It ain't necessarily so ..... the yield curve, inversion and recession
ref :- "How central banks distort the yield curve's predictive power" , Megan Greene of Manulife Asset Management in the Financial Times 4/9/18
We should apologise for taking so long to get to this piece ..... it was just that kind of week last week. Still, it's easily accessible online (just Google the title) and of course remains as pertinent as ever. It's certainly not aimed at gnarled professionals, but in taking us all the way from what the yield curve represents to why what's going on right now may be a bit different to what's gone before, it'll have something for both newcomers and hopefully those who have been around a little longer.
We know that in the normal way of things the yield curve as represented on a graph runs in an upward sloping trajectory from bottom left to top right ..... in other words, the longer term of the debt, the higher the return required by investors. That seems only reasonable ..... it intuitively makes sense that all other things being equal (growth, inflation), you would want a bit "extra" to lock up your money for longer (the term premium).
For a reference on what the yield curve is doing, investors most commonly look at the difference (the spread) between the yield on the 2yr and 10ry US Treasury Notes. By way of providing context, at the start of 2014 the 2yr/10yr spread was trading above 260 basis points, i.e. one received 2.6% more for investing your money for 10 years rather than 2 years. Ten days ago or so, the spread was trading at just 18 basis points (or .18%). Today we are knocking around the 24bp level, and technical traders would tell you that the historical chart of the 2yr/10yr spread shows a classic downtrend which, assuming it continues, points to shorter-term yields actually moving higher than longer-term equivalents. That would be called yield curve inversion and is a big deal in bond markets and beyond.
It's big because the inversion of the yield curve has a strong track record as a precursor to economic recession. Ms Greene points out that each of the seven recessions since the 1970s was preceded by a yield curve inversion, and it is a more reliable guide to predicting downturns than the witterings of mere economists. (We always feel the need to add a small but important caveat at this point : in recent times every recession has been preceeded by a curve inversion, but not every inversion was followed by recession).
Anyway, in terms of the fundamentals why should it be the case that curve inversion presages a downturn ? In a time of growth, a central bank would naturally be lifting short-term rates as a measure against rising inflationary fears but has very little control over long-term yields which are decided by (amongst other things) expectations of short-term rates in the future and by the level of those fears about inflation. Remember : Long bonds HATE inflation as it eats into fixed returns, so rising inflation expectations push prices lower and therefore yields higher. If longer yields fall (and prices rise) in relation to short-term rates, it suggests a reining back of expectations of growth (and therefore inflation). If they fall so much that they end up trading lower than short rates .... well, that's taken to be a pretty reliable sign of impending recession.
We're not there yet ..... but even if the curve does invert there's a decent argument to be made that it need not be the pessimistic signal that many will make it out to be. People making that argument might point out a flaw in tracking the yield curve's inversion (or not) by referring to the 2yr / 10yr spread. If you want to compare short rates to long rates, wouldn't it be better to follow the yield on a genuinely short instrument like a 3-month Treasury Bill and compare that to the 10yr ? That spread is still out above 80 basis points. But there are other factors specific to this moment in time that also cloud the picture :
1. Mr Trump's tax cuts may have been good for growth and for equity prices, but in the short-term at least they leave a much bigger deficit in government finances that has to funded by issuing debt. The Treasury has chosen to do so by issuing proportionately more short-term paper, thus pushing short rates higher in relation to long-term yields
2. Quantitative Easing (QE) across the globe has not only soaked up huge tranches of government debt but created enormous demand for high quality debt that actually offers a respectable yield. The returns on offer for 5 - 10yr US Treasuries make them seem particularly attractive in comparison with the negligible (or even negative) yields available elsewhere. Big demand means higher prices, and lower yields, even as Fed raises short-term rates as a response to strong growth and tight labour markets. Thus yield curve flattening is the result of financial considerations rather than economic ones.
3. We keep expecting inflation to rear its ugly head (as a function of strong growth and tight labour markets) but contrary to expectations it remains decidedly under control. It could be that the dynamics that drive inflation have changed for ever. It's not just the "Amazon Effect" keeping a lid on prices, it's also other forms of automation and robotics and labour practices. In fact, demand for longer bonds has been boosted by deflationary fears with investors hedging their sky-high valuations of equity portfolios . Deflation would of course be good for bonds and very bad for stocks.
For what it's worth, we actually believe that things have changed enough to support the theory that an inversion of the yield curve wouldn't necessarily be the result of factors pointing to a coming recession. But that's not the same thing as saying that an inversion won't presage a downturn on this occasion. If enough people believe something to be true, then more often than not it becomes true .... it's a self-fulfilling prophesy in other words, regardless of economic fundamentals. That's just the way markets are, and right now a worrying amount of people still subscribe to the old inversion/recession theory.