Economic forecasters should put their $ where their mouth is

By Razib Khan | January 3, 2012 1:24 pm

Happy Days Are Here Again! Don’t believe the naysayers: An economic recovery is right around the corner.:

Economic forecasting is a mug’s game. There are simply too many unknowable factors that affect “the economy” for anyone to make accurate predictions. The Fukushima earthquake and nuclear disaster, for instance, had a noticeably negative macroeconomic impact around the world, and nobody knows what lurks inside the hearts of central bankers. Plus, if I did possess the secrets to the future, I’d be making a fortune as a speculator, not telling you about it.

There are plenty of financial types who have funds where investment is contingent upon expectation of macroeconomic conditions. You know what they think because you know how they invest, and they’ll tell you what they think as well. What’s the point of journalists and academics even offering predictions if they don’t have “skin the game”? You can basically just say anything to be contrary, perhaps like in a publication such as Slate.

Here’s another anti-pessimism piece from Slate, No Pessimists Allowed! America’s economic recovery will be twice as big as experts predict. Date: January 2nd, 2010. With hindsight this was obviously unwarranted optimism. But how sure was the author, Dan Gross, about his prediction? It would be nice to get a sense if Gross had some put real money down on his predictions and made the values transparent. The amount of money would have given you a sense of the individual’s confidence in their proposition.

Of course, as Matt Yglesias says above why would he even publish his predictions instead of investing them? As Yglesias has said explicitly and implicitly he’s not a blogger for the money, he likes his job a great deal. He likes making predictions and performing rapid analyses. So that’s one reason why a Harvard grad with a degree in philosophy didn’t go to law school or into finance to make more money. But another issue here is that most journalists are probably not big enough individually to move the market. Therefore, their “small bets” would be a good gauge at least of how seriously to take any given prediction for their readers. Though obviously most readers aren’t interested in this stuff for truth. They just want to be entertained….

CATEGORIZED UNDER: Economics, Human Evolution
MORE ABOUT: Economics
  • Darkseid

    i find it funny how many from the Forbes/WSJ crowd acts smug and pretends to know everything about economics yet most can’t predict anything of any value, on the macro scale, to the average person. yet they still constantly make predictions all while claiming the economists aren’t supposed to predict anything but that they’re simply there to evaluate what’s already happened. what is it about majoring in economics that makes a person so terribly cocky? or is it self selecting? the arrogance and condescension wafts right off the page when i read, etc. It’s almost intolerable. NYT business section and a few others are toned down enough, though.
    anyway, interesting point about it being pure entertainment. i’d never thought of it quite like that. how could i have missed that idea after seeing Jim Cramer’s show??

  • Chris T

    Though obviously most readers aren’t interested in this stuff for truth. They just want to be entertained….

    It really should be drilled into everyone’s heads that writers are hired for their ability to draw readers first and foremost and not for knowledge or analytical ability. A writer who can spin a good story will win out over sound logic and statistics every time.

  • johnq

    Not sure what the point is here. As best I can tell, there’s no evidence that major investors are any better than random chance at predicting public markets, despite the fact that they’re risking significant money. Successful investors make money on inside information (either illegally or in private markets), or by extracting terms that other investors can’t get (Warren Buffett), or by charging other investors fees to manage money, or by privatizing gain and socializing risk (hedge funds, investment banks) or pure luck. No one consistently beats the public markets, at least for any significant period of time.

    There’s no reason to believe that any journalist has a better than random chance likelihood of predicting market moves, but I’m having trouble understanding why having them put their money where their mouth is would improve things. You’d certainly get a better sense of their level of confidence in the predictions, but what’s the use of knowing the level of confidence in predictions if the predictions are worthless? I suppose you could ask astrologers to put skin in the game to back their predictions, but would that make astrological predictions any more believable?

  • Laeeth Isharc

    A very fair comment, Razib.

    But one can actually take your observations one step further. Many financial types who appear on TV ostensibly to inform viewers regarding their outlook for the economy and for markets do not actually make their living that way (by getting the outlook correct), and if they do, are not necessarily investing in accordance with their stated view.

    Bill Gross for example, is a voice that understandably carries great weight with many professionals and quasi-professionals. But betting on the economy is not, from what I understand, how Pimco has actually made money over the years. They made money rather by exploiting structurally cheap sources of excess return, such as mispriced volatility (often too expensive versus realized outcomes), and credit spreads for certain bonds that were wider than justified by the true fundamentals. And after many conversations with people who as marketmakers take the other side of Pimco’s trades, what Pimco actually does is not always exactly consistent with their stated bigger picture outlook. It would be overly cynical, and hard to prove, to suggest that they use liquidity created by public expression of a view to create liquidity to maneouver in the opposite direction in their portfolios, so I shall certainly not suggest that.

    A big problem with the world of commentary is the short attention span – both in the failure to understand that economic fundamentals move relatively slowly (unfolding over months to years), in the context of which one economic number contains far more noise than signal; and in the failure to consider continuity. Everyone is asked what their views for the coming quarter are, but almomst nobody is asked well last time you were here you had these five ideas; two worked, three didn’t – why were you right, and why were you wrong. So the incentive is to say something eye-catching and clever-sounding (in a Tyler Cowenesque way), but not too far out there (lest you not be invited back).

    When it comes to macro (which is largely the topic you refer to), the markets are so big and liquid that even if you are running a billion dollars leveraged, the public expression of your views is really likely to have little impact on the market in the medium term (really, looking past a day). But, given the regulatory environment, most large firms don’t like you talking too much and portfolio managers are prohibited from speaking to the press. But it is hard for smaller firms to devote the time and resources to PR to getting invited on the shows. So as a result one hears from a certain group of people, mostly selected adversely with regards to what one really wants.

  • simplicio

    A lot of prediction pieces don’t tell you how sure they are, but they do tell you why they think what they think, and you can assess how sure that makes you. Trying to treat writers like oracles that just give outcomes is sort of silly, even if they were actually right most of the time. The point of the first article linked is that unemployment is down, the US fed is holding interest rates down, GDP is growing, etc. and thus growth will return.

    The point isn’t how convincing the person who wrote those arguments finds them, the point is how convincing the reader finds them.

  • John Emerson

    You hardly ever see someone cleaning up because they knew something no one else did. That’s because the key factors can’t be known, and the factors which can be known have already been taken into account in pricing. If someone figures something out and wins, others will copy him, cut into his advantage, and eventually adjust the market to the new state of knowledge.

    On the other hand, you have some gamblers who, knowing that they’re gambling, are more successful than others.

    What I just said is a version of the efficient markets hypothesis, and the EMH isn’t foolproof . But I think that it explains what we’re talking about pretty well.

    People who wrote and sold horserace tip sheets didn’t try for repeat sales unless their tips panned out. If someone won big on the first tip sheet, they’d probably buy several more from the same guy. But most didn’t win, so the tipster, like ponzi schemers, always needed new customers.

    A lot of finance scuttlebutt is on a par with the crudest superstitious beliefs of the pre-modern age.

  • Frank Howland

    Are you (Razib) saying that we should not take your predictions seriously unless you put some skin in the game and let us know who much you’ve invested? One can certainly make money betting which way things will go in science and politics.

  • Razib Khan

    There’s no reason to believe that any journalist has a better than random chance likelihood of predicting market moves,

    i’m not talking about markets in a narrow sense. i’m talking about the macroeconomy or politics or anything.

    Are you (Razib) saying that we should not take your predictions seriously unless you put some skin in the game and let us know who much you’ve invested?

    yeah. i try to entice people into bets, and am not generally successful. they look at me like i’m crazy. and when i’ve brought up trying to make a bet in the comments again there are no takers. so if you’re implying i’m not willing to “put up or or shut up,” i’ve put up. no market in that 😉

  • ziel

    It’s kind of funny that Yglesias puts forth as “evidence” that the economy will boom next year is this assertion: “Total bank credit, which collapsed during the crisis, is growing again and will keep growing.”

    This on the same day we here that “Bank of America Corp., under pressure to raise capital and cut risks, is severing lines of credit to some small-business owners who have used them to stay afloat.”

    If I believed what Yglesias believe I’d be all in the stock market. But he uses as an out that the recovery “will temporarily push inflation above normal levels and increase pressure on the Fed to tighten money and nip the boom in the bud.” So that would be his excuse for not betting – that he (and Obama) can’t control what the stupid Fed will do.

    He’s in a way probably right about that last bit – every time optimism starts to take hold, commodities spike hard, dampening the recovery. Wall Street had a good day today – but oil is now above $100, and that tends to keep a lid on things.

    The other thing he’s confused about is the employment picture – the drop in the unemployment rate is deceptive. There are indeed 1.7 million more people working than a year ago, but that hasn’t even kept pace with the increase in the working age population. The drop in the unemployment rate is primarily due to stagnant labor-force participation.

  • Darkseid

    out of the infinite internet betting market, i used to follow this one:
    it’s a little grittier than the politics betting site but it combines some political strategy in there. and i’m a hockey fan so i’d check
    for the up-to-the-minute playoff chances (this one’s for the NBA cuz I know no one likes hockey;)

  • Darkseid

    this was also relevant from today:
    Krugman was actually rated as one of the most accurate predictors in econ.

  • Nandalal Rasiah

    a Simon-Ehrlich type bet would be great fun to track. I think anyone of your 2012 predictions (scaled down of course) would work out. Those not involving genetics would probably draw wider interest but those that do would probably be more significant in the long term.

    In a way Yglesias has some skin in the game–he did buy a house (albeit in the anomalous DC market)

  • Konkvistador

    This reminds me:

    “Why do futurists make the same mistaken predictions over and over? The same reason politicians abandon campaign promises and switch principles as expediency demands. Predictions, like promises, are sold today and consumed today. They produce a few chewy bites of delicious optimism or delicious horror, and then they’re gone. If the tastiest prediction is allegedly about a time interval “3-5 years in the future” (for AI projects) or “6 months in the future” (for Iraq), then futurists will produce tasty predictions of that kind. They have no reason to change the formulation any more than Hershey has to change the composition of its chocolate bars. People won’t remember the prediction in 6 months or 3-5 years, any more than chocolate sits around in your stomach for a year and keeps you full.

    The futurists probably aren’t even doing it deliberately; they themselves have long since digested their own predictions. Can you remember what you had for breakfast on April 9th, 2006? I bet you can’t, and I bet you also can’t remember what you predicted for “one year from now”.”

  • Chris T

    Darkseid – Rated most accurate by who?

  • Darkseid
  • John Emerson

    If you think of all the social sciences as historical sciences with an unknowable future, the way evolution is a historical science, you’ll get a better grip on what they’re actually doing and why their predictions don’t work out. Also take into consideration that all social scientists are players in history, whether or not they want to be (and they usually do). Every prediction that’s not kept completely secret becomes one of the factors deciding the future.

    Can evolution be predicted? Does anyone try? Do we demand that biologists predict evolution? In very small ways, maybe, but usually it will be in the conditional form “unless this species evolves to adapt to this new environment, it will probably become exist”. What the new adaption will be, or whether it happens, is almost absolutely unpredictable.

    Social and human sciences claimed to be exactly predictive into the sixties and maybe the seventies, or at least they set that as a goal, but as time went on these claims and that goal came to be seen as unreasonable. But there still are holdouts, and advocates of all kinds will claim predictivity in order to gain the advantage.

    This isn’t to say that there’s no predictivity, either. But as the complexity of the phenomenon increases, the predictivity diminishes. Physiological psychologists in controlled conditions can predict human behavior remarkably well, but what they’re predicting is atomistic components of actual behavior in an uncontrolled context.

  • Jacob Roberson

    I’m generally on the Dem side but this is one thing they demagogue that I can’t stand: Of course I want to “privatize” SS, how is that a dirty word? Of course I want someone making money when my retirement makes money. Look what the mouths do when there’s no money there (Congress).

  • ziel

    Perhaps the most parsimonious explanation for Yglesias’s “prediction” is that this is the start of the 2012 election year and as a Journolist alum he’s just trying to do his part for the cause.

  • Chris T

    Darkseid – I actually took a look at the predictions (on the excel spreadsheet) and am not impressed. Most were trivial or vague, and some weren’t even predictions (ie: “We believe Hillary Rodham Clinton can and should be the nominee”).

  • ohwilleke

    “if I did possess the secrets to the future, I’d be making a fortune as a speculator, not telling you about it.”

    Proves too much. Lots of people don’t have enough money to make enough money to live on speculating on their own account, even if they are very, very good, and not everyone has the ability to secure other people’s money to invest on their behalf. Marketing and networking involve very different skill sets than predicting markets. Many of the most important finds in economics have been made by people who didn’t become particularly wealthy applying that knowledge.

  • gcochran

    “why their predictions don’t work out”

    Well, for one thing, they’re pinheads. Prediction is sometimes possible.

  • Mustapha Mond

    The fact remains that investing in equities via an index fund or surrogate portfolio with reinvested dividends still produces the best inflation adjusted return over any given working lifetime (40 years.) This can be particularly true of tax deferred accounts.

    The sharks may win or lose billions during that time frame but the average working stiff can still win with this long term lifetime strategy.

    Many of the best blue chip companies have direct stock purchase plans which automatically re-invest dividends at no charge and may be an even better way to put together a portfolio that will even beat the indexes as you are benefiting from compound interest on dividends.

    And many of these bluechips sell potato chips rather than computer chips.

  • Clark

    It’s interesting as I think something is only a science to the degree it can make reasonably accurate predictions. Of course there may always be some other phenomena interacting in some cases. But if you do worse than just guessing it’s a sign you don’t know much. Unfortunately I think economics has more of its share of that – which wouldn’t be so bad except so many economists are pretty dogmatic!

    John, wouldn’t you say that evolution makes predictions about the types of fossils we should expect? It’s quite reasonable to allow vague predictions if one allows that ones knowledge is vague. I think the element of chance in evolution leads to vague predictions but also allows a quite reasonable set of predictions which I’d argue are very well matched. No one is saying we have to know everything about the state of what we are analyzing. Not even physicists claim that typically.

  • Clark

    Ziel – I’m skeptical the economy will do better next year if only because I don’t think the place of Europe is clear. I think it reasonable to think the central bank there may screw things up. That said there are some interesting data suggesting improvement such as this post at Modeled Behavior. It’s just that there’s still a lot that’s negative as well.

  • John Emerson

    Predictive determinism was the model for science for a long time. Even in physics (quantum uncertainty and turbulence / chaos, etc.) it’s not generally possible any more, but I don’t think that the ghost has been exorcised.

    Prediction: lots of rough margin-of-error prediction is still possible. But during the great age people claimed exactness and certainty of prediction. They weren’t bragging about vague or rough prediction. A fair amount of vague or rough prediction is possible just by common sense, without any science. They claimed a big step beyond that.

    Predicting fossils isn’t predicting the future, it’s filling in middles.

  • Clark

    I don’t think I’m suggesting predictive determinism (since I don’t believe in it). But even within chaos (say a double pendulum such as we teach chaos to students with or as you say turbulence) there are predictions you can make. The predictions are limited but real. Which is why I called it vague knowledge.

    Certainly you are right that in the 19th century more was claimed and many problems were brushed under the rug. But by and large ones knowledge was determined by what you could predict about future measurements. Within reason – the problem of some degree of holism undermining theoretical predictions is a real one. You can always claim some other phenomena is at work – and of course that’s the problem in economics. Typically some other phenomena is at work.

    I disagree that predicting fossils isn’t predicting the future. If I say the range we should expect future discoveries to follow that’s a very real prediction. There are lots of gaps but the empirical discoveries filling in those gaps take place in the future and our prediction of what the gaps ought look like are a prediction about the future.

  • John Emerson

    Not in every case, but in most, with fossils we have a before and an after. Knowing that, we can have a pretty good idea what the intermediate forms might be.

    Actual prediction:

    Case 1: You have a fossil which seems to be the end of the line, with no successor / descendant. To what degree can you predict what its successor will be like, if we do find a successor which we have not discovered yet?

    Case 2: You have a fossil which has one successor / descendant. To what degree can you predict what other forms it took, if other forms did exist which we haven’t discovered yet?

    That would be predicting actual unknown futures, and I don’t think that it can be done. Predicting intermediate steps is a much easier case.

    In any case: rough prediction isn’t what scientists ever were talking about, because people without science can make predictions which aren’t that bad. What was claimed was some sort of scientific surplus, where predictions are much more accurate and reliable than commonsense predictions. And a lot of science attained that goal in certain areas, and it became the standard for all science. But in the social sciences, that goal was attained much less than was claimed. And generally, the less trivial the thing predicted, the more doubtful the prediction.

  • ziel

    Clark – I think that Modeled Behavior post is a load of hooey. He’s attempting to show that the 2nd derivative of employment growth is really really positive and we’re supposed to get excited about that. Take a look at the numbers at the FRED site yourself and note what employment growth was like in 1983 after a deep recession that drove the unemployment rate to over 11%, and then look at the pathetically anemic growth this year. Like I said, employment growth has barely kept up with population growth – that far outstrips any significance of his “% change in employment minus % change in population” statistic. He’s basically arguing that job growth is faster this recovery than it was in the last recovery – you know, the one that was known as the “Jobless Recovery.” He’s a liberal economist carrying water for the Obama administration.

  • Divalent

    I’m thinking more along the same line as Johnq (#3) above. In many respects, it’s like evaluating preditions of what number will come up on a roulette wheel based on whether the predictor actually bet on the number they predicted. It may be that someone knows it’s rigged to produce a specific outcome, but chances are everyone who placed a bet is just a gambler looking for entertainment.

  • Naveen

    Is there something hardwired into human nature that makes people over-indulge in making predictions (astrology, your next babies gender, economics…blah, blah)? I’ve no idea, but I just mean to point out that ‘having skin in the game’ would likely only distort/manipulate the predictions even further, rather than make people stop an think about what they’re actually saying!
    The ‘talking heads’ on e.g. Bloomberg TV often smack of some institution desperate to sell a potentially profitable prediction.

  • gnome

    Economics predictions are not exactly like the weather or similar, they can be causative too. How people are influenced by the media and what their economic experts tell them or what they predict will in turn change their behavior which would in turn influence the economy.

    As an example, in an economic slump, by simply generating broad optimism about the economy through positive predictions we could encourage more spending, hiring, offset deflationary pressures and in real terms actually influence things in that direction. We aren’t only predicting, we are influencing.

    Of course, there are a million other factors, I just thought I’d throw that in there.

  • david g swanger

    Late to the discussion, but Philip Tetlock (perhaps best known for his work on moral dumbfounding) wrote a book about this called “Expert Political Judgement”, for which he did studies of political predictions and their accuracy. As you might guess, the accuracy was very low.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar