June 20, 2015

Peter Lynch was right - you can't predict the economy

This is a quote by the famous investor Peter Lynch
I spend about 15 minutes a year on economic analysis. The way you lose money in the stock market is to start off with an economic picture. I also spend 15 minutes a year on where the stock market is going.
According to Peter Lynch, it's easy to overestimate the skill and wisdom of professionals who is trying to predict the economy. If an "expert" on television with a fancy job title is predicting the economy, it's easy for those who haven't studied Peter Lynch to believe the expert knows something that you and me don't understand. The truth is that the expert knows as little as everyone else does, even though many experts believe they know something. 
Peter Lynch describes it more deeply in his book One up on Wall Street. According to the book, there are 60,000 economists in the US. Many of them are employed full-time trying to forecast recessions and interest rates. If they could do it successfully twice in a row, they'd all be millionaires by now. But the truth is that they aren't. 
The idea that it's impossible to predict the economy might first sound strange. I myself had a hard time before I accepted this fact. The question is why so many experts are wasting their time, even though they are not always aware of it, while the amateurs continue to listen to them.
Because Peter Lynch is not describing exactly why in his book, I had to wait a few years to really understand why it's impossible to predict the economy. I finally found the answer in the book The signal and the noise by Nate Silver. That book includes several chapters on why so many predictions fail, including why it's impossible to predict earthquakes, why US failed to predict Pearl Harbor, and why some people didn't trust those who predicted the Hurricane Katrina. Another chapter in the book is about why it's unnecessary to listen to statements like:
  • The economy will create 150,000 jobs next month
  • GDP will grow by 3 percent next year
  • Oil will rise to $120 per barrel

The secret truth about economic forecasts
According to Nate Silver, economic forecasts are blunt instruments at best, rarely being able to anticipate economic turning points more than a few months in advance. In fact, these forecasts have failed to "predict" recessions even once they were already under way: a majority of economists didn't think we were in one when the three most recent recessions were later determined to have begun. 
Peter Lynch has the same ideas as Nate Silver. Again a quote from his book:
Nobody called to inform me of an immediate collapse in October [1987], and if all the people who claimed to have predicted it beforehand had sold out their shares, then the market would have dropped the 1,000 points much earlier due to these great crowds of informed sellers.
Every year I talk to executives of a thousand companies, and I can't avoid hearing from the various gold bugs, interest-rates disciples, Federal Reserve watchers, and fiscal mystics quoted in the newspapers. They can't predict the markets with any useful consistency, any more than the gizzard squeezers could tell the Roman emperors when the Huns would attack.
One of the latest examples showing how inaccurate experts are when predicting the economy is the Credit Crisis of 2008. About one year earlier, economists in the Survey of Professional Forecasters expected the economy to grow at a just slightly below average rate of 2.4 percent in 2008. And they thought there was almost no chance of a recession as severe as the one that actually hit the world in 2008 where GDP shrank by 3.3 percent. It was a scenario the economists thought would happen with a probability of just 3 percent. 
The above isn't just one case that confirms the rule. In a report with data from 1968 up until now, predictions by the Survey of Professional Forecasters for GDP fell outside the prediction interval almost half the time. So it is clear that the economists weren't unlucky - they fundamentally overestimated the reliability of their predictions.

Why is it so difficult to predict the economy? 
According to Nate Silver, economic forecasters face 3 fundamental challenges:
  1. It is very hard to determine the cause and effect from economic statistics alone. There are millions of statistical indicators in the world, and a few will happen to correlate with stock prices and GDP, even though it's just a coincidence. For example, ice cream sales and forest fires are correlated because both occur more often in the summer, but there's no causation. With so many economic variables to pick from, someone will find something that fits the past data, even though it's just a coincidence. 
  2. The economy is always changing. If the economists have predicted an upcoming recession, the government and the Federal Reserve will take steps to soften the recession. So forecasters have to predict political decisions as well as economic ones. Also, the American economy has changed from an economy dominated by manufacturing to one dominated by the service sector, which will make old models that used to work obsolete. 
  3. The data the economists have to work with isn't that good. With different governments with different policies, the past data will be subjected to these different policies, making it difficult to use old data and say that the data will be the same as today. Also, the data available may be limited. If you look at data between 1986 and 2006, which is 20 years, but the problem is that these years contained just two mild recessions.  
So what's the solution?
We know that Peter Lynch's solution is to spend about 15 minutes a year on economic analysis. Nate Silver, on the other hand, argues that if you have to listen to predictions of the economy, you should listen to the average or aggregate prediction rather than that of any one economist. These aggregate forecasts are about
  • 20 percent more accurate than the typical individual's forecast at predicting GDP
  • 10 percent better at predicting unemployment
  • 30 percent better at predicting inflation. 

No comments:

Post a Comment