Astrology and Economic Forecasts

John Galbraith once said that “The only function of economic forecasting is to make astrology look respectable.” Although many of us are avid readers of economic forecasts issued by the OECD, the IMF, and the EU (the government’s forecasts attend to suffer from a general lack of creditability), it is questionable if our confidence in them is well founded. In my opinion, which is based on my experience, it is not.

Firstly, a large number of economic models try to predict the future by extrapolating the past. The current crisis, like so many others, has highlighted the folly of this method. Such models also predicted in the 1950s that the USSR would become the world’s most powerful economy (its economic growth rate was three times that of the west at the time), and the same was said about Japan (remember the best seller Japan as Number 1, by Harvard Professor Ezra Vogel?), the Asian dragons in the 90s, and now it’s China’s turn. The logic of projecting past growth rates onto the future is an intellectual and economic fallacy, as stated by Paul Krugman in his excellent paper “The Myths of Asia’s Miracle.” For Vogel’s followers, perhaps this BoA-ML graph will serve to illustrate how much his forecasts were worth:

Secondly, when making forecasts economists tend to favor consensual thinking, which allows slight deviations from the established average. It is, however, very unusual to make forecasts based on methods that are far removed from a series of parameters, given that it is easier to explain a mistake due to consensual thinking than a mistake based on a completely out of consensus prediction. For example, before 2007 the IMF extolled the merits of the “originate and distribute” credit model, which is the cause of numerous current problems. It praised the solidity of the Icelandic banking system, and in the summer of 2008 stated that “The worst of the crisis had passed.” Moreover, a brilliant UBS economist (George Magnus) correctly predicted in March 2007 the financial meltdown (or Minsky moment, as it is known, named after economist Hyman Minsky who warned of the dangers of speculation by financial institutions), while other UBS economists stated in May 2007 that “The risk of global recession was 0%,” (a statement that if I’m honest I have to admit I would have agreed with at the time) and the bank itself was accumulating structured products on its balance sheet.

---

---

Thirdly, when an economist issues a forecast that differs substantially from consensual thinking (as in the case of Roubini, who had been predicting the crisis for years), the prediction that is so different to that of consensual thinking is often made for so many years that in the end it has to finally happen. If an investor heeds the warning every year they will probably have been ruined by the time it actually happens, which only goes to prove that Keynes was right when he said “Markets can remain irrational longer than you can remain solvent.”

Fourth, very few economists (and I have to admit that I am not one of them) have a deep knowledge of financial markets, about bank accounting standards, about the valuation of derivatives on banks’ books and the systemic risks that link commercial banking and investment banking. Without this essential knowledge, economic forecasts in a financial economy like ours would have very little worth. Hence, it is interesting that the Fed initially tried to bail out Merrill Lynch with Wachovia, ignorant of the dire situation Wachovia was itself (and the FED was Wachovia’s regulator, remember). Moreover the two large Irish banks that left Ireland needing a bailout from the IMF and the EU “passed” the stress test carried out in the summer of 2010.

Not convinced? The black line in the following graphs show the forecasts for economic growth and inflation in the US for the year ahead based on the consensus of economists. The red line shows what really happened.

So who should we listen to? Personally I am far more interested in the surveys on future activity published in the purchasing manufacturers index (PMI) and its US equivalent (ISM). People who are in the front line taking investment and contracting decisions tend to make better predictions of the future of the economy, and it is they who supply the information for these surveys.

Whatever the case, economists remain the only people fortunate enough to get paid twice for the same thing. First, when they issue a wrong prediction, and then again when they explain why they got it wrong.

---

–Professor Ignacio de la Torre, Academic Director, Master In Finance Programs, IE Business School, Spain

Comments

[Milo Jones]

Great piece! You see exactly the same phenomena among both producers and consumers of intelligence and geopolitical forecasts (which is to be expected, as both involve decision-making about non-linear phenomena under conditions of uncertainty). See: http://www.jstor.org/pss/2626667

A former member of the National Intelligence Council (NIC) reminisced about one key difference, however, between business/academic and intelligence forecasts: intelligence analysts are usually further hobbled by the "cult of secrets". He recalled, “At the NIC, we used to quip that if academics sometimes did better than intelligence analysts, it was because the former weren’t denied access to open sources!"

[Dan Jordan]

Economists relying on a Keynes / Hicks model of the world seem to have been very accurate in recent years. The correct information is out there if you know where to find it.