Wealth Taxes

by
Robert Waldmann
My articles My siteAbout meMy booksMy videos
Follow on:Twitter

There is an interesting discussion among smart, expert, thorough economists about wealth taxation. It is clearly stimulated by Warren’s proposal to tax wealth. Gabriel Zucman, Roger Farmer, and the much less famous but also super smart Noah Smith are debating the issues. I’m sure AngryBear readers can benefit from their discussion (to which I don’t link cause I just saw one tweet).

I am also sure that it will be a waste of time to click “more” and read my thoughts on the topic. Caveat lector.

I am not going to do any research. So I will have fresh thoughts untainted by data analysis or familiarity with the theoretical discussion.

1. How is introducing a wealth tax different from an increase in capital gains and estate taxes which produces revenue of the same present value ?

One issue is that the Roberts Court might declare it to be unconstitutional. I won’t discuss that.

Another is that to assess a wealth tax one has to measure wealth. Moving wealth to assets whose value is not easily assessed is one way to avoid a wealth tax. One class of such assets is art — thin market, amazing valuations, hard to assess. I am going to assume that this is not an insolvable *new* problem. Hidden ownership and non fair market valuations are also used to hide income and capital gains. My guess is that the problems are similar. I am very not expert on this, and won’t go on.

Another issue is that the tax starts immediately. Note the key weasel phrase “present value”. If one assumes that the government has an intertemporal budget constraint and maximizes some assessment of social wealfare (or national goals or whatever) subject to that constraint, then the present value of tax receipts is the only thing that matters. But this is not the way politics actually works. If governments did that, then they would issue bonds and buy risky assets making huge risky returns on the carry trade in exchange for automatically smoothing business cycles.

There is no rational reason why only a few states do this (see Norway, Singapore, and I guess some petrostates). When we discuss policy, we must consider the policy makers we have not the policy makers we want. Without understanding the nonsensical choice, we can’t figure out how reforms affect future policy. I think that collecting the tax now (just pulling the revenue forward) will affect meaningless numbers such as estimated deficits and national debt. These numbers are not economically relevant (the budget constraint also depends on future tax revenues and unfunded pension liabilities, and non financial assets owned by the public sector, oh an financial assets owned by the public sector whose value is assessed without assuming that the public sector has the same risk bearing capaicty as private investors).

A model of a rational policy maker is of very little real world relevance. Sorry I mean has negligible real world relevance.

Consider the cases of Fannie Mae and Freddie Mac and the bailout. This was by far the most profitable trade in human history. The gain of hundreds of billions (in present value) was estimated as a cost of hundreds of billions. The reason was that the standard calculation of the estimated effect on the expected national debt after 10 years showed a profit. And that can’t be, because public ownership of Fannie and Freddie is socialism which can’t be profitable and efficient. So the books were cooked (in the opposite of the usual way) and publicly owned assets were valued at market price not at the expected flow of revenues discounted at the Treasury rate.

Barney Frank complained that this was unfair to him (it was). He was ignored. The bottom line is now clear — in any case it is clear that it is to be written with black not red ink. Nonetheless it is argued that socialist Fannie and Freddie are no good and they must be privatized. Ideology has defeated arithmetic. Economists have to face this fact when they discuss policy.

2. Realized capital gains, restarting on inheritance and stuff. Under the current system capital gains aren’t taxed at all if they are not realized before the owner’s death. The assets pass to heirs and future capital gains are assessed compared to the value on the date of inheritance. This is an important feature of the current tax code. In effect it makes capital gains taxes avoidable in a way which a wealth tax wouldn’t be.

I will consider the effects of the current tax code. I think it really, in present value, penalizes the realization of capital gains. This reduces trading volume and makes assets less liquid. I think this is a good thing. I think that a financial transactions tax would be good policy and the incentive to leave capital gains unrealized until death has a very vaguely similar effect. It makes people think of the very long term. It is a distortion. If one assumes that market outcomes are Pareto efficient and maximize money metric welfare, distortions are dead weight losses. If one doesn’t make those absurd clearly false assumptions, then one has to consider each distortion with an open mind. I think anything which reduces trading is probably good.

3. Foundations. The really rich don’t just avoid capital gains taxes by holding assets till they die. They also avoid estate taxes by giving their wealth to a foundation and making their heirs officers. The cost is that they can’t spend that wealth (or if they do as Trump did they can be penalized). This is not a cost at all for the super wealthy who are physically incapable of spending their wealth anyway.

In FRED there are not numbers on the total wealth of households. All refer to households and non-profit corporations. The approximation is that most assets of non profit corporations are effectively the property of their founders or their founders’ heirs. Similarly, Forbes counts assets of foundations when it tries to figure out how rich the super rich are (not giving to the Bill and Melinda Gates foundation did not affect the Forbes rank of Bill, Melinda, or Warren (oh that is Warren Buffett the one with immense wealth not Elisabeth Warren the one who wants to tax wealth).

A wealth tax absolutely has to confront this issue (and I know that E Warren has a plan for that).

I think the solution is pretty simple. It would have two parts. First the founder and any relative of the founder must not be an officer of the foundation. This would be like anti-nepotism laws (but with real money at stake). Second that the founder and all relative of the founder must not have any private, ex parte, communication with the foundation. Any communication with any employee of the foundation must be public.

That would mean that wealth which is given away is actually really away. This would be an extremely radical change. For one thing, it wouldn’t be a “taking” to apply this rule retroactively. The founders of foundations claim that they have no personal interest in the foundation. If they claim that forbidding them to communicate with the foundation is taking their property, they confess that they have committed tax fraud. I’m pretty sure Roberts et al would declare my proposal unconstitutional (no personal interest doesn’t mean no personal interest). I am pretty sure no Congress would approve it.

One might argue that, with my proposed rules, rich people wouldn’t give their wealth away and wouldn’t found foundations. I don’t see that as a problem. Bill Melinda and Warren can say that the assets belong to them personally to spend as they please and also give them to people fighting AIDS and Malaria and such like. Foundations currently are often like the personal property of the founders. I don’t see a big cost in forcing them to make that explicit (and pay 2% per year when they do).

The idea of forcing charities to act as charities and not as the property of their major contributors is too radical to implement. I am almost afraid to blog about it.

But without that, a wealth tax is just an invitation to even more fraud. I think this is an important issue.