A Prophet Who Dares Admit the Limits of Prophecy

Predicting the future has its limits (iStockphoto)

Fill in the blanks. I dare you. And then read the rest of the post to see who wrote this, and what it’s about.

We’re currently finishing up an hour-long radio program called “The Folly of Prediction,” and I have to say that it was bracingly refreshing to read this paragraph and the column that follows it.

After more than a quarter-century as a professional _________, I have a confession to make: There is a lot I don’t know about _________. Indeed, the area of __________ where I have devoted most of my energy and attention — the ups and downs of _________ — is where I find myself most often confronting important questions without obvious answers.

It’s from Greg Mankiw’s monthly New York Times column:

After more than a quarter-century as a professional economist, I have a confession to make: There is a lot I don’t know about the economy. Indeed, the area of economics where I have devoted most of my energy and attention — the ups and downs of the business cycle — is where I find myself most often confronting important questions without obvious answers.

And here is another good bit from it:

By its nature, punditry craves attention, which is easier to attract with certainties than with equivocation. But that certitude reflects bravado more often than true knowledge.

In my view, the sooner we admit how poor we are at predicting the future, the better we will become at (incrementally) predicting the future.


donzelion

It shouldn't be hard to admit limitations of future predictions: weather patterns ought to be far, far simpler than economic developments to predict, yet what system correctly predicts the temperature 90 days in advance at any particular location, or the path of a hurricane with precision? Do we have economic models more rigorous than models intended to measure the effect of heat on gas molecules?

James

Yet some things about the future can be predicted with quite impressive precision. It is, for example, quite possible to launch a spacecraft (Mercury Messenger) in August 2004, predict that it will arrive at its destination 6 1/2 years later, and be off by only a few seconds. Even where we can't make accurate long-term predictions, as with weather, it's possible to know why we can't make them (chaotic physics and those darned Brazilian butterflies), and even to quantify pretty exactly just how inaccurate the predictions are going to be.

None of this seems (to my admittedly limited understanding) to apply to economics: it's only useful when explaining what has happened.

pablo

He seemed so much more certain about things in the early Bush years . . . .

DaveyNC

As he notes, he has aged and learned the limits of his knowledge.

Medical Quack

Analytics can tell us a lot and just for blogging purposes I review the emails I get from Recorded Future, the program that Google and the CIA have invested in and of course there's the ever most important interpretation:) It's funny I made a post a few months ago about the next potential 12 step program that we may see, data addiction and abuse:) I know it sounds funny but without balance and intelligent interpretations, it's goes wacko at times and we see it in the news too:) Will this be the next frontier where we have admitted that we are helpless to change the code that programmers and develop and as end users have to work with? Will developers admit to the fact that if it's there, code and substantiate and by all means get investors at any rate:)

Can those who write for "desired" versus accurate results break the habit? Who knows but the algorithms of data addiction are quickly growing upon us everywhere:)

http://ducknetweb.blogspot.com/2010/08/data-addiction-and-abuse-up-and-coming.html

Read more...

Dov

For once I agree with Mankiw. When I ran an institutional investment research group I always warned my team that we stood foursquare to the wind and predicted the past, might hazard a guess on the present but stayed the heck away from the future at all costs. Demography is destiny in the property markets and it's the one area where you can see the future walking around today and make reasonably good demand forecasts. Construiction takes a long time so we can also make reasonably good supply forecasts out two to four years.

Herman Brodie

We knew it all along.
http://www.unexpectedutility.com/economics/forecasting-yesterday

smartalek

Apologies for the OT; really not trying to thread-jack here (and it is this post that triggered this, as it's an excellent example), but...

I wonder if you could please consider -- at least for a while, til most of us can habituate and internalize the New Way -- putting up a tag (possibly something akin to the traditional "spoilers" tags) when linking to anything on the New York Times website?
For those of us who have (at least for the time being) chosen not to buy one of their subscriptions, and thus face a strict monthly numeric limit on access to NYT items, it would be helpful to know before we get there that clicking this or that link will carry a cost, requiring a determination of relative marginal utility.

(Huh -- maybe this isn't so OT after all)

Yes, of course we can see, before clicking, where links will take us if we have (a) chosen to show our browser's status bar, and (b) remember to check it every time before we click any links. But, as I suggested, it might take a while for some of us to get used to this new world of NYT scarcity...

Thanks for all your brilliant, instructive, and entertaining work.

Read more...

Shane

Inspired by the likes of Freakonomics and Nassim Nicholas Taleb's The Black Swan, I posted a thread on a politics discussion forum dedicated to predictions a few years ago.

We used it both to repeat old, incorrect (and forgotten) predictions and to venture our own. It has been useful and every so often someone ressurects the thread to say "Hah! I was RIGHT about Iraq!" or "Hah! You were WRONG about unemployment!" Apart from cheerful point-scoring I hope it's made us a little more humble as our predictions so often go wrong.

john

So you're predicting that prediction will remain difficult?

Eric M. Jones.

Check this guy: http://www.agreeley.com/author.html

In 1979 or 1980 Father Andrew M. Greely wrote in the Book of Predictions (Wallace, Wallace and Wallechinsky):

“Before 1990, the present communist government in the Soviet Union will be overthrown either by a violet internal revolution or more likely by a “social democratic” faction within the party. Some of the constituent republics (the Ukraine, for example) will obtain authentic separate status. The Soviet colonies in Eastern Europe will then go the same route.”

This would have been an astonishing prediction even six months before the USSR collapsed in December 1991!

Read the1980 "Book of Predictions" for much delightful foolishness that drives home the point that predictions are fraught with peril. All except for Father Greely....!

Barry Ritholtz

This is a well known phenomena -- see this young whippersnapper's column from back in 2005: The Folly of Forecasting: http://www.thestreet.com/story/10226887/apprenticed-investor-the-folly-of-forecasting.html

Excerpt:
"Unfortunately, investors all too often give these "predictions" in print or on TV far more weight than they should. It's very easy for a confident-sounding analyst, fund manager or professor to say something on TV that can throw off the best laid plans of investors.

I wish an SEC-mandated disclosure accompanied all pundit forecasts: "The undersigned states that he has no idea what's going to happen in the future, and hereby declares that this prediction is merely a wildly unsupported speculation."

Don't hold your breath waiting for that to happen."

Frank

http://prorevnews.blogspot.com/2011/05/krugman-rated-best-pundit.html

Some are much better than others at making predictions/projections in economics and politics. It all depends on the soundness of your understanding and the basic model you have constructed. Krugman is a stand-out. And he was most recently on the button with inflation.