A Bridge Too Far?

There is much current angst on the difficult problem of how to escape a liquidity trap. Paul Krugman points out that in Japan, the ratio of debt to GDP is growing, leaving little room for a further tame fiscal expansion. He favors something more aggressive.

Tony Yates argues instead for a helicopter drop. Print money and give it to Japanese citizens. The benefit of that approach is that it does not leave the government with an increase in interest bearing debt. 
Simon Wren Lewis looks more closely at the technical aspects of this idea.

What are the differences between aggressive fiscal expansion financed by debt creation; and printing money and giving it to citizens? There are two.

First, an aggressive fiscal expansion, as envisaged by Keynesians, would be spent on infrastructure. A money financed transfer would be spent by citizens.

Second, an aggressive fiscal expansion, as envisaged by Keynesians, would be financed by issuing long term bonds. A money financed transfer would be financed by printing money.

While infrastructure expenditure is sorely needed, at least in the U.S., I see no reason to give up on sound cost benefit analysis to decide which projects are worth pursuing and which are not. That’s why I favor giving checks to citizens over building a bridge to nowhere.

Once we decide how the fiscal expansion is to be distributed, we face the second question: how should it be financed? Print money? Or issue long term debt. Standard Economic models tells us that it doesn’t matter. At the zero lower bound, money and three month T-bills are perfect substitutes. And financing expenditure by three month T-bills has the same effect as financing it by thirty-year bonds because the composition of the government’s liabilities is supposed to be irrelevant. That of course, is nonsense. The composition of government liabilities matters. And it matters a lot.

Why does the composition of debt matter? Because the asset markets are incomplete. Our children and our grandchildren cannot participate in asset markets that open before they are born. And none of us can sell our human capital or buy the human capital of others. Once you realize that the composition of the governments portfolio matters, it is a short step to recognize that it is all that matters.

Why be wary of building bridges that are financed with 30 year bonds? Because the yield on these bonds is low; but it is not yet zero. A big increase in public sector borrowing, at the long end of the yield curve, will drive up rates and crowd out some private investment. A big increase in public sector borrowing at the short end of the yield curve will not crowd out private sector investment because rates at the short end of the yield curve are currently zero.

That observation suggests a third alternative to building bridges or to a helicopter drop. Buy back long term government debt and refinance it by printing money. That strategy would, one hopes, lower yields at the long end of the yield curve and stimulate private companies to invest in new capital projects.

I prefer private sector investment over government sector investment. But there are also good arguments for more public infrastructure projects. Build a bridge if it is needed; but make sure that it goes somewhere first. More importantly; finance the project by printing money: not by issuing thirty year bonds.