Residential Sales Down 55% Since Bubble Burst

People are always asking me, "How's the market?" Here's the simplest answer I know. Take a look at the chart, below, which shows the annual value of residential sales in San Francisco over the last fifteen years.

Sale prices didn't peak until the spring of 2007, but sale volumes started falling in 2004 (see this earlier blog entry). Stated in that way, the historical trajectory of the market may be difficult to grasp. The story becomes simpler, however, when prices and volumes are combined: After peaking at $8.0 billion in 2005, residential sales have fallen in every year since. Through the first eight months of 2009, sales are running at an annualized rate of only $3.6 billion. In percentage terms, that amounts to a 55% decline since the housing bubble burst.

That's how the market is doing.

Job Seekers Outnumber Openings Six to One

Today's New York Times has a discouraging story about the job market. According to the Labor Department, there are 14.5 million people officially unemployed, but only 2.4 million full-time permanent jobs open. That means that job seekers outnumber openings by a ratio of six to one. By comparison, the jobs-to-openings ratio never even reached three-to-one following the dotcom bust. (Unfortunately, the Labor Department didn't start tracking job openings until 2000, so we don't know what the ratio might have looked like during previous recessions.)

As icing on the cake, the story adds that many companies have pared back working hours for their remaining employees. When demand begins increasing once again, they'll be able to expand output without creating any new jobs.

Stories of idle capacity at American companies are rife. Take a look at the chart, below, which shows manufacturing capacity utilization over the last 60 years.

Capacity utilization is running at a 60-year low of 65.8%. That's 15 percentage points below the historical average of 80.9%. Manufacturing represents only about 12% of GDP, but considering the many other examples of idle capacity, I'd say that the Times story is right. It's likely to be quite awhile before unemployment falls appreciably.

Home Prices and Land Values

At the risk of stating the obvious, San Francisco housing is expensive because San Francisco land is expensive. Take a look at the chart, below, which shows historical sale prices for single family homes and vacant residential lots in four of San Francisco's ten real estate districts.

The blue bars show average sale prices for residential lots ranging from 1,500 to 4,500 square feet. (I used the average instead of the median because the average seemed to give a truer picture of what was happening to 'typical' lot values.) The dashed purple line shows median sale prices for single family homes. The four districts represented in the chart accounted for 86% of San Francisco's residential lot sales over the last fifteen years (i.e., as far back as my data goes).

Evidently, lot prices are correlated with home prices. That's what you would expect. Construction costs don't change much from year to year, so the opportunity to build a new home becomes more valuable as home prices rise. That causes developers to bid up the price of buildable lots. (Conversely, the high price of buildable lots prevents would-be homeowners from avoiding high home prices simply by buying vacant land and building their own houses.)

Through the first eight months of 2009, the average residential lot price was $440,000, compared to a median home price of $600,000. In other words, 73% of the value of a typical home (in one of the four districts listed above) is attributable to the underlying land. At the peak of the housing market, land accounted for almost 80% of the value of a typical home.

The price difference between a home and a comparable vacant lot can be interpreted as the market value of the existing structure (i.e., the thing that we usually refer to as a 'house'). Take a look at the chart, below, which plots the difference between the median home price and the average residential lot price for the same four districts as before.

The series is volatile, but averages out to around $175,000 over the fifteen-year period represented in the chart. That's the value that the market is implicitly assigning to existing single-family structures.

$175,000 doesn’t go very far in San Francisco. Even the small, 1,250 square-foot houses that typically are found in these districts would cost more than that to rebuild. Why does the market seem to value them at less than replacement cost?

The answer is that they would not be replaced today. If it costs $300 per square foot to build a house that will sell for $600 per square foot, it’s clearly more profitable to build a big house than a small house. That’s exactly what developers have been doing. In the process, they’ve driven lot prices up to the point where building a small, old-fashioned house would be a money-losing proposition.

So how valuable is San Francisco land? It’s so valuable that you probably wouldn’t even consider rebuilding your current house if it burned down. You’d be destroying a great deal of value if you did.

Commercial Real Estate Is Experiencing Its Own Hangover

The housing market wasn’t the only sector of the economy to experience a bubble earlier this decade. The commercial real estate market went through a bubble of its own. Take a look at the chart, below, which shows the NAREIT price index for equity REIT’s.

A REIT is basically a real estate holding company. Provided that it distributes at least 90% of its earnings in the form of dividends, it is exempt from corporate income tax. The NAREIT index is an index of REIT prices, similar to the S&P 500 stock index.

REITs typically finance their investments with a mixture of debt and equity. Their share prices measure the market value of their equity, so they are imperfect proxies for the value of their underlying real estate holdings. On the other hand, because the shares are actively traded on public exchanges, the quoted prices provide a current snapshot of what’s going on in the commercial real estate market. (In contrast, popular housing market indicators such as the Case Shiller indexes are published only once a month, with a two-month lag.)

REIT prices rose by more than 100% over a four-year period beginning in 2003. Following the peak in early 2007, prices fell steadily for the next two years. During that period, REIT earnings actually increased at a healthy rate. Prices fell because investors were concerned that future earnings might be lower. In the fall of 2008, they were proved right in a spectacular way. Over the course of just nine months, earnings (and prices) fell by almost 50%. Take a look at the chart, below, which shows an index of dividends for the NAREIT index.

Dividends didn’t peak until mid-2008. Since then, they have fallen by 46%, and are now at their lowest level in more than 20 years. Remember, REITs are required to distribute at least 90% of their earnings in the form of dividends. So the decline in dividends will have been closely paralleled by a decline in earnings.

The dramatic fall in earnings suggests that REITs will have increasing difficulty servicing their debts. Here in the Bay Area, for instance, the delinquency rate on commercial mortgages rose from 1% in the second quarter of 2008 to 4% in the second quarter of this year. Nationally, the default rate on commercial mortgages rose from 1.2% in the second quarter of 2008 to 2.9% in the second quarter of this year. Forecasters are calling for a continued rise in the default rate, to around 4% by year’s end.

Even if REITs successfully meets their debt service requirements, they may still face serious financial problems as a result of declining earnings. Unlike residential mortgages, commercial mortgages typically have brief terms of only 5-10 years. When a commercial mortgage is due, it must be paid off in its entirety. The common solution is to obtain a new mortgage and use the proceeds to pay off the old one. If the value of the property has fallen, however, banks may refuse to lend as much as the borrower needs to pay off the old mortgage. In that event, the borrower may face foreclosure, even if he hasn’t missed a mortgage payment.

That’s the situation that owners of commercial real estate are facing now. I don’t have hard any numbers, but the press if full of stories about borrowers in good standing who aren’t able to rollover the mortgages on their buildings. (In this respect, REITs are actually better off than most other owners of commercial real estate, since they generally use less leverage.) The result is likely to be an increased level of forced sales, which will put additional downward pressure on property values.

The NAREIT index touched bottom in February, and has risen 44% since then. As with the housing market, however, the recent bounce is primarily attributable to a collective sense of relief that the world isn’t descending into a second great depression. After all, REIT earnings are still falling rapidly, and the credit crunch hasn’t shown many signs of easing. Commercial real estate values are likely to be under pressure for some time to come.

Haircuts and Consumer Spending

Last week, I used haircuts as an example in a posting about consumer spending cutbacks. It turns out that consumers really are trying to save money by getting fewer haircuts. According to today's Wall Street Journal, roughly 70% of salons have seen a dropoff in revenue, while sales of hair clippers have risen by 10%. The next time that the Federal Reserve decides to pump up the economy by buying bonds, maybe they should pay with haircut vouchers instead of cash.

Bay Area Home Prices Up Again

Case Shiller released their home price indexes for June on Tuesday. Prices rose in 18 of the 20 markets that they cover. The Bay Area was one of the strongest markets, with prices rising by 3.8% compared to the prior month.

Case Shiller publishes its indexes with a two-month lag. If you don't want to wait that long, Dataquick provides median sale prices, which they publish with only a one-month lag. They released their results for July even before Case Shiller had released their results for June. And while Case Shiller publishes only two indexes for entire the Bay Area, Dataquick publishes a median sale price for every zip code.

Median sale price is a poor way to measure price movements for a diverse market like the Bay Area. It is sensitive, not only to price movements, but to changes in the mix of homes as well. July's results are a case in point. The best-performing county in the Bay Area was Santa Clara, where the median sale price rose by 10% compared to June. Yet the median sale price for the entire Bay Area rose by 12%. The explanation for this odd result is that sales volumes fell in some of the less expensive counties, and rose in some of the more expensive counties. The Case Shiller indexes were specifically designed to eliminate this problem.

Having said that, I'm usually too impatient to wait an extra month for the Case Shiller numbers. I'm mainly interested in county-level numbers anyway. Here's a chart of median sale prices for San Francisco County.

If you put any stock in median sale prices, the San Francisco market bottomed out in January, and has risen 14% since then.

By the way, the commonly quoted Case Shiller indexes are for single family homes only. Focusing on that statistic can cause its own problems if you're interested in a market like San Francisco, where condos comprise such a large fraction of the housing stock.

Job Market May Be Stabilizing

The job market is showing signs of stabilization. After hitting a 25-year high of 9.5% in June, the national unemployment rate fell back to 9.4% in July . Bay Area unemployment rose by less than 0.1% in July, finishing the month at 10.8%.

Most economists are expecting a slow, multi-year recovery in the job market. It would be nice to think that we've finally begun that recovery. If so, then Bay Area home prices may finally have hit bottom.