The Corner

The one and only.

The Not-So-Curious Capitalist


In his blog at, Stephen Gandel seems to feel he has scored some points attacking Arthur Brooks’s WSJ piece. In fact, he makes himself appear to be a sloppy thinker. I corresponded with Gandel yesterday to discuss the well-established link between the size of government and economic growth. I was pretty stunned to see what came of our correspondence.

At his blog, Gandel writes

I e-mailed Brooks and the AEI to ask them about all this, and they said that there is plenty of economic literature to suggest that higher government spending leads to lower growth. He pointed me towards a study, which was published by AEI last year, that concluded that on average a 10 percentage point increase in government spending, which is what we have had over the past five decades, would lead to 1/2 to 1 percent lower growth rates. But even the study acknowledges this isn’t a rule. There are some countries that seem to do quite well with high levels of government spending.

In my correspondence with Gandel (I am simultaneously “the AEI” and “he,” apparently), I passed along a seminal paper by Robert Barro from 1991, along with a link to an AEI book that came out last year on the subject by Andreas Bergh and Magnus Henrekson. Andreas and Magnus also have a paper in The Journal of Economic Surveys on the same topic.  

In that paper, they echo what the book concludes when they write on page 14, “In general, research has come very close to a consensus that in rich countries there is a negative correlation between total government size and growth. It appears fair to say that an increase in total government size of 10 percentage points in tax revenue or expenditure share of GDP is on average associated with an annual lower growth rate of between 0.5 and 1 percentage point.”

The authors go on to consider why some countries, such as Sweden, have experienced high growth despite big governments and identify a number of mitigating factors. Quoting from their WSJ piece on the subject:

We also investigated the claim that Sweden is proof that big government does not harm the economy. While Sweden has done very well compared to other developed countries in the last 15 years, it has also implemented sweeping pro-market reforms. Examples include a national system of free school choice based on vouchers up through senior year of high school, a financially stable public pension system that can adjust payouts if contributions to the system fall for some reason, and comprehensive tax reform that has lowered marginal tax rates tremendously…

Sweden’s recent growth is thus the result of opting for free-market solutions instead of growing government. By comparison, the U.S. already has a relatively free economy, and therefore does not have as much potential for further market-based reform in order to offset the negative growth effects of a larger government.

This discussion does not, as Gandel implies, invalidate the scientific conclusion drawn from the authors’ literature review. It simply addresses an observation that might puzzle a reader about recent Swedish growth.   

It is common for tendentious journalists to misuse such examples, and this is a classic case. For example, if the academic literature found that economists tend not to get very many dates, and we observed that there is one economist who has lots of dates, we might be puzzled. The observation that the particular fellow is unusually handsome might then lead us to conclude that his physical attractiveness overwhelms the negative effect of his economics degree. His presence does not in any way undermine the conclusion that the marginal impact of a degree in economics is negative.

So let’s be clear. At the margin, the surge in the size of our government is a negative for U.S. growth. That conclusion is based on a broad consensus from an extensive literature.


Sign up for free NRO e-mails today:

Subscribe to National Review