It has almost become a ritual. In the United States, the new year has seen disruptions in activity, as it began with a spell of abnormally hot and dry weather in the Southwest, and exceptionally humid and cold weather in the Northeast. A possible avatar for climate change, the adjustment for seasonal variations has become a major challenge for statisticians, and a source of risk when evaluating gross domestic product (GDP). After peaking at nearly 3% in 2018, GDP growth undoubtedly slowed in first-quarter 2019. The partial government shutdown and the trade war with China obviously didn’t help.
Yet most observers of the US economy think it would be only a temporary slowdown. This is notably the position of the members of the Federal Reserve (Fed), who are forecasting a “solid rebound” in the economy in the second quarter[1]. Indeed, a few indicators such as world trade and employment have picked up after slumping during the winter months. The equity markets are also looking perky again.
But other trends have not really turned around yet. After the euphoria of tax cuts in 2018, corporate earnings growth continues to return to normal levels. Certain surveys, like the one by the Federal Reserve of Philadelphia, suggest that industrial leaders’ expectations were more mixed in March. Known for its reliability, the OECD’s leading economic indicator is still declining.
Thus we cannot be absolutely certain that the economic slowdown is over. Judging by the quasi inversion of the yield curve[2], before things start getting better again, the US economy could be heading for worse in the months or quarters ahead. Yet for certain observers, this early-warning signal must be kept in perspective.
This time is different
The most common argument is that the yield curve, the distribution of US interest rates by maturity, is biased by the Fed’s securities purchasing programme, also known as quantitative easing (QE). Without QE, long-term rates would be higher and the yield curve would not have inverted. To defend this point, they regularly cite a study by Bonis et al. (2017), which evaluates the impact of QE on the term premium of Treasuries at 100 basis points (bp)[3]. Since this effect cannot be observed directly, however, their estimate must be taken with caution, especially since it comes with a rather high margin of error (+/- 50 bp). Moreover, the Fed is disengaging from the programme. Since October 2017, it has reduced its securities holdings by roughly USD 500 bn, which means it is less directly weighing on the long end of the yield curve.
It is not so much the central bank’s quantitative policy as its message to the markets that has pulled down bond yields in recent months, to the point that they are now lower than short-term rates. By acknowledging the deterioration of the economic outlook and announcing a pause in its monetary tightening, the Fed has facilitated, if not encouraged, the shift in expectations. At the recent meeting of the Federal Open Market Committee (FOMC), members seemed less worried, but market participants continue to give a 50% weighting to the probability of a key rate cut by the end of the year. This signals that in their eyes a further slowdown in economic activity and prices is still possible, and the 3-month forward rate in twelve month time continues to stand below the spot rate.
In the end, the inversion of the US yield curve in 2019 is no less significant than in 1989, 2000 or 2007. The regularity with which it predicts downturns in the economic and monetary cycles is sufficient proof that it should not be underestimated. An analysis by the Federal Reserve of Saint Louis even claims that inverted yield not only predict recessions but also have the power to cause them (Wheelock, 2018)[4]. By undermining transformation conditions, an inverted yield curve can lead to greater loan selectivity and curb activity (chart 2).
The risk of freezing up credit would not come as much from banks as from alternative market financing channels, one of the main vectors behind the swelling of US debt. Real Estate Investment Trusts (REITS), which use short-term borrowing on the repo market to make long-term investments in the MBS compartment, are among the shadow banking entities that seem to be most vulnerable to a flattening of the yield curve (IMF, 2013[5]).
Yet the segment that is currently drawing the most attention is leveraged lending activities. This market, which is also dominated by non-banking intermediaries - finance companies, Collateralized Loan Obligation (CLO) vehicles - has expanded very rapidly in recent years. Showing deeper internationalization and complexity, this market is also subject to greater risk taking: 80% of new operations are “covenant lite”, which means they are exempt or nearly exempt from guarantees (see chart 3 and box 4). The IMF regularly devotes several pages of its Financial Stability Report to the leveraged loan market, often for underlining excesses[6].