Treasury Repo

We've covered what Treasury repo is and how it works, both in text and in a video series, but it's such a crucial piece to the modern financial system that it can't be reiterated enough. Fortunately, it's rather simple.

A repo (short for repurchase) transaction involves the sale of assets (collateral) with an agreement to buy them back (repurchase) on a specified future date (usually the next day) and at a prearranged price. By selling your collateral and agreeing to repurchase it, you are just borrowing cash. There's no need to make it any more confusing than that. The pawn shop analogy is accurate:

For example, at a pawn shop you may "sell" $1000 worth of gold (ex: a gold ring), and accept $950. The reason the pawn shop "dealer" would be willing to lend only $950 for $1000 in collateral is because he knows that gold has been quite volatile lately, and if gold's price were to fall 5% (the collateral value falls to $950) or more, he may lose money on the transaction. The 5% "premium" is called a haircut, and is a layer of insurance.

At the end of the transaction, which in repo is usually the next day, you would purchase the gold back from the dealer for $950 (plus any interest). You could also choose to extend the loan ("roll over the repo").

But in reality, you're not selling the gold item per se, because you're agreeing to repurchase the item back from the dealer at the pawn shop. If you fail to do so, the dealer keeps the collateral — hence it is "secured lending."

While "gold repo" is in theory totally possible, it serves better only as an example. Treasury repo, where Treasury bills/notes/bonds are the collateral, is a daily operation of several trillion dollars, with unreported venues comprising another estimated ~$2 trillion in daily volumes.

The exhibit above is a very basic model of the dollar flows in repo. The collateral (Treasury securities) flow in the opposite direction of cash, obviously. And while there are four major repo venues (Tri-Party repo, GCF & DVP interdealer repo, and uncleared bilateral repo), the only differences are a third-party custodian holding the collateral (in tri-party repo) and whether the transaction is cleared by the FICC (in interdealer repo).

But at the end of the day, a repo is mechanically identical across all venues.

2/6

The Cash-Futures Basis Trade

In early 2018, a string of events formed an exploit in America’s sovereign debt market. Following a surge in Treasury issuance and regulatory reforms, asset managers (pension funds, mutual funds, and insurance companies) began to shift out of cash bonds and into long positions of their associated Treasury futures contracts.

A Treasury futures contract is a standardized agreement to buy Treasury securities at a predetermined price on a specified future date. Unlike option derivatives, which provide a right but do not obligate, futures contracts involve a binding obligation to transact on or before the contract matures.

The reason the basis (the difference between the cash Treasury price and the Treasury futures price) exists is because said asset managers began piling into Treasury futures, thus raising the futures' price relative to cash Treasuries. They prefer the futures over cash Treasuries because futures are operationally simpler and have less impact on their expense ratios. Most asset managers are not set up for repo, for example.

The cash-futures basis trade, or simply "the basis trade", is a three-legged arbitrage trade that seeks to exploit the basis, spanning three crucial financial markets: the cash Treasury market, where investors purchase Treasuries today; the Treasury futures market, where investors agree on a fixed price to pay for Treasuries they will receive in the future; and the Treasury repo market, where investors leverage their cash Treasury purchases.

Basis trades buy cash Treasuries and short Treasury futures to construct a payoff that depends on the two prices converging as the delivery date approaches (see Figure 2 in the third image).

This is similar to a long/short equity strategy, and convergence is virtually guaranteed: at the delivery date, cash and futures prices must be equal because otherwise on that date a trader could just buy a Treasury in the cash market and immediately deliver it into the futures market for an instant profit.

Shorting a Treasury futures contract means entering into an agreement to sell the underlying Treasury at a future date and at a predetermined price. It is an obligation.

To "deliver on a futures contract" means to fulfill that obligation by transferring the underlying Treasury to the buyer on the expiration date.

The key is that, so long as futures prices keep rising markedly above the price of their underlying Treasury securities, traders would buy bonds at a discount to what they’d receive when delivering these securities into futures contracts. If the basis were to narrow (or, potentially, invert), the trade would no longer be profitable, and this marginal buyer for Treasuries would vanish.

Only certain futures contracts and Treasury securities are used in basis trading. On any given date, there is just one Treasury security that basis traders want to own for each contract to make a particular deal as profitable as possible, called the “cheapest-to-deliver” Treasury.

The CTD ("cheapest-to-deliver") is simply the Treasury security with the cheapest value that is eligible to be delivered onto a futures contract.

Otherwise-similar Treasuries will differ in whether they are deliverable into a futures contract. A conversion factor attached to the futures price is meant to account for the desirability of individual Treasuries (the CME provides updates on conversion factors).

Due to these conversion factors, only one Treasury will be cheapest-to-deliver into each futures contract. But which that is can change over the life of a contract.

4/6

Implementing Basis Trades

Understanding sources of risk for basis trades and where stress can manifest requires understanding the technical details of how these trades are implemented.

The basis — or the profitability of a cash-futures basis trade — is characterized by the implied repo rate (IRR), which reflects the cost of carrying the security (including financing costs) until the futures contract's expiration.

When the implied repo rate is greater than the actual repo rate, basis traders borrowing in the repo market can profit by buying the cash Treasury and shorting the corresponding futures. At delivery, the trader will earn the spread between the IRR and the repo rate. When the actual repo rate is greater than the implied rate, a "long basis" trade is not profitable.

The IRR is closely related to the yield on a Treasury bill because the cash flows from the basis trade replicate those from a Treasury bill maturing on the futures delivery date.

In particular, in times of relative illiquidity and high balance sheet costs, the implied repo rate has deviated significantly from the rate of return on bills.

One example of these deviations occurred following the collapse of Lehman Brothers in 2008 (see Figure 4 above). Immediately after that collapse, as liquidity dried up in financial markets, implied repo rates collapsed deeply negative across contracts. The IRR decline reflected a flight to safety in Treasury markets.

Because the futures price and the cash price of the Treasury are known to the basis trader, provided he also knows the repo rate, profits on these bets at delivery are guaranteed.

The basis trade does not, however, offer risk-free profits. Several risks threaten the profitability of the basis trade, and thus create potential consequences for financial stability.

5/6

The Greatest Credit Event of All

In just the last few weeks, we've seen two major trades — the Yen carry trade and the short volatility trade — "blow up", even though neither lasted nor took the street down with it. What about the basis trade?

We alluded at the start of the year that the "mother of all credit events" would be a disorderly rise in bond yields leading to dollar debasement. This is the "de-dollarization" that keeps Treasury and Federal Reserve officials up at night — not means of trade being re-routed off the dollar and onto other forms of settlement.

A credit event occurs when a borrower can no longer meet debt obligations, leading to a default, bankruptcy, or restructuring. For example, an insolvent bank being unable to pay depositors in a bank run.

There are two legs to the basis trade: the asset managers who express Treasury exposure via buying futures, and the hedge funds who repo finance cash Treasury purchases. While hedge funds are the marginal buyers of cash Treasuries, it's the asset managers who buy Treasury futures that ultimately hold the risk.

The marginal absorber of cash Treasuries remains the basis trade, but it can run into limits either from a regulatory crack down on hedge funds or limitations on dealer repo financing. New research from TBAC suggests that changes in the credit environment could also threaten the trade.

What asset managers do is effectively take the opposite side of the basis trade by selling cash Treasuries outright and using the proceeds to finance their credit investments. They then add back the Treasury exposure through futures.

That positioning also more directly links the Treasury & credit markets together, as potential losses on credit may lead to deleveraging in Treasury positions as managers de-gross.

It could also lead to liquidity squeezes as managers sell Treasuries to meet redemptions, as credit markets may not be liquid to enough to raise cash. Outside of a systemic or apparent credit event, note that credit spreads have spiked this month.

In the wake of such a risk-off credit event (think: a bank failure), monetary authorities are limited to only a few modes of easing. But the rise of foreign, non-official, unhedged accounts as the marginal buyer for Treasuries means that they are particularly vulnerable to dollar devaluation that results from a policy of easing.

Aggressive rate cuts in the name of providing economic support for example may therefore paradoxically be ill advised insofar as it weakens the dollar against other currencies (like the euro), because said non-official accounts would likely firesale their Treasury holdings as they try to avoid realizing losses, spiking repo financing costs and repo rates as dealer warehousing capacity is pushed to the edge.

That was the case in March 2020, when risk-off paradoxically led to a spike in yields, as Treasury holders aggressively sold their securities for cash. Although this time, it may be to preserve foreign capital against unhedged Treasury losses.

(Disclaimer: this is a theoretical but plausible example).

6/6