Wednesday, March 11, 2009

The failure of macroeconomics?

I assume anyone reading this blog is heavily exposed to the rest of the econoblogosphere, and are thus aware that a lot of really big names have issued serious complaints about the state of modern macroeconomics, which is often described as Chicago, or Minnesota style. If one was to complete the trinity, Rochester would probably be the apt choice. One and a half semesters towards becoming a macroeconomist myself, basically all I've been exposed to is rational, optimizing agents, an economy driven by real shocks - government spending and productivity - so on and so forth. It wasn't until a couple of weeks ago that 'money' appeared.

Basically, what I'm saying is that you could get through a year of macro at Rochester - and then do metrics or micro - hence completing a PhD in economics without every seeing 'money' mentioned, with only a small change in the present curriculum. It appears Dani Rodrik thinks this would be a problem.

Now, I won't pretend to be conversant with the current macroeconomics literature, but let's take Rodrik's word:
The bad news is the world could really use some practical, relevant macroeconomic theory at the moment. Brad DeLong and Paul Krugman are doing a superb job of reminding us of the continued relevance of Keynesian thinking. But they are hampered by the absence of micro-founded models that plausibly deliver the Keynesian remedies they advocate.

I am, however, fairly confident with the theory of supply and demand. Also with the desire of many tenure track economists to receive the accolades of Krugman and others.

So, there appears to be a large demand for these micro-founded models (based on the actions of aggregated rational individuals), but very little in terms of supply, consistent with the belief that a large reward awaits those who can produce. And yet, we don't see these models being produced. I am under the impression that this is because when you try to craft these models, they just don't match the data at all, to paraphrase one of the professors here from the other day.

Note that in real business cycle models you can construct government spending multipliers greater than one, but the ones I've seen have this effect result from the government spending making people feel poorer and having them 'purchase' less leisure as a result. And I don't think that's how Keynesians percieve their stimulus.

Now, perhaps there is some insight out there that can rescue Keynesian models and make them approximate the data to a reasonable degree, or perhaps reconciliation simply isn't possible without seriously changing the definition of Keynesianism. Given the effort that's been exerted, let's assume for the purpose of argument that it's the latter. Can we still advocate Keynesian positions and call ourselves social scientists? Don't think so, but I can't see people giving up, either.

POSTSCRIPT: For the record, I continue to believe that the real side of the crisis is roughly approximated by a destruction of a large chunk of the capital shock. I can't really say anything about the monetary matters.

I also had an interview for a summer RA position the other day, which I failed to secure. Which is probably deserved, since the girl that did get it was running 95%+ in macro I, while nobody else in the class was threatening to break 80%. Still, being approached by a prof for the position makes one feel good.


Gabriel said...

Uh oh... don't go there.

What do you mean by empirical success? If you follow conventional criteria (2nd moments and IRFs) then Smets-Wouters (2003) model will kick the ass of any model with flexible prices. Then again, you can find all sort of things to complain about w.r.t. that model...

Please don't become one of those people who approach macro as a religious thing, with a hint of condescending towards those that don't share the particular modeling and theoretical preferences of those that taught one macro. Enough said, I think.

Many of the "Keynesians" out there err because they lack humility about what they really know and can argue in terms of how the economy works. Don't try to outdo them in unwarranted smugness.

Also, like it or not, there's a lot of interesting (and pretty much true) macro knowledge out there that won't fit into what Sargent does in his textbook or in dynamic general equilibrium models. Too bad. You can either pretend that those ideas don't exist and handicap yourself or you try to integrate them.

I'd rather see the current situation as a failure across the board, no one is above the problem.

Gabriel said...

P.S. In a baseline RBC model output can go down ONLY if either factor inputs (or utilization rates) go down or if TFP goes down.

Congratulations for being able to fit anything into this framework (when you say that capital output went down) but the goal is not to rationalize observations to make them fit your model but rather to have something insightful to add to our understanding of observables, right?

On the other hand, I think it's better to see this as a drop in TFP because, in an aggregate model, shocks to intermediate goods sectors (e.g. financial markets) show up as depressed TFP (cf. oil prices shocks). :-)

When I heard that there's a recession my first knee jerk reaction was to think: what shock of shocks had bad realization(s) so as to cause this, but then I had to ask myself what exactly I was trying to do and who I was trying to convince of what... :-(

Curtis said...

OK dumb undergrad alert:

How was there "destruction of a large chunk of the capital shock"(stock?)...

Capital was suddenly less productive one day? I suppose thats what a stock market crash implies...

Im confused

Andrew said...

Gabriel: I must've written poorly, because what I approached the post thinking was precisely that we shouldn't adopt a specific brand of macro as religion.

You're obviously much more versed on the macro literature than I am, but even I am discontent with pure RBC models. Measuring TFP is a dodgy exercise, and there's lots of simple adjustments one can make to eliminate a ton of the volatility in the residual, e.g. adjusting for the intensity of capital usage.

What I was simply trying to convey is that there seems to be a lack of models that people who support stimulus can use to argue their points, and lots of incentive to create such models. So it's strange that we don't see a lot more models supporting high multipliers and whatnot. I'm not basing this on my knowledge of the literature, just Rodrik's remark.

Curtis: I tend to view this as a capital shock because for the last decade, at least, large parts of the world have devoted the best minds to developing financial gimmickry that turned out to be much less valuable than expected.

For example, suppose we found out that we could grow trees bearing cell phones. This would be pretty excellent, so society provides incentives for lots of people to manage cell phone plantations.

Now, suppose a scientist comes along a couple of years later and concludes these cell phones spread cancer. How much are those investments in cell phone farms and cell phone farming techniques worth now?

Gabriel said...

Here are some models where there is a fiscal policy multiplier. The issue is, of course, what assumptions deliver what magnitude:

Since we will get to see what "fresh macro" people are doing because they are at our schools (well, more at yours than in other places :-(, ) I guess it pays to make an effort to look into what the Sith is doing. Just sayin'.

During the last 3 days or so I heard or read about 7 high-profile people in the profession saying that DSGE is useless crap and so on... I disagree, but I'd like to understand their claims in a charitable manner.

Tessa said...

I recently came accross your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.