Arnold Kling  

Francis Fukuyama on Public Administration

Democracy: Worse Than I Though... Bastiat's Insight on Governmen...

He writes,

anyone who has spent time in government realizes that the real questions that preoccupy officials have to do with implementation, or rather, the impossibility of implementing many desirable policies because of the huge number of constraints under which modern governments work.

I think he correctly identifies a problem, but I am not sure that he is right about the cause.

The problem is that policies are poorly implemented. Fukuyama suggests that our system of checks and balances, rooted in a fear of government, is the cause.

I have been thinking about this a lot lately. I think that the problem is fixable without getting rid of checks and balances or without losing our anti-government bias. I have been composing some essays trying to take the perspective of a management consultant on government.

I think that policies are poorly implemented because the people who make policy know nothing about implementation. The policy wonk view seems to be, "I design policy. Implementation is a detail that someone else can worry about."

The classic example of this is TARP. The original proposal was to buy toxic assets. This is when I first identified the "suits vs. geeks" divide. The "suits" proposed buying toxic assets, and when asked how these would be priced they said, "that's just plumbing." (Eugene Ludwig said exactly these words.) Those of us on the geek side of the divide thought that the pricing problem would be much more difficult. In fact, it proved insurmountable, and TARP in practice took a totally different direction: instead of a program to restore liquidity to the mortgage securities market, it became a pure bailout of large banks.

Another example would be mortgage modification. A lot of prominent economists and other policy wonks have pushed mortgage modification plans. Some of these have been tried by the Obama Administration. As far as I can tell, none of the planners has a clue about how mortgage origination and servicing work as business processes. I guess they think "that's just plumbing." If you had asked me, or anyone else with any industry background, we could have told them that it would take years to implement any of their hare-brained schemes.

If you want policies to be implemented effectively, then policy needs to be made by people who understand something about implementation. Lawyers, economists, and policy wonks have no training or experience in managing large organizations. As a result, policies are enacted without any thought given to implementation issues.

The most important thing to remember about government policies is that they involve large numbers of people. This means that there are management issues.

So a key element in improving public administration is getting public officials to understand the difference between making a pronouncement (or passing a law or writing a regulation) and achieving results. Of course, my inner Hansonian is telling me that public policy is not about achieving results, but that is another story.

Comments and Sharing

COMMENTS (12 to date)
Shangwen writes:

I think it depends on what is meant by implementation. If you ask a lot of public service managers and execs, they'll say that implementation plans are built into a lot of policy discussion. But what they mean by that is orienting civil servants, revising paperwork, publicizing policy changes, and some kind of monitoring.

But this is all done in a bubble where it's axiomatic that the policy is correct. What public servants are discouraged from considering is the public choice element, despite the fact that all policies, large or small, are of disproportionately high interest to certain private groups, and those groups are good at piecing together the benefits of the new policy with other existing ones. This is partly what makes policies fail (from a public good point of view).

You don't have to take a full-bore Tullockian perspective on this to accept that the public choice aspect is destructive. You just have to acknowledge--as many public administrators do privately--that not allowing any room for cynicism, an experimental approach, or the possibility of "diminishing returns to scope", is the enemy of policies that are intended to produce a real public good.

RPLong writes:

Anyone who has actually worked as a consultant to government officials knows that the bureaucrats make it up as they go along. The people who write policy offer the bureaucrats absolutely nothing except a mandate, an authorization for the bureaucrats to go forth and create a new monolith.

I cannot tell you how many meetings I have been in during which massive, expensive bureaucratic implementations are created largely out of thin air as some project manager rambles on about how she thinks it might kind of sort of be good to do something. Next thing you know, it's a reality.

Bureaucracy is an addiction and a disease. It's not as if there is any such thing as "good policy." Politicians create mandates and bureaucrats create red tape. Ordinary people suffer.

All we end up with at the end of the day is an obstacle that ordinary people must first overcome in order to do what they want to do. No one ever said, "It's a good thing I have to fill out a 10-page application to put a billboard up for rent on my corn field."

Mark Michael writes:

An excellent point re: TARP. Recall that TARP not only didn't buy those toxic assets, the Fed later bought them on a dollar-for-dollar basis quietly behind the scenes. All of that discussion on TV at those congressional hearings about those clever means to determine a price for those assets appears to me to have been just a bunch of "fool the rubes in Peoria that we actually will try to look out for the taxpayers' dollar" chatter. If Bernanke had said, "We're going to pay the full, face-value price for those MBSs" all heck would have broken loose.

Let's see, the Fed isn't directly on the hook when those toxic assets go bust, but most are tied back to Fannie or Freddie. Oh, wait, we taxpayers have taken over Fan & Fred, too. So it's us again. Drat. Note how the Fed seems to look out for the banking system first and foremost.

John Samples writes:

In the aftermath of the Great Society, political scientists got interested in how programs are carried out. A major work appeared in the 1970s by Aaron Wildavsky and Jeffrey Pressman entitled Implementation: How Great Expectations in Washington Are Dashed in Oakland; Or, Why It's Amazing that Federal Programs Work at All, This Being a Saga ... on a Foundation. (Here's brief summary: Wildavsky and Pressman was widely adopted and went through three editions by 1984. Some public policy people still do implementation, I think, but not as many as you might expect. Close attention to the details of implementation does not encourage the view that public policy can achieve great changes in society. But the students entering public policy programs hold such views on average. Constraining the illusions of your customers is perhaps not a winning business plan. In any case, the study of implementation does not seem to be a major part of political science these days.

Phil writes:

Isn't this just a more subtle case of "goals vs. policies"? Legislators have a goal in mind, but legislate a policy that may or may not be compatible with the goal. The problem is knowing how to navigate to the goal from the policy, if it's possible.

It's like legislating "cure cancer". The problem for the bureaucracy is *figuring out how to cure cancer* so you can figure out the details of the implementation.

asdf writes:

[Comment removed pending confirmation of email address. Email the to request restoring this comment and your comment privileges. A valid email address is required to post comments on EconLog and EconTalk. We have attempted to contact you. This is our final warning.--Econlib Ed.]

JoeFromSidney writes:

Your comment reminded me of this old joke.

A centipede suffered from arthritis in each of his hundred legs. Seeking advice, he went to the owl for help.

The owl pondered a bit, then said, "I recommend you convert yourself to a quadruped. That way you'll have only four legs hurting."

The centipede replied, "Sounds like an improvement, but how do I go about it?"

The owl replied, "I only set policy. I can't be bothered with details."

Having spent half my adult life in the implementing part of a government bureaucracy, I'm fed up with people who can't be bothered with details.

Floccina writes:

Somethings cannot be efficiently done by written rules or laws. Either the laws need to be very complex and changing all the time which would be a great burden, or people will game the rules or route around them. A Government employee cannot just say those loans look to risky to me, he has to find they break some rule.

Foobarista writes:

The problem is there's no penalty for the bureaucrats if their policies "fail". The assumption is a sort of schoolyard assumption that if you think and plan carefully, you'll come up with the "right answer". If it fails, it's never the fault of the bureaucracy.

There is also effectively no public recourse against the bureaucracy as nobody has a vote on whether to retain a particular bureaucrat or not, except at the very top levels of the bureaucracy. Sure, Congress may have an ultimate vote, but there's simply too much distance from Congress and much of the bureaucracy for Congress to sensibly supervise it (even if we ignore public-choice issues).

This is why one of my little rules is the bigger the government, the less democratic it will be, for the simple reason that big government does a lot and it can't be governed by the relatively small number of government officials.

If you try to introduce more voting and accountability into the system, you end up with something that starts to look increasingly like a market structure, not a classic government hierarchy.

Foobarista writes:

In above, change "small number of government officials" to "small number of elected officials". We have no shortage of government officials :)

Chris Koresko writes:

Arnold Kling: The problem is that policies are poorly implemented. Fukuyama suggests that our system of checks and balances, rooted in a fear of government, is the cause.

I didn't get that from his article. Instead, Fukuyama is concerned that the electorate correctly sees government as wasteful, and that this leads them to reject the higher taxes he thinks are needed to do many worthwhile things.

I think he sees government as a competitor with the private sector for the business of doing good. And government is losing out badly because its implementation is so poor. Fukuyama is advocating that the government team work harder on that so they can compete more effectively and win more business.

NB: The comments on this post are really insightful. Thanks to all!

Bill Drissel writes:

The grasshopper, feeling the cool weather, was in a panic. Winter was coming and ...

He went to his all-knowing friend, the roach, "What am I gonna do? Winter is coming."

The roach replied, "Turn yourself into a roach and spend the winter with me in the kitchen hot-air duct."

"How in the (world) can I do that?"

"Look, grasshopper, that's an implementation problem. Policy is my bag!"

Comments for this entry have been closed
Return to top