From LOUIS RUKEYSER February 2005:

Markets Look Absolutely Expensive? Think Relatively

Phil DeMuth

Phil DeMuth is President and Principal of Conservative Wealth Management (www.phildemuth.com), a registered investment adviser to high net worth individuals and their families, and co-author (along with Ben Stein) of Yes, You Can Time the Market!

Today's markets remind me of many politicians: bloated and
overvalued. Whether we turn to stocks, bonds, real estate or commodities, valuations are inflated compared to long-term historical averages. What is an investor to do when everything is expensive, and therefore likely to offer subpar returns going forward?
That is the $64 question at the start of 2005. Even cash is a
guaranteed losing investment after taxes and inflation. There's no
Free Parking anywhere, and yet we have to put our money somewhere.
In Yes, You Can Time the Market, Ben Stein and I argued that it was vital to value the market before plunging in with your hard-earned dollars. The theory we attempted to debunk was simple: the stock market went up 10.6% a year over the long run, so it really didn't matter when you bought. If true, the whole concept of "price" would have no meaning when it came to the stock market. Ben and I thought this was impossible. How could price be crucially relevant in every other marketplace in the world, but somehow be irrelevant as far as the highly efficient stock market was concerned? Well, we went through a century of data and discovered that price was extremely relevant to the S&P 500: people who bought when the market was low did vastly better than those who bought when it was expensive. It didn't make any difference for short-swing trading, but it sure did for long-term investors. While we looked at a number of different measures, including dividend yields, price/earnings ratios, the so-called Fed Model (earnings versus bond yields), it turned out that the humble price of the market, even taken all by itself, proved to have considerable predictive validity.
Yes, You Can Time the Market was a book about the S&P 500, but its implications can be extended to all asset classes. Price always matters. When it comes to stocks, the lower the price, the higher the expected return. It's that simple. But just because it's that simple doesn't mean it's that easy. Jack Nicklaus' golf swing is simple, too, but just try to do it yourself.
How can we apply this insight? Just because diversification is the investor's best friend, it does not follow that we should divide our assets exactly evenly among various types of investments. In situations like today, when all major markets are overvalued, the sensible thing to do is pit one market against the others, or one part of the market against another. We can do this by comparing relative valuations, and buying more of what's cheap and less of what's expensive. Asset allocation in the present becomes market timing in practice. Even if everything is overvalued, some market segments are more overvalued than others. Take the U.S. stock market as a
whole, but dissected into four parts. This is captured beautifully by four
indexes constructed by the Frank Russell Company: the Russell 1000 Growth Index of the largest 1000 growth stocks; the Russell 1000 Value Index, which does the same for value stocks; the Russell 2000 Growth Index, which captures the returns of small-capitalization growth stocks; and finally, the Russell 2000 Value Index, which tracks small-cap value stocks. Put them all together, and you have the Russell 3000: virtually the entire U.S. stock market.
Under ordinary circumstances (which is to say, never), a diversified
investor might spread his bets equally across the four corners of the U.S. stock market, as shown in "Equal Distribution." This way, he's got the bases covered.

EQUAL-WEIGHTED US MARKET
Large Value 25%
Large Growth 25%
Small Value 25%
Small Growth 25%

But now, let's apply the lessons of Yes, You Can Time the Market. We know that small-cap stocks and value stocks have been on a tear lately. Their prices are way up. Meanwhile, the large-cap growth stocks of the S&P 500 were pummeled when the Internet/telecom bubble burst, and many haven't really recovered. If we compare where each of the four Russell indexes is today with what has been its average level since 1995, we discover that three of the four are selling for more than usual. Here's by how much (as of 12/31/04):

MARKET OVER/UNDERVALUED
Large Value 34%
Large Growth -1%
Small Value 74%
Small Growth 19%

Small-cap value stocks are priced 74% above where they have been, on average, since 1995. (In other words, the level of the Russell 2000 Value Index is 74% above its average level since 1995.) But large-cap growth stocks (such as those in the S&P 500) are actually priced 1% less than their average price since 1995. If the way to make money in the markets is to buy what's cheapest, maybe the smart course is to look at large-cap growth stocks right now. If you are an individual stockpicker, this is the pool where I would saw a hole in the ice
and drop a line. "Big Growers" below, shows the Russell 1000
Growth Index's top 10 holdings. You may have heard of some of them. If you don't care for these, there are 990 others vying for your attention.

BIGGEST GROWTH STOCKS
Pfizer
Microsoft
Johnson & Johnson
Intel
WalMart Stores
Cisco Systems
Procter & Gamble
Intl Business Machines
Dell
American International Group

If, like me, you prefer buying whole indexes, I would allocate my holdings as shown in Index ETFs, below. Fortunately, iShares has an exchange-traded fund that tracks each of these quadrants. Each ETF's ticker symbol is shown next to the weighting.

VALUATION-WEIGHTED US MARKET
Large Value 24% (IWD)
Large Growth 37% (IWF)
Small Value 9% (IWN)
Small Growth 30% (IWO)

By allocating your dollars according to these percentages, you distribute them inversely to the degree of over- or under-valuation of each segment. Even if the market gets clobbered this year (and don't say it couldn't happen), you will still be better off buying more of what's cheaper than too much of what's overpriced. If you can't buy low, then at least buy as low as possible. -- Phil DeMuth