Should taxation be regressive?

Tyler Cowen said,

The best kinds of redistributions are those that, say, educate poor children or fix up their health or end malnutrition. And those also boost wealth. So if you follow the lode star of boosting wealth, you’ll be led toward the better rather than the worst redistributions.

. . .I’m way less enthusiastic about just taking money from rich people and giving it to poor. If the incentive on both sides is to work less and create less then I generally don’t wanna do it. I would just say stick with wealth enhancing redistribution. There’s plenty of that we can do, it’s very powerful.

If I were to try to summarize his latest book, Stubborn Attachments, standing on one foot, it would be this:

1. The future will include many people not yet born, and we should care about their welfare. To a first approximation, we should care about future people as much as we care about people who are alive today.

2. The way that economic growth compounds, the welfare of people in the future depends mostly on the growth rate.

3. Therefore, economic growth is really important.

Let’s push on that a bit. Suppose that there is a regressive tax scheme that enhances economic growth. It’s not hard to imagine. Tax consumption, but don’t tax income or savings.

You could use some of those tax proceeds to redistribute income to the poor. But let’s push a little harder. Suppose that a poor person will spend marginal dollars on wasteful purchases, while a rich person will put marginal dollars into investment. In that case, lower taxes and less redistribution could promote growth.

So it is possible that regressive policies today could improve welfare tomorrow.

27 thoughts on “Should taxation be regressive?

  1. Say I want to cross the street. A rational calculation is that this isn’t very dangerous. I multiply low odds of being hit times my lifetime, and get a trivial expected loss.

    But now suppose I have a 0% discount rate and regard future people as just as valuable as the present. My odds of being hit crossing the street don’t change, but now I have to multiply that by the value of my own life, my descendents and their descendents, etc. Given enough generations, I can make the expected loss from the risk of crossing the street huge.

    And so I should never take any risks, because I endanger huge numbers of possible descendents. Yet, if I never leave the house, I won’t have any descendents…

    How do you come up with sensible risk calculations if you use a zero discount?

    • If you are unable or unwilling to produce (more) children, go ahead and cross the street! And even if you are able and willing, maybe your premature death will incentivize someone else to have more children, so future population will not be reduced. Or maybe you have had many of your sex cells frozen, and your production of children will not be reduced by your early death. Or maybe the human population is, or will be, too large, and your production of offspring was going to detract from, not add to, the general welfare.

      • And how do you estimate any of those factors? I’m serious in my question. A zero discount rate is like some SF novel where all the future hangs on every little action of the time traveler.

  2. What is the ROI on educating poor children versus educating rich children? Let’s just say we can use a test to identify any diamonds in the rough among the poor (like, uh, Stuyvesant did). Does it make sense in Tylers logic for NYC to spend $24k/year educating its students with the least potential? Certainly daycare as daycare would be less expensive, maybe even better at teaching the three Rs.

    I could come up with similar arguments on healthcare (why do we pay so much to keep druggies and the unemployable alive). Along with other expensive programs.

    It seems unlikely to me that leftist demagogues are on the verge of increasing taxes to 80% to fund a gigantic UBI. However, one can easily see them increasing taxes to soak to rich in order to “invest it” in healthcare, education, and the like. That these things will never have an ROI worth justifying it won’t be enough to stop it. Too much crime think to know why.

    I think the best argument you can come up with these things is “political stability”, but Tyler is also pro-immigration which inevitably means people with low ROI moving from polities where they aren’t entitled to these things to polities where they are. That’s hardly “stabilizing” or good for economic growth.

    Let’s leave aside the fact that the sectors above will have to be addressed at some point and just talk about funding them.

    Who can most afford to pay taxes and least likely to be able to dodge them because they earn trackable W2 income? Childless professional urbanites.

    Who isn’t doing their part to create “wealth” for the next generation? People with high IQs that aren’t replicating their high IQs.

    I’d increase taxes but offer huge % based deduction to income taxes for each kid someone has (in wedlock). Since it’s a % of income it would in some sense be “regressive”, but anyone that has kids would benefit. This probably wouldn’t be a huge incentive for people at the bottom to have too many kids (but would be a help for families, and help greatly with some of those high marginal tax incidence rates the working/middle class face and promote families), while it would provide major incentive for those at the top to replicate themselves (by far the greatest investment in the future they could ever make).

    P.S. One of his commentators notes the problems with a zero discount rate. Uncertainty about the future is a big part of it. Feels weird to argue against greater discounting when I feel there should be less myself, but the nature of the future impacts that discounting a lot.

    • Who isn’t doing their part to create “wealth” for the next generation? People with high IQs that aren’t replicating their high IQs.

      Is this Bell Curve or Idiocracy nightmare really happening? This felt more true in 2000s but the births in the US dropped by 400,000+ of which almost 300,000 is single motherhood. And the Hispanic-American and African-Americans are dropping at the most since 2008. (Teen pregrancy is at an all time low) And we still the slowdown in the Flynn effect but it is still happening in the US. (And it is believed that minorities IQ have jumped the last generation that even Charles Murray agreed to it.) And there are a lot of jobs where you a dedicated person with good intelligence but you are not looking for high IQs.

      Living in SoCal, population slowdown is hitting one area the most which is small business because they pay the least. (And I still say the variable hurting the Japanese economy the most is population slowdowns.)

      Having a lot and growing population:
      1) Increases your nation power
      2) Increases both long term AD-AS curves.

      • If the bottom has 2.0 kids and the top has 1.0 kids, what happens?

        I agree that fertility of low IQ first worlders not a big deal (given its close to 2.0, problem is with the top). Global IQ and fertility big deal given immigration.

        According to Flynn Effect our ancestors were a bunch of retards (in the context in which you are arguing for the Flynn Effect). You don’t believe that do you? Even Flynn says the Flynn Effect doesn’t mean what you think it means.

        • I assumed the biggest reason of the Flynn Effect during the 20th century was the developed world learned to improve the basics of people lives. So if you have basic education, nutrition, cleanliness, and healthcare you simply are improving the quality of young people ability to think and take test. So I don’t believe our ancestry were retards, but there was a lot the classic Monty Python joke line from the Holy Grail. (How do you know Arthur is King) “Because he was the only one not covered in shit.” Life was very harse, nasty, and brutish to 95% of the population and we forget how harsh it was before 1800s.

          Here are my simple thoughts:
          1) Africa birthrate is declining although slowly. I suspect this will continue with economic development.
          2) Without Africa the global population would be very near/at Replacement level fertility.
          3) In SoCal the level of border hoppers is relatively low (less than anytime I can remember since 1980) and it is because there are a lot less people coming from Mexico. The Central American Caravans can somewhat replace Mexicans but they can not reach the totals crossing of 2001 ever.

          In terms of birth it really seems contradictory that our society can afford so much but people feel like they are limited in their ability to reproduce. And I do believe:
          1) I think a lot of Cowen Great Stagnation is coming the limit of cheap labor makes small business harder to be profitable.
          2) Again without population increases, it is impossible to long term increase the AD-AS curves. People are input to both and the number have effects on both.

          • Nutrition is a thing, but it doesn’t explain the Flynn Effect on its own. The Flynn Effect has happened in OECD countries well past the point of universal nutrition.

            I will leave it to you to research the Flynn Effect if you care. We know that it doesn’t show up in ‘g’, it doesn’t show up in specific skills tests (hard mathematics etc) it doesn’t show up in marketable skills, and it doesn’t close any relevant “gaps”. It seems to be limited to a subset of specific questions on specific IQ tests, and not a more general sign of increasing “intelligence” in the sense of more people being able to engineer nuclear reactors and such. If it did then the 1SD increase in “Flynn IQ” since the 1950s would have yielded so many Einsteins we would be living in a Sci Fi novel, but we aren’t.

            1) Even optimistic UN projections for African fertility are ominous.

            2) Population is declining in high IQ countries. What does China look like with two more generations of this rock bottom fertility.

            More importantly, there are already billions of low IQ people in the world. If fertility in the developing world were 2.0 but they all immigrated here it would still be an apocalypse.

            3) There are already too many Latinos in America. This kind of immigration should have been stopped in the 1990s like was written in The Bell Curve (and probably before that, Reagen’s amnesty was the worst mistake in American history). As they come of age we will have one party politics, even without immigration. A majority of babies born in the USA today are non-white.

          • Judging by my answer on the Flynn Effect, I fine with my analysis that IQ have gone up several generations due a variety improvements with society factors. And those improvement are hitting diminishing returns for developed nations.

            And sorry, as Socal it is hard to convince most of residents there is too many Hispanic-Americans. My two best friends in college were Mexican-Americans and two of my teenage sons Mexican-Americans are their best friends. (One of the friends is mixed whose mother was an immigrant.) We have good neighbors here. etc. (Note I almost married an Asian Immigrant as well.) I believe the main reason the border states are the least concerned with hard Immigration is because the residents have great experiences with Hispanic-Americans.

          • You are entitled to your anecdotal, emotional, non data driven opinion, which is wrong…

            Such ends all political debates. Exactly where we started because nobody is punished for being wrong, which is the only way anyone ever learns and accepts anything (some of the time).

  3. 1. The future will include many people not yet born, and we should care about their welfare.

    My guess regressive tax gains would and give to the proceeds to poor would:
    Cut down the number of people not born. That is a lot more under $100K families than over $100 families and if you increase regressive taxes you are also increasing vastly in the middle. And having kids is consumption. (Yes you can increase land deregulation but like don’t California rich cities to drop anymore than 10%. Again, who will invest in land that has significant decreasing value and homeowners that might be satisfied with current house. People not moving is one reason for house shortage.)

  4. Sadly, despite the best advise of economists, humans tend to care a lot more about relative wealth and status than objective improvements in material standard of living.

    Few things cause more of a negative impact on economic growth than the wars, revolutions, and social unrest that result when too many people become too dissatisfied with their relative wealth and status. Especially if those too dissatisfied are middle class.

    It seems that most people have forgotten that the government sponsored social safety nets that were built in western democracies after WWII were not based on a theory of economics except to the extent they were based on these observations. They were based on a theory of history. History that was a lot more easily remembered in the wake of the Great Depression and the rise of totalitarianism.

    • Present tense: Social unrest exists in Germany now. There’s the rise of totalitarianism too. That’s the AfD. Merkel’s legacy.

      Humans care a lot about status. But what kind of status does any voter have when voting is irrelevant? Voters in Britain and France and Italy and Greece are dissatisfied with unaccountability per se. Taxation without representation: Doesn’t that describe the western democracies?

      I get the argument for government sponsored social safety nets, but an election is a safety net. Representative government is a safety net. Democracy and open debate. The marketplace of ideas. As opposed to open disdain, which is an insult to people’s sense of status in Germany, Britain, France, Italy, Greece.

  5. With socialistic schemes like Basic Income as an alternative, the biggest push on redistribution is to make sure all those willing to work have some kind of job.

    Job Guarantees so that all those getting gov’t cash are also doing something to “earn” it — even if the market value of their $24k guaranteed job is only $20k, or even if only $10k.

    And those with jobs will be generally best positioned to educate their poor kids, or take care of their health, or insure a good diet — exactly those “best kinds” of redistribution.

    Incentives matter. What civilization needs is more incentives for marginal folk to be marginally better / work more, be more responsible, be married more (less promiscuous), less drinking, less smoking, less toking, less video game addictions.

    Offering some job to all supports the growth of self-respect and individual responsibility. Freedom for all depends to a good extent on each person being somewhat, or very, responsible.

    • I agree, but get ready for a knock-down drag-out fight from the public sector unions.

  6. I have not watched the whole thing yet, but didn’t that recent Cato documentary on Sweden mention that their tax system is regressive? Maybe there’s something to it.

  7. Heck, there are other arguments for regressive taxes – not high taxes on the poor, but lower taxes on high incomes than on middle incomes.

    Consider the assumption of declining marginal utility, combined with the fact that the vast majority of value created by an individual in a market economy is usually captured by others (for example, the value captured by Bill Gates is a tiny fraction of the value captured by the users of Microsoftware). Socially, we want highly-productive members of a market economy to spend all their time working, redirecting none of it to personal consumption such as watching tv, socializing, vacationing, engaging in hobbies, or talking with family members. Personally, however, they have incentivizes to do all those things (they enjoy them), at great cost to the rest of us.

    The last thing we should do, then, is tax the value they capture from their market activities, creating a (to us) perverse incentive that encourages them to spend their time on their own personal, private interests rather than creating value for society.

  8. I’m about halfway through Stubborn Attachments. So far, I am mostly enjoying it.

    The one exception is his advocating for a near 0% discount rate with regard to future persons. He pretty much fails to engage with the basic arguments for discounting that I remember learning from graduate school in the late 80s/early 90s. I had at least three classes where we went over some aspects of the problem and, while my memories are degraded by time, I feel pretty confident each of those professors would devastate Tyler’s arguments (as presented) with just their default lecture material.

    This is particularly disappointing because, when you work through math, the discount rate is pretty much the whole ballgame in terms of making societal tradeoffs. Maybe that’s not quite right. Perhaps it’s better to say the discount rate determines which ball game you’re playing.

  9. Suppose a poor person spends the transfer payment on wasteful activity.

    Someone is producing that wasteful activity and earning an income from it.

    So you have one round of wasteful spending going into a second round of income for someone producing output that the market say is what is wanted.

    But what is wasteful? I suspect that whatever the poor person buys will generally be considered wasteful but when the person who earns that money spends it on the same thing it would not be considered wasteful. If a poor person buys a TV it is wasteful, but if a middle or upper income person buys the same TV it is not wasteful

    How do you get away from these value judgements.

  10. “Should taxation be regressive?”

    Stubborn Attachments was a bit disappointing in that it did not really address optimal taxation. I searched the book for “optimal taxation” and got 0 hits. There are two hits for “taxation.” The first is in reference to Nozick’s objections and the second is in a footnote mentioning an author who suggests some countries are able to have high welfare spending because they package it with growth-enhancing low taxation on capital income. Searching just on “taxes” gets just three hits, one that says they are “a complicating factor,” another to observe that voters prefer to shift taxes out to the future, and another that suggests redistribution is good because receivers are more likely to pay taxes. I get an additional hit just using “tax” and here he states that higher taxes may discourage entrepreneurship.

    Because Cowen says we should redistribute wealth only up to the point that it maximizes the rate of sustainable growth, I suspect he would answer theregressive taxation question by saying he wants taxes to be growth maximizing too. Cowen does go so far as to include a section addressing the question of whether money should be redistributed to the wealthy in situations in which doing so would maximize growth. But he then comes up with various reasons why redistribution to the wealth may not be desirable. I may have missed it in the book, perhaps in the footnotes or appendices, but I just don’t find anywheere where he makes any practical observations on methods of improving possible trade offs between tax losses and redistribution gains. No doubt he would offer that the growth maximizing trade off is the correct one. There are interminable economic models for such, but if there are any that are practical or relevant to actual policy praxis I’ve yet to encounter them.

    In the end, I think Stubborn Attachments is about normalizing the normative. Benevolent autocrats will hand down enlightened policies that we must accept as being for the greater good. The two Buchanan-inluenced books that were posted about here yesterday seem to cover the same ground but, give the impression of setting out far more positive statements rather than normative statements, and to be process oriented rather than utopian, affording authentic individualism. I suspect Buchanan’s positive micro-concerns and small rules are much more likely to lead us to a society that lets individuality, happiness, and autonomy flower to their maximum extent than macrolevel normative goal pronouncements.

  11. A flat rate tax on consumption isn’t regressive with respect to consumption but is likely to be regressive with respect to income. Did God ordain that income is the proper basis for taxation?

  12. That makes a huge difference if you expect the future economy to deliver real exponential growth over the long term. That’s a pretty dumb expectation. Exponential growth always flattens out and becomes sigmoidal. Assuming we didn’t overshoot stability, which we almost certainly did, in which case a negative correction is likely.

    Note that in this case, “negative correction” is probably a euphemism for something utterly horrible.

  13. “we should care about future people as much as we care about people who are alive today.” Does this make economic sense? Let’s apply this reasoning to people in 1918. Should we have lowered economic well being in 1918 to benefit people today? No. People today are much better off. We should assume growth will continue and people in 2118 will be much better off than we are.

Comments are closed.