The Cash-Futures Basis Trade and the Leverage Underneath It

Treasury futures and the underlying cash bonds are close substitutes, and their prices are linked by arbitrage. When futures trade rich relative to the deliverable bond, the trade is to buy the cash bond and sell the future, capturing the difference as the two converge at delivery. The spread is typically measured in a few basis points.

A few basis points is not a business. It becomes a business through leverage. The cash bond is financed in the repo market, and because the position is hedged, the risk appears small, which allows very high leverage against it. The trade is therefore not really a bet on the spread. It is a bet that the financing remains available at a stable rate for the life of the position.

Where the risk actually sits

The convergence itself is close to certain, since delivery mechanics enforce it. The risk is entirely in the financing and the margin. If repo rates spike, the cost of carrying the position exceeds the spread being captured, and the trade becomes loss-making despite the underlying relationship performing exactly as expected. If futures margin requirements rise following a volatility event, the position requires additional capital that may not be available.

Either condition forces an unwind. An unwind means selling cash Treasuries and buying back futures, simultaneously, by many participants holding similar positions. The selling hits the cash market at precisely the moment when dealer balance sheet capacity to absorb it is most constrained, which converts a financing problem into a market functioning problem.

Why this matters beyond the participants

The Treasury market is the collateral foundation of the global financial system, and its functioning is a public good in a way that an equity market functioning is not. A leveraged trade that is individually rational and collectively capable of impairing that market poses a supervisory question rather than merely a risk-management one, which is why the size of these positions is now monitored and publicly discussed by financial stability authorities.

For an investor with no involvement in the trade, the relevant implication is that Treasury market liquidity is not a constant. It depends partly on the balance sheet capacity and positioning of leveraged intermediaries, which means the safest asset in the system can become temporarily difficult to transact in size for reasons that have nothing to do with credit.

Reading the Position Data

Aggregate positioning is partially observable through the futures commitments data, where the leveraged funds category holds a large and persistent short in Treasury futures that is generally interpreted as the futures leg of this trade. The corresponding long in cash bonds is not directly observable, which means the inference requires the assumption that the short is hedged rather than directional.

The assumption is reasonable given the size and persistence of the position, since an unhedged directional short of that magnitude would have produced losses inconsistent with the funds continuing to hold it. The useful reading is directional rather than precise: a growing leveraged short in futures alongside stable repo conditions indicates the trade is being put on, and a rapid reduction alongside repo stress indicates it is coming off in a hurry.

The 2020 dislocation and what it revealed

In March 2020 the cash-futures basis widened sharply as leveraged holders began unwinding positions to meet margin calls. The unwinds were forced and simultaneous, which meant enormous quantities of cash Treasuries hit the market at the moment when dealer balance sheets had least capacity to absorb them. Treasury market liquidity, which is normally taken as the deepest and most reliable in the world, deteriorated to the point where the Federal Reserve intervened with direct purchases to restore market functioning.

The episode demonstrated that the collateral foundation of the global financial system can become temporarily impaired for reasons that have nothing to do with credit quality. That vulnerability has structural rather than transitory character, since the leveraged intermediation the trade depends on is now a permanent feature of the market microstructure, and the conditions that produced the 2020 unwind can recur.

Why individual investors care about a strategy they never trade

An Australian household allocation, an SMSF portfolio, or a family office does not run leveraged basis trades. The relevance of the trade to those investors is that the safest asset in the system, the security most portfolios hold as their defensive anchor, has a functioning liquidity risk that is separate from any credit risk. During a repeat of the 2020 episode, an investor who needs to sell Treasuries or bond funds to raise cash could face wider bid-ask spreads and materially worse execution than the historical distribution suggests. Sizing a defensive allocation on the assumption of perfect liquidity in the underlying market is a specific case of assuming a stationary market microstructure. The microstructure is not stationary, and the deviation shows up at exactly the moment liquidity is needed most.