Why Does a Yield-Curve Inversion Suggest a Recession?

Last month (August 14, 2019), the spread between the 10 year treasury yield and the 2 year treasury yield inverted. If you believe the headlines, this traditional recession indicator caused the markets to sell off with the Dow Jones closing with an 800 point loss (-3.1%) with the S&P 500 also losing 85 points (-2.9%).

So what is the yield curve and why does everyone worry that it signals a recession?

The Bond Market

To get into yields, we need to understand a bit about bonds. Bonds are essentially loans that governments and corporations make to investors. If you buy a 10 year treasury bond, this means that you are loaning the US government money for 10 years. As with any loan, when the maturity date is reached (in this example, 10 years from today) you will have been paid off in full, plus some interest (or yield, the amount of cash you get from owning the bond).

[RELATED: “Inverted Yield Curves, Recessions, and You” by Paul Cwik]

10 year treasury bonds are the most commonly reported on in the media, although the treasury also offers shorter term bonds ranging from 1 month to 30 years. Investors anticipate receiving a higher interest rate (yield) for longer dated bonds because of time preferences and opportunity costs associated with lending money. In other words, having $1,000 today is better than having $1,000 tomorrow. The longer you have to wait, the more you expect to be compensated for waiting, and thus the higher the interest rate you expect to receive on the loan.

The Yield Curve

Given that this is the normal state of affairs, we can plot the yield for the bonds offered by the US government across different maturities and build what’s known as the yield curve .

(A normal, upward sloping yield curve. Data from US Dept. of Treasury. All plots are a result of the author’s own calculations.)

The plot above shows a typical and well-behaved plot of the yield curve. It slopes up and to the right as we would expect, meaning lending the government money for one month provides less yield than for 3 months, which is less than for 1 year, and so on. The trouble is that this isn’t always the case, sometimes we get yield curve inversions , as shown in the plot below.

(An inverted yield curve. Data from US Dept. of Treasury.)

Here, the curve is all over the place. The 6 month yield is higher than all of the longer term yields apart from the 20 year bond which has the highest rate. If you, as an investor, show up to the bond market looking to buy today, you would most likely look at that 6 month bond and grab that — after all, you get your money back sooner, and with higher interest than say a 2 or 10 year bond.

We can view inversions as snapshots in time such as in the plot above, but it’s more typical to plot the spread between two maturities. This is simply one yield minus the other yield. Because we expect the longer maturities to have higher rates, we subtract the longer yield from the shorter, and when the difference drops below 0 (shown in red below) we have a yield curve
inversion.

Long-short term spread. The one year rate used for short term prior to 1976 when the two year treasury was introduced. The 10 year rate is used for the long-term rate over the entire time
period. (Data from St. Louis Federal Reserve.)

History Points to Recession

Looking back at the 2 year and 10 year spreads, we can see that the curve has had sustainable inversions eight times since the early 1960s and led to a recession seven times.
The one time a recession didn’t follow an inversion was in the mid-1960’s, but even this period saw a contraction in output — just not enough to be labeled a recession.

We have daily 10-year rates minus 2-year rates (except for before 1976 where only one year rates are available). The gray bars indicate official recessions, while the blue dots show the first date of a sustained inversion of the yield curve.

There was also another, short-term inversion in 1998 that didn’t see a recession in the United States, but coincided with the implosion of the massive hedge fund Long Term Capital Management and was a few months ahead of the Russian debt default in August of that year. In other words, any time the curve has inverted, financial and economic trouble has followed.

By my data, an inversion precedes a recession by nearly a year on average, with a median of 9.5 months. As can be seen below, there is a reasonable amount of variance with some recessions coming in about 6 months and others waiting nearly 20 months. Of course, this is just looking at a small sample size from history (only seven recessions); this on its own doesn’t guarantee a recession will hit in the next year or so, but it should encourage some additional caution.

Median time to recession is 292 days from the first date of the inversion between the 10-2 year treasury bonds.

Why Does the Yield Curve Invert?

Bond yields move opposite to their price. For example, if you purchase a 1 year bill from the US Treasury for $950 that entitles you to $1,000 one year from now, you are receiving a 5.3% rate
on that bond. If the price of that debt increases because more investors want to hold US treasuries, you may now pay $975 to get $1,000 in one year, meaning your rate now dropped to 2.6%. This is the basic mechanism that drives yields and relates them to bond prices.

During an inversion, there is too much short term debt relative to long term debt which pushes short yields up and long yields down.

Inversions and Recessions

We can best explain yield curve inversion and subsequent the recession busing Austrian business cycle theory (ABCT). In a nutshell, the theory argues that unsustainable booms are set in motion through monetary manipulation which artificially lowers interest rates, causing a boom in wasteful investment (termed “malinvestment”) and projects. Eventually this is
followed by an inevitable bust. This is easy to see when you consider that a shortage of credit is just the inverse of an excess of debt in the bond market; not enough people are willing to lend money at the prevailing rates forcing bond prices down and yields up.

During the bust phase, ABCT predicts that this malinvestment must be liquidated. The unprofitable firms begin struggling, are unable to roll-over their debt, and rates rise to compensate.1 This leads to the inversion and provides insight as to why the yield curve is such a powerful predictor of economic downturns.

• 1. This certainly isn’t the only cause of the rise in the yield curve, nor the only one consistent with ABCT, but one that is consistent and expected given the theory and relatively easy to
  conceptualize. It would be interesting to corroborate the inverted yield curve with other data that may show short-term reduction in business lending, reductions in payroll numbers, and so forth, as businesses often rely on short-term lending to make payroll, replenish inventory, and manage their operations.