Month: February 2005

Robert Shiller writes:

According to a study by the Insurance Information Institute, expenditures on non-life insurance in 2003 amounted to only 0.83% of GDP in Indonesia, 1.19% of GDP in Thailand, and 0.62% of GDP in India, compared with 5.23% of GDP in the United States.

And his bottom line?

Foreign aid is no substitute for insurance. Charity inspires, reassuring us of our humanity, but it is often capricious. You wouldn’t want to rely on it. Indeed, when deciding how much disaster aid to offer, countries often seem to be influenced mainly by their leaders’ concerns about how others will view them. Charity responds to attention-grabbing events, often neglecting less sensational disasters.  Insurance, on the other hand, is a reliable and venerable institution, its modern form dating back to the seventeenth century.

Could the micro-credit revolution be followed by a micro-insurance revolution?  Shiller argues valiantly but I am not convinced.  Economists miss one of the biggest problems with insurance.  We are blinkered by adverse selection models, which imply that the dangerous prospects most want to buy insurance.  The opposite is more often true.  If you are an irresponsible driver, you are likely to be irresponsible in other spheres as well and not buy auto insurance.  On the whole many insurance markets show positive rather than adverse selection.  In a development context, this means that the people who most need insurance will be the least likely to buy it.

Rachel Soloveichik writes:

I completely agree [with my earlier post] that parents aren’t perfectly altruistic toward their children, and parents are happier when their children are better behaved rather than happy.  But if society shifts toward giving more power to children parents are much less likely to have children.  The small effect on children’s utility from more toys may be hugely outweighed by fewer children being born.  That’s probably part of the reason religious parents have more children, because they have a set of norms that justifying pushing their children to good behavior rather than being happy.

My question is where the limits lie, once we make this a moral question.  Say you could cut a deal with a fetus or would-be fetus.  You agree to raise a child, under the condition that it remain your slave for life. You also make sure that the life of that slave remains better than no life at all, although just barely.  Should we, as believers in the Pareto principle, approve of such deals?  I’ll say no, which is one reason why I don’t put much stock in Nozickian libertarian rights.  But I find this view difficult to defend.  After all, everyone is better off (forget about secondary consequences, if you wish to make the hypothetical example airtight).  Should we allow parents, or corporations, for that matter, to make such deals?  And would we need actual consent from the child-to-be, or is it enough to recognize that in expected value terms the child-to-be would likely prefer to be born, despite the onerous terms of the deal?

I suspect that yes, it is better for the deal to be made.  More happy people is a good thing, all other things equal.  But it is also better for the deal to be broken, ex post.  Some ways of treating people are just wrong, even if they were agreed to ex ante.

Oddly, perhaps libertarians should welcome this conclusion.  Arguably their ancestors consented to rule by the U.S. government when they moved to this country.  Or perhaps you offer a kind of implicit consent from your consumption of public goods or from your dealings with broader society, nearly all of which favors the idea of government power.  But libertarians still can object to their current treatment, no matter what explicit or implicit agreements were made in the past.

What if parents agreed to have kids, but only on the grounds that those kids would not freeze their social security benefits?

The blogosphere has been involved in a huge debate over the Baker-Krugman position that high stock returns are inconsistent with a low growth rate of the economy (thus either social security privatization won’t work or, if economic growth is high, it’s unnecessary).

Brad DeLong offers a useful summary laying out the possible scenarios.  I think, however, that he and everyone else have missed one scenario which may be important  (see also Tom Maguire and Andrew Samwick).

Everyone has assumed that economic growth and social security privatization are independent variables but suppose that social security privatization raises economic growth then one can consistently assume that economic growth will be low under the current system but stock market returns will be high if we privatize.

Can social security privatization raise economic growth? Certainly the answer is yes.  It’s a big reform which (done right) can significantly increase savings and reduce labor market distortions.

Can the increase in economic growth be enough to solve the Baker-Krugman problem?  I think not in the long run but I am not so sure about short-run dynamics, even a small increase in growth rates can increase prices significantly.  I turn this one over to better macro-economists than I.

Last week I examined the ethics of gradually indexing social security benefits to prices rather than wages.  It has never been obvious to me that the elderly are the deserving recipients of greater largesse.  Don’t we have better ways of spending the money, such as on biomedical research, immigrants, or foreign aid abroad?  Why might you think that the elderly deserve a greater share of resources in society?  Here are some options:

1. We should be egalitarians, and evaluate individual well-being in a "time-slice" rather than lifetime sense.  Yes the elderly have lived a long time, but right now they are the ones closest to death and often poor as well.  We should transfer resources to those who are not so well off.

2. The relevant political alternative is lower marginal tax rates for the well-to-do.  Transfer resources to anyone but them, whenever you can.

3. True, the relatively wealthy American elderly are not our highest priority, all things considered.  But in political equilibrium taking care of the elderly is unlikely to prove a substitute for other forms of charity.  By cementing our self-images as being caring people, it renders other forms of charity more likely, not less likely.

I suspect that some version of #2 is what motivates most liberals.  But consider a few points.  First, is this the relevant political trade-off today?  Can we not imagine a Republican-led Congress, for better or worse, determined to make the Bush tax changes permanent?  Alternatively, you might think the tax changes are headed for repeal in any case, bringing us back to no trade-off.  Might the real policy trade-offs involve Medicaid, immigration reform, and discretionary spending, not to mention the looming fiscal disaster known as Medicare? 

Second, marginal tax rates for the wealthy cannot be the relevant political trade-off forever.  It would be quite simple to favor repealing the Bush tax changes, and also wanting to price index social security benefits.  Note that price or wage indexing is supposed to continue for the indefinite future.  So we could arrive at your ideal tax policy — whatever that is and whenever that happens — and still face the social security issue.

I actually find #3 to be the stronger argument.  Note that the welfare-exorbitant Scandinavian states also spend relatively high amounts on foreign aid per capita.  #3, however, does suggest that the cost of helping the truly deserving is much higher than we had thought.  Not only do we have to spend the money, but we must invest significant resources in bringing around our social consciences.

Addendum: Here is more from Alex.