Be careful what you wish for. The concern about the flattening of the US treasury curve suggests people would be relieved if the flattening would stop. However, the historical experience shows this would give even more reason for concern. Past recessions in the US have been preceded by a yield curve flattening followed by an inversion but this knowledge is of little help when producing forecasts or, as an investor, doing market timing. Indeed, the time lag between a flat to negatively sloped curve and the start of a recession is long and variable (see chart). With the historical relationship between the slope and economic downturns in mind, one is inclined to hope that at some point the curve would steepen again. This could happen either because the 10-year yield rises more than the 1-year yield or because the latter would drop more than the former. In the first case, the curve steepens and shifts upwards. Under the assumption that the term premium* does not change, this would require a considerable increase in the long-term growth outlook which would pull upwards the entire path of future short-term interest rates. Given the mature stage of the business cycle, it seems hard to count on such a steepening for a prolonged period, if only because it would force the Federal Reserve to tighten policy much more aggressively. This would weigh on the long-term growth perspectives and cause the curve to flatten again.

US treasury curve in the run-up to a recession

In the second case, the entire curve shifts down and steepens because yields at the short end decline more than at the long end of the curve. This would reflect a downward revision in the growth outlook which triggers an anticipation of an ending of policy tightening and an imminent start of monetary easing.

The charts below show the slope of the yield curve and the federal funds rate in the months before a recession (the last month on the horizontal axis corresponds to the start of the recession based on the NBER recession dates; the recession of July 1981 is not shown because the money supply targeting strategy of the Federal Reserve implied a high degree of interest rate volatility). In most cases, when the curve starts to steepen, a recession follows quite soon thereafter. This means that the lead time is shorter and less variable than the time between the curve inversion and the subsequent recession.

The charts also show that the steepening of the curve mostly occurred in anticipation of a policy easing. From this perspective, assessing when the Fed will stop tightening in this cycle seems a question of bigger importance than focussing on the slope of the yield curve.

*The term premium is the extra yield a bond investor receives for taking duration risk, that is for buying a long-dated bond rather than continuously rolling over short-dated paper. In the US this premium has actually been negative most of the time in recent years.