HOW ECONOMIC FORECASTING CAN'T HELP

After we ran yesterday's article on forecasting, many readers wanted to know not only about our Peruvian underwear, but also about our forecasts of the market. To all of those that asked about the future, we are humbled by the fact that you care enough about what we think. Honestly though, despite having a formal background in economics and being involved with the markets for 25+ years - we have no idea. In fact, that was our point yesterday. Despite having large staffs, huge budgets and the data of the largest banks on earth, major economists came in with more than an 80% variance and were directionally right only about 50% of the time. Our point wasn't that these economists are bad, only that the economic prediction business, like earthquake prediction, is currently beyond human abilities. The analogy to earthquakes is apt, as we know exactly why earthquakes occur, yet no one has successfully predicted a major earthquake in the history of mankind. In similar fashion, we know why markets move after the fact, but there are too many variables that interact in a complex fashion to currently be predictive with any accuracy. To date, there are no standard economic forecasting models that display any useful predictive capacity. As we showed yesterday, economic predictions each year follow a predictable pattern that mimics non-economic forecasts such as earthquakes, sporting events, etc. These findings are tightly grouped and biased to the upside. In other words, even in the quantitative field of economics, all the modeling work gets overridden in either design or interpretation of the output and a behavioral bias is injected. The result is accuracy similar to a coin toss, dice roll or wild guess. So, if we are all just guessing, how does that help bank management increase earnings in the future? For starters, since predicting the economic future is a random event, bankers shouldn't waste their time worrying about a forecast. Unfortunately, economic prediction is a cost free activity so anyone can make a prediction that has very little ramifications if they are wrong. Further, to get ahead in the economic prediction business, game theory shows that the way to do so is to be an outlier on either side of the spectrum. Thus, mathematically speaking, the more wild the prediction, the lower probability that it will occur. As such, it is safe to take all predictions with a grain of salt. This is not to say that economists are worthless. While they aren't necessarily good at forecasting, they are good at understanding the forces of the economy. Thus, while attending a state bank association economic briefing won't help you find out where markets are going, it will help you understand components such as housing, business formation, spending and other trends. Further, understanding economic forecasts help bankers do one very crucial thing - role play. A good test of any management team is to ask, "How will the bank's strategy and tactics change in 2012, should an outlier economic prediction come to fruition?" This "stress testing" is one way for your bank to prepare for the economy to either take off or go into another deep recession and handle things, even if it is the higher probability middle path. Finally, all the extra resources that used to go into worrying about what duration should be targeted or what the stock market is going to do, can be devoted to three endeavors that matter most in managing a bank - profitable customer management, leverage and asset allocation. Choosing the right allocation for each lending and investment line; managing that allocation for geography; picking the right leverage or the desired risk/reward tradeoff and working on building a portfolio of profitable customers is multiple times more important than getting the level of interest rates right. All that said, if you are looking to make an equity forecast, markets rise about twice as much as they fall (likely an outcome of basic human overconfidence). If you don't have a clue what you're doing, it's a pretty decent bet to guess markets will rise. That said, if you live near an active fault as we do, always predict an earthquake and someday you will be right.