|
|
Thursday, 25th April 2024 |
| The bond yield curve is a valuable forecasting tool |
Back |
| At present in bond markets there’s lots of fancy jargon being used, some of it reasonably, self-explanatory but some of it quite confusing, says David Green. When we say that a bond yield curve is steepening, what we mean is that yields at the long-end (say 30 year bonds) are increasing at a faster rate / falling at a slower rate than yields at the short-end (say two year bonds). |
The ‘Liquidity Premium Theory’ suggests that, in normal circumstances, the yield curve should be upward-sloping. This makes intuitive sense - we know ourselves that if investors lend money for longer periods, they tend to demand a higher return because of all the greater uncertainties associated with the longer time-frame. What other terms are commonly used? Well, for example, if two year yields rise by 0.05 percent but 30 yields rise by 0.20 percent, then the yield gap between two year bonds and 30 year bonds has increased (or steepened) by 0.15 percent. This would be known as a ‘bear steepening’ since all yields are increasing. A ‘bull steepening’ would occur in the case of two year yields falling by 0.20 percent but 30 year yields only falling by 0.05 percent - the gap steepens by 0.15 percent but it’s bullish because the entire yield curve has moved down.
When we say that a curve is ‘flattening’, we mean that long-dated yields are increasing at a slower rate/falling at a faster rate than shorter-dated yields. Again, a ‘bull flattening’ would be where two year yields fall by 0.05 percent but 30 year yields fall by 0.20 percent, and a ‘bear flattening’ would be where two year yields increase by 0.20 percent at a time when 30 year yields only increase by 0.05 percent.
Lastly, when we say the curve has inverted we mean that shorter-dated bonds yield more than longer-dated bonds or that the curve is downward-sloping.
So enough of the jargon. What about history ?
There has been plenty of economic research looking at the yield curve as a predictor of future economic growth and the evidence is powerful - the yield curve is a valuable forecasting tool. However, much of the work has focussed on using the curve as an early-warning indicator of a slowdown or recession. And its true : a flat or inverted curve has typically been prior to a recession by somewhere between two and six quarters. In addition an inverted yield curve has preceded every recession in the last 25 years and the yield curve has inverted nine times since 1954 and in all cases, except one, a recession has followed.
The method by which the curve moves from upward-sloping to flat to inverted is also fairly predictable. In response to above-trend growth or fears of higher inflation for example, the Federal Reserve (the Fed) start to increase interest rates. This causes yields at the short end of the bond market to increase. But as the Fed increase rates further, investors in long-dated bonds anticipate that higher rates will cause growth to slow in the future and interest rates (yields) to be cut and therefore inflation will be lower; and so those investors buy bonds at their current attractive yields. And because they are all trying to buy long-dated bonds, bidding up the price causes yields to fall and the curve to eventually flatten or even invert.
A great recent example of this was in 2000, when with the US economy careering along at a growth rate of over 5 percent, and the Federal Reserve busy increasing interest rates to 6.50 percent, the US yield curve inverted early in the year. Now undoubtedly some of the fall in long-term rates was a reaction to the US Treasury’s buy-back program but it was also, in hindsight, a very accurate predictor of the current slow-down, and at a time when few foresaw the economy grinding to a halt.
But does the opposite apply as well ? If an inverted curve is a recession predictor, does it imply that a sharply-upward sloping yield curve is a predictor of good times ahead ? Thankfully, the answer is yes. And the steeper the curve the more pronounced the recovery should be - the logic being that the reason the curve is so steep is that short rates have fallen to very low levels and should therefore be stimulatory for future growth. The yield curve is always positively-sloped in advance of an economic recovery but the lead-time has ranged from one to thirteen months. Academic studies prove that over the last 30 years, the yield curve provides one of the best forecasts of real growth four quarters into the future. For example, those of you who were in financial markets should cast your minds back to 1987 and the aftermath of the stock market crash. The Fed aggressively cut interest rates and the curve steepened, suggesting good growth in 1988, yet many forecasters were predicting a recession based on the international stock market crashes. In fact, growth in 1988 was 4.2 percent.
The reason for the history lesson is that the US yield curve is presently at its steepest level in almost nine years - since the US economy emerged from its last recession in the early 1990s. If the yield curve is as good as history suggests, then by this time in 2002 growth in the US should be back to or above trend - a level we estimate to be around 3.50 percent. The path we travel to get to that level of growth might be a bit rocky along the way but the road is stretching out ahead of us. And that’s pretty much the view that we have here in KBCAM - growth in the biggest economy in the world steadily increasing throughout each quarter of 2002 so that although our year-on-year forecast looks miserly at 1.8 percent, quarter-on-quarter growth in Q3 and Q4 2002 will be 4.50 percent or more. That, in turn, will underpin solid earnings and equity returns, but make life difficult for fixed interest managers. |
David Green is senior asset manager at KBC Asset Management.
|
| Article appeared in the February 2002 issue.
|
|
|
