Does Romer & Romer mean that tax cuts are generally good for growth?
U.S. federal tax returns are due tomorrow, and the subject seems to be on the mind of bloggers. Greg Mankiw cites a recent study by Christina and David Romer, which finds a strong inverse association between exogenous tax changes and economic output:
Our baseline specification suggests that an exogenous tax increase of one percent of GDP lowers real GDP by roughly three percent.Greg cites the paper without comment, leaving readers to draw their own conclusions, and it appears that many readers have drawn conclusions that may not be warranted (and in some cases, clearly are not warranted) by the paper. The difficulty arises from taking the above finding out of context while failing to consider another important finding about the response to a tax increase:
…inflation appears to fall substantially. The point estimates show the impact reaching 2.2 percentage points after ten quarters, then spiking down to 3.1 percentage points in quarter 11 before returning to 2.2 percentage points in quarter 12.There are two things to notice about this inflation impact. First, it goes in the opposite direction from what would be expected based on a supply-side explanation of the impact of tax changes: if tax increases reduced supply, the inflation rate should rise, not fall. Second, if we take the point estimates seriously, the effect on inflation is huge. Anyone expecting a tax cut to increase growth is going to have to take into account the fact that Ben Bernanke is also likely to read this study – and we all know what Ben is likely to do when he expects inflation.
To put it bluntly, Romer & Romer’s results are quite consistent with a world in which any future tax cut (unless it is offset by a spending cut) is likely to reduce growth, once the expected monetary policy response is taken into account. Their results do not allow us to decompose the tax impact into a supply-side effect and a demand-side effect, but they are at least consistent with the complete absence of a supply-side effect, whereas they are not consistent with the absence of a demand-side effect. If there is no supply-side effect, then there is no change in the economy’s potential growth rate in the short run. Therefore there will be no change in the Fed’s target level of output. Therefore (barring a liquidity trap) the Fed will act to fully offset the short-run impact of tax changes on output. And to do so, in response to a tax cut, the Fed will have to raise interest rates, increasing financing costs for businesses and discouraging investment. Unless the marginal propensity to consume is zero or the Fed makes a mistake, the path of investment will end up lower than what it would have been in the absence of the tax cut. Therefore the capital stock will end up lower, and long-run growth will end up lower.
As I said, the results do not allow us to rule out a supply-side effect, and it’s quite possible that the supply-side effect will be enough to offset the damage to growth from the Fed’s response to the demand-side effect. But I wouldn’t be in too much of a hurry to cite this study as a reason to extend the Bush tax cuts.
Labels: economics, inflation, interest rates, macroeconomics, Mankiw, monetary policy, public finance, taxes
posted by knzn at 4/16/2007 01:46:00 PM
4 Comments:
Knzn, I havent read the paper, but from what you say, I dont necessarily see any inconsistency. If theres a tax hike, sure, a supply sider would claim reduced supply and, yes, higher inflation would be a priori evidence against the theory. So far so good. However, isnt it likely a tax hike (via the permanent income-life cycle hypothesis), makes people feel poorer and so reduces aggregate demand. And that reduces inflation enough to overwhelm any initital adverse supply side effect. Thats right: expansionary supply side effects can initially be inflationary, and vice versa. Remember: Greenspan raised rates in the nineties since he feared higher productivity would increase demand so much so as to raise inflation. Same idea here. See? So, buddy, take that in your pipe and smoke it.
By the way, knzn, I have to compliment you on your new "cuddly bear" image thing. It conributes so much to your caring, liberal image.
mvpy, Essentially, you’re saying that I’m attacking a straw man by saying that the study rules out a pure supply-side effect, since nobody ever expected a pure supply-side effect in the first place. Which is a good point. I would just emphasize again the size of the inflation effect, which suggests a very large demand-side effect. Based on the point estimate, it’s hard to escape the conclusion that the demand-side effect is most of the story. Then again, the estimate is quite imprecise, so you could argue that the demand-side effect is actually smaller than it appears. The paper doesn’t really settle much between Keynesians and supply-siders, but my main point is that it certainly doesn’t settle things in favor of the supply-siders: Keynesians who don’t like tax cuts (because they crowd out investment) have no reason to change their opinion.
Let me add that the supply side effects might take a good while to kick in. Ive always contended the nineties boom had a lot to do with the tax cuts of the eighties. And, by contrast, the demand side effects, sure enough, kick in right away in theory, via the PIH. Relatedly, Ive always been intrigued by the "expansionary fiscal contraction" thing; i.e. gov cutbacks set off a consumer boom, since people think theyre richer. So, my dynamic has more than anecdotal empirical support.
And, by the way, if you oppose tax cuts, I think the crowding out thing is the wrong way to go. I think this has little, if any, empirical support (think: Japan), in an international context.
Well. Happy tax day, cuddily bear -I know this is your fav day of the year, so dont want to be spoiling it on you like this!
Most people would regard Japan as a special case because of the deflation and the near-zero interest rate – which arguably prevented monetary policy from being an effective stimulus to bring Japan up to full employment. The more general argument against crowding out is that foreign capital will finance almost as much domestic investment as the domestic capital would have, and the typical cite would be the US during the 1980s. One could counterargue that there is global crowding out: e.g., US borrowing crowds out investment in the third world. And one could also object to having one nation run up a large foreign debt, because the resulting imbalances are risky, or because future generations will have to pay back the debt, or because the domestic fisc won’t get the tax revenue from those capital returns. There are counterarguments for all of these, as I have discussed in earlier posts (see the July 2006 archives). But in the case of the US, I think there is one argument that trumps all of these: Medciare.
Post a Comment
<< Home