« Is a Bonus Culture Ruining Africa and Our Financial System? | Main | Jobs »

Thursday, December 11, 2008

"Measuring the Effect of Infrastructure Spending on GDP"

Susan Woodward and Bob Hall say the multiplier for infrastructure spending is likely to be around 1.0:

Measuring the Effect of Infrastructure Spending on GDP, Woodward and Hall: The Obama administration’s focus on infrastructure spending raises the natural question of the effect of government purchases on total GDP. Does government spending stimulate other categories of spending, especially consumer spending? Or does government spending displace other categories, so GDP rises by less than the amount the government spends?

Valerie Ramey has written a paper with the results of her recent work on the question and with a full bibliography of earlier work. Her answer is that consumption and other categories stay about the same when the government spends more. In other words, the increase in GDP is about equal to the increase in government spending. To focus on changes in government spending that are not themselves responding to conditions in the economy, she considers military spending. She finds that GDP rises by about the same amount as an increase in military spending. ... [see original post for graphs]

If you think that the Obama administration is ambitious in spending a trillion dollars over several years on infrastructure projects, note that military spending maxed out at $7 trillion per year during the war, rescaled to the current size of the economy. During the expansion, GDP rose pretty much the same amount as did military spending. ... Notice, however, that when military spending fell after the victory, GDP did not fall nearly as much. Consumption and other components expanded rapidly to take up the resources freed from military activities and there was little sign of adverse effects from the lower military spending. ...

We believe that the one-for-one rule derived from wartime increases in military spending would also apply to increases in infrastructure spending in a stimulus package. We should not count on any inducement of higher consumption from the infrastructure stimulus but we should also not worry that infrastructure spending might displace consumption and other categories of spending.

Looks like a trillion dollars may not be enough.

Update: Paul Krugman:

Don’t know much about history: ...Bob Hall and Susan Woodward argue against the multiplier effect of infrastructure spending by pointing out that GDP and military spending rose by about the same amount during World War II.

Um, rationing?

With the onset of World War II, numerous challenges confronted the American people. The government found it necessary to ration food, gas, and even clothing during that time. Americans were asked to conserve on everything. With not a single person unaffected by the war, rationing meant sacrifices for all.

[Bangs head against the table]

Update: Greg Mankiw:

Spending and Tax Multipliers: ...In their new blog, Bob Hall and Susan Woodward look at spending increases from World War II and the Korean War and conclude that the government spending multiplier is about one: ... Similarly, the results in Valerie Ramey's research suggest a government spending multiplier of about 1.4. ...

By contrast, recent research by Christina Romer and David Romer looks at tax changes and concludes that the tax multiplier is about three: A dollar of tax cuts raises GDP by about three dollars. The puzzle is that, taken together, these findings are inconsistent with the conventional Keynesian model. According to that model, taught even in my favorite textbook, spending multipliers necessarily exceed tax multipliers.

How can these empirical results be reconciled? One hypothesis is that that compared with spending increases, tax cuts produce a bigger boost in investment demand. This might work through changing relative prices in a direction favorable to capital investment--a mechanism absent in the textbook Keynesian model.

Suppose ... that tax cuts ... take the form of cuts in payroll taxes... This tax cut would reduce the cost of labor and, if labor and capital are complements, increase the demand for capital goods. Thus, the tax cut stimulates demand not only by increasing disposable income and consumption spending (the textbook Keynesian channel) but also by incentivizing more investment spending. A similar result might obtain if the tax cut included, say, an investment tax credit.

This hypothesized channel seems broadly consistent with the empirical findings of Blanchard and Perotti, Mountford and Uhlig, Alesina and Ardagna, and Alesina, Ardagna, Perotti, and Schiantarelli. The results of all these authors suggest you need to go beyond the standard Keynesian model to understand the short-run effects of fiscal policy.

My advice to Team Obama: Do not be intellectually bound by the textbook Keynesian model. Be prepared to recognize that the world is vastly more complicated than the one we describe in ec 10. In particular, empirical studies that do not impose the restrictions of Keynesian theory suggest that you might get more bang for the buck with tax cuts than spending hikes.

Romer and Romer do not look at government spending multipliers, so we don't know if the government spending multiplier is smaller/larger than the tax multiplier if they are estimated in the same model (note, though, the the confidence band for Romer and Romer's tax multiplier is 1.3 to 4.7, so a multiplier smaller than Ramey's estimated value of 1.4 for government spending - though not strictly comparable since it comes from a different model - is within the confidence bounds of these estimates; confidence bounds for Ramey's estimates are not reported). I should also note that Romer and Romer isolate different types of tax cuts, the multiplier of 3.0 (plus or minus about 1.7) is for exogenous tax changes intended to promote economic growth, not tax changes intended to stabilize the economy. Their results for tax cuts used for stabilization purposes are not encouraging:

The behavior of output following countercyclical tax changes ... suggests that policymakers' efforts to adjust taxes to offset anticipated changes in private economic activity have been largely unsuccessful.

Finally, many estimates of the Keynesian government spending multiplier can be criticized on the same grounds that the tax cut advocates like to cite. The installation of infrastructure increases economic growth in the future, but these dynamic effects are missing from the standard Keynesian analysis (and using history as a guide is not very helpful since we have had very few surges in government spending devoted to infrastructure spending, and we have little data on the effects of fiscal policy - or any policy - in severe downturns since they are so unusual). If the dynamic effects were to be included in the infrastructure spending mulitplier, the multiplier would be larger.

Update: From Free Exchange:

Tax or spend?, Free Exchange: Greg Mankiw is getting nervous thinking about all that infrastructure spending. He's down with stimulus, but like all good conservatives, he recognises that tax reductions are preferable to spending increases. And he has evidence! Sort of:

In their new blog, Bob Hall and Susan Woodward look at spending increases from World War II and the Korean War and conclude that the government spending multiplier is about one: A dollar of government spending raises GDP by about a dollar.

That's not very good. The government spent gobs during the Second World War and consumer spending didn't rise a bit. But Paul Krugman takes a closer look...

Um, rationing? ...

That's right, the government was doing its best to crowd out consumer spending, so as to reduce resource competition, which would have increased the prices of commodities and munitions. But Mr Mankiw goes on:

By contrast, recent research by Christina Romer and David Romer looks at tax changes and concludes that the tax multiplier is about three: A dollar of tax cuts raises GDP by about three dollars. The puzzle is that, taken together, these findings are inconsistent with the conventional Keynesian model. According to that model, taught even in my favorite textbook, spending multipliers necessarily exceed tax multipliers.

That's somewhat true. The Romers do write:

Recall that we find that a tax increase of one percent of GDP lowers real GDP by about 3 percent, implying a substantial multiplier.

But there are plenty of questions about the extent to which this is applicable to countercyclical tax cuts, particularly since the authors specifically eliminate countercyclical tax changes from their sample in order to eliminate bias generated by omitted variables. If Keynesian tax changes are fundamentally different, then we can't learn anything about fiscal stimulus from this paper. And this may be why Barack Obama is happy to have Christina Romer in his administration.

And this may also be why Mr Mankiw's position no longer enjoys the support of his fellow conservative, Martin Feldstein, who wrote in August:

Those of us who supported this fiscal package reasoned that the program would boost consumer confidence as well as available cash. We hoped the combination would cause households to spend a substantial fraction of the rebate dollars, leading to more production and employment. An optimistic and influential study by economists at the Brookings Institution projected that each dollar of revenue loss would increase real GDP by more than a dollar if households spent at least 50 cents of every rebate dollar.

The evidence is now in and that optimism was unwarranted. Recent government statistics show that only between 10% and 20% of the rebate dollars were spent. The rebates added nearly $80 billion to the permanent national debt but less than $20 billion to consumer spending. This experience confirms earlier studies showing that one-time tax rebates are not a cost-effective way to increase economic activity.

Now there are other ways that a tax change might boost economic activity than consumer spending, and Mr Mankiw describes some potential alternative mechanisms. A tax cut in this environment certainly won't do any harm, and Mr Obama will likely include some tax reductions in his initial stimulus package. But it's difficult to make the case that a spending stimulus won't be as effective, at least, as a tax cut. Time to get those shovels in the ground, I think.

    Posted by on Thursday, December 11, 2008 at 01:08 AM in Economics, Policy | Permalink  TrackBack (0)  Comments (63)

          

    TrackBack

    TrackBack URL for this entry:
    http://www.typepad.com/services/trackback/6a00d83451b33869e2010536524367970b

    Listed below are links to weblogs that reference "Measuring the Effect of Infrastructure Spending on GDP":


    Comments

    Feed You can follow this conversation by subscribing to the comment feed for this post.