Category Archives: Stimulus

Is Infrastructure Spending Stimulative?

Wyatt Andrews of CBS News writes:

When Moody’s studied the 2009 stimulus package, infrastructure spending rated high. For every dollar spent, $1.44 was returned to the economy.

The problem with this is that it assumes that infrastructure projects will be executed in exactly the way that Keynesian theorists say that they ought to be (“timely, targeted, and temporary” in Lawrence Summers’s words).

That might work on a blackboard or in an (incomplete) computer model, but not in the real world. In the real world, infrastructure projects involve planning, bidding, contracting, construction, and evaluation. All of this takes time, especially if you want to make sure the money is spent wisely (remember, it also must be properly “targeted” or else it won’t work).

And, indeed, as an emperical fact of life, it does seem to take time. According to the CBO:

[F]or major infrastructure projects supported by the federal government, such as highway construction and activities of the Army Corps of Engineers, initial outlays usually total less than 25 percent of the funding provided in a given year. For large projects, the initial rate of spending can be significantly lower than 25 percent.

When macroeconomists account for the delays that are inherent in these types of projects, they arrive at exactly the opposite conclusion of Moody’s. For example, a recent International Monetary Fund paper by Eric Leeper, Todd Walker and Shu-Chun Yang found: Implementation delays can produce small or even negative labor and output responses.” Moreover, these “Implementation delays can postpone the intended economic stimulus and may even worsen the downturn in the short run.”

This helps explain why Lord Keynes himself became a skeptic of these types of projects later in life.  In 1942 he wrote:

Organized public works…may be the right cure for a chronic tendency to a deficiency of effective demand.  But hey are not capable of sufficiently rapid organization (and above all cannot be reversed or undone at a later date), to be the most serviceable instrument for the prevention of the trade cycle.

“Where are the jobs from the Bush tax cuts?”

So asked Senator Franken (D-MN) in a press conference this week.

It is a good question.  From 1981 through 2000, real GDP grew at an average annual rate of 3.3 percent.  But from 2001 to 2008 (even before the Great Recession began), real GDP grew at only 2.1 percent per year.  Why did growth seem to slow after the Bush tax cuts?  There are a number of different plausible answers.  I’ll take a stab at two:

1.  Yes, all things being equal, some people believe that a deficit-financed tax cut should improve the economy.  But all things were not equal in the 2000s.  In just about every aspect other than taxes, economic policy in the 2000s moved in an anti-market direction.  One could cite monetary policy as John Taylor does.  One could cite regulatory policy as Mark and Nicole Crain do.  One could cite trade policy.  One could cite the increased reliance on counter-cyclical stimulus efforts (there were 4 such packages during the Bush years).  Lastly, one could cite the large increases in spending that attended two wars, the prescription-drug benefit, the farm bill, and other policies.  Indeed, according to the broadest measure of economic freedom over time, the “chain-linked” EFW score by Gwartney, Lawson and Hall, U.S. economic freedom steadily improved up until around 2000, whereupon it precipitously fell:

2.  Another reply might be: we didn’t have a tax cut.  By this I mean that a tax cut without a spending cut is not a tax cut; it is a tax deferral.  When I wrote earlier that some people believe that a deficit-financed tax cut will improve the economy, I linked to John Maynard Keynes.  This is because Keynesian models predict deficit-financed tax cuts will spur economic growth (they also predict that deficit-financed spending increases will spur growth).  But there are other schools of thought.  One theory holds that if uber-rational, forward-looking consumers know that Deficits Are Future Taxes (Professor Mike Munger uses the helpful acronym “DAFT”), then they will save for those taxes today, which means that they won’t spend, completely offsetting whatever government spending that the deficits pay for.  But as I’ve explained in a previous post, you don’t need to believe in such an extreme model of human rationality to arrive at the conclusion that deficit-financed taxes will fail to spur growth.  This is because deficits lower the nation’s capital stock, which over the medium-to-long-run also harms economic growth (see the link for more details).

The bottom line: Though he almost certainly didn’t intend it this way, Senator Franken’s observation that the Bush years were not massive growth years actually bolsters the case for both spending cuts AND tax cuts.

How Much did the Stimulus II Spend to Get Broadband Into Homes?

Nick Schulz writes:

Eisenach and Caves looked at three areas that received stimulus funds, in the form of loans and direct grants, to expand broadband access in Southwestern Montana, Northwestern Kansas, and Northeastern Minnesota. The median household income in these areas is between $40,100 and $50,900.  The median home prices are between $94,400 and $189,000.

So how much did it cost per unserved household to get them broadband access?  A whopping $349,234, or many multiples of household income, and significantly more than the cost of a home itself.

Here is a link to the Eisenach and Caves paper.

Keynes vs. Hayek, Round II

Many readers of this blog are probably already familiar with Fear the Boom and Bust. It is a YouTube video produced by John Papola and my colleague at the Mercatus Center at George Mason University, Professor Russ Roberts. As far as rap videos about macroeconomics go, it is the best I’ve ever seen. It has certainly passed the market test with over 2 million views.

Today, Russ and John released a follow-up called Fight of the Century: Keynes vs. Hayek Round Two. Watch it and marvel (then forward to all of your friends):

The Backdoor Bailouts

The Washington Post reports:

States that have borrowed billions of dollars from the federal government to cover the soaring cost of unemployment benefits would get immediate relief from the Obama administration under a plan to suspend interest payments for the next two years.

According to White House Press Secretary Robert Gibbs, the President’s proposal, “prevents future state bailouts, because in the future, states are going to have to rationalize what they offer and how they pay for it.” I’m not convinced.

First, a little background:

The unemployment system is jointly administered by the states and the federal government. To finance the program, both states and the feds tax the first $7,000 of wages paid to each worker, while some states choose to tax income earned beyond that first $7,000.

As the Post reports:

In tough times, states routinely borrow from the federal government to pay benefits. But when states have an outstanding balance for at least two years, federal law triggers an automatic increase in the federal tax to repay the loan. Such tax hikes already have taken effect or are imminent in Michigan, Indiana and South Carolina.

What the Post doesn’t mention (but Bloomberg does), is: “From 2009 until this year, the loans had been interest-free under a provision of the economic-stimulus program.”

Now the President wants to go further, suspending any interest payments the states owe to the federal government for the next two years. In 2014, he would then change the tax base so that instead of taxing the first $7,000 of wages, the feds and the states would each tax the first $15,000. 

It is this aspect of the proposal that the press secretary, evidently, believes “prevents future state bailouts.” This might be true if we assume that policy makers won’t respond to the extra tax revenue by increasing spending. But more fundamentally, it seems to me that the press secretary is glossing over the fact that the first part of the plan—suspension of interest payments—is a bailout.

For historical context, I turned to Robert Inman. In the first chapter of Fiscal Decentralization and the Challenge of Hard Budget Constraints, he writes (p. 57):

The first major wave of lower government defaults occurred during the 1840s, when eight states (Arkansas, Illinois, Indiana, Louisiana, Maryland, Michigan, Mississippi, and Pennsylvania) and the Territory of Florida defaulted.

Maryland Representative William Cost Johnson (you can’t make that name up!) led the effort. As Inman explains, the rest of Congress didn’t agree with Mr. Cost; they refused to bail out the states (p. 57):

Importantly, opponents of a bailout stressed the strategic implications of such a policy; bailouts would signal an accommodating central government and encourage future deficits, defaults, and ultimately inefficient local governments. Congress said no, and there have been no state defaults since.

This marked a turning point in federal-state relations: through recessions, depressions and countless state fiscal crises, the strong no-bailout rule has survived nearly two centuries.

This made the US federal system unique. Unlike local governments in other countries, US states could not run up huge bills and export the costs to their neighbors. As Inman explains, other countries are not so fortunate to have such a strong no-bailout rule (p. 35):

The recent financial crises in Argentina and Brazil, largely precipitated by excessive local government borrowing, are prominent recent examples of how a fiscally irresponsible local sector can impose significant economic costs on a national economy.

The strong no-bailout rule in the US, however, has not prevented the federal government from increasing its role in state finance. Over the years, federal grants to state governments have steadily grown. Now, there are over 1,120 federal programs that are designed to aid the states. Today, federal funding now pays for nearly 1/3rd of all state spending.

What I find particularly alarming, however, is the recent growth in ad hoc state aid programs that are designed to offset short-term fiscal crunches. To me, these look an awful lot like bailouts. Consider the $135 billion in state aid in the stimulus which included:

  • A state fiscal stabilization fund designed to shore up deficits
  • A temporary increase in the federal Medicaid matching formula (FMAP)
  • Grants for various local projects from teachers to firefighters to police
  • The aforementioned interest-free loans for unemployment insurance
  • And much more

On top of that, the President successfully lobbied for an extension of the “temporary” FMAP increase and an extension of the federal-state unemployment insurance program (he was less-successful in last summer’s attempt to wrangle another $50 billion in state and local aid).

If somehow they could see this, I suspect that the senators and representatives who stopped a state bailout over 170 years ago might wonder if their “no bailout” stance really still stands.

Shovel Ready or Not

Last week, I opined that too-often, economic debates are characterized as a simplistic “markets work” vs. “markets don’t work” debate. This, I said, misses the point that even if markets don’t work, one must determine whether or not government is capable of improving on the situation. Public choice economics, I believe, is a powerful tool to answer that question (often the answer is “no; government cannot improve the situation”). 

In a similar vein, I (and others) have wondered whether the problems with the stimulus might have more to do with government’s inability to wisely spend than with any flaw in the Keynesian logic. 

Now, the president, it seems, is coming around to my view. In an interview with the New York Times Magazine, the president wondered whether:

He realized too late that “there’s no such thing as shovel-ready projects” when it comes to public works.

Remember, all resources that the government spends are borrowed or taxed out of the private economy. The Keynesian story relies on the idea that—during a recession—these resources aren’t doing anything productive in the private economy anyway (they are “idle”). So, according to Keynesians, it is okay for government to remove these resources from the private economy and put them to work. But in order for this to be a net positive, government must know how to effectively put these resources to more-productive use than the private sector would have. It must know:

  • Which spending items provide the most value to people, and
  • How to effectively deliver this value

But unlike private businesses, the government cannot rely on the price mechanism to help discern what people value. Since it doesn’t actually sell anything, the government cannot rely on consumer-feedback and consumer-price sensitivity to determine whether what it is doing is valuable. Instead, responsible governments must undertake time-consuming “cost-benefit” analyses to guestimate which “shovel-ready” projects are worthy and which are just pork.

So here is the rub: stimulus spending is only effective if it is timely and targeted toward high-valued projects. But by its very nature, government cannot determine which projects are high-valued in a timely manner. We can get timely, or high-value. But as Eileen told the Washington Post in February of 2009, “you can’t have both.”

Market Failure vs. Government Failure

Too often, economic policy questions are characterized as an over-simplified debate over whether or not markets work.

On one side you have Chicago School Economists or “fresh water economists” (so called since, historically, they have tended to be located away from the coasts). These thinkers articulate the reasons why markets can be expected to work. 

On the other side are various market-failure theorists, sometimes called “salt water economists” or Keynesians (at least when they are talking about macro phenomena). These thinkers point out that, ahem, markets sometimes mess up big time.

But as Harold Demestz pointed out over 40 years ago, it is not enough to condemn market failure and call for some government intervention to correct the error. One also needs to analyze the government’s actions and see whether or not it is prone to failure. After all, if the government cure is worse than the initial disease, maybe we are better off with the disease. Yellow fever sounds pretty nasty, but that doesn’t mean Benjamin Rush was right to prescribe bloodletting.

This, in a nutshell, is the raison d’être of the public choice school of economics. Public choice economists have varying beliefs about the efficacy of markets. Some think markets work quite well most of the time, others take the idea of market failure seriously. But they don’t assume—as most Keynesians seem to—that government corrections will be flawlessly executed. Instead, they study the ways in which government intervention actually plays out. And more times than not, they find that real world policy is far less equitable, less efficient, and less just than the Keynesian model assumes. 

Oddly, even prominent Keynesians regularly observe and complain about government failure (Google Krugman and the Iraq War). Why, then, do they persist in counseling more intervention? Do they assume that, miraculously, this time will be different?

All of this was going through my mind when I read:

The federal government last year sent about 89,000 checks of $250 each to dead or incarcerated people through the Obama administration’s economic stimulus program, according to a watchdog report.  

Is it Possible that the President thinks Economists Agree That Spending is the Answer?

[W]ith respect to aggregate demand, I don’t know any economist — including, I think, Martin — who would argue that we are more likely to get a bump in aggregate demand from $700 billion of borrowed money going to people like those of us around this table who I suspect if we want a flat-screen TV can afford one right now and are going out and buying one.

If we were going to spend $700 billion, it seems that we’d be wiser having that $700 billion going to folks who would spend that money right away if we were going to boost aggregate demand.

That is President Obama, responding to a question from Martin Feldstein about extending the Bush tax cuts. Note, first of all, that though the president is talking about whether or not we will allow taxes on high-income Americans to rise, he easily slips into the language of spending. In his vernacular, we are “spending $700 billion” when we choose not to raise taxes by $700 billion.

More to the point, however, the quote suggests that the president is under the impression that economists unanimously believe spending increases are wiser than tax cuts. It would be a shame if the president’s advisors gave him this impression.

In 2009, the University of Chicago’s Harold Uhlig reviewed the most-recent literature on this question. By Uhlig’s count, the following studies conclude that, in terms of boosting GDP, tax cuts have a larger impact than spending increases:

Shortly after Uhilg wrote this, Alesina and Ardagna (2009) also found that stimuli based on tax cuts tend to be more effective than stimuli based on spending increases.   

Aside from the spending boost vs. tax cut debate is the question of magnitude. On this score, the Administration’s assumptions are well-outside of the range found in the most-recent studies. According to Uhlig:

With the exception of Gali [,Lopez-Salido, and Valles] (2007), the fiscal multipliers for government spending also typically seem to be considerably more modest than the [Administration assumes].

To this list, I would add Barro and Redlick (2010).

This is not to mention work such as the recent IMF study by Freedman, Kumhof, Laxton, Muir, and Mursula (2010) that finds short-run positive effects from stimuli, but medium-term deleterious effects.

It would be one thing to know about this evidence and to dismiss it. The president, however, seems not to have heard of it.

Is Boosting Teacher Pay a Legitimate Use of the Stimulus?

That is the question raised by a recent headline from Fairfax County:

“Fairfax Teachers Fight for Raises Funded by Stimulus”

Now I don’t know whether teachers in Fairfax are underpaid or overpaid. And there may very well be good reason to boost their salaries (or at least to boost those of the top performers). But it makes little sense to argue that a salary boost right now would serve the original purpose of the stimulus (which was intended to stimulate the economy, for those who are still paying attention). 

Let’s try to look at this the way a Keynesian would. As Brian Caplan explained a few months back, an important element in the Keynesian model is the notion of wage rigidity. Here, according to Keynesians, is how it works:

  1. Trouble begins when aggregate demand falls (usually because animal spirits have caused a sharp decrease in investment).
  2. As this happens, firms have less revenue with which to pay salaries. They might be able to maintain employment, however, if they could convince their employees to take a real wage cut. 
  3. The problem, however, is that nominal wages are “sticky.” Employees do not want to accept lower wages and even if they did, employers would rather fire some people than lower everyone’s wage and have a disgruntled workforce.

The Keynesian solution to this dilemma is to somehow boost aggregate demand to get people spending again. As Brian has more-recently pointed out, an alternative solution that is completely consistent with the Keynesian diagnosis of the problem would be to make sure that all prices—including wages—are as fluid as possible. If we could reduce nominal wage stickiness and get people to accept lower wages during a recession, then we could lower the unemployment rate.

Somehow in practice, however, Keynesian policy usually ends up increasing nominal wage stickiness and the Fairfax teacher story is a case in point.

Addendum: My colleague, Veronique de Rugy, weighs in at NRO, noting that the Administration changed the rules in February to permit stimulus funds to be used for such a purpose.

What if Stimulus Works, But Government Can’t Get the Timing Right?

As of September 3, 2010, about $154.8 billion of the approximately $282 billion of total funds made available by the Recovery Act in 2009 for programs administered by states and localities had been paid out by the federal government.

That’s the conclusion of a new GAO report, out this week. Similarly, the Wall Street Journal reports this morning that:

[S]pending stimulus dollars fast has turned out to be surprisingly hard.

This reminds me of a point that Megan McArdle raised a few weeks back:

[W]hat if Keynesian stimulus works, but no one can ever actually afford to do it, short of something like World War II, where the government can tap into a patriotic outpouring of national savings by issuing bonds with negative real yields.

She was talking about the sheer size of the stimulus. But we could ask a similar question: What if Keynesian stimulus works, but the machinery of government is so slow and inept, that it is impossible to effectively implement it in time to be effective?

This, of course, was the (near) consensus view among macroeconomists just a little over a decade ago. Writing in the American Economic Review in 1997, Martin Eichenbaum wrote:

[T]here is now widespread agreement that counter cyclical discretionary fiscal policy is neither desirable nor politically feasible.

Perhaps the current struggles to effectively administer stimulus will one day cause that consensus to re-emerge.