Arnold Kling  

Some Questions About Government

Sins of Omission: What's Wrong... Is Bank Regulation This Easy?...

These are rhetorical.

1. Is government subject to diminishing returns to scope? In the business world, it is usually considered a better strategy to stick to one purpose rather than to constantly get into new lines of business. The thinking is that if you try to combine too many businesses, you end up being ineffective. Does this consideration apply to government? If not, why not?

2. Are government monopolies efficient? In theory, in the business world monopoly is efficient, because it eliminates duplicate overhead. (Monopoly is inefficient in theory because the monopolist charges a price that is too high, but we might suppose that government will not do that.) In practice, however, monopoly is inefficient because without the pressure of competition, business practices tend to stagnate. Is government immune from this stagnation problem, and if so, how?

3. Most new businesses disappear within a few years. Most government programs persist. Does this persistence indicate that government is more effective than the private sector at choosing carefully which initiatives to undertake, less effective at choosing which initiatives to terminate, or both?

4. Because of the profit and loss system, businesses are accountable to some extent for keeping their promises. (There are weaknesses in accountability mechanisms, to be sure. Most notably, an executive with a short-term focus can gain personally while making decisions with adverse long-term consequences.) In government, the main accountability mechanism is an election. But most government workers are not subject to elections, and elections are very crude expressions of voter preferences. Overall, is the accountability mechanism in government nearly as effective as that in business?

I do not believe that government is necessarily evil. I do not believe that liberty is the only good. I do not believe that individuals are always rational. I do not believe that markets are perfect. What makes me lean libertarian is that I have no enchantment with big institutions in general or with government in particular.

As corporations become large, they become more powerful in some ways, but they also become clumsy. As their behavior deteriorates, competitors will provide me with alternatives. As government gets large, I see the clumsiness. What I do not see is any tendency for the clumsiness of government to be corrected by competitive forces. I think that if instead of romanticizing government or treating it as an abstract solution ("when problem X occurs, we need government to fix it"), we need to evaluate it in terms of its institutional reality.

Comments and Sharing

CATEGORIES: Political Economy

COMMENTS (14 to date)
david writes:

Do any of these arguments apply to redistributive transfers and coerced risk-sharing instead of government-provided or -monopolized services, like the Post Office? These do, after all, account for the bulk of non-military state spending in most states. Services do not.

#2, #3, and #4 should vanish if the state "neoliberalized" and subject most state-funded services to competition, in a democratic way like Sweden or in an undemocratic way like Singapore. The problem in the US has always been doing so in a way that preserves the pre-existing social bargains over the degree of transfers.

Peculiarly, libertarian rage in the United States is most focused at the federal level, even though many key non-transfer programs are legislated at a local level. This doesn't say anything good about the American brand of libertarianism (insert snide remark about Rothbard's malevolent influence here, esp. compared to Friedman's). Compare the (successful!) movement toward devolved authority in Canada or the United Kingdom.

Arnold Kling writes:

These arguments may not apply to an abstraction called "redistributive transfers," but they do apply to Social Security, Medicare, and Medicaid. These real-world programs are badly structured on both efficiency and equity grounds, and they are part of government's entrenched monopoly.

I have plenty of local-level libertarian rage. As I have written many times on this blog, my county government is a wholly-owned subsidiary of the teachers' union. In this case, "devolved authority" is not a solution.

sabre51 writes:

Might be my favorite Kling. The thing that makes me libertarian is that we DON"T just rally for "liberty" or against "evil government," even though we are often accused of doing so. (Humorously, in my opinion, since Dems and Reps almost never get called out for simply rallying around "equality" and "the self made man/traditional values," respectively) It is more about questioning the common perceptions of things, like "when problem X occurs, we need government to fix it." You won't see these types of questions asked by Dems/Reps. Even though libertarians have widely varying beliefs, I think they are very good about focusing on things that actually are debatable (i.e. "what is the correct level of redistribution of wealth in society?"), whereas Dems/Reps often endorse positions which are clearly inferior to others. (Capital gains taxes vs. simple increased progressive taxation, tariffs and subsidies vs. free trade)Thanks for the post, +1

JKB writes:

The clumsier government becomes the more likely it is to eliminate or move to tightly regulate private sector competitors. Or as we see in the US, the federal government effectively comes to control state agencies through funding and mandates.

Milton Friedman covers most of the points in this short talk on socialized medicine. I like the idea of "Theory of bureaucratic displacement," in which useless work replaces useful work when a bureaucracy takes over.

As to your points,

1. As the government pie is fixed, expansion of programs leads to a diffusion of resources (in the short and medium terms) and all chasing the latest initiative even if weakly linked to their mandate for funding leading to a muddling of purpose. Google "homeland security fisheries" for examples.

2. Government can retard stagnation through the military system of a steady stream of incoming graduates, constant churning of assignments, and an up-or-out process that inhibits homesteading. Even that doesn't protect them from sudden competitors which forces a complete revision of doctrine and processes. The system is however incompatible with current civilian employment law and is facilitated by a probably unsustainable retirement system.

Seth writes:

Good questions.

I think a lot people intuitively believe that election accountability is more effective or safer than profit and loss accountability. I believe because it seems more transparent to them. They get to make an explicit choice.

However, they rarely ever consider how satisfied they are with what they get from the profit and loss system in relation to their satisfaction level from the election system.

Or, put another way, we don't have Sunday morning talk shows about iPods and what our grocery stores stock.

wd40 writes:

"What I do not see is any tendency for the clumsiness of government to be corrected by competitive forces.."

There is competition for running the government. There is competition between the Republicans and the Democrats and there competition within each party to gain the nomination. There is competition between levels of government and there is competition between government services and privately provided services. Voters can choose bigger or smaller government. So your monopoly analogy is misplaced.

joeftansey writes:

Can't resist. Here is the entire philosophy of government.

(Problem)? Government should fix (problem).

That's right folks. All this hubbub about existential freedom to make any sentence you want! At least its grammatically correct. Most of the time anyway.

Mike Rulle writes:

People tend not to get overly angry if their life is okay. Hence, they can live with most government inefficencies even if it is something they rationally disagree with.

In my experience, Democrats and the right have two different reactions to your "Bastiatian" rhetorical questions, and one common one.

The common one---no one understands or believes in the concept of opportunity cost and inefficiencies.

The Democrat One---Government has good intentions and the private sector does not---and they tend to be direct beneficiaries more so that GOPers. Plus big government types are pro choice and that seems to matter a lot.

The Republican One---This party is really Democrat light. The "left" has won the moral argument of Government has good intentions---hence the weird phenomenon of billionaire big government types. Republicans simply like status quo---whatever that is. They have been the proverbial frog being slowly boiled for 50 years.

Troy Camplin writes:

Rhetorical, perhaps, but worth answering:

1. Yes. We can see this in the prisons, where murderers and rapists are released to make room for drug dealers and users. The few legitimate reasons for government -- the areas in which coercion is valid -- are done increasingly poorly as government expands to do more and more things that do not require coercion to get done.

2. No. Without competition, there is no discovery procedure, and thus new and more efficient ways of doing things are not discovered -- at least, not as rapidly as they would be under the pressures of competition. This can be satisfied if there are competing governments, as the U.S. was initially conceived. But the more governance is taken over by the federal government, the less competition there is.

3. Without profit and loss, effective choice of programs is impossible. Instead, the worse the program is, the more money is thrown at it, under the misperception that the problem is lack of funding. Philanthropy faces similar problems -- though if givers do not believe the programs work, the charity will lose money, and eventually shut down. So there is at least some accountability. Buried in bureaucracies, government programs have practically no accountability, and the bureaucrats have the incentive to keep their programs, to keep their jobs.

4. No. See #3.

MMJ writes:


I think Guinevere Nell's book "Rediscovering Fire: Basic Economic Lessons from the Soviet Experiment" answers some of these questions (and very well, too).

I have no affiliation with the author or publisher, I simply enjoyed the book and consider it relevant.

Becky Hargrove writes:

Great post, had to take a full page of notes and think it over. What seems relevent to me as a pro-active libertarian is that government is not so much the problem as the way that government gets expressed in society: something that exists outside of ourselves. Instead of holding town hall meetings to debate future direction of resources and individuals, we hold town hall meetings to see if we can keep out whatever we do not personally like, regarding someone else's plans.

In other words, government has taken the relevant issues of our lives and externalized them, in the failed attempt to simplify them. In some respects(entitlements) it has tried to do this culturally. However, should we take back cultural traditions, that also means a new dialogue is needed: how do we replace the economic freedoms government once gave us, if we do those things ourselves?

Just as importantly, government has tried to rely upon experts to provide appropriate and adequate answers to many aspects of our lives. However, the fact that those experts do not agree with one another most of the time needs to be taken into account at a very basic economic level. Again such responsibilities need to be internalized at the individual level and we have the technology to do it. The experts do not have to agree with one another. However, the public needs to understand what the experts are saying to a degree that the public can also vote on the outcomes that the experts disagree upon. In fact, this is the most important education of our times: the very dialogue of the experts that we need to understand, so that we can vote - anonymously - on those outcomes ourselves. That way, no expert need undermine their own views by aligning with other experts on public solutions, and the spontaneous alignments by the public on such matters can be far more effective than voting on presidential candidates.

If government is ever going to be about us, it has to be internalized, pure and simple. Otherwise all those years of education we get are almost completely squandered.

EBK writes:


what if we introduce market signals to goverment programs? For example, competitive bidding for medicare. Also, what about requiring each program to get renewed after a specified period of time?

Maurice Halton writes:

OK then, for Steve Davies I offer my socialist comments.

Government’s role is not analogous to a business, unless a large national park can be described as a commercial enterprise, which it clearly should not be. A national park is run for the benefit of its inhabitants – all of them – and those who run it are custodians whose job it is to maintain a balance, so that one species doe not take over sections of the park, or destroy so much of the environment that other species suffer as a result. Businesses are by nature and by definition selfish and greedy.

Government monopolies, where they exist, usually do so because they provide a service or product that private enterprise cannot exploit profitably. To suggest that private firms could compete to run the police and the armed forces would be a novel, but absurd, proposition. The Friedman inspired Chilean privatisation of education is an interesting case: has it worked?

The comparison between government and business in terms of life span is bogus. Although Government lasts forever (unless someone can point to somewhere that anarchy prevails), governments or administrations rarely last more than a few years.

Businesses are accountable only to their shareholders, and only to some of them. They are in fact more directly accountable to the markets they purport to serve. In a truly ‘free market’ system – and the USA certainly is not one of them – there would be many more company failures than in fact there are. Are private company employees subject to some kind of election? Again, the comparison seems rather contrived.

One man’s liberty is another man’s tyranny.

Maurice J. Halton (happy new year)

Michael Strong writes:

Great post, pretty much sums up my perspective. Sometimes I try to explain that I'm against "large scale regulatory monopolies" but that is not very sexy. On the other hand, if I say that I'm a "libertarian" then some people think that I believe all sorts of ideological Rothbardian nonsense.

For me, once one understands Hayek's Millian argument in "The Creative Powers of a Free Civilization" along with public choice theory/public ignorance, then it is hard not to land pretty close to where Arnold is. I'm always surprised when I find people who have been exposed to both sets of arguments (innovation requires freedom and public choice/public ignorance) and don't land here.

Maurice Halton's romantic notion of government is stunning: "A national park is run for the benefit of its inhabitants – all of them."

Really? Has Maurice not read the literature on how government has mismanaged Yellowstone, or the billion plus dollars that government spent subsidizing the destruction of Tongass National Forest, or the Army Corp of Engineers draining more wetlands in one year than all private developers combined in an ineffective flood control project, or for that matter the role of the Army Corp of Engineers in the Hurricane Katrina disaster, etc., etc., etc.? Maurice, there is a vast literature on how government actually works, and essentially unlimited examples of it NOT acting to benefit the public good. You can claim that on net it is still not evil - that is an empirical issue - but to claim that governments act in the public interest because a civics textbook said so is simply delusional.

Regarding privatizing education, I know it is possible to create far more effective education for inner city children, but that the bureaucratically controlled system will not allow it to spread. See this for a basic introduction,

Chile may have allowed some flexibility in schooling, but even their private schools are essentially government controlled. By some accounts there were improvements even in that system, but most nations that have some school choice (including New Zealand, Holland, and Sweden) have systems in which the state essentially dictates what the substance of "education" is supposed to be. In order to reap the benefits of educational innovation, we can't define education to be more or less identical to what government schools already do. The world of IT would not be innovative if IBM had been a government monopoly in 1970 and then "privatized" by allowing other companies to produce IBM mainframes to government specs. In such a universe there would be no Steve Jobs, no Bill Gates, no Bill Joy, etc. We'd have mostly bureaucratic types creating quasi-IBM bureaucracies, and everyone would say that "privatization" failed because it didn't result in innovation.

Comments for this entry have been closed
Return to top