Federal Reserve forecasts for GDP growth
Since 2007 the Federal Reserve have published a Summary of Economic Projections four times a year which include their forecasts for GDP, unemployment and inflation. Professional commentators and the markets often look at these forecasts to predict the path of future interest rates and also general clues as to when a "normal policy regime" is going to be with us once again.
But, if we take a look at how accurate these forecasts have been, we have to wonder if they are really worth the paper they are written on.
The following diagram shows the evolution over time of the Fed's forecasts for GDP growth for the years starting from 2011 through to 2016.
Notice any repetitive pattern? Pretty woeful, huh?
These are (supposedly) some of the brightest minds in economics, with access to the best information and data in the markets. Yet for almost every quarter since 2011, the Board of Governors and Federal Reserve Bank presidents have made the same mistake and revised down their GDP growth predictions, bar one or two exceptions, irrespective of which year they are forecasting for.
You would think someone in the Fed forecasting department would have piped up after a while and said, "um, we made this mistake last year, maybe we should think if things have changed from the past and this recovery is a little bit different from the past, maybe we need to lower our starting forecasts a wee bit". (If you're interested, have a look at an earlier piece I wrote to see how this recovery is indeed different from any past recoveries).
Now the reason for getting it wrong year after year (after year after year after ..) could be put down to any number of factors. There are behavioral finance reasons such as confirmation bias or anchoring, for example.
There is also an intrinsic reason - part of the Fed's dual mandate is 'maximum employment', which goes hand in hand with strong economic growth. Now if that's your mandate, then you're going to be putting into place policies that you think are going to achieve the mandate - and if that included strong economic growth, then you'd look pretty silly if you weren't forecasting that strong growth was indeed going to happen.
Indeed, we can see the extent to which the Fed have got it "wrong" through their famous dot-plot diagrams (well, famous in the interest rate policy-setting world at least).
In 2012 the majority of the Federal Reserve board members were predicting interest rates would have begun to increase in 2014 - the highest forecast being that interest rates would be 2.75% by the end of the year, which seems like fantasy world now. (Interest rates haven't budged since 2008 and are currently targeted at between 0 and 25 basis points - light years away from 2.75%).
In 2013 (just last year) one policy maker was forecasting interest rates at the end of 2015 would be a staggering 4.5% - whoever it was that made such a wildly optimistic (and wrong) forecast is probably very glad that the forecasts are not attributable to specific individuals.
So, enough scorn heaped on the Fed's economic forecasting abilities. Let's turn to the IMF economists.
IMF forecasts of GDP growth
This is a similar diagram to the one above, but with forecasts this time from the IMF.
The folks over at the IMF have a different approach. They start off waaaay too optimistic (like the Fed guys), then along comes some bad data and they decide everything is actually really really bad. And then finally they see that actually it's worse than the "waaay too optimistic" first forecasts but actually not quite as bad as the "really really" bad forecasts. It leads to a series of downward sloping revisions with an uptick as the forecast year approaches. Which would be okay if it was done once, or even twice. But after four consecutive times it begins to look less like misfortune, and more like carelessness (if I may be permitted butcher an Oscar Wilde quote).
The striking thing about both the Fed and the IMF is not that they make mistakes in their forecasting, but that they make the same type of mistake year after year. In the Fed's case - to forecast a return to 'normality' year after year, with the necessary revisions that this entails. In the IMF's case - to waver first between optimism then to pessimism before settling on an in-between forecast.
Professional Forecasters forecasts of GDP growth
So, I thought to myself, if we can't rely on the Fed or the IMF to break out of their individual mindsets and see a changing world, how about the guys that get paid to get these forecasts correct? The big money professionals, the economists at Goldman Sachs, JP Morgan, Barclays Capital and so on. The Survey of Professional Forecasters is a quarterly macroeconomic survey of a number of professional economists. Below, is the same diagram, this time showing the forecasts as published in the Survey of Professional Forecasters:
So, we're back to a Fed style forecasting error. Specifically that is: Step 1 - forecast optimistically that the US is returning to historical growth rates; Step 2 - then revise downwards when it's apparent that this is not the case; Step 3 - repeat error following year.
So, what are we to make from all this? Well, firstly, don't rely on the forecasts - they create a lot of headlines when they are published but the professionals are a long way from having a crystal ball. But perhaps the stronger message to take away is the persistence of errors. Once a person or an institution has an inherent belief, then there is a tendency to repeat the error on a consistent basis. And if you take anything away from this article (other than enjoyment and new found knowledge) that's a message worth remembering any time you make an investment that doesn't work out the first time. It ain't necessarily going to work the second time either.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.