I have been slowly reading Hillary Clinton's book, What Happened, and will probably finish it later today. The stuff that I want to talk about is in the first half of the book, so I'm ready to talk about it now. More or less at the same time I've been reading through the essays that are Sidney Award winners, David Brooks annual suggestions for interesting pieces to read. I'm done with Part I and still have a couple to go in Part 2. But I've read all the essays that are relevant for this post. And I've been reading other pieces as well, from the NY Times and the New Yorker for the most part, with a few that catch my eye which are recommendations from Pocket, and maybe something that a Facebook friend posts. I also want to specifically react to Paul Krugman's latest column, which I had some issues with.
Let me begin with a very high level and over simplified version of things, just to get us started. Many people in America are hurting, a big part of which is lack of economic opportunity, and that is compounded by adverse social consequences (opioid addition, for example) that are at least partly caused by the lack of economic opportunity. To remedy the situation, the rich need to contribute more in taxes to help provide economic opportunity for those who are hurting. I believe most Democrats can agree with this much. Getting more depth and detail is where a variety of issues arise.
1. Income redistribution is inherently ethical. Don't hide the ethical discussion. Bring it front and center.
Let me take on Krugman's piece here, which seems to offer a simple way to consider income redistribution without getting stuck in messy ethical questions. The problem is that the approach he used, based on marginal utility of income, is bogus in my view. When I was in grad school, we were taught that interpersonal comparisons of utility are not possible. For the moment, let's say otherwise. If you and I are allowed to have different utility functions and I'm richer than you, diminishing marginal utility in itself doesn't guarantee that my marginal utility of income is lower than yours. It could go the other way, in which case would we really want you to make an income transfer to me? Alternatively, if we all have the same utility function, then that would imply we should have equal incomes after the transfer. Do we really believe that?
Now it is true that the paper by Diamond and Saez that Krugman cites uses a social welfare function that does add up the individual concave utility functions. This is part and parcel of what the literature on optimal taxation does and Diamond and Saez are both eminent economists, so we shouldn't dismiss what they do lightly. I will get back to this below.
In my view of things, if people are hurting then you try to help them, but the motivation to help others is better conceived from other than an economic perspective. I like to talk about social conscience. Social conscience means that you help others who are in trouble. But, and this is where it differs from the optimal tax stuff, if you and I both agree that you are doing okay and my income is higher than yours, then I have no reason to give income to you. You should do fine on your own without the income transfer.
Now let's go a step or two backwards and consider the process by which the incomes are determined.
Philosophers may talk about designing a system and then appeal to something like Rawl's Veil of Ignorance to get at what is fair. If the process is fair, who then should quibble with how the income distribution turns out? Do note that the Rawls approach is maximin - it makes the worst off in society as well of as possible. If that worst off person is not in trouble, then social conscience would say leave well enough alone - no income transfers here. But if that worst off person is hurting, then there is case for income redistribution, even with a fair process.
Many conservatives believe in something more extreme, the just-world hypothesis, where the process is assumed fair (without testing whether that is true or not) and the outcomes are fine, no matter how badly the worst off person is doing. It's unclear how people who subscribe to the just-world hypothesis can have their world view changed. But let me offer this. Since I accept Daniel Kahneman's precept WYSIATI (What You See Is All There Is), it would be necessary not just to amass evidence of the unfairness of the process or that the worst off are hurting, but to make that evidence sufficiently visible that these people do see it. This seems like a tough sell to me, as it is far too easy for them to turn a blind eye.
Now let me get back to Diamond and Saez and optimal taxation. The methodology with a social welfare function that is maximized, subject to various incentive constraints, is useful for cases where the worst off definitely will be hurting, to elucidate other reasons for not equalizing income (to provide incentives for income generation). The methodology is something that economists are comfortable with. But the general public is not and I believe making the methodology evident to the general public is not helpful.
Let me illustrate. Occupy Wall Street gave us the language of the 1% at the top of the income distribution versus everybody else. In What Happened, Hillary Clinton talks about taxing the 1% (I don't believe she said by how much) and then use the proceeds to fund the various policy proposals (infrastructure investment, for example) that she discusses. But some policy suggestions she dismisses because "she couldn't make the numbers work." So, one wonders whether the rates on the 1% should go up even more, or if maybe it should be the 5% or the 10% who see their taxes raised, so other policy initiatives would become affordable. Indeed, if the redistribution is justified along ethical lines, then the right thing to do would be to identify the minimum income where households are not hurting and then raise taxes on all households about that minimum. Elsewhere I've suggested that boundary is about the 20%, where household income is roughly double median household income.
2. This is more than just ethics, however. This requires leadership from the politicians to get voters to accept the redistribution as a good thing, even if they pay more in taxes.
I believe that upscale Democrats have been co-opted by all the right wing anti-tax rhetoric, so one reason why Hillary Clinton talked about only taxing the 1% is that most of us voters aren't in that segment of the income distribution. It doesn't take much to support Robin Hood income redistribution if you're on the sidelines as neither a donor nor as a recipient. It takes more character to willingly participate if you're being asked to be a donor. This is the ethical argument I really wish Democrats would advance. Above that minimum threshold, the increase in taxes can follow standard progressive taxation principles, but now we're talking about a much greater number of households who will see their tax burden raised somewhat. There is clearly a political risk in advocating for this, as some of those households may not base their votes on social conscience, but instead focus solely on selfish considerations. In effect, what I'm arguing is that Democrats have to openly denounce such selfishness among the very comfortable, as well as among the uber rich. Real leadership would deliver that message in an effective manner.
3. There is a different issue with the optimal tax literature about defining the population that gets aggregated into the social welfare function.
Consider these two possible populations. The first is all households. The second is only those households who vote. If there were universal voting, these populations would be identical. But we are nowhere close to universal voting at present and household income is a predictor of who votes now. So instead of a Rawls solution, we get something like - maximize median household income. If you look at Democratic Party rhetoric, it champions the middle class, but says little about the poor. At present many in the middle class are hurting, which is why there is this urgency towards aggressive policies to help them. If and when the Democrats are able to produce consistent majorities, maybe then we'll be able to return to LBJ's War on Poverty. Now, perhaps it is possible to implement policies that help both the poor and the middle class. Beyond that, political feasibility rules out further income redistribution.
4. A different political reality is that the electorate seems to swing one way, then another. Even if the Democrats gained control of both houses of Congress and the White House too in 2020, how long will it last?
Obamacare passed without a single Republican vote, and the Republicans have worked hard to undo it since. Should it be anticipated that whatever the Democrats do when they have control will be undone when the Republicans are next in power? Or might there be a way to provide for some durability of policy even in the event of regime change?
I confess not to know the answer here. But the optimal tax literature doesn't worry about this sort of thing. It might be argued that less drastic changes in tax policy have a better chance to endure party regime change. At a minimum, I think this sort of thing should be argued out. Stability of a policy is a good in itself. The regime changes that induce cycles in policy are to be discouraged.
5. Finally, and I don't believe this is controversial at all, while it is fairly easy to find information about household income distribution online, for example Wikipedia provides data from 2014, and this site provides more recent and more granular data, I don't believe you can find similar information that includes after tax income as well. Note that these sites rely on data provided by the Census, while the after tax income data would have to be provided by the IRS. If we're going to talk about tax policy, we really should see what the current situation looks like.
Warren Buffet famously had a lower average tax rate than his secretary, who had orders of magnitude lower income. The marginal tax rates have progressivity to them, but long term capital gains are taxed at a lower rate than earned income and the fraction of income that is unearned rises with income. Also note that capital gains are taxed only when the asset is sold. If the household retains the asset but the value goes up, there is income generated but no tax at all. Rich people time the sale of their assets to minimize their tax burden. Likewise, rich people may have access to a wide variety of tax deductions that the rest of us can't claim.
So there is a policy discussion that needs to occur about these issues and the best way to raise taxes on the rich. This issue should be considered in light of the discussion in point (1). Any such discussion should be informed by knowledge of where we currently stand. There is a sense that the current system is unfair, but that belief is supported by anecdotal information only. Lets get better data so we understand the situation more clearly.
Conclusion
One reason I wrote this post is that a friend from high school posted a link to Krugman's piece in Facebook and because he is Nobel Prize winner, implicitly asserted that everything he says must be true. Usually I think Krugman is good on taking economic issues and making them understandable to non-economists. But equating marginal utilities in a social welfare function is not the way to get non-economists to think about income redistribution. Talk about the ethical imperatives directly. People will understand this on an ethical dimension. That brings in philosophy as well as economics.
Then, politics matters too. It matters quite a lot. So while an economic analysis a la Diamond and Saez can be helpful, it should not be taken as the last word. The issues need to looked at from several different perspectives.
I want to bring up one more point and then close. When I read about Hillary Clinton's proposals during the campaign, I had a sense that they were right in the type of programs she came up with, but they were insufficient in their intensity to address the issues at hand. This issue of intensity of programs is related to how much taxes are raised on the rich. Recently Krugman had a different column where he argued we shouldn't rule out deficit finance of some programs and I agree with that. Clinton never proposed deficit financing at all. But, truthfully, we don't know how intensive the programs need to be to get all the people who are hurting now to be functioning well again, and we're in this odd world where unemployment rates are low, but so is labor force participation, so I believe there needs to be learning by doing in program implementation to find out what works. That will require some patience, especially for those households who have seen their taxes go up. I hope that people don't conclude that a quick fix is out there, but I'm afraid the political rhetoric encourages that.
No comments:
Post a Comment