In yesterday's post, I discussed effective marginal tax rates and the marriage penalty in the US social security system. The social security system can create incentives (and disincentives) for activities unrelated to working and income (in that case, marriage). However, most of the time when we talk about incentive effects in social security, we are talking about decreased work incentives.
I was reminded (by a note I left myself) that Abhijit Banerjee and Esther Duflo discussed these effects in their book Good Economics for Hard Times (which I reviewed here). In particular, Banerjee and Duflo note that there are both income effects and substitution effects associated with the social security system. Specifically:
...for people near the point between being takers from and payers into the system, there is potentially a strong disincentive to work. In other words, in addition to the income effect (I do not need to work if I have enough money to survive on already) that most policy makers worry about, such schemes can have a substitution effect (working is less valuable since what I make in extra income is taken out as reduced welfare payments).
The latter point relates to the effective marginal tax rate (the amount of the next dollar of income a taxpayer earns that would be lost to taxation, decreases in rebates or subsidies, and decreases in government transfers (such as benefits, allowances, pensions, etc.) or other entitlements). To see how these disincentives work, consider the diagram below. The income curve shows the distribution of income without any transfers. The poverty line (representing the minimum level of income necessary in order to lead a comfortable life) is M*. Now consider a perfectly targeted transfer, that would raise the income of any person whose initial income is below M*, exactly up to M*. People with initial income of M* or higher receive no transfer at all.
With this perfectly targeted transfer, poverty would be eliminated. However, what we are interested in is the incentive effects. Consider a person with income of M0. They initially have zero income (probably they are not working at all), and their income is raised to M*. They do very well from the transfer. Now consider a person with income of M1. Their transfer is less, and their income is raised to M*. However, they were working (perhaps part-time) and earning M1. Comparing themselves to the person with income of M0, the person with income of M1 might realise they don't need to work so hard and can still end up with an income of M* after the transfer. This provides a disincentive to work. This is the income effect described by Banerjee and Duflo. The person with an income of M1 couldn't be incentivised to work a bit more either. Every dollar of income above M1 is eliminated by a reduction in the perfectly targeted transfer (the effective marginal tax rate is equal to 100 percent). This is the substitution effect that Banerjee and Duflo described. Now consider a person with income of M2. They are earning more than M*, but they might also look favourably at the person with income of M0, and decide to leave work. The disincentive effects don't just apply to those below the poverty line, who are initially eligible for the transfer.
Finally, Banerjee and Duflo probably understated how wide the disincentive effects are. They note that they apply "for people near the point between being takers from and payers into the system" (which is people near the income of M*). However, I think they probably apply to everyone with income below M*, as well as some people with income higher than M*.
I noted in yesterday's post that the custodians of the social security system need to understand the unintended consequences that the social security system creates. They also need to understand how the system affects work incentives.
Read more:
No comments:
Post a Comment