Saturday, October 13, 2012

Government Debt & Inter-Generational Distribution

There is currently a big debate between on the one hand Nick Rowe of the Worthwhile Canadian Initiative (who has written about it first here and then here and here) and several leftist bloggers including Brad DeLong, Dean Baker and Paul Krugman.

The basic argument of the leftists have been that government debt doesn't create a burden for the economy because one person's, or institution's (like the government) debt is another person's or institution's asset and so higher debts also means higher assets. And this also means that for the nation as a whole there might not be any inter-generational redistribution (though it is likely to create some intra-generational redistribution) as the next generation doesn't just inherit more debts, it also inherits more assets.

Rowe's counter-argument is essentially that the next generation might not inherit more assets if the previous generation sells their bonds and consumes the proceeds rather than pass it on to the next generation. Assuming that the buyer is someone from the coming generation, this will indeed need inter-generational redistribution as the older generation gets to consume more than whey earn, while the younger generation must reduce their consumption to get the assets matching the debts that the older generation passes on to them.

Krugman's response to this is to concede that debt financed spending can redistribute between generations, but to argue that this doesn't mean that the country as a whole will be poorer. So who is right? Actually both sides are right, sort of, when it comes to the direct distributional effect. Rowe is certainly right that deficit spending that goes to consumption will redistribute from the generation that consumes that buys the bonds that finances it to the generation that consumes. Krugman is however right when he notes that this still doesn't make the country poorer because this just mean that during the period that the debt is created, the group that buys the bond will reduce their consumption by the same amount as the generation that receives the benefits will increase their, leaving overall net savings and therefore also net worth and economic output unchanged.

Nick Rowe attempts at creating another impression with an example of the government buying apples from cohort A in an axchange for an IOU that they receive, and then gives the same people those apples back as a transfer payments. Cohort A then eats the apples and sells their IOUs to cohort B. Cohort B then sells their IOUs to cohort C and uses the proceeds to buy apples. At that point, the government decided however that the debt level is unsustainable and decides to tax cohort C to buy the IOUs back. But all this shows is that there was a redistribution between the generations. At no point was total consumption changed. At point 1, when cohort A received that transfer payment they were forced to first give the government the apples in exchange for an IOU they later received as a transfer payment. At point 2, when cohort A sells their IOU to cohort B, they can consumer more, but cohort B are forced to consume less by the same amount to buy the IOUs. Similarly, when cohort B at point C sells the IOUs to cohort C, cohort B can increase their consumption but cohort C must decrease their consumption by the same amount. And finally at point 4, when cohort C is taxed tp finance the repurchase of the bonds, this won't affect total consumption either because the negative cashflow for cohort C from the extra taxes is matched exactly by the positive cashflow from selling the bonds. At none of these four points in time were total consumption either higher or lower. It is true that cohort A got to consumer more as a result of this procees while cohort C got to consume less, but since cohort A:s extra consumption happened at the same time when cohort B reduced theirs and cohort B:s compensated extra consumption happened when cohort C consumed less, total consumption at all points in time was never changed.

 Basically the dispute seems to be what is most important: the issue of whether or not debt redictributed between generations, which it clearly does, or whether or not it at any point in time changes aggregate net wealth of a nation, which it doesn't, at least not directly, assuming debt is financed domestically.

Note however, the disclaimers "directly" and "financed domestically". If foreigners finance the deficit spending directly or indirectly, then the country as a whole will become poorer because the increase in debt is larger than the increase in assets held by residents. And to the extent it crowds out, or promotes, investments it will reduce or increase growth. Similarly, to the extent this process affects incentives it could also reduce or increase growth.