One way to make some progress in reducing income inequality is to significantly increase redistributive transfers and public services. I’ve suggested that it will be difficult to fund that solely by heightening taxes on those at the top of the income distribution. Robert Waldmann asks, quite reasonably: Where’s the math?
Here’s an answer. I’ll use numbers for 2006, since that’s the most recent year for which we have good income and tax data from the Congressional Budget Office.
Suppose we need to increase tax revenues’ share of GDP by 5 percentage points. As the following chart shows, that would still leave us near the bottom among the world’s rich countries. But if the money were used well, it would be a notable advance.
GDP in 2006 was approximately $13 trillion; 5% of that is $0.65 trillion ($650 billion). President Obama has pledged to not increase taxes for the bottom 95% of Americans, so let’s presume the added revenue will come from the top 5%. In 2006 this group, 5.9 million households, had an average pretax income of $564,200. Their total pretax income was thus $3.3 trillion. The $0.65 trillion needed in order to boost tax revenues by 5% of GDP amounts to 20% of that $3.3 trillion in income. Thus, the effective tax rate (taxes paid as a share of pretax income) on the incomes of the top 5% of households would need to be increased by 20 percentage points.
The following chart shows the effective federal tax rate on the top 5% of households going back to 1960. The data from Piketty and Saez begin in 1960; the CBO data begin in 1979. I use the federal rate not only because data are available, but also because these taxes — mainly individual and corporate income — are the ones most likely to enhance the progressivity of the tax system (also included are payroll and excise taxes).
Incomes are higher in the top 1%, so what if we focused on that group? In 2006 the average pretax income among those 1.1 million households was $1,743,700. Their total income was thus $1.9 trillion. The effective federal tax rate on this group would have to be raised by 34 percentage points in order to increase tax revenues by $0.65 trillion, or 5% of GDP. Here’s what that would look like in historical context.
Whether desirable or not, increases of this magnitude strike me as unlikely. It’s worth thinking about additional potential sources of revenue.
Let me emphasize that my aim isn’t to discourage increases in taxation of the richest. I favor doing that. Rather, it’s to encourage the American left to think beyond heightened tax progressivity when considering strategies for inequality reduction.
Note: I’ve corrected an error in the earlier version of this post.
A fantastic system is a highly progressive consumption tax, as advocated by Cornell economist and New York Times Economic Scene columnist Robert Frank. For details see a brief Washington Post Column on this by Professor Frank.
Such a plan may not be politically feasible now, but it can be in the future if first the academic community becomes educated about it and strongly in favor of it; this can lead to opinion leaders, including pundits, and politicians favoring it, and then the public. This process has happened many times before.
Also, the exact plan may not be politically feasible now, but some aspects of it to some extent may, and these may be very valuable.
Professor Frank is famous for his tireless advocacy of considering positional/context/prestige externalities, the pink elephant of economics. Smart consideration of these monumental externalities could massively increase total societal utility. You will read more about this in the above linked article, but I strongly believe that every economist and sociologist should read his book, “Luxury Fever” (1999), or at least his very brief book, “Falling Behind” (2007).
Another conclusion from this calculation might be that inequality reduction in the US is impracticable without reducing military expenditures. At least, it is impracticable with the strategy you are recommending.
I notice that the total household income in 2006 calculated by adding up total pre-tax household income from all income quintiles is only some $10.6 trillion. Where is the other $4 trillion, and how (if at all) is this presently taxed?
Richard: I agree; I like Robert Frank’s take on this.
Gordon: The rest of GDP — around 2.7 trillion in 2006 — is investment, depreciation, and a few other things.
So the chart shows the increase would approach what we had in the middle class growth years of 50’s and 60’s. Ok, no political will to do it.
But, what seems to be forgotten in this distribution issue is also back then we had more of the income going to more of the people via pay for their labor. That is what made the difference.
You could not raise the rate as high as these charts suggest, if we return to policies that made it more profitable for companies to pay their help. Even unionizing fits with this tax issue. Tax a broader distribution of income means a lower overall rate for everyone AND, more money for the state to invest in all those things that business likes and needs to make money.
Time to broaden the discussion. (as I keep noting at Angry Bear).
Also, I have pointed out, that the amount of money needed to be made up in income to the 99% of the population to equal the share of 1976 is $1.4 trillion dollars. You just can’t do that with taxes. You have to do it by paying the help more for their labor.
DolB: Just to be clear, the charts here suggest we’d need top-end effective tax rates far higher than we had in the 1960s in order to increase tax revenues by 5% of GDP.
Lane,
I think the most relevant line out of the article was, “Whether desirable or not, increases of this magnitude strike me as unlikely.”
There are so many (possible) solutions out there, but the problem is that they all have their flaws.