Convergence of crypto perpetual futures

Question

Perpetual contracts are supposed to track the spot prices through the funding mechanism. Typically, if the future has traded above the spot in the last averaging period used to compute the funding, then the funding will be positive, and the long will pay the short exactly at the end of the averaging period. It is enough to only hold the asset at the very end of the period to receive the cash (like dividends on an ex date). I understand that this is to discourage investors to be long the future. But this does not make sense to me from an arbitrage perspective.

Indeed, to prevent the following arbitrage around the funding payment:

• shorting the future just before the funding payment,
• taking the funding (almost precisely known due to the averaging period),
• buying back the future,

it is required that the price of the perpetual moves UP (opposite direction to the convergence since we supposed perp>spot) by the amount of the funding. Same phenomenon with dividends: the spots decrease by the value of the dividends when going through the ex-date.

So that's my point: the required price correction that should arise from this payment is opposite to what I was expecting. It looks to me that it should create a divergence in price.

Obviously it's not the case, but what am I missing?

Answer 1

In theory there seems to be an arbitrage. But in practice if you want to build a model to benefit from it, you should take into account the following factors:

• The rate is small so your position size should be relatively large
• A positive funding rate means the price is going up, but you are taking a short position which is opposite to the market. The price is already moving against your position.
• Fees
• Slippage

Funding rates are high during the volatile markets, but so is slippage.