Ed Driscoll, subbing for Insty, links to this pjmedia post–The Lovitz Curve. It’s cute and all that, but commits a fundamental and all-too-common error. (Read on only if your tolerance for wonkery is high today.)
Here’s the famous Laffer Curve, which depicts one stylized version of the necessary relationship between tax rates and tax revenue. What’s obvious is that no revenue at all is generated by either a 0% tax rate (duh!) or a 100% tax rate (who the hell works for zero take-home pay? Besides bloggers, that is.) It follows as a matter of logic that there’s a tax rate somewhere between these extremes where tax revenue is at a maximum. (It needn’t be exactly in the middle, as depicted below, and is probably somewhere around 60 or 70 percent in the US.)
The mistake made by “Zombie” in the pjmedia post is to call the revenue-maximizing tax rate the “optimal” rate. It is properly called the “revenue-maximizing tax rate,” and is in fact a disastrously stupid tax rate that is way, way beyond any likely value of the optimal tax rate.
The reason the Laffer Curve gets less steep as revenue increases (in the region to the left of the maximum) is that people increasingly take measures to avoid taxes as the tax rate rises. In econojargon, the economic losses incurred by these tax-avoidance measures are called “deadweight costs.” They are “deadweight” because they represent the losses from the decline in mutually-beneficial economic transactions. Such costs are an unavoidable consequence of taxation.
Levying taxes, therefore, involves a tradeoff between the deadweight costs from raising tax revenue and the benefits of the government activity that the tax revenue is required to finance. The optimal tax rate is the one that strikes the correct balance between these costs and benefits. It is the rate at which the increased deadweight costs of another dollar of tax revenue are exactly offset by the amount by which the benefits of another dollar of government spending exceed a dollar.
Here’s a numerical example: When the tax rate is low, deadweight costs are negligible; a dollar of government revenue costs private individuals scarcely more than the dollar they’ve lost–maybe a penny or so. Then any use of that dollar that’s worth at least $1.01 to people is a justification for this level of taxes. As the tax rate rises, the deadweight cost of taxation rises much faster. So once the tax rate hits 20% or so, the deadweight cost of taxation is not at all negligible; when you get to 40%, the losses can be staggering. Careful estimates from New Zealand data indicate that raising the top income-tax rate from 39% to 40% is only justifiable if the benefits from additional government spending are worth $8 for every additional dollar spent. Does anybody think that’s even remotely possible?
By the time a government has reached its revenue-maximizing tax rate, there isn’t anymore revenue to be had from higher rates. But there are higher deadweight losses, which means that the incremental cost of revenue at the top of the Laffer Curve is infinite. Just below that point, it’s not infinite–just unbelievably high.
In a country where peoples’ responses to tax rates are similar to those of New Zealanders, even a top rate of 40% is pretty certainly indefensible, if not just plain crazy. In all likelihood, nothing much above 25% makes much sense at all.
So the next time your lefty friends yap about how we should cover our massive deficits by raising taxes, tell them why this would be a terrible idea even if it could be done.