Sunday, January 5, 2020

Wealth and Taxes, Part III

(This is Part III of a series.  Part II and Part IV   See the overview for a summary. The punchline comes in Part V. )

So, why do we care about the distribution of wealth? -- Especially,  as we learned in part I that wealth is poorly defined and poorly measured, and we learned in part II that much of the distribution of "wealth" reflects higher market prices for the same assets, which do not increase their owner's ability to consume over a lifetime? Why so much anger, even from commenters on this blog?
  • Why wealth inequality not income inequality or consumption inequality?
There already has been much ballyhoo about income inequality. Why worry, separately, about wealth inequality? Why worry, especially, given that "wealth" is measured as income / discount rate, so it is income inequality? Well, not really -- it is only certain kinds of income inequality, and different kinds of income get multiplied by different large numbers (1/r). But why do we casre about this particular kind of weighted income inequality rather than broad income inequality?

Why do we worry about wealth inequality or income inequality rather than consumption inequality in the first place?  If you're worried about inequality of lifestyle, inequality of who is using the planet's resources, and so forth, you want to think about consumption inequality.

In the popular imagination these are all the same.  In fact and data they are vastly different. Income varies a lot from year to year, especially among the risk-taking wealthy. One year's income is a very distorted measure of lifestyle inequality. Taxes, transfers and savings buffer consumption from income. Most wealthy people leave their wealth invested in companies or give it away slowly through charitable foundations, so wealth does not translate to consumption, both in fact and in the theory I outlined last time.  The popular imagination just doesn't comprehend how immense a billion dollars is -- it's really impossible to "consume" any substantial fraction of this much wealth.  So, in fact, most super-wealthy people reinvest almost all their income in their  businesses or in new businesses just by default, and then give it away. Why is this a crime?

But consumption inequality is vastly less than income inequality, which is vastly less than wealth inequality. And I know of no evidence that consumption inequality is increasing.   So why worry about wealth inequality, or income inequality, above and beyond consumption inequality?

So, what is the question to which wealth inequality, as defined by Saez and Zucman, and wealth taxation, as advocated by Saez and Zucman and company is the answer? Saez and Zucman even write:
...carefully measuring its wealth is important. The public cares about the distribution of economic resources,
Well, the "distribution of economic resources" is consumption or lifestyle inequality, not wealth inequality.

 The question cannot be concern over lifestyle inequality.  If you care about lifestyle inequality it makes no difference at all whether a high-consumer gets that consumption from inherited wealth, self-made wealth, from annual income, or from transfers from family members. Wealth inequality, above and beyond consumption inequality, is meaningless. So what is it?

A cynic would say Saez and Zucman focus on income inequality first,  wealth inequality later, and make the measurement choices they do, because at every stage of the game these choices make the numbers look more unequal and divergent over time. We saw that last time with the choice to use the same absurdly low interest rate for all fixed income investments, even though the wealthy earn much higher returns -- and how this boosted their measured wealth by a factor of 4.

But in service of what are they putting an elephant's foot on the scale? Well, obviously a wealth tax. They're up front about that. And the larger wealth "inequality" can appear to be, the larger their argument for a wealth tax. But given that we are not addressing lifestyle inequality with the wealth tax, just what is the argument? So as usual in much political economics the answer comes first and the question keeps changing.

Spoiler: The answer is not that well hidden.  Saez and Zucman's answer is explicitly political. They want a wealth tax in order to take away what they perceive as excessive political power of wealthy people. Big wealth inequality numbers serve that cause. This has nothing to do in the end with economics. We ought to be more upfront and debate the real point.

We'll get to all that in parts IV and Vt. Let us return to asking why of each of the choices made by Saez and Zucman, of which Smith et al only quibble with measurement.  Why accept these rules of the game?
  • Why take the present value of “capital income” only 
But suppose a lawyer, superstar surgeon, or CEO earns $1,000,000 per year directly. Why multiply partnership income by 1/r to call it "wealth" but not multiply this income by 1/r and call it "wealth" too? Why not include human capital as wealth? Especially since many small businesses and partnerships are worthless without their owners or partners participation?

Saez and Zucman define wealth as
"the current market value of all the assets owned by households net of all their debts...Our wealth concept excludes human capital, which, contrary to non-human wealth, cannot be sold on markets."
Smith et al. echo this definition:
For aggregate wealth, we follow Saez and Zucman (2016) in defining wealth as total assets minus liabilities of individuals at market value, excluding durable goods, unfunded defined benefit pension plans and Social Security, non-profits, and human capital.
  • OK then, why not just “marketable" wealth?  
Well, there is perhaps a certain logic to that, especially given the eventual political agenda. Although people don't want to sell their assets, perhaps it's interesting to know how much they could get if they decided to do so.

But the "marketable" test vanishes quickly. Already they include funded pensions, and other claims backed by assets which people can't sell. And then they start capitalizing private business and partnership income, which looks economically just like labor income.

In addition, as we have seen time and again, the separation between "capital income" and "labor income" is slippery, and people are really good at reclassifying one as the other in response to changes in the tax code.
  • Why take the present value of pretax income?
All of these calculations take the present value of pre-tax income! If you earn $100,000 per year, you pay $50,000 in taxes each year, what possible sense does it make to say your "wealth" is $100,000/r not $50,000/r?
  • Why ignore transfers and taxes?  

Phill Gramm and John Early made the above graph in the Wall Street Journal contrasting the distribution of pre and post tax and transfer income.
"Government transfers provide 89% of all resources available to the bottom income quintile of households and more than half of the total resources available to the second quintile.
In all, leaving out taxes and most transfers overstates inequality by more than 300%, as measured by the ratio of the top quintile’s income to the bottom quintile’s. More than 80% of all taxes are paid by the top two quintiles, and more than 70% of all government transfer payments go to the bottom two quintiles."
Or, from Gerald Auten and David Splinter

Well, obviously, leaving taxes and transfers out makes the "problem" look a lot worse. And maybe we're moving on to wealth inequality because the rising income inequality narrative is falling apart.

But leaving out taxes and transfers is  even more insidious:

  • Leaving taxes and transfers out of distribution measurements creates a "problem" that cannot be solved, no matter how much taxing and transferring the government does.

Imagine that the next President declares a national emergency and raises taxes and transfers to the point that the "Net income" line in first graph is perfectly flat, and consumption is exactly equal for everyone in the US. Yet, suppose that nobody reacts to incentives, as most on the left seem to believe, and the pre-tax income line is the same. Then measured income inequality, and wealth inequality capitalizing such income remains exactly the same!

How much redistribution should we do? If we accept this question, the answer is always "more!"

The only way taxes and transfers can reduce a measure of pre-tax inequality is by, in fact, so lowering the incentive to work or save that it kills the top end of the pre-tax income distribution. I.e. by destroying the economy.
  • Why do we include unfunded government debt and not unfunded pensions as "wealth?" 
Saez and Zucman interestingly do not include the present value of unfunded defined-benefit pensions as "wealth." Most of these are government employee pensions. It will indeed be interesting to see if these are not in the end bailed out by the general taxpayer. That seems like a reasonable assumption, though again it boosts inequality by removing a lot of middle class wealth.

Likewise, their answer to not capitalizing social security, medicare, medicaid, etc. is that the government has made these promises with no idea how to pay for them. Though again that removes a source of lower-income wealth that's a lot more stable than partnerships!

But the US federal government has just as little idea how it will pay back Treasury bonds. They too are unfunded? If we do not include unfunded promises as wealth, why not these too? Well, the cynic says, that would make inequality go down since rich people hold treasuries and lower income people hold social security and pension promises.
  • What harm does invested wealth and reinvested income do?
If you are measuring and worrying about wealth inequality, not consumption inequality, you are by definition worried about invested and reinvested wealth. A typical Uber-wealthy person leaves his or her money invested either indirectly through stocks and bonds, or directly in private businesses and partnerships, and takes most of their capital "income" and reinvests it in those or new companies. That's why the distribution of "wealth" is so much more extreme than the distribution of consumption.

But just what harm are they doing? Really, the uber wealthy are providing the vast bulk of national saving and investment funds. If we start taxing wealth, and they respond by consuming it, giving it away, or giving it to political candidates, are we better off? Why is high income thrift a social problem?

There is no sensible economic question I can see for this. There is a political one. You read often from inequality critics the complaint that the wealthy "control" too much of the world's assets.   I have not seen Saez and Zucman use this word, and rightly so because it shouts out to a sort of conspiracy nut-job view of the world.
  • Why is wealth inequality a “problem” requiring “policy-maker” attention?
How can we care about a problem that is nearly impossible us to define or to measure, let alone person on the street to notice?

Homelessness is a problem, and you can see it just walking down the street in San Francisco. Unemployment, when it is large, is a problem, and you can see it readily. How is wealth inequality, so invisible that talented economists can't begin to coherently define it or measure it a problem? Why does that kid in Fresno care if venture capitalists fly in turboprops or jets?

As above, Saez and Zucman offer because
"the public cares about the distribution of economic resources,"
But the public cares about a lot of things, and not always for wise reasons. Good government ignores most things that large swaths of the public cares about. Envy used to be a deadly sin, not a guide for sober public policy.

Why is any "inequality" a "problem" requiring "policy-maker" attention? We should ask this question rather than just accept it. The minute you say inequality is a problem, you say that society is made better off by making someone a lot worse off and someone else a little worse off, so that we are more equal. No squirming, this is an inescapable conclusion, the only question left being how much.

I think many of us (me) worry about lack of opportunity, and the many barriers to advancement on the lower end of America's economic spectrum. I think society as a whole is better off if the bottom end rises. This worthy impulse is, I think, what many people mean when they say they worry about "inequality." But they recognize that the life of a poor kid from Fresno is completely untouched by whether a venture capitalist in Palo Alto upgrades from a turboprop to a private jet, and will be made no better off when confiscatory wealth taxation forces him or her to drive.  If you're worried about opportunity, mobility, left-out and left-behind people and areas, good for you -- but "inequality" is a poor word to describe this worry. Use better words that do not empower disastrous economic policies.

Even if you're worried about inequality, then why should you worry only about inequality within a country, and not across countries? Libertarians and left-wingers used to agree that the country that you are assigned to at birth should not matter to your rights as a human being. Global inequality has gone down tremendously in the last decades.
  • The answer 
Well, why to all these questions? The only sensible answer is that none of these sensible alternatives would look bad, or seem to justify a wealth tax. We are once again searching for questions that justify a given answer.

So why do we want a wealth tax? Join me in Part IV, wealth taxes


  1. "If you care about lifestyle inequality it makes no difference at all whether a high-consumer gets that consumption from inherited wealth, self-made wealth, from annual income, or from transfers from family members."

    Leisure is a form of consumption. Person A earning and spending 100k from a full time job and person B earning and spending 100k from investments do no have the same total consumption.

    1. Interestingly, evidence suggests income effects dominate at the high ends of the income bracket

    2. Yes, agreed. Wish there were a simple solution.

      A technical one is to tax complements to leisure, like skis, at a higher rate. As I have only skied once in my life, I like this principle. :-)

    3. And job satisfaction is a form of leisure.

      Down this path lies madness.

    4. @ Frank. We shouldn't stop there. Lots and lots of studies show attractiveness and height confer additional earnings advantages. Athletic ability enables thousands of sports figures to earn tremendous salaries. Even higher IQ's lead to individuals in high earning professions. Or perhaps the biggest one of all: individuals who have the fortune of having healthy genes are able to live longe, letting them live and enjoy life more than the rest of us.

      Since none of these attributes were worked hard for but rather conferred from birth(much like a big fat inheritance) - we should be taxing them too no?

    5. The highest satisfaction a human being can get is from being loved. Being this the case, a tax should be imposed on religious communities (obviously subject to a governmental seal of aproval) since it is in those communities where more love can be enjoyed by the individuals.

      The tax should also be progessive, so the higher the level of love to each other this particular religious community promotes, the higher the tax rate.

  2. " it's really impossible to "consume" any substantial fraction of this much wealth."

    Yachts, airplanes, and women.

    1. “I spent half my money on gambling, alcohol and wild women. The other half I wasted.”

      ― WC Fields

    2. It has been attributed to a highly paid athlete that he said in his later years that he spent 90% of his then income on booze and women. The rest was wasted, he said. :-)

    3. The "How Distribution Works" graph is brilliant!

  3. I propose a poker tournament. We seek 100,000 players each with a $10,000 buy in so the total pot is $1,000,000,000. The catch is that the winner gets $50,000,000 and the government takes the other $950,000,000 - because no one needs more than $50,000,000 and it makes no difference to the losers since they lose their money either way.

    Saez and Zucman believe that the confiscatory tax on the winner will not discourage anyone from buying in and playing. ¯\_(ツ)_/¯

    1. In fairness, they do not believe that at all. They just want other people to not do that math.

  4. Burned 18 mil in 2 months once. It's not hard.

  5. Considering the incentives, a wealth tax is preferable over a capital gains tax. A capital gains tax taxes those capitalists the most that employ their capital the most productively, which is not ideal. In contrast, a wealth tax does not punish capitalists for deploying their capital productively and should thus be preferred.

    I also commented on this here earlier:

    1. I don't understand your comment, but then I'm also not an economist. Nevertheless it seems to me that capital gains are only realized when you sell an asset. So how is selling an asset employing "their capital most productively?" Likewise a tax on wealth is a tax on investments that have not been liquidated, investments that are funding businesses and jobs. Your comments seem exactly backwards to me, but again this is far from my field of expertise. Maybe you can better enlighten me.

  6. This was an excellent sequence of posts. Really enjoyed reading and thinking through the argument. It changed my mind in a number of ways.

    One - not very important - point. You are correct to point out that simple measures of consumption inequality did not rise anywhere near as much as income inequality. However, this may be a data problem. There is fairly good evidence that consumption inequality in fact did mirror income inequality:

  7. A cynic might also say that the government is more likely to ultimately default on Treasuries than to default on Social Security since the default on the debt is less likely to lead to the riots on the streets.

  8. The more fundamental issue; A wealth tax is an assault on one’s personhood. Wealth issues from one’s talent, assumption of risk and assiduousness. As one has dominion, authority and jurisdiction of their person, that is to say, property in their person, that person has equal dominion, jurisdiction and authority over their wealth. If the law is violence then a wealth tax is the very threat of violence if one refuses to comply. A case for implied force can be made for adjudicating contracts. Not so for the arrogance of a government that threatens and intimidates when arbitrarily taking what rightfully belongs to an individual. If Warren, Sanders, Saez and Zucman believe it’s okay to confiscate wealth in the interest of the “greater good,” they do so at their own peril as we are not slaves.

  9. Nice post, but I disagree. Few points:

    a) it is not so clear that consumption inequality has not increased. See the Aguiar and Bils NBER paper.

    b) if you get definitional about wealth, then you should about consumption as well. Is running your own foundation part of consumption of the extra wealthy? As Piketty once said, the extra wealthy consume many things, such as politicians, economists ...

    c) I agree though that the main objective is political (and that is why I am in favor of a wealth tax even though I would prefer a higher estate tax instead).

    d) even if consumption inequality did not go up, and leaving definitional issues aside, consumption inequality is not all. Suppose you and I consume the same amount each year because, out of your benevolence, each year you take your income and give me half of it while I earn nothing instead. While I would be very thankful for this, I would feel completely at your mercy: if next year you wake up and decide not to give me anything, I would be stuck with C=0.

  10. >But they recognize that the life of a poor kid from Fresno is completely untouched by whether a venture capitalist in Palo Alto upgrades from a turboprop to a private jet

    What on earth could have possessed whoever built that private jet to spend their time building private jets rather than making sandwiches to feed the poor kid in Fresno? I guess we will never know.

    1. "What on earth could have possessed whoever built that private jet to spend their time building private jets rather than making sandwiches to feed the poor kid in Fresno? I guess we will never know."

      I know! The lack of a wealth tax! In countries with wealth taxes there is no aerospace industry and the free-sandwich manufacturing and distribution industry is booming. If only we had a wealth tax so we could take some of Bill Gates' loot and redistribute it to poor people in Africa.

  11. > Envy used to be a deadly sin, not a guide for sober public policy.

    Love this!


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.