Measuring fiscal policy and evaluating its results

There are two measures of fiscal policy: the sum total of everything a government does and the "ramp-up-the-spending-quickly" component which gets labeled "stimulus" by politicians and the media.

In the blogosphere, economists argue mostly about the latter yet it is the former which is more important.  (On this question, among others, I credit Yglesias for being ahead of most of the economists.)

The most important, most effective, and least controversial forms of fiscal policy are the automatic stabilizers.  Let's say you have two countries, A and B.  In country A government spends 50 percent of gdp, mostly on a well-designed welfare state.  When the downturn comes, there is only enough extra borrowing to make up for the lost revenue, and there is no designed "stimulus" per se.  In country B, government spends 25 percent of gdp, mostly not on a well-designed welfare state.  When the downturn comes, country B does an extra three, four, or even five percent of gdp "ramp-up" borrowing and spending.

Which country has a better, more active, and more AD-stabilizing fiscal policy?  Well, it depends on the details and the numbers but I would encourage you to consider country A for this honor.

I don't agree with Krugman's recent interpretation that "Korea and China both engaged in much more aggressive stimulus than any Western nation – and it has worked out well."

In South Korea the welfare state is smaller and more of the government spending is corporate and military, compared to Western Europe, plus government spending is lower overall.  In China central government spending is 19.9 percent of gdp (beware those numbers, though) and the social welfare state institutions, and automatic stabilizers, are very weak, both in terms of quantity and quality. 

For purposes of contrast, in Germany government spending is about 44 percent of gdp and you'll find similar or higher number across Western Europe.

I think of Korea and China as having more "ramp-up" stimulus to make up for what is initially a weaker fiscal policy all things considered (you can add Russia to that list, which also has a high measured "ramp-up" stimulus).  Overall, despite the bigger "make-up." it is quite possible Korea and China are doing less with fiscal policy to stabilize aggregate demand than are the more substantive welfare states.  (Also see Sumner's comment on Krugman here.)

Or think about the fiscal variation within Europe.  "Ramp-up" spending measures of zero vs. 1.5 percent of gdp are portrayed as big differences, but as a chunk of the broader metric — the variation in AD-stabilizing properties of the public sector – it's not such a big deal, especially once you take "crowding out" into account.  Whichever quantity of ramp-up spending you prefer, it's not reason to preach doom and gloom for the welfare state countries with the smaller ramp-up plans.  (Have I mentioned that ramp-up spending is often of lower quality?)

The best fiscal policy — for cyclical and not just growth reasons — is a steady stream of permanent and high-quality government expenditures.  That's true no matter what you think the absolute size of this flow should be.

Sometimes you will see this point acknowledged, such as when U.S. federal transfers to the states are praised as the most effective part of ARRA (which they were).  But again, that's just a constant stream of spending.  And that's in the particulars, not just the aggregate, and it matters because you don't want the federal spending having to rehire the people the states laid off.  The praise is correct, but rarely are its complete implications thought through and presented.

Let's say we measure the efficacy of fiscal policy correctly.  The countries with the most stabilizing fiscal policies would be the big government welfare states with high quality governments.  That means Sweden, Denmark, Germany, Netherlands, and so on, the usual list.  In addition to their relatively high quality governments, they also have relatively strong real economies and healthy institutions.

When this whole episode is over, I would not be surprised to see that same list of countries as having had relatively good recoveries (of course we don't know yet).  A regression could positively correlate "good final outcomes from the crisis" with "total fiscal policy, properly measured."  Such a regression also would be picking up the quality of the institutions and of the real economy.

One way of reading those (potential) numbers would argue that strong real economies, strong governments, and healthy institutions all go together and that such combinations help drive healthy recoveries.  That's actually not so far from real business cycle theory, a favorite whipping boy but in its more sophisticated forms alive and well.  

In the blogosphere, most of what you hear about "fiscal policy" — pro or con — is misconceiving and mismeasuring the concept and then drawing incorrect conclusions.  There's no good reason to focus our economic attention, or perform the informal (or formal) econometrics, on the "ramp-up spending" component.  The ramp-up spending attracts a lot of symbolic interest in the more partisan political debates because it has Obama's or Merkel's or whosever name on it, but it is better to see through such labeling.

Keep your eyes on the ball(s): high quality governments, stabilizing long-run expenditures, well-designed welfare states, robust real economies, and healthy institutions.  In principle there's plenty of room for those concepts in Keynesian economics, but right now they're getting…crowded out…in the intellectual debate.


"There's no good reason to focus our economic attention ... on the "ramp-up spending" component"

Except as of now, the components you stress are given, and the ramp-up component is the available short-run policy lever. Which is why it makes sense to focus on the question of how hard to yank it.

Tyler, read this (in Spanish) related to EU's stress tests and the risk of sovereign default

Why so long-winded Dr. Cowen? In introductory macro we learn this lesson easily by explaining the government does not save. Higher taxes increases the AD (and please don't interpret this to mean raising taxes). I cannot find the explaination in your long explanation!

Thank you.

So, what you're saying is that because of government spending being 44% percent of spending and because of its automatic stabilizers, Germany did not experience as severe a recession and recovered more quickly.


I like your article very much, thanks for you said,hope you like what.

From a commenter at Sumner's:

The role of Govt in a Financial Crisis is to:
1) Prevent Panic
2) Prevent Indigency
The two are related, in that the Fear of Indigency increases Panic.

Automatic Stabilizers have the following positives:
1) They combat Uncertainty (which increases panic ).
2) They're Temporary. They'll go down when the economy improves.
3) They're less open to lobbying and favoritism.
4) They're targeted at people who will spend the money.
5) They don't reduce competition.

When facing a Panic, you want to signal that the Govt is a Lender of Last Resort in the hope that such backing will reduce the panic and the damage to the economy, and so reduce the likelihood that the Govt will actually have to spend money.

Milton Friedman's A Monetary and Fiscal Framework for Economic Stability is a plan based upon Automatic Stabilizers. If you want to keep Govt Expenditure to a minimum in a Financial Crisis or Downturn, then such a plan is the best hope of doing that.

Tyler: "Keep your eyes on the ball(s): high quality governments, stabilizing long-run expenditures, well-designed welfare states, robust real economies, and healthy institutions. "

I could not agree more. I have been trying to enunciate this for years. The whole area of study needs to be broken wide open. Most people don't even have an acquaintance with the most general concepts, composed as an interlocking set of ideas.

"Institutional design" should fully understand the reasons why institutions exist, and the designs that make each one most efficient. Markets are a subset, though an important and commanding subset, of institutions.

The reasons why institutions exist are because (1) there is surplus GENERATED in the gains from trade among specializations, and (2) there is surplus RELEASED by the reduction in transactions costs.

The design that makes institutions most efficient is the two-way street between the user and the institution: focus and accountability. "Focus" means the tightness of focus of the institution upon the problem it is meant to address, and "accountability" means the responsibilities of the users and the institutions to each other. To take two institutions: Markets work by price competition, governments by votes.

"Institutional economics" has come to mean the study of contract law in markets. But this is only the beginning of the discussion.

"I could not agree more. I have been trying to enunciate this for years. "

You agree with, and have been trying to convey for years, an utterly empty, banal set of observations? Good for you! (Sound of clapping.)

Oughtn't you indicate a way in which at least one of them is empty or banal, and disprove Tyler's previous comments while demonstrating your superior knowledge of the topic?

"Oughtn't you indicate a way in which at least one of them is empty or banal, "

Uh, see indianjim's comment above; that pretty much says all that needs to be said. I'd only add, this is
another good example of Tyler posing as a deep thinker without actually saying much.

"Who could disagree with having "high quality governments"?"

"And what exactly are "healthy institutions"? They are Margarettaville; just as Jimmy Buffet defines they could be anything you want them to be."

Although I am inclined to give a pass to anyone including a Jimmy Buffett metaphor, I'll tell you- Krugman disagrees with high quality governments. He wouldn't admit to it, but anyone who just wants to "spend! NOW!" is not concerned with quality and only wants quantity. Maybe now is the time for quantity, I don't know, but I'm not a Keynesian. "Employment at any price!"

I'm not sure exactly what healthy institutions are, but I know what they aren't- anything that grows from a hole piled to the top with money. Again, maybe this is the time for such measures, but normally, making it rain is not a good recipe for high quality institutions.

Speaking in vague platitudes is not difficult; the devil is in the details. This is the reason that the more above 15 to 20% of GDP that government expenditure constitutes, the worse the prospects of economic growth and national survival.

Some commenters have been promoting this post as thoroughly repudiating Cowen's observations. All I can say is: WTF? I know many people around here already believe this. It basically functions as a logical prior for them. But it doesn't mean it's true. It doesn't mean we all agree. And it is certainly does not prove itself. If you want to make this claim, you will have to provide evidence. For example, given the relative post-WWII circumstances of the United States and Germany, is it really possible to claim, unequivocably that the United States' approach has served her better than Germany's has? Or more absurdly, that Germany's national survival is in doubt? These claims that generous well-fare states are inevitably insolvent... are not as well supported by the evidence as those who make them appear to believe.

The big problem in the United States on this issue is that much of the well-fare state (and government services) are funded by state governments which cannot run deficits. And it wouldn't be reasonable for the federal government to just hand out money to them blindly except as a desperation measure. Either you federalize the policy and the spending or you localize both. You can't localize the policy and federalize the spending. That's what you might call moral hazard (or worse I think?).

Comments for this post are closed