How Hard a Hit from Housing?

S&P's chief strategist says the slump probably won't tilt the economy into recession, thanks to a proactive Fed

by Sam Stovall From Standard & Poor's Equity Research

In the past few weeks, news and forecasts surrounding the housing decline—which began in 2005 and is projected by S&P Economics to bottom out during the first half of 2008—appear to have shaken investors awake and generated a renewed worry. At a recent meeting of S&P's Investment Policy Committee, someone said these investors must have awakened from a coma and are only now pressing the panic button multiple times.

Maybe the consensus expected the decline to be a lot worse than it will likely be. David Wyss, S&P's chief economist, has been forecasting for nearly a year that the slump in housing would subtract 1% from U.S. gross domestic product in 2007 and 2008. What's more, Wyss thinks the median price for homes across the U.S. will decline 11% from the peak in 2005 to the projected trough during the first half of 2008, and that—as we are less than halfway there—things will get worse before they get better.

What could be generating increased agitation among investors, in our opinion (besides the 10.2% plunge in September's U.S. housing starts to a 1.191 million pace, the lowest level in 14 years), is history, which has demonstrated a high correlation between declines in residential construction and recessions. I was alerted to this fact by my friend Jim Stack of InvesTech Research in Whitefish, Mont., and was alarmed to see that every recession since 1960 has been accompanied by a year-over-year decline in residential construction, which includes new home construction and home improvements.

During the current housing cycle downturn, S&P and Global Insight, an economic forecasting service, sees the year-over-year decline in residential construction bottoming in the first quarter of 2008 at –20.8%, and becoming positive once again in the second quarter of 2009. Obviously, with such a high correlation between construction declines and recessions, no wonder investors worry about recession and question the validity of our stated 33% risk of recession in 2008—equal to the risk percentage held by former Fed Chairman Alan Greenspan. But we think we will escape recession, despite weak housing, as we did twice since 1960.

Lessons from the Past

The reverse of this proposition is less alarming because even though all recessions have been accompanied by construction declines, not all construction declines resulted in recessions. Twice the U.S. experienced year-over-year declines in residential construction—1967 and 1995—but did not slip into recession. In 1995, the reason could have been that the housing decline was the second shallowest of all 10, posting a 7.5% drop-off during the second quarter of 1995. The construction decline was also fairly short, lasting only four quarters, as opposed to the decline of 1973-1974, which lasted nine straight quarters, eight of which featured double-digit declines. (The housing recession of 1980-1982 saw 14 months of declines, but was separated by one positive quarter.)

Another reason the 1995 housing slump may not have triggered a recession, in our opinion, was that the Federal Reserve acted fairly quickly to reverse the effects of the seven Fed funds rate increases from February, 1994, through April, 1995, that raised rates from 3% to 6%. The Fed reversed course five months later and cut rates from July 1995 through January, 1996, bringing the Fed funds rate down to 5.25%. Also, U.S. real (i.e., unadjusted for inflation) GDP slowed to a 2% year-over-year gain in the December, 1995, quarter and was surrounded by 2.5% gains in the September, 1995, and March, 1996, quarters.

The housing slump from the fourth quarter of 1964 through the first quarter of 1967, which saw a 22.9% year-over-year decline in residential construction, but did not lead to a U.S. recession, was also incorrectly anticipated by the S&P 500 as it slipped into a bear market mode from February, 1966, until October, 1967, that shaved 22% off of the S&P 500's value. One reason could be that the economy again slowed, but did not enter into a recession. In March, 1966, the U.S. economy peaked at an 8.