Bay Area Mortgage Defaults Holding Steady at High Levels

Bay Area mortgage default rates have fallen from the peak levels recorded in 2009, but seem to have stabilized at relatively high levels.

During the third quarter of 2010, there were 483 default notices recorded in San Francisco, and 12,690 recorded in the nine-county Bay Area. Those figures are more than six times higher than the corresponding figures from the third quarter of 2005.

The recent stability in default activity is at least partially due to forbearance on the part of lenders, who have been under political pressure to help homeowners avoid foreclosure. (Remember, we're not talking about actual defaults, but rather about notices of default, which must be recorded by lenders.) A substantial part of the stability, however, is probably due to the stabilization of Bay Area home prices.

I've talked about the relationship between price declines and default rates in earlier postings, for instance, this one. (The key point is that being underwater on your mortgage makes it tempting to surrender your home to foreclosure.) Now that prices seem to have stabilized, the number of underwater homeowners should begin to fall, mainly due to foreclosures and short sales. That should shortly translate into lower default rates.

Homeowners Are Still Badly Stretched

I said in my last post that homeowners are financially stretched. Indeed, even after three years of economy-wide debt reduction – unprecedented in modern history – households are still carrying financial burdens that would have seemed astronomical before the housing boom. Take a look at the chart, below, which shows the ratio of aggregate household debt to aggregate household income.

Until 1999, household debt had never exceeded 80% of household income. When the housing boom began, however, mortgage debt expanded dramatically, driving the debt-to-income ratio as high as 116% in 2007. The ratio has fallen since then (mainly because of mortgage defaults), but is still more than 30 percentage points higher than at any time before the boom.

Another way to look at debt burdens is to compare debt servicing costs to household incomes. (The chart above uses the total amount of debt rather than the cost of servicing that debt.) Mortgage rates are at 50-year lows, so it’s possible that homeowners can carry substantially higher levels of debt without greater difficulty.

Take a look at the chart, below, which shows the ‘Financial Obligations Ratio’ for homeowners.

The Financial Obligations Ratio is the ratio of aggregate housing costs (including property tax and insurance) to aggregate disposable income. Although not as far out of line compared to historical levels as the debt-to-income ratio, the financial obligations ratio is still higher than at any time before the housing boom, including the peak of 1980’s housing cycle.

Mortgage rates are likely to remain low until a sustainable economic recovery is underway. They're bound to rise eventually, however, driving the financial obligations ratio higher and increasing the pressure on home prices just as the economy begins to recover. All things considered, I’d say that the housing market is likely to remain in the doldrums for years to come.

Note: The ratios shown in the charts are based on aggregate figures, and are only indirectly representative of the debt burdens of typical households. The debt-to-income ratio for a typical first-time homeowner would have been far higher than 116% at any point during the thirty year period shown in the charts.

Too Many Homeowners?

One of the most remarkable facts about the housing bubble was the large number of renters who became homeowners. Take a look at the chart, below, which shows the nationwide homeownership rate.

The data comes from the Census Bureau, which defines the homeownership rate as the percentage of households that are owner-occupied.

The homeownership rate began increasing rapidly in 1995 and reached an all time high of 69.0% in 2004. The peak value was almost five percentage points higher than the pre-bubble average (64.3%), and more than three percentage points higher than the previous record (65.8%). The homeownership rate has declined steadily since the bursting of the housing bubble but the current rate (66.9%) is still more than two percentage points higher than the pre-bubble average.

We can't rule out secular changes that might be driving long-term increases in the rate of homeownership (although it's clear that the rapid increase between 1995 and 2004 was not attributable to long-term factors). Perhaps the aging population is causing a secular rise in the rate of homeownership. If so, then perhaps the rate will settle down at a level that's higher than the pre-bubble average. Considering the state of the economy and the over-stretched finances of US homeowners, however, I think we'll be lucky if the homeownership rate doesn't fall substantially below 64.3% during the next few years.

San Francisco Apartment Valuations Are Still Falling

San Francisco apartment building valuations have been falling since 2007. Take a look at the chart, below, which shows the median price-per-unit for 5-15 unit buildings.

The median price-per-unit actually rose in 2010, and is only about 10% lower than in 2007. How is that consistent with my assertion that valuations have been falling?

Despite the downturn, rents have remained fairly strong in San Francisco. Take a look at the chart, below, which shows the median rent-per-unit for 5-15 unit buildings.

San Francisco rents actually have been trending upward since the depths of the financial crisis, and appear to be in line with the long-term trend. As a result, if you compare apartment prices to apartment rents, you get something like the chart below, which shows the median 'gross rent multiple' for San Francisco apartment buildings.

The gross rent multiple (GRM) is the ratio between the building price and the gross annual rent. It's analogous to the price/earnings (P/E) ratio for a stock. If the GRM is falling (as it has been since at least 2007), that means that buyers are paying less for every dollar of rent that they expect to obtain from the building. That's what I mean when I say that valuations are falling. (Interestingly, the number of transactions began falling well before there was a noticeable decline in valuations. The same phenomenon occurred in the market for single family homes and condos, as I pointed out here.)

There are many possible explanations for the observed decline in GRM multiples. Ultimately, they all can be attributed to diminished optimism about the future of San Francisco real estate.

San Francisco Lot Prices Still Falling

San Francisco lot prices peaked in 2006 and have been falling since then. The median sale price through the first nine months of 2010 is more than 60% lower than the 2006 peak.

The chart shows the median sale price for all lots that are listed in the MLS and that have 1,500-4,500 square feet. Most of them are residential lots. Unfortunately, many listings do not include zoning information so I simply included all lots.

Land prices are highly sensitive to home prices. Why? Building costs don't change much from year to year. If the price of a home rises by $100,000, the price of the vacant lot next door is likely to rise by the same $100,000 increment. On a percentage basis, the increase in the lot price will be much larger than the increase in the home price. That's why lot prices have fallen more than 60% while home prices have fallen only 10%-20%.

Keep in mind, too, that most lots are priced under the assumption that they will be developed with multiple units. That's why the median lot price has fallen by roughly $500,000 while the median home price has fallen by only around $150,000.

Office Rents Back to 2006 Levels

Here's another quick snapshot of the business climate in San Francisco. Asking rents for office space have fallen 24% from the recent 2008 peak.

The chart comes from Loopnet; you can find more like it here. Ideally, the chart would extend back another ten years to encompass the dotcom boom period. I don't have a free source of commercial property data, however, so I have to settle for odds and ends like this chart.

Judging from the chart, the office market stabilized early this year, at a level that hardly seems catastrophic -- asking rents have returned roughly to the same level as in 2006, when there wasn't even a whiff of financial crisis. That observation is particularly surprising in light of the fact that San Francisco unemployment was running in the low 4%-range during 2006, and is now close to 10%.

Bay Area Housing Permits Near 30-Year Low

Permits for construction of new Bay Area housing units hit a 30-year low in 2009. Activity has rebounded since then, but 2010 is still on track to be the second-worst year for Bay Area housing construction since at least 1980.

I annualized the July year-to-date figures to get an estimate (roughly 7,000 units) for calendar 2010. That's less than half of the 30-year average of around 19,000 units. Although I didn't show it in the chart, the pattern of permit activity is nearly identical in each of the three main metropolitan areas of the Bay Area, i.e., Oakland, San Francisco, and San Jose.

Hotel Room Rates at Lowest Levels in Twenty Years

Here's a quick snapshot of the business climate in San Francisco. The average daily rate for San Francsico hotel rooms is at its lowest level (in inflation-adjusted terms) in twenty years.

Note: I've used the Consumer Price Index to convert historical room rates to 2010 dollars.

Occupancy rates have actually been running well above average. In that kind of environment, room rates usually rise quickly. My guess is that the mix of hotel guests has changed during the recession, with price-sensitive tourists replacing high-paying business travelers. If so, then San Francisco's business climate is soft indeed.

Foreclosure Activity Has Limited Impact on Home Prices

Homeowners who are underwater on their mortgages often make ‘strategic’ decisions to walk away from their homes. Take a look at the chart, below, which is an updated version of one that I first presented in June 2009.

The vertical axis shows the annualized default rate for the three month period ending in August, for each of the nine Bay Area counties. The horizontal axis shows the percentage change in price through May, measured relative to the three-year period from July 2005 to June 2008 (i.e., the bubble years).

Note: I reversed the horizontal axis to make the chart easier to read. Ideally, it would show the percentage of homes that are underwater. I don’t have detailed data on underwater homes, however, so I used the change in price relative to the bubble years as a proxy for the percentage of homes that are underwater. You can read more about this proxy for underwater homes here and here.

The relationship between the two variables is clearly strong, and supports the idea that underwater homeowners are making cold-blooded decisions to default, simply because they believe it’s in their interest to do so.

On the other side of the coin, there seems to be a widespread belief that heavy foreclosure activity depresses sale prices. If that’s true, then falling prices can lead to a self-reinforcing “doom loop” where underwater homeowners walk away from their homes, which then hit the market as foreclosures and depress prices even further.

Perhaps that story holds true in other markets, but there isn’t much support for it here in the Bay Area. Take a look at the chart, below, which compares recent price changes to REO sale activity for the nine Bay Area counties.

The vertical axis shows the change in median price over the last twelve months. The horizontal axis shows the number of bank-owned homes (REO’s) that have been sold during the same period.

REO’s accounted for more than 30% of all sales in five of the nine Bay Area counties. All but one of these counties outperformed San Francisco, where REO’s accounted for only 15% of sales. I wouldn’t call this an ironclad case but I don’t think Bay Area homeowners need to worry about foreclosures depressing property values.

Inventory Model Suggests California Housing Market Has Stabilized

Historically, the inventory of unsold homes has been a good predictor of near-term price movements. I first wrote about the correlation between these two series in October of 2008. At the beginning of that year, the unsold inventory of single family homes in California had been equivalent to 14.3 months of supply. That was well above the historical median of 6.3 months, suggesting that 2008 would be a weak year.

In fact, 2008 turned out to be an abysmal year, largely because of the developing financial crisis. Home prices already had fallen 27% by August (the most recent month available at the time of my posting) and went on to record a 41% decline for the year.

I’ve updated my October 2008 chart through July of 2010. The chart compares the beginning-of-year inventory of single family homes in California to the change in median real price during that same year. As you can see, with the notable exception of 2008 and to a lesser extent, 2007, the beginning-of-year inventory has been a very good predictor of subsequent price movements.

The California housing market seems to have returned to equilibrium over the last year or so, at least as far as the balance between supply and demand. 2010 began with an unsold inventory equivalent to only 4.1 months of supply. That's somewhat lower than the historical median, suggesting that 2010 will actually see modest price increases. In fact, the median price of a single family home in California had increased by around 5% through the end of July (the most recent month available). Even with the recent drop in sales activity (following the expiration of the first-time buyer tax credit) the unsold inventory of homes in July was still only 5.8 months. That's in line with the historical median, suggesting that the near-term outlook for California home prices is reasonably good.

Note: I plan to do a similar analysis for San Francisco, but it will take some time. The unsold-inventory data that I had been using for San Francisco appears to be unreliable.

San Francisco Rents Are Firming Up

Rents in the City appear to be firming up. Take a look at the chart, below, which shows a history of the median asking rent per room in San Francisco.

The chart comes from SFRentStats. Unfortunately, I don't have a free source of actual (contract) rents, but asking rents should provide a reliable proxy.

The Bay Area job market stabilized in the spring of 2009. It's no surprise, therefore, that the rental market is also showing signs of stability. And because prospective home buyers generally make cost comparisons between renting and owning, stability in the rental market tends to foster stability in the purchase market. That's one of the reasons why I don't think home prices will fall appreciably unless the job market takes another dive.

Bay Area Home Prices Are Within 15% of Long-Run Equilibrium

My favorite way to assess 'fair value' for home prices is to compare them to rents. On that basis, Bay Area home prices appear to be about 15% higher than they 'should be'. Take a look at the chart, below, which shows Bay Area home prices, rents, and median household incomes for the period since 1975.

Note: You can find descriptions of the data series here.

After briefly reaching a level consistent with long-run equilibrium, Bay Area home prices began climbing again in the spring of 2009. Prices have increased by about 18% since then, and are now somewhere between 10% and 15% above equilibrium.

In my last post, I said that prices probably won't fall appreciably as long as the unemployment rate holds steady. How is that consistent with prices being 15% above equilibrium?

There are two ways for the relationship between prices and rents to adjust: prices can fall or rents can rise. Historically, rent increases have been the dominant adjustment mechanism. For instance, between 1990 and 1996, the price-to-rent ratio fell by 21 percentage points but price declines were responsible for only 8 percentage points of that adjustment; rent increases accounted for the rest. I expect a similar pattern of adjustments this time around.

Will Home Prices Fall Again?

The question I'm hearing most often at the moment is, "Do you think home prices will fall again?" My answer is, not unless the job market takes another dive. Take a look at the chart below, which shows the Case Shiller home price index and the unemployment rate for the Bay Area.

The current recession is unusual because it was brought on by falling home prices. Financial institutions and households both were highly leveraged at the peak of the housing market. As a result, their balance sheets suffered serious damage when home prices fell. Economic demand naturally fell, leading to the current recession. (As you can see from the chart, home prices began falling rapidly before unemployment started climbing. That's consistent with the idea that falling home prices caused the recession.)

That's not the normal pattern. Normally, job losses lead to falling home prices, as in the last recession. When the job market stabilizes, home prices tend to stabilize as well. That's what happened in the last recession and it appears to be happening again. The job market stabilized around June of 2009 and since then, Bay Area home prices have increased by roughly 15%.

I wouldn't bet on continued price increases, but as long as the unemployment rate holds steady, my guess is that home prices won't fall appreciably from here.

Note: I didn't have time to make this case as strongly as I'd like. I'll have more to say about it soon.

New Home Sales Hit Record Low, Builder Confidence Suffers

Sales of new single family homes fell to 276,000 units in July, on a seasonally adjusted annual basis. That's the lowest sales total on record, going back to 1963.

Sales in the west region came in at 44,000 units in July, on a seasonally adjusted annual basis. That's also a record low, going back to 1973. Sales had been improving until the end of April, when the home buyer tax credit expired.

Note: Sales of new homes are recorded when contracts are signed. For existing homes, however, sales are recorded when transactions close. That's why existing home sales fell only modestly in May (following the April 30 expiration of the tax credit), while new home sales fell abruptly.

Not surprisingly, builder confidence is suffering. After hitting a record low of 8 in January 2009, the NAHB Housing Market Index had risen as high as 22 in May 2010. The index has fallen back sharply since then, however, to it's current level of 13.

Evidently, builders' hopes were buoyed by tax-credit-induced sales. Did they not see the end of the tax credit coming?

Expiration of Tax Credit Hits Bay Area Housing Market

The National Association of Realtors announced yesterday that existing home sales were 27% lower in July than they had been in the prior month. The annualized sales rate of 3.83 million units was the lowest since the NAR began keeping records in 1999.

Sales had been rising rapidly as the April 30 expiration date for the home buyer tax credit approached. The subsequent drop-off in sales was therefore widely anticipated. (The same pattern was observed with the original home buyer tax credit, which expired on November 30, 2009.) Still, the scale of the decline came as a surprise. In a survey of economists by Bloomberg News, nobody had predicted such a large decline.

Housing is far more expensive in San Francisco than in other parts of the country. Consequently, the tax credit might have been expected to have a relatively small impact on demand. Surprisingly, however, home sales fell almost as much in the City as they did in other regions. Between June and July of 2010, home sales fell roughly 21% in San Francisco. The July 2010 total (452 units) was 17% lower than the corresponding figure for July 2009 (543 units).

Housing Market Slows Following Expiration of First-Time Buyer Credit

The housing market has slowed dramatically following the April 30th expiration of the first-time home buyer tax credit. Take a look at the chart below, which shows the NAR's Pending Home Sales Index.

The index tracks the number of purchase contracts that are signed in each month. It is normalized that the average for 2001 is 100.

In order to qualify for the tax credit, buyers had to be in contract by April 30th. There was some question as to how large an impact an $8,000 credit would have on buying activity. Now we know. Nationwide, pending home sales fell 30% from April to May. In the western region, pending home sales fell 21%.

Anecdotally, the San Francisco housing market seems to have slowed as well. I'll address this in an upcoming posting.

Mortgage Defaults May Have Peaked

Almost 20,000 Notices of Default (NOD's) were recorded in the Bay Area during the second quarter of last year. Default activity has fallen rapidly since then. In the fourth quarter (the latest available), the number of NOD's came in more than 30% lower, at roughly 13,600.

San Francisco County has experienced a similar decline in default activity. There were 607 Notices of Default recored in the City during the third quarter of 2009. The fourth quarter figure came in almost 25% lower, at 465.

Note: The abrupt decline in NOD's during the third and fourth quarters of 2008 resulted from lender uncertainty in the face of legislative changes. Keep in mind that the chart shows recorded Notices of Default, not actual instances of default.

Two quarters don't make a trend. But the recent stability in home prices and employment suggest that mortgage default activity will continue falling.

Bay Area Housing Market Recovery Continues

Bay Area home prices touched bottom in April of last year, and now appear to be almost a year into recovery. According to the Case Shiller indexes, single family home prices are 15% higher than they were at the bottom of the market. They've increased in every month since then, albeit at a slowing pace.

The Bay Area condo market also has bounced back, but its recovery has been less impressive. Condo prices are only about 7% higher than they were at the bottom of the market.

Job Market Stability Brings Relief to Housing Market

Since peaking at 11.0% in August 2009, Bay Area unemployment has been trending slowly downward. January's reading of 10.3% is the lowest since May of last year. Nationwide unemployment peaked at a (seasonally adjusted) rate of 10.1% in October 2009. January's reading came in slightly lower at 9.7%.

Most economists expect unemployment to remain high for the remainder of 2010. That's not good news for the housing market (especially in areas with a lot of underwater homeowners), but stable unemployment is far better than rapidly rising unemployment. Take a look at the chart, below, which shows the recent trajectories of home prices and unemployment in the Bay Area.

Bay Area home prices remained stable until the job market began to deteriorate. Prices then fell rapidly as unemployment rose, but stabilized once again when unemployment plateaued. (A similar course of events played out during the 2001 recession, although the current price decline has been far more severe.)

Some smart economists believe that the U.S. economy is not yet out of the woods, and could slip back into recession later this year. Let's hope that it doesn't work out that way. If so, then Bay Area housing prices may have reached a point of stability.

End of Fed Program Doesn't Necessarily Imply Trouble for Mortgages

The Federal Reserve has taken emergency measures to keep the recent financial crisis from spiraling out of control. One such measure, announced last March, was to purchase almost a trillion dollars worth of mortgage-backed securities from Fannie Mae and Freddie Mac. The financial crisis had severely impaired banks' capacity to provide credit, so the Fed stepped in to support the housing market.

The announced program of Fed purchases is almost complete, and is scheduled to end in March. The credit markets have stabilized over the past year, but housing market observers nevertheless are anxious about what will happen to mortgage rates when the Fed withdraws its support. The anxiety is understandable. The Fed purchased 73% of the mortgages that Fannie Mae and Freddie Mac turned into securities during 2009. Withdrawal of that kind of support seems likely to cause at least some disruption.

As I’ve pointed out in the past, mortgage rates are closely tied to Treasury rates. Take a look at the chart, below, which shows the interest rate spread between 30-year fixed-rate mortgages and 10-year Treasurys.


In the years leading up to the crisis, the mortgage spread remained close to its historical average of 1.6 percentage points. It rose dramatically as the crisis unfolded, but has since fallen to less than 1.3 percentage points. A casual analysis would therefore suggest that mortgage rates are likely to rise when the Fed program ends in March.

At least one major bond market player (PIMCO) is betting that mortgage rates will indeed increase soon. Bill Gross (PIMCO's managing partner) has an outstanding record of accurate interest rate forecasts. That's why PIMCO has become the world's largest bond fund manager; if you're looking for investment advice, take theirs. Before you decide that mortgage rates are headed for an abrupt increase, however, take a look at the chart, below, which shows a longer history of the mortgage spread.


Note: This chart uses annual averages, which smooth out some of the volatility that is evident in the monthly chart above. That’s why the maximum spread shown in this chart is lower than the maximum shown in the monthly chart.

The mortgage spread remained below 1.6 percentage points from 1991 to 1997. (It’s worth pointing out that this period coincided with the last significant housing market slowdown, and followed the S&L crisis of the late 1980’s.) The average spread during this period was 1.35 percentage points, which isn't far from the current spread. Perhaps PIMCO is correct about the likely course of mortgage rates, but I wouldn't look for a dramatic near-term increase.

(I’m not saying that you should wait to refinance. Today’s mortgage rates probably will look like a bargain a few years from now.)

Builders Are Optimistic - as Usual

On Tuesday, the National Association of Home Builders released the February edition of its Housing Market Index. The HMI rose by two points compared to January, but remains in record-low territory.


Despite the protracted downturn in the housing market, builders have remained steadfastly optimistic. Indeed, they've actually been expecting conditions to improve for the last three years. Take a look at the chart, below, which shows the 'current conditions' and 'expected conditions' components of the HMI.


The 'expected conditions' component of the HMI (which measures builders' expectations for the next six months) has remained roughly ten points higher than the 'current conditions' component since October 2006. Throughout most of the period in question, however, conditions actually deteriorated. Evidently, builders keep expecting the future to be better than the present, but the future keeps receding.

Take What "They Say" with a Grain of Salt

I love this story. Three years ago, the Mortgage Bankers Association announced plans to buy a new, $90 million headquarters building in Washington, D.C. According to Jonathan Kempner, then president and CEO of the Association, "...we have come to the inescapable conclusion that owning our own building was the smartest long term investment for the Association." The Association moved into its new headquarters in June 2008.

On February 5th, 2010 - less than two years after moving in - the MBA sold their building to CoStar Group for $41 million. In so doing, they effectively acknowledged a colossal investment blunder. Plenty of people overpaid for real estate during the boom. But the MBA was supposedly a group of financial experts, who had access to a wealth of intelligence about the real estate market. That's why I cringe whenever someone begins a sentence with the phrase, "They say..."

The Blog Is Back

The blog is back. It's been more than two months since my last posting but I expect to be posting regularly once again.

Since this is my first posting of 2010, let's start with prices. Take a look at the chart, below, which shows home price indices for San Francisco as well as the nine-county Bay Area.

The San Francisco index comes from Dataquick and includes single family homes as well as condos. The Bay Area index comes from Case Shiller and includes single family homes only.

San Francisco prices plateaued in the summer of 2005 and remained fairly stable until the summer of 2008. Then, during the six-month period ending in January of 2009, prices fell by more than 25%. They bounced back strongly in February and have remained fairly stable since then, at a level that's only about 15% below the peak.

Prices in the greater Bay Area also plateaued in the summer of 2005, but remained at peak levels for only two years instead of three. The subsequent price decline was both longer and more dramatic than in the case of San Francisco. Over the 18-month period ending in March of 2009, Bay Area home prices fell by a whopping 45%. Home prices (along with stock prices) bounced back strongly when it became clear that the economy wasn't sliding into depression. Even so, Bay Area home prices are still about 35% lower than they were at the peak.

Perhaps there's an economic justification for the strong relative performance of the San Francisco market, as compared to the greater Bay Area . I suspect, however, that the main reason why San Francisco has held up better than the greater Bay Area is that San Francisco homeowners generally have more financial staying power than their counterparts in the greater Bay Area.