How Wrong are the Experts?
Very wrong, it turns out.
Henry Blodget, who runs the excellent Business Insider, tells us -- and shows us -- exactly how off the mark economists and business analysts are:
It's prediction time again--the time when analysts and economists tell everyone what they think will happen next year.
And so it's a good time to remind everyone that analysts and economists have no idea what will happen next year.
Well, okay, not no idea. But pretty much no idea.
(And I say this as a former analyst. And it's not some huge revelation or secret. Most professional analysts and economists, if they've been around a while, have learned the hard way that economic forecasting is like driving fast at night. Thanks to your headlights, you can see what's coming a few hundred feet in front of you, but you can't see beyond that. And if you're going too fast, by the time you see the unexpected curve or deer [black swan!], it might be too late).
And he points us to a couple of excellent graphs. The first, from Bloomberg, is something called the Citigroup "Economic Surprise Index." (Not making that up, by the way.) It shows the difference between what economists said was going to happen and what actually happened. When the index is zero, the economists were correct. The index doesn't hit zero a lot:
The other interesting graph is from James Montier at GMO, and is even funnier. This graphs the average estimate, from the experts, for the coming year's GDP against the actual GDP:
Of course, it isn't news to us -- or it shouldn't be -- that the consensus estimates of approved experts is often, maybe even mostly, wrong.
But it always seems to take them by surprise.
Why is that?
- Comment (15)
- · Quote
- · UnfollowFollow (2)
Comments:
Jun '10
Re: How Wrong are the Experts?
Six out of ten portfolio managers can't beat the average of the major indices and yet they make their living prognosticating. Might as well slaughter some chickens and examine the entrails for all it's worth. Wasn't there a South Park episode where they did just that?
Jun '10
Re: How Wrong are the Experts?
They could predict, if history repeated itself, but history never exactly repeats. Even if the financial variables look the same, the array of investors is always different, so the triggers will vary and the pace will vary, both affecting each other.
Jul '11
Re: How Wrong are the Experts?
Daniel Kahneman's new book addresses this topic very nicely. Makes a great gift!
Mar '11
Re: How Wrong are the Experts?
Very interesting that the consensus forecast remains within a such a narrow band for such a long period of time.
Dec '11
Re: How Wrong are the Experts?
Fatal conceits abound among the predictors!
The BEST portfolio managers don't make macroeconomic bets. They manage their portfolio like they owned each company entirely, prioritizing sound business practices, great cash flow, low debt, barriers to competition. They also buy the stock at a discount to the value of the company (recognizing that their predictions can be wrong and they need a buffer). Value investing seems like a very conservative philosophy to me!
Edited on December 20, 2011 at 4:50pmOct '10
Re: How Wrong are the Experts?
The problem for forecasters is that unpredictable, unforeseen events have real impacts, but by definition, they are unpredictable.
That is why forecasters sometimes argue that they would have been right had certain unforeseen events not transpired. But our question should then be, then what good is your forecast?
Edited on December 20, 2011 at 5:01pmFeb '11
Re: How Wrong are the Experts?
I don't know, by my count in the second graph they were right 23 times -- when the prediction line intersected the graph of what actually happened. That's not so bad in 39 years. That's almost being right once a year.
In the first graph, they were right 19 times over 4 years. That's being right over 4 times per year. Those guys are brilliant compared to the first lot.
Re: How Wrong are the Experts?
Thanks! It's here, just bought it on Kindle...
May '10
Re: How Wrong are the Experts?
This is why my cousin relies strictly on a company's recent history measured by objective statistics (what he calls "fundamentals") to choose his investments. He beats the index every year.
But, like Kudlow, my cousin is always bullishly optimistic about the economy regardless of the circumstances. His macroeconomic predictions are based on desire, not reality.
There was a Financial Times article in late October which included the following in just the opening paragraphs:
Europe's leaders delayed fiscal doom by stealing from their own countrymen and inviting massive investment (with conditions) from China. And our investors had a banner day.
"Animal spirits" indeed.
Aug '10
Re: How Wrong are the Experts?
Rob,
Since I'm sure you've read William Goldman's books on screenwriting (if for nothing more than the anecdotes), then you must know that Nobody Knows Anything. That applies in politics and business as much as it does in the arts. We are all just guessing.
May '10
Re: How Wrong are the Experts?
When I first scrolled over this story, I assumed those graphs were related to climate science. One could write the same post on that subject.
Apr '11
Re: How Wrong are the Experts?
Monetarist and Keynesian theories give insufficient (or no!) weight to capital structure. Take the simplest thing, say a paper napkin. It reached your hand by decades of capital investment. Someone had to plan the tree farm long ago. Someone had to invest on the pulping machinery long ago. Transportation relies on long chains of capital investment.
No one really understands this complicated structure. No one really can, but traditional macroeconomics attempts the impossible. Monetarism and labor-base theories accomplish this by fibbing. They hide uncertainty and lack of knowledge behind blind, semantically empty statistical aggregates.
This is wrong. Macro ought to work out the logic of things, not the measure of things. I like Garrison's book, TIme and Money.
Aug '10
Re: How Wrong are the Experts?
If the forecasters were "usually" correct, then command economics would be possible. We would be able to accurate predict performance and demand, and we'd all be practicing Psycho-Historians.
May '10
Re: How Wrong are the Experts?
BI is occasionally enlightening, particularly regarding some of the WS bank stuff. But it is a consistently left-wing group on most matters.
May '10
Re: How Wrong are the Experts?
Given the "never before seen to this magnitude" structural problems facing the global economy today, I don't know how anyone can project/predict with a straight face. In today's world, it would take a very "brave" person to integrate economic consensus projections into their investment plans. Given the 100 - 300+ swings that happen daily with the Dow (that show no signs of abating) picking the days for comparison would be an art form rather than science.
With the cloud of black swans blackening out the sun, we are currently on the highway with neither ambient light nor headlights. good luck to us.