Much ado about austerity

There’s been a lot of talk recently about austerity as a spur for growth. Much of it has revolved around the work of Alberto Alesina, a Harvard economist. Bloomberg summarizes his position this way:

“Alesina argues that austerity can stimulate economic growth by calming bond markets, which lowers interest rates and promotes investment. In addition, he says, deficit-cutting reassures taxpayers that more wrenching fiscal adjustments won't be needed later. That revives their animal spirits and their spending. Alesina says that as a way to shrink deficits, spending cuts are better for growth than raising taxes.”

Bruce Bartlett highlights other work in this vein in The Fiscal Times.

Like so many arguments made by respectable economists, on one level this logic is perfectly fine. I have no doubt that fiscal conservatism can spur growth. I would be surprised to find anything else, in fact. But there is an important difference between:

(1.) Increasing GDP growth rates and spurring recovery, and
(2.) Spurring recovery in an economy with depressed aggregate demand and spurring a recovery in an economy suffering from other problems

For some reason, time after time, people act as if all downturns are created equal. If an economy is suffering from a demand shortfall, removing more demand exacerbates the problem. What isn’t a particular concern in an economy that doesn’t exhibit a notable shortfall in demand (say, the American economy of the early 1920s) becomes a major problem in an economy where demand is already quite weak.

The primary mechanism through which austerity operates for Alesina is the bond market. Austerity calms the bond markets, “which lowers interest rates and promotes investment”. No Keynesian would quibble with the importance of this mechanism. The point they raise (highlighted in the Bloomberg article) is that interest rates are already quite low. If nothing else, this suggests that bond markets aren’t concerned about government debt (removing the primary “confidence” argument that Alesina utilizes). So, perhaps cutting spending would lower rates somewhat more – but who honestly believes that government borrowing is propping up interest rates right now? If interest rates are a problem, the source is either (1.) a zero lower bound on interest rates, or (2.) lack of investment demand.

To believe Alesina you have to believe that fear of the U.S. government’s ability to pay back its debt is a bigger factor than unwillingness to make investments, and you have to believe that a zero lower bound on interest rates is not a serious constraint right now (perhaps because you think inflation will make that zero lower bound non-binding as a constraint on real interest rates). Both of those cases seem untenable to me, which is why I really don’t think this argument makes sense at a time like this, at least (it would certainly hold true under different conditions). The primary problem is investment demand. Cutting demand further doesn't address this primary problem. If there were another primary problem, I probably wouldn't have these sorts of doubts.

The other version of the Alesina argument is not a bond-market story but a taxpayer story, and it’s known as “Ricardian Equivalence”. Robert Barro suggested that taxpayers would recognize that increasing deficits would be mean higher taxes down the road, leading them to curtail spending in anticipation. To a certain extent this amounts to assuming his own conclusions. The only reason why they would anticipate higher taxes down the road (as opposed to the same taxes imposed on a larger economic pie) is if they expected deficits not to increase output. In other words, Ricardian Equivalence essentially says “deficit spending cannot stimulate the economy if it does not stimulate the economy” (because if it did stimulate the economy, taxpayers would expect that the government could pay off the debt with lower taxes on a broader tax base). So it’s somewhat nonsensical (or at least tautological) on that point alone. But Paul Krugman points out another problem. He writes:

“If the government introduces a new program that will spend $100 billion a year forever, then taxes must ultimately go up by the present-value equivalent of $100 billion forever. Assume that consumers want to reduce consumption by the same amount every year to offset this tax burden; then consumer spending will fall by $100 billion per year to compensate, wiping out any expansionary effect of the government spending.

But suppose that the increase in government spending is temporary, not permanent — that it will increase spending by $100 billion per year for only 1 or 2 years, not forever. This clearly implies a lower future tax burden than $100 billion a year forever, and therefore implies a fall in consumer spending of less than $100 billion per year. So the spending program IS expansionary in this case, EVEN IF you have full Ricardian equivalence.”

This gets back to my initial point that the impact of deficits on GDP growth rates is different from the impact of deficits on recovery. Even if you believe in Ricardian Equivalence, even if you believe there is exactly zero stimulus effect of government spending (i.e. – a multiplier of 1). Hell – even if you believe there is a multiplier of less than one, you still have the consumption smoothing effect of counter-cyclical deficits! They may do nothing for the net present value of wealth, but they can help out with the roller coaster ride that is the modern industrial economy.
.
So if we’re thinking in terms of interest rates and bond markets, Alesina’s point doesn’t seem to apply to this particular downturn very well at all. If we’re thinking in terms of tax payers and public debt, Alesina’s point doesn’t seem to apply to any downturn. After those two, distinct critiques what we’re left with is that keeping a lid on public debt can improve long-term secular growth.
.
Big whoop – you don’t need a Harvard economist to figure that one out.

Note: If you're a fan of George Mason economics, you might be interested in James Buchanan's critique of Robert Barro's argument here (sorry, can't find an ungated version). His concerns are along somewhat different lines.