# The Real Multiplier (II)

In truth, the Keynesian multiplier is a mathematical fiction, as explained here, and government spending is in fact destructive of economic growth, as discussed here and in some of the posts listed at the end.

From “The Real Multiplier“:

My own analysis of post-WWII statistics (for 1947-2010) yields a multiplier for G of 0.8. That is to say, every additional dollar of government spending has led to a 20-cent reduction in non-government spending. To put it another way, the real effect of additional government spending is the shrinkage of the private sector. (Which makes sense, when you stop to think about it.) Thus the irrelevance of the multiplier: What good is more government spending if it shrinks the private sector, where real products and services are produced?

I may have been too generous. Valerie A. Ramsey, of UC San Diego and NBER, writes:

…For the most part, it appears that a rise in government spending does not stimulate private spending; most estimates suggest that it significantly lowers private spending. These results imply that the government spending multiplier is below unity. Adjusting the implied multiplier for increases in tax rates has only a small effect. The results imply a multiplier on total GDP of around 0.5. (“Government Spending and Private Activity,” January 2012)

Ramsey’s thorough analysis trumps my back-of-the envelope calculation. The government-spending multiplier is 0.5 — and don’t you forget it, Paul Krugman.

UPDATE: Daniel J. Mitchell offers much additional evidence about the high cost of government spending.

# Why the “Stimulus” Failed to Stimulate

This post examines practical reasons for the failure of “stimulus” to stimulate and the “multiplier” to multiply. The deeper truth is that the Keynesian multiplier is a mathematical fiction, as explained here, and government spending is in fact destructive of economic growth, as discussed here and in some of the posts listed at the end.

I spell out the reasons in “A Keynesian Fantasy Land.” There are six of them, including the timing-targeting problem (number 3).

[In an earlier version of this post, I also mentioned Ricardian equivalence, which is an aspect of reason number 2, the disincentivizing aspects of government borrowing and spending. I referred to a post by Steven Landsburg, which I had read hastily and misinterpreted as a discussion of Ricardian equivalence. When Dr. Landsburg graciously pointed out that I had the wrong end of the stick, I deleted the brief discussion of Ricardian equivalence from this post.]

A key component of the timing-targeting problem is the strong possibility that “stimulus” money will be spent on already-employed resources, thus bidding up their prices but doing little or nothing to stimulate real economic activity. Tyler Cowen recaps two papers that document the misdirection of “stimulus” money:

My colleagues Garett Jones and Daniel Rothschild conducted extensive field research (interviewing 85 organizations receiving stimulus funds, in five regions), asking simple questions such as whether the hired project workers already had had jobs.  There are lots of relevant details in the paper but here is one punchline:

…hiring people from unemployment was more the exception than the rule in our interviews.

In a related paper by the same authors (read them both), here is more:

Hiring isn’t the same as net job creation. In our survey, just 42.1 percent of the workers hired at ARRA-receiving organizations after January 31, 2009, were unemployed at the time they were hired (Appendix C). More were hired directly from other organizations (47.3 percent of post-ARRA workers), while a handful came from school (6.5%) or from outside the labor force (4.1%)(Figure 2).

One major problem with ARRA was not the crowding out of financial capital but rather the crowding out of labor.  In the first paper there is also a discussion of how the stimulus job numbers were generated, how unreliable they are, and how stimulus recipients sometimes had an incentive to claim job creation where none was present.  Many of the created jobs involved hiring people back from retirement.  You can tell a story about how hiring the already employed opened up other jobs for the unemployed, but it’s just that — a story.  I don’t think it is what happened in most cases, rather firms ended up getting by with fewer workers.

There’s also evidence of government funds chasing after the same set of skilled and already busy firms.  For at least a third of the surveyed firms receiving stimulus funds, their experience failed to fit important aspects of the Keynesian model.

The Keynesian model is deeply flawed because it is a simplistic model based on simplistic assumptions about the behavior of human beings and human institutions. I say in “A Keynesian Fantasy Land,” models are supposed to mirror reality, not the other way around. The Keynesian model — or the version embraced by Paul Krugman and his fellow leftists — is a version of the reality that they would prefer: a reality in which government runs the economy.

# The Keynesian Fallacy and Regime Uncertainty

In “A Keynesian Fantasy Land,” I gave six reasons for the failure of “stimulus” spending to stimulate the economy, despite the insistence of leftists and left-wing economists that economic salvation is to be found in bigger government. The reasons, which I elaborate in the earlier post, are these:

1. “leakage” to imports

2. disincentivizing effects of government borrowing and spending (regime uncertainty)

3. timing and targeting problems (spending that is too late and misdirected)

4. reversed causality (lower aggregate demand as symptom, not cause)

5. the negative consequences of bail-outs

6. the unaccounted for complexity of human behavior

An article by Casey B. Mulligan, “Simple Analytics and Empirics of the Government Spending Multiplier and Other ‘Keynesian’ Paradoxes,” underscores the futility of “stimulus” spending. These are among Mulligan’s conclusions:

From a partial equilibrium perspective, it would be surprising if government purchases did not crowd out at least some private consumption, and that a reduction in factor supply did not result in less output. Yet some “New Keynesian” models, not to mention much public policy commentary, claim that today’s economy has turned this partial equilibrium reasoning on its head, even while it might have been historically valid. Among other things, individual firms and the aggregate private sector are alleged to leave their production invariant to changes in factor supply conditions during this recession. This paper shows how the government spending multiplier and the “paradox of toil” are related in theory, and examines evidence from this recession on the output effects of factor supply…

This paper does not contain a numerical estimate of the government purchases multiplier. However, its examination of data exclusively from the 2008-9 recession suggests that sectoral and aggregate employment and output vary with supply conditions in much the same way they did before the recession. The results contradict Keynesian claims that the government purchases multiplier would be significantly greater during the recession than it was before 2008, suggesting instead that historical estimates of the effects of fiscal policies are informative about fiscal policy effects in more recent years. Moreover, the supply incentives created by government spending cannot be ignored merely because 2008 and 2009 were recession years; rather incentives mattered as much as ever. Government purchases likely moved factors away from activities that would have supported private purchases. Unemployment insurance, food stamps, and other expanding means-tested government programs likely reduced employment and output during this recession, in much the same way they did in years past.

Compounding the futility of “stimulus” spending is the general climate of economic fear that Obama’s policies have engendered; for example:

As John Steele Gordon points out,

[t]he greatest periods of American economic growth came when taxes were very low—such as in the 19th century—or being lowered and simplified, as in the 1920s, 60s, and 80s. Inescapably, to tax wealth creation is to discourage it. But there is a large and politically potent segment of the population that, because its interests are now aligned with those of the government, seek to promote dependency through entitlements. This segment favors ever higher taxes (although they disguise the fact by demanding that only “the rich” pay their “fair share.”) But, as with regulation, high taxes inevitably produce low growth—and low growth threatens entitlements in the long term. If the United States remains in the doldrums for several more years without hope of a real turnaround, Medicare as it is currently constituted will go bankrupt in 2019. Raising taxes to prevent that will only slow overall growth, and that will actually defeat the purpose of saving Medicare.

So there is really no alternative to pursuing policies that encourage economic growth through private action by liberating the forces of the free market. A presidential candidate who finds a way to ground his economic policies in this core truth—and harnesses the idea to a larger and more optimistic understanding of the United States, both past and future, and resists the take-your-medicine tone that dominates the conservative policy discussion of the present moment—will be able to draw a sharp and effective contrast with the failures of the Obama years. (“Growth: The Only Way out of This Mess,” Commentary, July 2011)

But there is no point in cutting taxes unless government spending is cut — and cut drastically — for government spending, along with regulation, is the real drag on the economy. Only in the left’s magical thinking is government spending a good thing. In reality, it is a destructive force — even during recessions and depressions.

# A Keynesian Fantasy Land

This post examines practical reasons for the failure of “stimulus” to stimulate and the “multiplier” to multiply. The deeper truth is that the Keynesian multiplier is a mathematical fiction, as explained here, and government spending is in fact destructive of economic growth, as discussed here and in some of the posts listed at the end.

“Liberal” economists and pundits complain incessantly that the recovery from the Great Recession is weak, and in jeopardy, because the federal government hasn’t spent “enough” money. (See this for some examples of the “liberal” view.) How much is “enough” for Paul Krugman et al.? It is always more than the government spends, of course.

Why should that be? The blindingly obvious answer — but not obvious to Krugman and company — is that demand-side fiscal policy (i.e., government “stimulus” spending) is ineffective. If the economy depends on government spending, how does one explain the decades after the Civil War, when government spent less than 10 percent of GDP (vs. today’s 40 percent), while America’s economy grew faster than at any time in its history? It took World War II and regime change (the disruption of the New Deal by the war) to end the Great Depression. Mr. Roosevelt’s adoption of Mr. Keynes’s hole-digging prescription (the Civilian Conservation Corps and similar make-work projects) had nothing to do with it. Mr. Roosevelt may have been an excellent marketeer, but he was a dismal economic engineer.

This is not to reject supply-side fiscal policy: tax-rate reductions. When tax-rate reductions are prospectively permanent — as opposed to one-time tax rebates and “holidays” — they can and do spur economic growth. Christina Romer, former chair of Obama’s Council of Economic Advisers, once proved it — though she developed a convenient case of amnesia when she became a proponent of “stimulus.”

As any reputable economist will tell you, however, the best that one can expect of a temporary increase in government spending is a temporary increase in economic activity; it is a stop-gap until the economy recovers on its own. (And a reputable economist, unlike Krugman, will also tell you that a permanent increase in government spending diverts resources from productive uses — uses that yield economic growth and satisfy actual economic wants — toward less-productive and counter-productive ones, including the creation of paper-shuffling, regulatory bureaucracies.)

Despite the promises of Obama, Romer, and company, the “stimulus” has evidently failed to do much — if anything — to alleviate the Great Recession and its lingering aftermath. (See this, this, and this, for example.) Thus the wailing and gnashing of teeth by Krugman and company — who want to replicate the failure on a grander scale.

WHY THE “STIMULUS” FAILED TO STIMULATE: GENERAL OBSERVATIONS

What went wrong? Anthony de Jasay offers a piece of the explanation:

…In Keynesian parlance there is the multiplier effect and it is greater than 1. As long as there is spare capacity (unemployment) in the economy, the government ought to go on spending more, working through the multiplier, because the extra private saving takes care of the government dissaving and the extra consumption is, so to speak, a welcome windfall gain. Timidly refusing to generate it is criminal waste.

Despite truculent voices to the contrary, the Keynesian logic is faultless in that the conclusions do follow from the assumptions. Why it does not really work and why it singularly failed to work in 2009-2010 and maybe beyond, is that other things do not remain equal. Part of the extra spending stimulus fails to stimulate domestic income because as much as 0.3 of the multiplier might leak out through extra imports. Much of the rest may be offset by industry taking fright of the rising budget deficit and reducing investment, and consumers striving to reduce their indebtedness producing some saving to balance the government’s dissaving. The total effect of higher imports and lower investment might be a multiplier barely higher, or maybe even lower, than 1 and the stimulus stimulating nothing except the national debt. This is not the fault of Keynes but of those whose macro-economics exist in a fantasy land. (Library of Economics and Liberty, “Micro, Macro, and Fantasy Economics,” December 6, 2010)

Generally,

[t]he available empirical evidence does not support the idea that spending multipliers typically exceed one, and thus spending stimulus programs will likely raise GDP by less than the increase in government spending. (Robert J. Barro and Charles J. Redlick, “Stimulus Spending Doesn’t Work,” WSJ Online, October 1, 2009)

(For more on the subject see Barro’s “Government Spending Is No Free Lunch,” WSJ Online, January 22, 2009.)

WHY “STIMULUS” FAILS: SPECIFIC REASONS

Altogether, there are six reasons for the ineffectiveness of Keynsesian “stimulus.”

1. The “leakage” to imports, as indicated by de Jasay.

2. The disincentivizing effects of government borrowing and spending, to which de Jasay alludes.

As de Jasay suggests, industry (and the high-income earners who finance it) are being cautious about the implications of additional government debt. As I say here,

the sophisticat[ed] … institutions and persons who have the greatest interest in government’s actions [are] large corporations and persons in high-income brackets. They will react to government borrowing as if it would affect them and their heirs (corporate and individual).

That is to say, even if additional debt does not crowd out private-sector borrowing to finance business expansion, it will nevertheless inhibit investments in business expansion. This inhibiting effect is compounded by the reasonable expectation that many items in a “stimulus” package will become permanent fixtures in the government’s budget.

3. The timing-targeting problem.

The lag between the initial agitation for “stimulus” and its realization. In the extreme, the lag can be so great as to have no effect other than to divert employed resources from private to government uses. But even where there is a relatively brief lag, “stimulus” spending is essentially wasted if the result is simply to divert already employed resources from private to government uses.

The timing-targeting problem is one that strident Keynesians and their unsophisticated disciples in the media seem not to understand or care about. (They are happy as long as government “does something,” regardless of the cost.) The problem arises from the fundamental flaw in the Keynesian analysis: Economic output is portrayed as a homogeneous commodity, one that can be characterized  in terms of aggregate demand (AD) and aggregate supply (AS). Accordingly, in the Keynesian orthodoxy, all it takes to stimulate AD is to pump in some additional government spending (dG), and the rest takes care of itself.

Arnold Kling calls it “hydraulic” macroeconomics:

Once upon a time, Joe lived in Keynesiana, where he was a representative agent.

Joe worked in a GDP factory, making GDP. Every Monday morning, he went to work, and he worked five days a week. He was paid \$1 for every 24-minute segment he worked, and he worked 100 segments (40 hours), so he earned \$100 a week. Every Friday afternoon, Joe cashed his paycheck and went to the GDP factory outlet, where he spent it all on GDP.

One day, Joe decided that he needed to accumulate some savings. He made up a rule for himself. Knowing that he needed to consume at least \$40 of GDP each week, he decided that his rule would be to save 20 percent of everything he earned over and above that \$40. So the first week, that meant saving 20 percent of \$60, or \$12. So he cashed his \$100 paycheck, but that Friday afternoon he only spent \$88.

Next Monday, morning, Joe’s boss had some news. “A funny thing happened last week. We sold 12 percent less GDP than usual. So this week, we’re gonna put you on a short week. You work 88 segments, instead of 100.”

Joe was disappointed, because this meant he would only be paid \$88 this week. Sticking to his new rule, he resolved to save 20 percent of \$48, or \$9.60. So that Friday afternoon, he cashed his \$88 paycheck and spent \$78.40.

Next Monday morning, Joe’s boss said. “Well, golly, it looks like we sold even less GDP last week. I’m afraid we’ll have to cut you back to 78.40 segments this week.” Still following his rule, Joe resolved to save 20 percent of \$38.40, or \$7.68. So he spent only \$70.72 at the GDP factory outlet that Friday.

Seeing where this was going, the country asked Krug Paulman, the famous economist, what to do. He said, “The stupid people are saving too much. We need government to spend what the idiots are not spending.” So the government borrowed \$29.28 from Joe and spent it at the GDP factory outlet.

Now, when Joe came to work on Monday morning, his boss said, “Good news, we sold 100 percent of what we used to sell, so you can work 100 segments this week.” Sticking to his rule, Joe saved \$12 on Friday afternoon. But the government borrowed the \$12 and spent it at the GDP factory outlet. They all lived happily ever after. (Library of Economics and Liberty, “Hydraulic Macro: A Fable,” August 30, 2009)

But in reality, economic activity is far more complex than that. One very important part of that reality the vast variety of goods and services changing hands, in response to constantly shifting tastes, preferences, technologies, and costs. The real economy bears no resemblance to the “hydraulic” one in which the homogeneous “fluid” is units of GDP. For “stimulus” — an increase in government spending (dG) — to generate an real increase GDP significantly greater than dG, several stringent conditions must be met:

a. dG must lead directly to the employment of resources that had been idled by a downturn in economic activity (or newly available resources that otherwise would lay idle), therefore eliciting the production of additional goods for delivery to consumers and businesses.

b. Accordingly, government functionaries must be able to distinguish between unemployment that occurs as a result of normal (and continuous) structural changes in the economy and unemployment that occurs because of a general slowdown in economic activity.

c. To the extent that the preceding conditions are satisfied, dG may be used to restore employment if government functionaries do the following things:

• Ensure that dG is used to purchase goods and services that would have been produced in the absence of a general slowdown in economic activity.
• Ensure that dG is used by those persons, businesses, and governmental units that have become unable to buy those goods and services because of a general slowdown in economic activity.
• Allowing for shifts in tastes, preferences, technologies, etc., adjust the issuance, allocation, and use of dG so that goods and services are produced in accordance with those shifts in taste, etc.
• Reduce dG as the demand for unemployed resources rises, in order to avoid the distorting and disincentivizing effects of inflation.

To the extent that dG is less than on-time and on-target, there is “leakage,” which causes the multiplier to recede toward a value of 1. It can easily slide below 1 — as Barro has found — because of the “leakage” to imports and the disincentivizing effects of government borrowing and spending.

The fourth reason for the failure of the “stimulus” to stimulate is that it is does not address the cause of the drop in AD. A drop in AD usually is caused by an exogenous event, and that exogenous event usually is a credit crisis. Pumping money into the economy — especially when it results in the bidding up the prices of already employed resources — does not reinflate the punctured credit bubble that caused the slowdown.

If a credit crunch arises from a sharp rise in the rate of home-mortgage defaults — as in the case of the Great Recession — the obvious way to “solve” the problem is to prop up the defaulting borrowers and their lenders, and to do so quickly.

But, in practice, the propping up is hit-and-miss, and the misses have drastic consequences. Consider, for example, the decision not to bail out Lehman Brothers and the effects of that decision on financial markets.

Which leads into the fifth reason…

5. Inequity, moral hazard, and their consequences.

Any kind of “stimulus” that targets particular individuals and firms, in an effort to rectify their failures of judgment, has adverse political and economic effects.

Favorable treatment of defaulters and failing companies generates considerable popular resentment, which — in the present instance — has found a vocal and politically potent outlet in the Tea Party movement. Favorable treatment of defaulters and failing companies also creates moral hazard; that is, it encourage unwise risk-taking that can (and probably will) spark future crises, leading the government to assume more obligations and impose more regulations, in a futile effort to change human nature.

All of this adds up to a climate of political contention and financial pessimism — conditions that militate against consumer confidence and business expansion.

6. The human factor.

The preceding five reasons for the ineffectiveness of Keynesian “stimulus” point to a sixth, fundamental reason: the human factor.

Models are supposed to mirror reality, not the other way around. Those who cling to the Keynesian multiplier would like the world to comply with it. But the world does not because it is filled with people, whose behavior is not determined (or described) by a simplistic model but by their responses to incentives, their political predispositions, their informed and reasonable skepticism about the consequences of government intervention in economic matters, and — above all else — their fallibility. And, believe or not, government officials and bureaucrats are no less fallible than the “ordinary” citizens whose lives they would like to organize.

The human factor is an inconvenient truth. But “liberals,” in their usual arrogance and ignorance prefer magical thinking to reality. Belief in the Keynesian multiplier is a prime example of magical thinking.

# Does World War II “Prove” Keynesianism?

In “How the Great Depression Ended,” I say that

World War II did bring about the end of the Great Depression, not directly by full employment during the war but because that full employment created a “glut” of saving. After the war that “glut” jump-started

• capital spending by businesses, which — because of FDR’s demise — invested more than they otherwise would have; and
• private consumption spending, which — because of the privations of the Great Depression and the war years — would have risen sharply regardless of the political climate.

That analysis is by no means an endorsement of simple-minded Keynesianism (as propounded by Paul Krugman, for example), which holds that the government can spend the economy out of a recession or depression, if only it spends “enough” (which is always more than it actually spends). But there is no point in pumping additional money into an economy unless the money elicits productive endeavors: business creation and expansion, leading to net capital formation and job creation.

Pumping additional money into government programs results in the misdirection of resources, at best, and in the discouragement of productive private activity, at worst. Discouragement takes two forms: crowding-out and active interference (usually through regulatory inhibitions).

The answer to the question of this post’s title is that World War II has nothing to do with Keynes or Keynesianism, as it is widely understood. Employment and output (measured in dollars) rose sharply during World War II, but most of the additional output was devoted to the war effort. Huge increases in government spending did not lead to huge increases in the material well-being of Americans, most of whom were working harder while being deprived of the fruits of their labors, through rationing.

If anything, the post-war recovery “proves” the folly and wastefulness of efforts to stimulate an economy through government spending. It was not government spending that re-started the U.S. economy after World War II, it was private spending on capital investments and consumer goods. Some of that private spending was encouraged by the end of regime uncertainty. That end was brought about by the curtailment of New Deal initiatives (until the 1960s) because of the war and FDR’s death. Private spending — which was boosted by wartime saving — would have been purely inflationary had businesses not been willing and able to create jobs and expand output.

# Toward a Risk-Free Economy

If the real economy — which produces goods and services — could be disconnected from financial markets, the Great Depression (and thus the New Deal) and the Great Recession (and thus TARP and “stimulus”) would not be part of history. The problem is that financial markets are a necessary part of the real economy — unless your idea of an economy is one that functions without money, banking (as we know it), credit, and risk-pooling (e.g., insurance companies and corporations).

Money is the root of all financial crises because it eases the buying and selling of goods and services. That sounds good, but money also enables its holders to more readily change their minds about what and when they buy and sell. When Farmer Joe trades wheat to Farmer Jake in exchange for butter, he does so, in part, because wheat isn’t nearly as portable as money. If Farmer Joe gets money for his wheat, there’s no telling what he’ll do with the money from one day to the next. He might even decide to save some of it, thus depriving Farmer Jake of sales that he was counting on and triggering a Keynsian rollback in aggregate demand.

Banks would be okay, as long as they are warehouses for goods and are not in the business of holding money and lending it out. Instead of paying interest, banks would charge customers for storage services.

Why shouldn’t banks lend money? Because lending by banks is a form of credit, and credit is to be eschewed. If money is the root of all financial crises, credit is the thing that allows money to do its dirty work. When borrowers don’t repay their loans, banks (and other lenders) go belly-up, which just triggers another kind of Keynsian rollback in aggregate demand. Government actions to make lenders whole simply transfer the risk of lending from particular depositors and investors to taxpayers at large, whose natural reaction is to spend less now because they can see higher taxes in their future.

Risk-pooling goes hand-in-hand with credit. People who pool their money to underwrite risky propositions (e.g., business ventures) do so knowing that not all propositions will succeed. Obviously, the thing to do is to back only those propositions that are ensured of success, but there’s no way to do that. Solution: Don’t allow risk pooling because it’s too, well, risky.

So there you have it, a prescription for a risk-free economy: no money, no credit, no banking, no risk-pooling. Just plod down the road to Farmer Jake’s place and trade some of your wheat for some of his butter. And don’t worry about the fact that you live in a thatched hut with a dirt floor, drive a rickety cart which is pulled by a rickety donkey, dig potatoes out of the ground, and eat those potatoes (with a little butter) by the dim light of a few home-made candles.

Wait a minute! There’s still the risk of bad weather, which could stunt or ruin your wheat crop. I guess there’s no such thing as a risk-free economy, is there? But don’t tell that to the regulators, you’ll spoil their fun.