FRB Dallas: The Yield Curve is Better at Predicting Slumps in the Durable Goods Sector Than at Predicting Recessions

Evan Koenig of the Dallas Fed with the national economic outlook, comments about the usefulness of the yield curve as a predictor of recessions, the purchases–product price ratio, and the strength of consumer spending going forward:

National Economic Update, The Yield Curve: Better at Predicting Slumps in the Durable Goods Sector Than at Predicting Recessions, by Evan F. Koenig, FEB Dallas: The yield curve has been a powerful, but imperfect, predictor of recessions over the past 50 years. It's done a better job of predicting declines in durable-goods purchases by households and firms. In Chart 1, the slope of the yield curve—measured by the difference between the 10-year and one-year U.S. Treasury yields—is shown in blue; the contribution to GDP growth from investment and household durable-goods purchases is shown in red; and recessions are indicated by shaded bars. Note that every slump in durable-goods purchases has been preceded by a negatively sloped yield curve, and every negatively sloped yield curve yield—including the most recent—has been followed by a slump in durables purchases. Usually, the durables slump is severe enough to trigger a recession. The notable exception occurs in 1967, when other components of spending remained strong enough to offset the negative contribution to GDP growth from durable goods.

Table 1 compares the current slump in investment and consumer-durables demand with two prior episodes, both of which ended in recession. Specifically, it looks at how much each of several different expenditure categories contributed to real GDP growth during and just prior to the slump periods. Notice that we've seen considerably less weakening of GDP growth so far in the current episode, despite a similar combined deterioration in investment and consumer-durables spending. Most of the difference is due to relatively steady consumer purchases of nondurables and services. (Actually, in the current episode, the growth contribution from consumer durables has also been steady. Weakness has been concentrated in investment.) Net exports have been a stabilizing influence during the current episode, but not much more so than in the past.

Another Powerful Influence on the Economy: The Purchases–Product Price Ratio
When the prices of the goods you buy go up relative to the prices of the goods you sell, you are worse off, all else constant. For the U.S. as a whole, this phenomenon occurs when the price index for gross domestic purchases goes up relative to the price index for gross domestic product. Such shifts act as adverse supply shocks, temporarily reducing the economy's noninflationary growth potential.

Chart 2 shows the history of the purchases–product price ratio for the United States over the past 50 years (blue line). During the 1960s and 1990s, shifts in the ratio were generally favorable (down), and achieving high output growth and low inflation simultaneously was a relatively easy task. The 1970s were a different story altogether (mostly as a result of increases in the price of imported oil), and the 1980s a mixed bag. Since the 2001 recession, shifts in the price ratio have been generally quite unfavorable to the United States. The past two quarters are the exception. However, recent monthly data on the ratio of import to export prices (purple line) suggest that adverse shifts in the purchases–product price ratio may have resumed.

Chart 3 shows the relationship between the changes in the purchases–product price ratio and changes in the inflation rate facing consumers. It shows the relationship between changes in a relative price and changes in the growth rate of the general price level. The strength of this relationship is affected by monetary policy. Typically, Fed policymakers have partially accommodated supply shocks—they have allowed such shocks to bleed through to consumer price inflation. Notice, however, that the amplitude of the swings in inflation (blue line), relative to those in the price ratio (red line), is smaller now than it was during the 1970s. This change suggests that a reduction in the degree of monetary accommodation of supply shocks has contributed to the stability of inflation that we've seen over the past 20 years.

Consumption Gets a Boost, But for How Long?
The decline in the purchases–product price ratio over the past two quarters, evident in Chart 2, helps explain why consumer demand has held up so well during the current investment slump. As shown in Chart 4, from mid-2004 through mid-2006, slower growth in nominal spending (purple line) and higher inflation (green line) squeezed real growth in consumer spending (the difference between the two lines). Had the purchases–product price ratio continued to increase, this squeeze on real household spending would very likely have intensified. In fact, though, oil import prices dropped sharply in September 2006. As a result, the purchases–product price ratio reversed course, headline consumer inflation slowed, and U.S. households and the U.S. economy were given a reprieve.

The downward pressure on nominal household spending growth evident in Chart 4 seems unlikely to abate without substantial revisions to the long-term economic outlook. (Evidence from both surveys and financial markets suggests that consumer-expenditure inflation will average no more than 2 percent over the long term. These inflation expectations are sensible, given real potential growth of 3 percent or less, only if annual nominal household spending growth eventually slows to under 5 percent—considerably below where it is now.) Hence, the recent resurgence of the price of oil imports raises concerns about a near-term deterioration in prospects for real consumer spending.

Long Enough, Maybe
A weakening of real consumer spending growth is fine, of course, provided other contributors to GDP growth pick up the slack. In this regard, it's encouraging that economic prospects in our major export markets are bright. Over the near term, too, there is some reason to think that the growth drag from residential investment will diminish. Chart 5 shows that there is a stable long-run historical relationship between the level of real residential investment (blue line) and the level of single-family building permits adjusted for changes in the price of single-family homes relative to the overall deflator for residential investment (orange line). Whenever a gap between the series opens up, it subsequently tends to close, which means that the gap is helpful for predicting future growth in residential investment. Recent permits data signal that residential investment will remain a drag on the economy, but less of a drag than we've seen over the past several quarters.