Tobin's Q and Capital Cycle theory

From Edward Chancellor's Capital Account (Marathon Asset Management letters)

p50
"In fact, the overvaluation of assets by the stock market is probably one of the key drivers of the capitalist process. After all, companies should convert cash into assets only if they believe they can add value.
Thus, when Tobin's q ratio is in excess of one, there is a strong incentive for firms to increase capacity. All too often, this encourages undisciplined expansion, which in turn leads to excess capacity and falling profitability, causing share price to tumble. This process of regression to the mean is always at work in the stock market. In the US, however, the stock market does not appear to revert to the mean except in the very long run. For a period of more than 15 years to the mid-1980s, share sold for well below replacement value as measured by Tobin's q. This suggests that the current phase of full (or over) valuation could also last for an extended period."

I agree with the thesis to a point. But do corporate managers think "Wow, I can generate $2 in market value by investing $1 into my asset?" I don't think so. They are mostly looking at profitability and pro-forma ROIC statements. I don't think anyone is that naiive to think that P/B ratio will hold when they increase investment. That's like Phil Fisher's comment on how high P/E ratio stocks have more torque per dollar in earnings increased because it will add that much market value to the shares.

I think the real reason behind overinvestment is that the managers tend to be extremely optimistic when the market is. They overestimate their ROIC, just as the market overestimates future profitability of the firms.

Still, one important tool that can be used when evaluating possibility in oversupply can be gathered - you can look at not only the competition entering the industry but what the industry aggregate CAPEX is doing. Could be interesting, especially when you have oligopolistic industry structure such as the Canadian Telecoms.