I am so used to the asymmetric incentives of the financial industry, where if you make a lot of money for the firm you get rich, and if you lose a lot of money for the firm you (1) get fired, (2) get to keep the riches you made previously, and (3) easily get another job because you have proved yourself as a risk-taker. These incentives, obviously, encourage risk-taking, which in some very broad sense is what the financial industry wants; investment banks and hedge funds are trying to identify and train skilled risk-takers. One can go too far with this, and over the last decade or so there has been a lot of conversation about dialing back these incentives, about bonus caps and clawbacks to reduce the asymmetry and discourage excessive risk-taking.
But the incentives of the corporate treasury industry are the opposite. Your job is not to lose money. If you succeed in not losing money, you receive your modest, stable rewards. If you fail and lose money, you are fired. If you succeed beyond your wildest dreams and make a lot of money, (1) you do not get any extra rewards (because that was not your job) and (2) if your bosses are smart, they fire you anyway, because the fact that you made a lot of money means that you took a lot of risk, which was very much not your job.
The conclusion here is probably that corporate treasurers are never going to decide to buy Bitcoin, at least not until it becomes broadly normalized. How can it become normalized if treasurers never decide to buy it? Well, corporate treasurers are not necessarily the final decision-makers about the corporate treasury. Corporate chief executive officers are chosen and trained and incentivized to take risks; their pay does go up if the company is worth more. Some of them are really into Bitcoin. If Elon Musk tells his corporate treasurer to buy Bitcoin, Tesla Inc. is going to buy Bitcoin. “Ehh I just work here,” the treasurer will shrug; “they don’t pay me to take risks, and telling Elon not to do this is a risk I’m not willing to take.”
Extremely Pleasing Tax Case
The way corporate income tax works is pretty much that you add up the company’s revenue, you subtract its expenses, and you’re left with its net income. Then it pays taxes as a percentage of that net income. This is basically straightforward, though sometimes people fight over what should count as an expense. (In the U.S., interest on debt is tax-deductible, but dividends on stock are not, which arguably encourages excessive debt, etc.)
The way personal income tax works is not like that. You add up all your revenue, but you do not subtract your expenses, generally; if you are paid $100,000 a year and spend $93,000 on food and clothes and rent and stuff, your taxable income is $100,000, not $7,000. Well, that’s not quite right either; there are a bunch of expenses that are “deductible”—they reduce your taxable income—and sometimes people fight over what expenses should be deductible. (In the U.S. these fights tend to be over things like mortgage interest, state taxes, charitable contributions, etc.)
But broadly speaking, in the U.S., most of the expenses of life are not deductible, and people get taxed on their gross income, not the income they have left over after buying stuff. But what is income? Obviously if you have a job and get a salary, that’s income. If you buy a lottery ticket and win, your winnings are income. Most of the time, if you get money, that’s income. But if you buy some shoes and don’t like them and return them, and the shoe store gives you a $100 refund, surely that’s not income. Your shoe purchase was not tax-deductible; you couldn’t reduce your taxable income by $100 by buying shoes. So when you return the shoes and get your money back, that should not increase your taxable income by $100.
Similarly if you buy a product and send back the mail-in rebate coupon and get back $5, that's not income. Or if you buy a product on your credit card that gives you 5% cash back, that 5% cash back is not income; it’s just a reduction in the purchase price. There is a “longstanding Internal Revenue Service practice, which says credit-card rewards are usually nontaxable rebates. In other words, buying a pair of shoes for $100 and getting a 5% reward is really a $95 purchase, not $5 of income.” 4
What if the product that you buy is $6.4 million worth of grocery-store gift cards that you then exchange for money orders that you then use to pay off your credit card bill, allowing you to keep the 5% cash reward offered by the credit card company? Then:
1. You are amazing, my friend, truly, hats off to you.
2. But is it taxable income?
The person who did this is named Konstantin Anikeev, and he is an experimental physicist. The Wall Street Journal, which tells the story of his gift-card experiment, calls it “an inquiry far outside his field,” but I disagree; he built a perpetual motion machine, which has fascinated experimental physicists for centuries. He was able to “exploit the difference between unlimited 5% rewards and lower fees on gift cards and money orders” to just shuffle money around, risklessly, routinely, in a way that left him with more money than he started with:
His American Express card offered unlimited 5% rewards at grocery stores and pharmacies after he had spent $6,500. So Mr. Anikeev used his AmEx card to buy prepaid Visa gift cards at grocery stores, routinely stopping during his commute and purchasing the maximum allowed per day at a store. He often used the gift cards to buy money orders, then used the money orders to make deposits in his bank account, then used that money to pay his credit-card bill.
In a $500 transaction, the 5% rewards would yield $25—more than enough to cover gift-card fees of about $5 and the $1 fee on the money order.
He made about $310,000 doing this.5 As an exploit of credit card rewards it is fairly straightforward, though ambitious in its sheer scale. (An American Express spokesman “said the company uses a ‘combination of strategies’ to police the rules of rewards programs that don’t allow purchases of cash equivalents.” 6 )