This article was originally published in The Journal of Federal Agency Action, November–December 2026, Vol. 4, No. 6.

In this first part of a two-part article, the authors analyze whether trading a political event contract is conduct or speech, whether political prediction markets should be considered as “metaspeech,” and three converging First Amendment interests. In the conclusion of this article, to be published in the next issue of The Journal of Federal Agency Action, the authors will discuss, among other things, the campaign finance analogy and its limits, the concept of “expressive association,” whether the “gambling” label can resolve the constitutional question, and content-based restrictions and strict scrutiny, before concluding with the case for tailored regulation.

Introduction

Prediction markets are fast becoming one of the most closely watched barometers of modern life. An event contract or swap is a bilateral financial agreement between traders on a trading platform over whether an event will occur, which is resolved entirely by the objective outcome of the event. These platforms enable users to trade on virtually anything: whether a team will win the next World Cup game, whether Costco will raise the price of its infamous hot dog combo, whether a redistricting referendum will pass, and whether Karen Bass or Spencer Pratt will win the Los Angeles race for mayor. The platforms offering these trades now process roughly $24 billion in monthly trading volume.[1]

This marketplace is attracting regulatory scrutiny at all levels of government. Prediction markets focused on elections, in particular, have drawn criticism as they have expanded ahead of the 2026 midterms, with opponents complaining they are unseemly or asserting that they pose a risk to the democratic process.

Minnesota recently became the first state to prohibit prediction market platforms altogether, though the federal government moved swiftly to block the ban before it became effective.[2] In defending the law, Minnesota Attorney General Keith Ellison cited election trading specifically, warning that prediction markets “invite people to gamble on not just sports scores, but on the outcomes that shape our democracy.”

Arizona Attorney General Kris Mayes took a more aggressive stance, filing criminal charges against platform provider Kalshi—including four counts alleging that the company violated state law by accepting trades on Arizona elections. That prosecution was enjoined by the U.S. District Court for the District of Arizona on grounds that the authority of the Commodity Futures Trading Commission (CFTC) to regulate these markets preempts the state’s action.[3]

That threshold jurisdictional dispute has now produced a circuit split. In KalshiEX LLC v. Flaherty, the Third Circuit, reviewing a preliminary injunction, held that Kalshi had demonstrated a reasonable likelihood of success in showing that sports-related event contracts traded on its CFTC-regulated designated contract market are “swaps” and that the Commodity Exchange Act preempts New Jersey’s contrary gambling laws. By contrast, in KalshiEX, LLC v. Assad, the Ninth Circuit affirmed the dissolution of preliminary relief against Nevada, concluding that sports event contracts were likely not “swaps” within the Act’s definition—and, independently, that Kalshi’s self-certification of those contracts violated the CFTC’s own regulation categorically prohibiting the listing of gaming-related contracts on designated contract markets, 17 C.F.R. § 40.11(a)—such that the contracts were not insulated from state gaming regulation by the CFTC’s exclusive jurisdiction. The Ninth Circuit remanded for the district court to consider Nevada’s challenges to Kalshi’s election contracts in the first instance.[4] Petitions for certiorari arising from both sides of that split are now pending before the Supreme Court,[5] although neither appellate decision addressed whether a subject-matter restriction on political event contracts independently burdens First Amendment interests.

On Capitol Hill, Senator Jeff Merkley and Representative Jamie Raskin introduced the STOP (Stop Trading On Predictions) Corrupt Bets Act, which would prohibit trading on elections altogether.[6]

The Wisconsin Elections Commission issued an advisory informing residents that trading on an election could invalidate their ballot.[7]

And 16 consumer groups have urged the CFTC to ban election trading outright.[8]

These legislative and enforcement initiatives share a common rhetorical frame: critics describe prediction market trades as “gambling” or “betting” on democracy, and argue that categorical prohibition is appropriate.[9]

But these regulatory battles have largely overlooked a fundamental constitutional question: When the government singles out political event contracts for prohibition—while permitting functionally identical contracts on other subjects—does that prohibition burden First Amendment interests?

This two-part article argues that it does. Trading in political event contracts is not mere financial speculation dressed up in constitutional garb, it is the next frontier of political speech and expressive association. Categorical bans on political event contracts burden at least three distinct First Amendment interests:

1. the trader’s interest in expressing political judgment through financial commitment,
2. the platform’s interest in publishing market-generated political information, and
3. the public’s interest in receiving information about matters of public concern.

Any effort to categorically restrict or ban political event contracts must therefore clear the highest level of constitutional scrutiny.[10]

The first part of this two-part article analyzes whether trading a political event contract is conduct or speech, whether political prediction markets should be considered as “metaspeech,” and three converging First Amendment interests. In the conclusion, to be published in the next issue of The Journal of Federal Agency Action, this article will discuss, among other things, the campaign-finance analogy and its limits, the concept of “expressive association,” whether the “gambling” label can resolve the constitutional question, and content-based restrictions and strict scrutiny, before it concludes with the case for tailored regulation.

The Threshold Objection: Speech or Conduct?

The strongest objection to First Amendment protection is straightforward: trading a political event contract is conduct, not speech. On this view, the government is not prohibiting anyone from endorsing a candidate, criticizing a referendum, publishing election odds, or debating public policy. It is merely regulating a financial transaction that pays out based on a future event.

That objection cannot be dismissed. Not every transaction that reveals a belief becomes protected speech. A person who buys oil futures may be expressing a view about geopolitical instability; a person who shorts a company’s stock may be expressing a view about management. Yet those transactions remain subject to extensive regulation without implicating the First Amendment.

This point deserves serious engagement, not dismissal. A person who buys oil futures because he or she expects a supply disruption is expressing a view about geopolitical instability. A person who shorts a company’s stock is expressing a view about management competence. A person who purchases credit default swaps is expressing a view about sovereign debt risk. Yet no one contends that those transactions are constitutionally protected speech. If the First Amendment attaches to prediction market trades because they “express” views about political outcomes, a skeptic might reasonably ask why it does not also attach to every financial transaction that reveals a belief about a future state of the world—a result that would constitutionalize securities regulation, derivatives law, and the entire apparatus of financial oversight.

The answer lies in the constitutional significance of the government’s basis for regulation. Oil futures, credit default swaps, and short sales are regulated because of market structure concerns—systemic risk, fraud prevention, capital adequacy, investor protection—that have nothing to do with the content of the views those transactions may incidentally reveal. The government does not ban oil futures because they concern the Middle East, or permit them on some commodities while prohibiting them on others based on the political sensitivity of the underlying subject matter. But that is precisely what a categorical ban on political event contracts does. It singles out contracts because they concern elections, referenda, and legislation—while permitting functionally identical contracts on every other subject. The regulatory distinction is drawn by reference to what the contract communicates. That is what transforms an otherwise permissible exercise of market regulation authority into a content-based restriction on political expression, triggering constitutional scrutiny that ordinary financial regulation does not.

But the objection proves too much. The First Amendment protects flag burning, parade marching, and campaign contributions—all conduct—because each contains a sufficiently expressive component and because the government, in regulating them, targets that component.[11] The relevant question is not whether an activity is “speech” or “conduct” in the abstract, but whether the regulated activity contains a meaningful expressive or informational element and whether the government’s restriction targets that element.

Political event contracts differ from ordinary financial transactions in a critical respect: their subject matter is the political process itself. A contract on crude oil futures conveys information about commodity supply and demand. A contract on the likelihood of a Senate confirmation conveys collective political judgment. That distinction is important because the Supreme Court has consistently held that political speech occupies the highest rung of First Amendment protection.[12] When a state bans contracts specifically because they concern elections, candidates, referenda, or legislation—while permitting contracts on every other subject—it identifies political subject matter and removes it from an otherwise available marketplace of information. That is content-based targeting of political expression, and it triggers constitutional scrutiny.

Application of the O’Brien Test

Even if one characterizes prediction market trading as expressive conduct rather than pure speech, the Supreme Court’s framework in United States v. O’Brien requires that government restrictions on such conduct satisfy a four-part test:

1. the regulation must be within the constitutional power of the government,
2. it must further an important or substantial governmental interest,
3. the governmental interest must be unrelated to the suppression of free expression, and
4. the incidental restriction on alleged First Amendment freedoms must be no greater than is essential to the furtherance of that interest.[13]

A categorical ban on political event contracts fails this test—particularly prongs 3 and 4.

As to the third prong, the government’s interest cannot be “unrelated to the suppression of free expression” when the very basis for the ban is the political subject matter of the contract. A regime that permits contracts on sports outcomes, entertainment awards, weather events, and economic indicators—but prohibits contracts on elections and legislation—is not regulating market mechanics. It is targeting political content. The distinction is drawn solely by reference to what the contract communicates. That is the hallmark of a restriction related to the suppression of expression, and it removes the regulation from O’Brien‘s more lenient framework altogether, triggering strict scrutiny instead.[14]

Even assuming arguendo that the government’s interest is unrelated to suppression—perhaps justified by election integrity or consumer protection—the ban still fails the fourth prong. The incidental restriction on First Amendment freedoms is far “greater than is essential” when precisely targeted alternatives exist. Anti-manipulation rules can address market integrity. Disclosure requirements can protect consumers. Conflict of interest restrictions can prevent insiders from trading. Foreign participation bans can address national security concerns. Position limits can prevent concentrated influence. Each of these measures addresses the asserted governmental interest without eliminating an entire category of political information. A categorical ban does not merely incidentally restrict expression; it eliminates the medium entirely. O’Brien does not permit such overbreadth.[15]

Thus, whether one applies strict scrutiny (as content-based restrictions on political speech require) or the intermediate scrutiny of O’Brien (as the most favorable framing for the government), a categorical ban on political event contracts cannot survive constitutional review.

Political Prediction Markets as “Metaspeech”

Before turning to the First Amendment rights of individual traders, it is worth appreciating that the prediction market itself is a speaker—one whose voice deserves independent constitutional protection. Each trade placed on a prediction market platform does not merely reflect the private judgment of an individual participant.

Aggregated across thousands of traders, those individual expressions merge into a market price that encodes the collective knowledge of the entire trading community at a given moment in time.[16] The market price “aggregates[traders’] opinions into a number. That number is the prediction about what is going to happen. In doing so, the market articulates a viewpoint and predicts the outcome of a future event.”[17]

The capacity to distill dispersed individual predictions into a single, observable collective judgment distinguishes prediction markets from almost every other medium of expression. A newspaper editorial reflects the views of its editorial board. A political poll reflects the views of a sample. But a prediction market reflects the aggregated, incentive-aligned judgments of everyone who has been willing to back a belief with money. The resulting price signal is more than the sum of its parts. It is a form of what Miriam Cherry and Robert Rogers aptly term “metaspeech”: an expression generated by the market that carries independent informational and expressive value beyond any single trader’s position—an expression worthy of First Amendment protection.

As Justice Oliver Wendell Holmes Jr. observed, the First Amendment’s core purpose is to safeguard “a marketplace of ideas” in which truth emerges through the competition of opposing views.[18] Political prediction markets make that metaphor literal. They are not merely analogous to the marketplace of ideas; they are the marketplace of ideas rendered tangible, where participants assign quantifiable confidence to political propositions, and the resulting market price communicates that confidence to the world.

Unlike a poll, a prediction market requires participants to back their assessments with something of value. Unlike an editorial, a market price reflects the aggregated judgments of everyone willing to accept financial consequences for their views. That financial commitment is central to the market’s informational function—a person may casually say that a candidate “has no chance” or that a bill “will definitely pass,” but a trader who risks money on that view has an incentive to be more careful. Prediction markets thus create a mechanism for testing political claims against financial consequences.

Indeed, studies show that prediction markets are better at predicting political results than traditional forms of polling. David Rothschild’s analysis of the 2008 election cycle found that debiased prediction market forecasts “provide more accurate probabilities of victory and more information than debiased poll-based forecasts,” particularly early in the cycle and in uncertain races—the races typically of most interest.[19] Rothschild further demonstrated that prediction market data contained independently valuable information not encompassed by even the most sophisticated poll-based forecasts. These findings built on over a decade of earlier empirical work demonstrating that prediction markets outperformed polls in predicting vote share, and subsequent studies based on more recent data from the 2024 presidential election continue to confirm this phenomenon.[20]

Three Converging First Amendment Interests

The Trader’s Expressive Interest

The Supreme Court has recognized for more than 50 years that capping the amount of money that can be donated to or spent by a political campaign “impose[s] direct and substantial restraints on the quantity of political speech” and limits the very political expression that is “at the core of our electoral process and of the First Amendment freedoms.”[21] Political speech “does not lose First Amendment protection ‘simply because its source is a corporation[,]'” making clear that neither the identity of the speaker nor the medium of expression can justify content-based suppression of political speech.[22]

In Buckley v. Valeo, the Court observed that restrictions on political expenditures impose “direct and substantial restraints on the quantity of political speech.”[23] In Citizens United, the Court held that “[r]apid changes in technology—and the creative dynamic inherent in the concept of free expression—counsel against upholding a law that restricts political speech in certain media or by certain speakers.”[24] And in FEC v. Cruz, the Court went further, holding that any law placing a “drag” on the expenditure of money to communicate political ideas is presumptively unconstitutional,[25] explaining that personal loans to a campaign “may be a useful tool to signal that the political[candidate] is confident enough in his campaign to have skin in the game, attracting the attention of donors and voters alike.”[26]

Political event contracts apply this principle in a new medium. A trader who takes a position on whether a candidate will win, a referendum will pass, or a policy will be enacted is expressing political judgment through financial commitment. Like many other forms of speech, political event contracts require money to proliferate ideas. But this is nothing new; “virtually every means of communicating ideas in today’s mass society requires the expenditure of money.”[27] These contracts are tools that increase “the number of issues discussed, the depth of their exploration, and the size of the audience reached.”[28] Courts “must decline to draw, and then redraw, constitutional lines based on the particular media or technology used to disseminate political speech from a particular speaker.”[29]

The Platform’s Publishing Interest

A prediction market platform does more than match buyers and sellers. It defines contracts, maintains order books, displays prices, reports trading volume, publishes market odds, and disseminates the resulting information to users and the public in real time. Those acts have communicative significance.

The integration between the press and prediction markets has accelerated rapidly. In December 2025, CNN announced a formal partnership with Kalshi—a leading prediction market platform—making Kalshi its official prediction market partner and granting CNN real-time API access to Kalshi’s data for use across its television, digital, and social channels.[30] CNN’s Chief Data Analyst Harry Enten now cites Kalshi data during live coverage. Polymarket, a decentralized prediction market, secured a formal partnership with X (formerly Twitter) in mid-2025, combining its market data with Grok’s real-time analytics capabilities to deliver prediction market insights to millions of users. That major news and technology organizations have partnered with prediction market platforms to offer up-to-the-minute information confirms the role these markets play as twenty-first-century instruments of expressive communication.

As the Court observed in Buckley, “[t]he electorate’s increasing dependence on television, radio, and other mass media for news and information has made these expensive modes of communication indispensable instruments of effective political speech.” Prediction markets, which now process tens of billions of dollars in monthly trading volume and are embedded in real-time news coverage by the nation’s largest broadcasters, are today’s indispensable instruments of political communication and deserve the same constitutional protection. A prohibition on political event contracts does not merely regulate financial instruments—it shuts down a publisher of vital political information.[31]

The Public’s Right to Receive Political Information

The First Amendment protects not only the right to speak but also the right to receive information and ideas.[32] Political prediction markets contribute to that interest by giving voters, journalists, campaigns, researchers, and policymakers access to real-time assessments of political events. When journalists cite prediction market prices during election coverage, when analysts compare market odds with polling averages, or when voters use market signals to understand the perceived likelihood of a political outcome, the information has entered the public square.

The government may not suppress a source of political information simply because the information might influence voters. Polls influence voters. Endorsements influence voters. Fundraising totals influence voters. Editorials influence voters. Influence is not a constitutional defect—it is often the very purpose of political speech. Suppressing prediction markets because their probability signals might affect voter behavior is precisely the kind of government control over political discourse the First Amendment emphatically prohibits.[33]

---

Notes

Editor’s note: This article will conclude in the next issue of The Journal of Federal Agency Action.

Rainni Crutchfield and Lauren Lamp, J.D. candidates at the University of Richmond School of Law, assisted in the preparation of this article.

[1] Nathan Goldman, Prediction Markets Hit $24 Billion a Month. States Are Fighting Back, Forbes (July 1, 2026), https://www.forbes.com/sites/nathangoldman/2026/07/01/prediction-markets-hit-24b-a-month-now-states-are-fighting-back/.

[2] See United States v. State of Minnesota, No. 26-cv-2661 (KMM/DTS), 2026 WL 2150211 (D. Minn. July 27, 2026) (granting preliminary injunction prohibiting the state from enforcing Minn. Stat. § 609.7615 against CFTC-registered contract markets).

[3] See Kalshi v. Mayes, No. CV-25-00753-PHX-DGC (D. Ariz. 2025) (granting preliminary injunction on CFTC preemption grounds).

[4] KalshiEX LLC v. Flaherty, 172 F.4th 220 (3d Cir. 2026) (2–1 decision affirming preliminary injunction after concluding that Kalshi demonstrated a reasonable likelihood of success on its CEA preemption theory); KalshiEX, LLC v. Assad, No. 25-7516 (9th Cir. Aug. 28, 2026) (affirming dissolution of preliminary injunction as to sports event contracts and remanding for consideration of election contracts).

[5] Petition for Writ of Certiorari, Flaherty v. KalshiEX, LLC, No. 26-299 (U.S. filed Sept. 2, 2026); Petition for Writ of Certiorari, Robinhood Derivatives, LLC v. Dreitzer, No. 26-338 (U.S. filed Sept. 10, 2026).

[6] STOP Corrupt Bets Act of 2026, H.R. 8123, 119th Cong. (2026) (introduced by Sen. Jeff Merkley and Rep. Jamie Raskin); see also Press Release, Jeff Merkley, New Merkley, Raskin Legislation Bans Gambling on Elections, Sports, War, and Government Activity (Mar. 26, 2026), https://www.merkley.senate.gov/new-merkley-raskin-legislation-bans-gambling-on-elections-sports-war-and-government-activity/. Consumer groups have jumped into the fray, calling upon the CFTC to ban trading on elections altogether. See Letter from Sixteen Consumer Organizations to Christopher J. Kirkpatrick, Sec’y, Commodity Futures Trading Comm’n, Re: Prediction Markets, RIN 3038-AF65 (July 27, 2026), https://bettermarkets.org/wp-content/uploads/2026/07/Prediction-Markets-Comment-Letter.pdf.

[7] Don’t Bet on Your Ballot: WEC Warns Voters Against Election Gambling, Wis. Elections Comm’n (July 21, 2026), https://elections.wi.gov/news/dont-bet-your-ballot-wec-warns-voters-against-election-gambling.

[8] See Letter from Sixteen Consumer Groups, supra note 6.

[9] More than half of U.S. states restrict or ban gambling on elections entirely. Shifra Dayak, More Than Half of States Restrict Betting on Elections, Pew Rsch. Center (June 23, 2026), https://www.pewresearch.org/short-reads/2026/06/23/more-than-half-of-states-restrict-betting-on-elections/.

[10] While outside the scope of this article, the reasoning may extend to other trades on prediction markets, which may qualify for protections available to commercial speech, subject to intermediate scrutiny under Central Hudson Gas & Electric Corp. v. Public Service Commission, 447 U.S. 557 (1980). Non-misleading commercial speech about lawful subject matters can be restricted only if the restriction directly advances a substantial government interest and is no more extensive than necessary to serve that interest.

[11] See Texas v. Johnson, 491 U.S. 397 (1989) (flag burning is speech protected by the First Amendment); Buckley v. Valeo, 424 U.S. 1, 16-17 (1976) (campaign contributions); Hurley v. Irish-American Gay, Lesbian and Bisexual Group of Boston, 515 U.S. 557 (1995) (parade marching).

[12] Buckley, 424 U.S. at 14 (1976) (“Discussion of public issues and debate on the qualifications of candidates are integral to the operation of the system of government established by our Constitution.”).

[13] United States v. O’Brien, 391 U.S. 367, 377 (1968).

[14] See Reed v. Town of Gilbert, 576 U.S. 155, 163 (2015) (holding that laws targeting speech based on communicative content are subject to strict scrutiny, not the intermediate scrutiny of O’Brien).

[15] See O’Brien, 391 U.S. at 377 (requiring that “the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest.”).

[16] See Miriam A. Cherry & Robert L. Rogers, Prediction Markets and the First Amendment, U. Ill. L. Rev. 833 (2008) (developing the concept of prediction market prices as “metaspeech”).

[17] Id. at 850 (explaining that the market price “aggregates[traders’] opinions into a number. That number is the prediction about what is going to happen. In doing so, the market articulates a viewpoint and predicts the outcome of a future event”).

[18] Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting).

[19] David Rothschild, Forecasting Elections: Comparing Prediction Markets, Polls, and Their Biases, 73 Pub. Op. Q. 895, 895-916 (2009).

[20] See Joyce Berg et al., Results from a Dozen Years of Election Futures Markets Research, in Handbook of Experimental Economics Results (2008); Laurie E. Cutting et al., Are Betting Markets Better than Polling in Predicting Political Elections? (Vanderbilt University, Working Paper, July 11, 2025), https://arxiv.org/abs/2507.08921.

[21] See Buckley, 424 U.S. at 14 (citation omitted); Citizens United v. Federal Election Comm’n, 558 U.S. 310, 339-40 (2010).

[22] Citizens United, 558 U.S. at 342 (quoting First Nat’l Bank of Boston v. Bellotti, 435 U.S. 765, 784 (1978)).

[23] Buckley, 424 U.S. at 39.

[24] Citizens United, 558 U.S. at 364.

[25] Federal Election Comm’n v. Cruz, 596 U.S. 289, 304-05 (2022).

[26] Id. at 304 (explaining that personal loans to a campaign “may be a useful tool to signal that the political[candidate] is confident enough in his campaign to have skin in the game, attracting the attention of donors and voters alike.”).

[27] Buckley, 424 U.S. at 19.

[28] Id.

[29] Citizens United, 558 U.S. at 326.

[30] See, e.g., Kalshi to Become CNN’s Official Prediction Market Partner, Kalshi News (Dec. 2, 2025); Aisha Malik, X Names Polymarket as Its Official Prediction Market Partner, Techcrunch (June 6, 2025).

[31] See Moody v. NetChoice, LLC, 603 U.S. 707, 716 (2024) (“To the extent that social-media platforms create expressive products, they receive the First Amendment’s protection.”).

[32] Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, 425 U.S. 748, 757 (1976) (“Freedom of speech presupposes a willing speaker. But where a speaker exists . . . the protection afforded is to the communication, to its source and to its recipients both.”).

[33] First Nat’l Bank of Boston, 435 U.S. at 776.