Supply Constraints Won't Save San Francisco

"San Francisco prices will never go down because you can't build here." That was a popular theory when the housing market was hot. Even now, many people seem to believe that supply constraints will protect San Francisco from the nationwide housing bust. Never mind that San Francisco prices have already fallen almost 20% from their peak. The theory never made sense in the first place because it was based on muddled economics. Prices are determined by the interplay of supply and demand, not supply alone. You don't have to go far from San Francisco to see what I mean.

Between 2000 and 2007, the San Francisco housing stock increased by 11,300 units, or roughly 3.3%. During the same seven-year period, the housing stocks in Alameda, Marin, and San Mateo Counties increased by 4.5%, 3.2%, and 2.4% respectively. Those are small numbers, all of them in the same ballpark as San Francisco. If supply constraints alone were sufficient to prop up housing prices, we'd expect these other counties to have held up as well as San Francisco.

Here's what actually happened. According to the California Association of Realtors, as of September of 2008, the San Francisco market had fallen by 19% from its peak. Relative to their own peaks, Alameda, Marin, and San Mateo Counties had fallen by 37%, 29%, and 25%, respectively. In other words, despite similar rates of new supply, these markets all performed worse than San Francisco, Alameda much more so.

There's no mystery about what happened to the Alameda County market - it was swamped by a wave of foreclosures. (Foreclosure sales accounted for almost a third of Alameda County resale activity during August.) But that happened even though the supply of new housing in Alameda County was almost as limited as in San Francisco.

I'm not saying that San Francisco is due for a fall. But the notion that supply constraints will protect it from the housing bust is wishful thinking.

Note: The housing stock data were obtained from the Census Bureau.

Consumer Confidence Takes a Hit

Turmoil on Wall Street doesn't usually filter down to Main Street, but few people living have seen the kind of financial market upheaval that we've experienced in the last couple of months. It seems to be having an impact on Main Street. The Conference Board reported yesterday that consumer confidence fell to an all-time low of 38.0 in October. The one-month decline in the index (-23.4 points) was the third largest on record. You can read the press release here.

Funny Forecasts and Vested Interests

In my posting on October 23rd ("The Future(s) of San Francisco Real Estate"), I said I'm skeptical of real estate experts because they often have built-in incentives to present the market in a favorable light. Here's a good example, which I'd forgotten about until today.

In February of 2005, David Lereah, chief economist of the National Association of Realtors, published this little gem of a book: "Are You Missing the Real Estate Boom?: The Boom Will Not Bust and Why Property Values Will Continue to Climb Through the End of the Decade - And How to Profit From Them".

For the record, new home sales peaked less than six months later, in the summer of 2005. The Case-Shiller 20-market composite price index peaked the following summer, in July of 2006. I've made my share of bad predictions but Lereah's was spectacularly bad and spectacularly self-serving.

It's not that I don't trust any experts. I read and respect the work of serious academics. All others, I take with a grain of salt.

Housing Inventory Drives Price Movements

In the long run, housing prices are driven by fundamental variables such as income and population growth on the demand side, and land availability and construction costs on the supply side. These variables generally move in predictable long-term trends. The recent housing bubble suggests, however, that long-term trend variables are not the only factors that drive housing prices. If you're thinking about buying or selling a home within the next year or so, it would be nice to have a near-term price forecasting tool.

Take a look at the chart below, which compares the beginning-of-year inventory of single family homes in California to the change in median real price during that same year.

Inventory is stated as the number of months required to sell the current stock of for-sale housing, assuming that selling activity remains at current levels. I plotted it using an inverted scale in order to highlight the correlation with price changes.

The apparent strong relationship between the two series will be easier to understand if you consider what the inventory variable is really measuring. Months-of-inventory is a ratio, whose numerator is the number of houses currently for sale; in other words, the numerator is a measure of housing supply. The denominator is the current monthly rate of sales; in other words, it's a measure of housing demand. The combined ratio provides a snapshot of the balance between supply and demand. We shouldn't be surprised, therefore, to see a strong relationship between this ratio and subsequent near-term price movements.

Real estate practitioners often assert that the housing market is equally balanced between buyers and sellers when there is six months worth of inventory. The chart bears this out. In years when the beginning-of-year inventory was less than six months, the average increase in median real price was 11.6%. In years when the beginning-of-year inventory was greater than six months, the average increase was -1.7%.

This isn't a recipe for quick riches. For starters, housing bubbles are rare occurrences; it may be awhile before we see another year of 20% price increases. Remember also that buying or selling a home involves large transaction costs. When you account for those, a buy/sell strategy based on months-of-inventory is unlikely to yield substantial profits. Nonetheless, if you're planning to buy or sell a home within the next year, it seems wise to keep this basic supply/demand snapshot in mind.

The inventory of single-family homes in California was roughly 12 months at the beginning of 2008. The most recent figure was 6.7 months, as of August (although that’s not a seasonally adjusted figure). In light of what's happened in financial markets since then, my guess is that inventory has increased significantly. That suggests that home prices will be soft in the coming months. (Before you panic, remember that the average real price decline in high-inventory years was only 1.7%.)

By the way, you may be curious about the relationship between months-of-inventory and long term price changes. Months-of-inventory is sensitive to short-term factors such as interest rate fluctuations, recessions, and stock market turmoil. That’s why it correlates with near term price changes. For precisely the same reason, however, it is unlikely to predict long term returns.

The Future(s) of San Francisco Real Estate

What does the future hold for San Francisco real estate prices? I'm skeptical of 'expert' opinions, especially since real estate experts often have built-in incentives to present the housing market in a favorable light. (Real estate agents and mortgage brokers make a lot more money when the housing market is strong.) There's at least one group of opinion makers, however, who are financially incentivized to make accurate predictions about home prices. I'm talking about the people who gamble in the housing futures market.

The chart above shows the recent history of home prices for San Francisco and the greater Bay Area. The San Francisco data series (solid blue line) comes from Data Quick, and runs through August, 2008. The Bay Area data series (dashed purple line) comes from S&P/Case-Shiller, and runs through July, 2008. The purple diamonds come from the CME futures market, and represent futures market predictions for Bay Area home prices; that series runs through May, 2009.

Futures markets enable investors to sign contracts today for financial transactions that they intend to execute in the future. A typical futures contract locks in the price and terms of a transaction, but specifies that the money and goods are to be exchanged at a future date. You might enter into a futures contract if you were a farmer who wanted to lock in the sale price for some crops that you had just planted. If market prices fall between the time that you sign your contract and the time that you deliver your crops, you will have 'won' your futures bet. (Your contractually agreed sale price will be higher than the market prices that prevail at harvest time.) On the other hand, if market prices rise, you will have 'lost' your futures bet. (If you hadn't signed the contract, you could have sold your crops at a higher price.)

The housing futures market works much the same way. People who transact in this market are essentially making bets on the direction of housing prices. That doesn't make them experts. (The recent turmoil in the banking system should persuade you that even experts can be badly wrong about asset prices.) However, futures market players do have strong financial incentives to make accurate predictions about the direction of housing prices. Furthermore, these predictions take the form of publicly displayed and continuously updated futures prices, rather than fluffy language couched in caveats.

The latest futures prices imply that Bay Area housing prices will fall by another 10% between now and May, 2009. That would take housing prices back to the same level as at the beginning of 2002. (For what it's worth, that's largely consistent with other 'expert' opinions.)

By the way, the chart shows that San Francisco home prices have held up well by comparison with the rest of the Bay Area. All kinds of reasons could be offered to justify this state of affairs; I won't get into that right now. But before you dismiss what's going on in the rest of the Bay Area and say it-can't-happen-here, ask yourself if that doesn't sound like wishful thinking.

Note: The S&P/Case-Shiller indices can be found by clicking here, and the CME futures prices can be found by clicking here.

The Hidden Cost of Renting

I often hear people (especially real estate agents) say, "As a homeowner, you get the benefit of price appreciation." Technically, that statement is true but it needs some interpretation. For starters, how do you benefit from price appreciation if you never sell your house? If you live to 100 and die in the house that you bought 50 years ago, does that mean that you lost the homeownership gamble?

Not by a longshot. As a homeowner, the bulk of your housing cost (namely, interest on your loan) is fixed for life. As a renter, on the other hand, you can expect steady rent increases for the rest of your days. (Yes, I know San Francisco has rent control, but what if you decide to move?)

The chart below compares median rent to median household income for the Bay Area. There is obviously a strong correlation between the two data series. That's to be expected in a densely populated region like the Bay Area. As incomes rise, people naturally demand more housing. Zoning restrictions limit the supply of new construction, however, so higher demand leads directly to higher rents.

Over the 38-year period represented in the chart, rents and incomes both increased by more than 600%. In contrast, if you had been a homeowner for that same 38-year period, your only cost increases would have been associated with marginal items such as insurance and property tax. Those are not insignificant items, but they'll start to look that way after 20 or 30 years of 5% rent increases.

Now to be honest, rents probably won't increase at a 5% annual rate in the future. The Federal Reserve is unlikely to repeat the mistakes of the 1970's, so inflation is likely to be lower than it was in the past. Let's assume it will be 2.5%. If that were the end of the story, we would conclude that rents are likely to increase by 2.5% per year. However, median incomes in the Bay Area have generally grown faster than inflation, by about 1% per year. (That makes sense. Improvements in technology lead to higher productivity, which leads to higher income even after accounting for price inflation.) All told then, rents are likely to increase by 3%-4% per year. Over a 20-year period, that translates to a cumulative increase of somewhere between 80% and 120%. Over 30 years, the cumulative increase will be between 140% and 320%.

That's the hidden cost of renting. It's also the true benefit of owning, at least in a financial sense. While owners get to lock in the bulk of their housing costs when they close escrow, renters can look forward to a lifetime of rent increases. When projected out over 20 or 30 years, the impact of those rent increases will be measured in multiples, not percentages.

Stock Market Fallout

On October 13th, I said that stock market turmoil would force many prospective home buyers out of the market. It turns out, on the same day I wrote that, the online discount broker Redfin announced that they were laying off 20% of their staff. Here's what they said:

Today Redfin laid off roughly 20% of our employees.

Unlike other startups, our industry’s recession started a year ago, when home prices first plunged.

Since then, we’ve fought like starving animals, and with some success: while industry-wide transaction volumes dropped 33%, we grew revenues by nearly 50%. Traffic grew more than 300%.

Even a month ago, we were raising 2009 revenue projections. All our markets, now including Chicago, contributed profits.

But the past few weeks have seen a major reversal. As the stock market wiped out prospective down-payments, tours and offers dropped 30%. Transactions that were done came undone. October will still be pretty good, then we’re headed for a big dip.

http://blog.redfin.com/blog/2008/10/a_very_tough_day.html

Let's hope the stock market stabilizes soon.