### The Elusive Laffer Curve

Mark Thoma at Economist's View takes exception with the miraculous appearance of the Laffer curve on top of a chart of tax revenues and corporate tax rates for a set of industrialized countries:

Mark Thoma rightly describes Norway as an outlierâ€”a data point that doesn't fit the pattern displayed by the other data. The anomalous data point for Norway strongly suggests that another significant factor has entered the picture.

It's a mistake to draw the curve right through Norway; to do so is to say that the data from this one country is more important than the data from all the others represented on the chart. Why not draw the curve through Iceland or Luxembourg or Canada or France? Instead, the only data point the curve actually touches is Norway. It is a principle of data analysis that one should try to use the simplest curve possible to represent the data. A straight line, such as the one Thoma draws in a second version of the chart, is preferable to a curve.

Over at Maxspeak, You Listen!, Max Sawicky actually did the math, by running a linear regression analysis using similar data from another year:

The Wall Street Journal says Kevin Hassett has discovered the Laffer curve, but I think these data might say something else.The curve presented in this chart in a Wall Street Journal editorial purports to show that lower corporate tax rates generate higher tax revenues. Imposing the Laffer curve on the data isn't as delusional as pointing to an apparition of the Blessed Virgin in a grilled cheese sandwich, but it is intellectually dishonest. Here's why.

Mark Thoma rightly describes Norway as an outlierâ€”a data point that doesn't fit the pattern displayed by the other data. The anomalous data point for Norway strongly suggests that another significant factor has entered the picture.

It's a mistake to draw the curve right through Norway; to do so is to say that the data from this one country is more important than the data from all the others represented on the chart. Why not draw the curve through Iceland or Luxembourg or Canada or France? Instead, the only data point the curve actually touches is Norway. It is a principle of data analysis that one should try to use the simplest curve possible to represent the data. A straight line, such as the one Thoma draws in a second version of the chart, is preferable to a curve.

Over at Maxspeak, You Listen!, Max Sawicky actually did the math, by running a linear regression analysis using similar data from another year:

For fun I did a regression line for similar data (same year, same countries). I threw out those annoying communist outliers at the top (Luxembourg and Norway) -- countries with high tax rates and revenues. (By the way, when you include them without forcing a zero intercept, there is no significant relationship between the corporate tax rate (combined, central and sub-central govs) and corp rev/GDP. Very slight negative relationship. Throw out the two outliers and the same non-result (insignificant positive relationship).Mark Thoma sums up:

I know how much supply-siders want to find a Laffer curve, they've become frustrated going this long without success. But if they really think one exists they'll need to keep looking because they haven't found it yet.So the data in the chart don't yield a strong relationship of any kind, let alone the elusive Laffer surve. Like an upside down Chesire Cat, the data have once again disappeared, leaving the curve hanging there like an disappointed, lopsided frown. Maybe next time.

## 4 Comments:

With my very basic training in statistics - I can look at that data without doing the regression analysis and see that the curve is a joke.

Even without being taken over by Rupert Murdoch, the WSJ is off its rocker when it comes to supply side economics.

If Mr. Murdoch wants to talk about buying TommyWonk, I'd be willing to consider an offer. For the right price, I'm sure we could come to terms on editorial independence.

People respond to incentives, I don't need a curve to tell me any more.

So the elusive Laffer curve is superfluous as well...

Post a Comment

<< Home