Comment of the month

You may need to read back over the post, which is thoroughly worthwhile in itself (for eco geeks, or anyone with an interest in social science) but I lerved this comment.

There are no simple mistakes in applied macro, Nick! Unless one counts asking, on a public forum, provocative and far-reaching questions such as: “Is the entire field of applied Macro junk?” Hopefully the next faculty lunch isn’t too awkward.

In my opinion, the simplest refutation of applied macro is that hedge fund managers and other assorted punters have zero interest in the field. If macro models had even the slightest ability to forecast any better than a closed eye and a thumb held at arm’s length, these guys would be drooling over them.

Unfortunately, a lot of careers and prestige currently depend on macro modeling continuing to be taken seriously, and you have to spend a LOT of time studying the friggin things to be able to confidently say that they are crap. Sadly, honest eyebrow raising like this notwithstanding, I fear it may be some time before we can finally snap our MatLab CD’s in half. One funeral at a time, as they say.



So Zdeno, the Troppo Mercedes Sports is yours for this month and is being helicoptered to you as we speak.  Here at Troppo, we want you to know we appreciate your efforts. Enjoy.

This entry was posted in Economics and public policy, Philosophy. Bookmark the permalink.
Notify of
Newest Most Voted
Inline Feedbacks
View all comments
11 years ago

An alternative possibility is that hedge fund managers have to make more money than the next guy, so if things like macro policy are known by everyone (which I assume they are), then there is no advantage in using it.

James Farrell
James Farrell(@james-farrell)
11 years ago

I thought that the first comment on that thread, from Bob, was more to the point than Zdeno’s streetwise sarcasm. Structural models are primarily for policy analysis. If two or more people are discussing the likely effects of this or that policy change (or other exogenous event), and the the reasoning involves complex causal chains with feedbacks, it’s more efficient to do it with an explicit model in front of them. Otherwsie they will constantly find themselves talking at cross purposes. If they disagree about the consequences for variable Z of an increase in variable A, looking at the model together will help isolate the assumptions that separate them. They can do simulations for a range of values and see ehther it makes much difference to the result. In other words, models are cognitive devices.

But let’s consider the point about forecastoing anyway. When Rowe says

Instead of sticking a curvy ruler on the time-series for X, extrapolating out, and plugging the forecast for X(t+1) into the model, why not just stick a curvy ruler on the time-series for Y?

— and this is evidently the sentence that’s supposed to clinch his argument — he seems to be ignoring the fact that, when it comes to forecasting, structural models aren’t used mostly to extrapolate from current trends, but to calculate the effect of particular shocks and policy changes. You can’t even pose the problem in those terms with a VAR, which doesn’t distinguish between exogenouys and endogenous variables.

The question is not whether structuarl macro models aree good for ‘forecasting’, but whetehr there is any better tool for identoifying and quantifying the likely effects of a defined future change. And as for hedge fund managers, when did they become synonymous with best practice in forecasting?

11 years ago

What use is a macro model for a hedge fund? That is a serious question. If I run a hedge fund focused on the buyside of credit derivatives in Europe why would I need a complex macro model when I would have modelling done that is far more specific on the type of investment I am undertaking?

Labor Outsider
Labor Outsider
11 years ago

It is not a strong point because structural models were not developed for forecasting over the very short time frames on which hedge funds (often much less than 3 months) and the like trade on. A structural model gives you a framework for thinking about the economy and how the shocks that hit the economy may play out over a medium term horizon, or understanding what sort of shocks best explain the economy’s path over some horizon of interest to the practitioner. They were never meant to provide superior short-term forecasting to other tools. Developing a DSGE model for example would simply be a waste of resources for a hedge fund. Even central banks, which have much longer horizons, only use them as one of a suite of tools and models used for policy analysis and forecasting. One way to think of the issue is that a policymaker is often most interested in the fundamental value of an asset price (say an exchange rate) and then guaging how the current market price differs from that fundamental value. Structural models can help you understand fundamentals, but they don’t tell you when a reversion to the fundamental will take place. The fundamental price may be of limited interest to most traders unless they think there will be some reversion to that fundamental within their trading horizon.


[…] that there isn’t a better way of quantifying the effect of the stimulus. (This, after all, is what these models are for.) We can ask what would have happened if there had been no fiscal stimulus, a different kind of […]

11 years ago


Economic factors form components of financial models but a macro model of the DSGE (amongst others) will most likely not be used. Don’t forget that hedge funds are a catch-all phrase for any LLC/LLP that can trade in anything from investing in equities to equity derivatives to FX, buy-side RMBS etc etc. Credit derivatives can be horrendously complex and so specific to a part of an economy that a macro-model might not be useful but economic indicators do form part of model inputs and they do influence investment decisions.