The Bond Market’s Problems Aren’t All to Do with Donald Trump

There can be little doubt that the 90-day pause on tariffs for most countries that Donald Trump announced abruptly on April 9th was driven by worries about the bond market. A jump in yields in America’s Treasury market after the tariffs were first unveiled had, even the president conceded, made people “queasy”. The heavy sell-off in Treasuries came amid weak demand at auction, high price volatility and even talk of “fire sales” in the market.

It is hard to think of anything more alarming for global financial markets than the prospect of Treasuries, long seen as the risk-free asset to which investors flock in turbulent times, losing their status as the ultimate haven. If they are not perceived to be safe, what is? What would the loss of their gold-plated reputation mean for the borrowing costs of the American government, whose interest payments on $36trn of Treasuries already consume about a fifth of its tax revenues? And what about all those households and businesses whose debt costs are tied to the interest rate on Treasuries?

The last time the Treasury market faced such a constellation of factors was in the pandemic of 2020. To stabilise the market then, the Federal Reserve bought massive quantities of Treasuries, including $362bn-worth in a single week.

From the Fed’s perspective things may be even more complicated this time, since in addition to the obvious downside risks to growth and employment there is now also the risk of spiking inflation. In this context, initiating an enormous purchase programme would probably be perceived as an extremely dovish monetary-policy signal—which the Fed would want to avoid. Is there another option?

Our research suggests that part of the disorderly selling may be due to a deeper source of fragility in the way the Treasury market operates. The story starts with the observation that pension funds, life-insurance companies and corporate-bond funds (which we collectively call “asset managers”) are all heavy users of Treasury derivatives such as futures and interest-rate swaps. They take long positions in these derivatives along with shorter-term corporate debt in order to establish an asset mix that has both some extra yield (from the corporate debt) and a maturity profile that suits them. If they could, they would just buy long-term corporate debt, but there isn’t enough of it. So they use derivatives as a substitute. Lots of them: we estimate that they had about $1trn invested in long positions in Treasury futures alone earlier this month.

For the Treasury derivatives market to clear, someone has to take the other side of the trade. That someone is hedge funds, which leaves them with a short position in futures. The hedge funds, however, do not want to make an outright bet on interest rates. So, in addition to going short futures, they buy cash Treasury securities. They can earn a premium on this trade, known as the “basis” trade, but not a very big one. So, to make the trade more profitable, they juice it up with leverage, borrowing to buy the Treasuries. They might borrow, say, $49 for every dollar they invest.

The hedge funds’ need for leverage brings in the last two actors in the story: money-market funds and the Treasury market’s “primary” dealers (Goldman Sachs, JPMorgan and the like). The money-market funds typically lend to the dealers, who lend the money on to the hedge funds.

This ecosystem has several weak points. One is its reliance on the money-market funds, which can lose their appetite to lend, for instance if they are facing withdrawals during market turmoil. The second risk is the dealers, who may scale back lending to hedge funds, even if they have enough to lend, in order to reduce risk.

The third and probably biggest risk comes from the hedge funds. When volatility rises, they may step out of their highly leveraged trades to protect against losses. That may be prudent risk management for any single hedge fund, but if a lot of them do it at once prices need to fall to tempt others to take their place. There are some signs that this kind of behaviour contributed to the latest turmoil triggered by Mr Trump’s tariffs.

Another destabilising factor is that when volatility increases, futures exchanges increase the margin requirements—to post collateral—that they impose on the hedge funds. Such margin calls may also force hedge funds to unwind basis trades they would rather stay in, forcing them to sell their cash Treasuries, even at fire-sale prices.

The situation over the past week does not seem to have been as dire as in March 2020. But the market remains edgy. If things were to deteriorate further, and the Fed had to intervene again, how should it act?

The problem with buying Treasuries outright, as the Fed has done in the past, is that it can look like “quantitative easing”—an easing of monetary policy through purchases of government bonds and other assets—which is not the signal it would want to send right now. Instead, it should set up a special-purpose vehicle (SPV), perhaps in collaboration with the Treasury Department, that would take over the hedged long-Treasuries-plus-short-futures bundle that the hedge funds would be dumping.

This would have three advantages over unhedged outright purchases. First, it would make it clear that this is an action designed to address market function, not an easing of monetary policy. Second, the Fed would only lend on a collateralised basis to the SPV and the SPV would own the hedged bundle, which would be immune to interest-rate changes. The taxpayer would therefore not be exposed to any losses from such changes, and the SPV would self-liquidate as conditions normalised and the futures contracts matured. Finally, the purchases could be made via an auction, which would put a lower-than-normal price on the long-Treasuries-short-futures bundle. This would ensure that hedge funds that had managed risk poorly and had to rely on the SPV suffered some losses, thereby mitigating moral hazard.

A century and a half ago, in “Lombard Street”, Walter Bagehot, then the editor of The Economist, laid out the terms on which a central bank should act, as the “lender of last resort”, to stabilise the financial system in times of stress. The Treasury market of today—a focal point of a much more complicated financial landscape—requires its own version of the backstop. One commentator has dubbed our proposal the “synthetic lender of last resort”. Whatever you call it, a new approach is needed to deal with the dysfunction at the heart of the world’s most important financial market.

Anil K Kashyap is the Stevens Distinguished Service Professor of Economics and Finance at the University of Chicago’s Booth School of Business. From 2016 to 2022 he was an external member of the Bank of England’s Financial Policy Committee. Jeremy C. Stein is the Moise Y. Safra Professor of Economics at Harvard University. From 2012 to 2014 he was a member of the Federal Reserve’s Board of Governors. Their research on the Treasury market was conducted with Jonathan Wallen and Joshua Younger.

© The Economist Newspaper Limited, London (April 11, 2025)