« Back to the future | Main | Bodysnatchers »

Yaa Boo Lefties, the Laffer Curve exists!

Pharyngula: The stuff of legend
This is turning up all over the place — at Brad DeLong's, Crooked Timber, and this pair is from Cosmic Variance — it's the most sublimely, awesomely, wickedly stupid example of fudging a curve ever. The two graphs below have exactly the same data points, and the only difference is the curve that was 'fit' to the distribution. Which one looks plausible to you?


The one on the left looks sensible and simple, and looks like it was actually drawn with some consideration of the data. The one on the right … not so much. I have no idea how anyone could think that particular curve belongs in there.

Yeh, yeh, yeh ... the lefties are all excited because the WSJ drew a silly illustrative line on a graph to indicate the principle of the Laffer curve. But is their straight line a better fit?

If it was true then it would show that the higher the tax rate the higher the take, presumably up to and beyond 100% (and yes there have been over 100% tax rates). And that is crap.
Purely by eye I sliced the chart on bands of rates and estimated the mean point. If anyone wants to do it more accurately I would be pleased to see it. (Norway is a special case in that they are hoovering up North Sea Tax revenue, but even without it the curve is similar.)



There are two outliers, Norway and the UAE. All the rest are the sort of meaningless splodge of data that "social science" so commonly produces. You could draw a cicle around them for all it matters. Bah.

U.A.E. can't be an outlier, if you have a tax rate of 0.0%, then tax revenue will be 0%. So whatever curve you choose to use U.A.E. will be precisely on it.

This is ridiculously belated, but, hey, if I stumbled on this page, someone else might as well.

You know, there are better methods for interpolating curves than "draw a hilarious 'illustrative' curve", "draw a straight line" and "draw a yecchy squiggle". If you scan the comments at the Pharyngula post, you'll notice that I've provided interpolations; the quadratic ones seem to look the nicest.

The simply linear fit isn't good to extrapolate past the bounds of the provided data. But the pseudo-Laffer curve isn't even good for extrapolation within the plausible range. It's being used to argue for a position that the provided data stands against. And plus, it's really damned funny.

Post a comment