# EMD, Meet the Debt Ceiling – Again

### Our research suggests that the recent market rally has been driven by a liquidity injection from the US Treasury; what happens when debt-ceiling machinations reverse that flow?

Don’t say we didn’t warn you – US Treasuries and the coming bond market selloff.

### Introduction

2022 was a bad year for most assets. That much is by now well known, but somewhat lost amid the negative calendar-year numbers was a powerful rally from market lows in mid-October that has carried over into 2023. 10-year US Treasury yields have fallen by 77 basis points since then, while the MSCI World Index is up by 16%, the euro has gained 13% against the US dollar, and the JPMorgan Government Bond IndexEmerging Markets Global Diversified (GBI-EM GD) has returned 15%.1

The consensus explanation of this strength can be broadly summarised in two parts:

1. Risk assets had become substantially oversold; and
2. Expectations of “immaculate disinflation”, or only a limited recession, have increased.

We think that expectations might have become a bit too optimistic.

We agree with the first argument, but are not as convinced by the second. It is difficult for us to believe that the market, which until September last year regarded central banks as having lost control of the inflationary process, has taken just four months to price in a return to 2% annual inflation (if one accounts for term premium) for the next 30 years (Figure 1). This is only one area where we think that expectations might have become a bit too optimistic.

##### Figure 1. Evolution of the US Treasury Breakeven Inflation Curve, 2022-2023

Source: Man GLG and Bloomberg; as of 25 January 2023.

Indeed, a 5y5y US Treasury breakeven currently at 2.26% is more characteristic of a market pricing in a meaningful probability of deflationary outcomes. For its part, the Federal Reserve (‘Fed’) views 2.7% 5y5y breakevens as a level consistent with its 2% inflation target after accounting for term premia. However, neither equities nor credit spreads seem to be pricing in such a scenario.

The rally in risk assets between October and now was, in our view, driven to a significant extent by the liquidity injection produced by the depletion of the US cash balance of the Treasury at the Fed.

### An Injection The Market Didn’t See Coming

We have an alternative explanation: the rally in risk assets between October and now was, in our view, driven to a significant extent by the liquidity injection produced by the depletion of the US cash balance of the Treasury at the Fed between then and midJanuary. This was not expected by those market participants who tend to be aware of this dynamic, as back in October the US Treasury did not project to draw down its cash account at the Fed.

We have written about this phenomenon in the past. When the US Treasury is unable to issue net new debt due to the debt-ceiling restriction, its cash withdrawals from its account at the Fed are injected into the economy. When the Treasury pays its bills, individuals and companies receiving the payments end up with that cash deposited in their bank accounts. This in turn forces the banks to rebalance their portfolios to bring the duration of their assets closer to where they were prior to the liquidity injection (i.e. by buying Treasury bills). The sellers of those Treasury bills then turn around and buy slightly longer bonds (to keep their durations closer to what they were initially), and this triggers a portfolio-adjustment process that pushes investors to riskier assets on the margin than they held prior to the liquidity injection.

In Figure 2, the blue line shows that between 25 October 2022 and 13 January 2023, the Treasury’s cash balance fell by $349 billion, equivalent to an annualised rate of$1.59 trillion. If, as we believe, this liquidity injection – which was unexpected by the market – operated like QE, then we are looking at an annualised rate of QE close to $1.6 trillion. Put in context, this is$400 billion more than the maximum possible amount of annual quantitative tightening in the US. In our opinion, this Treasury liquidity injection was the main driver of the drop in US yields and – with a lag – the rally in risk assets.

##### Figure 2. 10-Year US Treasury Yield Versus Treasury Cash Balance Held at the Federal Reserve

Source: Bloomberg; as of 25 January 2023.

In the case of emerging market (‘EM’) hard-currency credit spreads, we found back in 2017 that there is roughly a six-week lag between the moment the Treasury cash balance changes and the impact on EM credit spreads. Figure 3 shows how EM spreads compressed on the back of the fall in the Treasury cash account at the Fed.

##### Figure 3. JPMorgan Emerging Bond Index Global Sovereign Spread over Treasuries Versus Treasury Cash Balance at the Fed (Advanced Six Weeks)

Source: Bloomberg; as of 25 January 2023.

In the case of EM local bonds, Figure 4 shows the evolution of the Treasury cash balance at the Fed versus the EM local bond index (inverted). The lag with which changes in the cash balance affect EM currencies and rates is roughly four weeks.

##### Figure 4. JPM GBI-EM GD Versus Treasury Cash Balance at the Fed (Advanced Four Weeks)

Source: Bloomberg; as of 25 January 2023.

As the charts show, the extraordinary measures that the US Treasury began implementing in January have reversed the evolution of its cash balances at the Fed. In late January, the Treasury cash balances at the Fed increased, producing a reduction in liquidity (operating as quantitative tightening). This coincided with the incipient reversal in Treasury yields.

Given how positioning has changed from oversold to overbought in the past four months, the risk of a short-term correction is significant.

We therefore believe that until mid-March – and quite likely until May or June – the US Treasury cash balance is unlikely to be a source of liquidity for the market. Hence, for the rally in EM currencies, rates and spreads to continue, the “immaculate disinflation” arguments will have to prove to be right. Otherwise, given how positioning has changed from oversold to overbought in the past four months, the risk of a short-term correction is significant.

For a sense of how such a correction may look, Figures 5-7 illustrate the experience of the 2017 debt-ceiling rollover. On that occasion, there was a temporary three-month extension of the debt ceiling in October, with the final long-term extension occurring in the second half of January 2018. As we can observe, the patterns are quite similar to the ones in the charts above.

##### Figure 5. 10-Year US Treasury Yield Versus Treasury Cash Balance Held at the Federal Reserve, 2017-2018

Source: Bloomberg; as of 25 January 2023. Period shown is 10 February 2017 to 1 July 2018.

##### Figure 6. JPMorgan Emerging Market Bond Index Global Spreads over Treasuries Versus Treasury Cash Balance at the Fed (Advanced Six Weeks), 2017-2018

Source: Bloomberg; as of 25 January 2023. Period shown is 10 February 2017 to 1 July 2018.