Valuing Wall Street: Protecting Wealth in Turbulent Markets [Paperback]

Smithers and Wright use as a measure of valuation for stocks a statistic called "q" (or Tobin's q, named after Nobel laureate and Shiller colleague James Tobin). q represents the value of equities divided by the cost of replacing the underlying capital stock. So you might expect the stock market to be worth somewhere near q=1, where companies are worth what it cost to build them; historically, the average value of q is near 1.

Smithers and Wright show that changes in q and equity prices are almost identical, since the cost of replacing the capital stock changes so little. They also show that high values of q are associated with terrible subsequent returns. They show how a simple strategy of selling when q rises to 1.5 and buying again when q falls below 1 * trounces * a buy-and-hold strategy. And they top it all off by showing that today's level of q, around 2.5, is unprecedented. So SELL!

The reason the book is so much better than Shiller's is that Smithers and Wright give a coherent, fact- and theory-based argument for why q should be used to value stocks, not just P/E, stock earnings yield compared to bond earnings yield, or other popular measures. Shiller just used P/E and told us to sell due to today's high P/E; he did not even consider, not to mention try to debunk, other theories of valuation.

Smithers and Wright point out, for example, that in the early '30s, P/E was very high due to the depressed depression-era profits of companies, but that q was very low, providing the buy signal of a lifetime that would have been missed by looking at P/E alone.

The only negatives of this otherwise excellent book are: (1) Like most finance books, this one would gain from adding computations of after-tax returns, which shift us away from fancy trading strategies and towards buy-and-hold in taxable accounts. (2) They should admit that there are significant differences between today's economy and economies of the past. For example, in an economy such as ours where intellectual property is paramount and provides barriers to entry, firms' values may stay above the cost of replacing capital.

The best service Smithers and Wright have provided in VWS is in demonstrating that contrary to popular wisdom (and books by other authors), a "buy and hold" approach to investing in the stock market has not necessarily provided a sure path to riches (or even a comfortable retirement). How well an investor has done depends crucially on, 1) what year he/she began investing in stocks, 2) the duration over which he/she invested, and 3) the year in which he/she retired. Over many periods, investors would have earned a much higher risk-adjusted return by holding bonds or cash instead of stocks. Thus, the authors conclude that with stocks set to decline by over 50% in the near future, any investor with less than, say a twenty year time horizon, would be well-advised to sell their stocks now.

Of course, Tobins q only serves as a reliable indicator of future stock market performance if the factors responsible for its accuracy in the past are still valid today. Some of the other reviewers here have done a good job of pointing out the problem with measuring corporate net worth exclusively in terms of real assets. If US corporations were to capitalize (rather than expense) all of their investments in intangible assets, net worth would be significantly higher and Tobins q would not appear quite so top-heavy. Smithers and Wright attempt to discredit this point by suggesting that for every firm that adds value to its balance sheet (through intangible assets that perform well), there is another firm that destroys value (through intangibles that perform poorly). Thus, their argument goes, in the aggregate corporate net worth is no higher than what is reflected by real assets in the denominator of q. The problem with this defense is that one could logically make the same argument for real assets. For every factory or machine tool that adds value to a company by contributing to profits, there is probably a factory or machine tool at another firm which contributes to loss making. The logical (but absurd) conclusion of this line of reasoning is that in the aggregate, the real net worth of US corporations is ZERO.

Smithers and Wright are supposedly economists by training but they have forgotten the first rule of economic valuation: sunk costs are irrelevant. The reason Tobins q is unlikely to accurately predict fair value in the stock market is that intrinsic value is not calculated by adding up total net assets but rather by discounting to the present time, the future cash flows which those assets can be expected to produce. Clearly, today's investors in the US stock market believe that is a very large figure indeed. Perhaps for their next book, the authors can assess the probability that these expected cash flows will actually materialize. That would tell us whether or not the stock market is overvalued and if so, by how much.