Our Outlook on the Housing Slump--Page 2
Notwithstanding recent turmoil, evidence may indicate the worst is over.
In terms of vacancies, the country reached a breaking point in the 2005-06 timeframe as the number of owned homes for sale started its meteoric rise. Now at 2.9% of owned homes, a more appropriate vacancy rate is closer to the 10-year average of 1.7%. According to this analysis, about 1 million homes need to be absorbed before the market is back into in equilibrium. More on this later.
Affordability Is a Bright Spot
Prices are declining and inventories are high--that much is clear. However, if one digs deeper into the data, some very encouraging trends start to appear. One such area is affordability. We keep close tabs on this metric for all of the cities featured in the Case-Shiller 20-city Index, as well as more than 330 metros covered within the Office of Federal Housing Enterprise Oversight (OFHEO) series. Both indicate the same thing: Affordability is rapidly coming back into line with historical averages. The blue line on the chart below depicts the percentage of annual pretax income a median earner in Phoenix would have to commit to service the mortgage of a median-priced home (based upon the Case-Shiller Index) over the past two decades.
It's clear that affordability in Phoenix is no longer meaningfully below historical averages. Based upon Case-Shiller paired sales indexes, the same can also be said for San Francisco, San Diego, Denver, Las Vegas, Minneapolis, and Tampa, all of which now reside relatively close to fair value. Even better, cities such as Dallas, Atlanta, Cleveland, and Detroit are approaching bargain pricing. In fact, only one city in the 20-city composite, Portland, Ore., currently sits more than 20% above our fair value estimate, and only four are more than 12% above that metric. Compare this to mid-2006, when 12 of the 20 cities in the study were more than 20% overvalued, nine of them by more than 30%. Although a few MSAs still have a way to go before their homes are reasonably priced, it's pretty evident that affordability has gotten appreciably better across most of the country.
Is it possible for prices to overshoot our estimate to the downside? The answer, of course, is yes, especially if current dislocations in the banking industry don't abate over the next few quarters. But we wouldn't count on a sustained downward trend in pricing once a region regains affordability. Aside from regions with widespread systemic job losses like Detroit and some markets in Ohio, home prices have proved to be sticky to the downside according to studies such as the two referenced below.
As median-priced homes in more regions converge on fair value, declines in the Case-Shiller indexes are slowing materially. Although the annual declines are still bone-chilling, the month-to-month drops are a fraction of what they were a few months ago. For instance, the June decline in the 20-city index of 0.5% is the smallest decline in more than a year, and the trend is improving. June declines were a bit more than half of May's decline, and less than a quarter of the steep declines suffered from November 2007 through last March. What's more, home prices in nine of the cities actually increased in June, while seven went up in May. Eight metros increased in April, up from only two in March, and zero in the prior four months. Although some of the increase appears to be normal seasonality, we suspect something bigger is currently happening. As a result, we expect annual declines to meaningfully abate this fall when the index starts to lap easier comparables, provided the economy doesn't completely disintegrate before then.
Inventories remain a problem; there's no doubt about it. A look at the chart below shows the current overhang may even be more than our 1-million-unit estimate derived from the census data, as the current 3.9 million listed single-family units is 1.6 million units above the upper bound of a normally functioning real estate market, and as much as 2.4 million above the lower boundary.
But even here, there are some mild bright spots, as inventory growth has actually slowed to a crawl. Importantly, the chart below shows that the growth rate in inventories peaked in July 2006, more than a year before the 2007 peak in resets of the 2005 and 2006 cohorts of the now infamous "2/28" variable-rate subprime loans.
Local data also indicate that inventories have ceased their upward spiral. In fact, as the chart below shows, year-over-year inventories in September were reduced in almost 66% of the 55 MSAs tracked by housingtracker.net. This is up from just 6% in April 2007 and about 11% at the beginning of this year. Of course, seasonality may have something to do with it as sellers decide to take their homes off the market in the seasonally slow summer period. Even so, the large year-over-year decreases indicate there's likely more at work here.
So what explains slowing inventory growth in the face of mushrooming foreclosures? Well, it may be plausible that a good chunk of the "for sale" homes counted in the census and National Association of Realtors (NAR) data are, and have always been, those that were going to be foreclosed upon anyway. The net effect is nil once the bank reposes then relists the property. It's also worth noting that the dramatic drop in home starts, pictured below, has had a material countervailing effect to the crush of foreclosures and bank sales coming back to the market over the past several quarters. In fact, recent work done by Dr. Karl Case, Katherine Coman, and A. Barton Hepburn indicates that homebuilders have pulled back more this cycle than in past ones. They state that gross residential investment (all capital put into residential housing) was 3.1% of GDP in 2008's second quarter, a value that's lower than the 3.5 to 3.6% spent at each of the 1975, 1982, and 1991 cyclical bottoms.