Tuesday, June 15, 2021

How the Ultra-Wealthy Get Away With It (Paying Ultra-Low Taxes): In Part, Because It Is Truly Complicated

by Neil H. Buchanan
Last Thursday, I wrote a column here on Dorf on Law discussing ProPublica's recent blockbuster report on the incredibly low taxes paid (in many years zero dollars) by the 25 wealthiest people in the world.  As a percentage of accurately measured annual income, these wealthy people pay what ProPublica calls "true" tax rates in the low single digits.

What I hope was my main contribution with that piece was to emphasize that the non-news -- extremely rich people get away with paying low taxes -- is accompanied by good news in the way that the report changes the terms of the conversation.  Specifically, ProPublica's true tax rate calculation substitutes "change in wealth" in the denominator for the more standard "taxable income."  (Even the broader "gross income" leaves out most changes in wealth.)  As I stated there, and as I will say at more length here, that is a non-obvious move and can be misunderstood, but it is genuinely accurate.
Most importantly, it forces the conversation into an area that wealthy people hope to avoid, which is how they actually accumulate wealth.  They would much prefer that we continue to be distracted by the complications of "gross income," "adjusted gross income," "taxable income," and all of the deductions and credits that make people's eyes glaze over.  The more they can muddy the water, the better able they are to get the public (and thus members of Congress) to say, "This is all a mess.  I know the rich are getting away with something, but it's all too hard to figure out.  I guess there's nothing we can do."

In expanding on my points from last Thursday, I will draw from two very sensible contributions on the comment board, one from a reader who goes by Unknown, the other from a reader who goes by Michael C. Dorf (who, unless Professor Dorf's identity has been taken over by a sock puppet, is the proprietor of this blog).  Both comments raise questions about how we measure and could decide to tax wealth.  Both enhance the discussion considerably.

Unknown wrote:
Professor, I am unsure I can agree with your classifying wealth growth as income. I can't directly trade, at least no reasonably, the increase in price of my Apple stock for food, transportation, or medical care, for example. I have to sell some of that stock, at which point I am assessed a tax on the net increase in market price. Meanwhile, the rest of my wealth in such a stock is only theoretical at most. Contrary to your assertion at the ned of your essay, recognition of this lack of income is not a "redefinition" of the concept. 
This raises an important conceptual matter.  To be clear, I am not making a novel argument (nor is ProPublica) that income is properly measured by including changes in wealth.  That is, in fact, the most basic definition of income out there, and it has been for more than a century.  (The incorrect re-definition comes when people claim that "realized income" and "income" are one and the same.)   The so-called Haig-Simons definition of income, which is also called "the economic definition of income," says that income in a given year is equal to the sum of two things: the fair-market value of a taxpayer's consumption, and the change in a person's wealth.

To be clear, I am neither hiding behind someone else's authority nor saying that we should always believe economists.  (Frequent readers of this blog know that I have a great many differences with my economics colleagues.)  I am, however, saying that there is nothing controversial in using change in wealth as part of the definition of income.  This definition makes sense because the result of receiving income is either to increase one's consumption or to increase one's wealth.  That is why, for example, consumer loans are not income, because they give people cash with which to consume but they are offset by a decrease in wealth (higher liabilities).
As an aside, the "theoretical" nature of the gains means that an asset's value could go down in the future.  That can be handled by allowing tax credits for declines in value (negative income) in subsequent years.  That is probably the strongest practical reason to stick with a realization regime, but I happen to think it is not strong enough, given the benefits of dumping realization from our system.  Similarly, for non-rich people who have truly non-liquid assets like owner-occupied homes, an exemption or special rule would make sense.
My point is that we are giving up quite a lot of taxation of unrealized gains when we look at these practical problems and conclude that we must simply abandon taxation of all unearned income.  The tax law casebook that I use, by the way, includes a chapter subsection with the title: "Realization: The Root of All Tax Evil."  Quite so.

So Unknown's point is important in questioning whether we would want to tax all income after we have accurately measured it, which is not the same as saying that income should not be measured by including changes in wealth.
The comment notes, for example, what we commonly call the liquidity issue, using the example of Apple stock that cannot be used to buy goods and services.  That is an important issue, but to belabor my previous point, one cannot argue that receiving Apple stock is not income, nor could one argue that an increase in the value of one's existing Apple stock is not income, because (among other things) people readily accept both of those things as income in their own accounting.  That is, if I owned stock and my employer told me that she would make the value of my stock go up rather than paying me a salary, I would view that as an acceptable substitute.  Whether I receive salary or receive increased wealth, I have received income.

If I want to buy goods and services and I have no other cash, I can (as Unknown says) sell some shares of stock and then buy goods.  Under current tax law, that is the time when the person pays tax, that is, when the gain is "realized."  (I can also choose to realize assets with losses, if they exist, strategically allowing myself to consume without paying taxes.)  But although we currently do it that way, there are other ways to set up the tax system.  We currently say that a person who has received cash salary and wants to put it into financial assets has to pay taxes in the year earned, even though that means that people are not able to buy as many financial assets as would be possible if they were not taxed in the year that the salary was received.  Having a person liquidate some Apple stock to pay taxes on the appreciation of Apple stock leaves him in the same position as a person who paid taxes on cash salary before buying Apple stock.

To be clear, this is all a matter of making sure that substance dominates form, which is essential because as soon as we tax-favor certain forms of income, we give people an incentive to receive their income in those tax-favored forms.  And that is the story of the ultra-rich, who have the ability to become richer in forms that (they have made sure) are not currently taxed.

Michael C. Dorf's comment says, in part: "I agree that the realization rule is a huge giveaway."  To emphasize the point, making sure that we receive income in tax-favored ways is something that the rest of us can only dream about.  From a purely pecuniary standpoint, the ideal situation is to learn that one's income is not being taxed.  But a very good next-best situation is to learn that, although one's income is going to be taxed, the taxpayer herself has the ability to decide exactly when she will pay tax on that income.  That is why the timing of taxes is such a big part of any tax law class.

To clean up a related point, I should mention that these wealthy people, even though they often spend lavishly on consumption items, do not need to realize income in order to pay for that consumption.  Their appreciated assets are rock-solid collateral for loans at near-zero rates, so they are able to produce liquid cash as needed.
But to return to the reason that the realization rule is such a giveaway, what we are in fact doing is providing an interest-free loan to people who receive their income in unrealized form.  Public policy aside, I would rather pay taxes on this year's income twenty years from now, so long as my amount owed is not charged a rate of interest.  The time value of money means that interest-free loans are advantageous.  Paying $1 million today or $1 million in twenty years is an easy call.  (There is always the risk of one's tax rates rising, but rates can also fall.  That is a different matter entirely.)
The commenter whom I will presume to be the real Professor Dorf (in part because the question is so trenchant), raises one point that is too technical to go into here but then raises an essential policy issue: so-called stepped-up basis, which wipes out all unrealized gains for tax purposes upon death.  Not only do people who have the necessary financial freedom (and acumen, and access to expertise) receive interest-free loans from the government even when they eventually pay taxes on realized gains, but they might never pay taxes on those gains at all.  That, in fact, was the hidden unfairness that the now-all-but-repealed estate tax was supposed to partially (and imperfectly) mitigate.

Which brings us back to remembering why the answer to Unknown's question carries such a key insight.  Super-rich people become wealthier and wealthier every year, and even if there are occasional bear markets, their unrealized gains are very much income.  Otherwise, we would have to believe that it is possible for someone to become richer and richer for decades on end without ever having received income.  We can redefine words if we wish, but "got richer" and "received income" seem like they ought to mean the same thing.

In the end, as the title to this column indicates, the deeper problem is that accurately measuring income is not intuitively obvious, because there are so many forms in which income can be received.  "It's not liquid" does not mean the same thing as "It's not income," and liquidity problems are a good reason not to tax only some people, none of whom are the ultra-wealthy.  But unlike Unknown's questions, which lend themselves to clarification and possible policy adjustments, most people do not get nearly that far.  They merely think something like, "That doesn't sound like income to me," followed by, "Boy the IRS is just inventing ways to take people's money."  And Elon Musk and the rest say, "Yeah, that's right!  Now let me get back to my spaceship."


Michael C. Dorf said...

I am not a sock puppet (but of course, that's what I would say if I were a sock puppet, so we're back where we started).

David Ricardo said...

I will break my silence because this topic is an area where I have both academic credentials and real world credentials, and because the post and related comments illustrate the issues between an academic approach and a real world approach.

There is nothing incorrect about any of the analysis here. It is cogent, consistent and sound. But it never addresses a core issue, which is why is ‘realization’ is used in taxing gains. Of course, one reason which is correct is that high net worth individuals use the concept to avoid taxes. In fact, the realization concept combined with estate tax law, which allows a step up in basis results in taxes on gains being never taxed

But the main reason why realization is used is that valuation of assets for which there is no active market, such as privately held businesses, intellectual property, real estate etc is that recognition without realization is a terrible costly exercise in both time and money, and it is a terribly inexact science. Trust me on this, or don’t. If you don’t then read about the many years it takes to value an estate with a large number of assets for which no active market exists. Years for sure, and sometimes decades.

A problem is that the public does not understand this. News reports continually estimate the value of Bill Gate’s net worth, or Donald Trump’s real estate holdings, or any other mogul who holds non-traded assets and treat the estimate as exact. It is not.Even if the bulk of one’s assets are public traded stock it does not mean that the value of the holdings are the stock price times the number of shares. If one has 1,000 shares, and the market price is $20.00 per share it is reasonable to value the holdings at $20,000. If one has ten million shares, such a large block might have to be sold at a discount, or might be sold at a premium.

To get back to Trump, and who doesn’t want that, he made a conservation easement for some property and deducted the value of the property. That value is under investigation and potential litigation, both civil and criminal. How long will it take to conclude? A decade is not a high estimate, it is a low one.

Valuation theory and practice is a large part of tax and finance. But the answer to ‘what is this worth’ has no definitive answer. It is what valuation experts say it is worth, and highly educated and experienced valuation experts have honest disagreements. Don’t think it’s a problem, how happy are people with their house valuation for real estate taxes?

So the major reason why realization is used rather than recognition is that realization gives a definitive result to value. The problem is not using realization, the problem is in the tax code which allows tax avoidance even when there is realization.