Forget all those ludicrous dot-com yardsticks. Here's how to find real value in a stock.
EARLY LAST YEAR, the financial press was full of stories about how to value stocks in the Internet economy. Conventional yardsticks such as price-to-book and price-to-earnings were for fuddy-duddies. Instead, serious people — well, at least Wall Street analysts and on-the-make economists — talked dreamily about a whole new world of financial metrics. Web-content companies traded based on monthly page clicks. For dot-com retailers, the magic number was gross margins. Or, if they were negative, ad spending.
This sounds ludicrous today — more like Tulipomania or the Mississippi Bubble than serious investing. With prices of nearly 250 Internet companies off by 90% or more since last April, I think it's time for a different take on how to measure a stock's value. So this month I'm dusting off one of the oldest, most conservative stock-picking methods around. It's called the dividend discount model, which asset allocation guru William Bernstein calls "the closest thing investors have to the Law of Gravity."
In what follows, I'll explain the dividend discount model, or DDM, and tell you how to use it — which is easier than ever with help from a new Web site. I'll introduce a Boston economist who may have influenced the way we think about the stock market even more than value-investing legend Ben Graham, though he's virtually unknown. And finally, I'll zero in on a group of high-yield real estate investment trusts that look like bargains today in DDM terms.
This story begins when John Burr Williams graduated from Harvard in 1923. Williams went to work on Wall Street and made lots of money. Then the market crashed, and he headed back to Harvard to study economics. Williams wasn't humbled by his losses as much as he was curious about how to avoid them next time. He wanted to see if there was a theoretical "right" price for stocks.
By 1938, Williams was sure he had the answer. He packed his ideas into a doctoral thesis that he called "The Theory of Investment Value." But instead of waiting for a Ph.D., he fired off manuscripts to two big-name publishers — Macmillan and McGraw-Hill. Back came rejection notices: No one will read a book with math symbols. So Williams convinced Harvard to publish his work, paying part of the cost himself.
Pride of Ownership
These nine real estate investment trusts combine high yields with higher-than-average expected returns.
| COMPANY | PRICE* | 52-WK
HI-LO | YIELD | EARNINGS
GROWTH** | DESCRIPTION |
Apartment Inv. & Mgt.
(AIV) | $47.31 | $50-$37 | 5.92% | 10.46% | Apartments |
AvalonBay Communities
( AVB) | $47.00 | $50-$34 | 4.77% | 10.25% | Upscale apartments |
Boston Properties
( BXP) | $41.63 | $45-$30 | 5.09% | 10.14% | Upscale office & hotel |
Duke-Weeks Realty
( DRE) | $24.69 | $26-$18 | 6.97% | 9.26% | Office & retail |
Equity Office Properties
( EOP) | $30.63 | $33-$23 | 5.88% | 10.06% | Office |
Kimco Realty
( KIM) | $43.31 | $45-$33 | 6.65% | 8.52% | Shopping centers |
Liberty Property
( LRY) | $27.81 | $29-$22 | 8.20% | 9.00% | Suburban office |
Spieker Properties
( SPK) | $51.31 | $59-$39 | 5.46% | 13.20% | Office & industrial |
Vornado Realty
( VNO) | $36.19 | $41-$30 | 5.86% | 11.75% | Office & retail |
S&P 500 median
( N/A) | $40.06 | $51-$26 | 1.09% | 13.08% | N/A |
* Prices as of 1/26/01
** Analysts' consensus estimate of annual rate for next three to five years.
Data: Research Wizard 3.1; Zacks Investment Research
Williams died more than a decade ago after a low-profile career managing money in Boston. But his book is still in print, and his ideas underpin much of modern finance. Williams liked dividends. He pounded home the idea that any stock is ultimately worth no more than what it will throw off in current or future income. Buying for any other reason means you're a mere speculator — paying 30 times page clicks today because someone will surely pay 32 times tomorrow.
Obviously, few people are paying attention. Only 20% of all companies bother to pay dividends today, down from nearly 70% in the late 1960s. And even companies that cling to payouts earmark a steadily declining portion of their earnings for shareholders — just 30% last year, a record low. But don't dismiss Williams too quickly.
The second part of his legacy has to do with the link between time and money. Williams figured out how investors could value dividend payments they might not receive for a long, long time. Think of it this way: What would you pay to get $1,000 annually forever? If the long-term interest rate (which economists call the expected return or cost of capital) is 5%, that so-called perpetuity is worth $20,000 ($1,000/0.05).
The time-tested dividend discount model is 'the closest thing investors have to the Law of Gravity,' notes one market guru. |
|
| |
Consider a company with a $1 annual dividend. If it doesn't go bankrupt, interest rates stay the same and the payout never changes, the share price ought to be $20 a share ($1/0.05). What if the dividend is increasing, say by 3% annually? Williams's model says the stock is now worth $50. But getting to this value involved all those equations that turned off Williams's potential publishers: Value=Dividend/(Expected Return-Growth Rate) or $50=$1/(0.05-0.03).
The math here is complex, unless you're a whiz at integral calculus. So let's look at things another way: When dividends are rising, the expected return equals the current yield plus the dividend growth rate. In the previous example, when you adjust the stock price to allow for dividend growth, the current yield falls to 2% ($50/$1). But the expected return doesn't change: It's 5% — either from a fixed dividend that yields 5% now or from a 2% dividend growing at a 3% annual rate. What we now have is a wonderful tool for comparison-shopping among stocks.
Another wrinkle makes the dividend discount model even more useful. Suppose two companies have similar long-term prospects, but one business is steady while the other has lots of ups and downs. Unless you're a dunce, you'll want a higher return before you invest in the more volatile business. A simple measure of this risk is called beta. If shares in each of my hypothetical companies have 5% expected returns (current yield plus long-term growth rate), the one with the lower beta (less risk) is probably the better investment.
Enough theory. The classic dividend discount model works best when valuing mature companies that pay a hefty portion of their earnings as dividends — like the real estate investment trusts in my table. But Williams's ideas can be applied more broadly. The most common wrinkle is substituting cash flow for dividends as an input. The idea here is that it's important to know how much income a company is likely to generate over time, even if you don't know how much will go directly to shareholders. Some analysts even build multistage models that assume growth companies will start paying dividends when they mature, and economists use DDM to compare different types of assets. For now, I'll stick with the basics — looking for stocks with above-average expected returns, based on high current yields and growing dividends. The lower the beta, the better.
As you might expect, the pickings are rather slim. In an initial screen, I searched for companies with market values of over $1.5 billion and yields greater than 3%. I was left with only 150 qualifiers out of 6,000 stocks. Then I added tougher criteria. I set my expected return minimum at 15%. (Over time, dividend growth and earnings growth should be roughly the same, so DDM mavens add current yield to earnings growth projections to get this number.) I wanted historical dividend growth to be at least 6% annually, and — to minimize the risk of dividend cuts — I wanted current earnings to be at least 25% above current dividends.
Running those numbers left me with 22 companies — and a surprise: a few big names (Caterpillar (CAT), Dana (DCN), Philip Morris (MO)), but more than half my finalists were REITs. And when I looked for stocks with the lowest betas, the REITs moved to the top of the list.
Instead of being nervous, I'm reassured. I like REITs. (I wrote about them in this space back in September 1999. On average, that portfolio of 11 stocks has gained about 13%.) They're unique, with characteristics of both stocks and bonds — which is probably why they're so appealing in DDM terms today. But the REITs in my table don't just have high yields. Compared with the S&P medians, they also come with higher expected returns, analysts like them better, and they carry significantly less risk. If you prefer a mutual fund, Vanguard has a REIT index offering that combines diversification with low fees.
And about that DDM Web site: dividenddiscountmodel.com is a convenient place to apply some of the ideas I've discussed, and links to the Zacks database let you analyze any stock in DDM terms. There's good value here — even if the sponsor of the site is Duke-Weeks, one of the REITs in my table.
