The arithmetic of hedging with perps

It appears that the CFTC is weeks away from bringing perpetual futures to US onshore exchanges, en masse. Every institution we speak to hedges with CME. Not one uses perps. The reason is arithmetic.

Funding rate drag across Binance, dYdX, Hyperliquid, GMX, Bybit runs 8-22% annualised depending on venue and position size. That is the explicit cost. It does not include bid-ask friction (2-8bp per leg on thin venues versus 0.5bp at CME), implied borrowing costs (8-15% versus 5.5% Fed rates), or persistent basis bleed (1-3% on crypto perps versus less than 0.5% at CME).

Add them up. 12-30% annual hedging cost on a perpetual future. CME quarterly futures: 2-3%.

Perps were not designed for hedging. They were designed for leveraged speculation. Using them for institutional risk reduction is an expensive structural mismatch. Now that regulated perps are coming to the US, the question is whether the cost structure changes or whether the same funding mechanics follow. I am not seeing any reason they will. They haven’t changed materially in design since 2015, when we launched them on Eurex as CMFs. I think these were a little better in roll cost calculation than the current version, but still suffered the same issue and therefore lower adoption.

If you’re running a book that hedges with perps, DM me. We modelled venue-specific cost breakdowns across five exchanges and will share the thresholds. It will come as a repo.

This first appeared on LinkedIn on 30 March 2026. If you want to comment or discuss, that’s the place.