Are Homebuilder Stocks Actually Cheap?

THIS IS A QUESTION I’ve been considering over the course of this decline in homebuilder stocks:


After all, a real estate bull might expect that such a move would be a mere consolidation in the context of an ongoing bull market that began with the death knell of the tech bubble in 2000. The baton was passed almost seamlessly from tech stocks to the real estate market. Here is a chart of Toll Brothers, one of the best-performing stocks of the 2000-2005 bull market. Toll bottomed right around the March 10, 2000, top in the Nasdaq:


There is no doubt that homebuilding stocks provided phenomenal returns up until summer 2005 and have been “falling knives” since then. Where there is doubt is whether these stocks are bottoming in preparation for a huge rally or whether they are drawing bargain hunters into a “value trap” right before falling another 50%.

My Whiskey colleague Mish has done a great job cataloging the slow-motion demise of the housing market in this newsletter, most recently in last month’s post called “A Dose of Reality.” I agree with him regarding the unpleasant fate awaiting real estate speculators in the most overheated markets. Also, those expecting a short, painless correction in housing prices after such a long and incredibly speculative run are likely to be disappointed. But as an investor, it’s important to consider all sides of the macro debate before committing capital to a trade or an investment.

Toll’s Latest Quarter: A Bottom or a Preview of Next Year?

On Aug. 22, Toll Brothers released some pretty terrible results in comparison with the stellar numbers the company had been posting over the past few years. It was not really a surprise, considering that the company has repeatedly made headlines this year with profit warnings accompanying lower “guidance.” Diluted EPS in the quarter ending July 31 came in at $1.07, down 16% from last year’s record 3rd-quarter result of $1.27. But far more important than the quarterly results was Toll’s deteriorating backlog:


Even more disconcerting, Toll’s order book is deteriorating much faster. It is more forward-looking and indicates the magnitude of the decline likely to befall Toll’s 2007 earnings. The key Northeast and Mid-Atlantic markets, highlighted in blue, have seen orders dry up. The result will be further contraction in backlog as Toll delivers finished goods (houses) out of its backlog faster as it replenishes the backlog with new orders:


In a statement accompanying the press release, CEO Robert Toll stated:

“The continuing malaise in the housing market, we believe, is the result of an oversupply of inventory and a decline in confidence. The speculative buyers of 2004 and 2005 are now sellers; builders that built speculative homes are trying to move them by offering large incentives and discounts ; and some anxious buyers are canceling contracts for homes already being built. This overhang in supply and the aggressive discounting of many builders is undermining consumer confidence and keeping buyers on the sidelines as they continue to worry about the direction of home prices.” (Ed.: emphasis added)

The idea that real estate speculators have chosen preconstruction homes in the $700,000 price range as the objects of their speculation seems preposterous to me. But then again, I cannot hope to identify with the mind-set of a house-flipper. A far more reasonable excuse, and one that we will likely hear again in future quarters, relates to price competition from smaller, private builders. There will probably be quite a few private builders who cannot sell the “spec” homes they constructed for a profit. Raw material suppliers and bankers could force liquidations below cost, including even more price cuts that do not show up in statistics. These include free upgrades, below-market mortgage rates, and even free cars (it’s not unheard of).

Going back to the issue of declining backlog, the decision that Toll executives make regarding capital allocation over the next few quarters will be key to shareholders’ long-term return. Should they hit the brakes hard, slash marketing, complete deliveries out of their backlog, walk away from options to purchase more raw land, and buy back stock with the free cash generated by inventory destocking over the next year? I’d vote in favor of this “batten down the hatches” decision, but Wall Street certainly wouldn’t like it.

The market, especially in today’s environment, usually prefers companies to take a more aggressive growth stance. Also, public sentiment toward real estate has only started to change, so such a move would be viewed as excessively bearish and radical. So Toll executives are likely to continue reinvesting as planned. If that is the case, then the stock has probably not established a durable bottom. At least right now, it appears that Toll will go valiantly down with the ship. While the easy money has already been made on the short side in homebuilders, there still exists plenty of opportunity for nimble traders to profit from the fallout of a housing bust in other sectors.

Housing Bulls Should Not Extrapolate the Past Into the Future

Probably the single greatest argument advanced by housing bulls is the near-religious devotion baby boomers have toward real estate as an attractive asset class. Over the course of their entire lives, boomers have never seen a year-over-year decline in real estate values at the nationwide level, so most cannot imagine prices ever declining due to this strong “anchoring” effect. Compound interest can produce jaw-dropping multidecade returns when there are no negative years over a lifetime.

But past experience in real estate returns resulted from a combination of a few key favorable trends that are either weakening or have reversed: The U.S. essentially “dollarized” its international trading partners; energy prices declined in real terms over several decades; monthly household budgets were given an enormous one-time boost as women entered the work force; and, finally, primary residences (not including rental properties) have come to be viewed as “investments,” rather than the physically deteriorating, cash-consuming structures that they are.

The “dollarization” of its international trading partners stands out as the most significant factor behind the post-WWII housing boom/bubble. The implementation of the Bretton Woods international monetary standard set up the world to trade exports for paper dollars that could be issued without limit.

The banking system was revolutionized and mortgage growth has proceeded virtually without limit. There has not been a need to fund mortgages with domestic savings for decades. As long as foreign creditors keep recycling billions of U.S. dollars back into the Treasury and mortgage-backed security (MBS) market, mortgage rates should remain low. But these foreign creditors are communicating — through the press and with their wallets — that there are far more pressing domestic needs for savings and investment than financing American profligacy.

We are not the only country in the world that can think and invent, and to assume otherwise is a grave mistake. The arrogant “we think, they sweat” mind-set is overly simplistic and gives no heed to politics, wealth distribution (after all, not all Americans can “think” for a living), and the distribution of debt and savings (leverage may be fine for the wealthy, but can be the path to bankruptcy for the poor).

I am not pointing this out due to political bias — I am an avid supporter of free trade and very limited government — but I hear a lot about debt-to-GDP statistics and “national housing wealth” in the housing market debate and very little about the distribution of debt among those who can really tolerate it and those who think they can, but really can’t when a recession occurs. To quote Warren Buffett: “It’s only when the tide goes out that you learn who’s been swimming naked.”

So this will be no run-of-the-mill “correction” in the housing market. One should not extrapolate past housing market conditions well into the future when considering an investment in a homebuilder.

Value Investors Are Attracted to Low P/Es and High Returns on Invested Capital

Some well-known contrarian investors with phenomenal track records — most notably Bill Miller of Legg Mason — have been accumulating homebuilders like Beazer, Centex, Pulte, and Ryland since spring 2006. After major declines, these four stocks are trading for an average trailing P/E ratio of 4.7. This is incredibly cheap in the current market, but trailing earnings represent the very peak of the most speculative housing market in history (in other words, 2007 earnings are likely to decline significantly, making the forward P/E ratio potentially double or triple the trailing ratio). Your macro outlook for the housing market over the next couple of years will determine whether you think these stocks are bottoming or just pausing before another round of declines.

While Mr. Miller has often acted a little too early before major turns in sentiment toward an unpopular sector, he is often proved right a few years after buying near the lows. His holding period often spans several years and his return on capital/cost of capital approach is proven to produce market-beating returns over the long-term.

A simple way to think about return on invested capital (ROIC) is the profit a business owner expects to receive for an incremental investment in capacity expansion. The formula for ROIC is similar to ROE (net income divided by equity), but is more useful for comparative purposes, since ROIC does not factor in financial leverage. Taking on a ton of debt leverages the equity holders’ investment and boosts ROE, but it also raises the risk profile substantially.

Google provides a good example of a high-ROIC business. There is no “inventory” to tie up capital. There are no large outlays for maintaining plant and equipment. The sustainable return on capital for Google can be estimated as nearly infinite, because the actual value of “invested capital,” the divisor in this formula, is very subjective; it depends on the perspective of the person estimating its value. While Google’s invested capital can be estimated by using last quarter’s book value, it constantly changes and consists of claims on intangible assets such as goodwill, brand name, and customer loyalty. This is where the concept of accounting as an art, rather than a science, comes into play.

For illustration purposes, let’s assume that last quarter’s balance sheet and annualized operating income are used. I calculate Google’s annualized return on invested capital to be about 17%. If you assume that no more investment on the part of shareholders is needed to finance Google’s future growth (a reasonable assumption) and that the past few years’ income growth continues for the next few years, the result will be ever-increasing ROIC. Mathematically, this results in a growing numerator (return) and a flat denominator (invested capital). Such a business model would be worth practically any price, including the current multiple of 55 times trailing earnings, right?

Well, when the prospect of future competition enters the picture, the water becomes muddied. Over the course of capitalism’s history, the length of a technology company’s life has largely depended on its success in fending off competition through continual product development and innovation. Google fits the mold of a company that continually innovates, gaining an edge over the stodgier competition with each innovative product launch.

But Google does not have a monopoly on ideas. Perhaps Microsoft will develop the next great search engine technology that relegates Google to the dustbin of history within 5-10 years. All of a sudden, a high return on invested capital business model is shattered and ruins the premise behind owning the stock in the first place. The probability of such a threat emerging far in the future must be estimated fairly quickly and accurately and the Google investor must be early in acting on it by selling before the crowd — hardly an easy task.

Therein lies the challenge of buying and holding a stock in an industry characterized by short product life cycles and creative destruction. The “moat,” or barrier to competitive entry, must be deep, wide, and sustainable to make outsized profits by buying and holding a technology stock.

Return on Invested Capital of Homebuilders

Now let’s apply this ROIC framework in estimating the intrinsic value of homebuilder stocks. Using the example of Toll Brothers, what sort of “moat” surrounds its business? The first thing that comes to mind is its brand name. It is associated with luxury, prestige, and quality — at least in the minds of its customers. Secondly, Toll’s large size gives it a slight customer service advantage, and the company’s purchasing power gives it more leverage over its suppliers when compared with small, private builders. The homebuilding industry remains very fragmented, with the largest 10 builders accounting for only about 25% of total new homes sold in the U.S. Lastly, Toll has nailed down several years worth of future raw land needs through outright purchases and options.

This seems like a significant “moat” when you consider these three components in unison. But it’s vital to consider how well the moat holds up through industry downturns. If the prestige and popularity associated with owning a “McMansion” loses its luster among the wealthy, the value of Toll’s brand name will be diminished. In fact, it may not require abandonment on the part of customers; community associations are likely to continue being a “fly in the ointment” for both Toll and its prospective customers. In his March 26 article posted on MSN, “The Swelling McMansion Backlash,” Christopher Solomon describes the tactics that community activists and local governments are using to block the construction of houses above a certain size. A sustained expansion of its legal staff will not be favorable for the brand name component of Toll’s “moat,” or its long-term return on capital, for that matter.

The second component of Toll’s moat — its size advantage and its bargaining power over suppliers — must be considered in light of a large, lengthy downturn. These advantages work wonders for profit margins and ROIC in an uptrending market, but can serve as handicaps in a nasty environment for new home construction. With great size and scale comes a similarly sized overhead expense structure. A talented sales force and top-notch contractors do not come cheaply, but they are necessary to maintain a sterling reputation.

As Mish pointed out in his piece last month, management talent has not come at a cheap price for Toll shareholders. They have given a huge chunk of future upside to executives in return for leadership that carries a highly debatable value. Millions of stock options have been granted to executives and employees (the top three executives received 31% of the 2.7 million stock options issued in FY2005).

Bargaining power over suppliers is another benefit that can offset much of the pain of rising lumber, drywall, concrete, and steel costs in a hot construction market. Negotiating volume discounts is an advantage that is often pressed by first-movers who have grown their way to the top of the building business. But as a hot construction market turns cold, raw material inflation is likely to level off, taking away most of the benefit from volume discounts. In fact, as projects around the country are canceled, a glut of building supplies could pressure prices, perhaps encouraging Toll to include raw materials as part of its future inventory charges.

The last component of Toll’s “moat” that I will address is its large land position. Toll considers one of its key competitive advantages to be its foresight in beginning a “land grab” at the outset of the inflation in raw land prices. As a result, the company now has enough land at its disposal to cover about six years worth of future building. There’s one key assumption about this “advantage” — it only remains an advantage as long as raw land prices keep appreciating year over year. If land prices fall, not only will Toll have to take a charge against inventory, but its decision to “hoard” land will look foolish in hindsight. They are “not making any more land,” but that doesn’t prevent local prices from falling, especially in a recession.

“Inventory” Is Not Just Finished Houses

Here is a detail of Toll’s most recent balance sheet:


Considering that inventory is the only asset of significance — at 85% of total assets — I thought it relevant to include a section from Toll’s 10-K describing its inventory accounting policies:


In accounting, there is considerable leeway allowed between capitalizing and expensing for companies in project-oriented businesses. For example, software companies “capitalize” hours of programming done by engineers on a large project (i.e., they account for programming as an asset, rather than as an expense). Similarly, the most accurate way to account for something like a housing subdivision is to “capitalize” most of the labor, materials, and development costs over the course of construction. Since these costs are not expensed through the income statement until the homes are delivered, this practice has the effect of overstating earnings at the peak of the building cycle. As a result, Toll will likely have to keep taking quarterly write-downs to its inventory. Wall Street usually ignores inventory charge-offs, but if an undeniable string of inventory charges develops, then Toll’s accounting practices would come under suspicion.

No more detail is given regarding how Toll’s $6.2 billion inventory position is allocated among land, raw materials, capitalized development, and finished houses. One key disclosure the company makes is that it normally takes about five years to fully complete a housing subdivision, indicating that the houses Toll is delivering this year sit on land that is valued on the books at 2001 prices. This makes profit margins higher than they would otherwise be if 2006 home deliveries included 2006 land costs. There is no easy way to estimate how much raw land inflation has padded earnings since 2001. It certainly is significant, and I expect it to transform into an earnings head wind over the next few years. Toll will likely continue to write down land and walk away from land purchase options in the worst local markets, leading to a multiyear compression in profit margins and downward-trending EPS. Over the past three years, Toll aggressively doubled its land position at inflated prices — from 41,000 home sites in October 2002 up to 83,000 in October 2005:


So Are Homebuilder Stocks Actually Cheap?

Just because several homebuilders are selling near book value and have business models with high historical returns on capital doesn’t mean they are bottoming yet. Return on capital can depend highly on favorable macro conditions, as I have shown with a few examples. The measure of book value for most homebuilders will be a moving target in the future, as further inventory charges and margin compression is very likely. So the argument that homebuilders are cheap rests on shaky accounting and extrapolation of the past into the future. That adds up to a “value trap,” in my opinion.

Good investing,
Dan Amoss, CFA
September 1, 2006