Benjamin Studebaker

There’s broad agreement among the political left in most developed states that we should raise taxes on high earners, if not now then after the economy recovers. The justifications vary somewhat, depending on how one comes at one’s leftism, but in most cases it can be boiled down to the principle of diminishing returns, which holds that the more money you have, the less utility additional money buys you. A homeless person almost always benefits more from a dollar than does a rich person, and if we are seeking to maximize welfare, it is reasonable to redistribute wealth from the rich person to the homeless person. Even some right wingers agree to this, in theory. The trouble is that there is much disagreement as to the extent to which we ought to redistribute empirically. That’s the question I’m going after today.

Answers to the question of the extent of redistribution range from “no redistribution at all” to “until there is perfectly outcome equality”. What accounts for the wide-ranging difference of views? There is disagreement about when redistribution becomes counterproductive. It is conceivable that, at a certain point, raising the tax rate on the wealthy might reduce their incentive to be productive, causing them to produce less and actually shrinking the amount of money available for redistribution. In this way, raising the top tax rate can theoretically reduce revenue. Here’s an example:

In this hypothetical example, raising the top tax rate from 35% increases the amount available for redistribution (as in Keynesopolis), but raising it too high brings about a collapse in incentives, reducing output. The question is at what point does that tax rate becomes too high? What is the optimal top rate of tax?

There have been many philosophical arguments made by right and left wingers arguing that it’s lower or higher than it presently is. Most of these arguments rely on gut feelings or feed personal prejudices and biases. Thankfully, quantitative research has been done on the issue. The research focuses on something called the Elasticity of Taxable Income (ETI), which measures the relationship between increases in the tax rate and decreases in the amount of taxable income people earn. A high ETI indicates that changes in the tax rate decrease incentive significantly; a low ETI indicates a smaller impact. Researchers look at cases in which tax rates have been moved up and down and measure the response in reported income, controlling for other extraneous factors. The work that has been done typically predicts an ETI for the rich of around 0.25. This means that for every 1% the tax rate rises, the rich report 0.25% less income.

An ETI in this neighborhood implies an optimal total top rate of income tax (local, state, federal, national, all together) of between 73 and 80%. This would be roughly twice what it presently is in the United States, and a significant increase in most other developed countries. The research tries to take into account under-reporting, tax avoidance, reclassification of income, and the various other ways in which the rich can earn the same amount while making it seem smaller for tax purposes. It’s possible that the research has not fully accounted for this, however, in which case the top rate might be higher still.

Of course, applying such high top rates of tax in the real world still carries with it some empirical difficulties. In countries in which emigration to a tax haven is convenient, the rich may not work substantively less hard, but they may flee the country, taking their potential taxable income with them. The shock of a large, immediate hike in the rate in a place from which emigration is not challenging could produce brain drain. A resolution to that difficulty requires either a reduction in labor mobility (which decreases economic efficiency) or agreement on the rate of tax among multiple countries via treaty–a kind of “tax trust”. Otherwise, tax havens can use low rates to poach rich workers and the revenues they produce.

One might also note that even as tax revenue is maximized by a high rate, the total output at any given moment may still be reduced, shrinking the pool of money available for investment. This could have a deleterious effect on growth and future incomes, at least in theory, which could eventually leave the poor worse off than they might be without the system of redistribution. This is a popular argument among the right, but there is an effective retort. Many of the investments the state makes into the future economic potential of the economy are significantly more profitable in the long run than private sector investments are. This is because the state is not constrained by the short-term profit motive that drives private enterprise. The state can lose money on an investment for years, decades even, with no significant repercussions. By contrast, the private sector, charities excluded, must demonstrate immediate profitability quarter to quarter, year to year.

In consequence, the state can invest revenues in long-term projects like public education, health, space exploration, scientific research, and so on. These kinds of programs have very high rewards for the economy over extended time frames; they are not especially suitable for short-term private sector investment, and when the private sector does engage in these areas, the effect is usually to maximize immediate profits rather than long-term societal utility. While the total amount of money available for investment is reduced by the redistribution system, the efficacy of the investment in promoting growth may nonetheless increase insofar as the state is better at investing in certain kinds of economic goods than the private sector is. Redistribution can consequently be seen as a sustainable policy tool–it may reduce total economic investment now, but it will increase the amount available for investment by both the private and public sectors in future because what investment we do have will be more efficiently distributed.

For an example of the principle in action, we need look no further than the 50’s and 60’s in the United States, in which top tax rates exceeded what our modern researchers believe to be the optimal top tax rates yet, despite this, economic growth was consistently much higher than it has been in recent decades. The research and investment committed by the American state during that period sparked the modern revolution in information technology, which still drives much of the economic and technological progress we see today. In contrast, the extra money made available for private investment by the lowering of the tax rate in recent decades has produced a series of bubbles (tech, stock market, housing, and so on). These bubbles consist of wasted, inefficient investment into the wrong sectors of the economy. They create recessions, instability, and large-scale human suffering.  The strength of the economy of the future will be in no small part dependent on our ability to realize we have committed a grave inefficiency and to enact, through international treaties if necessary, the necessary readjustments.