Thanks for the effort that went into this post. However, I thought there was a conspicuous lack of any discussion of Optimal Taxation Theory.
Quoting from Mankiw’s excellent review article, we can see why this part of economics is highly relevant to the issue: it is directly concerned with what type of tax system maximises utility:
The standard theory of optimal taxation posits that a tax system should be chosen to maximize a social welfare function subject to a set of constraints. The literature on optimal taxation typically treats the social planner as a utilitarian: that is, the social welfare function is based on the utilities of individuals in the society. … one would not go far wrong in thinking of the social planner as a classic “linear” utilitarian.
I’m not sure I could put it better than he does, so I hope you forgive the repeated quotations. One of the main findings of this field is that taxes on capital should be zero:
Perhaps the most prominent result from dynamic models of optimal taxation is that the taxation of capital income ought to be avoided. This result, controversial from its beginning in the mid-1980s, has been modified in some subtle ways and challenged directly in others, but its strong underlying logic has made it the benchmark.
Why? There are several reasons, and I encourage you to read the whole article, but the third justification he lists should be especially appealing to longtermist EAs: capital taxation reduces investment, which makes everyone poorer in the long run: even those who do not own any capital.
A third intuition for a zero capital tax comes from elaborations of the tax problem considered by Frank Ramsey (1928). In important papers, Chamley (1986) and Judd (1985) examine optimal capital taxation in this model. They find that, in the short run, a positive capital tax may be desirable because it is a tax on old capital and, therefore, is not distortionary. In the long run, however, a zero tax on capital is optimal. In the Ramsey model, at least some households are modeled as having an infinite planning horizon (for example, they may be dynasties whose generations are altruistically connected as in Barro, 1974). Those households determine how much to save based on their discounting of the future and the return to capital in the economy. In the long-run equilibrium, their saving decisions are perfectly elastic with respect to the after-tax rate of return. Thus, any tax on capital income will leave the after-tax return to capital unchanged but raise the pre-tax return to capital, reducing the size of the capital stock and aggregate output in the economy. This distortion is so large as to make any capital income taxation suboptimal compared with labor income taxation, even from the perspective of an individual with no savings. [emphasis added]
There has been a lot of work on the subject since then—for example here and here—but I think of Chamley-Judd as being a core result that the rest of the field is responding to. Some find that capital taxes should be positive or high, and some find that they should be negative—that we should subsidise investment—but the negative effects of capital taxes on investment, growth and aggregate welfare is clearly an important topic that can not be dispensed with without comment!
The above is concerned with capital taxation, but corporate taxes specifically are I think even worse. They essentially function as capital taxation, but typically allow interest expense to be deducted, hence distorting financing decisions away from equity and towards debt—contributing to systemic risk. (This problem was partly addressed in the US by the 2017 tax reform). To the extent that they only apply to legal corporations, and not other types of entity, they also distort organisational choice, which is also bad.
As a result, it seems that corporate taxes are harmful, and it would be better for the world (and the long term future) if they did not exist. Unfortunately they do exist—probably due to exactly the problems with institutional decision making that longtermist EAs are concerned about (e.g. short planning horizons, high discount rates, and capture by special interests). Fortunately, international tax competition provides something of a remedy, by encouraging countries to lower their corporate taxes to closer to the ideal level. Contra your suggestion that it ‘damages both “winners” and losers’, it acts as a beneficial check on the ability of countries to institute harmful policies. We should be supporting tax havens and praising their effects, not seeking to destroy them.
Despite having a section on ‘Objections’, the article does not really address this argument. You do sort of get at this issue here:
Developing Countries
Tax havens are necessary structures in encouraging investment in developing countries[35]. …
But the response misses the point:
Response: Agreed -- developing countries need to build both legal and tax system capacity. Development Financing Institutes and other investors require developing countries to honour and enforce contracts and to refrain from arbitrary seizure of assets.
Getting rid of tax havens degrades our ability to resist arbitrary seizure of assets. This is no small deal—many of the worst disasters in history have been intimately tied with governments’ seizures of assets and resultant damage to productive capacity. If we get rid of one check on this problem, we should have something else in place that can serve a similar job. The mere threat of losing access to financial markets for a while is insufficient. There are possible alternatives—once upon a time the west used gunboat diplomacy to this effect—but we should not remove our current solution without first instigating a new one.
Indeed, I think this article actually showcases the problem to a small degree. You write:
[tax havens] cost governments worldwide at least $500B/year in lost tax revenue
It is true that current investments, if subject to a higher level of taxation, would lead to higher tax revenues for governments (in the short run). But these investments were made by individuals and companies who were expecting to pay lower taxes! If taxes had been higher, fewer of these investments would have been made. To point out now that there is a lot of capital out there that could be taxed more if we changed the rules is precisely the sort of ex post asset seizure that people are worried about.
This section also sort of hints at the problem:
Growth
Tax havens promote economic growth in high-tax countries, especially those located near tax havens. US multinationals’ use of tax havens shifts tax revenue from foreign governments to the US by reducing the foreign tax credits they claim against US tax payable. As a result of the 1996 Puerto Rico tax haven phaseout mentioned above, employment by affected firms dropped not just in Puerto Rico, but in the US as a whole; affected firms reduced investment globally.[36]
But again the response misunderstands:
Response: If curbing tax havens reduces growth and taxes in developed countries for the benefit of developing countries, that is likely a trade-off many EAs would be willing to make (see below). Abbott Laboratories and other multinationals affected by the Puerto Rico phaseout may have reduced global investment, but increased investment and jobs in developing countries such as India. Given that US dollars go a lot further in less developed countries, a reduction in global investment by specific firms could also reflect better value for money.
The problem is not so much that getting rid of tax havens will reduce investment in the west specifically, but that this will result in a global increase in effective tax rates, and as such will reduce investment globally.
Mankiw’s review article on Optimal Taxation Theory is really great. This is exactly why EA needs more economists.
I find this intriguing:
it is directly concerned with what type of tax system maximises utility
Does “utility” as used by economists diverge in important ways from “utilitarianism” as used by moral philosophers? If the social planner optimizes for minimal negative utility, I assume that could change the optimal tax structure.
In the Mankiw article, the first justification of Capital Income Ought To Be Untaxed seems clear:
because capital equipment is an intermediate input to the production of future output… it should not be taxed
However, I’m not sure if capital income and capital equipment are interchangeable concepts. Perhaps they are.
The second justification seems a bit murkier:
Second, because a capital tax is effectively a tax on future consumption but not on current consumption, it violates the Atkinson and Stiglitz (1976) prescription for uniform taxation. In fact, a capital tax imposes an ever-increasing tax on consumption further in the future, so its violation of the principle of uniform commodity taxation is extreme.
I’m not clear how that logic would be different if applied to labour. Probably I need to read more about this to fully understand it. I assume the idea is I can reinvest capital income into more capital, but I can’t reinvest labour income into more labour. The article mentions “human capital”, so maybe I can reinvest my labour income into “human capital” such as education.
Finally, the third justification:
This distortion is so large as to make any capital income taxation suboptimal compared with labor income taxation, even from the perspective of an individual with no savings.
Has the premise:
In the Ramsey model, at least some households are modeled as having an infinite planning horizon (for example, they may be dynasties whose generations are altruistically connected as in Barro, 1974).
I’m not sure how realistic it is for households to have an infinite planning horizon. Moreover, it seems undesirable and dangerous to optimize for what could potentially be a totalitarian dynasty. Perhaps I’m misunderstanding the argument here.
The Mankiw article made me curious:
How easy is it to distinguish labour income from capital income? For example, a salaried plumber receives labour income, but a plumber who is paid as an independent contractor presumably receives capital income.
Is a laptop purchased by a business considered capital, but the same device purchased by a family considered consumption?
If tax havens’ low tax is beneficial, could the secrecy they provide to illicit financial flows be considered a harmful externality?
The article raises democratic legitimacy as an important constraint:
When Margaret Thatcher, during her time as the Prime Minister of the United Kingdom, successfully pushed through a lump-sum tax levied at the local level (a “community charge”) beginning in 1989, the tax was deeply unpopular.
Optimal Taxation Theory contains some really interesting ideas. Tax policy, in general, seems like a neglected area of EA.
Thanks for the effort that went into this post. However, I thought there was a conspicuous lack of any discussion of Optimal Taxation Theory.
Quoting from Mankiw’s excellent review article, we can see why this part of economics is highly relevant to the issue: it is directly concerned with what type of tax system maximises utility:
I’m not sure I could put it better than he does, so I hope you forgive the repeated quotations. One of the main findings of this field is that taxes on capital should be zero:
Why? There are several reasons, and I encourage you to read the whole article, but the third justification he lists should be especially appealing to longtermist EAs: capital taxation reduces investment, which makes everyone poorer in the long run: even those who do not own any capital.
There has been a lot of work on the subject since then—for example here and here—but I think of Chamley-Judd as being a core result that the rest of the field is responding to. Some find that capital taxes should be positive or high, and some find that they should be negative—that we should subsidise investment—but the negative effects of capital taxes on investment, growth and aggregate welfare is clearly an important topic that can not be dispensed with without comment!
The above is concerned with capital taxation, but corporate taxes specifically are I think even worse. They essentially function as capital taxation, but typically allow interest expense to be deducted, hence distorting financing decisions away from equity and towards debt—contributing to systemic risk. (This problem was partly addressed in the US by the 2017 tax reform). To the extent that they only apply to legal corporations, and not other types of entity, they also distort organisational choice, which is also bad.
As a result, it seems that corporate taxes are harmful, and it would be better for the world (and the long term future) if they did not exist. Unfortunately they do exist—probably due to exactly the problems with institutional decision making that longtermist EAs are concerned about (e.g. short planning horizons, high discount rates, and capture by special interests). Fortunately, international tax competition provides something of a remedy, by encouraging countries to lower their corporate taxes to closer to the ideal level. Contra your suggestion that it ‘damages both “winners” and losers’, it acts as a beneficial check on the ability of countries to institute harmful policies. We should be supporting tax havens and praising their effects, not seeking to destroy them.
Despite having a section on ‘Objections’, the article does not really address this argument. You do sort of get at this issue here:
But the response misses the point:
Getting rid of tax havens degrades our ability to resist arbitrary seizure of assets. This is no small deal—many of the worst disasters in history have been intimately tied with governments’ seizures of assets and resultant damage to productive capacity. If we get rid of one check on this problem, we should have something else in place that can serve a similar job. The mere threat of losing access to financial markets for a while is insufficient. There are possible alternatives—once upon a time the west used gunboat diplomacy to this effect—but we should not remove our current solution without first instigating a new one.
Indeed, I think this article actually showcases the problem to a small degree. You write:
It is true that current investments, if subject to a higher level of taxation, would lead to higher tax revenues for governments (in the short run). But these investments were made by individuals and companies who were expecting to pay lower taxes! If taxes had been higher, fewer of these investments would have been made. To point out now that there is a lot of capital out there that could be taxed more if we changed the rules is precisely the sort of ex post asset seizure that people are worried about.
This section also sort of hints at the problem:
But again the response misunderstands:
The problem is not so much that getting rid of tax havens will reduce investment in the west specifically, but that this will result in a global increase in effective tax rates, and as such will reduce investment globally.
Thank you!
Mankiw’s review article on Optimal Taxation Theory is really great. This is exactly why EA needs more economists.
I find this intriguing:
Does “utility” as used by economists diverge in important ways from “utilitarianism” as used by moral philosophers? If the social planner optimizes for minimal negative utility, I assume that could change the optimal tax structure.
In the Mankiw article, the first justification of Capital Income Ought To Be Untaxed seems clear:
However, I’m not sure if capital income and capital equipment are interchangeable concepts. Perhaps they are.
The second justification seems a bit murkier:
I’m not clear how that logic would be different if applied to labour. Probably I need to read more about this to fully understand it. I assume the idea is I can reinvest capital income into more capital, but I can’t reinvest labour income into more labour. The article mentions “human capital”, so maybe I can reinvest my labour income into “human capital” such as education.
Finally, the third justification:
Has the premise:
I’m not sure how realistic it is for households to have an infinite planning horizon. Moreover, it seems undesirable and dangerous to optimize for what could potentially be a totalitarian dynasty. Perhaps I’m misunderstanding the argument here.
The Mankiw article made me curious:
How easy is it to distinguish labour income from capital income? For example, a salaried plumber receives labour income, but a plumber who is paid as an independent contractor presumably receives capital income.
Is a laptop purchased by a business considered capital, but the same device purchased by a family considered consumption?
If tax havens’ low tax is beneficial, could the secrecy they provide to illicit financial flows be considered a harmful externality?
The article raises democratic legitimacy as an important constraint:
Optimal Taxation Theory contains some really interesting ideas. Tax policy, in general, seems like a neglected area of EA.