Rationality in government.

This post is a response to Jason Antrosio’s comment on my previous post, about the persistent conviction that success in business is an indicator that you’d be good at governing, too. He references Richard E. Hartwig’s book on the politics of transportation infrastructure in Colombia: “The author argues that the whole idea that you can find ‘efficiencies’ or ‘rationality’ in government such as in business is misguided. Because government encompasses everyone in the population (or should), it operates on a very different rationality than that of the consumer.”

You can talk about rationality and efficiency in government, but you have you to be careful what you mean by those terms.

I find it useful to think about two distinct types of efficiency:

1. Are you pursuing the right goals?
2. Whatever goals you’re pursuing, are you achieving your end at the least cost?

(And if you’re wondering about what quality you’re getting for your low cost, the level of quality can be included as part of the choice of goals in the first point.)

Idolaters of the market assume that markets achieve both types of efficiency through the wonders of competition.

For a given goal, if your company reaches that goal at less cost than mine, it can offer its product or service at a lower price, so I will be forced to copy you or go out of business.

And if your company is offering products and services that people want more than the products or services that my company offers, you’ll have more revenue than I will, and again, I will be forced to copy you or go out of business.

I think it’s almost tautological that the second type of efficiency is an unmitigated good. Whatever you’re doing, do it with the least inputs possible (while still meeting your chosen standard of safety/quality), so that you have additional means to pursue additional goals as well.

And in principle, competition should move us in that direction, though it can be a messy process. For example, naïve theory says that an unregulated, uninspected food system should offer safe food because of the damage to your reputation from killing your customers. In practice, as the 19th century showed, this is a very ineffective—one might almost say “inefficient”—mechanism for cleaning up the food supply.

The problem with the choice of goals is starker.

I phrased the market-idolater’s case for goal choosing in terms of companies thriving if they offered the products or services that people want. But that’s hand-waving across an important distinction.

There are things that we value—e.g., street lights, clean air—that can’t be provided at a profit, generally because they’re “non-excludable”: if I provide street lighting for Jill, I have no practical way of keeping Jack from having it too if he happens to be in Jill’s vicinity. These are, of course, the things typically referred to as “public goods.”

Within the set of things that aren’t public goods, markets do a good job of catering to whatever it is that people want, with the caveat that the wants and needs of a person with a lot of disposable income count for more than the wants and needs of a person with little spending power.

But markets are structurally incapable of providing public goods. If a profit-seeking firm tries to provide a public good as part of what it offers, it will generally be at a disadvantage to a firm that offers only the “excludable” portion, without the public good.

So the only ways to provide public goods are either through volunteerism or through government action.

“Efficiency” sometimes seems to be used to mean “it makes money,” in which case the provision of public goods will always be “inefficient.”

But if we step back to the original question of, “Are we pursuing the right goals?”, then we can see that leaving all choice of goals to the market is in fact inefficient.

The attraction of leaving it to the market is that it allows us to avoid potentially messy discussions of values and priorities, because the category of “public goods” is much larger than obvious cases like street lighting and clean air.

Health insurance obviously can be provided by the private market on its own, but the imperatives of profit guarantee that if you are old, or have any one of a range of pre-existing conditions, any insurance available to you will either be worthless (not covering the things that matter to you) or financially unbearable.

Affordable insurance that covers people with low incomes, people with various health conditions, people with a lot of years under their belts, can only be provided through government action. It doesn’t have to be provided by the government, but in that case the government still has to create a set of rules guiding the actions of consumers (e.g., the mandate to buy insurance) and of insurance companies (e.g., the essential health benefits that have to be covered), and the government has to cover some or all of the cost for households below some income threshold.

(I lay this out more extensively here, here, and here.)

But once we concede that role for the government, we have to have a discussion about what counts as an “essential health benefit,” and we have to decide what size subsidies go to people at what income levels, etc.

In other words, we have to get into the nitty-gritty of policy, of how it works (or doesn’t), of how it affects people. We have to decide that A is more important than B, and maybe we have to convince people that A is more important than B.

It’s so much simpler to just say, “Let the market do it,” and that feels like we’re not expressing any preferences at all; we’re just letting people make the choices that they think are best.

But we are making a choice when we “leave it to the market.” We’re choosing not to have public goods. And that is often a bad choice.

(This is related to a long-standing conservative miscomprehension about the role of government, as discussed here.)

What about having the public sector choose the goals, but then letting the private sector provide the product or service? That way, in theory, you get the efficiency of the private sector while not neglecting public goods.

The practice is much more complicated than the theory.

Take the example of running a system of water treatment and water delivery pipes.

You could have a fully private system, which would have the predictable result of serving wealthy neighborhoods, presumably with a pretty high level of service, while ignoring poor neighborhoods.

So instead the government ensures that poor neighborhoods will be served, but contracts the work out to a private company. “We want water service provided to the city, including these poor neighborhoods. Make your bids, and we’ll choose the least expensive one that meets the following criteria for extent and quality of service.”

At this point you’re faced with a fundamental conundrum. The private sector achieves efficiency by finding different ways of doing things (the second element of efficiency described at the top of the post), or by finding different things to do (the first type of efficiency).

As soon as the government specifies what its private-sector partner needs to accomplish, it has cut off the first type of “efficiency,” because it has chosen the goal. And the more narrowly it specifies that goal, the more it reduces the private firm’s scope for finding the second type of efficiency. But if it backs off and gives the private firm less guidance as to what needs to be accomplished, it increases the odds that the private firm will reduce costs by providing the service in a way the government actually finds unacceptable.

On top of that, there’s a catch-22. In addition to being charged with inefficiency and incompetence, government can be corrupt and unfair: the city runs the water system, but it puts lots of money into serving wealthy neighborhoods while neglecting poor areas. How much better it would be to remove the corrupt, incompetent, unfair government from providing a vital service like water supply and turn it over to those efficient folks in the private sector?

But what entity is it that decides on the parameters of the bid requests that it puts to the private sector?

The government.

What entity is it that chooses the winning bidder?

The government.

What is the entity that has to examine the work of its chosen partner to see if its meeting its contractual obligations, and to fine or dismiss the company if it is failing to live up to its commitments?

The government.

Doing all those things well requires a government that is reasonably competent, reasonably fair, and not too corrupt.

But if you had a government that met all those criteria, it could probably do a pretty good job of running the water system itself.

So either you have a decent government that can provide needed public services, or you have a bad government that is both incapable of providing the service and unable to or uninterested in overseeing a privatization process that would actually achieve a good outcome.

In the end, there’s no getting around the need for a reasonably competent, clean, efficient government if society is to be able to function in a somewhat efficient, rational way.