Robert Barro's Ricardian equivalence theorem was one nail in the coffin. This theorem says that stimulus cannot work because people know their taxes must rise in the future … We cannot return to mechanically adding up today's consumption, investment and export demands, and prescribe the government demand necessary to attain some desired level of output. Every economist now knows that to get stimulus to work, at a minimum, government must fool people into forgetting about future taxes, an issue Keynes and Keynesians never thought of.
Franco Modigliani was generally thought to be a Keynesian and Barro’s theorem is in part a reformulation of the Ando-Modigliani life-cycle theory of consumption. Barro added the sensible proposition that the present value of future taxes must cover the sum of the current government debt and the present value of government spending – but this proposition does NOT lead to the conclusion that any increase in government purchases will be completely offset by a reduction in consumption as Kevin Quinn notes:
Ricardian equivalence, it is true, implies that deficit-financed tax cuts cannot affect demand. Deficit-financed temporary increases in Government spending, on the other hand, can. Consumption falls today, because the present value of future taxes is higher by the amount of the spending increase, but not by as much as G rises. The reduction in the present value of life-time income implies that the [present value of the] sum of reductions in current and future consumption will be equal to the increase in G, so the reduction today will be small.
Kevin’s point has also been recently made by Paul Krugman and yours truly. And yet – Cochrane ignores this aspect of what Brad DeLong rightfully refers to as “Ricardian Consumers and Fiscal Policy Once Again”.
Mark Thoma has a thoughtful post on why the Barro-Ricardian proposition about the ineffectiveness of tax cuts might fail, that is, why giving a tax cut to a borrowing-constrained households might still lead to an increase in aggregate demand. My only misgiving with Mark’s post was his lead:
This discussion at Brad DeLong's makes the point that Ricardian equivalence fails for deficit financed temporary changes in government spending. But what's not clear from the discussion is that there's no reason to expect Ricardian equivalence to hold in any case in practice, even for deficit financed tax cuts where it can be true in theory.
Even if the Ricardian model did fit the real world, Cochrane’s proposition that increases in government purchases did not impact aggregate demand is wrong, that is, simply a failure of his ability to understand the implications of this particular theory.