Last week Bloomberg featured an article
about Finnish economist Katja Taipalus and her model for detecting asset market
bubbles which she explains in her paper, “Detecting asset price bubbles with
time-series methods.” Boiling down the 217 page paper to its essentials, she is
trying to determine when stock market and housing prices diverge from their
“fundamental” value.
Along with most mainstream economists she
errs in thinking that assets have an objective value determined by the net
present value (NPV) of future income. However, as Austrian economists know
there is no objective NPV. Earnings forecasts and the discount rate used to
calculate NPV are subjective. Future earnings will vary depending on the
optimism of the forecaster. The discount rate changes according to each forecaster’s tolerance for risk.
Taipalus’ model uses the log of the yield
of a stock market index to identify bubbles. If the yield falls at a rate set
by her test statistics, then the model signals a bubble alarm. Tests of the
model show that collapses in asset prices follow within a year of the alarm and
recessions follow the collapse.
Experienced investors know that dividend
yield is the inverse of the price-earnings (PE) ratio if all profits were
distributed as dividends. Yield is E/P whereas price-earnings is P/E. So yield
can fall if prices rise faster than earnings and yield will rise if prices rise
slower than earnings. Economists love dividend yields for some reason while
most investors tend to follow PE, but they tell us essentially the same story.