I refute it thus

Although things have picked up recently, the recovery from the 2007-8 recession has been slow. Why?  One answer is offered by Edward Lazear, a well-known economist at the Stanford Graduate School of Business:  "Threats of higher taxes, the constantly increasing regulatory burden, the failure to pursue an aggressive trade policy that will open markets to U.S. exports, and the enormous increase in government spending all are growth impediments."   Obama's economic policies have been aimed at "promoting social agendas" rather than "creating conditions that are favorable to investment."   Of course, other equally distinguished economists, like Paul Krugman, say that this explanation is all wrong.  You might hope that people would have measured the sort of things Lazear talks about and analyzed their relation to growth rates over time, but by and large that hasn't happened.  For example, I don't know of any serious attempt to measure the extent of the "regulatory burden" or the "threat of higher taxes."*

However, it seems safe to say that there's a systematic difference between Democratic and Republican administrations.  Ever since the 1930s, Democrats have stood for more social spending and regulation of business, and Republicans for less.  So Lazear's explanation implies that growth will tend to be higher under Republican administrations--for example, George W. Bush's, which had the benefit of Lazear's advice as the chairman of the Council of Economic Advisers.  As I observed a couple of posts ago, that's not true--it's been much higher under Democratic administrations.  

But as I also observed, growth is affected by "luck," so maybe Republicans have had better policies but worse luck.  I took a closer look at that possibility by distinguishing three categories:  Democrats substantially better, Republicans substantially better, and both about the same.  I defined "about the same" as a difference of less than 0.1% a year.  Using the prior distribution discussed in that post, we get a 71% probability that the Democrats are substantially better, 23% chance that the two parties are about the same, and only a 6% chance that the Republicans are substantially better.    Those conclusions depend on the particular distribution, but if you start with a symmetrical distribution (that is, no initial bias in favor of one party) the chance that the Republicans are substantially better is invariably small.  

Blinder and Watson also look at Canada, France, Germany, and the UK.  Although specific policies differ, the general distinction between policies of the left and right applies in all of them.  They find that growth is higher under governments of the "left" in Canada, but conclude that probably reflects the influence of the US (Canadians happen to have had Liberal governments when the American has grown more rapidly).  They find no significant difference in the other nations.    Taking everything together, there is room for doubt about whether governments of the left are better for economic growth or whether left and right are about the same.  But it's pretty safe to say that governments of the right are not substantially better.   

*Actual marginal tax rates on top incomes are a lot lower today than they were 50-60 years ago,  but I understand Lazear as saying that proposals for higher tax rates deter investment, even if they are not eventually enacted.