Marginal Income Tax Rates and Economic Growth, Ctd.

Building on yesterday’s post about Mike Kimel’s data crunching of top marginal rates vs. economic growth, today Kimel has created a handy bar chart of t+1 real GDP growth vs. top marginal income tax rate.

Personally, I’d like to see some numbers run on effective tax rates vs. GDP growth (not to mention some inclusion of state and local tax data), but Kimel’s use of the data shows pretty effectively that cutting the top marginal income tax rate does not, in fact, lead to improved economic growth–if there was a cause and effect relationship between low taxes and improved economic growth, there would definitely be a stronger correlation between the two–but there simply isn’t. This is contra the opinions of pretty much all mainstream economists, and defiitely contradicts one of the fundamental talking points of conservative politicians. It also, I might add, goes against what my intuitive expectations would lead me to believe. But the numbers are what they are, and as Robert Heinlein so eloquently put it:

What are the facts? Again and again and again — what are the facts? Shun wishful thinking, ignore divine revelation, forget what “the stars foretell,” avoid opinion, care not what the neighbors think, never mind the unguessable “verdict of history” — what are the facts, and to how many decimal places?

A useful bit of advice for more than just politics, but especially applicable here, I think.

FILED UNDER: Economics and Business, Quick Takes, Taxes, US Politics
Alex Knapp
About Alex Knapp
Alex Knapp is Associate Editor at Forbes for science and games. He was a longtime blogger elsewhere before joining the OTB team in June 2005 and contributed some 700 posts through January 2013. Follow him on Twitter @TheAlexKnapp.


  1. Andy says:

    Marginal tax rates are pretty much meaningless when it comes to what people actually pay. You’re right that someone should look at effective tax rates, but the political focus is always on the marginal rate.

  2. This would only be correct if marginal tax rates were exogenous. People have pointed out over and over that all the Presimetrics stuff relies on an incorrect assumption that the variables are exogenous and independent, when they’re not; they’ve simply been ignored, or accused of NOT WANTING TO SEE THE TRUTH because they are wearing IDEOLOGICAL BLINDERS!!!! Nor has he discovered some fantastic data series that economists are unaware of–believe it or not, he’s not the first guy who thought of looking at how growth performs when marginal tax rates change. The question is exceedingly complicated, and includes things how much of your budget is deficit financed. The consensus is that ceteris paribus, there’s a small effect, that probably becomes a large effect over time because of the miracle of compound growth–but that ceteris is rarely paribus.

  3. Alex Knapp says:


    I’m less interested in the causation numbers and much more interested in what appears to me to be a rather clear demonstration that the top marginal income tax rate and GDP growth don’t appear to have much correlation at all in the long term.

  4. Alex Knapp says:

    Er, causation argument, I mean.

  5. john personna says:

    It was a campaign issue just this last November, that cutting taxes created growth, and jobs.

    Here in California we had our governors’ debates. When asked how to make jobs, Meg Whitman said cut taxes (this in a state with a budget set to explode).

    So no, I don’t agree that this affect is not claimed to be large and “functional.”

    (FWIW, I also disliked Jerry Brown’s answer: “mumble, mumble, green jobs.”)

  6. Alex, I don’t understand why you are more interested in whether or not two gross macro statistics appear to move together on a graph. There’s a reason that social scientists do more than just eyeball excel charts.

    Moreover, you don’t appear to simply be claiming that there’s no gross correlation; you seem to be claiming that one can therefore draw inferences about (lack of) causation. This is simply false. Christy Romer did some interesting relatively recent work in which she attempts to control for endogeneity and finds some pretty strong and persistent effects.

  7. Davebo says:

    “Presimetrics”, “exogenous” ?

    Sorry Meghan, but using a rarely cited word such as exogenous or a word made up by an author such as Presimetrics doesn’t change the fact that you aren’t an economist or even a person well informed on economic theory.

    Seriously, give up the vocabulary training wheels and just try to make an argument.

    Preferably without the caps lock.

    Now an MBA may confer a certain level of competence, but I’m sure you will agree it’s really just an Endogeneity.

  8. Alex Knapp says:


    Alex, I don

  9. Alex Knapp says:

    Oh, great. WordPress ate my entire response.

    I don’t have time to write it all out again, so let me nutshell it:

    If, as pundits and politicians claim, lowering the top marginal income tax rate led to improved economic growth, then observing the data should show, on average, that this is true. (Economic performance is multivariate, so expecting 100% correlation is unreasonable–but expecting average improved performance isn’t.)

    We don’t see improved economic growth, on average.

    Conclusion: lowering the top marginal income tax rate does not appear to cause improved economic performance.

    Policy conclusion: Let’s try something else.

  10. Alex, not necessarily. If the changes in marginal tax rates are caused in part by the underlying performance of the economy, then the causation will be masked. Moreover, a lot of the argument by the more sophisticated economists relies on compounding, which means you’d need a long lag to detect the effect (I agree that GOP pols often make moronically simplistic arguments, but they are no more moronically wrong than the arguments Democratic pols make about stimulus; obviously, neither side has even a rudimentary grasp of the theory espoused by their respective economic teams). Then there are confounding factors. Then there’s the question of whether you should posit ricardian equivalence, and look at spending rather than tax rates.

    I think you *can* make the statement that obviously, there is no absolutely ginormous effect of the size posited by the more enthusiastic supply siders; that *would* show up. But that’s not the same thing as saying there’s no effect, and when you consider the compounding effects of even small changes in the growth rate, you have to take that seriously.

  11. john personna says:

    Megan, should we really get excited about effects we cannot see?

    That’s the bottom line. This is sold as something big and life-changing. Your defense is that it might be true is some subtle and invisible way.

    Ah … seriously?