How to Fix Our Tax Problem
The former chief economist for the Fed told me I’m a lunatic. But here's what we do: Tax the rich. And the poor.
But, you’re right. The bulk of the tax would not come from income. Instead, it would come from a tax on the following type of transaction, which is not currently taxed. My firm just negotiated a $45 million loan for a client. Under a 1 percent transaction tax, the lender would pay a tax of $450,000 on the loan (which, of course, it would pass on to the borrower). The borrower would pay a 1 percent tax on all the loan repayments, meaning a total tax (with the transaction tax also being charged on the interest paid) of almost $1 million.
Now we’re talking. But would that kill the economy? I don’t think so. The tax would be only half of the “facility fee” that the bank charged for the trouble of making the loan to my client. With all the broker, attorney, and accounting fees my client paid, the extra $450,000 would be barely a blip. And with the interest rate as high as it was, the extra one point on the repayment wouldn’t have fazed him. (And let’s not forget that he’s no longer paying an income tax on his business profits, so he has that extra money.)
But wouldn’t a 1 percent tax on stock trades kill the stock market? I’m involved in a small hedge fund, so I’m not entirely ignorant of how the stock market works, and I certainly have an interest in the outcome if the stock market were to tank (that is, tank worse than it did in the internet bubble of 2001 and the subprime debacle of 2008 and the European debt crisis of 2011, etc.).
I called a friend who works at a large investment bank.
“It would kill the stock market,” he said.
“If you tax something, you get less of it.”
“Well, we have to tax something,” I said, “and if we’re taxing income, then by your thesis we’re getting less income.”
“The stock market is bigger than that. It would destroy the efficient allocation of capital.”
By “efficient allocation of capital,” he means high-frequency trading where stock traders make thousands of trades per hour in the same stock, exploiting minute spreads in the bid and ask price. In other words, he meant it would kill his business.
But fluctuations in a company’s stock price do not necessarily affect the company’s business. It’s not like the company is getting the money being made by the persons buying and selling its stock. Does Apple make fewer iPhones when its stock drops $8 per share? Does it really need its stock price to be “efficiently allocated” down to the penny on a second-by-second basis?
I called a portfolio manager I know who runs a hedge fund. He has about $75 million of capital under management.
“It would kill performance across the industry,” he said. “Most funds turn over their portfolios between four and five times per year. A point to sell and a point to buy equates to a 10 percent tax on the value of the portfolio, assuming five turns. That’s a $7.5 million tax for us.” (He
can do math pretty quickly in his head.)
“So if that’s the kind of tax that came down the pike, would you have to fold the tent?”
“No, the market would adjust. It would kill high-frequency trading, of course, but there would still be trades. A lot of stocks fluctuate by more than 1 percent per day anyway. Plus, reported performances would be normed for taxes.
“Trading as much as I do means that most of our gains are short-term taxed at 35 percent. Assuming five turns of the portfolio every year, I would have to produce a 28 percent return to equate to a 1 percent transaction tax. Pretty tough to do on a consistent basis, but I would just trade less, which would reduce the tax hurdle. We would be fine.”
A pretty quick retreat.
What about the broader economy? Would people stop buying? Of course not. No one cares about the little extra that comes on a bill. I thought about the custom floor mats for my car that I bought online for $149. The “shipping and handling” charge was $12.99—8.7 percent!—but I didn’t think twice. Think about all the extraneous charges on your phone bill. Think about any bill. No one cares about small transaction costs.
So it was time to test my idea on some smart people, people with more august titles than just “stock trader.” I called Michael Cox, a fellow professor at SMU. He’s the director of the O’Neil Center for Global Markets and Freedom, and he holds the distinction of having been the only man ever to hold the title of chief economist for the Federal Reserve System.
“It’s the height of lunacy!” he said. “It would be horrible for the economy.”
“It’s the equivalent in the surgical world to operating on someone with a butter knife.”
“But why? What part specifically?”
“The general principal of taxation is to administer taxes in such a way that they are the least discouraging of good behaviors—things we
want people and businesses to do. A bad tax is one which discourages things that lead to the production of GDP. Education, saving, investment, productivity, making something, starting a business, hiring people, inventing, innovating, and so on—these are things you don’t want to tax. A ‘good’ tax is one that falls on consumption, especially if that consumption behavior is undesirable. Like cigarettes. A transactions tax would tax all things equally, which in no way would be optimal.”
“There’s research that assessed the optimization of taxes?” I said. “Can I see it?”
“I’ve done economic research for 40 years,” he said. “I know a crazy idea when I see one. There are thousands of crazy ideas that I don’t spend time taking seriously.”
“So there’s been no research?”
“Has anyone done any serious research on space aliens traveling to Earth to create crop circles?”
To be fair, he mentioned early in the call that he had just returned from Disney World with his small child. That will put even the sturdiest horse off its feed, so I decided not to mention that I totally disagreed with his thesis about using the tax system to encourage good behavior and discourage bad. In my view, the ideal tax system would not be in the business of deciding what’s good economic activity and what’s bad. Who is to decide? Moreover, you’ll immediately fall into the conundrum of using the tax code to encourage the construction of cigarette factories—and then taxing the heck out of them to discourage their consumption. It was clear that the good professor would not be in the mood for such an argument.
But I knew I was on to something.
The editor of this magazine sent my idea to someone he knew at the Federal Reserve Bank of Dallas. The Fed staffer emailed back to state that a limited version of the transaction tax had been proposed by Pennsylvania Representative Chaka Fattah (also noted by professor Cox), but it didn’t include stock sales. He said that the idea had been debunked by Snopes.
Snopes? Really? (To be fair, the staffer may have just dashed off the email without giving the topic a lot of thought, and he explicitly stated that his observations were not the position of the Fed. But still.)
His other objections were (i) a transaction tax is less efficient than a VAT because it can tax the same productive activity more than once; (ii) it’s more regressive than a sales tax; and (iii) some currently public transactions would be driven underground.
We’ve already shown that a 1 percent transaction, even assuming a tax on 100 percent of wages and 100 percent of spending/savings is hardly regressive—but that just shows the innumeracy of human thinking (although surprising to witness in a Fed staffer). One percent is simply very small. Otherwise, no one would withdraw money from an ATM and pay a $2 transaction fee (which, for any withdrawal under $200, is far more than 1 percent).
Some public transactions being driven underground? You mean, more than the $2 trillion that’s operating down there already? Also, right now under-the-table income is not ever caught in the federal tax web. However, all transactions eventually appear at the cash register or at the leasing office or in the stock market, in which case the federal government would get its sliver.
Finally, taxing the same productive activity more than once? The income tax is the worst on that score. Consider $100 of earnings at ExxonMobil. It pays $35 in corporate income tax. Let’s say it distributes a dividend of the remaining $65 to its shareholders. They pay a 15 percent income tax, leaving $54.4. Let’s say they spend all $54.4 at an ExxonMobil gas pump. ExxonMobil’s wages are approximately 20 percent of gross revenue, meaning $10.88 of that $54.4 is subject to 35 percent income tax at the employee level, or another $3.81 in tax. Finally, ExxonMobil’s income is approximately 8 percent of gross revenue, meaning an additional $4.35 of that $54.4 is subject to 35 percent corporate income tax, or another $1.52 in tax. Bottom line, $100 in ExxonMobil earnings is stripped down to less than half remaining after the repeated bites at the tax apple. No way can a transaction tax even approach that level of tax drag.
So unless you’ve got something better than Snopes and space aliens, I believe the transaction tax works—as long as the dollar amount of all transactions is at least $370 trillion annually (to fully balance the $3.7 trillion budget). Sounds kind of steep, though, given our economy is only $14 trillion. Is the money there? Let’s look again to our favorite whipping horse, the stock market. Average daily dollar amount of trades in stocks and bonds is about $900 billion, or $329 trillion annually. Add total wages, total consumer spending, total business spending, total property sales, total inheritances, and total credit (to name a few)—and I think we more than get there. Even assuming a drop-off in securities trades, I think the transaction tax is less than 1 percent.
Put that on your butter knife and spread it.
Write to firstname.lastname@example.org.