Jan 10 2014

Do tax cuts promote growth? Part MCXIV

Posted by: Nicholas Shaxson in: Thoughts

Cross-posted from the Tax Justice Network.

The world’s economists have been arguing over this question for decades. Which is odd, because the evidence is rather clear on this point. Corporate tax cuts, or overall tax cuts (or hikes) just seem to have a neutral effect on economic growth.

Which is quite unsurprising, given that (as we never tire of reminding people) taxes are not a cost to an economy, but a transfer within it. Why should an internal transfer necessarily change overall growth. It might, but there’s no obvious reason at the outset why it should. But whole swathes of the learned economics profession start out with this ‘fact’, and then create studies to prove it. The evidence-tickling out there is a sight to behold.

Let’s start with a little graph we just made, which updates earlier work by others. This takes the prosperous western democracies, above a certain income level (so that we’re not comparing apples with oranges), and compares their per capita GDP growth with their absolute levels of tax in GDP. Here it is.

That trend line is, for all intents and purposes, as flat as a pancake. Ireland is an outlier, as is Luxembourg – but these are small tax havens with their own special dynamics so we should probably have excluded them. Anyway, the flatness of this graph is quite striking: despite truly massive differences in tax revenues per capita from below 30 percent to nearly 55 percent, they’ve nearly all grown at roughly the same speed.

This isn’t definitive proof of anything, but it’s a bulwark against the tax-cutters’ arguments. Perhaps an even bigger bulwark – though still far from definitive proof – is that economic growth during the ‘golden age of capitalism’ quarter century after the Second World War, a very high-tax era, was much higher around the world than in the tax-cutting era that followed. For example in the 1950s top marginal tax rates in the U.S. were 90 percent and real per capita GDP increased annually by 2.4%; in the 2000s that tax rate had fallen to 35 percent and real GDP per capita was less than one percent.

But there is plenty more. Not least our two recent blogs on the importance of the corporate income tax, Part 1 and Part 2 which provide a whole slew of insurmountable factors that simply have to be taken into account in any study of the effects of corporate tax-cutting – and which conveniently aren’t considered in any of those studies that purport to show that tax-cutting is the golden elixir of growth.

Now take this latest article from Citizens for Tax Justice in the U.S., veterans of this scene. (We have just quoted them on a rebuttal they did to an appalling campaign in the New York Times claiming — on the basis of a huge and unworldly mess of an economic model — that it’s a great idea to abolish the corporate income tax.) CTJ point to new reports from the nonpartisan U.S. Congressional Research Service (CRS), which “summarizes the evidence on the relationship between tax rates and economic growth” and finds “little relationship with either top marginal rates or average marginal rates on labor income.” It also finds that work effort and savings are “relatively insensitive to tax rates.”

Some of the reports they cite aren’t online, though this one is. CTJ pokes fun at of some of the tax-cutters’ ridiculous and unworldly models and notes:

“While many advocates of tax cuts claim that a high top marginal personal income tax rate hinders investment by the wealthy, the report finds that “periods of lower taxes are not associated with higher rates of economic growth or increases in investment.”

So how much did the Bush tax cuts help? They cite CRS:

“The models with responses most consistent with empirical evidence suggest a revenue feedback effect of about 1% for the 2001-2004 Bush tax cuts,” meaning the effects that the tax cuts had on the economy and on behavior of taxpayers offset just 1 percent of their total cost. And much of this effect may have taken the form of taxpayers changing how many deductions they take, and other tax planning changes, rather than actual economic growth.

The reports debunk (yet again) the laughable Laffer curve (pictured, at top), and also refutes the ill-advised ‘incidence’ argument which is wielded by the shriller tax-cut shills as in their campaign to abolish the corporate income tax. (Or, as one blogger summarises: “CRS teaches the American Enterprise Institute how to do panel regressions”.)

CTJ also makes another important point about bamboozlement:

These studies are often mind-numbingly complicated and it is rare that policymakers or their aides have the time and ability to go through these studies to understand whether or not they actually make sense. Thankfully, the Congressional Research Service has done that job for everyone.

In fact, there’s so much in the short CTJ article worth reading, that we would urge you simply to go there.

5 comments so far

La Chupacabra 1th January, 2014 11.23 pm

It appears that you are simply re-cycling an article you wrote in The Guardian a while back, the shortcomings of which were already comprehensively exposed in the comments section here (http://treasureislands.org/heard-that-countries-should-compete-on-tax-wrong/),

You continue to compare things that are not statistically comparable, i.e. average revenues (first order data series) versus growth in GDP/capita (a first order derivative). You select your sample with a distinctive bias against low-tax/high-growth countries like Singapore, Taiwan, Korea, Chile (maybe even Peru and Colombia) and of course China. You also should really look at (post-tax) real disposable income/capita instead of GDP/capita.

One final thought about this idea that the ‘golden age of capitalism’ was some time in the fifties: during that period, global GDP/capita grew by less than 0.4% per year, almost exclusively in the West and Japan, while 80% of the world’s population had barley enough food to survive. Since 1980, global GDP/capita has grown between 1.5% per year (the eighties) and 2.4% per year (the noughties). Everywhere (ok, maybe not in North Korea or Cuba) extreme poverty is declining rapidly.

Nicholas Shaxson 1th January, 2014 3.01 pm

Oh dear, dear dear. If go and take OECD countries. Remove those below a certain GDP per capita level (from memory, it was 20k per annum) otherwise you get those playing catch-up, muddying the data. And you will get the same sample as I chose. Granted, I should probably have removed Ireland and Lux, because they’re tax havens and the dynamics are completely different. But I left them in, and it didn’t help my case. So what exactly are you saying? I should cherry pick eduardo’s favourite economies, and make a new graph? You are free to do that on your own blog. It’s pretty much an update of this graph from Martin Wolf

Nicholas Shaxson 1st January, 2014 1.18 pm

I decided to check the exact numbers following the extraordinary assertions in your final para, that global gdp per capita growth was higher since the 1980s than in the postwar golden age. . . . and the results are in. Check the graphs out on p81

Or you could try those extreme leftists at the American Enterprise Institute, who note that world gdp per capita halved from 2.92% annually from 1950-73 to 1.41% in 1973-2001.

So I was wondering if you had identified some new and unnoticed regions of the world (pacific islands? Polar regions? Lunar bases?) where the results were so stupendously different that despite their small weighting they were able to flip the entire picture of the world as a whole?
Answers on the back of a postage stamp please . . .

La Chupacabra 1th January, 2014 4.48 pm

Nick –

Since you are unable to even define a tax haven except to say that it is any jurisdiction that ‘Shaxson disapproves of’, it is a good idea to leave Luxembourg and Ireland in this graph.

But your idea to leave out fast-growing economies out of your analysis is truly myopic:

It is a little like looking at the stock market since 1980, but ignoring Apple, Microsoft, Google and focusing instead on HP and IBM. Or ignoring Amazon and just looking at Barnes & Nobles.

You are selecting your sample to fit your conclusions.

You are right one on thing though: it is your prerogative to focus on the countries that bother reading what Martin Wolf has to write. Although I still cannot see anyone reading the FT in the US.

Nicholas Shaxson 1th January, 2014 5.00 pm

so comparing the growth of, I dunno, Peru and Germany, or Singapore and the United States, is a case of comparing apples with apples, is it? I just don’t see why my quite transparent and clear sample should somehow be twisted to incorporate some anonymous commenter’s favourite economies, which don’t fit? I don’t get it. And no, it isn’t anything remotely like leaving out tech companies from a stock market analysis.

Leave a comment