The Case for Regressive Taxation

Please bear with me as this is more of a thought experiment than a manifesto… We’re talking equality again. Here’s Bain Capital executive Edward Conrad, putting forth the case for inequality, and here’s TAS Alum and all around genius extraordinaire Jim Manzi, guest-blogging for Friend of the Scene Megan McArdle and arguing that inequality is a bug, not a feature. Conard has set off a firestorm of discussion, and while I’m not sure I agree with him I think he is at least not easily discounted.

With that in mind…

Believe it or not, I’m quite sympathetic to the idea of progressive income taxation. The moral argument according to which, when one earns a lot of money, one should give back more (more proportionally, even), is one which I am wholly untroubled by and even endorse.

Some will point out that progressive taxation is not “giving back” more, it’s “being forced to hand over” more, and to these I say: humbug! Elizabeth Warren’s right: if you become rich in a developed country, you are the beneficiary of the common goods established by a nation (education, enforceable contracts, infrastructure and so on) and financed by taxation, and you are a party to a social contract that says that in exchange for these common goods if you use them to make lots of money you shall hand over lots of it. From a moral and legal perspective, the state is not the same as society, no, but it is a representative of it which is legitimate through elections and the enforcement of the rule of law and certain fundamental rights. It is therefore entitled to as much as your earnings and wealth as it wants until that share shades into expropriation, and while we can argue about where the border to expropriation lies, I don’t think you can straightfacedly argue that it lies westerly of any progressive tax.

I’m even untroubled by the idea that taxation can have moral goals beyond economic policy. As long as you’re not collectivizing the kulaks, it is legitimate (desirable is another question, but legitimate) for a polity to say “Even if we get slightly less economic growth, we think it’s morally right to use redistributive/progressive taxation as a means to reduce income inequality, because we believe income inequality is intrinsically wrong.”

In fact, it seems to me that there is a strong presumption for progressive taxation. From an efficiency perspective, the correlation between low tax rates and prosperity is, at best, unclear. From a moral perspective, the case for progressive taxation, at least the weak case (“If you make lots of money you should give back more”, as opposed to the strong case “Redistribution/high income reduction is desirable in itself”) seems to have a lot going for it.

These long prolegomena as a way of saying that I am not approaching the question of how progressive taxation should be from the perspective of a rabid anti-tax activist.

Now, if we accept that there is a strong presumption for progressive taxation, what could the case for regressive, or at least flat, taxation be?

For such a case to be convincing, one would have to show that the “spillover effects” of great wealth are indeed massive to overcome this presumption.

There are two basic kinds of spillover effects you can argue: there’s “trickle down” effects, and then there’s what I’ll call “the Model T” effect.

Let’s take trickle down first. The very mention of “trickle down” is likely to elicit hilarity since the idea of beneficial “trickle down” wealth is seen to have been ridiculed as a political ploy to redistribute wealth to the rich.

That being said, the case for trickle down effects cannot be wholly dismissed. The basic economic case for wealth trickling down is unassailable: the rich do spend and invest money and that money, in turn creates lots of economic activity. The question is not whether wealth trickles down, it’s how much, and whether this effect is sufficient to overcome a presumption against progressive taxation.

Based purely on anecdotal gut instinct, I’d say that trickle down wealth is pretty awesome. If you look at the great cities of the world, these are cities where there is a lot of concentrated wealth trickling down. From my perspective, New York is a much greater city than London which is a much greater city than Paris, and it is no coincidence that the better cities have large concentrations of rich people. By “better”, I mean with more cultural amenities, better restaurants, nicer neighborhoods, better business and employment opportunities and all the stuff that makes city life worth living.

When I look at Brooklyn, for example, I see a resurgence of a crafts, shop-class-as-soulcraft economy which I view as a great harbinger of the future of the economy, with small-scale, global-reach entrepreneurship displacing conventional employment. While most of the actors of this economy probably view themselves as embarked on a progressive endeavor, it seems obvious to me that this economy could never have seen the light if there wasn’t a high concentration of rich people with plenty of disposable income in New York, willing to pay higher margins for pickles and artisanal cheese. It’s hard for me not to find a very strong correlation between the awesomeness of a place and the number of superwealthy people who live there.

Same with culture and arts: one strong argument in favor of progressive taxation is the idea that rich people with too much money will engage in negative-sum competition to acquire positional luxury goods, and that it’s therefore better to just redistribute that money away. But while it’s hard to make the case for (lavishly subsidized, by the way) multiple megamansions as socially productive positional competition, positional competition among the superrich also takes the form of patronage for the arts, culture and well-being, either through the for-profit market (ie buying fine art or pursuing angel investments) or through non-profit markets (ie donations to various charities). In fact, one might argue that socially wasteful positional competition occurs among the merely rich, while socially productive positional competition occurs among the superrich: while you might impress a fellow multimillionaire with your huge new house, you’re definitely not going to impress a fellow multibillionaire with your private jet—but you might if you fund lots of charity. If indeed moderate wealth leads to socially destructive positional competition but large wealth leads to socially productive positional competition, the counterintuitive solution to socially destructive positional competition among the wealthy might be to cut their taxes even further.

It also seems obvious to me that what is thought of as deleterious effects of concentrated wealth are really deleterious effects of regulation: concentrated wealth drives up New York rents, but the reason why the rents are too damn high is because of regulation rather than just wealth concentration. Another problem with New York wealth is education: mediocre public schools on the one hand, absurdly expensive private schools on the other; but I hope I don’t have to spell out for frequent Scene readers why it’s the gummit that’s to blame here.

This isn’t by itself an argument against progressive taxation. You’d have to find out whether and how much spending by rich people creates these positive externalities, and whether and how much those externalities are countered by the virtues of progressive taxation. And I don’t think there’s a way to “prove” it one way or another, because these things can be measured and modeled in all sorts of different ways. But my strong feeling is that the benefits of trickle-down tend to be severely discounted.

Ok, so that’s trickle down. The other benefit to low taxes on the rich is what I call the “Model T” effect.

What’s that?

I take it after a quote by Silicon Valley investor Paul Graham: “You need rich people in your society not so much because in spending their money they create jobs, but because of what they have to do to get rich. I’m not talking about the trickle-down effect here. I’m not saying that if you let Henry Ford get rich, he’ll hire you as a waiter at his next party. I’m saying that he’ll make you a tractor to replace your horse.”

I think this is a crucial insight. It’s both accepted and misunderstood.

It’s accepted because, as we saw at the death of Steve Jobs, most people will actually not begrudge the wealth of someone who built wonderful products.

We have a model where, in essence, Steve Jobs is like Adam Smith’s baker: we know he’s selfish, and we don’t necessarily like that he’s selfish, but his selfishness has positive spillover effects that make the world a better place, so we should let him be selfish. Okay.

Another point which is important is that this effect happens because of “what people have to do to get rich.” In some societies, like Nigeria or France, the way you get rich is by getting cozy with people in government who will redistribute advantages to you. In other societies, possibly, the way you get rich is by providing useful goods and services to a marketplace and thereby make the world a better place. (I will leave it as an exercise to the reader to determine in which category financial wizardry falls. (Hint: the former, mostly.)) So it’s very important that we have good institutions and the rule of law and so forth. So the case for free market capitalism is, at minimum, a case for good government and the rule of law. It might even be a case for small government, insofar as the less things are controlled by government, the less unjust advantages it can confer. Maybe.

Anyway, the Model T aspect cannot be discounted. You don’t have to fall into John Galt worship to recognize that entrepreneurship, and more generally a vibrant free market economy, creates massive positive externalities in the form of consumer surpluses and products and services that improve people’s lives.

The question is, again, whether and to what extent high taxes deter this activity. Whether high taxes would have deterred Henry Ford.

This dynamic is related to the rise in income inequality around the world. Let’s stipulate this for starters: income inequality has increased around the world because of some combination of technology and globalization, with taxes playing only a marginal role. (Inequality increased in most developed countries; it increased more in the US, arguably because of taxes, but it increased everywhere.)

If this is true that technology and globalization are driving more inequality, and that we don’t want to destroy technology and globalization, what should tax policy be?

One answer, which I’ll call the Yglesias answer, is that precisely because of this effect there should be more progressive taxation. The “excess” gains that our galtian overlords get is not due to their increased talent or winsmanship, but to a global economic context, and so those excess gains can and should be shaved away by taxation.

But there’s another way to look at it: because of technology and globalization, what happens is that entrepreneurial outcomes follow a power-law rule, where the winners win very big, and the “second-best” or “third-best” actually don’t win out that much. Thus Facebook is worth $100 billion, MySpace was worth half a billion, and everyone else is worth zero.

This can be viewed as “unfair” and therefore more argument toward redistributive taxation. But there’s a flip side to that coin: you want these power-law outcomes.

They’re best regarded as a lottery. The rational response to a lottery is to not play, because the statistical value of a lottery ticket will always be negative. But in the power law lottery, we want people to play, and what reality teaches us is that more people play when the gains are truly outsized: witness the success of the recent mega millions lottery.

In theory, the response to a lottery with huge gains and a lottery with ginormous gains should be the same. In practice, it’s not. In practice, you want the possibility of truly outsize gains to motivate people to join the lottery.

And this is not just true of entrepreneurs—again, we shouldn’t become Galt-worshippers—but more importantly of the ecosystem around them, such as angel investors, institutional investors, risk-taking purchasers and so on.

The pro-market case for progressive taxation paints a picture of a world where progressive taxation leads to a “same, but less” world. Sure, well-meaning advocates allow, you might get slightly less innovation and risk-taking at the margin, but on the whole you’re going to get roughly the same. You will still have Jobses and Zuckerbergs, and you will have a better safety net.

Except that if we’re in a power-law world, that risks being much less true. It’s obviously not a Manzi-level controlled experiment, but it’s striking to look at the fates of technological innovation in Europe and in the US. In the US, there are dozens of publicly traded internet companies worth more than one billion dollars. In Europe, there are three, and up until a couple years ago there were zero. It’s a massive difference. None of these European companies has global reach. All three are part of protected markets; one is Betfair, which deals in online gambling where US companies are prohibited from competing and the other two are Yandex and Mail.ru, Russian companies that no one doubts benefit from implicit government protections. Its not “same, but less.” It’s completely different.

And even though Silicon Valley is, at a first approximation, the only place in the world which invests in innovation, you can make an argument that it dramatically under-invests in innovation, as Peter Thiel among others has argued. Institutional venture capital has dramatically failed at its mission.

When we talk about billionaires and Model Ts, it’s striking to note that most of the new space ventures, for example, are by and large funded, not by institutional funds whose job it is to take risks with other people’s money, but by individual billionaires taking risks with their own money. In theory it is the institutionals who should have more incentive to take bigger risks; in practice, they don’t. But it’s not hard to see that Jeff Bezos might not fund so many space companies and robot companies if he was a “mere” billionaire and not a deca-billionaire.

Solving the Model T problem in a power-law world might indeed require regressive or at least flat taxation.

Now I want to reiterate what I’ve said at the outset: my gut moral instinct is to support progressive taxation; I’m not also convinced by this case for low taxes on the rich, though I think that this case is understated and poorly understood.

There are also many devils in the details: for example, the ways in which the tax code subsidizes Big Finance, which is also a big source of inequality. In my preferred financial system, you might see less superrich financiers, or at least for sure less superrich non-entrepreneur financiers.

Another important devil is that if more inequality is indeed desirable, this increases, not decreases, the impetus to “look after” the “losers”, and to design systems where, to coin a phrase, rising tides lift all boats.