Dissing GDP

This article from The Economist is  music to my ears, as I have been disparaging the importance of GDP for a couple decades now. Time flies.

But I don’t think it goes far enough, mainly because it focuses on technical issues (like quality advance), rather than fundamental issues (like the difficulty of aggregating utilities across individuals).  At high estimated national incomes in particular, real GDP is not just flawed. It is meaningless.

I would add three points, in particular, to those raised in The Economist article.  First, from a markets perspective, I don’t know of a single question the answer to which would have much to do with GDP.  To cite just one recent, simple and glaring example of this point: the early recovery from recession during 2009-12 was very weak when measured by the GDP, but actually quite strong when measured by corporate GDP.


Simply by excluding from the GDP calculation the part of the economy that is not directly relevant to equity prices, the whole complexion of the early recovery is shifted – and in a direction that makes the tripling of the equity market less “ironic.”  This is not to say that corporate GDP is the answer. It is to say that standard GDP is not.

The equity tripling was not just about margin expansion or bubble dynamics.  The domestic corporate top line soared, although from a deep cyclical compression, and so too therefore did the market.  People who obsessed about record-weak GDP missed this, obviously, and then whined endlessly about it, as you may recall.  The S&P and GDP are totally different things!

Second, it is tempting to speculate on why Wall Street economists (admittedly decreasingly) obsess over GDP.  I don’t really know, but having spent just over a couple decades around Wall Street macro, I have a suspicion, which is only partly tongue in cheek.  The people who hire economists, luckily, do not typically come up through economics research. As a rule, they found something more useful to do for a career. So they don’t really know what economists do. But they figure that  asking economists to guess GDP seems like as good a task as any. Once the boss asks the economist to guess GDP, he can then measure if the economist succeeded. It all has the ring of objectivity, leaving aside that the exercise is largely useless.

To elaborate just a bit on the uselessness. Economic data matter, obviously.  The GDP calculation is one specific way of aggregating all the demand-side measures. Why would this specific way be optimal from the perspective of markets? That would be an amazing fluke.  It ain’t.

Finally, at high incomes, utilities are rivalrous, not only across individuals but arguably too across countries.  Not only do I want a bigger house, I want my house to be bigger than yours! That way I can lord it over you. And not only does America want a strong defense, it wants its defense to be stronger than China’s – or Gina’s, as Alec Baldwin would have Trump pronounce it.

This is an extremely simple and obvious point.  You would need  a PhD in economics not to understand it. And it basically disqualifies GDP as a measure of welfare.*

Some would retort, yes but GDP tends to correlate with what we care about.  First of all, challenge. GDP per capita has never been higher and yet people seem pretty pissed.  Moreover, if you know what the GDP correlates with that matters, then focus on THAT thing instead. For example, if you like GDP because you think it correlates with employment or life expectancy, then just look at employment and life expectancy. Take out the middle man.

But we can’t aim at employment and life expectancy, so trying to drive up GDP is good enough for government work, the success of which needs to be measured.

No it ain’t.

* Moreover, this is not some newfangled left-wing idea.  You can find it in Adam Smith, most compellingly in his Moral Sentiments.   If you have not read the first ten pages of Section IV, you might be quite surprised by what you find there. For example, his first reference to the “invisible hand” applied to economics is pretty hard on the vain motivations of those who seek to accumulate wealth. At the end, these motivations end up being all to the good because they drive progress. But the idea that people seek status, as opposed to some absolute measure of material well being, is central.

And now, event studies

Seeing as how we are in the business of dissing stuff this morning, let’s take the opportunity again to dis the logic of event studies applied to macro.  It failed on Brexit. It failed on Trump.  So how’s it doing on Italy?

Post the rejection of the Italian constitutional reform and Renzi’s resulting resignation, the US equity market is up about 0.5%.  This result was largely discounted ahead of time, but I am not sure to what extent precisely. Let me just make up a number: 85%.  So the odds of the result went up 15% on its actual realization.

Accordingly, on the logic of a macro event study, rejecting the constitutional reform was worth about + 3 1/3% on the S&P. Phew! Good thing that did not go through! It would have been awfully bearish had Italy turned out to be governable.

This is the same approach that allowed advocates of QE to “find” it had important and enduring effects on bond yields. As opposed to, you know, the state of the economy itself.