Taxes and Tradeoffs
January 22, 2014 3 Comments
I got several interesting emails about my post on Casey Mulligan’s paper on the impact of the ACA on labor markets, so I thought I would elaborate a bit.
If you tax something, you get less of it. So, there are only two reasons I can think of to levy a tax.
- you want less of something (excise taxes on tobacco)
- you want to raise money to redistribute via government to obtain an outcome that markets didn’t reach
The second point means there is always a tradeoff. Is the outcome I seek (increased health insurance coverage, a larger military, minimum social security benefit for the elderly, interstate highway system, cancer research, etc) worth the distortion or inefficiency that I will inevitably create by taxing and redistributing money worth it? Note that I am saying that every dollar of government spending is inherently redistributional. If not, there must be a tax that is levied and then returned in the exact share to those who paid it. I don’t think such a tax/spending pattern exists. This doesn’t mean you cannot dispute the legitimacy of a given tax or of a spending priority, but that first and foremost that is an argument that the distributional outcome to be obtained by governmental redistribution is not worth it, not that it is redistributional.
Casey Mulligan focuses on the marginal tax rate of labor income, and says that the labor market will not return to pre-crash levels until we lower this marginal rate. His work shows an increase in this tax rate due to the expansion of unemployment insurance during the economic crisis, and from the provision of subsidy to purchase health insurance in the ACA; all of this on top of the explicit tax code. At the conference, Casey noted that doesn’t mean that extending unemployment insurance is necessarily the wrong policy at a given point. This surprised me given my reading of his blogging, but it showed that he was engaging in the tradeoff; something could be warranted in the short run, that exacerbates a longer run concern. There is plenty of art involved in getting questions like this right.
What would be useful for policy making would be to take the detailed data/programming machinery that Casey has assembled and model different means of financing health insurance coverage expansions, and showing how they may differentially impact the marginal labor income tax rate. In particular, quantifying the degree to which the intuition in creating a policy like the ACA “we have to reduce the cost of subsidy” actually increases the impact on incentives for full time or part time work. What is the crossover point at which more subsidy is worth it when evaluating the policy from a labor market perspective?
Asking these questions and being interested in the answers doesn’t mean that you you to buy into the degree to which labor market behavior will be changed due to the ACA. The predictions from Casey’s paper should be made, but in the end, that is an empirical question that we should have more information on in the coming months and years.
update: I forgot I wrote this in November.