David R. Henderson  

Cowen on the Multiplier: Is it Really "Worth"?

Making You Smarter... Society of Lies...

Tyler Cowen recently wrote:

So, in these cases [a government giving aid to to another government], a multiplier of one means that a dollar of aid--the alternative to the fiscal consolidation--is worth a dollar. I find that easy to believe. It's not really a claim about fiscal policy or Keynesian economics.

It is a claim about economics. And it's a mistaken claim. "Worth" has to do with, well, what something is worth. Worth has to do with value, something economists have thought long and hard about. If a government gives another government a dollar and that dollar is totally wasted, that is, spent on something worth zero, then the value, which means the worth, of that dollar is zero, not one dollar.

I'm not claiming that all government expenditures are wasted in that extreme sense. What we do know, though, is that there is no market test and so the odds that the expenditure will be wasted are quite high. Also, economists in the economic development literature, starting, I believe, with the late Peter Bauer, have found that government to government aid is the least efficient form of aid.

Treating a dollar of expenditure by government as being worth a dollar, regardless of context, is an instance of what I call "GDP Fetishism."

Comments and Sharing

CATEGORIES: Fiscal Policy

COMMENTS (13 to date)
Ken B writes:

I assume, maybe wrongly, that Tyler is talking about transfers in today's Europe. Germany gives Greece $1. The Greek gov't will spend that somehow, rather than the German gov't spending it somehow (as an example). Isn't he claiming that more or less and on balance the Greek and German multipliers are the same so the "transfer multiplier" to coin a phrase is 1. If the Greeks had a multiplier of 5 and the Germans 1.5, then the German dollar would do more if spent by the Greeks, suggesting a "transfer multiplier" of 3.3.

Is that right?

Joe Cushing writes:

The multiplier: You might as well be writing intellectual debates about unicorns but since this is a debate that just won't die...

When the government spends money, production is destroyed. This happens because money that would have otherwise been spent on goods and services that pass a market test for value is diverted through debt and force to be spent on things the market has deemed to be not of value. Are there things the government does with money that a market would provide if it were given a chance to? Sure; but we can rest assured that the government, which faces no market check on its spending, will destroy value even in these areas.

I can't believe there is anybody out there would would suggest the multiplier is anything at or over one. It is obviously and empirically less than one. It is also an unknowable number. Some say the multiplier is less than 0. This means that not only is the production value every dollar the governments spends destroyed but additional production is also destroyed along with it.

There is some logic behind this less than 0 argument, although it is not a provable hypothesis at this time. Often when the the government spends money, it spends it on things that have no productive value. The salaries of the IRS employees or the web of "regulatory" agencies have no productive value for example. I put regulatory in quotes because the agencies don't actually regulate anything, they just write rules. Often the rules destabilize rather than regulate. In any case all of the salaries of many people who work for government are wasted and these people's jobs inherently reduce the ability of the private market to do its productive tasks. So not only is the money diverted to valueless tasks of government, it is diverted to tasks that sabotage the ability of the private market to create valuable goods and services.

Craig Richardson writes:

@Joe Cushing- although I am highly sympathetic to free markets and their relative efficiency, and in general David's claims, there are some things the government does where the multiplier effect is definitely over 1.0.

Consider a country where property rights are enforced through tax dollars providing a judicial system and policing system, that replaces communal property or ownership by theft and force.

As H. deSoto has rightly pointed out, there are huge multiplicative gains to small investments in government-enforced property rights. The network of economic effects starts with collateral, the building of capital, and most importantly, the building of social trust, i.e., "I will build this and it will not be stolen from me."

Where this runs awry, as in Zimbabwe, the effects are disastrous, as I have pointed out in a number of articles. The withdrawal of property rights led to Zimbabwe losing nearly half its annual GDP in just a decade.

If government enforcement of rules and regulations didn't matter there, the economy would have sailed on quite nicely.

Doug writes:

"When the government spends money, production is destroyed. This happens because money that would have otherwise been spent on goods and services that pass a market test for value is diverted through debt and force to be spent on things the market has deemed to be not of value."

Look, I'm an extreme radical libertarian anarcho-capitalist. But even I'm willing to admit that not all government spending is equally destructive, and as a simple corollary not all government is totally destructive.

The second thing that we have to acknowledge is that market failures do exist, so it's possible for government intervention to actually increase social utility even if it doesn't pass the market test.

Now the reason I'm staunchly against government intervention in the case of market failures is that political failures are almost always more problematic than any purported market failure they claim to solve. Using the state to fix a market failure is like trying to solve a rodent problem by releasing a bunch of snakes in your house.

Monopolistic behavior by some industrialists: let's create an anti-trust act that will be used to protect inefficient firms without regard to consumer welfare. Irrational consumer behavior with regards to addictive substances: let's throw 25% of all poor minorities in prison. Adverse selection in health insurance markets: Let's nationalize the entire healthcare industry.

But we must acknowledge that the occasional government intervention does in fact do more good than harm. Which isn't to say this is the outcome we'd expect ex ante. It's simply luck and good fortune that the process wasn't corrupted that one time. Trying to repeat new interventions by matching the few previously successful old interventions is like buying yesterday's lottery numbers.

Nor is it to say that the few successful government interventions wouldn't make the tradeoff of tearing down the leviathan state we have completely worth it. But it is saying that there are certain government activities that I'm far more comfortable with than others. And that even as an extreme anti-statist I'd probably leave some government activities alone at the margin.

I'd say spending $1 on the CDC is probably worth more than $1 in terms of net social utility.

On the other hand spending $1 on the EPA probably destroys close to $10 of net social utility.

Joe Cushing writes:


I agree that certainly not everything the government spends money on is equally destructive and of the many millions of things it has spent money on, some of the are assuredly to be productive. I'm speaking on averages.

Craig Richardson

"If government enforcement of rules and regulations didn't matter there (Zimbabwe), the economy would have sailed on quite nicely."

You don't necessarily have to have the state run enforcement of rules and regulations. What we see in places where the state collapses is that there is a vacuum in this area. In the US the government is 40% of the economy and it controls much of the private market as well. If the government collapsed, it would be chaos and anarchy would be blamed for it. Anarchy is not the problem, it's the sudden loss of needed systems that is the problem. Nearly everything the state does, the private market/voluntary activity could better. Nearly everything else does not need to be done.

People often point to Somalia as an example of how a state is important in bringing about prosperity. They say how terrible things are there without a state. I don't know that Somalia was ever prosperous but I have heard that some people are finding life easier without a state to get in the way of their business.

I'll say it again, the government does not write regulations. The government rites rules through law with the stated intent of regulating but there is no evidence that these rules make anything regular. Often stated regulations actually make things irregular. The FDA is a great case in point. So is the EPA.

Joe Cushing writes:

If the government ran all grocery stores and then the government collapsed, there would be a sudden loss of the ability for the country to distribute food. This would be huge chaos and a huge catastrophe. This would not be evidence that anarchy caused this problem and that the government is needed to be in the business of selling food. It would take a while before private actors could take over and set up new stores.

The same can be said of all the other things government does. The fact that government has a monopoly or near monopoly on police, fire, schools, roads, food safety, etc, does not mean that these could not be done in a voluntary society but if government collapsed, all of these services would suddenly be removed from society and it would be chaos.

Craig Richardson writes:

Joe writes, "People often point to Somalia as an example of how a state is important in bringing about prosperity. They say how terrible things are there without a state. I don't know that Somalia was ever prosperous but I have heard that some people are finding life easier without a state to get in the way of their business."

Yes, those "people" are the ones with guns, highly powered boats and lots of cash from ransom money. Would you feel comfortable taking a sailboat along the coast of Somalia to see how well the market is working there under a lack of government? You probably would learn from the hostages about free trade.

And I guess the lack of running water and sewage pipes is just the market valuing hostage running over sanitation.

I'm not sure if you have traveled to Africa, but I think it would help for you to go, to understand the importance of relatively good government (visit Ghana, for example) and how the quality of life differs from such despotic regimes such as Somalia.

I loved Ghana- its energy, entrepreneurial energy, and quality of life. A democratic government, with improving governance indicators (World Bank), foreign investment and safety because of honest police make this a place that will thrive in the future. Foreign investors look at all these things from the outside, and place their chips in areas with lower risk, i.e. stable governments. Lack of government represents the unknown, therefore higher risk.

Joe Cushing writes:

"Yes, those "people" are the ones with guns, highly powered boats and lots of cash from ransom money."

The post I read was talking about people running actual businesses. The pirates, are actually a form of government taking root by taxing shipping in an area of water they claim. That's what government is, a system of organized theft carried out by threats of violence. If this keeps up, perhaps some of the pirates will work out deals with the shipping companies to pay the tax without having to hold anyone hostage. They sell this by telling the shipping companies they will spend their time going after going after other pirate organizations, shutting them down. This is something else governments do, they go after thieves who are competing with them. The shipping companies would like this but they would still be shaken down by the one group of thieves that emerges on top--the government.

Once the government eliminate most of the other thieves they will need to come up with a something else for their patrols to do because they are already on the payroll and they have the power to overturn the government. Maybe they will take to imposing their rule over the land. They can collect taxes (protection money) from people on land too. Eventually they will have to buy the people off by providing them something for their tax money. Along the way, there will probably be wars for competing gangs (governments)and eventually one gang might emerge as the winner of the fight. That gang then be recognized as the government by outsiders.

All of this could be avoided if the governments of the rest of the world didn't outlaw the shipping companies from being armed. The ships should have cannons on them. The should have 50 caliber machine guns. How easy it would be to defend a massive ship against a tiny boat if the shipping companies had the right to keep and bear arms.

Doug writes:

Claiming that anarcho-capitalism could never because of Somalia, is like trying to say that electric cars are physically impossible because your car doesn't start if you don't put gas in it.

Shayne Cook writes:

@ David R. Henderson:

Your "GDP Fetishism" article is wonderful. I'll be forwarding the link to your article to lots of folks I know, and all my future students. Also valuable in that regard was Arnold Kling's references that the U.S economy is NOT just a "GDP Factory".

Thanks to you both. (I generally just emphasize that GDP is a metric, not a goal.)

David R. Henderson writes:

@Shayne Cook,
Thanks very much.

Bostonian writes:

Robert Batemarco wrote an essay "GNP, PPR, and the Standard of Living" (available online) expanding on Rothbard's ideas. Rothbard suggested that government spending be subtracted from GNP to get the private product remaining.

Do measures of national income that downweight government expenditures correlate better than GDP with how people rate economic conditions? If so, that would argue for emphasizing such measures.

John Papola writes:

The so-called "multiplier" as well as it's cousin, the marginal propensity to consume, seem to be overwhelmingly useless and nearly tautological ex-post constructions. They're also essentially impossible to measure in any sense that reveals useful information for understanding what will happen in the future. There's no mechanism to prove causality at all. The entire enterprise amounts to a child playing with the controls on an arcade game running the demo mode. He didn't put in a quarter. He's not moving the characters, even if his actions sometime correlate with on-screen movements. It's a joke. Only worse. The child doesn't hurt other people with his confusion. Economists do.

No doubt, as David notes, the lack of voluntary exchange and mutual gain renders our judgement of the true value impossible when government spends. But there's also the problem that we live in a dynamic world. People react. They change. Every single day they do. So to make assumptions about what everyone in the aggregate will do as a result of "spending", based on the near-worthless past "data" is fully worthless.

Economists was MUCH better when it was being done by people who weren't trained in economics, like Adam Smith, David Hume, John Stuart Mill and David Ricardo. We're not progressed in our understanding of macro beyond them. As Robert Skidelsky told Russ and I after his interview, economists is not a progressive science. Boy is he right.

Comments for this entry have been closed
Return to top