Educating Dean Baker

I never expected to type that. I have great respect for Dean Baker who, among other things, convinced Paul Krugman we were in a housing bubble. He is very smart and writes about important issues. But he has recently written two posts which contain the same elementary error.

I don’t want to be rude, but the posts make me think of Robert Lucas’s tirade against Christy Romer.
Baker wrote

Steven Rattner remains convinced that handing future generations trillions of dollars of government bonds imposes a burden on them and is very unhappy that I don’t see things that way. Let’s try this one more time.


At some future point, everyone who owns this debt today will be dead. They will have no choice but to hand this debt on to members of the next generation, either their own heirs or someone else’s.

Baker asserts that if someone owns something, he must keep it till he dies then leave it to his heirs. In fact, it is also possible to sell it and consume the proceeds. My objection to Baker’s post really is just that elementary.

I suppose one might argue that I have missed something. I don’t think I have and explain at gruesome length after the jump.

First I stress that Baker didn’t just assert that something will happen, he claimed that something must happen “have no choice.” That means his claim is a mathematical claim which can be refuted by an example. The example is presented in

Diamond Peter (1965) “National Debt in a Neoclassical Growth Model” AER LV December 1965 pp 1126-1150.

The way in which national debt can fail to crowd out capital formation is that the current generation can (and will) choose to leave bequests. If they do this for purely altruistic reasons and if they have rational expectations, then public debt can affect the economy only due to distortions from the taxes needed to service this.

Note that, if Baker is right, then tax cuts do not stimulate aggregate demand. If he is right that, if there were non-distortionary taxation,, deficits would do no harm when the economy is not in a liquidity trap, they would do no good when the economy is in a liquidity trap (although public spending would stimulate).
OK now I will address the argument. I will always discuss what the current generation can do. In particular I will ask what happens if we all decide that those yet to be born will get no financial assets except as wages for work performed (to add a hint of realism I assume that they will be given milk and such when they are newborn but no cash no bonds and no stocks). I don’t think we are going to suddenly turn that selfish. I am asking whether we have that choice or if we have n o choice but o leave bequests.

As always I start from a very simple model with very extreme assumptions.

The extreme assumptions are that our days are numbered and that we know the number. That is people know exactly when they will die. Such people can arrange to have zero wealth when they die. They can sell bonds to other people who will live longer and use the money to buy and consume goods and services. If people know when they are going to die, they just don’t have to leave bequests. Aggregating, the current generation sells the bonds to future generations and consumes more than it would if there were no such bonds. This is mathematically possible and would happen under the extreme assumptions (which are basically the assumptions Diamond made).
OK now assume that death is unpredictable except also assume that there will be plenty of not yet born people working before the currently alive die (still extreme). Those who don’t want to leave a bequest can sell their bonds to the young and buy annuities. These are really existing financial instruments which pay a stream of income so long as the owner is alive and then are worthless. The market for annuities is tiny, because people with assets want to leave bequests not because the have no choice.

Well that is still unrealistic. Death can come at any time. But if this generation is nasty enough, we can name an heir in this generation. That way money only passes to those not yet born when the very last person no alive dies. Long before that, the now living people can sell bonds to the not yet born and buy annuities.

It is absolutely possible for us to make public debt a burden on future generations. The math works fine. There have long been models in the literature (most with the super strong assumption) which have the property that public debt makes future generations poorer. The distinction between models with infinitely lived agents and models with overlapping generations is fundamental in the macroeconomic theory literature.

Theory might be pointless but not because we can tell that something is logically impossible even though it would occur if a formal mathematical model were reality. Baker is claiming we know something from logic alone (always wrong) when, in fact, we can tell from logic alone that his claim is false.

OK no one will read this far so I can be as rude as I want. Baker’s two recent posts remind me of Cochrane and Lucas. The funny thing is that the concept of Ricardian equivalence is involved in both huge errors. In his diatribe against Romer, Lucas made the exact same error Baker made asserting that we know that there is Ricardian equivalence. That it is not an implausible theoretical possibility but a necessary truth. The similarity is that in all three cases, economists who have done good work made elementary errors. All three asserted that economists know something to be true, when elementary lectures are devoting to proving that it is false.

(Dan here….some minor formatting was done for easier reading)