The US yield curve has historically provided an accurate barometer of the economic outlook. But it is wrong to conclude the recent flattening is pointing to a recession, argues James McAveley.
5 minute read
The shape of the US yield curve is closely watched by financial markets for good reason. After all, an inverted yield curve, where short-dated debt yields more than debt with longer maturities, has been an accurate harbinger of the last seven US recessions, going back as far as the late 1960s.
Although the US curve is not currently inverted, it has flattened appreciably in 2017 as yields on long-dated US bonds plunged to their lowest level in more than a year. As can be seen in chart 1, the slide in longer dated bond yields has driven the spread between the yield on two and ten-year debt to 77 basis points, close to its lowest level in a decade and down from a December high of 136 basis points in the aftermath of Donald Trump’s election victory. This has prompted many to conclude the bond market may be on the verge of inverting, signalling a recession may be looming.
Chart 1 – US government 10-year minus 2-year yield
Exceptional circumstances
Since historically yield curves have inverted when the economy is about to go into recession, it is understandable why investors might draw this conclusion. However, this analysis is flawed. In viewing the yield curve through a traditional lens, these investors are failing to take account of the exceptional circumstances that still prevail, a decade on from the financial crisis.
US government bonds continue to be distorted and manipulated by central banks. As such, it is unwise to read too much into what yield curves imply about the future, at this point in time. The information embedded in them, in terms of what could happen to the economy, is slim-to-non-existent.
That is not to deny fundamental factors have been a factor in recent developments. After the yield curve steepened appreciably last year in the belief Donald Trump’s election victory would spark stronger economic growth, this year’s reversal has been driven partly by markets beginning to challenge this view. This dynamic has been reinforced by softer inflation data.
Central banks trigger collapse in volatility
However, fundamentals only partly explain the curve flattening. Central banks, which have continued to expand their balance sheets, have arguably played an even bigger role. Ongoing quantitative easing (QE) by the European Central Bank and Bank of Japan has helped fuel private sector demand for US Treasuries from Europe and Japan. Other central banks have been mopping up Treasuries more directly. This year’s decline in the dollar has led to a big increase in the amount of capital flowing into emerging nations. As a result, many of these countries’ central banks have bought US dollars to prevent their own currencies from appreciating. Those dollars have in many cases been reinvested into Treasuries. Crucially, these central banks tend not to be sensitive to prices as they have nowhere else to reinvest their dollars.
At the same time, whereas The People’s Bank of China last year sold Treasuries to plug a gap in its capital account – which at the time was shrinking – to support its currency, this year it has been buying once again. So, in contrast to last year when overall demand for Treasuries from central banks declined, this year it has grown.
Although we have been in a low-volatility regime since the post-crisis period, the further collapse in volatility this year has been remarkable. It seems the sheer scale of price-insensitive buying from central banks, regardless of issuance patterns, has stopped the market from moving. This means it is unable to reflect any level of information being transmitted. Daily price changes are confined to two to three basis points, about half of what they historically were, and a fraction of what they were in the immediate post-crisis period when there was no quantitative easing.
Close correlation
The collapse in volatility ultimately explains the bulk of the flattening in the yield curve this year. As can be seen in chart 2, with the exception of the immediate post financial crisis period, volatility and the shape of the US yield curve (as represented by ten-year minus two-year yields) have historically been closely correlated.1
Chart 2 - Correlation between volatility and the shape of the US yield curve
There is a rational explanation for this. A flat yield curve is consistent with low volatility, since there is no difference between the ‘spot’ and ‘forward’ curves and yields are not expected to change over time. A steep yield curve, by signalling rising rates in the future, is consistent with higher volatility.
Although there is no way to formally arbitrage between the two, a low-volatility environment makes carry trades look relatively attractive on a risk-adjusted basis. When volatility is low, market participants effectively have an incentive to borrow short-dated funds at low rates of interest to invest in bonds with longer maturities.
This explains why in such an environment investors tend to allocate capital to bonds where carry is highest (five to ten year maturities), which in turn tends to bid up the price of such debt, forcing down its yield and flattening the curve.
A further gauge of the extent to which the yield curve is being distorted by the actions of central banks is provided by the term premium – the additional return an investor can expect over and above that determined by the expected path of overnight money. Even in a world where cash rates are expected to be unchanged, investors should normally expect a greater return for locking their money away for longer terms. This, however, is no longer true. As chart 3 shows, the term premium on ten-year debt has sunk since December and now stands close to an all-time low.
Chart 3 – US 10-year term premium
Risky carry trades
Many investors seem to assume this ‘free’ money is going to be around for ever, and as a result are borrowing and investing it in carry trades. However, that looks like an increasingly risky strategy with the Fed about to start shrinking its balance sheet, especially if the market re-evaluates the prospect of inflation returning to target.
In the absence of central banks’ desire to rig the market, the distortions that have been created should gradually unwind. While the Fed will be keen to prevent these distortions unwinding too abruptly, as was seen during the ‘taper tantrum’ of 2013, it will surely be happy to see a gradual rebuild of term premia by 50 basis points or so over the next two years. Although ordinarily the yield curve would be expected to flatten in a rising interest rate environment, we believe the opposite is quite possible as the term premium begins to normalise.
While economic growth has been reasonably strong this year, the bond market appears to be focusing more on inflation. However, we would caution against reading too much into recent inflation numbers. Like the Fed, we remain of the view that stronger growth will eventually cause inflation to pick up, albeit gradually.
With the labour market tight and the economy growing at a healthy clip, the Fed will eventually be forced into acting more aggressively than the market expects.
There are other factors forcing the Fed’s hand. Increasingly, the central bank is starting to talk more about financial conditions, which have continued to ease with the dollar falling, credit spreads tightening and equities going up. That is no surprise given that credit is now building up in pockets of the financial system that were largely dormant for most of the post-crisis period. The bottom line is that US rates are likely to rise for various reasons and not simply because of the current level of inflation.
As for what this means for the US yield curve, price-insensitive demand from foreign central banks has caused deep lasting scars on the market and has been the main force behind the yield curve flattening this year. Right now, however, the curve is much more reflective of where the term premium is and tells you little about the country’s economic prospects. While, with the economic cycle now mature, it is understandable why some investors are starting to bet on a recession, they are likely to be disappointed. We expect short rates to rise further and would not be surprised if the yield curve actually steepens.
References
1 When the financial crisis struck, the relationship broke down. Quantitative easing by the Federal Reserve pushed volatility down. At the same time the yield curve steepened as short-dated rates collapsed while longer-dated yields climbed due to worries about the inflationary consequences of money printing and the impact of soaring fiscal deficits.
