Our tax system (over which I predict we will be having a large debate in the near future) is more or less a debate over two questions, one technocratic and the other values-based.
The technocratic question: what is the efficiency of various forms of taxation?
The values question: how and how much should we redistribute income?
The second question is nigh unanswerable– Rawlsians would say that income redistribution should be almost 100%. Utilitarians (and most economists) would come up with utility functions for citizens and solve a Mirrleesian tax model. Libertarians would say no income redistribution should be done at all, and that taxes should be levied on the basis of benefits received.
The first question however, has an answer, and it is that raising taxes on 1) individuals with high income, 2) individuals with already high marginal tax rates, and 3) savings, all do more harm to the economy than raising marginal taxes on the poor. The poor do not change their behavior much in response to a tax increase, they’ve just got to soldier on. High income earners, by contrast, will cut back their working hours and take more time to enjoy their earnings if you try to take a slice from them.
The evidence on this point seems pretty overwhelming– I’ve yet to see any analysis of the elasticity of taxable income that estimates the elasticity of high earners as less than that of low earners.
There are three typical responses given by liberals.
The first, which is reasonable, is that poor people deserve more income. In other words, yes, progressive taxes reduce the economic pie… but it is worth some losses to cut the pie more evenly.
The second– which is deeply flawed but will not be discussed at length here– is that low taxes on the poor will increase demand, and thus drive the economy. In other words, they believe that the Keynesian paradox of thrift– a theory developed to explain recessions– holds even during normal economic periods. In short, the paradox of thrift is a temporary phenomenon, and thus can only serve as a temporary justification of policies.
The third, which is what I’d like to discuss a little and put to rest, is that we need to redistribute income, because if we don’t, poor people will commit crime or foment revolution or somesuch. In other words, if you’re a rich person, and you don’t care about the poor, you should still pay more in taxes, because if you don’t, you’ll get robbed.
This argument defies comprehension for three reasons.
1) It simply isn’t true.
Here, check for yourself:
Notice something? Income inequality has been steadily rising since about 1970. Over that same period, violent crime has fallen nearly in half. A decade ago it might be possible to cherry pick data and come to the conclusion that income inequality fueled crime, but these days virtually any time series panel regression will show no positive relationship between the two variables.
2) It’s hypocritical.
Every time there is a debate on an issue of national security, liberals tell us that Republicans are the party of fear, that they win elections by scaring voters and trumping up threats. But who are the fear mongers? “You better give money to the poor else they’ll mug you” is a threatening message.
3) It’s poor politics.
If you’re going to convince people that the poor need our support, you’re not going to get very far by painting them as some sort of crime prone extortionists, ready to rob your grandmother if they don’t receive their monthly bribe from the government. That’s exactly the sort of characterization that will compel people to avoid redistributive policies out of spite.
There’s plenty of room for debate on income redistribution. But let’s not pretend that such a policy would be win-win.