Deflation: You Can't Have Your Cake and Eat It Too

"What's so bad about deflation? I'd love to pay less for the things I buy." I'll admit, that thought has occurred to me lately. I'm glad that gas costs half as much as it did a year ago. I'd be singing a different tune, however, if I were in the business of selling gas. The trouble with deflation is that almost everyone is in the business of selling something. For every buyer who benefits from lower prices, there is a seller who suffers. It’s clear, therefore, that a general reduction in prices is not an unqualified boon.

The concept of deflation is difficult because few of us have ever thought about it. So let's start with inflation.

If inflation is running at 2% when you buy your house, you'll probably pay about 6% interest on your mortgage. If inflation increases to 10%, your lender will increase the interest rate it charges on new mortgages to 14%. The extra 8% interest is intended to compensate the lender for the additional 8% erosion that it expects (annually) in the real value of its principal.

What happens if you get your loan when inflation is running at 2%, and then inflation increases to 10%? In that case, you get a windfall. Your salary will start increasing at a higher rate, in accordance with the higher rate of inflation. (You'll spend more money for haircuts, but the barber will have more money to spend in your store.) Your mortgage payments will remain fixed, however, so they’ll take up a lower percentage of your income than you had counted on.

Of course, your salary increases aren't 'real'. After all, they ultimately result from the Federal Reserve's decision to print more money. We know intuitively that putting more money into circulation won’t make us all richer, but it can make some of us richer. We just saw that an increase in the inflation rate will make it easier for you to service your loan, so you'll be a clear winner. On the other hand, your lender will lose out in the deal because your 6% interest payments won’t even compensate for the 10% annual erosion in the real value of your loan balance.

In short, a sudden increase in inflation results in a significant transfer of wealth, from your lender to you.

Now let's consider the opposite case, where the inflation rate falls from +2% to -10%. As before, there will be a transfer of wealth, but this time, it will be from you to your lender. Your salary will decline at a 10% annual rate, making it increasingly difficult to service your mortgage. Conversely, your lender will be receiving 6% interest on its principal, even while the real value of that principal is increasing at 10% per year.

The problem gets worse. Because the value of your home is likely to be falling (just like everything else), your lender will be getting anxious about its collateral. If you default on your loan (an increasingly likely outcome considering the circumstances), the proceeds from a foreclosure sale may not be enough to pay off the principal. That will make your lender increasingly reluctant to extend new credit against your home. In short, while your creditworthiness is deteriorating and your likelihood of default is increasing, some banker will be losing his job because his employer isn't doing enough mortgage business.

And so on. There's more to the story, but by now it should be clear that deflation means more than discounts at Wal-Mart.

Don't Count on Foreign Buyers to Rescue San Francisco Real Estate

When the housing crash started grabbing headlines last year, real estate insiders were hopefully espousing the theory that foreign buyers would prop up demand for San Francisco housing. The idea was that the declining dollar made US real estate cheaper for foreign buyers, who naturally preferred world class cities like San Francisco and New York. Unfortunately, the theory makes little sense. The reason for any sudden drop in the dollar is that foreigners are trying to reduce their holdings of US assets. Why, then, should they suddenly increase their demand for San Francisco real estate?

The dollar has indeed declined over the last several years, but the decline has been gradual and modest. A careful look at the facts suggests that the resulting impact on foreign demand for San Francisco real estate has been marginal at best.

Currencies rarely move in lockstep with one another: while one currency is appreciating relative to the dollar, another one might be depreciating. In such cases, how can we say whether the dollar is getting cheaper or dearer? Economists address this problem by using trade-weighted exchange rates. The idea is to create an index whose value is adjusted from period to period by applying a weighted-average of the percentage changes in a representative group of exchange rates. The weights are chosen to reflect the amount of trade that each country does with the United States.

Take a look at the chart, below, which shows the inflation-adjusted value of the dollar, measured against a trade-weighted basket of foreign currencies.

Remember that this 'exchange rate' is actually an index: a 10% increase in the level of the index indicates that the dollar has appreciated by 10%, but the level of the index itself is meaningless. (I obtained this series from the Federal Reserve. I re-scaled it so that the average of the index over the last 20 years is 100.)

The index reached its peak value of 117.7 in February of 2002. It reached its recent low of 88.6 in March of 2008. The peak-to-trough change was approximately -25%, which is indeed a sizable decline. Keep in mind, however, that most of this decline happened gradually, over a six-year period. In other words, it's not as if San Francisco real estate suddenly went on sale. The index did fall at an accelerated rate beginning in late 2007, but the cheap-dollar/foreign-buyer theory had taken root long before then. And when the sudden fall did occur, foreign buying slowed, just as I suggested above. (The National Association of Realtors reported that fewer Realtors were working with foreign buyers in August of 2008 -- when the dollar was near its low -- than in August of 2007.)

The latest index value is 98.6. That represents a discount of 16 percentage points relative to the 2002 peak. Again, that's a sizable discount, but it's not exactly a half-off sale. I'm not saying that foreign demand for San Francisco real estate has not increased, but a discount of that size seems unlikely to have had anything more than a marginal impact. And once you reject that simplistic explanation, the notion that foreign demand will prop up San Francisco real estate starts looking like wishful thinking.

Mortgage Rates Seem Likely to Improve

Here's an encouraging piece of news for the real estate market: The Treasury Department is working on a plan to push mortgage rates down to 4.5%. The details are being worked out, but the essence of the plan is that the Treasury will act as a bank: They'll borrow money from private investors (by selling Treasury bonds) and lend the proceeds to homeowners. As long as the loans get repaid, the Treasury will actually earn a profit, determined by the spread between its lending rate (4.5% in this case) and its borrowing rate.

The Treasury's borrowing costs have been coming down lately, as investors all over the world dump risky assets and pour money into Treasury bonds. Since December of last year, the yield on 10-year Treasury bonds has fallen from 4.1% to 2.7% (as of December 5th). That's the lowest rate in the last 50 years, by a good margin.

Meanwhile, 30-year mortgage rates have remained stable. When the subprime loan crisis started making headlines in 2007, mortgage rates were in the low 6%-range. They've fallen since then, but are still within 0.75 percentage point of their pre-crisis levels. In contrast, 10-year Treasury yields have fallen by almost three times that amount.

With Treasury yields falling and mortgage rates remaining stable, the interest rate spread between the two classes of securities has increased to levels that haven't been seen in the last twenty years. (This is one of the primary indicators of distress in the credit markets.) From 1987 to 2006, the average spread between 30-year mortgages and 10-year Treasurys was 1.6 percentage points. Today, the spread is 2.8 percentage points. It hasn't been that high since the early 1980's, when inflation was running rampant.

Anyway, back to the Treasury's plan. Assuming that Treasury yields remain below 3% (a good bet when inflation is running at close to 0%), a mortgage rate of 4.5% represents a spread of at least 1.5 percentage points. That's consistent with historical averages. So even if the Treasury plan is abandoned, there are good reasons to expect that private lenders will push mortgage rates down to the same levels.

Banks will have to be recapitalized in order for that to happen, but I don't think we'll have to wait for home prices to stabilize. (During the housing market downturn of the early 1990's, the mortgage spread actually fell below 1.5 percentage points.) Even if the Treasury plan is abandoned, there's a good chance that we'll see 30-year mortgage rates below 5% within the next year.

A Temperature Gauge for the San Francisco Real Estate Market

How is the San Francisco real estate market doing? Unfortunately, that can be a difficult question to answer. Take a look at the chart, below, which shows historical sale prices for two-bedroom condos.

I'd summarize this chart by saying that prices plateaued in the summer of 2005, and haven't shown any obvious trend since then. So much for the last three years, you say, but what about the last three months? Consider the chart, below, which shows the annual rate of change for two-bedroom condo prices.

Unfortunately, I find this chart even more difficult to summarize than the first one. If you asked me how the market is doing and this was all I had to go on, I'd probably wind up reading data points straight from the chart. That doesn't seem helpful, particularly if you're not interested in two-bedroom condos. There must be a better way to characterize the state of the market.

You may remember that when the housing frenzy was at its peak, overbidding and multiple-offers were common. Take a look at the chart, below, which shows the percentage of two-bedroom condos that were sold above the asking price.

I've also included the change in average sale price (taken from the second chart, above). The two series track each other fairly closely, which suggests that either one can be used as a proxy for the other. The advantage of focusing on bidding activity instead of price changes is that it provides a cleaner snapshot of the market.

If I say that prices have fallen 5% in the last three months, you'll probably want to know how much they changed in the three months before that. Or maybe you'll want to know the level from which they started, and whether I believe that that level was appropriate. Pretty soon, we'll be far afield from the original question, which was, "How is the market doing?"

Okay, here goes: During the last few months, only about 20% of the housing units sold in San Francisco have gone for more than the asking price. In contrast, when the dotcom and housing bubbles were at their peaks, 80% of sales were for more than the asking price. Except for a few months following September 11, 2001, the recent rate of over-bidding activity is at its lowest level in the last ten years.

In other words, the San Francisco housing market has cooled dramatically.

Fewer People Are Planning to Buy Homes

This article in today's New York Times is full of bad news for housing. Among other things:

• S&P/Case-Shiller released its home price indices for September. The 20-market national index is down by 17% from last September's level, and is down by 22% from its peak (which occurred in July, 2006). The California markets fared even worse. From September of 2007 to September of 2008, San Francisco, Los Angeles, and San Diego fell by 30%, 28%, and 26%, respectively. Each of the three markets is now about one third lower than it was at its peak. Considering recent activity in the housing and financial markets, the October and November index values will likely come in even lower than September's.

• According to the latest Conference Board consumer survey, just 1.9% of households are planning to buy a home in the next six months. There are roughly 117 million households in the US, so if we annualize the Conference Board's estimate, that works out to about 4.4 million home purchases in the next year. To put that in perspective, the National Association of Realtors reported yesterday that the seasonally adjusted annual rate of home sales was approximately 5.0 million for the month of October. (That's for existing homes only; the rate would be somewhat higher if we included new homes as well.) In other words, if the Conference Board survey is to be believed, the rate of home sales is likely to fall substantially from its already depressed level.

Bay Area Home Prices Still Falling

Bay Area home prices are down 40% in the last twelve months. That's according to last Thursday's press release from Dataquick. San Francisco was the best performing market, with a twelve-month price change of minus 12%. San Mateo County was second best, at minus 22%.

Perhaps a more interesting statistic is that 45% of all (existing) homes sold in the Bay Area during October had been foreclosed on within the last twelve months. That's an astounding number. In San Francisco, previously foreclosed homes accounted for a surprising 11% of total sales.

Bay Area Home Prices: Where's the Bottom?

One of the purposes of this blog is to shed some light on the notion of "fair value" for Bay Area housing. Today's entry focuses on the relationship between home prices and household incomes.

I'll adopt the naive assumption that over the long haul, home prices and household incomes should move in tandem with one another. It's easy to come up with reasons why that assumption might fail, but it seems like a good starting point; in any event, it has worked pretty well in the past.

Consider the chart below, which shows home prices and household incomes for the three-county region that includes Marin, San Francisco, and San Mateo Counties.

The pink diamonds represent an index of median household income. The solid blue line is the OFHEO home price index. (OFHEO stands for Office of Federal Housing Enterprise Oversight.) It's not a perfect index (among other things, it includes only single family homes), but it correlates closely with other common home price indices, and it's a particularly convenient index for purposes of this analysis. Both series are indexed to a value of 1.0 in 1999.

As you can see, the two series tracked each other closely over the 25-year period ending in 2000. After that, prices climbed much quicker than incomes. By 2006, the ratio of home prices to household incomes exceeded its pre-1999 level by about 80%. The ratio has fallen dramatically since then, but is still about 45% higher than its long-term average.

Notice that I haven't talked about what the price-to-income ratio should actually be. Instead, I've pointed out that the ratio has been stable in the past, and that it's now far above its long-term average, even after the recent price declines. As I said, it's easy to come up with arguments as to why the "fair" level for the ratio should have risen. But I don't know if those arguments can stand up to 25 years of history.

Note: The household income figures were obtained from the Census Bureau. You can find the OFHEO home price index here.