How To Build a Recovery

by: Felix Salmon

Larry Summers has a long and pompous article in the FT today on building a financial recovery. If you can slog your way through the awkward constructions ("consideration should be given to whether the government should establish a mechanism for purchasing assets from stressed banks in return for warrants or other consideration"), you're likely to end up thinking that his diagnosis is pretty accurate, but that his proposed cure is far too vague to be useful.

The best part of the article is when Summers explains just how far-reaching the problem is:

Alan Greenspan has been fond of explaining that the resilience of the US financial system and economy results from reliance on two pillars: banks and capital markets. When the banks were in trouble, as in 1991, capital markets took up the slack; when the capital markets were in trouble, as in 1998, the banks took up the slack. Unfortunately, today both the banks and the capital markets show signs of crisis.

The point can be put in another way. Four vicious cycles are simultaneously under way: falling asset prices are forcing levered holders to sell, driving prices further down; losses at financial institutions are reducing their ability to finance investment, which in turn reduces asset values, causing further losses; the weakness of the financial system is reducing growth, which in turn weakens the financial system; and falling output is hitting employment, which in turn leads to reduced demand for output.

"Without active efforts to interfere with these mechanisms," he continues, "there can be no basis for confidence that the American economy will recover even in the medium term." That might or might not be true, but his idea of "active efforts" seems a little week, given the gravity of the situation: it seems to be centered on, um, building frameworks.

First, as the ad hoc nature of the policy response to Bear Stearns and the GSEs illustrates, we do not have a framework in place in which the authorities can do what is necessary to counter systemic risk when a systemically important institution gets in trouble and at the same time protect the interests of taxpayers and the broader financial system.

Second, there is as yet no framework in place for handling the large quantity of bad assets sitting on financial institution balance sheets.

This is a classic politician's solution: a way of passing the problem over to a framework-building committee or two and thereby being able to say that Something is Being Done. In reality, if such frameworks get considered "on a contingency basis", as Summers suggests, they'll never get nailed down until they actually need to be implemented. All crisis response is ad hoc and ex-post, as Summers knows better than most.

Summers's other big idea is spending lots of money on infrastructure.

Given the pressures on state and local budgets and the dramatic increase in some inputs (the cost of building highways has risen 70 per cent since 2004), there is now a substantial backlog of infrastructure projects that have been interrupted or put on hold. Allowing these projects to go forward on a significant scale would stimulate the economy and would channel demand towards the construction workers - mostly men with relatively little education - who have borne the brunt of the economic downturn and whose medium-term prospects are bleakest.

The problem here is the dramatic rise in costs. The increase in the cost of building highways is almost entirely a result of the increase in commodity and shipping costs, which means that a spike in highway building will, at the margin, increase the trade deficit, support high commodity prices, and send valuable taxpayer dollars out of the country -- all while perpetuating an automobile-based transportation infrastructure which is really part of the problem more than it is part of the solution.

Building oil refineries or nuclear power stations would be a better idea; wind farms and mass-transit systems and broadband infrastructure would be better still. For that matter, the US could do much worse than to revisit the Works Progress Administration of the 1930s, with its significant expenditure on the arts. If you want to get federal dollars to hard-working poor people, arts subsidies are one of the best possible ways of doing that: a much higher proportion of total expenditure goes straight to the workers than in things like highway building.

Summers is right that the end to this crisis is nowhere near in sight. Given that, a bit of imagination, rather than pro-forma calls for fiscal stimulus and framework building, might be the order of the day.