"The shame, the horror, the inequality!," proclaims the progressive economist, sending the preprint of his study in the New York Times. "Retirees, who have a lot more in their bank accounts than working folks, aren't paying any taxes! All the taxes are being shouldered by poor workers! We need to tax money people take out of banks!"
"Wait a darn-tootin' minute," says the retiree, having spilled half his beer. (Old people talk like that.) "I worked my whole life. I paid Federal income taxes, state income taxes, city income taxes, and payroll taxes on that money. My company paid sales taxes, corporate income taxes, property taxes, mitigation fees and so on out of every dollar I billed made for them. I paid vastly overinflated health insurance to cross subsidize medicaid, medicare, and indigent care. I saved for my retirement, rather than blowing it all when I was young. Then every year I paid taxes on interest and dividends, and capital gains when I rebalanced my portfolio. It's a miracle I still have this lousy $5 left to buy a beer. No, thank goodness there is not a take-it-out-of-the-bank tax. But I paid a heck of a lot of taxes on this money!"
I think this little story captures the essence of one of the many little -- well, Phillip Magness in in the AEIER, reflecting a little traditional conservative politeness, calls them "fibs" -- in the latest Saez-Zucman effort to prove that rich people pay less taxes than you and I, an equally ardent effort to prove that despite everyone else's numbers the rich really did pay a lot more taxes in the golden 1950s, reinforcing the trope cited even in the Democratic debates* with lightning speed.
Others have torn the numbers apart, including Magness, David Splinter at the Joint Committee on Taxation, Larry Kotlikoff in the Wall Street Journal, Robert Verbuggen in the National Review, Michael R. Strain at Bloomberg and many others. To usually sleep-inducing results. Hand it to Saez and Zucman to know how to tell the big fib and get your name in the headlights. So let me focus on two very simple problems that my little story highlights.
Income taxes: Really the problem starts by defining the question as income taxes -- what is the tax rate on this year's income -- and defining income to include "capital income," dividends, interest, and capital gains. Once you allow that definition, much of the game is lost. Yes, old and wealthier people tend to have more capital income, and young and less wealthy people tend to have more earnings. We tax earnings at a way higher rate, so duh. The fact that this "capital income" comes from savings, which comes from earnings that were taxed at the most progressive rates in the world then vanishes in this accounting.
"Income" is really a fairly meaningless concept. We do not live in the Ancien Regime, or a Jane Austen novel in which people are described for life by the annual income they receive. Income varies a lot over a lifetime, and ebbs and flows for many. And "capital income" is not the same as earned income. The broad consensus theory of taxation states that capital income -- the rate of return you get to induce you to save some income for future consumption rather than to blow it all right away -- should not be taxed at all. It really isn't "income" in any meaningful sense.
Consumption is a meaningful concept, and lifetime consumption in particular. A progressive consumption tax makes a lot of sense. But as the many commenters above point out, very rich people tend to consume much lower fractions of their wealth than the rest of us. They take this "income" and plow it back into businesses and mortgages, where it is out there hiring the rest of us and producing products for us. The horror of it all. So if you look at lifetime taxes paid as a fraction of lifetime consumption... the rich are paying much higher rates than the rest of us.
Corporate taxes: Of all the other heavy thumbs on the scale, let me point to the other biggest one, the treatment of corporate taxes.
To assess the progressivity of the tax system, in any sensible way, we have to include all taxes, and as many transfers as we can. It makes no sense to include only one tax and say A pays less than B, if B pays more than A of some other tax. If A pays $10 in tax but gets back $5 in cash benefits, he or she is really only paying $5. We also have to figure out who bears the tax, not just who pays it.
It is painfully obvious that "corporations" bear no tax. Every cent of corporate tax is paid by people, either from higher prices, lower wages, or lower payments to bondholders and stock holders. By and large most estimates find it's mostly higher prices and lower wages. Corporate taxes are not paid by lower electric bills -- you either pay that or they turn off the lights. For the same reason, corporate taxes are, in the usual accounting, not much paid by stock and bondholders. They just invest less or abroad until the after tax rate of return is the same.
Into this interesting economic debate wade Saez and Zucman and assume all corporate tax comes out of shareholders' pockets. And since the corporate tax rate has been coming down over the years, and since wealthy people hold more stocks, aha, the wealthy are paying less tax. (Despite higher statutory rates back then, there were lots more deductions, so even they grudgingly admit that wealthy people did not pay higher income tax rates in the halcyon 1950s.)
Note they do not assume all sales taxes come out of shareholders pockets. (Actually, I'm assuming here -- if someone has read in detail let me know if I'm right on this.) It's painfully obvious that sales taxes, though "paid" by corporations, are "borne" by you and me via higher prices.
But sales taxes and corporate taxes come equally out of the bottom line of a company's profits. Why does a company make up for its sales tax payments 100% by raising prices, but makes up for its corporate tax payments 100% by lowering interest and dividends (somehow, magically, without that lowering of payments also lowering the stock and bond prices so rates of return remain unchanged)?
Well, there is no reason at all to make such an assumption. Except that sales taxes, which are indeed borne by all consumers, have risen over time, while corporate taxes, which they alone assume are borne by the wealthiest, have gone down over time. So if you treat the taxes equally, you get the "wrong" answer.
Yes, figuring out the incidence of taxation is hard. The tax system is deliberately complex and opaque so that people can't figure out what they really pay, and can't figure out that indeed rich people don't pay nearly as much as the headlines say -- even if it is still quite progressive.
Economists should be spending their time studying the disincentives of the tax system, not its "fairness," on which we have little professional expertise. Yes, you heard it here, economics really has nothing to say about whether "the rich" should pay more or less taxes -- but we have a lot to say about what happens to economies that tax away the incentive to become rich, to study, move, start businesses, innovate, hire others, and produce great products along the way. But the (dis) incentives of the tax system are even more opaque, again deliberately. Not even every tax economist I have ever asked the question of can answer "what is your all-in marginal tax rate?" If it were clear, I suspect a lot more people would stop working!
* From the Transcript
Biden: Why in God's name should someone who's clipping coupons in the stock market make -- in fact, pay a lower tax rate than someone who, in fact, is -- like I said -- the -- a schoolteacher and a firefighter? It's ridiculous. And they pay a lower tax.[Poor Joe: I won't belabor that stocks pay dividends, not coupons, and "they pay a lower tax" is absolutely false. At least he said "rate" the first time.]
By the way, a tiny ray of common sense in that debate:
"YANG: ... The problem is that it's [wealth tax] been tried in Germany, France, Denmark, Sweden, and all those countries ended up repealing it, because it had massive implementation problems and did not generate the revenue that they'd projected."
"If we can't learn from the failed experiences of other countries, what can we learn from?
We should not be looking to other countries' mistakes. Instead, we should look at what Germany, France, Denmark, and Sweden still have, which is a value-added tax. If we give the American people a tiny slice of every Amazon sale, every Google search, every robot truck mile, every Facebook ad, we can generate hundreds of billions of dollars and then put it into our hands, because we know best how to use it."Well, "we [the Democratic party, when it takes over the Federal Government -- I presume he does not think President Trump needs 22% of GDP to spend!] know best how to use it" might be a bit open to question, but I'll take rays of sanity where I can find them.
Update: A good point from Matthew Lilley via Marginal Revolution What about sales taxes? They seem almost too regressive in Saez and Zucman. Why? Because SZ calculate the tax rate as tax paid / pre-transfer income. So if you pay $1000 in sales tax on $10,000 of purchases, based on a $1,000 income and $9,000 transfers, you have a 100% tax rate. That's nonsense. SZ drop those people from analysis rather than correctly define taxes on post transfer income. Once you do this, more transfers (the solution) just make tax regressiveness (the supposed problem) worse!
The real lesson though is not to quibble with the answers, it is to quibble with the question. To me, it emphasizes the central point -- it's not about the numerator (taxes paid), it's about the denominator -- relative to what? Once you think "income" is a meaningful concept, you give up most of economic common sense. I guess SZ don't want to count transfers as "income" because they want to show how ruthless the pre-transfer market is, and one can sympathize with that sentiment. Sentiment does not excuse skulduggery however.