“The bond market is very tricky, I was watching it,” Trump told reporters. “I saw last night where people were getting a little queasy.” – CNN (April 9, 2025)
“Bond vigilantes were screaming” – Ed Yardeni (April 11)
“There is still trouble in the bond market.” – The Wall Street Journal (May 11)
[Like the previous installment, this column has become perhaps too long. So again I offer an executive summary for busier readers. 🤔]
Executive summary
The Treasury Bond market went into convulsions last month following the “Liberation Day” announcement of broad new high-tariff policies (April 2). Because Treasurys play such an important role in the global financial system, it seemed briefly that the whole system might be on the verge of coming apart. The “queasiness” in the Treasury market was cited as the reason that tariff proposals were quickly put on hold (April 9). The markets recovered, and it seemed that the relationship between Cause (tariffs) and Effect (bond market turmoil) had been clearly demonstrated.
This is likely a false conclusion. For two reasons.
The first is scale. What was it – substantively – about the prospect of tariffs that so unsettled “the deepest and most liquid financial market in the world”? A market worth $29 Trillion, equal to 98% of U.S. GDP?
Initial assessments focused on the presumed negative impact of a tariff “tax” on consumers, with the heightened possibility of a recession or inflation or both. But these analyses, however superficially plausible, become unconvincing on closer examination, for one simple reason: the projected impact on either inflation or growth is quite small relative to the economy (as detailed in the previous column). The Effect comes to seem disproportionate to the presumed Cause.
Secondly, for several years the Treasury market has been behaving strangely, for reasons that have nothing to do with tariffs or trade policy.
There are five important Treasury market anomalies to consider.
The first three have to do with the yield curve, which describes the structure of interest rates (yields) in the Treasury market. An inverted yield curve is a sign of abnormal (and even illogical) market conditions where long-term Treasury debt pays less interest than short-term debt. Why should an investor receive less interest on a 30-year bond, exposed to many forms of risk over a very long period, than on a 90-day Treasury bill which for all practical purposes carries no risk whatsoever? But it happens and when it does an inversion is considered a very significant economic indicator, and an almost perfect predictor of recessions.
The anomalies are these.
- The Treasury market experienced its longest and deepest yield curve inversion ever – from July 2022 to September 2024 – three and a half times longer than the average of the four major inversions since 1982.
- Unlike all previous yield curve inversions for the past 60 years, this mega-inversion did not lead to or predict a recession. The economy boomed, and the “perfect indicator” failed.
- This inversion was very tight sync’ed with specific reversals (changes in sign, from loose to tight, or vice versa) in Federal Reserve monetary policy – which had not happened in previous inversions.
- Last September, the Treasury market broke a decades-long close relationship between the Fed’s official interest rates and yields on Treasury bonds. The relationship went from being almost perfectly positively correlated to almost perfectly negatively correlated. The implication is that the Fed’s interest rate policy lever may have ceased to function. Some say the Fed is now in “policy paralysis.”
- In the episode of turmoil in early April, the Treasury bond market deviated from the usual crisis-response pattern. The “normal” flight to safety response was interrupted and brutally reversed by a massive bond sell-off – which seems to have been the result of organic changes in the market that have developed recently, and which created the pre-conditions for a new kind of market crisis. Complex trading strategies like swap-spreads and “basis trades” have changed the tenor of the market in the direction of greater speculation, increased leverage, rising volatility, sometimes intensely crowded trades, and heightened sensitivity to external shocks – shocks such as a new tariff policy announcement.
In short, it appears more likely that tariff talk was a trigger rather than a cause.
A discarded cigarette may start a fire that burns down the house. But if the garage where the cigarette landed was full of old newspapers and gasoline-soaked rags, a full understanding of the cause of the fire would have to be much more comprehensive. It would appear that the Treasury market has been crammed full of gasoline-soaked rags – preconditions for the flare-up, which had nothing to do with tariff policy.
The Liberation Day gambit was a minor and transitory phenomenon compared to the vast scale and importance of the Treasury market. It may have been a trigger, but in the big picture, it is the somewhat unhealthy evolution of the Treasury bond eco-system which should command our attention. The bond vigilantes of 2025 may be a rougher crowd than their predecessors in the 1990s. The current concerns over the deficit may give fresh indications of this potential for violence. Meanwhile, tariffs per se are more or less a sideshow.
The Tariff Storm of April 2025
The “Liberation Day” tariff shock set the stock market ringing like a bell. It precipitated the worst two-day decline in history, with the Dow down 4000 points (4/2 to 4/4) – followed a few days later by a 3000 point gain (4/8 to 4/9), and then more back and forth. Volatility exploded, spiking 10 standard deviations above the long-term average.
But experienced investors know that the stock market can be “emotional” – occasional turbulence is to be expected and endured. The real danger signal in early April emanated from the bond market, which is supposed to be more sober and predictable.
The 10-Year Treasury Bond yield rose 64 basis points in two days – “one of the biggest two-day increases on record.” Rising bond yields mean falling bond prices, and the losses were savage. Treasurys lost more than 5% in value in one week. Bond market volatility, as measured by the MOVE Index, surged more than 3 standard deviations above its recent average, reaching the highest level since the pandemic shock in March 2020.
Technical symptoms of bond market stress appeared. Along with rising volatility, there was a “significant” deterioration in liquidity.
A surge in selling overwhelmed market-makers. Buyers pulled back. The “depth” of the order book dropped 75%.
Bid-ask spreads expanded. Price discovery became difficult. Intraday trading ranges – the difference between high and low trades of the day – widened. On April 7, the 10-year Treasury traded over a range of 35 basis points, four times above the long-term average – “the wildest [intraday] swing for 10-year yields in more than two decades.”
Contagion spread.
“Heightened volatility in the US Treasury market has spilled over to global bond markets, increasing volatility there as well. For example, yields on two-year German Bunds jumped by twenty basis points before subsiding. At one point in the past week, ten-year Japanese government bonds rose by thirteen basis points. These are unusually large rises for these government bond markets.” – The Atlantic Council (April 11)
The turbulence stirred fears that the Treasury market could seize up, as it did following the pandemic shock in March 2020. The situation was “potentially catastrophic” (quoth CNN). The market deviated from the established crisis-response script. Normally, in moments of market stress, investors dump equities and seek safe haven in bonds, driving bond prices up and yields down – known as the Risk-Off scenario. This happened at first, but then it un-happened. Bond prices fell – investors were dumping Treasury bonds, supposedly the safest of safe assets. The dollar weakened, also contrary to the normal Risk-Off scenario.
There was speculation that the Federal Reserve might have to intervene. The outlook seemed suddenly dire, “a seismic shift in the global financial order.” Some saw it as a loss of confidence in U.S. fiscal management and the reliability of U.S. government debt, in U.S. assets generally, the dollar itself, and even a threat to national security.
“The searing sell-off in Treasuries this week in response to tariffs caused dislocations in the world’s biggest bond market” – Reuters
“The dollar as the world’s global reserve currency is undergoing a process of rapid de-dollarisation.” (CNBC, April 11)
“Trump’s Tariffs Upend the Bond Market: The bedrock of the financial systems trembled…The Treasury market’s erratic behavior has raise fears that investors are turning against U.S. assets.” – The New York Times
“The market has lost faith in U.S. assets.” – Deutsche Bank analysts
“We have damaged the U.S. brand.” – Mark Rowan, CEO of Apollo
“Confidence in the US economy is plummeting as investors dumped government debt amid growing concerns over the impact of Donald Trump’s tariffs.” – BBC
“…threatening the dollar’s supremacy and revealing deeper vulnerabilities in the US macroeconomic outlook.” – London School of Economics
“… undermining U.S. economic security.” – Center for Strategic and International Studies
Normalcy Returns
Then – as quickly as it came on, the storm blew over.
The Vix returned to normal. The stock market recovered its losses rapidly, gaining 21% (as of May 16) from the April 8 low. The Treasury bond market stabilized. Daily trading volatility returned to normal. The yield on the 10-year Treasury fluctuated on average just 4 basis points (bps) per day. No move exceeded 9 bps, and there was no real directional trend. The 10-year yield remained well below its January high, about even with its pre-QE level. Even the dollar was up 3%, above its 5-year and 20-year averages. The MOVE Index was back to its pre-April 1 average. The normal economic indicators – jobs, inflation, retail spending, manufacturing activity – are coming in with generally positive readings. Corporate earnings have been strong. The Fed reported that “economic activity has continued to expand at a solid pace.”
And yet… despite the appearance of a return to normal, the Bond market remained jumpy. Tariffs seem to be still implicated (“Tariff Shock Reverberates in the Bond Market” – WSJ headline, May 11).
[Late Update: Market “queasiness” has returned in the last week driven by concerns about the Budget Bill. Yields have risen and the dollar has declined moderately. But that is another story. Tariffs meanwhile had been drifting towards the “old news” category — but Trump’s announcement in May 23 of potentially higher tariffs on the EU helped return the topic to the market’s front burner.]
All this uncertainty, seemingly tariff-related, has confused policy-makers, unsettled investors, and soured public opinion.
The Question of Proportionality
Aristotle understood that causality is a complicated concept, and identified four different types of causes. And it is easy to say that a particular event may have many causes. Which to pay most attention to?
Proportionality holds that the most important cause will generally be of the same scale as the effect. It is an intuitive principle that can help identify the most pragmatically significant causal factor.
The real impact of tariffs – if they are actually implemented – would not likely be large enough to unhinge the massive Treasury bond market. The imposition of tariffs would amount to a scattering of one-time taxes across the economy (a “transitory” phenomenon, as Fed Chairman Powell has described it). Short of an all-out trade war, which seems not to be the endgame here, tariffs actually play a limited role in the economy. The Tax Foundation – a nonpartisan think tank – calculated that the 2018 tariffs resulted in tariff duties of about $200-300 per household annually– about 0.3% of the median household income. Another oft-cited study of the 2018 tariffs by the Bureau of Economic Research found that “the aggregate real income loss was $7.2 billion, or 0.04% of GDP.”
Multiply that by several times to gauge the potential impact of higher rates proposed today, and the impact is still small. The Peterson Institute estimated that Liberation Day tariffs would cost consumers about $1200-1500 per household. That would be $150-200 Bn across 130 million households – or about 0.5% of GDP and about 1.5% of the median household income. The Penn Wharton Budget model forecast a similar reduction of about 0.4% of GDP in 2030. The Yale Budget Lab’s maximalist estimate of the impact of the tariffs included in the Liberation Day proposal was higher – $4800 per household, or about 2.1% of GDP. (Yale recently updated this to $2800 per household, or 1.2% of GDP.)
These numbers – if they turn out to be accurate – would not be insignificant, but they are quite small relative to the size of the $30 Tn U.S. economy.
What is not small relative to the economy is the Treasury bond market, currently valued at 95% of U.S. GDP.
The Treasury Bond market is an awesome thing. It is the deepest, the most liquid, the most efficient financial market in the world. It is often described as the “cornerstone” of the global financial system.
It embodies an immense power to determine the price of credit everywhere. Treasurys set the rate of return on trillions of dollars of bonds held in the reserves of all major central banks (including China), and trillions more in the portfolios of investors here and abroad – almost $29 Tn in all. The market trades more than $1 Tn per day – much more than the dollar volume of trading on the U.S. stock exchanges. Treasury rates influence borrowing costs throughout the economy, from consumer products like mortgages (98% correlation), auto loans (95%), and other consumer loans (97%), to U.S. corporate bonds (95-98%), even European bonds (94%). Treasurys are the foundation of America’s balance sheet. All questions related to financing the annual federal deficit ($1.9 Tn projected this year) and the cost of servicing the nation’s debt ($952 Bn on $36 Tn in debt) lead back to the Treasury market.
In perspective, a small transitory impact on consumer purchasing power – even if it materialized fully (which would be unlikely, since consumers and producers would take steps to mitigate the tariff taxes – as described in the previous column) is not the logical culprit to have caused the multi-trillion dollar up and down re-valuations in the financial markets that were seen in April.
On the other hand, there is something strange going on in the Treasury market. There are major anomalies that are far more significant than even the worst projections of the tariff impact.
A Parade of Treasury Market Anomalies
Anomaly 1: A Record Inversion
The yield curve is a classical financial market indicator. It charts the yield on Treasurys of various maturities – from 30 days to 30 years. Typically, the longer the maturity, the higher the rate – which is “logical.” The longer the term of the bond, the more the bond holder is exposed to risks of all sorts, which should require a higher rate of interest to compensate.
Sometimes, the Treasury market becomes “illogical” – and short-term bonds will pay more than long-term bonds. The curve is said to be inverted. Inversions are also viewed with alarm by investors, because they are reliable predictors of a forthcoming recession.
On April 1, 2022, the yield curve inverted, but only for two days. Then, on July 6, it inverted again – and the Treasury market remained in a state of inversion for the next 791 days, until September 2024. (Keep those dates in mind)
This was the longest yield curve inversion on record, three times longer than the average of major inversions since 1982.
The inversion of the 10-Year vs the 30-day Treasury note from 2022 to 2024 was even more pronounced, deeper, and just as long lasting. For more than two years, an investor could earn more owning a 3 month Treasury bill than from the 30-year Treasury bond. (Which is a bit crazy, no matter what the explanation. A 3 month T-bill carries almost no risk of any kind, whereas a 30-year Treasury is exposed to inflation risk, interest rate risk, and perhaps other risk factors.)
Anomaly 2: The Missing Recession
Inversions of the 10-Year/3-month spread have preceded every U.S. recession in history (except for one false positive in 1966). Inversions of the 10-Year/2-Year spread have a perfect record predicting recessions since the 1970s.
Generally, the longer the inversion lasts, the more severe the associated recession. An inversion as deep and long as the 2022-2024 episode should have been a powerful recession signal. Yet no recession has occurred.
Eventually – inevitably – a recession will occur, but the huge 2022-2024 inversion will not have predicted it by the traditional rules of interpretation.
Anomaly 3: Fed Policy Shifts Precisely Aligned with the Inversion
The initial tremor in the Treasury market was the brief two-day inversion on April 1, 2022. This occurred 2 weeks after the Federal Reserve announced the first rate increase in the recent cycle, of 25 basis points, on March 17.
The onset of the long-lasting inversion episode on July 6, 2022 followed by 3 weeks the rare “jumbo” rate increase of 75 basis points on June 16 – the first increase of that size in 28 years (since November 1994) – a strong market signal.
The alignment with the exit from the long inversion was even more dramatic. The Yield curve flipped from strongly inverted to rapidly steepening within a week of the Fed’s first rate cut in 4 ½ years.
The so-called term premium – which reflects the uncertainty associated with long-term bonds – also reversed its trend and jerked upwards on September 16, doubling in the next four months.
The period of the inversion (791 days) matches almost exactly the period of the Fed’s interest rate tightening program (823 Days).
In sum, the inversion was correlated with directional changes in Fed policy. It began when the Fed started raising rates for the first time in 3 years, and it ended when the Fed lowered rates for the first time in 4½ years.
Anomaly 4: The Treasury Market Goes Rogue
If it appeared that the Federal Reserve was controlling the yields in the Treasury bond market prior to September 2024, what has happened since must be astonishing. Instead of market rates following the Fed downward after the rate cuts began, Treasury yields have surged up — and have taken other important long-term interest rates up with them. The 10-Year yield rose 117 basis points.
As the Fed has cut rates further, the gap with the 10-year Treasury yield has open up to more than 200 basis points – and has done so faster than at any point since the 2008 financial crisis.
The reversal is stark. Prior to September, as noted, the markets and the Fed were in lockstep. Since then, they have moved in opposite directions.
Anomaly 5: The Gasoline-Soaked Rags in the Treasury Market’s Garage
When Liberation Day was announced, it unquestionably constituted what economists call an exogenous shock. The normal response to such an event is a “flight to safety” – investors sell off risk assets (equities) and buy less risky assets (bonds). The surge in demand drives bond prices up and yields down.
In this case, the initial response was typical. But then it abruptly reversed.
“Initially, Treasury yields dropped, in a classic flight-to-safety pattern. After a few days, however, longer-term Treasury yields started to rise sharply.” – Roberto Perli, New York Fed
Did the market’s assessment of the tariff crisis shift suddenly and hard from safety-first to run-for-the-hills just 48 hours later? Had the econometric modelers suddenly experienced a new epiphany regarding the economic impact of tariffs?
I don’t think so. I think the turnabout was driven by organic factors in the Treasury market.
The Treasury Market is not a simple environment anymore. For example, recently the phenomenon referred to as the basis trade has surged in volume. Without getting into the mechanics of this technique here, let us note that it involves creating positions that are sensitive to tiny price movements and amplified with massive amounts of leverage (“sometimes up to 100 times”). Torsten Sløk, chief economist of Apollo, describes the danger.
“Why is this a problem? Because the cash-futures basis trade is a potential source of instability. In case of an exogenous shock, the highly leveraged long positions in cash Treasury securities by hedge funds are at risk of being rapidly unwound. This could lead to a significant disruption in market functions … and liquidity issues.”
The volume of the basis trade has grown by a factor of five or more in the last several years, estimated to have reached $800 Bn to $1 Tn in March of this year (just prior to the Liberation Day announcement). If Sløk’s scenario is valid, the unwinding of that position would be a significant force for driving up Treasury yields.
Perli offers a different description of the combustibles stored in Treasury’s garage. He cites so-called swap-spread trades –
One factor that appears to have contributed to this unusual pattern is the unwinding of the so-called swap spread trade. … swap spreads are defined as the swap rate minus the Treasury yield, leveraged investors were making a directional bet that swap spreads would increase. However, on the heels of the tariff announcement, swap spreads started to decline and made the swap spread trade increasingly unprofitable. Because this trade is usually highly leveraged, prudent risk management dictated that the trade should be quickly unwound, which is what appears to have happened. The unwinding involved selling longer-term Treasury securities, which likely exacerbated the increase in longer-term Treasury yields.”
Unlike Sløk (and others), Perli claims there was no evidence of an unwinding of the basis trades.
The problem is that the information about the scale and shifts of these speculative trades is itself speculative. But it is clear that both basis trades and swap-spread trades are attuned to small changes in Treasury bond prices. Both are said to be highly sensitive to external shocks (like the Liberation Day announcement). Both are said to use a lot of leverage. This would create an inherent instability in the market in the event of sudden price movements or unexpected macroeconomic news.
Whether the rags in the garage were soaked in gasoline (basis trades) or kerosene (swap spread trades), the preconditions for a flare-up were in place.
Implications
What does all this mean? And what does it have to do with tariffs?
First, the bond market is responding — forcefully, albeit anomalously – much more to the Fed policy than to tariff talk. No surprise there, but worth emphasizing in connection with the tariff panic.
Second, the Federal Reserve seems currently unable to move market interest rates in the desired policy direction. Reducing the Fed Funds rate by 100 basis points from September to December has not eased credit conditions. Instead, bond yields and mortgage rates increased by 100 basis points over the same time frame.
Third, it would seem likely that “preconditions” in the bond market – especially the unwinding of the “basis trade” – were the major contributors to the market turbulence. Tariff talk was a trigger, but the large-scale market-clearing of Treasurys in a short time feels much more like a crowded trade unwinding than a response to the substantive economics of tariffs. And in any case, the rapid recovery in bond prices and the steep reduction in volatility suggests that tariff policy (where uncertainties remain in play) was not the main causal factor.
So – net net…
As far as the bond market turmoil goes, tariffs may well turn out to be a very red herring. Their economic impact, even under the worst-case scenarios short of all-out trade war, is not proportional to the scale of the Treasury market storm in early April (let alone devastation in the stock market). The real uncertainties in the bond market are lodged in the larger complexities and flammable preconditions resulting from highly leveraged speculative positions set to unwind on a hair-trigger.
For further reading, see:
Read the full article here
