My Home About Me

On Friday I reproduced a chart from the Wall Street Journal editorial page, demonstrating the Laffer Curve with respect to corporate tax rates across various countries. The chart has viciously attacked by the leftist economics blogosphere, led by Mark Thoma and Brad DeLong -- and I've been attacked by DeLong and others for "defending" it with my one-sentence blog posting Friday. Here's the original chart.

The essence of the matter is whether the trendline in the chart is legitimate. No doubt a graphic artist at the Journal took some liberties in drawing it through the highest datapoint, Norway -- but that said, the Journal editorial never said the line was a formal trendline, anwyay. But nothing justifies Thoma's simply taking out Norway, licking his finger, holding it up to the wind, and drawing ad hoc his own trendline -- he admits "I haven't actually run the regression" -- that he imagines fits the data better (or at least more to his political liking), one that shows no evidence of a Laffer Curve. Here's Thoma's chart.

Let's go to the source of the Journal's chart -- Kevin Hassett, a scholar at the American Enterprise Institute. Here's a similar chart he produced for a recent print edition of National Review, showing the relationship between corporate tax rates and corporate tax receipts as a percentage of GDP. The data set here is the United States plus the European Union, excluding Norway, Luxemburg and a couple tiny economies. It's pretty clear that the higher the tax rate, the lower the tax revenue. That's what you get when you really run the data, rather than just drawing a made-up line as Thoma did.

In the National Review article, Hassett even quotes economist Kimberly Clausing, writing in a recent Brookings Institution paper (yes, Brookings, liberal Brookings), that "the United States is likely to the right of the revenue-maximizing point on the corporate income tax Laffer curve."

It's the height of irony (and hypocrisy) that Brad DeLong should be jumping into this debate, centering on the question of what happens if you remove a single country from the data -- in this case, Norway. DeLong's had a little personal experience with just that problem. DeLong and Lawrence Summers co-authored "Equipment Spending and Economic Growth" in the Quarterly Journal of Economics in 1991, and "Equipment Spending and Economic Growth: How Strong is the Nexus?" in the Brookings Papers on Economic Activity in 1992. The papers were presentations of an empirical study showing higher social returns to capital equipment investment than would be predicted by the classic Solow growth model. But only a year later, "Reassessing the Social Returns to Equipment Investment," an NBER Working Paper revealed that DeLong and Summers were just flat-out wrong, and for the most embarrassing of reasons: their results fell apart with the removal of a single country -- Botswana -- from their data set:

"DeLong and Summers' own data fail to reject the Solow model for the OECD countries. The same is true even for their full sample of quite heterogeneous nations if we exclude just one country (Botswana). These results clearly refute DeLong and Summers' claim to have uncovered robust evidence of uniformly high social returns to equipment investment."

Oh, and irony of ironies -- the NBER paper was written by none other than (wait for it...) by Kevin Hassett (along with Alan J. Auerbach and Stephen D. Oliner). Their paper was published in 1994 in the Quarterly Journal of Economics -- the same Harvard-sponsored journal that had published the original erroneous DeLong/Summers work. Has DeLong been published in a peer-reviewed journal since?

bailout   Basic Materials SPDR (XLB)   bond trading   bond yields   capitulation   climactic   commodities   credit crisis   Currency ETFs   day trading   Deron Wagner   DJIA chart   economic data   Emini's Nas 100   emini's S&P 500   emini's S&P 500e   ETFs (exchange trade funds)   ETFs (exchange traded funds)   Gold Miners ETF (GDX)   Japanese Yen ETF (FXY)