On ROE, first and foremost we must recognize that any debt inflates ROE. Returns on assets, capital, etc... are more conservative.
Further on ROE, is high ROE itself the goal? Not at all. A high ROE is only worth something if it is sustainable, and only then as a marker that there should be market power and hence pricing power in the business. What a high return on equity/capital tells you is that there is an increased ability to create something extra out of a given level of investment/assets, thanks to a brand, moat, whathaveyou that allows either monopolist power when dealing with customers or monopsonist power when dealing with suppliers. As an example, WalMart's special attraction as an investment lies in its monopsonist rather than monopolist features.
Whenever I see a high ROE that appears sustained, I need to identify the moat before I get too comfortable. If the barrier isn't there, then that high ROE should be a warning sign. The problem with the midwest industrials is that the moat isn't readily identifiable. Hence, nearly by definition, a lot of their super-high ROE's are cyclical, and we are near that peak. So there will be a lot of favorable ROE vs. P/B comparisons as the stock price anticipates a downside and the current earnings still represent the upside.
There's no two ways about it for me. Either a high ROE means something because it is proven sustainable with an identifiable moat that gives it market power, or a high ROE means something else. Only the former attracts my interest on the long side, which is of course Buffettesque.
Finally, we have recently spoken here of costs of capital in the 8-10% range. Well, if we are assuming that the high returns on equity are meaningful and hence relatively unleveraged, then we should be assuming costs of capital quite a bit higher, as equity is of course typically more expensive than debt.
Good investing,
Mike |
