A Wealth Tax Would be a Mess

Democratic presidential candidate Elizabeth Warren has been advocating a wealth tax for a while now, and it seems like a bad idea to me.

Let’s start with the Trumpian level of dishonesty in the repeated claims of supporters that Warren’s wealth tax asks billionaires to pay a mere “two cents.” Obviously, it’s two percent — two cents on each dollar — and calling it “two cents” is a cheap trick to minimize it. And the wealth tax is not just for billionaires, as the 2% rate kicks in at $50 million. As for billionaires, the latest version of Warren’s proposal taxes their wealth at 6 percent. These are not trivial amounts of money.

It’s important to understand that a wealth tax is different from an income tax because a wealth tax is levied over and over again, year after year, on the same wealth. Your income is taxed only once, when you earn it. If you made $50,000 last year, you paid income taxes on that money when you earned it. But this year’s taxes will be based only on this year’s income. The IRS won’t make you pay taxes again on the money you earned last year.

Wealth taxes are different: You will pay a wealth tax on your wealth over and over, for every year that you are wealthy enough to meet the tax threshold. Given enough time, that’s a lot of money. Someone paying at the billionaire rate will lose half their wealth in just over 11 years.

That’s one of the problems with the wealth tax. According to estimates provided by Warren’s economic advisors,

If her wealth tax had been in effect since 1982, for example, Mr. Gates, who had made his first billion dollars by 1987, would have had $13.9 billion in 2018 instead of $97 billion.

Jeff Bezos, the world’s richest person, would have had $48.8 billion last year instead of $160 billion. And Michael Bloomberg, who is considering running for president himself, would have had $12.3 billion instead of $51.8 billion.

Now if you hate billionaires, that probably sounds awesome. But it’s not so good if you’re trying to raise government revenue. If we launch the wealth tax today, the first year will seize 6% of Bezos’s $160 billion fortune, or $9.6 billion, which a nice bit of change for the U.S. Treasury. However, in the hypothetical where the wealth tax started in 1982, the 6% wealth tax today would be applied to only the remaining $48.8 billion of Bezos’s fortune, producing about $3 billion in tax revenue. That’s a decline in tax revenue of more than $6.6 billion for Bezos alone, with more losses for all the other billionaires affected.

That’s what happens when your taxes come directly out of your tax base: Your tax base erodes, and your tax revenues go down. Every year, the billionaires have 6% less wealth, so revenues decline at 6% per year. The wealth tax will eat itself.

(That’s assuming that billionaires do nothing to minimize their tax burden, which seems an unlikely thing for billionaires to do. More on this later.)

These billionaires won’t just be paying taxes on their wealth, they’ll also be earning revenue on it. But from the billionaires’ point of view, all of their investments will now be earning 6% less per year due to the tax. Or to put it another way, if billionaires want to make money on their investments, they have to find investments that offer earnings greater than 6%.

In general, the world of investments offer an inverse relationship between safety and high-earnings. U.S. Treasury bonds are very safe, but they don’t earn very much. On the other hand, you could make a ton of money investing early in a high-tech startup, but you’re more likely to lose all your money. Most prudent investors diversify their portfolios, so that they can earn some money on the risky stuff, but they have a nest egg of safer investments to preserve their wealth in case of bad times.

The 6% wealth tax rips away that safety — anything earning less than 6% is a loser — there are no good safe investments for rich people any more. This seems likely to create shocks in the economy, as billionaires dump their safe investments and head for the riskier end of the market. Other investors will pick up some of the slack, but this seems likely to do a lot of damage to the bond market, especially for U.S. Treasury bonds and bonds issued by cities to pay for infrastructure, both of which are safe enough to pay far less than 6%.

It’s also not exactly clear how billionaires will pay their wealth taxes. The problem is that taxes have to be paid with money, but the wealthiest people don’t hold their wealth as money. It’s not like Jeff Bezos checks the Bank of America app on his iPhone Amazon Fire to see how much he’s worth today. Jeff Bezos owns owns stuff. Lots of stuff. And to pay a wealth tax, he’d have to sell some of that stuff. Depending what exactly he owns, that can be a problem. Many things are hard to sell quickly, such as real estate or business holdings, and sometimes the value of the holdings depends on who owns them.

For one thing, ownership is control, and control is important for an entrepreneur. Selling 6% of Amazon every year would mean Bezos giving up control of the company piece by piece. Using Warren’s estimate above, he would have lost 2/3 of his Amazon stock by now. He’d still be a very influential shareholder, but he would have lost control years ago. If that had happened, I doubt Amazon would be the same company it is today. Visionary companies need their visionary founders. Would Berkshire Hathaway still have the same credibility if Warren Buffet had lost control? Would Tesla be Tesla without Elon Musk? How much would the Harpo entertainment empire be worth without Oprah’s star power?

It’s hard to answer questions like that. In fact, it’s hard to answer questions about wealth in general. And if it’s hard to understand something, it’s hard to tax it.

Income is relatively easy for the IRS to discover, because income has to come from somewhere — there has to be a transaction with someone else — and that transaction is likely to be observable. It will almost certainly involve banks and other financial intermediaries who will keep records that the IRS can use. And in many business transactions, the receiving party’s taxable income is also the paying party’s tax deductible expense, giving the latter a strong incentive to see to it that the IRS is aware of the transaction.

Wealth, on the other hand, just sits there. You don’t have to do anything observable with it. So it’s a lot harder for the government to see it. Oh sure, the government can tell how wealthy I am, because my wealth is simple: I own a house and have some banking and investment accounts, all of which are easy for the government to discover. But not all wealth is held so simply.

To some extent, wealth can be inferred by looking at the flow in of income and expenses. If I know that over the entire course of your life you’ve earned income worth $4 million, and I know you’ve spent $3.9 million, maybe I can infer that you have retained the rest of the money, making you worth $100,000. Unfortunately, that only works if you spent the $3.9 million entirely on consumption goods that you used up. But what if you spent some of that money on something that retains value? If you spent $1 million on Amazon stock, that would still have value. It would be an investment, and it would count toward your wealth. You’d be worth $1 million (plus change).

That’s an easy case, though, because publicly traded stock has two important characteristics that make it suitable for assessing wealth: (1) Ownership is carefully tracked by independent parties, and (2) the value of the stock is easy to establish because it is traded so often. Not all wealth is so easy to understand. If you had purchased $1 million in gold and buried it in your back yard, you could be worth $1 million and almost no one would ever know. You could give it to someone or leave it to your children when you die, and then they’d be worth $1 million, and again, no one would know.

Even that’s a simple case, because it’s fairly easy to establish the value of Gold, so if someone (like the IRS) discovered you had the gold, they’d know how wealthy you are. The situation gets a lot murkier if you bought investment goods that are harder to set a value to, such as gemstones, artwork, or antiques. If you bought a painting for $1 million, even you wouldn’t really know how much it was worth unless you tried to sell it. Maybe some expert will pronounce it to be a fake, in which case it’s worth nothing. Or maybe it will appreciate to $10 million because someone made an Oscar-winning movie about the painter and all their stuff is incredibly hot. How would you put a value on that for tax purposes?

Or how about intellectual property? J. K. Rowling has made about $1 billion from selling Harry Potter books and movie rights and other merchandise, and that income has become part of her personal net worth. But in addition to what she has made from selling Harry Potter so far, she still owns the rights to keep selling Harry Potter stuff for the rest of her life, and to pass that ownership on to her children. It’s anybody’s guess what that’s worth right now, and I don’t know how you’d tax it, or how she could pay taxes on the net present value of future income she hasn’t earned yet.

Then there’s author Dan Brown who by 2003 had sold several books, including three novels, but remained pretty much an ordinary guy with ordinary amounts of wealth. And then in 2004 he published The Da Vinci Code which went on to become one of the most popular books of all time, boosting the sales of his first three novels, and inspiring a series of films starring Tom Hanks. With that single book, Dan Brown became a wealthy man. But when did Brown’s wealth first materialize? He sold the book for a ton of money, but doesn’t that mean that the book was worth a ton of money even when it was just sitting on his computer? Yet the value of the The Da Vinci Code was kind of murky until Brown sold it. But it wasn’t selling it that made it valuable. Selling it just priced it. It was already valuable. Dan Brown was already wealthy. It’s just that nobody knew it yet.

Since I started writing this last month, we’ve all heard about Maurizio Cattelan’s $120,000 duct tape banana, right? That thing is made from a few cents worth of duct tape and fruit, but so far the artist has sold three of them for $120,000 to $150,000. So when exactly did Cattelan become $390,000 wealthier? When he got the idea for the artwork? When he bought the bananas? I haven’t got a clue, and neither would the IRS.

These may seem like esoteric examples, but I chose them because they are vivid examples of the general problem with assessing wealth: A lot of wealth is held in confusing and intangible forms.

Think about the money in your bank account. We say it’s your money, and yet, at this very moment, you don’t have any of it in your possession. The bank has it. But we think of it as your money because the bank is required to give it back to you whenever you want it. In this way, your wealth consists not of money or goods, but of the bank’s promise to give you money. That promise is pretty reliable, because it’s enforced by contract, law, tradition, reputation, and the Federal Deposit Insurance Corporation. That money is so safely yours that it’s, well, “money in the bank.”

But let’s say you withdraw $100,000 of that money and loan it to me. Has your personal wealth changed? The amount of money you have in the bank has been reduced by $100,000, but you also hold a promissory note from me for $100,000, so arguably nothing has changed. On the other hand, I’m a bit less likely to pay you back than the bank was. I could go broke, and I’m not insured by the FDIC. You might only be able to recover a fraction of the $100,000 from the bankruptcy proceeding. Taking into account the risk of my being unable to repay you, that $100,000 promissory note is worth a bit less than $100,000 in the bank, and possibly a lot less if I’m in bad financial shape.

In other words, your wealth depends on my financial stability. So if the IRS wants to figure out how much wealth you have, they need to know something about my finances as well. (That would be especially complicated if the IRS couldn’t learn about my finances, perhaps because I lived in another country.)

Almost all investments work that way. Corporate and government bonds are essentially promises to pay, and their value is related to their face amount, adjusted for the risk that the corporation or government will default on the loan. This means that assessing your wealth means assessing the financial health of all the bond issuers who owe you money. As a practical matter, this is fairly easy if the bonds are publicly traded, because the value of the bonds can be assessed at whatever price you can sell them for in the market. But it’s a lot harder for privately-issued bonds and promissory notes that aren’t priced by the market.

Stocks are similar but more complicated. Stock in a company is a promise to be paid some portion of the future revenues of that company, which is even harder to figure out than whether the company will be able to repay a loan. Again, it’s easy if the stocks are priced on a public market, but a lot harder if the stocks are privately issued and traded, as they would be for a company owned by a family, a group of founding investors, or any limited investment arrangement.

Things get really complicated once we drift even further from tangibility. Consider out-of-the-money stock options. An option is a promise to allow you to buy (or sell) something at a specific price over a specific time period, and a stock option provides this for a company’s common stock. So suppose you have 1 million options to buy Amazon at $1800 per share for the next 12 months. As I write this, Amazon stock (AMZN) is trading for $1848 per share, so if you chose to exercise your options today, you would simultaneously buy 1 million shares for $1800 and sell 1 million shares for $1848, earning you a profit of $48 million. This means that your million options to buy Amazon at $1800 are worth $48 million today. Pretty sweet. And pretty easy to assess for tax purposes.

But suppose instead of a strike price of $1800, your million Amazon options have a strike price of $1900/share. It would be stupid to exercise them today, because you’d have to pay $1900 for shares that are only worth $1848. You’d be losing money. At first glance, it may seem like this makes your options worthless. However, because they don’t expire for a year, there’s a chance that Amazon stock will go over the $1900/share price, making it possible to profitably exercise your options at some point in the near furure. (AMZN was at $2000 in July, so $1900 is not unrealistic.) So how much are your Amazon options worth now?

It’s not easy to figure out. The value of a publicly traded stock is dependent on what investors believe the underlying corporation’s net cash flows will be in the future, so the value of the options depend on what investors believe that future investors will believe about those future cash flows. That’s a lot of speculation about the future, which is why determining the value of out-of-the-money options is a fundamentally hard problem.

(As with some of my other examples, determining a price is fairly easy if the options are publicly traded — as I write this, AMZN options for $1900 expiring in about a year are trading at $187, making those million hypothetical options worth $187 million — but figuring out the value of an option is much more difficult for a private contract.)

Options are only one example of a type of financial product called a derivative, in which the value of the derivative security is related to the value of some other asset, but does not represent a direct investment in that asset. The underlying assets can be things like stocks, bonds, currencies, and commodities, or they can be abstract concepts like market indexes and interest rates.

The value of the option described above is related to the value of Amazon common stock, but it isn’t actually Amazon stock. It’s a call option which conveys the right, but not the obligation, to buy a stock at a certain price. A put option is the opposite: The right, but not the obligation, to sell a stock at a certain price. Other types of derivatives are futures, which include the right and obligation to buy or sell at a price, and swaps, which create obligations for trading conditioned on future financial events.

Under normal circumstances, derivatives serve the purpose of allowing one party to assume risks on behalf of another party, in exchange for a payment. If you’re an airplane manufacturer and you sell a fleet of jets to an airline in India, deliverable in one year for a payment at that time of 100 billion Indian rupees, you might worry about the risk that rupees could lose value against the US dollar, resulting you receiving a smaller payment than expected. You could avoid that exchange rate risk, however, by paying a third party to guarantee the exchange rate, thus taking on the risk for you.

Farmers planning to take cattle to market and large restaurant chains that use beef can both ensure stable pricing by creating contracts to trade specific amounts of beef at specific prices at specific points in the future. Banks can reduce the risk that their creditors will not be able to repay the loan by buying credit default swaps that require the seller to make good on loan payments.

Derivative contracts are often very complicated — involving various thresholds, options, discontinuities, and triggers — and they can sometimes be traded independently of the original parties and bundled together into various pools, tiers, and tranches. There is a whole industry involved in creating these kinds of derivatives, and way too many of them are sold by shady brokers to naive investors.

(For example, there are so-called “principal protected” mutual funds which promise to protect your holdings against market losses that eat into your original investment. This can sound like a great idea for securing your retirement funds, but these are actually complex financial derivatives that are difficult to understand.)

Even perfectly legitimate derivative securities can be far too complex for most people to fully understand, and even large institutional investors have lost tons of money investing in derivatives. Remember the mortgage crisis that crashed our economy a decade ago? That was derivatives.

I’m sorry if I appear to have wandered too far from the subject of wealth taxes by discussing exotic financial instruments, but I wanted to show you a glimpse of the complexity involved. And I am by no means a finance expert. I’ve only scratched the surface. It gets much, much worse.

This brings me at long last to my point: If we decide to tax the wealth of rich people, then all of the wealth of rich people will begin to look like this, as they restructure their holdings to make it harder to tell how much they’re worth. An entire industry of lawyers and financial advisers will spring up to help them with this. Small island nations will rewrite their laws to allow American billionaires to own investments that are not reported to the IRS. Rich people will own complex and confusing portfolios filled with international derivative securities, secured by secret agreements that are enforced through the laws of nations that guarantee financial privacy.

As a result, it will be a lot harder to collect wealth taxes than it seems right now. Furthermore, all of this restructuring will be a wasteful drag on the economy, because avoiding taxes is not a productive activity. And finally, the complexity of all this restructured wealth will make it a lot riskier, not just for billionaires, but for the entire economy.

In short, I think the wealth tax will probably not accomplish its goals, I think it will waste American productivity, and I think it will make the economy less stable. If we want to raise tax revenues, there are better ways to do it.