Yield Curve vs GDP: Growth is Easy to Predict?

The Yield Curve vs GDP. You’ve no doubt read about the yield curve all over the blogosphere lately, one can hardly avoid it at any time on Bloomberg, the Wall Street Journal, or a myriad of independent sites. And for good reason! We’ve already shown on this blog that the yield curve is predictive for equities.¬†Yet perhaps even more interestingly, the yield curve actually predicts GDP as well.

Why you should care: because any metric with significant predictive power on recessions is hugely powerful in terms of avoiding the 30-50% equity drawdowns that accompany them.


Let’s standardize terminology. GDP is the real (e.g. inflation adjusted) dollar output of the American economy. The rate of change in GDP is simply the rate at which the economy expands or contracts.

The yield curve represents the interest rates payable on bonds of different maturities. Under normal circumstances, this is upward sloping: investors require more yield in order to lock up their money for longer periods of time. From time to time, however, the curve ‘inverts’ such that short duration bonds have higher yields than long duration bonds. This is called simply an inversion of the yield curve. The value between any two dates of maturities is called the yield curve spread.

I prefer to use the spread on the 10 year to 2 year Treasury bond because they provide useful insight of the expected behavior of the long and short term bond markets respectively. Some benchmark to the three month treasury, but as this moves largely in lockstep with whatever the Fed does, it moves in fits and starts whenever the Funds Rate is changed. I prefer the added information from the 2 year as it prices in market expectations of coming rate hikes/decreases. Historically, when the spread inverts, a recession generally follows shortly thereafter:

Yield Curve vs GDP

What I propose in this section is probably the simplest model of GDP expansion ever presented, and there are no doubt a bunch of dynamic equilibrium macro academics that are going to hate this model. But I’ve generally found simplest is best, and this model is plenty simple. The only inputs into this model are:

  1. An indicator (simply 0 or 1) if the 10Y/2Y spread inverted in the prior 2 years
  2. Time
  3. The change of the employment rate of the economy. I use this is a crude proxy for all of the stuff that happens during an expansion or recession that I don’t model. Unemployment falling and the economy is probably doing pretty good in real time. Unemployment rising and things are probably bad.

From those very simple inputs, it turns out we can predict nearly 70% of the year on year change in GDP over the last 40 years. All inputs are statistically significant at the 95% confidence level (most at the 99% level).

Yield Curve vs GDP: Results

There are two powerful results from this analysis. The first is the yield curve vs GDP. On average, a yield curve inversion causes GDP to fall about 4% 2 years later. This effect is robust to the inclusion of the unemployment rate, meaning it is truly predictive above and beyond what impacts we might observe contemporaneously in the real economy.

Second, the time trend for economic growth is falling by 0.04% per year. This might not sound like much, but it means the economy is growing by 1.6% less every year than it did in the late 1970s, holding all other variables constant. I’m sure there are a great many reasons for this, including stagnating productivity, a mature economy, or the simple mathematics that its much more difficult to grow a $20 trillion economy than a $5 trillion economy. It is nevertheless a real result which indicates we should expect more tepid growth, particularly as the US work force ages.

Yield Curve vs GDP: Conclusions

The more I study it, the more I conclude: if there was to be one solitary indicator to gauge the strength of the economy, it is simply the spread of the 10Y to 2Y treasury. It has predictive power on equity valuations, and it predicts economic growth quite accurately as well. If and when the yield curve inverts, it is time to consider managing your aggregate risk.

The yield curve is the financial Eye of Sauron that sees and knows all. It’s wisdom should be heeded.

1 Response

  1. TacoMan says:

    Dude – love bringing in Frodo and the like. Especially since the time horizon is so long in middle earth. A very convincing article. Super legit. I now plan to manage equity allocations with this metric in mind.