent win rates as a
measure for electoral competitiveness is not the best measure. Id. at 384:14-18, 498:9-13. He
reiterated that an incumbent only runs for reelection when that person is confident he or she can
win. Id. at 498:17-19. Specifically, he explained, because Dr. Spencer’s inquiry simplified
elections to wins and losses, his analysis does not provide information on how close the elections
were. Id. at 385:16-386:16. Dr. Cann admitted that by including state Senate primaries and
additional years of data, Dr. Spencer identified some years that were more competitive than what
Dr. Cann found. Id. at 502:19-25. Dr. Cann argued that, “even with those additional years and
looking at upper chambers, the general finding is that Colorado’s lower-chamber elections and
upper-chamber elections still have well below average levels of . . . primary contestation.” Id. at
503:7-11. But he did not elaborate further or use exhibits to show this conclusion. Further, Dr.
Cann believed Dr. Spencer omitted overall state competitiveness as a variable and limited his
analysis to individual contribution limits. Id. at 499:14-17.
Dr. Cann also disagreed with Dr. Spencer’s use of a fixed effects regression analysis,
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
52 of 142

53
which controls for “any unobserved variation from state-to-state” that is not captured by the
variables included in the statistical model. Id. at 500:1-8. He reran Dr. Spencer’s analysis using a
random effects analysis, which accounts for variation within and among groups of data; Dr. Cann
added the state competitiveness variable and used his single-index contribution limit variable. Id.
at 499:18-22. The results of this analysis led Dr. Cann to conclude that the “more stringent
contribution limits will be associated with higher incumbent win rates.” Id. at 502:4-5.
Findings on Dr. Cann’s Testimony
Although I have no reason to doubt the data on which Dr. Cann based his analyses, I find
his opinions are unreliable and not fully relevant to the facts of this case. For each of the analyses
he conducted, he tailored the data and variables in questionable ways. First, in analyzing primary
contestation rates, he omitted significant data that would have altered his results. I elaborate on
this problem in discussing the rebuttal of his opinions by Dr. Spencer below. Regarding Dr.
Cann’s three other opinions, I find his single-index contribution limit variable produced
conclusions that are not fully relevant to the circumstances here. This case challenges individual
contribution limits, not political action committee limits, labor union limits, or corporate limits.
In fact, as stated above, Colorado bans corporate and union campaign contributions. Dr. Cann
argued that including all four limits provided a more robust view of campaign finance regimes,
but I am not persuaded. I must decide whether Colorado’s individual contribution limits are
constitutional. The inquiry is not whether those limits, combined with other non-relevant limits,
are constitutional. Therefore, I will give his opinions that are based on his single-index
contribution limit variable only minimal weight.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
53 of 142

54
6. Dr. Douglas Spencer
Dr. Douglas Spencer is a professor of law at the University of Colorado Boulder. His
research is focused on voting rights and campaign finance regulation. Id. at 1073:15-18. This
includes peer-reviewed and non-peer-reviewed publications. Id. at 1074:2-16. Dr. Spencer is the
co-author of an Election Law Journal article with Dr. Alex Theodoridis about the corruption and
appearance of corruption standard in campaign finance. Id. at 1074:21-23. He presented three
opinions. First, he opined whether Colorado’s individual contribution limits prevent challengers
from running effective campaigns. Id. at 1080:14-16. Second, he evaluated Colorado’s
legislative elections to determine whether those races are competitive. Id. at 1080:18-19. Third,
he evaluated the extent to which political parties contribute money to Colorado campaigns. Id. at
1080:23-24.
Incumbent Reelection Rates
For his first opinion, Dr. Spencer studied incumbent reelection rates to determine whether
Colorado’s individual contribution limits impacted challengers’ ability to run effective
campaigns. He opined that if there is a concern “that a campaign finance contribution limit is an
incumbency protection scheme, then the way to evaluate that is to see whether incumbents are
winning more often or not.” Id. at 1082:22-25. He defined incumbent as the person who holds
the office being contested. Id. at 1084:10-11. Dr. Spencer agreed with Dr. Cann that incumbent
reelection rates are not the only metric to determine electoral competitiveness. Id. at 1083:13-14.
In arguing that incumbent reelection rates “are the primary metric of whether or not incumbents
are being re-elected,” he disagreed with Dr. Cann. Id. at 1083:16-18.
Using data from the state legislative election return database from 1968 through 2020,
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
54 of 142

55
Dr. Spencer compared incumbency election rates in Colorado state House and state Senate races
before and after Amendment 27.19 Id. at 1083:23-1084:3; see also Trial Exs. 322, 323. He did
not include variables for the number of votes or the closeness of the race. Trial Tr. at 1121:24-
1122:4. The data showed differing rates of incumbency reelection rates; the rates rose year-over-
year in some instances and dropped in others. Trial Exs. 323, 324. Dr. Spencer aggregated these
numbers and found that the incumbent reelection rate for state House elections, in which there
was an incumbent candidate, was 91.8% before Amendment 27 and 93.6% after Amendment 27.
Trial Tr. at 1120:22-24, 1121:18-20; Trial Ex. 322. In state Senate elections, the incumbency
reelection rate was 91.6% before Amendment 27 and 91.1% after Amendment 27. Trial Ex. 322.
Dr. Spencer argued that the differences observed between the pre- and post-Amendment 27 rates
were not statistically significant. Trial Tr. at 1085:17-18. He explained because of the variation
in the data, the confidence interval would be plus or minus 3%. Id. at 1086:8-13. Because the
differences between the incumbent reelection rates fall within that plus or minus 3%, they are not
statistically significant. Id. at 1086:14-16. Thus, he concluded there is “no evidence that there’s
any difference before and after” the enactment of Amendment 27. Id. at 1095:4-7. Based on the
evidence presented at trial, I accept this conclusion as a finding of fact.
Competitiveness of Colorado State Legislative Elections
Next, Dr. Spencer analyzed whether Colorado’s state elections were more or less
competitive compared to other states. Using the same deduplicated dataset used in his first
19 Mr. Engen identified an error in the dataset used by Dr. Spencer, which included duplicate
records from 2014 and 2016 for certain races. Trial Tr. at 1089:25-1090:3. These races were
counted twice in Dr. Spencer’s original calculations. Id. at 1090:2-3. After learning of this error,
Dr. Spencer issued a supplemental report in which he dropped the duplicative data, and his trial
testimony and exhibits reflect his revised calculations. Id. at 1090:6-11.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
55 of 142

56
analysis, see supra note 19, he created two graphs plotting the average of all of Colorado’s
legislative elections’ incumbent reelection rate and the average of all other states’ legislative
incumbent reelection rate. Trial Ex. 325. The data showed that the incumbent reelection rate in
Colorado state House elections was similar to the average of the rest of the United States from
1968 through 2016. Id.; see also Trial Tr. at 1088:12-14. Compared to the rest of the states in
state Senate races, Colorado was more competitive in the 1960s and 1970s, less competitive in
the 1980s, and nearly equal in competitiveness since the 2000s. Trial Ex. 325; see also Trial Tr.
1088:15-19.
Dr. Spencer then analyzed the relationship between state legislative incumbent reelection
rates and contribution limits. Trial Exs. 326, 327. This analysis only included Colorado and other
states with contribution limits. Trial Tr. at 1093:10-12. In state House election cycles from 2010-
2016, “Colorado elections [were] more competitive than the average of all the other states that
have a contribution limit.” Id. at 1092:7-9; Trial Ex. 326. In state Senate election cycles for that
same time period, however, Colorado elections were less competitive in 2010, 2012, and 2016
compared to other states that have contribution limits. Trial Tr. at 1093:3-6; Trial Ex. 327. Next,
Dr. Spencer analyzed whether the incumbent reelection rate in Colorado differs from those states
without contribution limits. Trial Exs. 329, 330. He found that for the state House election cycles
in 2010-2016, Colorado’s incumbent reelection rate was below the average of the states with no
contribution limits. Trial Tr. at 1094:10-14; Trial Ex. 329. In the state Senate election cycles for
that same period, Colorado’s incumbent reelection rate was higher, i.e., less competitive, in
2010, 2012, and 2016 than states with no contribution limits. Trial Tr. at 1094:15-18; Trial Ex.
330.
Dr. Spencer ran a regression analysis to see if contribution limits have measurable effects
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
56 of 142

57
on state legislative incumbent reelection rates. Trial Tr. at 1097:12-16. He combined state House
and state Senate elections from 2010, 2012, 2014, and 2016, id. at 1099:16-22, and studied 176
elections, id. at 1104:11-13. He argued it was proper to pool the state House and state Senate
races because they exist in the same political body, are subject to the same campaign finance
contribution limits, and must submit to the same rules on filing and running for office. Id. at
1100:4-11. He ran several analyses in which he controlled for different variables, including the
state’s use of multi-member districts, term limits, salary, asset disclosure rules, and state
population. Trial Ex. 331. Dr. Spencer, however, was unable to find data to allow him to control
for different characteristics of each challenger, including race, gender, prior work history, or role
in the state party. Trial Tr. at 1102:3-7. He included state fixed effects, or controls in the model,
in his analysis to address the concern that unobserved criteria from different states could impact
the predictor variables. Id. at 1105:6-14.
Dr. Spencer explained that he was not concerned by two of the state fixed effects dropped
out from his characteristics of the legislature regression analysis. Id. at 1105:18-22. He ran the
regression analysis with state fixed effects, without state fixed effects, with random effects, and
with a mixed model effect. Id. at 1106:18-20. In each variation, the data showed that the
individual contribution limit effect on competitiveness was the same regardless of the inclusion
of different specifications. Id. at 1106:24-1107:14. His regression analyses suggested there is no
relationship regardless of the controls used between individual contribution limits and the
incumbent reelection rate. Id. at 1103:2-9. Unlike Dr. Cann, Dr. Spencer did not study a
combination of contribution limits; he only studied the impact of individual contribution limits
on incumbent reelection rates, which is the challenged limit here. Id. at 1107:19-23.
Dr. Spencer reviewed a 2010 study by Dr. Thomas Stratmann, a professor at George
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
57 of 142

58
Mason University, that found races with an individual contribution limit of $500, compared to a
$2,000 limit, saw the likelihood of incumbent defeat increase by 10%. Id. at 1108:23-1109:1.
This was one of two studies he found supporting this finding. Id. at 1109:2-3. Thus, he
concluded, challengers would not win more often if Colorado abolished its contribution limits
because the challenger win rate in Colorado changes regardless of an individual contribution
limit. Id. at 1109:10-15, 1109:24-25.
As the last part of his competitiveness opinion, Dr. Spencer looked at the contestation
rates in Colorado general elections. Trial Exs. 332, 333, 334. He showed that the Colorado state
House and Senate elections followed a similar trajectory: In the 1980s, the proportion of
contested seats was lower compared to the 2000s. Trial Tr. at 1111:20-23; Trial Exs. 332, 333. In
the 2010s, that proportion began to decrease. Trial Ex. 333. He then plotted the proportion of
contested seats with the contribution limits for state legislators in 2014 and 2016. Trial Tr. at
1112:9-13; Trial Ex. 334. In 2014, for example, Colorado saw a higher proportion of contested
legislative elections compared to states with similar contribution limits. Trial Tr. at 1112:14-18;
Trial Ex. 334. While there are myriad ways to analyze election competitiveness, the evidence
presented by Dr. Spencer is directly relevant to the questions presented in this case. Therefore, I
accept his opinions on competitiveness as findings of fact.
Colorado Political Party Contributions
Finally, using data from TRACER, Dr. Spencer analyzed political party contributions in
Colorado. See Trial Exs. 335-339. He first calculated the number of Colorado candidates from
each party who reported any contributions from a political party. Trial Ex. 335. This number
likely does not cover all candidates running during an election cycle. See Trial Tr. at 1128:1-5.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
58 of 142

59
He then calculated the total dollars contributed by a political party to candidates. Trial Ex. 336.
In 2010, the parties contributed approximately $539,000 to candidates, and in 2022, they
contributed approximately $1,538,000 to candidates. Id. He divided the dollars contributed by
the number of candidates who reported political party contributions. Trial Ex. 337; see also Trial
Tr. at 1116:9-11. On average, the Democratic Party contributed $5,582 to candidates in 2010 and
$9,035 in 2022. Trial Ex. 337. The Republican Party, on average, contributed $1,251 to
candidates in 2010 and $4,692 in 2022. Id. These average contributions are all below Colorado’s
political party contribution limits. See Colo. Const. Art. XXVIII § 3(3)(d); 8 Colo. Code Regs.
§ 1505-6, Rule 10.17.1(i); Trial Ex. 351 at 20. Dr. Spencer also presented data that showed
political party contributions in the amount of $5,000 or $10,000 or more to candidates. See Trial
Exs. 338, 339. All of the political party contribution data led Dr. Spencer to conclude that
“candidates have access to fundraising that goes beyond individuals, and that fundraising can be
quite substantial.” Trial Tr. at 1118:9-11.
Rebuttal to Dr. Cann
Dr. Spencer rebutted Dr. Cann’s finding that incumbents were unlikely to face a primary
challenger. See, e.g., Trial Ex. 340. He testified that Dr. Cann’s numbers related to primary
election competitiveness were “misleading” because he only analyzed state House primaries and
did not include state Senate primaries or primaries for open seats. Trial Tr. at 1113:3-8. He
defined open seat races to typically include at least two challengers. Id. at 1119:8-14. By
including those two factors in his analysis. Dr. Spencer stated he was able to better understand
the rate at which incumbents run for legislative office and the rate at which challengers can
mount an effective campaign. Id. at 1113:9-16. Dr. Spencer’s data analysis, which included state
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
59 of 142

60
House, state Senate, and open seat primaries, showed that, for example, in 2022, 39% of the
primary races were for open seats and 14% of the incumbent primaries included a challenger.
Trial Ex. 340. Thus, 53% of Colorado’s 2022 state legislature primary races included at least one
challenger. Id. I agree that this analysis provides a better understanding of the landscape than that
provided by Dr. Cann’s analysis.
Findings on Dr. Spencer’s Testimony
Dr. Spencer’s opinions regarding incumbent reelection rates are relevant to a critical
inquiry in this case and his methodology is meticulous and reliable. Therefore, I accept Dr.
Spencer’s opinions and consider them in my Conclusions of Law.
II. CONCLUSIONS OF LAW
Plaintiffs presented two issues for me to decide at trial:
A. Whether Colorado’s individual campaign contribution limits, enacted by Colo.
Const. Art. XXVIII § 3 . . . , violate the freedom of speech and assembly clauses
of U.S. Const. amend. I. See Am. Compl., Count 1, ¶¶ 38-55 (ECF [No.] 46).
B. Whether Colorado’s asymmetric individual campaign contribution scheme,
enacted by Colo. Const. Art. XXVIII § 4 . . . , violates the freedom of speech and
assembly clauses of U.S. Const. amend. I. See Am. Compl., Count 2, ¶¶ 56-62
(ECF [No.] 46).
Pls.’ Trial Br. at 1, ECF No. 99.
In plain English, the first issue is whether the limits on individuals’ contributions to
candidates for Colorado state office are too low to be valid under the First Amendment to the
U.S. Constitution. The test for deciding this considers the governmental interest underlying the
individual contribution limits and the tailoring of the State’s campaign finance regulations. The
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
60 of 142

61
second issue is whether the asymmetric, individual contribution limits triggered by Colorado’s
voluntary spending limits scheme are impermissible under Supreme Court precedent.
Defendants contend that Plaintiffs’ precedents are inapposite and that § 4’s scheme
should instead be analyzed under the framework for states offering public financing options to
candidates. Defendants also challenge my jurisdiction to decide Plaintiffs’ claims. They argue
that Mr. House lacks standing to bring both Claim One and Claim Two, both claims are moot as
to Mr. Lopez, Senator Pelton lacks standing to bring Claim Two, and Claim Two is moot as to
Senator Pelton.
A. Campaign Finance Jurisprudence
The First Amendment to the U.S. Constitution safeguards “an individual’s right to
participate in the public debate through political expression and political association.”
McCutcheon v. Fed. Election Comm’n, 572 U.S. 185, 203 (2014) (plurality opinion); see also
Buckley, 424 U.S. at 15. Campaign finance regulations, like those challenged in this case, must
avoid impermissibly burdening these rights. What counts as a burden and the degree to which
these rights may be burdened has been extensively litigated and has resulted in a morass of
judicially created principles, frameworks, and standards plagued by ambiguity.20
1. Buckley and its Progeny
The “seminal”21 campaign finance case of the modern era, Buckley v. Valeo, articulated
20 As Justice Kennedy wrote, “[t]he universe of campaign finance regulation is one [the Supreme
Court] has in part created and in part permitted by its course of decisions.” Randall v. Sorrell,
548 U.S. 230, 264-65 (2006) (Kennedy, J., concurring in the judgment). He also posited that the
“new order” the Court has created and permitted “may cause more problems than it solves.” Id.
21 While I follow the custom of other jurists and use the term “seminal,” see, e.g., Ariz. Free
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
61 of 142

62
the foundational concepts that have informed the last fifty years of campaign finance
jurisprudence. Buckley involved constitutional challenges to the Federal Election Campaign Act
of 1971 and related sections of the Internal Revenue Code. The challenged provisions in those
statutes limited monetary contributions to political campaigns (i.e., campaign contributions),
restricted spending in support of or against campaigns (i.e., political expenditures), and
established a system for public financing of Presidential campaigns. 424 U.S. at 6-7. The Court’s
holdings rested on the principle that giving and spending money in connection with campaigns is
an essential component of political speech and, therefore, is entitled to constitutional protection.
Id. at 16, 19-23.22 Employing that principle, the Court drew a distinction between campaign
contributions and political expenditures. Id. at 19-23. The Court explained that “[a] restriction on
Enter. Club’s Freedom Club PAC v. Bennett, 564 U.S. 721, 757 (2011) (Kagan, J., dissenting)
(calling Buckley the Supreme Court’s “seminal campaign finance case”); Fed. Election Comm’n
v. Wis. Right to Life, Inc., 551 U.S. 449, 485 (2007) (Scalia, J., concurring in part and concurring
in the judgment) (describing Buckley as “the seminal case”), I think the term is ill-fitting.
Merriam-Webster defines the word “seminal” as “containing or contributing the seeds of later
development . . .” Seminal, Merriam-Webster, https://www.merriam-webster.com/dictionary/
seminal (last accessed Mar. 23, 2026). While later cases have certainly built onto the principles
announced in Buckley, whether the vines that now exist naturally grew from the seeds sown by
Buckley is not obvious. Judges and scholars have criticized later campaign finance precedent that
ostensibly has relied on Buckley as departing from Buckley’s stated principles. See, e.g., Citizens
United, 558 U.S. at 453 (Stevens, J., concurring in part and dissenting in part) (asserting that
Buckley could not “sustain [the majority’s] reading” of it); Randall, 548 U.S. at 266 (Thomas, J.,
concurring in the judgment) (“The illegitimacy of Buckley is further underscored by the
continuing inability of the Court (and the plurality here) to apply Buckley in a coherent and
principled fashion.”).
22 The Court in Buckley stressed: “In a republic where the people are sovereign, the ability of the
citizenry to make informed choices among candidates for office is essential, for the identities of
those who are elected will inevitably shape the course that we follow as a nation.” 424 U.S. at
14-15. The Court was focused on protecting the “unfettered interchange of ideas” and ensuring
that “debate on public issues [is] uninhibited, robust, and wide-open.” Id. at 14, 49 (citations
omitted). These are aims of the utmost significance, but in the digital age, their means of
protection may need to be reconsidered. See, e.g., Drowning Out Democracy, 137 Harv. L. Rev.
2386, 2390 (2024) (explaining that these foundations of democracy are threatened by the
potential for money to exploit contemporary media’s multifaceted deficiencies by “drowning
out” voices and ideas and through the spread of misinformation).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
62 of 142

63
the amount of money a person or group can spend on political communication during a campaign
necessarily reduces the quantity of expression by restricting the number of issues discussed, the
depth of their exploration, and the size of the audience reached.” Id. at 19. Accordingly,
restrictions on political expenditures can “limit political expression at the core of our electoral
process and of the First Amendment freedoms” and thus must overcome “exacting scrutiny.” Id.
at 39, 44-45 (citation modified).23
In comparison, the Court observed that a contribution limitation “involves little direct
restraint on [a contributor’s] political communication, for it permits the symbolic expression of
support evidenced by a contribution, but does not in any way infringe the contributor’s freedom
to discuss candidates and issues.” Id. at 21. Importantly, the Court considered the potential
impact of contribution limits on association rights as well, viewing the act of making a
contribution as “serv[ing] to affiliate a person with a candidate” and “enabl[ing] like-minded
persons to pool their resources in furtherance of common political goals.” Id. at 23. Contribution
limits may restrict “one important means of associating with a candidate or committee, but leave
the contributor free to assist personally in the association’s efforts on behalf of candidates.” Id. at
22. Based on that reasoning, the Court instructed that contribution limitations are permissible if
the state can “demonstrate[] a sufficiently important interest and employs means closely drawn
to avoid unnecessary abridgment of associational freedoms.” Id. at 25 (citations omitted).24
23 In addition to political expression, the Court discerned that spending limits can “impinge on
protected associational freedoms” by precluding associations from effectively amplifying the
voice of their adherents.” Buckley, 424 U.S. at 22.
24 Much of the controversy surrounding Buckley involves the Court’s differential treatment of
spending and contribution limits. See, e.g., Pamela Karlan, New Beginnings and Dead Ends in
the Law of Democracy, 68 Ohio St. L.J. 743, 747 (2007) (“Save for the notorious crack/powder
distinction, it’s hard to think of a line that has been subjected to more withering criticism over
the years than Buckley’s expenditure/contribution distinction.”); David Schultz, Revisiting
Buckley v. Valeo: Eviscerating the Line Between Candidate Contributions and Independent
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
63 of 142

64
Applying that standard, the Court held that the prevention of corruption or its appearance was a
sufficiently important interest to justify the Act’s restriction on the size of political contributions.
Id. at 25-29. It reasoned that, “[t]o the extent that large contributions are given to secure a
political quid pro quo from current and potential office holders, the integrity of our system of
representative democracy is undermined.” Id. at 26-27. It pointed to the “deeply disturbing
examples surfacing after the 1972 election” that “demonstrate[d] that the problem is not an
illusory one.” Id. at 27. “Of almost equal concern as the danger of actual quid pro quo
arrangements,” the Court observed, “is the impact of the appearance of corruption stemming
from public awareness of the opportunities for abuse inherent in a regime of large individual
financial contributions.” Id.
Buckley also involved a challenge to the Federal Election Campaign Act’s “limits on
expenditures by a candidate from his personal funds or the personal funds of his immediate
family, in connection with his campaigns.” Id. at 51 (citation modified). The Court held that the
Act’s “ceiling on personal expenditures by candidates on their own behalf . . . impose[d] a
substantial restraint on the ability of persons to engage in protected First Amendment
expression.” Id. at 52. It reasoned that “[t]he candidate, no less than any other person, has a First
Expenditures, 14 J.L. & Pol. 33, 70 (1998) (explaining that, by 1998, “half of the Justices who
ha[d] served on the Court since Buckley ha[d] rejected the contribution/expenditure distinction,
suggesting that significant debate surrounds th[at] distinction, and that Buckley’s continued
viability requires further inquiry and justification”); John C. Bonifaz, Gregory G. Luke, &
Brenda Wright, Challenging Buckley v. Valeo: A Legal Strategy, 33 Akron L. Rev. 39, 41 (1999)
(“More than 200 constitutional scholars from across the nation have signed a statement calling
for the reversal of Buckley’s prohibition on spending limits. The attorneys general for 26 states
and the secretaries of state or chief election officers for 21 states have gone on record seeking to
overturn the ruling. Members of Congress have introduced eleven bills since 1976 which would
establish campaign spending limits for federal elections and set the stage for revisiting Buckley.
Thirty-eight U.S. Senators have supported the call for the reversal of the ruling.” (citations
omitted)).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
64 of 142

65
Amendment right to engage in the discussion of public issues and vigorously and tirelessly to
advocate his own election and the election of other candidates.” Id. The Court stressed that “it is
of particular importance that candidates have the unfettered opportunity to make their views
known so that the electorate may intelligently evaluate the candidates’ personal qualities and
their positions on vital public issues before choosing among them on election day.” Id. at 52-53.
In considering whether a governmental interest might nevertheless justify the limit on personal
expenditures, the Court determined that the restrictions could not serve to prevent actual and
apparent corruption of the political process, as “the use of personal funds reduces the candidate’s
dependence on outside contributions and thereby counteracts the coercive pressures and
attendant risks of abuse to which the Act’s contribution limitations are directed.” Id. at 53. As a
result, the Court concluded that “[t]he ancillary interest in equalizing the relative financial
resources of candidates competing for elective office” was “clearly not sufficient to justify the
provision’s infringement on fundamental First Amendment rights.” Id. at 54.
The final relevant holding in Buckley relates to provisions of the Internal Revenue Code
that provide for the public financing of presidential election campaigns. Id. at 85. Pursuant to
these provisions, public funding was available for primary election campaigns, nominating
conventions, and general election campaigns, with levels of available funding varying based on
party status and affiliation. Id. at 86-90. The Court rejected claims that the public financing
provisions were facially unconstitutional, including arguments that the varying levels of funding
available for major and nonmajor parties and their candidates violated the Fifth Amendment. Id.
at 107-08; see also id. at 95-96 (“Congress enacted [the public funding provisions] in furtherance
of sufficiently important governmental interests and has not unfairly or unnecessarily burdened
the political opportunity of any party or candidate.”). In so doing, it endorsed four governmental
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
65 of 142

66
interests as justifications for the structure of the public funding program: “eliminating improper
influence of large private contributions,” “relieving major-party Presidential candidates from the
rigors of soliciting private contributions,” “not funding hopeless candidacies with large sums of
public money,” and avoiding the provision of “artificial incentives to splintered parties and
unrestrained factionalism.” Id. at 96 (citation modified). The Court regarded the public funding
program as “a congressional effort, not to abridge, restrict, or censor speech, but rather to use
public money to facilitate and enlarge public discussion and participation in the electoral process,
goals vital to a self-governing people.” Id. at 92-93. More generally, the Court described the
validity of public funding programs, stating:
Congress may engage in public financing of election campaigns and may condition
acceptance of public funds on an agreement by the candidate to abide by specified
expenditure limitations. Just as a candidate may voluntarily limit the size of the
contributions he chooses to accept, he may decide to forgo private fundraising and
accept public funding.
Id. at 57 n.65.
Ultimately, the Court in Buckley held the public financing program in the Internal
Revenue Code and the Federal Election Campaign Act’s contribution limits were constitutional,
but the Act’s spending limits, including those by a candidate from his personal funds, violated
the First Amendment. Id. at 143.
Post-Buckley, the Supreme Court “consistently upheld contribution limits in other
statutes.” Randall v. Sorrell, 548 U.S. 230, 247 (2006) (plurality opinion). Thirty years after
Buckley came Randall v. Sorrell. In Randall, the Court revisited when contribution limits might
become “too low and too strict to survive First Amendment scrutiny.” Id. at 243, 248. The case
involved Vermont’s 1997 campaign finance law which imposed stringent expenditure limits on
candidates as well as contribution limits on individuals, political committees, and political
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
66 of 142

67
parties. Id. at 237-39. The plaintiffs claimed that the cap on political expenditures and the
contribution ceilings violated the First Amendment. Id. at 240, 246-48. Both arguments proved
successful. Id. at 263. The plurality opinion concluded there was no justification for the Court to
reconsider or limit Buckley’s holding on spending limits. Id. at 243-46. It also held that the law’s
contribution limits were “unconstitutional because in their specific details (involving low
maximum levels and other restrictions) they fail[ed] to satisfy the First Amendment’s
requirement of careful tailoring.” In other words, they “imposed burdens upon First Amendment
interests that (when viewed in light of the statute’s legitimate objectives) are disproportionately
severe.” Id. at 236-37.
The analysis of the contribution limits in the plurality opinion is particularly relevant
here. See id. at 246-62. The opinion outlines a process for evaluating whether challenged
campaign contribution limits are viable under the First Amendment. That process starts, just as
Buckley did, with consideration of the government’s interest in imposing the contribution limit.
See id. at 247-48 (“[C]ontribution limitations are permissible as long as the Government
demonstrates that the limits are ‘closely drawn’ to match a ‘sufficiently important interest.’”
(quoting Buckley, 424 U.S. at 25)). When the government asserts a legitimate interest, such as
“preventing corruption and the appearance of corruption,” the review of the contribution limits
turns to whether there are “danger signs” that indicate the limits may be disproportionate. If such
signs exist, the next step is to determine whether the limits are “closely drawn” in light of the
five factors the plurality articulated. Id. at 248-62.
In 2019, in Thompson v. Hebden, the Supreme Court considered a challenge to Alaska’s
limits for individual contributions to candidates for political offices and to election-oriented
groups other than political parties. 589 U.S. 1, 2 (2019) (per curiam). The court of appeals had
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
67 of 142

68
declined to apply the framework set out in the plurality opinion in Randall. Id. at 3 (citing
Thompson v. Hebdon, 909 F.3d 1027 (9th Cir. 2018)). After briefly reviewing the concerns and
danger signs expressed by the Randall plurality, the Court remanded the case to the court of
appeals to “revisit whether Alaska’s contribution limits [we]re consistent with [the Court’s] First
Amendment precedents.” Id. at 6. This direction from the Court has been taken as an
endorsement of the Randall plurality’s test for the constitutionality of contribution limits.
2. Penalizing Candidates for Exercising their First Amendment Rights
Plaintiffs’ arguments regarding their second claim and the doubled contribution limits
available under Colorado’s voluntary spending limits scheme center on two additional Supreme
Court cases that were decided after Buckley and Randall: Davis v. Federal Election Commission,
554 U.S. 724 (2008), and Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett, 564
U.S. 721 (2011). In Davis, the Court scrutinized the so-called Millionaire’s Amendment to the
Bipartisan Campaign Reform Act. 554 U.S. at 729. Pursuant to the Amendment, if a candidate’s
expenditure of personal funds cause[d] the “opposition personal funds amount” (“OPFA”) to
exceed $350,000, the candidate’s opponent could “receive individual contributions at treble the
normal limit (e.g., $6,900 rather than the current $2,300), even from individuals who ha[d]
reached the normal aggregate contributions cap, and [could] accept coordinated party
expenditures without limit,” while the self-financing25 candidate remained subject to the standard
limitations. Id. at 728-29. “The OPFA, in simple terms, [wa]s a statistic that compare[d] the
expenditure of personal funds by competing candidates and also t[ook] into account to some
25 The Court referred to candidates whose expenditure of personal funds caused the OPFA to
exceed $350,000 as “self-financing candidates,” Davis, 554 U.S. at 730, so I do the same.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
68 of 142

69
degree certain other fundraising.” Id. at 729.
A former U.S. House of Representatives candidate from New York challenged the
Millionaire’s Amendment as an unconstitutional burden on “his First Amendment right to make
unlimited expenditures of his personal funds because making expenditures that create the
imbalance has the effect of enabling his opponent to raise more money and to use that money to
finance speech that counteracts and thus diminishes the effectiveness of [his] own speech.” Id. at
731, 736. The Court found that the Millionaire’s Amendment “requires a candidate to choose
between the First Amendment right to engage in unfettered political speech and subjection to
discriminatory fundraising limitations.” Id. at 739. According to the Court, the Amendment
imposed “an unprecedented penalty on any candidate who robustly exercise[d]” the First
Amendment right “to use personal funds to finance campaign speech” by “produc[ing]
fundraising advantages for opponents in the competitive context of electoral politics.” Id. Given
the Court’s conclusion that the Millionaire’s Amendment “impose[d] a substantial burden on the
exercise of the First Amendment right to use personal funds for campaign speech,” the
government was obligated to show the Amendment was “justified by a compelling state
interest.” Id. at 740 (citation omitted). The government’s proffered interest—“reduc[ing] the
natural advantage that wealthy individuals possess in campaigns for federal office”—was not “a
legitimate government objective.” Id. at 741. Consequently, the Millionaire’s Amendment
violated the First Amendment. Id. at 744.
In Bennett, the Court considered the constitutionality of Arizona’s campaign finance
program which provided matching funds to publicly financed candidates for qualifying money
spent by privately financed opponents and independent expenditure groups. 564 U.S. at 727-28.
To receive public funding, a candidate had to collect a set number of contributions from Arizona
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
69 of 142

70
voters and to agree to certain restrictions and obligations, including limiting his or her
expenditure of personal funds to $500 and adhering to an overall spending cap. Id. at 728-29.
Publicly funded candidates could receive additional matching funds for both primary and general
elections “when a privately financed candidate’s expenditures, combined with the expenditures
of independent groups made in support of the privately financed candidate or in opposition to a
publicly financed candidate,” exceed the respective election allotment of state funds to the
publicly financed candidate. Id. at 729.26 The plaintiffs, candidates and independent expenditure
groups, “argued that the matching funds provision unconstitutionally penalized their speech and
burdened their ability to fully exercise their First Amendment rights.” Id. at 732-33. The Court
determined that Arizona’s matching funds provision, like the Millionaire’s Amendment in Davis,
“impose[d] an unprecedented penalty on any candidate who robustly exercise[d] [his] First
Amendment right[s].” Id. at 736 (quoting Davis, 554 U.S. at 739). In fact, the Court found that
the benefit to the publicly financed candidate in Bennett—“the direct and automatic release of
public money”—was “a far heavier burden” on privately financed candidates than the burden
imposed on self-financing candidates in Davis. Id. at 737. Similar to Davis, the Court concluded
that “Arizona’s matching-funds provision substantially burden[ed] the speech of privately
financed candidates and independent expenditure groups without serving a compelling interest,”
and, therefore, violated the First Amendment. Id. at 728, 754-55.
These cases set the stage for the analysis below.
26 Matching state funds would be provided up to two times the initial authorized grant of public
funding, which would be when the publicly financed candidate had received three times the
initial allotment for the race. Bennett, 564 U.S. at 730, 732.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
70 of 142

71
B. Standing & Mootness
Defendants’ jurisdictional arguments must be addressed first. Article III of the United
States Constitution limits federal courts to adjudicating live cases and controversies. U.S. Const.
art. III, § 2; Citizens for Responsible Gov’t State Pol. Action Comm. v. Davidson, 236 F.3d 1174,
1181-82 (10th Cir. 2000). “For there to be a case or controversy under Article III, the plaintiff
must have a ‘personal stake’ in the case—in other words, standing.” TransUnion LLC v.
Ramirez, 594 U.S. 413, 423 (2021) (quoting Rains v. Byrd, 521 U.S. 811, 819 (1997)). Plaintiffs
bear the burden of establishing their standing to sue. Lujan v. Defs. of Wildlife, 504 U.S. 555, 561
(1992). A plaintiff’s standing to challenge one government action does not permit him to
“challenge other governmental actions that did not injure him.” DaimlerChrysler Corp. v. Cuno,
547 U.S. 332, 353 n.5 (2006). Thus, each plaintiff must show “standing to seek each form of
relief in each claim.” Bronson v. Swensen, 500 F.3d 1099, 1106 (10th Cir. 2007). As standing is
a jurisdictional requirement, it “remains open to review at all stages of the litigation.” Nat’l Org.
for Women, Inc. v. Scheidler, 510 U.S. 249, 255 (1994).
Constitutional standing has three elements. First, the plaintiff must have suffered an
“injury in fact,” defined as the invasion of a legally protected interest that is concrete and
particularized. Lujan, 504 U.S. at 560. Such an injury must be “actual or imminent, not
conjectural or hypothetical.” Id. (citation modified). To be imminent, the “threatened injury must
be certainly impending”; “[a]llegations of possible future injury are not sufficient.” Clapper v.
Amnesty Int’l USA, 568 U.S. 398, 409 (2013) (citation modified). Second, there must be a causal
connection between the injury and the challenged conduct—the injury cannot be “the result of
the independent action of some third party not before the court.” Lujan, 504 U.S. at 560 (citation
modified). Third, there must be a likelihood that the injury will be redressed by a favorable
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
71 of 142

72
decision on the merits. Id. at 561. Plaintiffs must establish each of these elements by pointing to
facts that are “supported adequately by the evidence adduced at trial.” Id. (quoting Gladstone,
Realtors v. Village of Bellwood, 441 U.S. 91, 115 n.31 (1979)). “For purposes of the standing
inquiry, the question is not whether the alleged injury rises to the level of a constitutional
violation.” Initiative & Referendum Inst. v. Walker, 450 F.3d 1082, 1088 (10th Cir. 2006) (en
banc); see also Warth v. Seldin, 422 U.S. 490, 500 (1975) (“[S]tanding in no way depends on the
merits of the plaintiff’s contention that particular conduct is illegal.”).
“[A]n actual controversy must be extant at all stages of review, not merely at the time the
complaint is filed.” Davidson, 236 F.3d at 1182 (quoting Arizonans for Official English v.
Arizona, 520 U.S. 43, 67 (1997)). This requirement—that there be a live case or controversy
throughout the case—is an additional constitutional prerequisite to federal court jurisdiction. See
Davidson, 236 F.3d at 1181-82. A claim is moot “when the issues presented are no longer live or
the parties lack a legally cognizable interest in the outcome.” Id. at 1182 (citation modified). In
determining whether a claim is moot, “[t]he crucial question is whether ‘granting a present
determination of the issues offered . . . will have some effect in the real world.’” Id. (quoting
Kennecott Utah Copper Corp. v. Becker, 186 F.3d 1261, 1266 (10th Cir. 1999)). As with
standing, mootness must be evaluated on a claim-by-claim basis. Smith v. Becerra, 44 F.4th
1238, 1247 (10th Cir. 2022).
Plaintiffs argue that their circumstances fall within an established exception to the
mootness doctrine, that “for disputes capable of repetition, yet evading review.” Fed. Election
Comm’n v. Wisc. Right to Life, Inc., 551 U.S. 449, 462 (2007). “The exception applies where (1)
the challenged action is in its duration too short to be fully litigated prior to cessation or
expiration, and (2) there is a reasonable expectation that the same complaining party will be
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
72 of 142

73
subject to the same action again.” Id. (citation modified). The second prong of this test “requires
a ‘reasonable expectation’ or a ‘demonstrated probability’ that ‘the same controversy will recur
involving the same complaining party.’” Id. at 463 (quoting Murphy v. Hunt, 455 U.S. 478, 482
(1982) (per curiam)). “[A] mere physical or theoretical possibility” does not satisfy the test.
Murphy, 455 U.S. at 482. While the defendant generally bears the burden of establishing that a
claim has become moot, see Smith, 44 F.4th at 1247, when a plaintiff asserts that this exception
applies, the plaintiff must show both prongs of the test are fulfilled, see Jordan v. Sosa, 654 F.3d
1012, 1235 (10th Cir. 2011).
Defendants successfully argued that Plaintiffs’ appeal of my Order Denying the
Preliminary Injunction was moot once the 2022 election had passed. See Lopez, 2023 WL
1960802, at *3. After the parties filed cross-motions for summary judgment in 2023, I prompted
them to address standing and the potential mootness of Plaintiffs’ second claim for relief.
08/02/2023 Minute Order, ECF No. 81. In their Supplemental Brief Regarding Mootness and
Standing, Defendants agreed that Claim One brought by Senator Pelton and Mr. Lopez was not
moot. Defs.’ Suppl. Br. Regarding Mootness & Standing at 1-2, ECF No. 83. However, they
argued that Claim Two was moot because Senator Pelton and Mr. Lopez had “not established a
reasonable probability that they will be injured by [§ 4(5)] again in the future.” Id. at 1.
Defendants also asserted that Mr. House did not have standing as to Claim Two “because it
causes him no injury as a political contributor.” Id. at 1.
Upon review of the parties’ supplemental briefing, I found that at least one Plaintiff—Mr.
House—“likely ha[d] standing vis-à-vis [Plaintiff’s] first claim, regarding § 3’s contribution
limits.” Order Denying Mots. for Partial Summary J. at 2 n.1, ECF No. 88. But I resolved that
the question of whether any Plaintiff had standing to bring Claim Two was too bound up with the
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
73 of 142

74
disputed facts and thus withheld my determination on that issue until the facts were fully
presented. Id. In their Trial Brief, Defendants then argued, in passing, that “Plaintiffs have a
threshold justiciability problem with [their second claim].” Defs.’ Trial Br. at 9, ECF No. 98.
Defendants’ post-trial briefing asserts that “[t]he only claim over which this Court now has
jurisdiction is Senator Pelton’s challenge to the contribution limits for state legislative office”
because Mr. House lacks standing to bring both Claim One and Claim Two, both claims are
moot as to Mr. Lopez, Senator Pelton lacks standing to bring Claim Two, and Claim Two is
moot as to Senator Pelton. Defs.’ Written Summation at 3-7, ECF No. 122.
1. Claim One
Defendants contend that neither Mr. House nor Mr. Lopez’s circumstances permit them
to challenge the Colorado contribution limit for statewide office. Defs.’ Written Summation at 3-
4. As noted, I have already found that Mr. House has standing to bring Claim One. See Order
Denying Mots. for Partial Summary J. at 2 n.1. I now address Defendants’ expanded arguments
and again find that Mr. House may pursue Claim One. Even if I concluded otherwise, however,
Mr. Lopez would still be able to maintain his challenge to Colorado’s contribution limit for
statewide office.
Mr. House’s Standing
Article XXVIII § 3 prohibits any person from contributing to a candidate committee in
excess of the contribution limit for the candidate’s sought office. As found above, Mr. House has
contributed to candidates running for Colorado statewide, legislative, and other offices since
2010. Trial Tr. at 138:24-139:10; see also Trial Exs. 108-110. Mr. House intends to continue his
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
74 of 142

75
financial support of various Colorado candidates for as long as he can, and, if permitted, he
would contribute more than the maximum amounts allowed by law. Trial Tr. at 152:24-153:1,
158:19-23; 160:1-14.
Nevertheless, Defendants argue Mr. House lacks standing to bring Claim One because he
“suffers no injury from Colorado’s limits.” Defs.’ Written Summation at 3. Defendants
emphasize that Colorado’s contribution limits still permit Mr. House to “associate with and
symbolically support his candidates of choice through contributions” and to “us[e] his own voice
and funds to express his support through avenues other than candidate contributions.” See Defs.’
Written Summation at 4. In support of their contention that contributors are not injured by
contribution limits, Defendants quote Justice Thomas’s concurrence in Randall, which accuses
the plurality of being “concerned not with the impact on the speech of contributors, but solely
with the speech of the candidates.” Defs.’ Written Summation at 3 n.1 (quoting Randall, 548
U.S. at 270 n.3 (Thomas, J., concurring)).27
The Supreme Court has been unambiguous since Buckley: Contribution limits restrict
political speech as well as political expression. See, e.g., Fed. Election Comm’n v. Colo.
Republican Fed. Campaign Comm., 533 U.S. 431, 440 (2001) (restating that contributing to
political candidates “fall[s] within the First Amendment’s protection of speech and political
association” (citing Buckley, 424 U.S. at 14-23)). Defendants’ attempt to decode an alternative
message in the case law is in vain.
27 Justice Thomas’s criticism relates to the plurality’s consideration of political party contribution
limits as a relevant factor for determining whether individual contribution limits are
constitutional. See Randall, 548 U.S. at 270. He cites only some of the plurality’s discussion of
political parties. See id. at 270 n.3 (not citing pages 256-59). Below, as part of my analysis of the
Randall factors, I review the full discussion, including the plurality’s concerns regarding the
potential impacts of contribution limits on the rights of individuals to engage in collective
political activity by donating to a political party. See infra Section II.C.3.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
75 of 142

76
Buckley explained that contribution limitations involve “little direct restraint on [a
contributor’s] political communication” for they still “permit[] the symbolic expression of
support evidenced by a contribution but d[o] not in any way infringe the contributor’s freedom to
discuss candidates and issues.” 424 U.S. at 21.28 Both Justice Thomas’s dissent and the plurality
opinion in Randall confirm that, pursuant to Buckley, contribution limits restrict a contributor’s
ability to engage in free communication, even if only marginally. Randall, 548 U.S. at 241
(“Contribution limitations, though a ‘marginal restriction upon the contributor’s ability to engage
in free communication,’ nevertheless leave the contributor ‘fre[e] to discuss candidates and
issues.’” (quoting Buckley, 424 U.S. at 20-21)); id. at 270 n.3 (emphasizing that “[e]ven Buckley
. . . recognizes that contribution limits restrict the free speech of contributors, even if it
understates the significance of this restriction,” and quoting the language from Buckley that
contribution limits “entail[] only a marginal restriction on the contributor’s ability to engage in
free communication” (quoting Buckley, 424 U.S. at 20)). A restriction on a constitutional right,
no matter how marginal, is still a restriction.
Moreover, Buckley clarified that “the primary First Amendment problem raised by the
[Federal Election Campaign Act’s] contribution limitations [wa]s their restriction of one aspect
of the contributor’s freedom of political association.” Buckley, 424 U.S. at 24. “Making a
contribution, like joining a political party, serves to affiliate a person with a candidate.” Id. at 22.
And contributions “enable[] like-minded persons to pool their resources in furtherance of
28 The Court ascertained that “[t]he quantity of communication by a contributor does not increase
perceptibly with the size of his contribution . . . .” Buckley, 424 U.S. at 21. “At most, the size of
the contribution provides a very rough index of the intensity of the contributor’s support for the
candidate.” Id. This is because other factors, such as the contributor’s financial ability and past
contribution history, must also be considered in interpreting the contributor’s financial
expression. Id. at 21 n.22.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
76 of 142

77
common political goals.” Id. Questions to be considered in reviewing the constitutionality of a
contribution limit include whether it “render[s] political association ineffective” and whether it
“render[s] contributions pointless.” Nixon v. Shrink Mo. Gov’t PAC, 528 U.S. 377, 397 (2000).
These questions contemplate the perspective of contributors, in effect asking whether a limit
discourages contributors from associating or contributing because doing so is futile. Supreme
Court precedent has recognized, then, that contribution limits restrict the contributors’ First
Amendment right to associate, as well as their ability to engage in free communication.
The purpose of the standing doctrine is to dissuade the querulous, not the resourceful. Mr.
House has shown that, in Claim One, he seeks to invalidate contribution limits that restrict,
whether minimally or otherwise, his First Amendment speech and association rights.29
Consequently, he has satisfied the first requirement for Article III standing: an injury in fact.
Accord Frank v. City of Akron, 290 F.3d 813, 816 (6th Cir. 2022) (holding that contributors who
“clearly intend to . . . contribute money in excess of the [at-issue] limitations” had standing).
Defendants do not argue that Mr. House has failed to show the other two requirements—a causal
connection and redressability. I conclude that the facts I have found establish those requirements.
Accordingly, Mr. House has standing to bring Claim One.
Mootness as to Mr. Lopez
Defendants insist Mr. Lopez’s challenge to the contribution limits for statewide office in
§ 3 of Article XVIII is moot. Mr. Lopez is a former Colorado gubernatorial candidate, who ran in
29 Although not mentioned by the parties in this case, the Supreme Court held in Buckley that
“each of the plaintiffs,” including “a potential contributor,” “alleged sufficient injury to his
constitutional rights enumerated in the [certified] questions to create a constitutional ‘case or
controversy’ within the judicial power under Article III.” 424 U.S. at 7, 12 n.11 (emphasis
added).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
77 of 142

78
2018 and 2022. Trial Tr. at 43:24-44:3. After the 2022 election, Mr. Lopez’s claim based on his
campaign for that election was moot. However, Mr. Lopez contends that the exception for
disputes capable of repetition, yet evading review, applies. As such, he must show: “‘(1) the
challenged action is in its duration too short to be fully litigated prior to cessation or expiration,
and (2) there is a reasonable expectation that the same complaining party will be subject to the
same action again.’” Wisc. Right to Life, 551 U.S. at 462 (quoting Spencer, 523 U.S. at 17).
Defendants argue that Mr. Lopez has not sufficiently demonstrated the latter requirement. I
disagree.
Mr. Lopez testified that he intends to run for statewide office again in 2026, Trial Tr. at
50:17-18, 102:2-4, and plans to fund his campaign through individual contributions, id. at 53:15-
20. He maintains an active Colorado candidate political committee, which has up-to-date
contribution and expenditure reports. Id. at 50:19-51:8, 102:13-14; see also Trial Ex. 19. His
testimony, while not detailed, is credible.
Defendants complain that Mr. Lopez has stated “only a vague intent to run for some
unspecified office in 2026,” that he has not filed a candidate affidavit, and is not actively raising
money for any race. Defs.’ Written Summation at 4-5. The fact that, at the time he testified in
2024, his candidate political committee had not accepted any contributions since 2022, see Trial
Tr. at 103:2-7, is not influential, as the 2026 election was still two years away. Considering the
nature of this challenge, the fact that his committee had not accepted any contributions could be
seen as prudent. Mr. Lopez intends, when this litigation is resolved, to receive larger
contributions, so accepting smaller contributions two years before the election could have
resulted in ineffective campaigning. Similarly, he should not be required to make a greater
showing that he is trudging farther up a mountain that he claims he has been prevented and will
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
78 of 142

79
be prevented from summitting because of the individual contribution limits. Article III does not
demand such an unproductive commitment from a plaintiff years in advance of an election.
Under the facts of this case, where Mr. Lopez has previously run for statewide office on
multiple occasions and has stated an unwavering intent to do so again, I conclude he has
established “a ‘reasonable expectation’ or a ‘demonstrated probability’” that his First
Amendment rights will be restricted by Colorado’s individual contribution limit for statewide
office. Wisc. Right to Life, 551 U.S. at 463 (quoting Murphy, 455 U.S. at 482). Therefore, he has
shown that his first claim falls within the exception to mootness for claims that are capable of
repetition, yet evading review, and thus that this Court continues to have jurisdiction over his
challenge to the individual contribution limit for statewide office.
2. Claim Two
Defendants contend that Mr. House lacks standing to assert Claim Two for the same
reasons that he cannot bring Claim One, that Senator Pelton lacks standing because he “was
never injured by the voluntary spending limits,” and that Senator Pelton and Mr. Lopez’s
“challenges to the voluntary spending limits regime were mooted with the end of the 2022
election.” Defs.’ Written Summation at 5, 7 n.3. I conclude that Senator Pelton has standing to
bring Claim Two and that the claim is not moot as to either Senator Pelton or Mr. Lopez.
Even though Plaintiffs contend that Mr. House has standing to bring Claim Two, their
merits arguments for their challenge to § 4’s voluntary spending limits scheme focus primarily
on the First Amendment rights of candidates, not contributors. See, e.g., Pls.’ Trial Summation &
Request for J. at 32, ECF No. 119 (“Asymmetric campaign contribution schemes burden a
candidate’s First Amendment rights per se and are subject to strict scrutiny.”); Pls.’ Proposed
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
79 of 142

80
Findings of Fact & Conclusions of Law (“FOFCOL”) at 29, ECF No. 119-1; Pls.’ Reply in Supp.
of Request for J. at 30, ECF No. 123.30 Because standing must be determined on a plaintiff-by-
plaintiff, claim-by-claim basis and because Senator Pelton and Mr. Lopez’s standing to bring
Claim Two gives me jurisdiction over that claim, I conclude that it is unnecessary to determine
whether Mr. House has standing to bring Claim Two. See Bronson, 500 F.3d at 1106; Rumsfeld
v. F. for Acad. & Institutional Rts., Inv., 547 U.S. 47, 52 n.2 (2006).
Senator Pelton’s Standing
Senator Pelton has run for state office three times and been successful in all three races,
serving two terms in the House and currently his first in the Senate. Trial Tr. at 105:24-25,
107:2-20. Senator Pelton accepted § 4’s voluntary spending limits in all three races. Id. at
133:10-24, 134:7-9. Defendants argue that Senator Pelton has lacked standing to bring Claim
Two from the inception of this case because he “was never injured by the voluntary spending
limits and cannot show that he faced impending harm when this suit was filed.” Defs.’ Written
Summation at 7. Defendants point to Senator Pelton’s testimony that, in his past races, he
accepted the voluntary spending limits, spent well below those limits, and still outspent and beat
his opponent. Id. (citing Trial Tr. at 126:9-129:22, 133:10-134:14).
Defendants misunderstand Senator Pelton’s claimed injury. They contend that injury is
30 Regarding Mr. House’s standing to bring the constitutional challenge to Colorado’s voluntary
spending limits scheme, Plaintiffs assert: “[Mr.] House wanted to contribute double the standard
contribution limits to his preferred candidates in past elections, just as donors to the opponents of
House’s candidates were permitted to do by Section 4.” Pls.’ Proposed FOFCOL at 2. This
claimed injury corresponds with an equal protection challenge more than a First Amendment
claim. Plaintiffs do not explore this injury elsewhere in their briefing, which confirms that their
theory for Claim Two revolves around the injuries to Mr. Lopez and Senator Pelton, not Mr.
House.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
80 of 142

81
being “subject to a lower contribution limit than [his] opponent.” Defs.’ Written Summation at 6.
But Senator Pelton has unambiguously alleged that his injury from the voluntary spending limits
scheme is that it forces him choose between two options that infringe on his First Amendment
rights. See, e.g., Pls.’ Trial Summation & Request for J. at 32-33 (“Section 4 offers candidates an
unconstitutional Hobson’s choice—no matter what they choose, the law violates their F[ir]st
Amendment rights.”); Pls.’ Proposed FOFCOL at 28 (“Section 4 gives a candidate only two
choices: “abide by a limit on . . . expenditures or endure the burden that is placed on [the right to
spend] by the activation of a scheme of discriminatory contribution limits.” (quoting Davis, 554
U.S. at 740)); Pls.’ Reply in Support of Request for J. at 5 (“The injury lies in having to choose,
because either choice burdens their First Amendment rights.”). “[W]here the plaintiff presents a
nonfrivolous legal challenge, alleging an injury to a protected right such as free speech, the
federal courts may not dismiss for lack of standing on the theory that the underlying interest is
not legally protected.” Walker, 450 F.3d at 1093. I must assume Plaintiffs’ claim “has legal
validity” when evaluating my jurisdiction. Id. at 1088. I cannot consider whether both options
presented by § 4 actually violate candidates’ First Amendment rights, as that would
inappropriately dive into the merits of Plaintiffs’ challenge. See id. I conclude only that Senator
Pelton has claimed a nonfrivolous injury, namely that § 4 has subjected him to an
unconstitutional choice.31
31 In a public financing case cited by Defendants, the First Circuit perceived a potentially broader
injury. See Vote Choice, Inc. v. DiStefano, 4 F.3d 26, 37 (1st Cir. 1993). The plaintiff in that case
was a previous gubernatorial candidate in Rhode Island who had rejected public financing. Id. at
29, 37. The district court had found that “the coerced choice between public and private
financing colored [the plaintiff’s] campaign strategy from the outset.” Id. at 37. The court
deemed this “impact on the strategy and conduct of an office-seeker’s political campaign” to be
“an injury of a kind sufficient to confer standing.” Id. Notably, the court did not limit the
plaintiff’s injury to the asymmetric benefits her publicly financed opponents received; it focused
on the impact of the public financing scheme on her campaign’s “strategy and conduct.”
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
81 of 142

82
Senator Pelton has established the three elements required for standing, and thus I have
jurisdiction over his § 4 claim.
Mootness as to Senator Pelton and Mr. Lopez
Regarding mootness, Senator Pelton and Mr. Lopez argue their circumstances fall within
the exception for disputes capable of repetition, yet evading review. I have already discussed Mr.
Lopez’s future campaign intentions. Senator Pelton intends to run for reelection in 2026, Trial
Tr. at 108:19-20, and testified that he does not intend to accept the voluntary spending limits
during his 2026 campaign. Id. at 135:2-4. He maintains an active Colorado candidate political
committee with up-to-date contribution and expenditure reports. Id. at 108:21-109:5; see also
Trial Ex. 20. As of the date of trial, Senator Pelton had raised $800 for his 2026 campaign. Id. at
109:6-9.
Defendants argue these facts are insufficient to show there is a reasonable expectation
that Senator Pelton and Mr. Lopez will be harmed by § 4’s voluntary spending limits in the
future. Defs.’ Written Summation at 7. According to Defendants, “any alleged injury [they]
might suffer in the future from [the] voluntary spending limits is dependent on the conduct of
themselves and unknown third parties.” Id. at 6. Defendants list a number of events they contend
must take place for Senator Pelton and Mr. Lopez to be harmed: Senator Pelton and Mr. Lopez
must run for office, they must decline voluntary spending limits, they must raise at least ten
percent of the applicable spending limits, and an opposing candidate must choose to accept the
voluntary spending limits. Id. at 6-7. Defendants contend this chain of events “is too attenuated
to establish a reasonable probability that Plaintiffs will be subject to asymmetric contribution
limits in the future.” Id. at 7.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
82 of 142

83
Again, however, Defendants misunderstand Senator Pelton and Mr. Lopez’s claimed
injury. There is no “speculative chain of possibilities” here. Cf. Clapper, 568 U.S. at 410 (listing
five events that would need to occur for the plaintiffs’ communications with non-U.S. persons to
be intercepted under § 702 of the Foreign Intelligence Surveillance Act as feared). Both Senator
Pelton and Mr. Lopez testified that they intend to run for state office in 2026 and have active
candidate political committees. Trial Tr. at 50:17-51:9, 102:2-4, 102:10-24, 108:19-109:5. In
entering their respective races, they face the same decision: whether to accept or reject the
voluntary spending limits, and they claim that choice is unconstitutional because their First
Amendment rights will be burdened regardless of which option they choose. Assuming their
claim has legal validity, Senator Pelton and Mr. Lopez have established “a ‘reasonable
expectation’ or a ‘demonstrated probability’” that they will experience this nonfrivolous injury.
See Walker, 450 F.3d at 1088, 1093; Wisc. Right to Life, 551 U.S. at 463 (quoting Murphy, 455
U.S. at 482).
In sum, the Court has jurisdiction over Claim One as brought by Mr. House, Senator
Pelton, and Mr. Lopez and over Claim Two as asserted by Senator Pelton and Mr. Lopez. I now
proceed with the merits analysis of these claims.
C. Claim One: The Constitutionality of Article XXVIII § 3
With their first claim, Plaintiffs contend the individual limits for campaign contributions
to statewide candidates and legislative candidates in § 3 of Article XXVIII of the Colorado
Constitution violate the First Amendment to the U.S. Constitution. Under Buckley and Randall,
the relevant inquiry is whether Defendants have demonstrated that the contribution limits “are
‘closely drawn’ to match a ‘sufficiently important interest.’” Randall, 548 U.S. at 247. If the
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
83 of 142

84
limits “prevent candidates from ‘amassing the resources necessary for effective [campaign]
advocacy’” or “they magnify the advantages of incumbency to the point where they put
challengers to a significant disadvantage,” they are too restrictive to survive this scrutiny. Id. at
248 (quoting Buckley, 424 U.S. at 21).
The first step is to determine whether the interest asserted by the state to justify the
contribution limits is sufficiently important. See id. at 247. Then, I must use the framework set
out in Randall to evaluate whether the contribution limits are closely drawn. See id. at 248-61.
This process begins with looking for “danger signs,” strong indicators that the challenged
campaign contribution limits may pose serious risks to electoral fairness and democratic
accountability. See id. at 249-53. In Randall, the plurality identified two specific danger signs:
(1) whether the contribution limits are lower than comparable ones found elsewhere in the
United States and (2) whether they are substantially lower than limits previously upheld by the
Supreme Court. Id. Later, in Thompson v. Hebdon, the Court referenced a third danger sign:
whether the contribution limit adjusts with inflation. 589 U.S. at 6. If such danger signs are
present, the court “must review the record independently and carefully with an eye toward
assessing the [limit]’s ‘tailoring,’ that is, toward assessing the proportionality of the restrictions.”
Randall, 548 U.S. at 249. Randall articulated five factors courts should consider in performing
that evaluation:
1. Whether the “contribution limits will significantly restrict the amount of funding
available for challengers to run competitive campaigns,” id. at 253;
2. Whether contribution limits on political parties “threaten[] harm to a particularly
important political right, the right to associate in a political party,” and “reduce the
voice of political parties . . . to a whisper,” id. at 256, 259 (citation modified);
3. Whether volunteer services or expenses are treated as contributions such that they
“impede a campaign’s ability to effectively use volunteers, thereby making it more
difficult for individuals to associate in this way,” id. at 260;
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
84 of 142

85
4. Whether the limits automatically adjust for inflation, id. at 261; and
5. Whether there is any “special justification that might warrant a contribution limit so
low or so restrictive,” id. at 261.32
In scrutinizing a particular contribution limit, courts must focus on whether “the amount,
or level, of that limit could make a difference.” Randall, 548 U.S. at 247 (quoting Buckley at 21).
“[D]istinctions in degree become significant only when they can be said to amount to differences
in kind.” Id. (quoting Buckley, 424 U.S. at 30). Courts possess “no scalpel to probe, whether, say,
a $2,000 ceiling might not serve as well as $1,000.” Id. (quoting Buckley, 424 U.S. at 30). In
other words, courts “cannot determine with any degree of exactitude the precise restriction
necessary to carry out the statute’s legitimate objectives.” Id. at 248.
1. Governmental Interest
The first question is whether the state’s interest in imposing the contribution limits in § 3
is sufficiently important. There can be little doubt as to what the official purpose of the § 3
contribution limits are. Article XXVIII § 1 explicitly states: “The people of the state of Colorado
hereby find and declare that large campaign contributions to political candidates create the
potential for corruption and the appearance of corruption” and that “the interests of the public are
best served by limiting campaign contributions.” Colo. Const. art. XXVIII, § 1.
Supreme Court “cases have made clear that the prevention of corruption or its appearance
constitutes a sufficiently important interest to justify political contribution limits.” McConnell v.
32 I note that the claims in Randall were brought by individuals who had run for state office in
Vermont, citizens who contributed to electoral campaigns in Vermont, and political parties there.
548 U.S. at 239-40. To some extent, these factors take into account the potentially broad impacts
of contribution limits on the First Amendment rights of each of those groups.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
85 of 142

86
Fed. Election Comm’n, 540 U.S. 93, 143 (2003), overruled in part on other grounds by Citizens
United v. Fed. Election Comm’n, 558 U.S. 310 (2010).
Corruption is a subversion of the political process. Elected officials are influenced
to act contrary to their obligations of office by the prospect of financial gain to
themselves or infusions of money into their campaigns. The hallmark of corruption
is the financial quid pro quo: dollars for political favors.
Fed. Election Comm’n v. Nat’l Conservative Pol. Action Comm., 470 U.S. 480, 497 (1985). The
Supreme Court once held a more traditional view of the corruption that states might properly
target with campaign finance regulations, recognizing that corruption extends beyond just its
“hallmark” of quid pro quo arrangements to the broader undue influence large contributors may
have on officeholders. Shrink Mo., 528 U.S. at 389 (“In speaking of ‘improper influence’ and
‘opportunities for abuse’ in addition to ‘quid pro quo arrangements,’ we recognized a concern
not confined to bribery of public officials, but extending to the broader threat from politicians too
compliant with the wishes of large contributors” (quoting Buckley, 424 U.S. at 28)); Fed.
Election Comm’n v. Colo. Republican Fed. Campaign Comm., 533 U.S. 431, 441 (2001)
(explaining that corruption is “understood not only as quid pro quo agreements, but also as
undue influence on an officeholder’s judgment, and the appearance of such influence” (quoting
Shrink Mo., 528 U.S. at 388-89)); McConnell, 540 U.S. at 143 (“We have not limited [the
interest in the prevention of corruption or its appearance] to the elimination of cash-for-votes
exchanges.”). However, in Citizens United v. Federal Election Commission, the Court narrowed
its view, declaring: “When Buckley identified a sufficiently important governmental interest in
preventing corruption or the appearance of corruption, that interest was limited to quid pro quo
corruption.”33 558 U.S. 310, 359 (2010); see also McCutcheon, 572 U.S. at 207 (“[W]hile
33 Existing campaign finance precedent is not fully in line with the Supreme Court’s now favored
originalist interpretive method, but I have no reason to conclude the outcome here would be
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
86 of 142

87
preventing corruption or its appearance is a legitimate objective, Congress may target only a
specific type of corruption—‘quid pro quo’ corruption.”); Fed. Election Comm’n v. Cruz, 596
U.S. 289, 305 (2022) (“[The Supreme] Court has recognized only one permissible ground for
restricting political speech: the prevention of ‘quid pro quo’ corruption or its appearance.”). So,
for Colorado’s anticorruption interest to be deemed sufficiently important, Supreme Court
precedent requires that the specific corruption targeted by § 3’s individual contribution limits be
quid pro quo corruption or its appearance.
The Supreme Court has described quid pro quo corruption as “a direct exchange of an
different even if that method were applied. Despite the adoption of originalist models privileging
history and tradition in other contexts, see, e.g., N.Y. State Rifle & Pistol Ass’n, Inc. v. Bruen,
597 U.S. 1, 26-31 (2022), the Supreme Court has largely refrained from framing its campaign
finance jurisprudence in such a manner, see, e.g., Citizens United, 558 U.S. at 432 (Stevens, J.,
concurring in part and dissenting in part) (“[O]ur campaign finance jurisprudence has never
attended very closely to the views of the Framers . . . .”). But the Court has applied history-
focused tests in other First Amendment contexts. See, e.g., Bruen, 597 U.S. at 24-25 (explaining
that, when the government restricts speech, it “must generally point to historical evidence about
the reach of the First Amendment’s protections” when it seeks to show that “the expressive
conduct falls outside of the category of protected speech”); Kennedy v. Bremerton Sch. Dist., 597
U.S. 507, 535 (2022) (“[T]his Court has instructed that the Establishment Clause must be
interpreted by ‘reference to historical practices and understandings.’” (quoting Town of Greece v.
Galloway, 572 U.S. 565, 576 (2014))). Reading the tea leaves, the Court might soon require a
similar historical review for First Amendment challenges to campaign finance regulations. See,
e.g., Nat’l Republican Senatorial Comm. v. Fed. Election Comm’n, 117 F.4th 389, 411 (6th Cir.
2024) (en banc) (Bush, J., concurring dubitante) (in a case challenging the Federal Election
Campaign Act’s limits on coordinated campaign expenditures by political parties, stating: “[I]t
appears that many, and perhaps a majority, of Justices would consider evidence of history and
tradition as relevant to analyze the constitutionality of the regulation at issue here”), cert.
granted, 145 S. Ct. 2843 (June 30, 2025). Were the Court to provide such instruction, there is
evidence suggesting that, at a minimum, the Founding Generation would have understood the
First Amendment to permit campaign finance regulations that target corruption. See, e.g., Jud
Campbell, Natural Rights and the First Amendment, 127 Yale L.J. 246, 314 (2017) (“Even
assuming that giving money to a campaign is expressive, or is an exercise of the natural right to
freedom of association, this activity was among the countless aspects of natural liberty subject to
regulations that promote the general welfare.”). The parties tailored their case to existing
precedent and thus did not craft the record for me to conduct a full historical review. As a result,
I do not undertake this analysis; I note only that I do not see this potential for jurisprudential
evolution as casting a shadow of doubt over the conclusions I reach here.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
87 of 142

88
official act for money” or “‘dollars for political favors’” McCutcheon, 572 U.S. at 192 (quoting
Nat’l Conservative Political Action Comm., 470 U.S. at 497). Criminal law provides some
additional insight into what might constitute quid pro quo corruption in this context. The
requisite “official act” for a federal bribery offense, i.e., the quo, is defined as “any decision or
action on any question, matter, cause, suit, proceeding or controversy, which may at any time be
pending, or which may by law be brought before any public official, in such official’s official
capacity, or in such official’s place of trust or profit.” 18 U.S.C. § 201(a)(3); see also McDonnell
v. United States, 579 U.S. 550, 562, 573 (2016) (clarifying that an agreement to engage in such
an act, even one that is not explicit, falls within the definition). “Setting up a meeting, talking to
another official, or organizing an event (or agreeing to do so)—without more—does not fit that
definition.” McDonnell, 579 U.S. at 574. However, a legislator’s agreement to vote a certain way
on a bill that is likely to be brought before the legislator in exchange for money likely would.
While the type of corruption that may be targeted is limited to the receipt of money for
some official act or favor, preventing the appearance of quid pro quo corruption is also a
sufficiently important governmental interest. See Buckley, 424 U.S. at 27. The Supreme Court
has broadly defined the appearance of quid pro quo corruption as “‘public awareness of the
opportunities for abuse inherent in a regime of large individual financial contributions’ to
particular candidates.” McCutcheon, 572 U.S. at 207 (quoting Buckley, 424 U.S. at 27). Thus,
Defendants must show that the purpose of Colorado’s individual contribution limits is to prevent
officials from accepting money in exchange for an official act or to prevent the public perception
that there are opportunities for such abuse to occur.
The history of individual campaign contribution limits in Colorado suggests that
Coloradans voted to enact Amendment 27 because they did not trust their legislators to set
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
88 of 142

89
individual contribution limits that would prevent such corruption or its appearance. Over a six-
year period, a “back and forth” occurred between the legislature and the citizens, ultimately
resulting in Amendment 27. Trial Tr. at 24:7-9. The efforts began when the legislature raised
Colorado’s long-standing individual contribution limits in 1996. See Trial Ex. 353 at 2-3, Trial
Ex. 374. The people of Colorado then passed a citizen’s initiative to substantially reduce those
limits in the Colorado statutes. See Trial Ex. 356 at 2, Trial Ex. 374. The legislature responded
by amending the statutes again and raising the limits to levels that were five to ten times as high
as those set by the citizen’s initiative. See Trial Ex. 354 at 1, Trial Ex. 374. The people of
Colorado had the last word by amending the Colorado Constitution and once more reducing the
limits to the level they thought was necessary to address corruption or its appearance. See Colo.
Ethics Watch, 269 P.3d at 1253. From this history, it is easily inferred that the contribution limits
were adopted because the public believed that the previously set higher limits for individual
contributions presented opportunities for abuse. There is no question that the anticorruption
purpose endorsed by the people of Colorado in Amendment 27 is legitimate and not
disingenuous or a post hoc justification.34
34 The Randall plurality observed that the Court ordinarily deferred to the legislature’s
determination of the appropriate level for contribution limits, as “legislators have ‘particular
expertise’ in matters related to the costs and nature of running for office.” 548 U.S. at 248
(quoting McConnell, 540 U.S. at 137). Because § 3 was enacted by popular vote, no deference
based on legislative expertise is warranted. However, the Supreme Court’s definition of the
appearance of corruption includes a subjective component—the public’s “‘awareness of the
opportunities for abuse . . . .’” McCutcheon, 572 U.S. at 207 (quoting Buckley, 424 U.S. at 27).
Considering that definition, deference should be given to Colorado voters regarding their
perception of opportunities for corruption and abuse in the financing of state officials’
campaigns. Cf. Shrink Mo., 528 U.S. at 394 (identifying the fact that Missouri’s contribution
limits were adopted by an overwhelming majority of the State’s voters as evidence that the
contribution limits “are necessary to combat corruption and the appearance thereof”). Colorado
voters perceived that there were opportunities for abuse with the contributions given to
candidates and that § 3’s contribution limits would address those opportunities.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
89 of 142

90
Nevertheless, Plaintiffs contend Defendants did not present sufficient evidence to
establish that this purpose of preventing corruption or its appearance relates to quid pro quo
corruption specifically. Plaintiffs also maintain that the legitimacy of the purpose is undermined
by the State’s voluntary spending limits scheme.35
35 Bracketed language within quotations is meant to bring clarity to the quoted language, not to
replace a court’s words with what the parties think could, should, or would have been said. There
are many ways to use language from a case to support an argument, but giving the false
impression that a court has already made a statement that it has not made is misrepresentative
and misleading. Plaintiffs take this liberty throughout their briefs. For example:
 Plaintiffs write: “‘Either [Colorado] is openly tolerating a significant risk’ of quid pro
quo corruption in exchange for candidates limiting their spending, or the contribution
limits ‘are in no real sense’ designed to prevent corruption.” Pls.’ Trial Summation &
Request for J. at 8 (quoting Fed. Election Comm’n v. Ted Cruz for Senate, 596 U.S. 289,
313 (2022)). But without any discussion of Colorado, the Supreme Court in Ted Cruz for
Senate actually stated: “Either the Government is openly tolerating a significant number
of ‘gifts’ far more generous than what it would normally think fit to allow, or post-
election contributions that go toward retiring campaign debt are in no real sense ‘gifts’ to
a candidate.” 596 U.S. at 312.
 Plaintiffs write: “And, unlike the Federal Government’s treatment of comparable
requirements, [Colorado] has not (insofar as [plaintiffs] are aware) created an exception
excluding such expenses.” Pls.’ Trial Summation & Request for J. at 24 (quoting
Randall, 548 U.S. at 259)). But the Supreme Court in Randall referred to the State of
Vermont and the Court’s own knowledge, stating: “And, unlike the Federal
Government’s treatment of comparable requirements, the State has not (insofar as we are
aware) created an exception excluding such expenses.” 548 U.S. at 259.
 Plaintiffs write: “Ultimately, the [Davis, AFE, and Riddle] law[s] failed because [they]
imposed different contribution limits on candidates vying for the same seat.” Pls.’ Trial
Summation & Request for J. at 32 (quoting Riddle v. Hickenlooper, 742 F.3d 922, 929
(2014)). But, discussing only Davis, the Tenth Circuit in Riddle stated: “Ultimately, the
law failed because it imposed ‘different contribution . . . limits on candidates vying for
the same seat.’” 742 F.3d at 929 (quoting Davis, 554 U.S. at 743-44).
 Plaintiffs write: “‘Much like the burden placed on speech in Davis, the [differential
contribution limits] “imposes an unprecedented penalty on any candidate who robustly
exercises [his] First Amendment right[s]”’ . . . .” Proposed FOFCOL at 28 (quoting
Bennett, 564 U.S. at 736). But the Supreme Court’s statement in Bennett was related to
the matching funds provision at issue there, not “differential contribution limits.” It reads:
“Much like the burden placed on speech in Davis, the matching funds provision ‘imposes
an unprecedented penalty on any candidate who robustly exercises [his] First
Amendment right[s].’” 564 U.S. at 736 (quoting Davis, 554 U.S. at 739).
This habit of using brackets to obscure case law has required me to review Plaintiffs’ arguments
and the cases they cited with considerable skepticism.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
90 of 142

91
The Showing Defendants Must Make
The parties dispute the standard Defendants must meet in establishing that the stated
interest is sufficiently important. After reviewing the cases cited by the parties, I conclude the
standard synthesized by Senior U.S. Circuit Judge Jane B. Stranch of the Sixth Circuit Court of
Appeals is appropriate:
When the Government defends a campaign finance regulation, it “bears the burden
of proving” that the regulation in fact addresses quid pro quos. McCutcheon, 572
U.S. at 210 (quoting United States v. Playboy Ent. Grp., Inc., 529 U.S. 803, 816
(2000)). That proof can come from record evidence, legislative findings, or
experience under the law. See Cruz, 596 U.S. at 307; McCutcheon, 572 U.S. at 219.
It is considered with an eye to context, minding the “difficulty of mustering
evidence to support long-enforced statutes,” [Colo. Republican Fed. Campaign
Comm.], 533 U.S. at 457, and recognizing “that no data can be marshaled to capture
perfectly the counterfactual world in which” the regulation does “not exist,”
McCutcheon, 572 U.S. at 219. For these reasons, the “quantum of empirical
evidence needed to satisfy heightened judicial scrutiny of legislative judgments will
vary up or down with the novelty and plausibility of the justification raised.”
McConnell, 540 U.S. at 144, (quoting Shrink Missouri, 528 U.S. at 391).
Nat’l Republican Senatorial Comm. v. Fed. Election Comm’n, 117 F.4th 389, 430-31 (6th Cir.
2024) (en banc), (Stanch, J., concurring in the judgment). Colorado’s contribution limits have
existed under § 3 for over twenty years, and “the dangers of large, corrupt contributions and the
suspicion that large contributions are corrupt are neither novel nor implausible” justifications,
Shrink Mo., 528 U.S. at 391. Thus, the quantum of evidence Defendants must marshal is not as
great as required for other campaign finance regulations courts have considered. See, e.g., Fed.
Election Comm’n v. Ted Cruz for Senate, 596 U.S. 289, 306 (2022) (considering the
constitutionality of a restriction on a campaign’s use of post-election funds to repay loans from
the candidate and viewing with skepticism the government’s asserted anticorruption interest
because the measure constituted a prophylaxis-upon-prophylaxis approach, as individual
contributions to the candidate were already regulated). When plaintiffs make a “showing of their
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
91 of 142

92
own to cast doubt on the apparent implications of Buckley’s evidence and the record,” however,
more extensive evidentiary documentation might become necessary. Shrink Mo., 528 U.S. at
394.
Pushing for the application of an even more stringent standard, Plaintiffs fixate on the
Supreme Court’s evaluation of the evidence in Federal Election Commission v. Ted Cruz for
Senate and attempt to raise the evidentiary requirement to an unreachable height. Citing Ted
Cruz for Senate, Plaintiffs insist that the question is whether Colorado’s contribution limits
“actually further” its stated interest. Reply in Support of Request for J. at 9-10 (citing Ted Cruz
for Senate, 596 U.S. at 309-13). But, as Defendants explain, the facts in Ted Cruz for Senate are
dissimilar from those in this case. See Defs.’ Written Summation at 14. In that case, the Supreme
Court considered whether the government’s anticorruption interest could justify a regulation it
had not previously considered that it perceived to be a “prophylaxis-upon-prophylaxis” measure.
596 U.S. at 306. The Court cited the fact that individual contributions to candidates were already
capped as a reason the regulation at issue might not be necessary. Id. Here, the challenged
regulation is a contribution limit, which is the first level of prophylaxis and for which the link to
reducing quid pro quo corruption or its appearance has been repeatedly established.
The pertinent question is whether Defendants have shown that Colorado’s interest in
preventing quid pro quo corruption or its appearance is sufficiently important to justify its
individual contribution limits. That inquiry does not depend on whether Defendants have shown
that quid pro quo corruption exists or existed in Colorado or that § 3’s individual contribution
limits have been successful in eliminating such corruption, even if related evidence is relevant.
Another sticking point for the parties is whether the specific amount of Colorado’s
individual contribution limits should be evaluated at this stage in the analysis or whether that
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
92 of 142

93
level is more appropriately considered at a later stage. In Lair v. Motl, the Ninth Circuit
explained that, initially, the proper inquiry “is whether the state has presented sufficient evidence
of a valid interest, not whether it has justified a particular dollar amount. The latter inquiry, if
ever appropriate, occurs in the second part of [the] analysis, in examining whether the restriction
is ‘closely drawn.’” 873 F.3d 1170, 1178 (9th Cir. 2017) (quoting Mont. Right to Life Ass’n v.
Eddleman, 343 F.3d 1085, 1092 (9th Cir. 2003)).36 Randall does not speak directly to this issue,
nor does Thompson. Thus, Lair’s take that the amount of the limits should be considered in the
second part of the test does not diverge from current Supreme Court instruction.
I conclude, as the court did in Lair, that consideration of the particular level of the
contribution limits should occur at later steps in the analysis. This is consistent with Randall’s
unique framework. In Randall, the plurality summarily accepted the State’s anticorruption
interest and then, in evaluating whether “danger signs” existed and whether the limits
“restrict[ed] the amount of funding available for challengers to run competitive campaigns,” it
specifically considered the level of the contribution limits. 548 U.S. at 247-56 (as part of its
danger signs analysis, instructing that lower limits, despite their ability to prevent corruption and
the appearance of corruption, are not always better because limits that are too low may “harm the
electoral process” and “reduc[e] democratic accountability”); see also id. at 247 (“[Buckley]
recognized that, in determining whether a particular contribution limit was ‘closely drawn,’ the
36 Plaintiffs argue Lair’s reasoning should be disregarded because it “relied on the same Ninth
Circuit precedent that the Supreme Court unanimously and summarily rejected in Thompson.”
Pls.’ Reply in Support of Request for J. at 6. But, unlike the Ninth Circuit’s opinion in
Thompson, the Lair opinion discussed here did not decline to apply Randall. Compare Thompson
v. Hebden, 909 F.3d 1027, 1037 n.5 (9th Cir. 2018) (“Randall is not binding authority because no
opinion commanded a majority of the Court.”), with Lair, 873 F.3d at 1186-87 (applying Randall
in the alternative and concluding the challenged contribution limits would survive scrutiny under
Randall); see also Pls.’ Reply in Support of Request for J. at 15 n.5 (acknowledging that the Lair
opinion “also discusses Randall”).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
93 of 142

94
amount, or level, of that limit could make a difference.”). Requiring the State to justify the
particular dollar amount of its limits at the first step of the Randall test would make later parts of
the analysis redundant or superfluous.
With the appropriate standard settled and taking into account the specific circumstances
in this case, I review Defendants’ evidence that § 3’s individual contribution limits are justified
by the Colorado’s interest in preventing quid pro quo corruption or its appearance.
The Showing Defendants Did Make
Defendants have shown that Amendment 27 was approved by 66.5% of Coloradans who
voted on the Amendment. See Trial Ex. 360 at 144-45. This great majority of voters approved
not only § 3’s contribution limits but also the purpose behind them, which was, as relevant here,
limiting “large campaign contributions to political candidates [that] create the potential for
corruption and the appearance of corruption.” Colo. Const. art. XXVIII, § 1. This specific
language used in the Amendment is significant because it mirrors the words of the Supreme
Court in Buckley, which the Court has determined relate to quid pro quo corruption. In Buckley,
the Court accepted the governmental purpose of “limit[ing] the actuality and appearance of
corruption resulting from large individual financial contributions” as a sufficient justification for
the contribution limits at issue. 424 U.S. at 26. This language is among the “comments in
Buckley that show quid pro quo was of central importance to the analysis.” McConnell, 540 U.S.
at 295-96 (Kennedy, J., concurring in the judgment in part and dissenting in part) (citing
Buckley, 424 U.S. at 26-27, 45); Citizens United, 558 U.S. at 359 (relying on Justice Kennedy’s
analysis in McConnell in concluding that the endorsed governmental interest in Buckley was
limited to quid pro quo corruption).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
94 of 142

95
Along with this anticorruption purpose, Amendment 27 indicated that the reforms were
also adopted to address the concern “that large campaign contributions made to influence
election outcomes allow wealthy individuals, corporations, and special interest groups to
exercise a disproportionate level of influence over the political process.” Colo. Const. art.
XXVIII, § 1. This purpose relates to the more general influence money might have in politics.
The fact that it was separately identified supports the conclusion that the anticorruption interest
relied on here is distinct and targets quid pro quo corruption, not any broader notion of
corruption that includes improper influence.
In addition to Amendment 27’s text and history, Defendants presented anecdotal and
expert evidence that the specific “corruption” Amendment 27 targets is quid pro quo corruption
and its appearance. Former State Senator Donovan testified about her experience meeting with a
group that conditioned future financial support for her campaign on how she might vote on a
specific anticipated piece of legislation. See Trial Tr. at 814:7-21. She stated: “[T]hey asked me
directly how I would vote on a hypothetical bill that would likely come up in the next session
and then directly implied that their endorsement and subsequent financial support hinged on my
answer to that question.” Id. at 814:9-13. Similarly, Mr. Buescher, former Colorado
Representative and Secretary of State, gave an example of a donor lobbyist informing a legislator
of issues the donor was concerned with and “made it clear” if the legislator voted the way the
donor wanted, he would have $500,000 for his race. Id. at 971:1-8. Mr. Buescher testified that
there were many examples of similar interactions. Id. at 971:10-20.
Plaintiffs argue this testimony is not relevant because the interactions Ms. Donovan and
Mr. Buescher described would not constitute quid pro quo corruption. Pls.’ Trial Summation &
Request for J. at 9. Plaintiffs claim that “[t]he examples that both witnesses identified as
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
95 of 142

96
potential quid pro quo corruption merely described routine lobbyist and constituent interactions”
and thus “[n]either witness identified actual quid pro quo corruption.” Pls.’ Proposed FOFCOL
at 5. According to Plaintiffs, “Senator Donovan described . . . an interest group explaining that
they would not endorse her if she did not support legislation that the group cared about.” Pls.’
Trial Summation & Request for J. at 10 (citing Trial Tr. at 830). However, Ms. Donovan testified
that the group was offering not just an endorsement but also the maximum financial contribution
the group could make—approximately $4,000.37 Trial Tr. at 814:9-13, 815:2-6. If Ms. Donovan
had agreed to vote as the group wanted on the bill that would likely come before her in exchange
for the group’s financial support, she would have engaged in quid pro quo corruption. And the
appearance of quid pro quo corruption stems from these opportunities in which individuals or
groups offer money for official acts, even if such exchanges are rare and couched in vague and
careful language. Because it is not clear that the interaction involved a specific piece of
legislation that was likely to be brought before the legislator or any other official act, Mr.
Buescher’s account of a legislator being offered $500,000 to support his campaign in exchange
for voting the way the donor wanted is less obviously quid pro quo corruption. Still, this
interaction reinforces the public awareness that large contributions create the potential for quid
pro quo corruption. On the whole, the testimony of Defendants’ witnesses confirms, as the
Supreme Court has previously found, “that the problem is not an illusory one.” Buckley, 424 U.S.
at 27.
To contradict this evidence, Plaintiffs offered their own testimony and that of Mr.
Wadhams. Mr. Lopez and Senator Pelton testified that they have followed Colorado politics for
37 Notably, this amount is greater than Colorado’s individual contribution limits, and Ms.
Donovan never had “a similar experience where the amount at issue was a $400 maximum
individual contribution.” Trial Tr. at 815:7-9.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
96 of 142

97
thirty-six years and twenty-five years, respectively, but were not aware of any occurrence of quid
pro quo corruption involving campaign contributions. Trial Tr. at 88:14-18, 125:24-126:3.
Similarly, Mr. Wadhams, the former Colorado Republican State Chairman, explained that, over
his entire life following Colorado politics, he was not aware of any appearance of quid pro quo
corruption involving campaign contributions. Id. at 668:6-11. This testimony does not negate the
experiences of Ms. Donovan and Mr. Buescher, who have spent significant time as elected
officials in Colorado and testified about opportunities in which such corrupt exchanges could
have occurred.
Even if Defendants could not provide recent evidence of quid pro quo crimes involving
individual campaign contributions, that does not mean the purpose of the individual contribution
limits must be something other than preventing quid pro quo corruption or its appearance, as
Colorado has long had individual contribution limits and § 3’s limits have been in place for over
twenty years. See McCutcheon, 572 U.S. at 219 (“recogniz[ing] that no data can be marshaled to
capture perfectly the counterfactual world”). On the contrary, this lack of evidence supports the
inference that the long-standing contribution limits are working.38
To further substantiate that Colorado’s interest in § 3’s individual contribution limits is to
prevent quid pro quo corruption or its appearance, Defendants offered the opinions of their
expert Dr. Wood. She studied national and Colorado-specific data and concluded that “the
Colorado electorate that enacted Amendment 27 and still the Colorado electorate . . . of today is
38 Plaintiffs also argue that every state anticipates the harm caused by quid pro quo corruption,
and because nearly 25% of them have no contribution limits, Colorado’s contribution limits are
not justified. See Pls.’ Trial Summation & Request for J. at 7 (citing Trial Ex. 1). Putting aside
that thirty-eight states do have contribution limits, see Trial Ex. 1 at 1-2, the fact that other states
may have chosen not to act despite an awareness that large contributions create an opportunity
for abuse is irrelevant to whether Colorado’s interest in its contribution limits is sufficiently
important.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
97 of 142

98
concerned with quid pro quo corruption and not only [undue] influence.” Trial Tr. at 988:22-25.
She ascertained that Americans and Coloradans considered exchanging money for votes as the
most corrupt action out of a range of actions associated with political corruption. Id. at 992:9-14;
Trial Ex. 318. She also found that Coloradans are more likely to believe that bribery is more
corrupt than other activities associated with political influence. Trial Tr. at 995:16-20; Trial Ex.
319. Dr. Wood thus determined that Colorado voters “were most concerned about quid pro quo
corruption.” Trial Tr. at 998:4-11.
To gain more insight into the Colorado public’s opinions about corruption before
Amendment 27 was passed, Dr. Wood also conducted a review of newspaper articles from 1998-
2002, searching for ones that mentioned “corruption” with “government” or “candidate.” Id. at
1003:16-1004:6. Of the 62 relevant articles she found, 17.7% of the articles discussed only quid
pro quo corruption whereas 6.5% discussed only improper influence. Id. at 1004:1-14, 1004:21-
1005:10. As a result, she concluded that most Coloradans around the time that Amendment 27
was adopted did not view corruption as only influence. Id. at 1003:23-1004:2, 1006:7-14. Her
opinions on these matters are reliable and persuasive. They reinforce the sincerity of the explicit
purpose in Article XXVIII and bolster the conclusion that the corruption § 3’s individual
contribution limits were intended to target was quid pro quo corruption specifically.
In an effort to show that Amendment 27 does not target the prevention of quid pro quo
corruption or its appearance, Plaintiffs presented the opinions of Dr. Primo. He testified that
“Americans think campaign contributions . . . [are] the source of corruption.” Id. at 317:20-23.
But according to him, “Colorado’s campaign contribution limits have no meaningful impact on
quid pro quo corruption or the appearance of corruption.” Id. at 292:5-7. More generally, he
opined that, “[i]f a state were to eliminate all contribution limits and public funding systems
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
98 of 142

99
today, it should expect nothing much to happen with regard to trust and confidence in state
government.” Id. at 274:7-10. Dr. Primo based his conclusion on the fact that Americans think
that “everything in politics is basically corrupt.” Id. at 317:21-25. As I found above, the analysis
on which Dr. Primo bases his conclusions, while rigorous, suffers from numerous flaws and
consequently his opinions are of minimal assistance.
The most severe flaw in Dr. Primo’s methodology is that he used “trust and confidence in
state government” as proxies for measuring public opinion on quid pro quo corruption. The use
of proxies in statistical analyses is often an inevitability, but the validity of a conclusion based on
a proxy depends on the proxy’s closeness to the variable in question. Here, the proxies Dr. Primo
selected are not equivalent or apparently close stand-ins for quid pro quo corruption. While I
suspect that trust and confidence in government and the absence of quid pro quo corruption are
related, Dr. Primo’s opinions and Plaintiffs’ related arguments rest on unsustainable assumptions
about the connection between the two.
Dr. Primo formulated his opinions after reviewing data gathered from responses to survey
questions about trust and confidence in government. The survey questions illustrate how his
analysis does not pertain to quid pro quo corruption in particular, or really even corruption
generally. For example, respondents were asked: “[H]ow much trust and confidence do you have
in your state government when it comes to handling state problems?” Id. at 250:20-23; Trial Ex.
305 at 2. This question implicitly recognizes that state governments must address many problems
and that individuals may view some of those problems as more important than others and may
have trust and confidence in the government to address some problems but not others. If, in
responding to this question, individuals were expected to consider the existence of opportunities
for state candidates to accept campaign contributions in exchange for official acts, that
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
99 of 142

100
expectation was buried so deep as to be indiscernible.
Plaintiffs contend that, “if someone thought the government is corrupt, she would also
not have confidence in the government’s ability to handle problems.” Pls.’ Reply in Support of
Request for J. at 13. But this point is not self-evident; while it is possible that an individual who
thinks the government is corrupt would not have confidence in the government, Plaintiffs have
not established that the presence of a problem within an institution necessarily leads to the
undermining of confidence in the institution as a whole. Indeed, Dr. Primo admitted that a lack
of corruption is not a necessary component of trust in government. Trial Tr. at 322:21-23. And,
as just discussed, individuals may value certain government functions so greatly that, as long as
those tasks are fulfilled, the existence of corruption would not reduce their trust and confidence
in government.
Even supposing Dr. Primo’s analysis may offer some insight regarding public perceptions
of corruption, it does not follow that equally reliable conclusions could be drawn about quid pro
quo corruption in particular. Plaintiffs justify Dr. Primo’s use of trust and confidence as proxies
for quid pro quo corruption by pointing to a sentence in Buckley that ties the appearance of
corruption to “confidence in the system of representative government.” Pls.’ Reply in Support of
Request for J. at 13 (quoting Buckley, 424 U.S. at 27). However, Plaintiffs ignore the complete
statement from the Buckley Court, that is: “Congress could legitimately conclude that the
avoidance of the appearance of improper influence ‘is also critical . . . if confidence in the system
of representative Government is not to be eroded to a disastrous extent.’” Buckley, 424 U.S. at 27
(quoting U.S. Civ. Serv. Comm’n v. Nat’l Ass’n of Letter Carriers, AFL-CIO, 413 U.S. 548, 565
(1973)). The statement clearly pertains to the broader concept of corruption, including improper
influence, not just quid pro quo corruption. It does not support the assumption that high levels of
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
100 of 142

101
confidence in government necessarily indicate that the public perceives no opportunities for quid
pro quo corruption. Because the evidence here must ultimately relate to quid pro quo corruption,
Buckley does not legitimize Dr. Primo’s opinions or Plaintiffs’ approach.
In effect, Plaintiffs argue out of both sides of their mouths by insisting that the only
acceptable government interest is preventing quid pro quo corruption or its appearance, while
relying on expert analysis that focuses on a much broader view of government functioning. This
position is untenable. If Defendants had presented expert testimony that Coloradans had little
trust and confidence in government at the time Amendment 27 was enacted, Plaintiffs
undoubtedly would have responded that the opinion was irrelevant as it did not provide any
insight into whether Coloradans were concerned about quid pro quo corruption specifically. So,
too, Dr. Primo’s analysis using trust and confidence in government as proxies cannot support his
opinion that § 3’s individual contribution limits have no meaningful impact on quid pro quo
corruption or the appearance of corruption, and his analysis ultimately does not help me
determine whether those limits were adopted to address or have been successful in reducing quid
pro quo corruption or its appearance.
The other criticisms of Dr. Primo’s analysis—that it does not examine variability in
contribution limit amounts, does not consider the precise levels at which the contribution limits
in Colorado are set, and fails to examine views held by Coloradans specifically—likewise
undermine his opinions, albeit to a lesser extent. See Defs.’ Written Summation at 19-21.39 Dr.
39 Defendants also contend that Dr. Primo’s analysis is flawed because he used imprecise and
inconsistent methods in creating his dataset. See Defs.’ Written Summation at 21. Specifically,
Defendants point to Dr. Primo’s conversion of every survey response into a 100-point scale. But
some responses were based on a four-answer survey while others were based on a five-answer
survey. See Trial Ex. 305. As explained in my Findings of Fact, this is problematic given the
different weight given to responses based on which survey the individual responded to.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
101 of 142

102
Primo testified that his opinions pertained to all levels of contribution limits. Trial Tr. at 353:7-
355:9. In other words, his opinion would be that no contribution limit would have any
meaningful impact on quid pro quo corruption or its appearance. See id. at 292:5-7, 357:18-
358:4. Common sense and Supreme Court precedent counsel that this is wrong. Additionally, Dr.
Primo’s opinions were based on nationwide data and did not consider the particular level at
which Colorado’s contribution limits are set or the potentially unique perceptions of Colorado
voters. See id. at 302:8-12 (explaining that he did not run an analysis of Colorado’s limits
specifically because he did not feel doing so would be consistent with sound social science based
on the data he had); id. at 309:22-310:6 (acknowledging that he did not know whether Colorado
voters’ attitudes about contribution limits and the appearance of corruption might differ from
those of voters in other states); id. at 311:15-312:2 (testifying that “based on [his] knowledge of
the states and campaign finance laws, it didn’t seem . . . that there would be a reason to think that
Colorado voters were somehow special or different”). National data could certainly be relevant
to whether contribution limits in Colorado might impact Colorado voters’ perception of the
opportunities for quid pro quo corruption. However, it cannot be presumed that the specific
circumstances in Colorado will mirror nationwide conclusions without some discussion of why
that is so. Dr. Primo focused his conclusions on Colorado even though he used nationwide data
and offered no basis for assuming that the circumstances in Colorado would mirror those
elsewhere. In contrast, using statistically acceptable methods, Dr. Wood used Colorado-specific
data to arrive at her Colorado-focused conclusions.
I also cannot overlook the fact that Plaintiffs again hold Defendants to a higher standard
than their expert: Plaintiffs criticize Defendants for failing to produce “any evidence whatsoever
of any kind of quid pro quo corruption in Colorado involving campaign contributions,” Pls.’
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
102 of 142

103
Trial Summation & Request for J. at 9 (emphasis added), while they simultaneously advocate
that Dr. Primo’s opinions based on nationwide data are conclusive in this Colorado-specific case,
see Pls.’ Reply in Support of Request for J. at 13. Finally, even if Americans think that
everything in politics is corrupt, see Trial Tr. at 317:21-25, that fact does not support the
inference that Coloradans did not intend to specifically target quid pro quo corruption or its
appearance in adopting § 3’s contribution limits. For all these reasons, I give Dr. Primo’s
opinions very little weight and conclude they do not undermine Defendants’ evidence that
Colorado’s individual contribution limits were adopted to prevent quid pro quo corruption or its
appearance.
Since Buckley, courts have held that this purpose is a sufficiently important governmental
interest to justify limitations on campaign contributions. See, e.g., Buckley, 424 U.S. at 26-27;
McConnell, 540 U.S. at 143. Plaintiffs baldly assert that they have made a showing that “cast[s]
doubt on the apparent implications of Buckley’s evidence and the record here.” Pls.’ Reply in
Support of Request for J. at 9 (quoting Shrink Mo., 528 U.S. at 394). I do not see how.
Defendants have presented valid evidence that aligns with the conclusions in Buckley, including
the history and language of Amendment 27, the testimony of Ms. Donovan and Mr. Buescher,
and the opinions of Dr. Wood. Plaintiffs’ related evidence, on the other hand, was minimally
relevant or immaterial. Considering that § 3’s contribution limits are long-standing and the
interest put forward by the State is “neither novel nor implausible,” Shrink Missouri, 528 U.S. at
391, Defendants’ showing meets the applicable standard. I conclude Defendants have gone
beyond “mere conjecture,” McCutcheon, 572 U.S. at 210, and have established that preventing
quid pro quo corruption or its appearance is a sufficiently important interest underlying
Colorado’s § 3 individual contribution limits.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
103 of 142

104
Section 4 & the State’s Interest
Plaintiffs contend that Colorado undermines this interest by allowing § 3’s contribution
limits to be doubled under § 4’s voluntary spending limits scheme. In Plaintiffs’ words:
Colorado cannot claim either its contribution limits or Section 4 furthers valid
anticorruption interests when it willingly abandons its contribution limits in
exchange for reduced candidate spending. This tradeoff shows that Colorado’s real
interest is in reducing money in politics, which is unconstitutional.
Pls.’ Reply in Supp. of Request for J. at 8; see also Pls.’ Trial Summation & Request for J. at 8
(“‘Either [Colorado] is openly tolerating a significant’ risk of quid pro quo corruption in
exchange for candidates limiting their spending, or the contribution limits ‘are in no real sense’
designed to prevent corruption.” (quoting Ted Cruz for Senate, 596 U.S. at 313)). Although
Plaintiffs raise a valid consideration, their argument is neither compelling nor persuasive.
Because setting the amount of a contribution limit is an imprecise science, Buckley and
Randall make clear that there is a permissible range for contribution limits. Courts possess “no
scalpel to probe, whether, say, a $2,000 ceiling might not serve as well as $1,000.” Randall, 548
U.S. at 248 (quoting Buckley, 424 U.S. at 30). Thus, the fact that § 4’s voluntary spending limits
scheme permits the § 3 contribution limits to be doubled does not mean the amount of the § 3
limits does not fall within the permissible range.
Plaintiffs cite Fulton v. City of Philadelphia, 593 U.S. 522 (2021), as an example of the
Supreme Court holding that the asserted government interests were insufficient because the
policy in question undermined those interests. Pls.’ Trial Summation & Request for J. at 7-8;
Pls.’ Reply in Support of Request for J. at 7. In Fulton, the City of Philadelphia discovered that a
Catholic foster care agency would not certify same-sex couples to be foster parents and, based on
a nondiscrimination provision in its foster care contract, stopped referring children to the agency.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
104 of 142

105
593 U.S. at 526-28, 530-31. In considering the agency’s challenge under the First Amendment’s
Free Exercise Clause, the Supreme Court applied strict scrutiny, evaluating whether the City’s
policy was justified by an “interest[] of the highest order” and was “narrowly tailored to achieve
that interest.” Id. at 541. The City’s relevant asserted interests were “maximizing the number of
foster parents . . . and ensuring equal treatment of prospective foster parents and foster children.”
Id. The Court observed that including the Catholic foster care agency “seem[ed] likely to
increase, not reduce, the number of available foster parents” and then concluded that the City’s
“creation of a system of exceptions under the contract undermine[d] [its] contention that its non-
discrimination policies [could] brook no departures.” Id. at 542. The facts here are entirely
distinguishable, and the applicable standards and guidelines are distinct. Additionally, evaluating
whether the interest in Fulton was undermined was straightforward; the question was whether
the policy would increase or decrease the number of foster parents. In this case, the inquiry is
much more nuanced, as it involves the actions of a multitude of independent actors and
measuring the public’s perception of corruption.
The doubled contribution limits under § 4’s voluntary spending limits scheme are
accompanied by a spending cap. Without pointing to any evidence, Plaintiffs assert that the
doubled contribution limits make corruption more likely because each contribution makes up a
greater portion of the candidates’ expenditures. See Pls.’ Trial Summation & Request for J. at 8.
This reasoning from the First Circuit’s opinion in a case involving public financing counters
Plaintiffs’ supposition and is particularly relevant in competitive elections when the spending cap
is meaningful:
[O]nce it is clear that a publicly financed candidate’s campaign can reach the
overall fundraising limits, any single contributor to that campaign becomes less
important because the contributor can be “replaced” at no marginal cost. In other
words, the fact that the campaign seems bound to reach the fundraising ceiling
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
105 of 142

106
means that a given contributor is occupying a contribution slot that could as easily
be occupied by someone else. With this distinction in the importance of individual
contributors comes a corresponding diminution in the risk of corruption and,
therefore, a diminished justification for stringent contribution limits.
Vote Choice, Inc. v. DiStefano, 4 F.3d 26, 39 n.15 (1st Cir. 1993).
Plaintiffs offer only speculation that the doubled contribution limits, with the spending
cap, undermine the State’s interest. I cannot conclude that the State’s interest is undermined
based on mere speculation, especially given that courts “cannot determine with any degree of
exactitude the precise restriction necessary to carry out the statute’s legitimate objectives.”
Randall, 548 U.S. at 248. Ultimately, Defendants have established that the § 3 individual
contribution limits were adopted to guard against quid pro quo corruption or its appearance,
which is a sufficiently important interest to justify those limits.40
2. Closely Drawn: Danger Signs
Next, I consider whether the danger signs identified in Randall and Thompson are present
here. Those danger signs are: (1) when the individual contribution limits are lower than
comparable ones found elsewhere in the United States, (2) when the limits are substantially
lower than those previously upheld by the Court, and (3) when the limits do not adjust with
inflation. Randall, 548 U.S. at 250-53; Thompson, 589 U.S. at 6. The presence of these signs
indicates that the individual contribution limits may be too low. Impermissibly low limits “harm
the electoral process by preventing challengers from mounting effective campaigns against
incumbent officeholders, thereby reducing democratic accountability.” Randall, 548 U.S. at 249.
40 Because the only interest put forward by Defendants is preventing quid pro quo corruption or
its appearance, I do not consider whether there may be other sufficiently important governmental
purposes the Supreme Court has not previously rejected that would justify the contribution
limits.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
106 of 142

107
Defendants concede that the lower tier of campaign contribution limits—for state
legislative candidates—bears danger signs because the amount that those candidates can receive
is “lower than limits in other states and lower than limits previously upheld by the Supreme
Court.” Defs.’ Written Summation at 10.41 The parties disagree, however, on whether the higher
tier of the contribution limits—for statewide candidates—also exhibits danger signs. I find that it
does. Although an inflation adjustment mechanism applies to the individual contribution limit for
statewide candidates, Colorado’s limit is either among the lowest in the nation or is the lowest in
the nation, depending on the office. See Trial Ex. 1 at 2; Trial Tr. at 149:14-150:7. Of the thirty-
eight states that have individual contribution limits, only two states—Delaware and Montana—
have limits close to Colorado’s. See Trial Ex. 1 at 1-2. “The lowest campaign contribution limit
[the Supreme] Court has upheld remains the limit of $1,075 per two-year election cycle for
candidates for Missouri state auditor in 1998.” Thompson, 589 U.S. at 5 (citing Randall, 548
U.S. at 251). According to Thompson, that amount was equivalent to $1,600 in 2019. Id. Thus,
Colorado’s limit of $1,450 for candidates for statewide office is less than any amount the
Supreme Court has found to be constitutional. I conclude, therefore, that both levels of
Colorado’s individual campaign contribution limits bear sufficient danger signs to necessitate
further evaluation.
3. Closely Drawn: Randall Factors
Because danger signs are present for both tiers of contribution limits, I proceed with the
analysis of whether each limit is “closely drawn.” In doing so, I “review the record
41 Defendants do not concede that the third danger sign—unmooring the limits to inflation—is
present.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
107 of 142

108
independently and carefully” and “assess[] the proportionality of the restrictions,” applying the
factors set out in Randall. 548 U.S. at 249. I focus on whether § 3 is sufficiently “tailored” to
ensure that candidates, and specifically challengers, are able to engage in effective campaign
advocacy and that contributors are not unduly restricted in their ability to associate.42 Colorado
does not cite any special justification for its individual contribution limits, see Defs.’ Trial Br. at
8, so I weigh only the first four Randall factors.
Factor One: Challengers and Competitive Campaigns
The first concern is whether the individual contribution limits impede the ability of
candidates challenging incumbent officeholders to run effective and competitive campaigns by
“significantly restrict[ing] the amount of funding available.” Randall, 548 U.S. at 253.
Incumbent candidates undoubtedly have certain advantages, including name awareness, learned
insights, personal connections, party relations, the ability to undertake longer-term projects that
require extended time in office, and the potential consequences of the incumbent candidate’s
decisions while in office. See, e.g., Trial Tr. at 378:18-379:4 (Dr. Bonneau explaining that
incumbents have name recognition, a well-established funding base, and the ability to benefit
and communicate with voters in the district). These advantages can create an asymmetry that a
challenger must overcome, and if contribution limits are too low, the limits could significantly
restrict a challenger’s ability to fundraise and run an effective campaign. The “critical question,”
as identified by the Randall plurality, “concerns not simply the average effect of contribution
42 Because the particulars for how the balancing should be performed remain unstated, the
Randall test leaves room for discretion. For example, it is not apparent whether the outcome is
dependent on the direction in which a majority of factors tilt, nor has instruction been provided
on whether certain factors may carry more weight under specific circumstances.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
108 of 142

109
limits on fundraising, but more importantly, the ability of a candidate running against an
incumbent officeholder to mount an effective challenge.” Randall, 548 U.S. at 255; see also id.
at 255-56 (explaining that the focus is competitive elections in which spending is likely greater
and not simply average elections43).
No witness in this case, who was or had been a candidate for office in Colorado, provided
credible evidence that the contribution limits prevented him or her from amassing sufficient
funds to run an effective campaign. Mr. Lopez estimated that he would need to raise at least $3
million and maybe north of $4 million to run a competitive gubernatorial campaign. Trial Tr. at
94:15-21. In his two gubernatorial campaigns, he raised far less than that amount. Id. at 89:2-4
(testifying that he raised a little more than $60,000 for the 2018 race); id. at 76:20-23 (explaining
that he raised approximately $155,000 for the 2022 race). However, Mr. Lopez received no
contributions at the statutory limit for his 2018 race and, for his 2022 race, received only twenty-
six at that limit, a small fraction of the total contributions to his campaign. Id. at 91:17-92:18.
Nevertheless, during his 2022 race, Mr. Lopez was able to and did campaign in all sixty-four
counties in the State of Colorado. Id. at 57:22-24. From these facts, no inference can be made
that, to the extent Mr. Lopez’s campaign was ineffective, the result was due to Colorado’s
contribution limits. Mr. Lopez was unable to secure many contributions at the existing limits,
and there is no reason to believe his campaigns would have been more effective were the limits
raised.
Senator Pelton, who successfully ran three campaigns—two for the Colorado House and
one for the Colorado Senate—agreed he was able to raise sufficient funds, even though he was
43 The plurality in Randall flagged the possibility that incumbent legislators might set low
contribution limits to make it more difficult for challengers to unseat them. See 548 U.S. at 261.
That concern is not present here because Colorado voters adopted the limits, not the legislature.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
109 of 142

110
subject to the contribution limits. Id. at 126:9-129:22. Similarly, Ms. Donovan testified that she
was able to raise enough money to run an effective campaign in 2018. Id. at 799:25-800:3. And
Mr. Buescher, who outraised his opponents even when he lost, also indicated Colorado’s
individual contribution limits did not hinder his ability to get out his message to voters. Id. at
947:13-19, 949:12-13, 950:23-951:1, 954:14-19, 955:22-25. The testimony presented by these
fact witnesses does not directly address Randall’s specific consideration of whether challengers
can meaningfully contest incumbents, as none of the witnesses ever ran against an incumbent.
But Mr. Buescher was defeated twice despite being the incumbent, and that fact suggests that
challengers are able to mount effective campaigns despite the contribution limits. Id. at 950:25-
951:4, 954:16-19. Based on his experience, Mr. Buescher testified that the contribution limits
simply required him to find more donors, which in turn gave him more advocates. Id. at 956:1-6.
As further evidence that Colorado’s individual contribution limits do not inhibit
challengers from mounting competitive campaigns against incumbents, Defendants presented
expert testimony from Dr. Spencer, a law professor who focuses his research on voting rights
and campaign finance regulation. I detailed his methodology and opinions in the Findings of Fact
and summarize his relevant opinions here.
Based on his review of data from the state election return database from 1968 to 2020,
Dr. Spencer concluded that there was no statistically significant difference in incumbent
reelection rates in the Colorado legislature after Amendment 27 was adopted compared to the
period before the Amendment 27. Id. at 1085:10-1086:16, 1095:4-7. Using that same data, he
created two graphs showing Colorado’s incumbent reelection rate for state legislators has been
similar to that of all other states since Amendment 27 passed. Trial Ex. 325; Trial Tr. 1088:9-19,
1089:2-5. He also compared Colorado’s incumbent reelection rates for state legislators to those
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
110 of 142

111
of states with contribution limits and, separately, to those of states without contribution limits
from 2010-2016. See Trial Exs. 326, 327, 329, 330. Those comparisons yielded mixed results:
Colorado’s incumbent reelection rates for the state House were lower compared to states with
contribution limits and those without, but Colorado’s incumbent reelection rates for the state
Senate were lower compared to states with contribution limits and those without for three of the
four election cycles studied. Trial Tr. at 1091:5-1092:14, 1092:21-1094:25, 1095:8-13; Trial Exs.
326, 327, 329, 330. Dr. Spencer explained that those differences were relatively minor, and there
was “really no effect on the incumbency re-election rate, generally.” Trial Tr. at 1094:21-25. Dr.
Spencer also conducted regression analyses to investigate how individual contribution limits
impact incumbent reelection rates for state legislative offices. Id. at 1097:1-6. He ran models
controlling for multiple variables, including the state’s use of multi-member districts, term limits,
salary, asset disclosure rules, and state population. Trial Ex. 331. His analyses “provided extra
evidence that there is no relationship between the individual contribution limit and the incumbent
reelection rate.” Trial Tr. at 1108:4-13. Lastly, Dr. Spencer performed a literature review and
found two articles that addressed “the relationship of an individual contribution limit and the
likelihood that a challenger or incumbent will perform better.” Id. at 1108:17-1109:9. In one
study, the author determined that a contribution limit of $500 compared to $2,000 increased the
chance of incumbents losing by 10% and, in the other, the authors reached a similar conclusion.
Id. Based on his research, Dr. Spencer opined that abolishing Colorado’s contribution limits
would not result in challengers beating incumbents more frequently. Id. at 1109:10-15, 1109:24-
25.
Plaintiffs criticize Dr. Spencer’s opinions because he used incumbent reelection rates to
assess whether challengers can mount effective campaigns against incumbents. They emphasize
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
111 of 142

112
that Randall “never mentions incumbent reelection rates as a means of testing whether
challengers have the funds necessary to run competitive campaigns or testing the
constitutionality of contribution limits.” Pls.’ Reply in Support of Request for J. at 15. While the
words “incumbent reelection rates” do not appear in the plurality opinion, their absence does not
support the conclusion that “incumbent reelection rates . . . are irrelevant under Randall.” Id. at
16. Under the standard definition of “incumbent,” any official who is seeking to remain in office
and is running against at least one opponent is an incumbent facing a challenger. Therefore,
when incumbents are reelected at higher rates, it necessarily means that challengers are being
voted into office at lower rates. This makes incumbent reelection rates a relevant and useful
metric.44
To rebut Dr. Spencer’s opinions, Plaintiffs presented the testimony of Dr. Cann, Dr.
Bonneau, and Mr. Engen. The opinions of these experts are either irrelevant or unreliable
because they do not refute Defendants’ evidence that challengers are able to amass sufficient
funds to run effective campaigns despite the contribution limits. An overarching problem with
these expert opinions is that Plaintiffs misunderstand Randall’s inquiry regarding what
constitutional contribution limits must not do. Randall does not stand for the proposition that the
44 Dr. Cann testified that incumbent success rates are “one measure of electoral competitiveness,
but not necessarily the best measure.” Trial Tr. at 384:14-18. He explained that, by collapsing the
inquiry down to wins and losses, the analysis does not provide information on how close the
elections were. Id. at 384:19-385:15. This clarification regarding Dr. Spencer’s opinions is valid.
However, it does not make those opinions any less relevant or reliable in this case, where the
critical inquiry is whether challengers are able to mount effective campaigns, see Randall, 548
U.S. at 253, 255. Effective may not always mean successful, but an incumbent loss means a
challenger’s campaign was successful and thus effective. Moreover, Dr. Spencer’s focus on
incumbent reelection rates would tend to undercount competitive races, as there are certainly
competitive races in which the challenger loses and the incumbent is elected. See also Trial Tr. at
498:14-21 (Dr. Cann discussing how incumbents usually only run if they are likely to win
reelection, so incumbent reelection rates are “known to be biased upwards in [an] important
way”).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
112 of 142

113
First Amendment must maximize competitiveness wherever and whenever possible. Rather, the
inquiry in Randall is whether the contribution limits meaningfully impair the ability of
challengers to oust incumbents at the ballot box. See 540 U.S. at 255. The plurality in Randall
commented that “information about average races, rather than competitive races, is only distantly
related to that question, because competitive races are likely to be far more expensive than the
average race.” It did not instruct, however, that the First Amendment requires anticorruption
reforms to maximize electoral competitiveness. In fact, “[t]he First Amendment prohibits such []
attempts to ‘fine-tun[e]’ the electoral process, no matter how well intentioned.” McCutcheon,
572 U.S. at 207 (quoting Bennett, 564 U.S. at 747).
Dr. Cann opined that Colorado’s contribution limits “negatively affect the overall
competitiveness of elections in Colorado.” Trial Tr. at 444:22-445:1. I detailed his opinions and
methodology in my Findings of Fact. Here, I summarize his testimony again and explain why he
did not successfully rebut Dr. Spencer.
Dr. Cann’s first opinion related to primary contestation rates. He found that the states in a
category he defined as “very strict”—those with $1,000 or less individual contribution limits—
averaged about a 14% primary contestation rate, while the states in the category he defined as
“not strict”—those with higher contribution limits—averaged between a 23% or 24% primary
contestation rate. Id. at 460:2-7; Trial Ex. 129. Thus, he opined that the contribution limits in the
“very strict” states caused the reduced primary contestation rate. Trial Tr. at 460:10-13. Dr. Cann
admitted that Colorado’s term limits for legislative offices create open seats and challengers are
more attracted to those seats than to races in which an incumbent is running for reelection. Id. at
514:5-14. Yet, he omitted from his analysis data on open seat races, as well as data for state
Senate races. Id. at 511:5-8, 513:1-3. Nor did he control for any variables or perform any stress
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
113 of 142

114
testing to determine whether other contribution limits would change his conclusions. Id. at
518:14-22. Dr. Spencer reran Dr. Cann’s analysis, including the omitted data on open seat and
state Senate races, and determined that, in 2022, 53% of Colorado’s state legislature primary
races included at least one challenger. Trial Ex. 340. In light of Dr. Cann’s insufficient
justification for the parameters he selected for his analysis and failure to control for any
variables, his opinion that contribution limits cause lower primary contestation rates is unreliable
and of limited value.
Dr. Cann provided three other opinions on electoral competitiveness in Colorado. He
found that states with stricter campaign limit regimes, like Colorado, resulted in 3% fewer
contested general elections compared to those states with less strict regimes. Trial Tr. at 470:4-
14; Trial Ex. 130. Dr. Cann also determined that strict contribution limits have a “substantial
effect” on the amount of money candidates in a lower state chamber raise, decreasing the amount
of funds state house candidates are able to raise by approximately $40,000 for challengers and
$70,000 for incumbents. Trial Tr. at 477:24-478:20, 494:19-21; Trial Ex. 131. He opined that
campaign spending provides greater marginal benefits to challengers than it does incumbents,
and thus, even if incumbent candidates are able to raise more money under less strict
contribution limits, challengers are advantaged because their spending is more effective. Trial Tr.
at 489:25-490:1, 493:15-19, 495:19-496:1. Dr. Cann concluded that, if Colorado had no
contribution limits for state House elections, an average challenger would raise $40,000 more per
election cycle, which would lead to an average decrease of 3.2% of the votes for the incumbent.
Id. at 495:4-6, 12-15.
There are two main problems with Dr. Cann’s opinions. First, as explained in my
Findings of Fact, he used a single-index contribution limit variable that considered contribution
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
114 of 142

115
limits that are prohibited in Colorado, including corporate contribution limits and union limits, as
well as political action committee contribution limits. Id. at 523:14-525:18. These types of limits
are not subject to challenge and should not have been included in his analysis. The inclusion of
these other variables obscures the effect the individual contribution limits, the only variable in
question here, has on the competitiveness of elections.
Second, because Dr. Cann focuses on average rather than competitive races, his opinions
shed little light on what the impact of Colorado’s individual contribution limit would be in
competitive races. His findings do not show that Colorado’s individual contribution limits
prevent challengers from mounting effective campaigns. At best, his findings support the
inference that Colorado challengers may be able to mount more effective campaigns without
contribution limits, but the First Amendment does not demand that contribution limits maximize
electoral competitiveness. Considering these flaws in Dr. Cann’s analysis, his opinions provide
little assistance in determining whether Colorado’s contribution limits prevent challengers from
running effective campaigns, and they do not offset the persuasive opinions of Dr. Spencer.
Plaintiffs also offered the testimony of Dr. Bonneau, who opined that Colorado elections
are not particularly competitive for non-incumbent candidates, likely because contribution limits
result in lower campaign expenditures. Id. at 377:12-15, 380:6-11. Dr. Bonneau also testified that
a challenger benefits more from campaign spending than an incumbent and thus that incumbents
benefit from contribution limits. Id. at 383:9-13, 383:22-23. However, he did not study whether
there was a direct link between contribution limits and the likelihood of a challenger defeating an
incumbent. Id. at 389:9-12. He also admitted in his testimony that existing research finds that
even states without contribution limits tend to favor incumbents over challengers. Id. at 401:16-
402:10, 403:13-17. These considerations diminish the value of his opinions, which are
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
115 of 142

116
outweighed by Dr. Spencer’s more targeted and reliable testimony.
Mr. Engen’s opinion is even more unavailing. He testified that, immediately after 2002
when § 3’s contribution limits were adopted, there was a marked decrease in the number of first-
time Colorado state senate candidates winning their general election races against incumbent
candidates. Id. at 604:15-19, 605:8-17. As I explained in my Findings of Fact, the principal
defect in Mr. Engen’s analysis is obvious: He defines “incumbent” to mean a candidate “who has
previously held a Colorado[-]level office, regardless of the” office or district represented. Id. at
608:24-609:1.45 Thus, I do not credit his opinion that Colorado’s individual contribution limits
caused a marked decrease in the number of first-time candidates for state office winning their
general election races.
No credible evidence in the record substantiates that § 3’s individual contribution limits
prevent challengers from amassing sufficient funds to run effective campaigns. Incumbent
reelection rates for Colorado’s state legislative offices remained the same after Amendment 27
was adopted. Id. at 1085:8-1086:16, 1095:4-7. And, as Dr. Spencer reliably determined, “there is
no relationship between the individual contribution limit and the incumbent reelection rate.” Id.
at 1095:8-13, 1108:12-13. The considerable evidence presented on this point shows that
45 Again, the traditional definition of incumbent is the candidate holding the office at the time of
the campaign. See, e.g., Trial Tr. at 426:8-11 Apart from the fact that this is certainly not the
definition that the Randall court had in mind, it counts candidates who have served in one state
office but are running for an entirely different office and excludes people who served in public
office but were not elected to their office. It is true that the only benefits that Randall mentions in
relation to being an incumbent are name-recognition and certain “reputation-related or media-
related advantages.” 548 U.S. at 248 (quoting Shrink Mo., 528 U.S. at 403-04 (Breyer, J., joined
by Ginsburg, J., concurring)). But I do not read Randall to limit the benefits of incumbency to
these attributes alone. As I identified above, name awareness, learned insights, personal
connections, party relations, potential to achieve longer-term projects requiring more time in
office, and real world consequences of the incumbent candidate’s decisions in that office are all
potential benefits that distinguish an incumbent from a challenger. Many of these are not
captured by Mr. Engen’s definition.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
116 of 142

117
Colorado’s individual contribution limits have not decreased the ability of challengers to mount
effective campaigns against incumbents. Consequently, I conclude that this Randall factor
weighs in favor of determining that § 3’s individual contribution limits are closely drawn to meet
their objective.
Factor Two: Limits on Contributions from Political Parties
The second Randall factor is the extent to which contributions from political parties to
candidates are limited. 548 U.S. at 256. The Randall plurality was concerned by the fact that
Vermont’s law imposed the same low limit on campaign contributions from individuals as those
from political parties. Id. at 258-59. “[B]y placing identical limits upon contributions to
candidates, whether made by an individual or by a political party,” Vermont’s law undermined
“the need to allow individuals to participate in the political process by contributing to political
parties that help elect candidates.” Id.46 The Court sought to ensure that the combined effects of
46 Plaintiffs frame this concern differently. They assert: “Randall held party contributions were
relevant because they ‘represent[ed] a significant amount of total candidate funding’ in
‘competitive races.’” Pls.’ Trial Summation & Request for J. at 22 (quoting Randall, 548 U.S. at
254). By no means was that a holding in Randall. The full sentence from which Plaintiffs quote
is: “For another thing, the petitioners’ expert witnesses produced evidence and analysis showing
that Vermont political parties (particularly the Republican Party) ‘target’ their contributions to
candidates in competitive races, that those contributions represent a significant amount of total
candidate funding in such races, and that the contribution limits will cut the parties’ contributions
to competitive races dramatically.” Randall, 548 U.S. at 254. The Court was simply describing
the petitioners’ evidence of the circumstances in Vermont. In their Proposed Findings of Fact and
Conclusions of Law, Plaintiffs again misinterpret the full significance of political parties to the
Randall analysis, stating: “Randall held political parties should be permitted to donate more
funding to their candidates than individual contributors to help the candidate run competitive
campaigns.” Pls.’ Proposed FOFCOL at 22 (citing Randall, 548 U.S. at 256-59)). The Court’s
consideration of this factor did not center on the impacts of political party contribution limits on
candidates; it was focused on how the limits might “severely inhibit collective political activity
by preventing a political party from using contributions by small donors to provide meaningful
assistance to any individual candidate.” Randall, 548 U.S. at 258 (using an example to illustrate
how low limits on party contributions would “thwart the aims of [thousands of] donors from
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
117 of 142

118
the party and individual contribution limits did not infringe the right to associate or “reduce the
voice of political parties . . . to a whisper.” Id. at 256, 259 (citation modified).
Colorado’s limits for contributions from political parties are vastly higher than those for
individual contributors. The contribution limits for political parties range from $20,500 for
candidates for the Colorado House of Representatives to $789,060 for candidates for governor
and are among the highest in the country. See Colo. Const. Art. XXVIII § 3(3)(d); 8 Colo. Code
Regs. § 1505-6, Rule 10.17.1(i); Trial Ex. 351 at 20, Trial Ex. 375 at 1. To say this is dissimilar
to Vermont’s scheme at issue in Randall is an understatement.
Plaintiffs claim that the higher political party contribution limits in Colorado “just
slightly help[]” Defendants’ case because practical limitations and political realities have made
Colorado’s political parties “nothing more than ‘a whisper.’” Pls.’ Trial Summation & Request
for J. at 21 (quoting Randall, 548 U.S. at 259). Regarding practical limitations, Plaintiffs stress
that “state parties do not have the financial resources to donate the maximum amount permitted
under Colorado law to all their candidates.” Id. at 22 (citing Trial Tr. at 164, 645-56).
Additionally, Plaintiffs observe that, even when parties do have the financial resources, they still
may actively oppose some of their candidates. Id. And parties often “avoid contributing to
candidates in races [the parties] deem uncompetitive.” Id. (citing Trial Tr. at 168-69, 184, 640-
43). That candidates cannot necessarily rely on party support is a complaint far afield from
concerns of constitutionality. Of course, “party contributions only benefit some candidates in
making a meaningful contribution to state politics by giving a small amount of money to the
party they support”); see also id. at 256-57 (describing the principle from Buckley, stating:
“contribution limits constitute ‘only a marginal restriction’ on First Amendment rights because
contributor[s] remain[] free to associate politically, e.g., in a political party, and ‘assist
personally’ in the party’s ‘efforts on behalf of candidates’” (quoting Buckley, 424 U.S. at 20-
22)).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
118 of 142

119
some circumstances.” Id. Party resources are inevitably limited, and parties have their own
decision-making authority. The Randall plurality viewed party contributions not only as a means
to ensure that elections are competitive but more broadly as furthering the right to associate. See
548 U.S. at 256 (discerning that Vermont’s low contribution limits for political parties
“threaten[] harm to a particularly important political right, the right to associate in a political
party”); id. at 258-59 (recognizing “the need to allow individuals to participate in the political
process by contributing to political parties that help elect candidates”). Colorado’s political
parties frequently provide financial support to candidates’ campaigns but do so using their own
discretion. See Trial Tr. at 183:12-185:25, 649:6-652:13, 810:6-812:20. They also engage in
general voter turnout initiatives. See, e.g., id. at 640:21-642:11. That § 3’s contribution limits do
not prevent parties from providing support to their chosen candidates or engaging in other
prodemocratic conduct relieves the concern that the limits might thwart individuals’ efforts to
associate in a political party. Randall, 548 U.S. at 258.
Another reason Plaintiffs assert this factor should be given little weight is the reality that
political parties’ relevance as campaign contributors is limited and diminishing. Plaintiffs
complain that “even if Colorado party contributions to candidates were unlimited, the practical
restraints on a political party’s ability and willingness to contribute to candidates would still
prevent them from taking advantage of this unlimited opportunity, which demonstrates the
reduced significance of party contributions.” Pls.’ Reply in Support of Request for J. at 23.47
Similarly, Plaintiffs emphasize that spending from independent expenditure committees
47 Plaintiffs also note that “both major political parties are traditionally not involved in primary
elections, making party contributions significantly less important in one-party districts that do
not have competitive general elections.” Pls.’ Trial Summation & Request for J. at 21 (citing
Trial Tr. at 170-71, 637-38). But again, this is a party choice, not a governmental restriction. It
provides no insight into whether the § 3 contribution limits are closely drawn.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
119 of 142

120
overshadows that of political parties. Pls.’ Trial Summation & Request for J. at 23. Plaintiffs’
lamentations on the diminution of political parties’ relevance are misplaced. Whether some
external factor, such as fundraising difficulties or independent expenditures, has diminished the
impact of political party spending is not a focus of the Randall analysis.
The § 3 contribution limits do not undermine “the need to allow individuals to participate
in the political process by contributing to political parties that help elect candidates,” see
Randall, 548 U.S. at 258-59, and if the voice of political parties in Colorado is reduced “to a
whisper,” it is not because of the contribution limits. See id. at 259. The political party limits
show Colorado has given due consideration to the need to allow individuals to engage in
collective political activity. Thus, the answer to the question considered in Randall is that the
State’s high limits on contributions from political parties support that § 3 is properly tailored.
Factor Three: Volunteer Services and Expenses
In addition to financial contributions, individuals may support a candidate by providing
volunteer services. The Vermont statute at issue in Randall explicitly excepted from the
definition of “contribution” all “services provided without compensation by individuals
volunteering their time on behalf of a candidate.” 548 U.S. at 259 (citation omitted). However,
the statute did “not exclude the expenses those volunteers incur, such as travel expenses, in the
course of campaign activities.”48 Id. The Court expressed particular concern that when
48 The Randall plurality provided a number of examples of volunteer expenses that could count
against an individual’s contribution limit such that the ability of campaigns to effectively use
volunteers would be impeded and it would be more difficult for individuals to associate with
candidates as volunteers. 548 U.S. at 260. The opinion discussed the following volunteer
expenses: mileage, the use of a volunteer’s house for a few dozen neighbors to meet the
candidate, the provision of coffee and doughnuts to the attendees of such an event, postage
supplied by volunteers, and pencils and pads used by volunteers. Id.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
120 of 142

121
contribution limits are low, such a broad definition of contribution could “impede a campaign’s
ability effectively to use volunteers” and could significantly restrict the ability of and means in
which supporters may associate with a candidate. Id. at 260. I conclude that, in practice and
theory, that concern is not substantial here.
Section 2(5)(a) of Article XXVIII of the Colorado Constitution defines “contribution” for
the purposes of § 3’s contribution limits. Under that definition, a contribution includes:
(I) The payment, loan, pledge, gift, or advance of money, or guarantee of a loan,
made to any candidate committee, issue committee, political committee, small
donor committee, or political party;
(II) Any payment made to a third party for the benefit of any candidate
committee, issue committee, political committee, small donor committee, or
political party;
(III) The fair market value of any gift or loan of property made to any candidate
committee, issue committee, political committee, small donor committee or
political party;
(IV) Anything of value given, directly or indirectly, to a candidate for the purpose
of promoting the candidate’s nomination, retention, recall, or election.
Colo. Const. art. XXVIII, § 2(5)(a). Article XXVIII, like the statute in Randall, excepts from the
definition of contribution “services provided without compensation by individuals volunteering
their time on behalf of a candidate.” Colo. Const. art. XXVIII, § 2(5)(b).49 Plaintiffs point out,
however, that the definition in Article XXVIII does not specifically exclude the expenses
volunteers incur in the course of campaign activities. Pls.’ Trial Summation & Request for J. at
24. Plaintiffs claim that, without such an exclusion, the definition here applies like the one in
Randall and threatens the same harm.
49 The relevant Colorado regulations do not specifically address volunteer expenses, but they
elaborate on the definition of volunteer services, stating “[i]f volunteer services yield a thing of
value, ‘contribution’ only includes the reasonable value of the materials involved, unless the
value is de minimis.” 8 Colo. Code Regs. § 1505-6, Rule 1.5.2(c).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
121 of 142

122
Initially, I conclude and underscore that the testimony at trial by current and former
candidates for Colorado state office established that Article XXVIII’s definition of contribution
does not “impede a campaign’s ability effectively to use volunteers” nor restrict the ability of
and means in which supporters may associate with a candidate. See Randall, 548 U.S. at 260.
Defendants presented testimony that volunteer expenses that “produce a thing of value” are
considered contributions but mileage, on the other hand, is not. Trial Tr. at 885:14-18, 889:21-
23; see Trial Ex. 359. Mr. Bouey, the Campaign Finance and Lobbyist Program Manager at the
Colorado Department of State, testified that, over the span of fifteen years, no one in his office
had received any inquiries concerning volunteer-related expenses and their potential incursion
into contributions. Trial Tr. at 888:4-13. Nor was he aware of any campaign finance complaints
claiming that volunteer gas mileage must be reported as a contribution. Id. at 888:17-20.
Although there might be instances of reporting volunteer mileage, see, e.g., Trial Ex. 50, this is
considered a choice to overreport and is not flagged in the reporting system. Trial Tr. at 890:14-
16. Mr. Buescher never had a volunteer be unable to contribute money to his campaign because
the volunteer had “maxed out the individual contribution limit by driving around the state and
volunteering” on his behalf. Id. at 960:15-19. Similarly, Ms. Donovan testified that she never
experienced a volunteer being unable to financially contribute to her campaign because that
person had maxed out their contributions by driving around the state. Id. at 813:7-11. This type
of evidence was not discussed in Randall. The plurality there determined, based on the text of the
Vermont statutes alone, that “the likelihood of unjustified interference” was sufficiently great to
find a lack of tailoring. 548 U.S. at 260. Here, however, there is strong evidence in the record
that the likelihood of unjustified interference is slim, if not nonexistent.
Plaintiffs attempt to sow confusion where there is none. They press me to find that
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
122 of 142

123
“Colorado ‘does not exclude the expenses [] volunteers incur, such as travel expenses, in the
course of campaign activities.’” Pls.’ Proposed FOFCOL at 9 (quoting Randall, 548 U.S. at 259).
Effectively, such a finding would require me to provide advisory opinions on the application of
Colorado constitutional provisions to hypothetical scenarios that are not at issue in this case and
for which proceedings by the Department of State or Colorado courts are absent from the
record.50 While I decline to engage in that exercise, I review the relevant law and determine that
Defendants’ positions regarding the application of the definition of contribution to volunteer
expenses are reasonable.
The types of volunteer expenses mentioned in Randall would implicate only the last three
provisions of Article XXVIII’s definition of contribution—“[a]ny payment made to a third party
for the benefit of any candidate committee,” “[t]he fair market value of any gift or loan of
property made to any candidate committee,” and “[a]nything of value given, directly or
indirectly, to a candidate for the purpose of promoting the candidate’s nomination, retention,
recall, or election.” Colo. Const. art. XXVIII, § 2(5)(a)(II), (III), & (IV). A Colorado Court of
Appeals opinion from 2015, Keim v. Douglas County School District, 399 P.3d 722 (Colo. App.
2015), sheds some light on these provisions.
In Keim, a school district paid a third-party entity to create a report that favorably
evaluated a reform agenda in the district and referenced an upcoming school board election
50 At trial, Plaintiffs sought to obtain similar opinions from Stephen Bouey, the manager of the
campaign finance and lobbyist disclosure programs at the Colorado Department of State. See,
e.g., Trial Tr. at 911:20-23 (Plaintiffs’ counsel cross-examining Mr. Bouey: “I know, from
reading past depositions and the preliminary injunction hearing, you hate hypotheticals. But
we’re lawyers, and we love them. So we’re going to give a few a try.”). Mr. Bouey is not an
attorney, and it is not his job to provide legal advice on these questions. Id. at 838:13-14, 913:23-
24, 914:14-16, 925:1-4. If a member of the public wanted guidance on a specific matter, he or
she could request an advisory opinion from the Department of State. Id. at 886:18-23, 913:25-
914:3, 923:1-2, 925:5-7.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
123 of 142

124
during which a “reform board” could be installed. 399 P.3d at 724-25. The district then sent that
report to approximately 85,000 residents of the county where the district was located. Id. at 725.
The case came before the court on the appeal of an administrative law judge’s decision finding
that the district had made a contribution in violation of Colorado’s Fair Campaign Practice Act,
Colo. Rev. Stat. § 1-45-117(1)(a)(I). Id. at 724-25. The definition of contribution for that law has
the same meaning as provided in Article XXVIII § 2(5)(a). See id. at 727. The court first
considered whether the provision of that definition in § 2(5)(a)(II) would apply to the district’s
conduct. Id. at 727. Offering little analysis, it determined that “there was no evidence in the
record that the District made a payment to a third party for the benefit of any candidate
committee, issue committee, political committee, small donor committee, or political party,” so
that provision was inapplicable. Id. The court then conducted an in-depth review of the proper
interpretation of the provision of the definition of contribution in § 2(5)(a)(IV). See id. at 728-30.
The court concluded that “the phrase ‘given, directly or indirectly, to a candidate for the purpose
of promoting the candidate’s . . . election’ requires that (1) a thing of value (2) be put into the
possession of or provided to a candidate or someone acting on the candidate's behalf (3) with the
intention that the candidate receive or make use of the thing of value provided (4) in order to
promote the candidate’s election.” Id. at 729. Ultimately, the court held that “the evidence does
not support the ALJ’s conclusion that the Hess Report was given, directly or indirectly, to a
candidate for the purpose of promoting that candidate’s election.” Id. In reaching that holding,
the court acknowledged that one could read the commissioned report “as favorable to the reform
agenda and therefore conclude that the [report’s] mass distribution might have provided an
incidental benefit to a pro-reform candidate.” Id. The court explained: “Concluding that such an
incidental benefit was sufficient to establish a ‘contribution,’ however, would stretch the
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
124 of 142

125
meaning of ‘given, directly or indirectly, to a candidate for the purpose of promoting the
candidate’s . . . election’ too far.” Id.
Plaintiffs correctly point out that Keim does not address volunteer expenses. Pls.’ Trial
Summation & Request for J. at 24-25; Pls.’ Reply in Support of Request for J. at 25. The case
nevertheless provides guidance on the interpretation of the relevant definition of contribution.
Regarding payments to a third party for the benefit of any candidate committee, Keim’s holding
makes clear that § 2(5)(a)(IV) requires there to be a concrete benefit to a specific candidate. See
399 P.3d at 727. Paying a third party to produce something that happens to benefit a candidate is
insufficient. A benefit must be more than incidental. Id. at 729. Additionally, to count as a
contribution under § 2(5)(a)(IV), the thing of value must be “put into the possession of or
provided to a candidate or someone acting on the candidate’s behalf.” Id.51 The intention must be
“that the candidate will eventually receive or make use of that thing of value.” Id. Based on these
principles, Defendants’ position that a volunteer’s vehicle mileage is not covered by Article
XXVIII’s definition of contribution is reasonable.52 Likewise, it is reasonable that Defendants
51 Plaintiffs claim the requirement from Keim that a contribution be put into a candidate’s
possession, if imposed to all the provisions in the definition of contribution, would exclude any
intangible gift, such as advertising, as counting towards contributions. Pls.’ Reply in Support of
Request for J. at 25. As I read the record and Defendants’ arguments, they did not represent that
the requirement that a contribution be put into a candidate’s possession is universal. For
example, Mr. Bouey testified that, if asked about a particular volunteer expense, he would point
the inquirer to the applicable rules discussing “the creation of tangible things” and would inform
the person that “there is potentially a component that that item of tangible value needs to be
placed directly into the possession of a candidate for it to be considered a contribution.” Trial Tr.
at 911:24-912:17 (emphasis added); see also id. at 914:7-9 (“Based upon my understanding of
Keim and looking at the totality of information about what constitutes a contribution.” (emphasis
added)).
52 Plaintiffs also note that Keim does not discuss the “fair market value” of property given or
“loan[ed]” to a campaign. Pls.’ Reply in Support of Request for J. at 24 (quoting Colo. Const.
Art. XXVIII § 2(5)(a)(III)). Plaintiffs suggest that this provision of the definition of contribution
could cover a volunteer who “‘loans’ her car to a campaign.” Id. If the volunteer temporarily
provides the car to a campaign for it to use the vehicle as it chooses, perhaps that provision
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
125 of 142

126
view the endless range of potential volunteer expenses with nuance.53
The Randall plurality feared that “a gubernatorial campaign volunteer who makes four or
five round trips driving across the State performing volunteer activities coordinated with the
campaign can find that he or she is near, or has surpassed, the contribution limit.” 548 U.S. at
260. Defendants have submitted credible evidence that this phenomenon is not a problem in
Colorado. While other volunteer expenses may be counted toward an individual’s contribution
limit, the evidence in the record supports that candidates’ use of volunteers is not frustrated nor
is the ability of supports to associate with candidates dampened. If the treatment of volunteer
expenses for § 3’s individual contribution limits produces any of the concerns highlighted by the
Court in Randall, it is to a much lesser degree. Thus, this factor does not show a lack of tailoring
in Article XXVIII’s definition of contribution.
Factor Four: Inflation Indexing
The final Randall factor relevant here is whether the individual contribution limits
automatically adjust for inflation. 548 U.S. at 261. A shared defect between the Alaskan
campaign finance framework at issue in Thompson and Vermont’s Act 64 in Randall was their
lack of any mechanism to adjust contribution limits in response to economic inflation. Id.;
would apply, see, e.g., Trial Tr. at 914:17-915:9, but it would be a stretch to say that an
individual who continues to use his or her own vehicle while volunteering for a campaign would
be loaning the car to the campaign.
53 Plaintiffs insist the definition of contribution is ambiguous and that the lack of clarity “chills
the exercise of First Amendment rights” and demonstrates a lack of tailoring. Pls.’ Trial
Summation & Request for J. at 25-26 (quoting Wyo. Gun Owners v. Gray, 83 F.4th 1224, 1248
(10th Cir. 2023)). Nuance is not equivalent to ambiguity. As Defendants observe, Plaintiffs did
not bring a vagueness challenge. Defs.’ Written Summation at 37. Moreover, the application of a
multi-provision definition to countless individual circumstances inevitably will be fact-specific
and nuanced. Plaintiffs’ claim that speech is chilled is unsupported by evidence and ignores that
individuals may request advisory opinions from the Colorado Department of State.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
126 of 142

127
Thompson, 589 U.S. at 6. The Court in Randall was concerned not only that a failure to have
limits adjust with inflation would mean “limits which are already suspiciously low will almost
inevitably become too low over time,” but also that the burden to remedy that situation would be
placed on incumbent legislators who might have little incentive to adjust the limits “to ensure the
adequate financing of electoral challenges.” 548 U.S. at 261. The Court reiterated these concerns
in Thompson. See 589 U.S. at 6.
Plaintiffs acknowledge that § 3 contains a system for indexing the contribution limits to
inflation but argue that the mechanism is methodologically flawed. As set out in my Findings of
Fact, the relevant provision of § 3 states: “Each limit on contributions described . . . shall be
adjusted by an amount based upon the percentage change over a four[-]year period . . . rounded
to the nearest lowest twenty-five dollars.” Colo. Const. art. XXVIII, § 3(13). Plaintiffs contend
this indexing mechanism necessarily leads to an impermissible undercounting of inflation such
that it does not satisfactorily address the concerns articulated in Randall. Plaintiffs highlight that
“[s]ince 2002, cumulative inflation has increased 64.7% . . . but the limits for statewide and
legislative office have only increased 45% and 12.5%, respectively.” Pls.’ Trial Summation &
Request for J. at 26 (citing Trial Tr. at 579-81, 580-83, 688; Trial Ex. 40). According to
Plaintiffs, “current statewide and legislative limits would be 14% and 47% higher, respectively,
if they were adjusted for inflation every four years without rounding down.” Id. Based on this
reasoning, Plaintiffs assert that Colorado’s individual contribution limits have experienced a
“decline in real value” and cannot keep pace with the increasing costs of campaigning. Id. at 26-
28 (quoting Randall, 548 U.S. at 261).
To be sure, the inflationary adjustments made thus far have undermeasured the
depreciating value of the U.S. dollar, see Trial Tr. at 874:8-10 (Mr. Bouey agreeing that
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
127 of 142

128
Colorado’s contribution limits have not perfectly matched inflation). And, as Plaintiffs fairly
point out, the contribution limits have not kept up with rising campaign costs, see Pls.’ Trial
Summation & Request for J. at 27-28; Trial Tr. at 606:2-8 (Mr. Engen testifying that increased
costs have made it more difficult for candidates to reach all the voters in their districts). This
divergence produced by the terms of the indexing mechanism and the increased costs of
campaigning is cause for genuine concern.
Nevertheless, § 3 contains an inflation indexing mechanism that fundamentally
distinguishes it from the contribution limits at issue in Randall and Thompson. Both tiers of
Colorado’s individual contribution limits have increased pursuant to that mechanism during the
pendency of this case. See, e.g., Trial Ex. 15 at 7, Trial Ex. 16 at 7. Additionally, I have
concluded that candidates, and specifically candidates challenging incumbents, have been able to
campaign effectively, even with the relatively low individual contribution limits. This suggests
that any defect in the indexing mechanism has not impacted the ability of challengers to run
effective campaigns. And, as Defendants point out, “other contextual changes . . . may enhance
candidates’ ability to amass resources in elections over time, such as growth in the size of the
electorate.” Defs.’ Written Summation at 38 (citing Trial Exs. 371, 372); Trial Tr. at 878:19-22,
879:4-9 (Mr. Bouey explaining that the pool of potential donors has increased while the total
number of state House and Senate districts has remained the same).
The other concern for the plurality in Randall was that, without an inflation mechanism,
the burden to remedy a contribution limit that was declining in real value would be placed on
incumbent legislators. See 548 U.S. at 261. Because Colorado’s contribution limits were adopted
by constitutional amendment, that risk is not present here. Plaintiffs contend that “requiring the
general public to stay abreast of the costs of campaigning and subsequently amend Colorado’s
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
128 of 142

129
constitution every few years is unthinkable.” Pls.’ Trial Summation & Request for J. at 28.
However, this assumption is belied by the swift actions of the people of Colorado in establishing
Colorado’s contribution limits. In 1996, after the Colorado legislature increased individual
campaign contribution limits, the Colorado voters promptly responded by passing Amendment
15, which revised the applicable statutes and reduced the contribution limits. Trial Ex. 353 at 2-
3, Trial Exs. 356, 374. When the legislature again raised the limits in 2000, the Colorado
electorate passed Amendment 27 within two years, lowering the limits once more. Trial Ex. 354
at 1, Trial Ex. 374. This history shows that the people of Colorado can be expected to spring into
action to protect their desired contribution limits. Coloradans may act preemptively or may
choose to monitor whether contribution limits begin to fall behind inflation to such an extent that
they prevent challengers from mounting competitive campaigns against incumbents. Either way,
it is not “unthinkable” to trust the public with this responsibility.
If I were in the business of policymaking, the indexing mechanism in § 3 might not be the
one I would use. And I recognize that it is not the mechanism that Colorado legislators have
selected in other contexts. See Pls.’ Trial Summation & Request for J. at 26-27. It is, however,
the mechanism that Colorado voters chose in this context, perhaps because there has been a
historical disconnect between the desires of Coloradans and the preferences of their elected
representatives. At present, candidates for state office in Colorado are not prevented from
amassing sufficient resources to campaign effectively. Given the facts in this case, I give
deference to the people of Colorado to ensure that § 3’s individual contribution limits remain
within the constitutionally acceptable range. Consequently, I conclude that the indexing
mechanism, despite its sluggish and imperfect reflection of inflation, is not so flawed as to lend
support for the conclusion that § 3’s individual contribution limits are not closely drawn.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
129 of 142

130
* * *
In summary, the overall effect of both tiers of the campaign contribution limits defined in
Article XXVIII § 3 of the Colorado Constitution is “merely to require candidates and political
committees to raise funds from a greater number of persons and to compel people who would
otherwise contribute amounts greater than the statutory limits to expend such funds on direct
political expression.” Buckley, 424 U.S. at 21-22. The contribution limits are not so low that they
“harm the electoral process by preventing challengers from mounting effective campaigns
against incumbent officeholders, thereby reducing democratic accountability.” Randall, 548 U.S.
at 249. Under the test set out in Randall, Colorado’s contribution limits are closely drawn to the
sufficiently important state interest of preventing quid pro quo corruption or its appearance and,
therefore, do not violate the political expression and association rights of candidates or
contributors that are guaranteed by the First Amendment to the U.S. Constitution.
D. Claim Two: The Constitutionality of Article XXVIII § 4
Plaintiffs also challenge the voluntary spending limits scheme set out in § 4 of Article
XXVIII of the Colorado Constitution. As described above, this scheme offers candidates for state
offices the choice to either spend without restriction and be subject to § 3’s generally applicable
contribution limits, which I have concluded are valid under the First Amendment, or to accept
the applicable spending cap and potentially benefit from an increased individual contribution
limit of double the standard limit. Plaintiffs contend the Supreme Court’s holding in Davis v.
Federal Election Commission mandates the conclusion that this scheme is unconstitutional.
Defendants argue that § 4 should be viewed, instead, as akin to public financing options in which
candidates may choose to accept a public benefit with strings attached. Under the public
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
130 of 142

131
financing framework, a regulatory scheme does not burden a candidate’s First Amendment rights
if it is not coercive, and thus it is not subject to heightened scrutiny.54 See Corren v. Condos, 898
F.3d 209, 227 (2d Cir. 2018) (“A candidate’s voluntary choice to accept public funds in lieu of
private contributions under the terms of the [public financing o]ption does not entail a burden on
the rights of the candidate . . . .”). I determine that Davis does not clearly apply to the
circumstances in this case, as it was aimed at protecting candidates’ right to freely spend their
personal funds in support of their campaigns and involved a scheme that was unquestionably and
automatically punitive. I conclude, instead, that the analysis for public financing cases is a better
fit for the facts here, and under that framework, that § 4 does not infringe candidates’ First
Amendment rights.
Although Plaintiffs rely primarily on Davis to support their claim that § 4’s voluntary
spending limits scheme violates their First Amendment rights, they cite two other cases they
contend confirm their position: Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett
54 Plaintiffs claim “[a]symmetric campaign contribution schemes trigger strict scrutiny because
‘imposing different contribution . . . limits on candidates vying for the same seat is antithetical to
the First Amendment.’” Pls.’ Trial Summation & Request for J. at 30 (quoting Davis, 554 U.S. at
743-744); see also id. at 32 (“Asymmetric campaign contribution schemes burden a candidate’s
First Amendment rights per se and are subject to strict scrutiny.” (citing Davis, 554 U.S. at 738-
40; Bennett, 564 U.S. at 746-49; Riddle, 742 F.3d at 929)). Here, I find § 4 does not burden First
Amendment rights, but even if it did, I am not convinced that strict scrutiny—and not some other
permutation of heightened scrutiny—would be triggered. The Tenth Circuit seemed to conclude
that the differing contribution limits at issue in the case do “not survive strict scrutiny.” Riddle,
742 F.3d at 929. Nevertheless, the court also discussed the difference between strict scrutiny and
the “less rigorous test for contribution limits, examining whether they are closely drawn to a
sufficiently important governmental interest.” Id. at 927-28. It then “assume[d] that this form of
intermediate scrutiny applies when contributors challenge contribution limits based on the
Fourteenth Amendment’s Equal Protection Clause rather than the First Amendment,” and
determined the “discriminatory limits [we]re not closely drawn to the State’s interest in battling
corruption or the appearance of corruption.” Id. at 928-29. Indeed, in his concurrence in Riddle,
Justice Gorsuch noted that “the Court has yet to apply strict scrutiny to contribution limit
challenges” and summarized the debate over which level of scrutiny should apply in the equal
protection context. Id. at 930-32 (Gorsuch, J., concurring).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
131 of 142

132
and Riddle v. Hickenlooper. In opposition, Defendants advocate for the application of Corren, a
relatively recent Second Circuit case involving public financing. I review the facts of these cases
in detail and conduct a gimlet-eyed comparison.
Davis, which I have already summarized above, considered the right of candidates to
spend money on their own campaigns. 554 U.S. at 729. The Supreme Court had determined, in
Buckley, that the Federal Election Campaign Act’s “ceiling on personal expenditures by
candidates on their own behalf . . . impose[d] a substantial restraint on the ability of persons to
engage in protected First Amendment expression.” 424 U.S. at 52. Davis’s analysis of the
Millionaire’s Amendment was a direct extension of this reasoning from Buckley. See Davis, 554
U.S. at 738-39 (“Buckley’s emphasis on the fundamental nature of the right to spend personal
funds for campaign speech is instructive.”). The Davis Court deemed the provision of
fundraising advantages to the opponents of self-financing candidates to be “an unprecedented
penalty on any candidate who robustly exercise[d]” the First Amendment right “to use personal
funds to finance campaign speech.” Id. at 739. While candidates in Davis had the choice to
refrain from spending personal funds in an amount that would trigger the fundraising advantages
for their opponents, the Court found that choice did “not provide any way in which a candidate
can exercise th[e] right [to make unlimited personal expenditures] without abridgment.” Id. at
740. According to the Court, wealthy candidates had “two choices: abide by a limit on personal
expenditures or endure the burden that is placed on that right by the activation of a scheme of
discriminatory contribution limits.” Id. at 740. The Court stressed that this choice was quite
different from that imposed by the public funding program in Buckley, where “a candidate, by
forgoing public financing, could retain the unfettered right to make unlimited personal
expenditures.” Id. at 739.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
132 of 142

133
The facts in Davis are distinguishable from those in this case in three significant ways.
First, candidates for state office in Colorado, like those in Buckley and unlike those in Davis,
may reject the voluntary spending limits and retain their unfettered right to make unlimited
personal expenditures. Second, Colorado’s scheme does not impose a clear penalty on candidates
who do not accept the spending limits. For an opponent to receive the benefit of higher
contribution limits, that opponent must also accept the spending cap. In Davis, opponents of self-
financing candidates who exceeded the threshold for personal expenditures received the benefit
of higher contribution limits without any downside or corresponding restriction. The third major
difference between Davis and this case is that the choices presented to the candidates, not just the
contribution limits, were asymmetric. Only candidates who could self-fund their campaigns to a
substantial extent faced the unconstitutional choice in Davis, whereas a candidate who could not
contribute considerable funds to his or her campaign would not encounter that choice. Here, all
candidates for state office in Colorado face the same choice. See Colo. Const. Art. XXVIII §
4(3); Colo. Rev. Stat. § 1-45-110(1).55
To be able to receive contributions that are double the standard contribution limit, a
candidate must accept the applicable spending cap and an opponent who does not accept that cap
must raise more than 10% of the cap. See Colo. Const. Art. XXVIII § 4(5). Plaintiffs make too
much of this fact. Focusing on the 10% fundraising trigger, Plaintiffs attempt to portray the § 4
scheme as one that directly penalizes a candidate’s right to spend funds, as the Millionaire’s
Amendment in Davis did. Pls.’ Trial Summation & Request for J. at 31; Pls.’ Reply in Supp. of
55 Calculations regarding the advantages of the options might change as candidates enter the race
and select their preferred path, but candidates who have accepted the spending limit may later
withdraw that acceptance if another candidate enters the race and decides not to accept the
spending limit, see Colo. Const. Art. XXVIII § 4(3), (4).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
133 of 142

134
Request for J. at 30.56 But the 10% requirement is inconsequential here. It does not alter the ways
in which § 4 is distinguishable from the Millionaire’s Amendment. A candidate’s acceptance of
the voluntary spending limits does not result in a clear penalty for non-accepting candidates, and
all candidates for state office face the same choice.
Plaintiffs quote Riddle for the proposition that “[o]nly ‘uniform contribution limit’ laws
can be ‘constitutional.’” Pls.’ Trial Summation & Request for J. at 32 (quoting Riddle, 742 F.3d
922, 929 (10th Cir. 2014)). This reading of Riddle is illogical and dishonest.57 The full sentence
from Riddle that Plaintiffs edited beyond recognition is: “Though a uniform contribution limit
would have been constitutional, the Court noted its difficulty in imagining how Congress would
advance its ‘anticorruption goals’ by creating more severe contribution limits for candidates
financing their own campaigns.” 742 F.3d at 929 (quoting Davis, 554 U.S. at 741). The Tenth
Circuit never indicated that only uniform contribution limits could be constitutional and certainly
did not in this sentence discussing Davis.58
In Riddle, a panel of the Tenth Circuit considered a Colorado statute that, in effect,
permitted candidates from parties with a primary to receive twice the amount from an individual
56 Notably, in 2021, Mr. Lopez’s fundraising did not trigger the higher contribution limits for
opponents who had accepted the voluntary spending limits; Governor Polis’s fundraising did.
See Pls.’ Br. Regarding Mootness & Standing at 8, ECF No. 82 (“Polis’s fundraising enabled all
Section 4 accepting gubernatorial candidates to receive double the maximum donation from their
individual contributors.”).
57 This proposition, if true, would directly contradict Plaintiffs’ repeated insistence that strict
scrutiny should apply to asymmetric contribution limits. See supra note 54. While strict scrutiny
is a high hurdle, it is still a test that can be met.
58 To be sure, the Supreme Court observed in Davis that it “ha[d] never upheld the
constitutionality of a law that imposes different contribution limits for candidates competing
against each other.” 554 U.S. at 738. And the Tenth Circuit quoted that language in Riddle. See
742 F.3d at 929. However, observing that the Court has never done something is not the same as
concluding that it cannot.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
134 of 142

135
contributor than candidates that did not participate in a primary could. 742 F.3d at 924-25.59
Individual contributors to a write-in candidate sued, claiming that the statute violated the First
Amendment and the Equal Protection Clause of the Fourteenth Amendment. Id. at 924. The
Tenth Circuit panel decided the contributor-plaintiffs’ claim brought under the Equal Protection
Clause and declined to address the First Amendment claims. Id. The court determined that the
Colorado statute resulted in “candidates running against each other with different contribution
limits.” Id. at 930. As such, it “follow[ed] the teaching of Davis” and held that the statute was
unconstitutional because it was “not closely drawn to the State’s anticorruption goal.” Id. While
Riddle followed Davis, it is not instructive in this case. The statute at issue in Riddle did not offer
candidates any campaign financing options.60 Moreover, Plaintiffs here did not assert a claim
under the Equal Protection Clause.
The third, and in my view most relevant, case Plaintiffs highlight is Bennett, which I also
summarized above. In that case, the Supreme Court invalidated the matching funds provision of
Arizona’s public financing program. 564 U.S. at 728, 754-55. Under that provision, publicly
funded candidates would receive additional matching state funds when a privately financed
candidate’s expenditures, which included money spent by independent groups, exceeded the
initial allotment of state funds to the publicly financed candidate. Id. at 729-30. The Court
59 As explained in my Findings of Fact, Article XXVIII § 3 of the Colorado Constitution sets
contribution limits for both the primary and general election. Riddle, 742 F.3d at 924. In 2004,
the legislature enacted a statute that would allow candidates from a party with a primary to spend
money they raised during the primary phase on the general election. Id. So, for example, a
Republican nominee for the state legislature could receive $200 for the primary and $200 for the
general election from an individual contributor and could spend the full sum on the general
election, while unaffiliated and write-in candidates could receive only $200 from an individual
contributor. See id.
60 The only choice candidates might have had in Riddle was to change their party affiliation or
status, which is unquestionably not a campaign finance choice comparable to the one candidates
face under section § 4.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
135 of 142

136
determined that this benefit of providing matching funds to publicly financed candidates, i.e.,
“the direct and automatic release of public money,” was “a far heavier burden” on privately
financed candidates than even the burden imposed on self-financing candidates in Davis. Id. at
737; see also id. at 742-43 (“The direct result of the speech of privately financed candidates and
independent expenditure groups is a state-provided monetary subsidy to a political rival. That
cash subsidy, conferred in response to political speech, penalizes speech to a greater extent and
more directly than the Millionaire's Amendment in Davis.”).61 For candidates in Arizona to
receive the initial public funding allotment, they had to agree to certain restrictions and
obligations, including limiting their expenditure of personal funds to $500 and adhering to a
spending cap. Id. at 728-29. One of the justifications for the matching funds provision that the
State offered was it ensured enough candidates participated in the public funding system and thus
helped to combat corruption. Id. at 752. The Court concluded this justification was insufficient
for the burden imposed, stating: “[W]e have never held that a State may burden political
speech—to the extent the matching-funds provision does—to ensure adequate participation in a
public funding system.” Id. at 753.
Like Davis, the penalty in Bennett is obvious, and indeed even more severe. “[T]he direct
and automatic release of public money” guarantees publicly funded candidates a campaign
financing advantage. Id. at 737; see also id. at 742 (referring to the matching funds as “a state-
provided monetary subsidy to a political rival” that is the “direct result of the speech of privately
financed candidates and independent expenditure groups”). Again, the facts here are
distinguishable because there is no clear penalty for candidates who do not accept the spending
61 Of great significance to the Court were the additional burdens on independent expenditure
groups, which it viewed as “worse than the burden [the law] impose[d] on privately financed
candidates.” Bennett, 564 U.S. at 738-39.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
136 of 142

137
limits under § 4. As the Court in Bennett noted, raising contribution limits is not as problematic
as the direct provision of public funds because a candidate “still ha[s] to go out and raise the
funds.” Id. at 737. And, while publicly financed candidates in Bennett did have to accept a
spending cap and other restrictions, the matching funds were a much more direct and certain
penalty than the doubled contribution limits available under § 4. This difference is underscored
by the percentage of Arizona candidates who participated in the public financing program
compared to the percentage of Colorado candidates who accept § 4’s voluntary spending limits
scheme. During general elections from 2002 to 2010, 52% to 67% of candidates for Arizona state
office chose to participate in the public funding program. McComish v. Bennett, 611 F.3d 510,
517 (9th Cir. 2010), rev’d sub nom., Bennett, 564 U.S. 721. In Colorado, from 2014 through
2022, 30% to 37% of eligible candidates chose to accept the voluntary spending limits under § 4.
Trial Tr. at 895:14-17; see also Trial Ex. 380. If the doubled contribution limits were a true
penalty for candidates who did not accept the spending limits, as the matching funds were in
Bennett, the percentage of candidates who accept § 4’s limits would be much higher.
Additionally, of the Colorado candidates for state office in 2022 who declined the voluntary
spending limits, 21% spent more than the limit. Trial Ex. 379. This statistic shows that
candidates face a substantial risk of being outspent if they accept the voluntary spending limits,
and thus the spending limits counterbalance the doubled contribution limits such that the latter do
not provide a clear campaign financing advantage.
In Corren, the Second Circuit considered whether Vermont’s public financing program
“unfairly and unjustifiably burden[ed] the speech and associational rights of [publicly financed
candidates], their supports, and political parties.” 898 F.3d at 212-13. To receive public funds
under the Vermont program, a candidate must have accepted certain restrictions on his or her
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
137 of 142

138
receipt and expenditure of contributions. Id. at 214-15. The court, in Corren, summarized the
general standard for considering such public financing programs, stating:
As the Court of Appeals for the Fourth Circuit observed, “[s]ince Buckley the
circuit courts have generally held that public financing schemes are permissible if
they do not effectively coerce candidates to participate in the scheme.” N.C. Right
to Life Comm. Fund for Indep. Political Expenditures v. Leake, 524 F.3d 427, 436
(4th Cir. 2008). That is because such schemes do not burden candidates’ rights if
they merely create another viable funding option rather than compel candidates to
choose public funding. See Daggett v. Comm’n on Gov’t Ethics & Election
Practices, 205 F.3d 445, 467 (1st Cir. 2000) (“A law providing public funding for
political campaigns is valid if it achieves ‘a rough proportionality between the
advantages available to complying candidates . . . and the restrictions that such
candidates must accept to receive these advantages.’” (ellipsis in original) (quoting
Vote Choice, Inc. v. DiStefano, 4 F.3d 26, 39 (1st Cir. 1993))) . . . .
Id. at 219-20. The court explained that a statute may “require[] an individual to choose between
two methods of exercising the same constitutional right . . . provided the statute does not
diminish a protected right,” or if the statute does diminish a protected right, the burden on that
right must be justified by a sufficient interest. Id. at 221 (quoting Republican Nat’l Comm. v.
Fed. Election Comm’n, 487 F. Supp. 280, 284-85 (S.D.N.Y. 1980) (three-judge court), aff’d
mem., 445 U.S. 955 (1980)); see also id. at 222 n.6 (“As long as an offer of public funding to
which an objectionable condition attached is not so advantageous that a candidate is effectively
compelled to take it, we could expect the candidate to reject such an offer, in which case the
condition would not burden the candidate.”). Applying that standard to the Vermont program, the
court held that “the voluntary decision to accept public funds and forgo the ability to accept
private contributions does not burden [publicly financed candidates’] First Amendment rights.”
Id. at 222.62
62 The Court also held that the restriction on contributions other than those covered by the public
funding program did not burden a candidate’s constitutional rights, even though it might restrict
the candidate’s ability to self-finance his or her campaign. Corren, 898 F.3d at 227 (“[I]f a
candidate voluntarily chooses to accept public funds in lieu of private fundraising—presumably
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
138 of 142

139
The scheme in Corren, like § 4 here, offered candidates a status quo option that was
constitutional. The alternative option in Corren, again like the one under § 4, provided
candidates with a fundraising benefit but required that candidates accept certain campaign
finance restrictions. The benefit offered and corresponding restrictions in Corren are comparable
to the benefit and restriction offered to candidates who accept the voluntary spending limits
under § 4; the Corren benefits and restrictions are simply of a greater magnitude. Although a
public funding program is not at issue in this case, the public funding framework fits because
there is a benefit tied to a restriction and that option is offered as an alternative to the
constitutional, status-quo campaign financing.
The through line in Corren, Davis, and Bennett is the consideration of how much
pressure the schemes place on self-financing or privately financed candidates to reduce their
spending or accept public funding. If the standard from Corren were applied in Davis and
Bennett, the regulations at issue in Davis and Bennett would be found to be sufficiently coercive
to burden candidates’ First Amendment rights and thus trigger heightened scrutiny. In contrast,
the spending limit and doubled individual contribution limits offered as the alternative option
under § 4 provide “a rough proportionality between the advantages available to complying
candidates . . . and the restrictions.” Corren, 898 F.3d at 220 (quoting Daggett, 205 F.3d at 467).
As explained above, the incentive provided by § 4—the doubling of contribution limits—is less
attractive than public financing options, as it does not guarantee any funding to candidates. The
spending limits restriction is proportional to the benefit of the doubled contribution limits, which
is apparent from the fact that 21% of candidates who declined the spending limits spent more
because the grant of public funds will expand her ability to speak—then prohibiting her from
using funds other than the public grant, including her own, does not burden her constitutional
rights.”).
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
139 of 142

140
than the limit and only 3% of candidates who accepted the spending limits spent up to the limit.
Trial Ex. 379. Candidates who do not accept the spending limits may still gain an advantage by
spending over the limits, while at the same time, candidates who accept the spending limits are
not severely restricted. Because of the spending limits, candidates are not incentivized by the
doubled contribution limits to such a great extent that it could be said they are coerced or
compelled into forgoing the status quo option. The fact that approximately two-thirds of
candidates declined the spending limits alternative from 2014 through 2022 is strong evidence
that candidates do not feel compelled to choose that option. Trial Tr. at 895:14-17; see also Trial
Ex. 380.
Moreover, nothing in the record suggests that candidates who choose not to accept the
spending limits under § 4 are less competitive than those who accept them. Instead, the
testimony from current and former candidates for Colorado state offices supports that candidates
make strategic decisions for their individual campaigns about whether to accept or decline the
spending limits. See, e.g., Trial Tr. at 135:7-9 (Senator Pelton testifying that he made a
“calculation about what’s in [his] strategic best interests” regarding his decision to decline the
spending limits for the 2026 election). Consequently, I conclude that the voluntary decision
candidates face to decline or accept the expenditure limits under § 4 does not burden their First
Amendment rights.
Given that conclusion, Defendants need only establish a rational basis for § 4’s voluntary
spending limits scheme. See Ysura v. Pocatello Educ. Ass’n, 555 U.S. 353, 359 (2009). Plaintiffs
do not argue that such a basis is lacking, and I conclude that it is not. The interest articulated by
Defendants is “increasing the speech of candidates who are unlikely to raise up to the spending
limit, thereby providing candidates with two speech-maximizing choices, and by ‘relieving [] the
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
140 of 142

141
burdens of private fundraising.’” Defs.’ Written Summation at 44 n.15 (quoting Corren, 898
F.3d at 226).63 Providing candidates with speech-maximizing options is a valid interest. As the
Supreme Court in Buckley stated, “facilitat[ing] and enlarg[ing] public discussion and
participation in the electoral process” are “goals vital to a self-governing people.” 424 U.S. at 92-
93.64
* * *
The § 4 voluntary spending limits scheme falls between two separate legal frameworks. I
conclude the public financing cases provide the best fitting legal standard because the facts in
Davis, Riddle, and Bennett are distinguishable and their reasoning does not undermine the
standard from Corren. The extension of Davis and Bennett to the circumstances here would
potentially conflict with Buckley’s endorsement of public financing programs. When no higher
court has addressed this conflict or indicated that Davis and Bennett have a broader reach, I
decline to make that leap, especially when doing so would overrun the will of the people of
Colorado. Candidates in Colorado have a status quo option that does not infringe on their First
Amendment rights. Section 4 offers them an alternative option they may strategically choose to
accept or reject. This scheme is not coercive, which is borne out by the participation numbers,
63 Plaintiffs argue that the Tenth Circuit has found the purpose of § 4 is “to encourage candidates
to limit their expenditures.” Lopez, 2023 WL 1960802, at *1. This statement by the panel is
dicta, made in an Order and Judgment dismissing Plaintiffs’ appeal as moot. Moreover, the court
did not find that § 4 coerced candidates to limit their spending nor that the only purpose of § 4
was to encourage candidates to limit their expenditures.
64 Colorado does not contend “that Section 4(g) furthers the anticorruption interest served by [its]
contribution limits.” Defs.’ Written Summation at 44 n.15. And I recognize that the purposes of
public financing endorsed by the Supreme Court in Buckley, including that such programs
“facilitate and enlarge public discussion,” work to eliminate “improper influence,” and relieve
candidates of the “rigors of soliciting private contributions,” 424 U.S. at 92-93, 96, 249, are not
applicable for § 4’s voluntary spending limits. Nevertheless, Defendants must only show a
rational basis underlying the scheme, and they have done so.
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
141 of 142

142
and it is supported by a rational basis. Thus, I conclude § 4’s voluntary spending limits scheme
does not offend the First Amendment to the Constitution.
III. CONCLUSION & ORDER
Accordingly, I conclude that Plaintiffs’ constitutional challenges to §§ 3 and 4 of Article
XXVIII of the Colorado Constitution fail, and Plaintiffs are not entitled to any relief. Judgment
shall therefore enter in favor of Defendants and against Plaintiffs on both Claims One and Two
in the Amended Complaint.
DATED this 26th day of March, 2026.
______________________________
JOHN L. KANE
SENIOR U.S. DISTRICT JUDGE
Case No. 1:22-cv-00247-JLK Document 128 filed 03/26/26 USDC Colorado pg
142 of 142