Hold a Forecast Contract on ForecastEx and two separate flows of money run through one position. One is the thing you came for: a dollar at settlement if the event question goes your way, nothing if it does not. The other arrives every month whichever way it goes, calculated on what the position is worth, and it is called an Incentive Coupon. On Kalshi the same idea appears as an annual percentage yield paid on cash and on the collateral behind open positions.
Neither payment is a trick. Money posted as collateral earns interest somewhere, and returning it to the person whose money it is beats the alternative. The problem is narrower: the category added a second source of money to the trading screen without writing down how to describe it. The exchange rulebook, the European disclosure document and the broker page each describe it differently, each correctly, and a trader reading one of them learns something the other two would qualify.
The only supervisory sentence on this feature answers a different question
On 3 July 2026 the European Securities and Markets Authority published public statement ESMA35-243228190-8148, two pages on whether event contracts fall inside the national product intervention measures on binary options. We took the classification half apart in our piece on the professionals only plan. Buried in its reasoning is the only thing a European supervisor has said in public about the coupon:
"In some cases, an investor may receive a 'coupon' or 'reward' representing the interest earned on the funds paid. The existence of such 'coupon' or 'reward' does not change the binary nature of the event contract itself."
Notice what that sentence is doing. It closes off an argument: that a product paying a running yield is no longer the fixed payout or nothing instrument the 2018 measures were written for. As far as we can find, nobody had made that argument in public. The supervisor saw the feature, decided it did not move the product between boxes, and moved on. It answers what the thing is and says nothing about how it is sold.
What the rulebook requires, and what it hands to the broker
The product side is better documented. In Notice to Members 2025-22, dated 14 April 2025 and effective on 28 April 2025, ForecastEx self-certified amendments to Rule 612 of its rulebook with the Commodity Futures Trading Commission under Commission Regulation 40.6(a). Rule 612(a) now requires the exchange to return all interest it earns on cash in customer and non-customer collateral accounts to its Members as a monthly coupon. Rule 612(c) then requires Members to pass a portion of that coupon to their customers, and sets the minimum daily rate as:
Coupon Rate = (EFFR minus 50bps) / 2
EFFR is the Effective Federal Funds Rate published by the Federal Reserve Bank of New York, and the notice adds that "Members may pay their Customers more than the minimum coupon rate."
Set that against the default: under Commission Regulation 1.29, a futures commission merchant or derivatives clearing organization that invests customer funds in permitted instruments is not precluded from "receiving and retaining as its own any incremental income or interest income resulting therefrom." In the ordinary futures world the interest on your collateral belongs to the intermediary and no rule makes it travel. Rule 612 is a venue writing itself an obligation the regulation does not impose, an exchange setting a floor above its own entitlement. What it does not do is describe how the number reaches the person trading. It sets a minimum and leaves the rest to the Member, and the Member is where the trader is.
Four published rates, and the formula lives on a different page
Follow the number through the documents and it changes at every step, legitimately, because it tracks a floating benchmark.
- The Federal Reserve Bank of New York published EFFR at 3.88 percent for 29 October 2026, the most recent reference rate available as we write on 2 November 2026. Put through the Rule 612(c) formula that gives a minimum of about 1.69 percent on the same annualised basis.
- Interactive Brokers announced the European expansion of Forecast Contracts in a press release dated 30 July 2025, which put the coupon at "3.83 percent APY" and noted that contracts "pay an interest-like incentive coupon based on the market value of open positions."
- The ForecastEx Key Information Document, dated 10 September 2025, uses 4.33 percent in its illustrated cost table.
- The Interactive Brokers prediction markets page, as it stood on 2 November 2026, states that interest "accrues daily at the current rate of USD 3.38% APY and is paid monthly" and that coupons are "based on the closing market value of the positions," with rates "subject to change."
Two of those figures line up. With EFFR at 3.88 percent, the quantity inside the rulebook formula, EFFR minus 50 basis points, is 3.38 percent, which is the figure on the broker page and exactly twice the rulebook minimum. On the published numbers the broker passes through the whole of EFFR minus 50 basis points rather than the half the rule obliges it to pass. That is better than the floor, arrived at by choice rather than by rule, and a trader cannot tell which of the two applies, because the floor and the formula sit in a notice to members while the rate sits on a marketing page. The Key Information Document puts the rest in six words: "Your distributor may deduct a spread from the Incentive Coupon." It does not say how much.
The base ESMA described is not the base the product uses
ESMA's sentence calls the coupon the interest earned "on the funds paid." The product documents describe something else. The Key Information Document says the investor "will earn income on the contract's daily settlement value based on the Effective Federal Funds Rate," and its footnote to the cost table adds that the coupon "is calculated daily on the notional value, which in turn will vary for a given number of contracts with fluctuations in price." The ForecastEx explanation of how Forecast Contracts work says each customer receives "an incentive coupon on the daily closing value of contract."
The funds you paid are a fixed number. The daily closing value of the position is not. Buy a No contract at 25 cents, watch the market move to 80 cents, and the coupon stops accruing on a quarter and starts accruing on four fifths. The accrual follows the mark, so it grows on the position that has already moved your way and shrinks on the one that has not. Both sides get paid: ForecastEx collects 1.01 dollars per matched pair and settles at 1.00, so the interest per pair is roughly fixed while the split between the two holders tracks the price.
None of that is hidden and none of it is wrong. It is simply not what a reader of the only supervisory sentence on the subject would picture, and it is the part a user needs to predict what lands in the account at month end.
The one standard form that exists files the yield under costs
Because Forecast Contracts reach European clients, ForecastEx produces a Key Information Document in the PRIIPs format, naming itself as PRIIP manufacturer and the Commodity Futures Trading Commission as competent authority. The template puts the product in risk class 7 of 7, the highest, then reaches the costs table. The illustration uses 20,000 contracts at 50 cents, a notional of 10,000 dollars, held to a 90 day expiration. Entry costs are 100 dollars of exchange fees. The ongoing costs row, headed "Management fees and other administrative or operating costs," reads "ForecastEx pays interest in the form of an incentive coupon payment" and carries a figure of minus 108.25 dollars. Total costs: minus 8.25 dollars.
The standardised European disclosure form for the highest risk class therefore reports, for this product, that the costs are negative. That is not sleight of hand by the manufacturer. The template has one row for ongoing costs, the coupon is the only recurring cash flow there is, and a form built on the assumption that holding a product costs money has nowhere to put a product that pays you to hold it. The same document carries both "Recommended (required minimum) holding period: None" and, three pages earlier, the statement that Forecast Contracts "incentivize long-term investment by distributing interest in the form of 'Incentive Coupons'." Both sentences are true. Together they are the gap.
It keeps paying while you are losing, and the venue says so plainly
The ForecastEx explainer makes the behavioural point better than any paraphrase. Its description of the basis of the contracts notes that even if the investor is wrong and the contract settles at zero, "they still receive coupon income over the life of the contract."
Take that sentence seriously rather than defensively. It describes a payment for elapsed time in a position, which continues while the position moves against the holder. The money is the holder's own interest coming back, so the intent is nothing like the gambling parallel, and the product is fully collateralised with no leverage, so the loss stays capped at the premium. The structural shape is nevertheless the same shape that deposit limits and cool off periods were invented to interrupt, and Directive 06 of our manifest, which expects revenue "from many informed participants over years, not a few ruined ones over months," is the right place to measure it. Nothing in that directive argues for removing the coupon. It argues for labelling it, because an accrual that rewards holding changes behaviour, and Directive 02 asks that such things be stated "on the surface where the trade happens."
On the US retail side the same feature arrives with a threshold
This is no longer one venue's quirk. Kalshi's help centre page on APY states a rate of 3.50 percent APY, variable and subject to change, and says users earn interest "on your cash and open positions, including the collateral behind your positions." Interest accrues daily on the portfolio's net value at midnight Eastern time and is paid monthly, on two conditions: the average monthly balance of cash and positions together must be at least 250 dollars, and the programmes reach only eligible users in the United States and its territories. The same page describes a higher tier for perpetual futures, where traders "can now earn 7% APY on the funds backing their open perps positions."
That last one is the cleanest case in the category: a yield that exists only while a leveraged position stays open. The 250 dollar floor is the clearest disclosure problem, because the promise is stated categorically, as a number on a page, while the payment is conditional. A trader with 180 dollars has read a true sentence and will receive nothing.
Whether any of this counts as an incentive programme is still open
On 12 August 2026 the Division of Market Oversight published CFTC Staff Letter 26-23, the advisory on self-certification of incentive programmes for prediction markets that we went through in our piece on bought volume. Its own description of the category is wide: incentive programmes, it says, are "intended to encourage increased market participation or trading volume in certain contracts by providing payments, credits, rebates, or other forms of compensation based on market participation or trading," and they reach the Commission through the same Rule 40.6(a) self-certification route. Footnote 15 recommended that exchanges revisit previously submitted programmes and file amendments by 14 September 2026.
Rule 612 travelled that route, and the manufacturer's own Key Information Document says the coupon incentivises longer holding. Yet the advisory never mentions interest pass-throughs or coupons, its worked examples are volume tiers and market maker arrangements, and we could not find any public statement, from the Commission or from a venue, placing a coupon of this kind inside or outside its scope. We are flagging that as an open question rather than answering it. What is not open is the disclosure hook: the Rule 612 filing itself invoked Core Principle 7, the availability of general information requirement at 7 U.S.C. section 7(d)(7) and 17 CFR 38.400, which asks that terms and conditions be public. A rulebook satisfies that. It is not the same as stated on the screen where somebody is deciding whether to hold.
The American rulebook for advertising a yield covers banks
One body of law tells an institution exactly how to compute and advertise an annual percentage yield, and it does not reach here. Regulation DD, 12 CFR part 1030, implements the Truth in Savings Act and exists, in its own words, to "enable consumers to make informed decisions about accounts at depository institutions." Its scope is depository institutions other than credit unions, and the advertising rules at section 1030.8 reach any person advertising an account offered by one. An APY quoted on an event contract position sits outside all of it. Nobody is breaking the uniform computation rule, because that rule was written for deposits. Four documents, four correct numbers, one floating benchmark, a spread permitted in writing without a figure, a threshold on one venue, and no standard for presenting any of it.
What the label would say if the category wrote it first
None of this needs a rule change and all of it could ship in a product sprint. A venue or broker paying a yield on open positions could publish, beside the position rather than in a monthly statement:
- The base, in words. Whether the accrual is on the cash paid, the closing value of the position or the collateral behind it, and whether it moves with the price.
- Gross and net, with the spread as a number. What the venue passes to the intermediary and what the intermediary passes on. "May deduct a spread" is a permission, not a disclosure.
- The floor beside the rate. Where a rulebook sets a minimum, print the minimum next to what is actually being paid, so a rate cut is visible as a choice.
- Every condition at the point of the promise. Minimum balances, eligible jurisdictions and account types belong in the sentence that states the rate, not two clicks away.
- When accrual stops. At settlement, on an offsetting position, on withdrawal, and what a part month pays.
- A gross figure wherever the yield is netted. If a cost table subtracts the coupon from fees, both numbers have to stay readable alone.
A yield paid on a losing position is not a scandal. It is the first feature this category has invented that pays for patience, which in a market built on information is arguably the most defensible thing on the screen. That is exactly why it deserves a label written by the people who built it, before somebody who did not writes a worse one for the whole category.