In Tobin's Q, you don't want to know what those assets cost, you want to know what their replacement value -- their current cost --is.
He is picking a nit with analysis by PIMCO's Bill Gross. Gross is arguing--and Salmon does not really disagree--that Tobin's q is really low today, which could mean that stocks are a screaming buy.
Tobin's q is the ratio of the market value of capital, as measured by stock prices, to the replacement cost of capital, as measured by the cost of replacing capital. It is one of my favorite concepts in economics. I think of it as an arbitrage indicator. It is high when it is greater than 1.0, and low when it is less than 1.0. When it is high, it means that paper wealth is dear and physical wealth is cheap. When it is low, it means that paper wealth is cheap and physical wealth is dear. The arbitrage principle is to buy cheap and sell dear.
When the ratio is high, the indicator is saying "Sell stocks! Buy physical capital!" In 1999, the indicator was saying, "Buy a web server! Create an Internet stock and sell it!" That is, the Tobin's q for Internet stocks felt like it was at least 100.
When the ratio is low, the indicator is saying, "Sell physical capital! Buy stocks!" Today,. that means it is telling firms not to make any investments. Suppose that you are a company in the unusual position of having cash or borrowing strength and a desire to expand capacity. You should not build new facilities. Instead, you can get them much more cheaply by purchasing a rival company on the stock market.
Most of us cannot arbitrage against today's low value of q. We cannot simultaneously sell physical capital and buy common stock. We can buy common stock, hoping that the way that q returns to 1 is by stock values rising. But the other way for q to get back to 1 would be for the replacement cost of capital to fall.