Housing and car sales slumps are the best business cycle predictors of recessions
Business cycle research seems to be making a resurgeance. As well it should given the compelling data set forth by UCLA economist Edward Leamer in a paper presented at Jackson Hole last week
(warning: pdf). According to Leamer, of the 10 recessions that have occurred since World War II:
We have experienced 8 recessions preceded by substantial problems in housing and consumer durables....
Residential investment consistently and substantially contribute[d] to weakness [in GDP growth] before [ these 8 ] recessions....
After residential investment as a contributor to prior weakness come consumer durables, consumer services, and then consumer nondurables. Those are all consumer spending items -- it's weakness in consumer spending that is a symptom of an oncoming recession.... The timing is: homes, durables, nondurables, and services. Housing is the biggest problem in the year before a recession... durables is the biggest problem during the recession [although consumer durables declined even more than housing before 2 of the 10 post World War II recessions]
In terms of timing, housing typically begins to decline 5 quarters before recession, with durables and nondurables hitting their peak 4 quarters before the recession, and gently declining until the recession hits. The only two times since World War II that a substantial housing decline did not presage a recession were in the midst of the Korean War and the Vietnam War. As an aside, a 1954 paper presented a graph showing that housing also peaked in 1925 and declined consistently throughout the rest of the 1920s, and then continued to decline all the way till 1933!
After a housing slump, the next part of the economy to turn negative before a recession is consumer durables, meaning big-ticket items that consumers purchase to last a long time. The biggest example of that is cars. As Leamer explains:
The same interest rate or other variables [mainly employment] drive both the housing cycle and the durables cycle.... It turns out that much of the ampitude inconsumer durables comes from vehicles not furniture
and he explains why this is so:
Both durable manufacturing and residential construction have four special features:
- Previous production of new homes and cars create a stock of existing assets that compete with current production....
- The asset prices of both homes and new cars suffer from downward price rigidities
In other words, the economy can only sustain X amount of cars and houses in any given period. Overbuilding in one period (a boom) means that in the next period, there will be underproduction (a bust). It is this absollute decline in activity that forms the basis for the recession.
Leamer actually believes, contrary to his own model, that this time we will escape recession. This is a total conjecture on his part that manufacturing employment, having never recovered from the 2001 recession, will rescue us from recession this time by dint of not declining further! I believe he is wrong here. He is essentially speculating that non-manufactuing employment (including construction employment, which boomed in the last 5 years) will not decline. Leamer's own thesis leads to the conclusion that, if recession is avoided, rather, it will be because, like the Korean and Vietnam War non-recessions, military spending will keep the economy as a whole from contracting. A look at Leamer's graph (on p. 6 of his paper), however, indicates that the Bush-Rumsfeld strategy of not providing enough troops to do the job in Iraq (or Afghanistan) also means that the percentage of GDP being spent on Bush's wars does not match the levels of the Korean or Vietnam Wars and so is unlikely to be enought for the eocnomy to avoid recession.
Before I discuss this week's housing and cars data, let's briefly note the other best predictor of recession.
An inverted yield curve is the best econometric predictor of recession
Last week the Wall Street Journal reported on a July study from the New York Federal Reserve Bank reiterating that an inverted yield curve is the best metric forecasting a recession; specifically, when 3 month interest rates are higher than 10 year interest rates:
Although ignored by most forecasters, a vocal minority places a lot of emphasis on an inverted yield curve as a predictor of recession. The inverted yield curve is an unusual occurrence in which short-term interest rates are higher than long-term rates. The study used the average spread between the yield on a 10-year U.S. Treasury note and the 3-month Treasury bill over the course of a quarter.
“We find that a simple model for predicting recessions that uses only real-time yield curve information would have produced better forecasts of recessions than the professional forecasters provided,” Rudebusch and Williams said.
Data this week showed continued housing and car sales weakness, and today's jobs data suggested recession may be near
This week, we got new information on both housing and cars. As to housing, the Census Bureau reported that, as shown in this graph of year-over-year change in residential construction shows it to have become negative for almost a year. And pending home sales have fallen off a cliff. We also learned that the rate foreclosures is now higher than in over 50 years!
As to vehicles, bloomberg reported that
the pace of car and truck sales in the U.S. dropped for seven consecutive months through July, the biggest string of declines in at least 31 years,
August sales rebounded slightly, but were still down 3.7% from August 2006.
So housing and cars continue to, well, suck wind, but what about interest rates. The yield curve has spent most of the last 2 years at least partially inverted, but specifically as to the 3 month vs. 10 year rates, those have recently "uninverted" and remain so.
So far the 4-week moving average of weekly jobless claims is at 325,000, at least 25,000 less than would be a signal that a recession may have started, and August retail sales remained strong, 5.0% ahead of last year.
BREAKING: This morning's August nonfarm payrolls number of -4000 is devastating. It is the number that says the housing and car sales have bled over into the overall economy, accelerating the gradual decline towards weakness since 2004 that is evident in this graph.
In conclusion, I recently argued that failure of long term interest rates to decline, requiring a consumer paydown of debt have been the reasons for the 2 consumer led recessions since 1980. In that regard, we will learn much more about whether the consumer began to cut back when the St. Louis Fed releases information on the consumer debt ratio of payments to household income for the April - June 2007 period, later this month. If the consumer is indeed retrenching, recession may be imminent.
Comments are closed on this story.