Why does Polymarket and the Fed Funds futures disagree? Right now, Polymarket places the odds of a Fed hike at 52% while the FF futures trade at a 58% breakeven rate. The Fed Funds futures currently imply a 58% chance of a 25 bps rate hike in September. We know these odds do not reflect the true probability of a hike because they include a risk premium. The risk premium exists because in the aggregate we are long duration risk and a rate hike will therefore lower aggregate wealth. Buying a Fed Funds Futures contract provides insurance against the loss of wealth from a rate hike and in equilibrium that insurance must be paid for in the form of higher break-even probabilities. I explain this risk premium in this blog post from just before the last FOMC meeting: How much is the Fed Funds contract overstating the true rate hike probability? We don't know exactly but we can form an educated guess. Fed research finds the premium is about 1 bps per month on average. This translates into about 1.5 percentage points of probability for the meeting in 12 days. However, that's the long run average risk premium and the premium does vary with uncertainty about the fed path and the sensitivity of the long end to Fed policy surprises. The path is both very uncertain right now and investors are telling us they fear this uncertainty by demanding a large term premium on the 10-year bond. The skew of SOFR options and swaptions also imply a much larger than normal risk premium. This is where the basis between polymarket and the FF futures can help us size the risk premium. Obviously, these market's aren't perfectly integrated. The arb trade of buying Poly and selling FF pays 6 bps but it exposes the trader to counterparty risk with poly and requires margin and transaction costs on both ends. Furthermore, the liquidity in poly is probably too small to let anyone with a futures account arb in size. I assume the segmentation between poly and the CME contracts also breaks down along duration exposure lines. Those with high duration risk look to the CME to hedge and are willing to pay the risk premium to do so, while those trading in poly are more likely to be less exposed to interest rate risk and trading when they think the price deviates from the true probability. If these assumptions are true, we can bound the risk premium at 6 percentage points (all of the basis). Of course, if you think poly traders are just uninformed then the basis can reflect both the risk premium and noise trader bias.