Short-termism just won’t die. Despite obvious counter examples like Amazon, Tesla and Netflix it clings to its long-term life. But beyond the counter examples, there is a bigger conceptual problem with the idea of short-termism, namely that it overlooks the obvious fact that all hard information is short term. Companies report their information quarterly. All the information from past quarters is old news. The only new hard financial data the market gets is this quarter’s performance. It never gets hard data about the “long term.” Of course, a lot of people have beliefs about what the long term looks like, in particular corporate CEOs. But those beliefs are probabilistic projections, not data. The only data are the short-term quarterly reports and possible other releases of financial data. But by definition all that information arrives “right now” – in the short term.
Thomas Bayes laid down the basis for understanding how information affects beliefs more than two centuries ago. Bayes was aware that with regard to the future all we can have are probabilistic beliefs. He called these initial beliefs the prior distribution. For example, I have a prior distribution of beliefs regarding what Tesla’s earnings will be over the next five years. When information arrives, such as Tesla’s quarterly report, Bayes shows how rational actors will revise their probabilistic beliefs. He calls the revised beliefs the posterior distribution.
Bayes theory makes it possible to have a more meaningful discussion of short-termism. In the context of his analysis, the market would be “too short-term oriented” if investors revise their prior distribution “too much” in response to quarterly information. This of course begs the question of how much is “too much?” The most common complaint that executives seem to have is that in response to negative financial information investors revise their long-term beliefs too far downward and as a result the stock price drops significantly. But the key thing to realize is that investors are still pricing the stock based on their expectations for long-term performance, they have just adjusted those expectations more than the CEO would like. All valuation is “long term.” What people disagree about is what that long-term looks like and how those long-term views should be revised as information becomes available.
Article by Bradford Cornell's Economic Blog
