Why does anyone support private macroeconomic forecasts?

This question has come up a few times lately in the blogosphere.  I've long found the market for such forecasts to be a puzzling practice.  Sometimes the forecasts are purchased and other times they are given away, but either way money is being spent. 

The first question is whether those forecasts are more accurate than naive "random walk" models.  On average probably not, although you could argue that in a recession mean-reversion gives an edge to structural models.  Still, the forecasts are paid for in both good times and bad.  Note also that those who predicted our last crisis were, for the most part, giving that knowledge away for free.

Second, there are many such forecasts.  Even if some forecasts are quite useful, what's the value of supporting a marginal or additional forecast?  Is the next forecast to come along so much better?  Forecasts would seem to be the classic example of a public good.

I would explore other models.  Under one possibility, outsiders pay for the forecast to join a more exclusive club of clients with other privileges.  It's a bit like how art galleries won't sell their best pictures to complete outsiders but instead ask that you "pay your dues" by being a loyal customer for years.  In other words, it's an arbitrary fee to enforce price discrimination, backed by some plausible pretext.

Under a related model, the firm pays for the forecast as a means of generating publicity, signaling its size, seriousness, and audience, and in general marketing itself to outside clients.  It is unclear who bears the final incidence of these expenditures, the firm or the clients, but still "forecasting isn't about knowledge," as Robin Hanson would have said to the oracle at Delphi.

Either way, I do not put much credence in what the forecasts say.  They do usually represent "standard macroeconomic knowledge" and in that sense they are not a complete fraud.  But the fact that they are being paid for does not, in my eyes, mean they are passing a market test in the traditional sense.

If anything, the persistence of the market in such forecasts should make you wonder about many of the other functions of these banks.


Ever heard these corporate truisms? "You never go wrong buying IBM." "A consultant is somebody you pay to tell you what time it is by looking at YOUR watch."

It depends on how concerned you are with the next quarters' forecast, vs. the next years' forecast. It's unlikely that the year-ahead forecast would be much better with one of these services.

However, a random walk model will not provide a good tracking estimate of GDP growth for the upcoming quarter. Macroeconomic Advisers model provides tracking estimates that are adjusted daily based on the economic data that comes out. For instance, when the imports data came out very weak for June, they sharply revised down their Q2 forecast and by the time the actual numbers come out, they are usually quite close. However, if you look back 2-3 months ago, they were much more bullish on Q2 GDP growth. I guess it just depends on what you think is important.

I think people forecasting the economy have similar incentives to those who blog. Bloggers don't get direct compensation usually but they get recognition or credibility which benefits them indirectly. From a firm's perspective they may get some "free" advertising

I agree with Gabe and Andrew. A lot of it is just plain CYA.

Similar considerations govern such things as hiring investment managers and advisors for pension funds and the like. You will often be better off just indexing, but try explaining that to a jury when the market crashes and you are the fiduciary who was trying to save a percent or so by not buying useless advice. Much better to be able to say, "We relied on the noted expert, "Mr. X."

We don't do well with randomness, and so look for explanations that seem plausible. Hence there is a market for explanations, that includes both horoscopes and much economic forecasting.

"Note also that those who predicted our last crisis were, for the most part, giving that knowledge away for free."

1. Who, exactly, "predicted" the timing, nature and trigger of our last crisis?

2. The folk who are currently riding the wave of the claim that they "predicted" the crisis (Roubini chief among them) did not exactly give away their knowledge for free so much as they continued their long history of calling for bears and got one right - which elevated their status. This is little different than any other doom-and-gloom peddler. Their motivations are hardly altruistic.

On average probably not

It sounds like you don't have a study or anything to back this up. Is this assertion a priori or is there more substance to it that you've withheld?



One can buy the output of a service that polls a number of economists at banks and the like. The consensus (mean) is pretty much what you get from a random walk with drift. However, what is more interesting is the standard deviation. This gives you a sense of what the 95% confidence interval is around that expectation, at least as far as known-knowns and known-unknowns are concerned. (Unknown-unknowns, aka surprises) are still possible. A fund manager can also look at subcomponents of GDP/GNP and see where the confidence interval is tight or loose. If the forecasts are tight for, let's say consumption, more research on one's own part is not likely to add much value. However, if forecasts are broad on Capex, it might be worth doing one's own research there. Furthermore, if one is an association economist, or the economist for a commercial or industrial firm, one can take the publicly-provided forecasts, and their variances, and use them as the basis for one's own forecasts of things of interest to one's firms, such as e.g., auto sales, or appliance sales, clothing sales, etc.

Cover your ass. That is rule number 1 for most people.

as an employee of a company that buys forecasts, I can tell you that one reason this industry exists is that there aren't many publicly available forecasts for markets outside the G-10. I agree without regarding US forecasts, but for other markets it really is a necessity unless you want to pay an entire staff of PhD's.

Very few people do pay for it, per se. Brokers give it away to generate commissions. Much of the analysis is interesting, and the forecasts are useful as a picture of the consensus.

A portfolio manager will probably keep his clients if he does what everyone else is doing and loses money. But if he goes out on a limb and loses money he is almost certain to lose his clients.

The risk-reward ratio for the manager -- portfolio or business -- is to do what the consensus is doing.
So the manager needs to know what the consensus is. that is what creates much of the demand for forecast.

In 1987 I started my own soft dollar equity strategy - economic service and in the summer I started saying a major bear market was ahead for the US market. I was correct-- although I did not expect it to happen in one day. But I still lost almost every client in early 1988 when renewal time rolled around.

Over some 20 years of running an independent equity strategy service based on current micro economic data -- not forecast -- I have lost more clients for being right than for being wrong.

Why would an economist ask why demand for economic forecasts results in a supply of forecasts using macroeconomic models?

Surely such an economist would rush to market with a different product to differentiate his product. How about a noodleconomics model forecast?


Nice one - I've had the same experience. Being right for reasons that are against the clients core beliefs loses clients. Most clients would rather lose money than be told they have foolish beliefs and then have that actively demonstrated to them.

Also, private forecasts worthwhile as a "reality check". When doing anything of sufficient complexity with the potential for large amounts of subjective interpretation, it is extremely useful to have a paid, objective yardstick of sufficient detail.

Most publicly available forecasts are not of sufficient detail to convince detail oriented experts. Paying for something compels you to pay attention to it. These two qualities create a useful objective reality check that you'll actually read and use.

Then tyler is missing the entire world of macro hedge funds. What are macro funds but high cost economic forecasts with logical investment positions attached to those forecasts? Why would someone pay 2 and 20 for private economic forecasts? Because the returns from even small informational advantages are huge.

I always enjoy watching academic economists talking about market behavior in sectors close to them. I've found that while they are very good at analyzing other people they become perplexed and confused when looking at themselves. This is perhaps why there is no widely distributed analyses of the appalling government subsidized oligopoly that College has become.

But to the point: Clients hire people they know and trust. Advisors know this and therefore engage in trust building activities, principally value adding conversations that are 'gifts' to the client that demonstrate commitment and expertise. The banks are always trying to find different topics that can jump start a substantive conversation. One topic the "economic weather" is a perennial "everybody talks about it but nobody does anything about it" it is the perfect jumping off point into a substantive discussion about whether a client's assets or strategic positioning are appropriate and what can be done about it. Coming to a client with someone else's analysis, no matter how similar it is will not enable the bank to get into the conversation and therefore will mean that they don't build trust and brand reputation for the inevitable day when assets move or deals or done.

Economists always forget that it's people who make decisions. Bankers do it because it works.

Wall Street economists tend to be skilled story tellers who craft stories that justify portfolio positions taken by the active fund managers who are key clients of their firms. The fund managers in turn need to be able to justify their positions to their own clients with compelling stories -- like the ones that the Wall Street economists tell so well. Compelling narratives tend to prompt fund manager action -- and commissions for the brokers -- far more reliably than random walk models or vector autoregressions. Trust me on this one.

The CYA motive mentioned in other comments is also highly relevant. The key relationship that makes money for most sell-side institutional brokers is the relationship between the institutional sales person and the active money manager. If the sales person independently makes a trade recommendation and it blows up on the client, he or she may lose the relationship and forfeit an attractive annuity stream for the firm. On the other hand, if the sales person makes a trade recommendation based on the view of one of his firm's economists or stock analysts, he or she has someone to blame if it goes wrong. Then both the sales person and the client can commiserate saying "we shouldn't have listened to that worthless economist/analyst." The valuable relationship with the client is preserved regardless of the forecast's accuracy. If the economist/analyst is wrong often enough to do damage to the entire franchise, that cog in the wheel is easily replaced.

In short, the value added of the economist is that there is someone to blame if the trade based on their view goes wrong. Since behavioral finance suggests that losses are roughly three times more emotionally costly than the psychic benefit from commensurate gains, the economist helps risk-averse clients and sales people cope with the emotional pain associated with making decisions under uncertainty. That is apparently worth something since the institution of broker-provided forecasts has survived so long. The same argument can be made to explain why active managers earn their fees. See Hersh Shefrin's "Beyond Fear and Greed" for an elaboration of this view.

Could you really keep the "premium" forecast secret?

"If experts are smarter than the market, why don't they just place their bets and get rich?"

Maybe because they happen to be very risk averse? Or maybe they just don't care about being rich?

All about sales. I once worked for a chief market strategist in the investment management dept at a Wall Street bank. We had a ton of salespeople who couldn't tell you the first thing about investment fundamentals. They would lure the clients in and the ones they could not hook right away they would set up on a call with the market strategist. A couple of guys who have been on CNBC can go a far way in convincing people to invest.
The market strategist would not offer any advice or really opinions. He could talk to a room of people for an hour telling them anecdotes and stories and each of them would think that they were gifted with profound knowledge that they wouldn't be able to put their finger on but would be much more confident in whatever notions they entered the room with.
Also we supplied financial advisors with marketing materials and stories. Giving the financial advisor all of the charts and stories(which are cheap to produce) makes her job 100 times easier and our stories were much more credible and intricate than anything they could come up with on their own. We would literally have a sales pitch for every type of asset class (large cap equity, small cap equity, market neutral, normal bond funds, long duration bond funds, high yield bond funds, hedge funds, real estate funds) so the advisor could pull out whatever story he thought his client would be inclined to hear. Our salespeople could then leverage access to the sales materials into sales. No sales from an advisor and we would cut them off from the good sales ideas around the latest economic/political stories.

BKarn beat me to it, but:

"Note also that those who predicted our last crisis were, for the most part, giving that knowledge away for free."

We're pretending this is real now? In other "news", the banks never needed the federal government's help, either. Really. Trust us.

Those who can, do and make a lot of money. Those who can't, make predictions for other people's money.


I'm a big fan.

Glad you read MR.

In my experience, forecasts of many types (beyond macro economy forecasts) are largely useful not for the specific forecasts, but for analyzing the downside. It's easy to live with the upside.

During the last crisis, the failure to consider the downside (e.g., housing prices would fall) led to complacency, which led to sorrow.

BTW, Michael Lewis's "The Big Short" told the story of selected individuals who saw the collapse coming and put their money on that possibility. It's a great tale of mass complacency and much rarer contrarian thinking.

To Paul, who posted the explanation of how financial advisors use charts and stories. please contact me via my blog www.underwhatconditions.com. I might be interested in your services for my firm.

Manchester Escorts, premium Escort Agency, Exclusive Company, based in North West England, providing professional outcall services.

Comments for this post are closed