Housing Market Projections

Our National Report is out this week. From the press release:

Housing Prices Down by 2.1% in January

Inventory levels

Mountain View, CA February 9 2009. The Altos 10 City Composite Price Index showed a decline in asking prices of 2.1% in January and 2.4% during the most recent three-month period. Prices of properties listed for sale fell in 23 of 26 major markets according to the Real Time Housing Market Report, jointly published by Altos Research, the premier source for real-time real estate data, and market analysis firm Real IQ.

Asking prices fell at the fastest rate in Las Vegas, down 5.4% during January, and 8.8% during the past three months. This marks the tenth consecutive month that Las Vegas has posted the fastest rate of declining prices among major markets. Listing prices rose at the fastest rate in Miami, up 1.5% in January. Price increases in these markets are functions of a shift in the class of properties rather than traditional buyer demand.

"Despite continuing decreases in inventory levels, asking prices remained on a steep downward path in January," said Michael Simonsen, CEO and co-founder of Altos Research. "The trends appear to be continuing. We can find no signs of a turn-around in the data as of the first week of February."

Inventory levels declined in most major markets with the exception of Portland, Seattle, Charlotte and Salt Lake City. Across the 10 City Composite Index markets, inventory declined by 3.3% in January and 9.7% during the most recent three-month period. Inventory fell by more than 5% in Detroit, Boston, Atlanta and Cleveland during January.

"Inventory levels generally increase in January as sellers anticipate the seasonally strong spring selling season," said Stephen Bedikian, partner and research director for Real IQ. "So far we are seeing the opposite as inventory levels are continuing the decline they began in the summer of last year. The big test will come over the next few months as any increase in inventory will exacerbate the already large imbalance of housing supply and demand by adding to the existing overhang of available homes."

The median days-on-market rose in all markets during January staying above 100 or more days-on-market in every major market. By far, the market with the slowest rate of inventory turnover was Miami at a median of 189 days-on-market or more than six months. Miami has experienced the slowest market turnover in every month since September 2007. Salt Lake City enjoyed the fastest rate of turnover with a median days-on-market of 101.

The report is free, register here to download.

Our latest article with Forbes.com is out. This one was a toughie. Their editors called and asked, "Where are the best selling suburbs for sellers right now?" It's a tough question because the answer, really, is no where.

Sure the best properties, properly priced, in the best neighborhoods, will find their buyers. In fact we documented surprising resiliency in some markets through the summer. But even though prices might be holding, demand continues to slip. By our Market Action Index, there are essentially no markets with demand levels high enough to call them "Seller's Markets". We settled on identifying ten suburbs whose demand trends (as measured by Market Action Index and days on market) simply weren't horrible.

My tongue-in-cheek quote for Matt at Forbes was,"If you have to sell your home, do it in one of these suburbs."

Full article at Forbes.com

For years I've written about the function of NASDAQ stock wealth on prime Silicon Valley real estate. I like to summarize with this phrase, "You don't buy a $2 million home in Palo Alto with a $1.8 million sub-prime loan. You buy it with $1.35 million in NASDAQ capital gains and a $650,000 in financing."

That fact has buoyed real estate demand and prices from Saratoga through San Francisco in the past two years - even as the lower end of the market has been pummeled.

Consider this: two companies (and their stocks), Apple (NASDAQ: AAPL) and Google (NASDAQ: GOOG), minted several thousand local millionaires between 2004 and 2007. (The back-of-the-envelope calculation works like this: Of GOOG's 16,000 employees assume 5,000 locals are mostly vested. Assume they collectively own maybe 8% of the company. At GOOG's 2007 peak of $711/share that's $18 billion in local wealth created since 2004. AAPL has a similar scale from an $11 stock in 2004 to $190 peak, spread over maybe 8000 vested local employees, is an additional $13 billion of local capital gains wealth created in the preceding four years.)

Many of those newly wealthy want to buy a home. The result is that simply from Google and Apple you have hundreds or even thousands of buyers at the high-end of the market.

Couple this demand with the peculiar small-supply situation of the local region (limited new construction yields only a handful of available properties) - and you get the recipe for resisting the bursting bubble right up through the first half of 2008.

Local Realtors know that the supply side of the equation is not likely to change drastically. There are simply no homes being added in close-in Silicon Valley unlike, say suburban Phoenix.

So the recipe depends on capital-gains-wealth-driven demand.

And guess what? Year to date for 2008, both GOOG and AAPL are down by half. In the last 9 months or so, $15 billion of local wealth in just those two stocks has evaporated.

The bubble burst has only started to hit us.

[note: I am aware this is an over simplified argument. Home prices do not correlate with the movements of any one stock. We've checked. In some of these markets, they do correlate with movements in the NASDAQ as a whole, and we all know this isn't an isolated dip.]

Here are the Case Shiller Index numbers released today for July. No big surprise. Down of course. One percent more this month. Down 17% from July 2007. As I've been saying in the press lately, we've seen no signs of a bottom. Even before the crisis of the last couple weeks, pricing through September continues to be down sequentially.

As we've already measured, the next three months of the Case Shiller numbers continue their fall at the same pace. Price decreases through September have not accelerated. There are no signs of market inflection points yet either. Days on Market is climbing and Inventories are flat at best. Contact us if you want specifics on the CSI for a given future date or a given MSA market.

MSA	CSI July 2008	Change from June 2008	Year over year change
Boston	162.58	0.17%	-5.35%
Chicago	149.6	-0.35%	-9.95%
Denver	132.67	0.77%	-4.71%
Las Vegas	154.15	-2.75%	-29.90%
Los Angeles	192.55	-1.63%	-26.18%
Miami	186.84	-1.60%	-28.25%
New York	192.92	-0.76%	-7.41%
San Diego	172.2	-1.81%	-25.02%
San Francisco	156.88	-1.85%	-24.81%
Washington DC	195.49	-1.07%	-15.80%
10-City Composite	178.46	-1.09%	-17.49%

Our National Report for September data comes out later this week with the current view on the market rather than the backward looking stuff released today. Stay tuned for details

Our latest National Housing Market Report is out. This one examines data through end of July 2008. You can download the PDF here.

Here's the press release:

Real Estate Report: National Housing Prices Down by 0.8% in July

Report measures current real estate data in major markets around the country. Data shows Las Vegas continuing to lead as the weakest US housing market with real estate prices down another 4% in July. Current real estate data shows some summertime price stability in Midwestern markets Detroit, Cleveland, Indianapolis, and Minneapolis

MOUNTAIN VIEW, CA August 5, 2008 - The Altos 10-City Composite Price Index showed a decline in asking prices of 0.8% in July and 1.3% for the past three months. Prices of properties listed for sale fell in 13 of 26 major markets according to the Real-Time Real Estate Report, published by Altos Research, the premier source for real-time real estate data, and market analysis consultancy Real IQ.

Asking prices fell at the fastest rate in Las Vegas - down 4.0% during July - and 7.5% over the most recent three-month period for an annualized rate of 30%. Listing prices rose at the fastest rate in Detroit - up 4.8% in July - followed by Cleveland where prices were up 2.7%. Prices were also up slightly in the Midwestern markets of Indianapolis and Minneapolis.

"While prices continue to fall in coastal and Western markets, prices appear to have stabilized in Midwestern markets that were previously declining," said Michael Simonsen, CEO and co-founder of Altos Research. "The real test will come in the fall when markets typically experience a seasonal slowdown which will be exacerbated by high job losses and foreclosures."

Asking prices fell at the fastest rate in Las Vegas - down 4.0% during July - and 7.5% over the most recent three-month period for an annualized rate of 30%. Listing prices rose at the fastest rate in Detroit - up 4.8% in July - followed by Cleveland where prices were up 2.7%. Prices were also up slightly in the Midwestern markets of Indianapolis and Minneapolis.

Listed property inventories declined with the 10-City Composite markets showing a decrease of 2.0% in July. Inventory rose in just 6 of 26 markets with the biggest decreases occurring in Detroit and Cleveland.

"Broadly declining inventory is a positive sign in the near-term, particularly for the Midwestern markets which all showed inventory declines," said Stephen Bedikian, partner and research director for Real IQ.

For the Altos 10-City Composite, the average days on market was 111, a slight increase from 109 in June. Thirteen of 26 markets had an average days-on-market of 100 or more. By far, the market with the slowest rate of inventory turnover was Miami at an average of 156 days-on-market, nearly a full month more than the next slowest market - Tampa. Austin led all markets with the fastest rate of inventory turnover at an average of 78 days-on-market.

Data in the Real-Time Housing Market Report is based on analysis of over one million properties currently listed for sale in 31 metropolitan markets across the country. The report is the most timely source of housing market data on current market activity.

The report examines housing pricing, inventory levels and market conditions in 31 major U.S. metropolitan statistical areas (MSAs): Atlanta, Austin, Boston, Charlotte, Chicago, Cleveland, Dallas, Denver, Detroit, Houston, Las Vegas, Los Angeles, Miami, Minneapolis, New York, Phoenix, Portland, San Diego, San Francisco, Seattle, Tampa, and Washington, DC. The Real-Time Real Estate Report is released every month.

Yesterday I wrote about Chicago's dubious distinction as the most socially regulated city in the US. I argue that trend does not bode well for the creative class, the city's future prosperity and ultimately its real estate values over the long haul. However today I came across a glowing article on Chicago in Fast Company, calling it "City of the Year". Indeed it's a city I love, so let's look at the positives.

Fast Company lauds some of the city's social restrictions as forward thinking "Greening" efforts. (Ironically they also posit that the city's position as anchor of 20th century architecture happened here because there was "no one to tell [the developers] to do it differently.")

Construction Booms
Most of what FC is happy about though derives from, Chicago's marvelous growth spurt. The city is the fastest-growing non-Sun Belt city in the US. The economy is growing faster than New York or LA. Immigration remains strong from all over the world.

While highrise condo development has slowed from it's frenzied pace a few years ago, I'm still excited about the Chicago's recent construction boom. The famed skyline is remains one of the most moving in the world. And its development continues. (I highly recommend the Chicago Architecture Foundation's boat tour if you have 90 minutes to spare while you're in town.) Between the 150-story Chicago Spire and the new Trump tower (who let him in there?) the city will remain on the forefront of skyscraper architecture.

With all this development, it's worth a look to see how the downtown Chicago condo market is holding up. Here's a handful of zip codes:


Let's look at demand rates also.


Like much of the country, the most desirable parts of town have (those with the higher prices already) show reasonably consistent demand and stable prices. This is not the case as you leave the hot neighborhoods.

So what's in store for the City? Construction, investment, and immigration warm my heart. Laws to dictate my diet chill me to the bone. The good news is that buildings last a long time. Bad laws can be as ephemeral as the foie gras they're restricting. Let's call this one a net positive.

Rock on Chicago.

Link: Chicago Real Estate Data

Checking out Andy Kaufman's blog this morning, I couldn't help but noticing the contrasting AltosCharts he's showing for Oakland and Berkeley California.

Oakland of course is the bigger city, but these next-door-neighbors share some parts that are virtually indistinguishable from each other. There are some spectacular parts of Oakland and sketchy parts of Berkeley and vise versa. But look at the home price trends over the last year.

Likewise look at the trends in inventory levels for the same to cities.

You couldn't have a clearer picture of the real estate pricing phenomenon we're seeing all over the country. It works like this:

• Demand is off everywhere and everyone knows it.
• So if you're a home buyer in the East Bay with good cash and good credit, you get your pick. In this case, Berkeley has generally better schools, more cachet, etc., so the buyers go there first.
• There's enough demand to keep the market afloat.
• If you don't have good cash and good credit, you get nothing. There is literally no spillover demand for Oakland.
• A few years ago, a home buyer might look to the nicer neighborhoods in Oakland for additional inventory or a lesser school-premium price. Not the case any more.

Actually, I'm sure we could dig through the zip code level data in Oakland and illustrate a similar phenomenon within the cities themselves. But this particular story jumped out at me this morning, so that's what gets covered.

Altos Links: Oakland real estate data, Berkeley Real Estate Data

The other day, I highlighted the announcement from Bob Shiller's MacroMarkets to list exchange traded funds on the housing market. I've now had a chance to investigate more deeply and I'm giddy like a schoolgirl. (Albeit an incredibly geeky schoolgirl, but giddy nonetheless.)

First, some foundation as to why this matters.

In all businesses you have risks you can control and costs you can't: food, energy, interest rates, etc. For those costs, it makes sense to hedge. Successful jet fuel hedges are a big reason Southwest Airlines is the strongest in the country. Consumer products (e.g. cheaper airline tickets, wacky mortgages) get created on the foundation of these tools. (i.e. derivatives are a good thing.) Speculators can also participate - they add potential return to their portfolio where a hedger removes risk. Speculators create liquidity for the hedgers. (i.e. speculators are a good thing.) Financial derivatives, futures, options, swaps, etc. exist in nearly every asset class to solve these problems for people.

Likewise, lots of people and companies have real estate "exposure". This is a $21 trillion asset class people. You should be able to hedge. Especially now, people realize housing prices don't always go up. But before 2006, there were no financial products that let you hedge your real estate risk. And the only way to speculate was to buy investment property.

In 2006, MacroMarkets introduced, on the Chicago Merc, housing derivatives. Unfortunately it turned out that there were practical limitations on the housing futures that prevented nearly all potential "end-users" from participating. (The big banks could trade amongst themselves, but how fun is that?) Namely, you need big capital requirements, special trading accounts, most of the time you need a broker-dealer on the other side of your trade, and the payoff is not significantly leveraged. Perhaps I was a bit harsh to characterize MacroMarkets as having "dropped the ball" but, as of today, mere mortals basically still can't hedge their real estate risk. So how do you eliminate these hurdles?

Enter Exchange Traded Funds

ETFs are securities that trade like stocks on stock exchanges. You can play the oil price trends or diversified stock market positions simply buy buying a single "stock".

Here's how MacroMarkets' new ETFs ("MacroShares" as they call them) work for the housing market:

• The shares will list on the NYSE Arca. You can buy them from any brokerage account, like you would any stock.
• They're designed to follow the Case Shiller Index - specifically the CSI 10-City Composite, the national index. more on the implications of this below.
• The shares are issued in pairs: ticker symbol NYSE:UMM goes up when housing prices go up, NYSE:DMM goes up when housing prices go down.
• Like any stock, there's a market specialist so you always have price liquidity. You don't need to line up Goldman or Merril to take the other side of your bet.
• The shares have a par value of $25/share so when the market is heavily bearish the DMMs will trade at a premium to the $25.
• They have a 10-year horizon so you can play the longer term cycles of the housing market.
• The underlying assets get parked in short term US Treasuries, so the securities actually return a quarterly dividend.
• So the very easy trade is if you're bearish, simply buy the DMMs. Think you're seeing the bottom? Buy the UMMs. Ahh... simplicity.

Exercise some caution however, because there are nuances of how these things will behave. Namely:

• They trade in pairs that offset each other. This means if home prices double, the DMMs will go to zero and the security closes out. This happened in MacroMarkets oil MacroShares this year. AMEX:UCR and AMEX:DCR. (Those securities have market values in the billions now, so people are obviously adopting them as useful instruments.)
• That means also that the upside is limited to a 100% move. You can improve your return with margin leverage like any stock.
• As these things approach close (because of A. time or B.100% market moves) the pricing will exhibit optionality. They'll behave less like the market and more like an option.
  

But is the Case Shiller Index Useful?

The remaining challenges for these products are oriented around the data. It's easy to bitch about the Case Shiller Index: doesn't include condos, or new construction, or flips, etc., etc., etc. Add in local market peculiarities and a lot of people wonder if the CSI actually measures the housing market.

My take on this argument is that Case-Shiller is not useful for making a home purchase decision. But that doesn't preclude its usefulness in financial instruments. The fact is that the CSI 10-City Composite peaked in June 2006, and that's widely regarded as the national turning point for this housing market cycle.

The classic example of the localness problem came when Brad Inman asked Bob Shiller on stage and his conference in Miami, "So let's say I bought a $2 million home in Sausalito in 2005. How would I hedge that?" Ill distill Shiller's 10-minute-Yale-finance-prof reply into two words for you: "You can't."

With Given that these new securities are based on the CSI 10-City Composite, which is down strong in the last 12 months, they're not going to be helpful to hedge in Sausalito, which is doing just fine, thank you.

But if you're a reasonably diversified investor, brokerage, lender, builder, supplier, or yes, even if you're a speculator, this is a great way to measure US housing broadly. Given success in the market, there's no reason why they can't list regional funds too at some point in the future, to get a little closer to home.

Finally, of course, the backward-looking nature of all typical housing market data presents opportunities for clients of the Altos real-time real estate data. Rock on.

This is big, folks. Huge.

I don't imagine that this innovation is going to save anyone from foreclosure. But we're looking at the only effective way to manage your real estate assets without physically selling off properties. Think about that. Won't that be amazingly useful?

Look for these to get listed sometime in Q3 or Q4 2008. You can be sure that we'll be watching, and of course publishing data to help you trade.

More coverage: Here, here, and here.

Jonathan Miller published the November Radar Logic RPX housing market report over the weekend. It's easy to spot that Sacramento is leading the nation down.

[aside: Radar Logic is cool. They measure Price Per Square Foot and try to do it across all properties, condos and single family homes, new and existing construction. No small task. The price per square foot approach is intended to measure value of the property regardless of the size of the property itself. In the real estate derivatives trading business, Radar is getting all the mindshare. Case Shiller is falling behind. more here.]

In the report just published for November, Sacramento saw a price-per-square-foot decline of 18%. Zowie. For those of you unfamiliar with California's central valley, this is an area dominated by lots of new construction, in huge projects, partly as Bay Area super-exurb. So the underlying economy in the Central Valley isn't nearly as dynamic as San Francisco, San Jose, or Los Angeles. Also much less dominated by the high-end, Sacramento is feeling the subprime fallout harder than most.

So Radar Logic is publishing for November. What are the real-time stats saying? Sacramento is not seeing any relief yet. Here's our price per square foot for Sacramento through February 1 2008.

Sacramento Housing Market Data Links:

Our free Sacramento Real Estate Research page

Here's a solid Sacramento Area Blog for more local flavor.

Some posts just write themselves.

Can't. Help. It. Must. Write. More.

Ugly? You betcha. Do these tell the whole story? Not a chance. In Silicon Valley, San Jose is the dominant market, of course. San Jose is a diverse community, with lots of sub-prime and other crazy loans in the past few years. But also some really great neighborhoods with prosperous, fully employed folks. Lots of big, but not-risky loans too. Here's how the market in a desirable part of town, Willow Glen, is holding up. I've done the price chart in Quartiles so you can see the trends at each price point in the market.

Link: San Jose Housing Market.

For the NAR conference this month I was a bit tardy in posting my travel schedule, and I missed some connections there as a result. So here's a heads-up for my conference travel schedule in January and February.

January 8-11 New York: Inman Connect conference. Brad Inman always throws a grand shindig. I make connections, friends, and new clients at every Inman conference. My meeting schedule is filling up fast, but I still have some time on Thursday and Friday. Give me a buzz and we'll get together. Scott will also be in attendance so we should have good coverage meeting old friends and new.

February 5-8 New York: O'Reilly Money:Tech conference. This conference is all about hacking Wall Street with new data and tools. It promises to be one of a kind. Love it. I'll be presenting on real-time real estate market analytics, plus our data that leads the Case Shiller and Radar Logic housing futures markets. We'll be demonstrating some never-before-public data goodies.

Lots of big shots presenting at Money:Tech too. Jim Cramer, Barry Ritholtz, Henry Blodgett, Bill Tancer, and more. Check it out.

As always, these trips are about meeting clients, partners, and readers. email mike at this url.com to get in touch.

When we started this company, one of the things we envisioned was creating a National Housing Market Report -- one that would be a right-now alternative to the laggard OFHEO reports and even the Case Shiller releases. Well today is the day, folks. The Altos Research Real IQ National Housing Market Report is here!

In it we look at 20 major metro markets, publish some key stats about pricing and supply and demand trends and we draw some conclusions about what's happening out there. The U.S. metropolitan statistical areas (MSAs) covered in the report include: Atlanta, Boston, Charlotte, Chicago, Cleveland, Dallas, Denver, Detroit, Las Vegas, Los Angeles, Miami, Minneapolis, New York, Phoenix, Portland, San Diego, San Francisco, Seattle, Tampa, and Washington, DC.

We're working in conjunction with Stephen Bedekian from Real IQ. Stephen is an industry leader, writer, and consultant and he helped bring this project together. This being the first issue, it took us a few extra days of editing. In future months, we'll be aiming to publish the report just a few days after month's end.

The latest version of the report is available for download on our financial institutions page. Or you can download the PDF here.

The Press Release headline this month isn't any shocking news - Surprise! Prices are still under pressure! But we're laying the foundation here to be the first to identify the eventual bottom of the market. From the press release:

REPORT: HOUSING PRICES
CONTINUE DECLINE IN NOVEMBER

Time-on-Market
Increased as Housing Market Demand Falls Faster than Supply

MOUNTAIN VIEW, CA
- December 13, 2007 – The listing prices on properties in 18 of 20 major markets across the U.S. fell during the month of November. San Diego experienced the steepest decline with listing prices falling 5.8 percent as the recent wildfires exacted a toll on demand. This information was presented in the newly-launched Real-Time National Housing Market Report*, published by Altos Research, the premier source for real-time, real-estate market research and Real IQ™, a market analysis consultancy.

The Real-Time National Housing Market Report is based on an analysis of data from over one million properties currently listed for sale in 20 metropolitan markets across the country and represents the most timely source of housing market data on current market activity.

“Real estate information tends to be highly latent and subject to a lot of revisions,” said Michael Simonsen CEO and co-founder of Altos Research. “It takes several months before the S&P/Case Shiller Index or the
OFHEO data is released for a given month. When you're making investment decisions or trading derivatives these lag times are simply killers.”

Another key trend noted in the new report was an increase in the time-on-market duration for homes on sale in virtually all markets. Miami experienced the longest time-on-market span with an average days-on-market of 137 in November. Minneapolis had the second highest average days-on-market at 125.

Listed property inventory levels displayed seasonal declines in many markets with the exception of Las Vegas where for-sale property listings increased 6.6% over the past three month. “While inventory levels declined in most major markets, the decline in supply could not keep pace with the rapid fall in demand,” said Stephen Bedikian, partner and research director for Real IQ. “We expect time-on-market will continue to lengthen and apply pressure on homeowner pricing decisions until buyers regain confidence and demand levels off. So far that point is not in sight.”

*The first report was published December 7, 2007 and will be released every month. Report downloads are available from Altos Research.

If you've watched the Case Shiller Index numbers this year, you might have noticed that Denver, Colorado is the lone market still registering positive housing price gains for 2007. (BTW: if you prefer Uncle Sam's OFHEO numbers, you'll have just this week discovered that the US housing market is under pressure.)

We've been watching Denver too, because that a market that's been bucking the trend. We recently opened our subscription service for real estate professionals in Denver, as well as home buyers and sellers there.

Here's what the median price looks like for the Denver MSA, in a rolling average over the past several months.

You can see from the chart, home prices in the Denver area have so far missed the bursting that's hitting most of the rest of the country. Up just fractionally, but steadily this year.

Why the resilience? A couple reasons probably: 1) Colorado real estate didn't have as much upside in the last few years to begin with. 2) The economy and investment levels are still strong. 3) Denver is just a little bit of a laggard economically behind the coastal cities. That implies the burst simply hasn't hit Denver yet.

Looking at the rolling average smooths out weekly noise, so the trend is easier to see. But using a three month rolling average, it'll lag the real-time market a bit. The next chart is the weekly sample.

In this image we're just starting to see the weeklies break below the 90-day rolling average. Too early to make a big bet, perhaps, but a noticeable change nonetheless.

Plus, if you look at the city of Denver the change is much more pronounced. Zowie! That's quite a drop. (caveat: don't discount the seasonal impact that's surely happening in some capacity here.)

Keep an eye on Denver. I'll be fascinated to see if that town is able to demonstrate real staying power, or if it's just a few months behind.

Links:

Free research for the Denver housing market

Free research for the Aurora housing market

etc.

Here's a couple of Altos Research mentions in the Seattle Post Intelligencer last week.

Altos client and blogger Sandy Kaduce is a contributor to the paper's blog site, wrote last week about the changing nature of the market.

On Friday, writer Aubrey Cohen called to get our take on the Seattle housing market numbers. The numbers that come out of the Northwest MLS are just starting to show year over year declines. (If you're an Altos watcher, you've seen this coming for a long time now of course.) Aubrey notes that the mix of property types being sold have an impact on the over all "median price" of a market. He uses one of my favorite qutoes to describe the 2007 (where the pain is starting low).

And the recent problems in the subprime mortgage market, which serves people with poor credit, have a disproportionate effect on cheaper homes, meaning those that do sell are more likely to be at the higher end of the market, buoying the median price, said Michael Simonsen, chief executive of Altos Research in Mountain View, Calif.

"You don't buy a $1 million waterfront property on a subprime mortgage," he said.