Monday, 9 January 2012

Landsburg getting it wrong about public debt.

   Steve Landsburg says 'When the U.S. government borrows from Chinese investors, it’s exactly as if U.S. taxpayers had borrowed from the U.S. government. (That is, the effect on Americans is the same either way.) Since the assets of the U.S. government are, ultimately, the assets of the taxpayers, it’s exactly as if the U.s. taxpayers had borrowed from themselves.'
   Landsburg is assuming there is no principal-agent problem. Taxpayers are the principal and the Govt is a perfect agent which has no agenda of its own and which does what is in the best interests of its boss- the taxpayer.
    Landsburg says- ‘debt, by itself, is not a terribly compelling reason either to cut entitlements or to raise taxes’. 
Even assuming zero agency dilemma, I think this statement is misleading. A Govt. debt level which is foreseen and budgeted for and universally accepted would not, in any meaningful sense, be Govt. debt any more than Shareholder Equity is considered debt in a limited Company. Indeed, the Govt can get such debts off its books simply by selling off the rights to its future income streams immediately. Under perfect information, no uncertainty or agency hazard,zero transaction costs etc. the Govt is always just one instantaneous market transaction away from having a balanced budget. It can't meaningfully be said to carry debt at any time.
However, where debt levels diverge from what they were predicted to be (which can only happen if some risk was overlooked or some information not properly discounted) then a compelling reason arises to cut entitlements and raise taxes just the same as transpires when Balance Sheet weakness compels a Business to cut dividends or restructure or issue a call on partly paid up Equity.
Landsburg concludes by saying- ‘As Krugman point out, this is not to say that all government debt is harmless. It is to say that if it’s harmful, it’s harmful for different reasons, and probably to a much lesser extent, than is commonly believed.’
I am curious to learn more of that transcendentally mystic state in which even Landsburg begins to find Govt. Debt to be harmful. 
I believe,  he will then have to invoke either uncertainty, or agency hazard or both. IN other words, he first banishes reality so as to say one silly and counter-intuitive thing but then, so as not to appear a complete nut-job- has to bring back that bogeyman to justify saying the reverse. Landsburg is a reverse Talmud Golem. He won't print my objections on his website- well, he did a couple coz he thought I didn't know from OLM- but when the going got tough he just shut me down and carried on regardless. 
Meanwhile, this is Landsburg being misguided by Ricardian Equivalence to come to the conclusion that Public debt isn't really a serious 'burden on our grand-kids'.
'Murphy objects to formulations along the lines of “government debt is not a burden because we owe it to ourselves” and offers a parable that he thinks illustrates all the key issues. I agree that his parable illustrates all the key issues, so let’s review it — and see what it really illustrates.
In Year One, Abraham (who is old) owns an old apple tree and Isaac (who is young) owns a young apple tree. Each tree delivers 100 apples to its owner. Shortly thereafter, Abraham’s tree dies and Abraham follows suit. In Year Two, Isaac’s tree delivers him another 100 apples, and then both the tree and Isaac die.
Now in Year One, Abraham’s government decides to give him a present of 10 extra apples, which it borrows from Isaac. As a result, Abraham gets to eat 110 apples and Isaac eats only 90. In Year Two, the government owes Isaac 11 apples (including interest). It gets these apples the only way it can, by taxing Isaac. Therefore Isaac pays 11 apples tax, receives an 11 apple bond payment, and eats 100 apples. Bottom line: The government policy has increased Abraham’s lifetime apple consumption at the expense of Isaac’s. Therefore, says Bob, it’s clear that the government’s debt constitutes a burden to Isaac.
Fine. Here’s my counter-parable. In Year One, Abraham’s government decides to give him a present of 10 apples, which it gets by taxing Isaac. In Year Two, the government does nothing. The government policy has increased Abraham’s lifetime apple consumption at the expense of Isaac’s exactly as in Bob’s story. The Landsburg-Isaac feels exactly the same burden as the Murphy-Isaac, even though there is no debt in the Landsburg world. Therefore debt cannot be the source of Isaac’s burden.
Indeed, the source of Isaac’s burden, plain and simple, is that his government decided to transfer resources from him to Abraham. Whether they do this via debt or via taxation is as irrelevant as whether they deliver the apples to Abraham by truck or by train.
If the apples are delivered to Abraham in a wheelbarrow, one could, I suppose, blame everything on the wheelbarrow and talk about the “burden of the wheelbarrow”. And in some very contorted sense, one could defend that position. But why would you want to?
Why is this silly? Well, we have a story about a guy who goes and buys Govt bonds which can only be retired by a lump-sum tax upon him self! In other words, buying bonds today is what determines how much tax you're going to pay tomorrow! Who would do that? Certainly not someone with Rational Expectations. But Ricardian Equivalence (which is what has misled Landsburg into buying into Murphy's  parable) is crucially dependent on people having Rational Expectations and being free to costlessly enter into financial transactions.
Ask yourself this question- suppose your kid comes to you and asks you to lend him 1000000 pounds. You ask how he is going to repay it. He shrugs his shoulders and suggests you just raise his allowance and deduct from that. Would you lend your kid the money?
In real life, the Govt isn't your 5 year old kid, nor is it a perfect servant. Murphy is assuming that the Govt. can levy a lump-sum tax at any time. He further assumes that people only know the lumpsum tax in the current time period, not what it will be in the next time period and also not that the lump-sum tax is dependent on how many bonds the Govt has to retire. But Murphy's assumptions violate Rational Expectations.  His parable is simply silly. He doesn't have a subsistence lower bound nor a satiation upper-bound- very silly in economies (like ours) where zero Marginal Cost commodities exist.
More ludicrously yet, he doesn't understand that if bonds are tradeable then no borrow/tax mix can hit a specific redistribution target - i.e. the Govt would have to be irrational to be using this policy instrument. In so far as, under his assumptions, present transfers are funded by tax payers hundreds of years in the future- this only happens because the Govt has created a new Econ good- viz. a hedge (not perfect) against starvation in the future, or method of saving- Still, Murphy is a happy camper coz his ludicrous model lets him have think he's proved something- even though it's the opposite of what he needs to show.

Murphy aint stupid, but, assuming people are stupid in a pre-specified way in your model lets you show anything you like.  And that's stupid. Landsburg doesn't point this out. No, because he has a wheelbarrow to bring in. The Govt takes Isaac's apples to give to Abraham in a wheelbarrow. Notice, if Isaac was stupid enough to have bought bonds with apples, he has only himself to blame. Not the malevolence of the wheelbarrow, but his own stupidity in lending to the Govt has harmed him. If Isaac learns to 'just say no' there is no lump-sum tax for him to pay in future years. He no longer suffers from 'fiscal illusion' because he now has Rational Expectations. True, the Govt can just take stuff from one guy and give it to another but, if agents are risk averse there will be a coalition which hedges against lump-sum tax/entitlement uncertainty such that within that coalition Govt. actions are immediately cancelled out.  Under plausible assumptions, this coalition will dominate and crowd out the Govt by making an equal or better offer to anyone with whom the Govt wishes to interact.
Landsburg really has no answer to points like this. He feels that, because certain sorts of assumptions are standard in 'over lapping generations' models of Murphy's sort, there is no need to use one's common sense. True, common sense isn't common, but consistency should be second nature to a Math maven. Landsburg isn't applying Rational Expectations consistently- incidentally, Rational Choice theory with Schutzian ideal types get round his objection that people in an identity class, in Murphy's OLG model, don't know what proportion of their bond purchases will be clawed back. This doesn't matter, so long as they have enough cognitive capacity to achieve a Rational Choice Verstehen- similar to my not buying into a pyramid scheme though not being sure whether I mightn't be one of the lucky ones who get rich off it. 
Landsburg says- 'deficit finance does not allow an entire generation to increase the burden on its descendants beyond what it could have done anyway (and in that sense is nothing like a time machine). It only allows some families to increase the burden on their descendants. The greatest damage one generation can inflict on the next is to consume everything in sight, and this is always possible without deficit finance.'
This is not true. Deficit finance can lead to hyperinflation, the collapse of the market for Govt bonds, collapse of confidence in the regime, the rise of extremist political parties, War, Genocide, Slavery.
Lansburg thinks the only way we can fuck up our grand-kids is if we eat up all the non renewable resources and let all capital goods depreciate away to nothingness. But, countries which have been invaded and denuded of all productive capacity and natural resources can still bounce back if they inherit Institutions with a reputation for, or even new found dedication to,  thrift, transparency, zero agency hazard etc. 
Indeed, under plausible assumptions and assuming factor mobility, they may well stand taller than the rest- as happened to West Germany and Japan- within a relatively short time.
To misread Ricardian Equivalence as justifying Landsburg's indifference to the level (especially in that it was unforeseen) of Public debt is to succumb to a particularly witless pedagogic, rather than truly academic, availability cascade.


Rajiv said...

I think you're missing the point. Landsburg et al are auditioning for Fox News, or to be Sarah Palin's go-to-guy, at a time when the Republicans are engaged in a stupider-than-thou hiding to nothing.
BTW, I thought you didn't liek Rational Choice theory.

windwheel said...

@ Rajiv- well, D'uh. But, here's my question. If interactions in Games create novel Utility fn. arguments such that what one is really looking at is a sort of fuzzy functor space is there something one can say in advance about their degenerate forms?