Two Cheers for Regressive Taxes

David Leonhardt had an excellent article the other day on the long-run need for more tax revenue than can be generated merely be allowing for the expiration of the Bush tax cuts. He makes mention of something called Wagner’s Law, which is related to what I was talking about here, and that basically states that as society gets wealthier it demands more and more of the sort of services that need to be provided by the government. That means that over time as the economy grows the share of the economy that becomes tax revenue goes up. This not only meets public demand for more public services, but provided the tax rate doesn’t go up too rapidly, people’s after-tax income is still rising so private consumption goes up along with public services. And as he observed “the problem can’t be solved just by taxing the rich” largely because there aren’t that many very rich people.

The moral of the story, I would say, is that the left will need to embrace some revenue enhancements that are not-so-progressive in their distributional impact. I think that means, in the first instance, taxes on behavior that’s undesirable—carbon taxes, alcohol taxes, congestion pricing, market pricing of parking, etc. (obviously some of this stuff would be for local government rather than the federal government)—and in the second instance the dread VAT.

But why would progressives want to embrace non-progressive revenue sources? Well, fortunately Lane Kenworthy did an excellent post on this a year ago that contains graphs I can steal. The first chart shows that if you look around the one at what it is countries do to mitigate income inequality, nobody is substantially equalizing things through the tax system, but many countries are substantially equalizing things on the spending side:

Not that progressive taxation is a bad thing, or meaningless in the contribution it makes, but clearly insofar as direct public policy interventions (as opposed to things like wider distribution of educational attainment) are going to reduce inequality, it needs to be done on the spending side. Now this raises the question how do you get the spending side to do more? Is it by “means testing” existing programs and creating new small-bore “targeted” programs aimed at the neediest? Well, not really:

The most important thing is to just have lots of tax revenue. Public expenditures are pretty progressive in their impact everywhere, and the difference between a very progressive and a not-so-progressive system is mostly that the more progressive ones are bigger. So while liberals have no reason to give in to conservative demands to make the existing revenue scheme less progressive—by adopting a flat tax, say, or replacing the income tax with a consumption tax—there’s very good reason to basically be looking for revenue by any means necessary. If it’s easier, politically, to get some center-right politicians on board for new consumption taxes than for higher income taxes, then it’s incumbent on progressives to walk through that door and take the revenue. At the moment, of course, that’s not an open door so there’s really no need to worry about it either way. But this is the kind of choice you can imagine progressive politicians and/or activists facing at some future point, and I think it’s important to start building understanding of the structure of the choice.