Halloween may have passed, but home construction stocks continue to build a house of horrors: They’re down nearly 40%, on average, from their early-2018 highs. Nevertheless, don’t be too quick to conclude that their misery spells the end of the bull market.
Of course, it’s understandable why home construction stocks’ weakness is so worrying, since the sector proved to be a good leading indicator of the 2008-09 Great Financial Crisis. A year prior to the 2007-09 bear market, for example, the iShares U.S. Home Construction ETF
, for example, had already fallen 25%.
By the time that bear market started in October 2007, furthermore, this ETF had fallen 56%. (See accompanying chart.)
No wonder investors are so anxious. The problem with anxiety in this case is that we don’t have enough data to be confident about it. An historical parallel based on one data point is hardly significant, after all.
To conduct a more robust analysis, I focused on the Case-Shiller U.S. Home Price Index, which tracks prices of U.S. residential real estate. I fed into my PC’s statistical package monthly values for the index, which extend back to the early 1950s, along with the S&P 500
. I looked for correlations between the index’s changes over the trailing one-, three-, six-, 12- and 24-months and the subsequent returns of the S&P 500 over the same period lengths.
There is virtually no correlation between prices in the housing market and in the stock market.
I came up empty. None of the correlations was significant at the 95% confidence level that statisticians often use when determining whether a pattern is genuine.
Robert Shiller, the Yale finance professor and Nobel laureate, says we shouldn’t be. In an email, he said: “I have calculated that there is virtually no correlation between prices in the housing market and in the stock market over long intervals of time (back to 1890). Also, little correlation with gold prices, bond prices, even farmland prices.”
There’s a compelling rationale for Shiller’s finding: Residential real estate investment, a category that encompasses “construction of new single-family and multifamily structures, residential remodeling, production of manufactured homes, and brokers’ fees,” averages roughly 3% to 5% of GDP, according to the National Association of Home Builders.
Even if we add in consumption spending on housing (gross rents and utilities paid by renters), the share of GDP rises only to 15% to 18% of GDP, according to the NAHB.
Though that’s a sizeable chunk, it still constitutes only a small fraction of the overall economy. So if a housing downturn guaranteed a stock bear market, it would be a relatively small tail wagging a much bigger dog.
Of course, it would be easy to concoct a story in which housing prices are nevertheless correlated with the stock market. If a story like this were to capture investors’ attention, then it would follow that a downturn in housing would lead to a stock market decline. But, at least for most of U.S. history prior to the Financial Crisis, Shiller notes, such a narrative either didn’t exist or at least was not widespread.
That doesn’t mean that such a narrative couldn’t capture investors’ attention going forward, Shiller added. Indeed, he allows that these beliefs might now be developing, as a direct result of the Financial Crisis.
Even if this shift were taking root in investor consciousness, it’s not clear we would know about it enough in advance to be able to exploit it for profit. As Shiller said in his presidential address to the American Economic Association, these narratives that capture investors’ attention “are deeply human phenomena that are difficult to study in a scientific manner.”
The bottom line? The recent downturn in U.S. home construction stocks might be foretelling a stock bear market, but it’s hard to know for sure in real time or far enough in advance to profit from it. Not a lot to go on.