Clifford Winston on Market Failure and Government Failure
Dec 28 2009

Clifford Winston of the Brookings Institution talks about the ideas in his book, Market Failure vs. Government Failure, with EconTalk host Russ Roberts. Winston summarizes a large literature on antitrust, safety regulation and environmental regulation. He finds that government regulation often fails to meet its objectives. While markets are imperfect, so is government. Winston argues that idealized theories of government intervention based on textbook theories of market failure are not the way regulation turns out in practice. He argues that special interest politics explains much of the disappointing outcomes of government regulation.

Cliff Winston on Transportation
  Cliff Winston of the Brookings Institution talks with EconTalk host Russ Roberts about his recent article in the Journal of Economic Literature on the U.S. transportation system. Winston argues that the while the United States has a very good...
Don Boudreaux on Market Failure, Government Failure and the Economics of Antitrust Regulation
Don Boudreaux of George Mason University talks with EconTalk host Russ Roberts about when market failure can be improved by government intervention. After discussing the evolution of economic thinking about externalities and public goods, the conversation turns to the case...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Per Kurowski
Dec 28 2009 at 7:23am

“He finds that government regulation often fails to meet its objectives”

Sometimes the problem is though that the objectives are not even specified.

In the current bank regulations produced by the Basel Committee there is not one word on the purpose of the banks… because stability or better phrased the avoidance of excessive instability cannot of course per se be an objective but only a desired feature.

Greg Ransom
Dec 28 2009 at 1:33pm

I note that Clifford and Russ have identified a market failure in the market for data on anti-trust … but the question become figuring out how much of this “failure” is government failure and how much of this failure is market failure. How do we distinguish this?

A difficult “identification” problem that applies to all cases of “market” and “government” failure.

I’m wondering if Winston has address this problem of identifying and distinguishing between what is due to “market” and what is due to “government” — in a world where these are hopelessly inter-incriminated.

Arnie Kriegbaum
Dec 28 2009 at 7:09pm

This is a useful exploration of the concept of Government failure, and is subtle in presuming that a market and a government are not the same thing. That may seem obvious to many here, but remember that this is not obvious to many people.

Still, I would go one step further than the conclusion that there should be some ongoing experiments with privatization of things like Midway airport. In the bus/transit arena, instead of awarding the whole system to one private operator, a city could award sectors of the city or certain lines to different companies and see who produced a happier bus passenger overall.

Mads Lindstrøm
Dec 28 2009 at 7:35pm

I am wondering when something is government regulation, and when it is something different. Is having a police force government regulation? Is having courts enforce contracts government regulation? Is patents law? Is copyright law? Is anti-drug laws?

I hope somebody has a concise answer to my question.

Thank you for a great podcast,


Richard Knox
Dec 29 2009 at 1:33am

I’ve read the book you discussed with interest. It reminds me of Fisher’s statement that the market had reached a permanently high plateau, on the eve of the 1929 crash. In the midst of a market failure of massive proportion and a governmental failure of equal size, you give a Panglossian picture of the world and praise the Fed as being one of the few good institutions of government.

As a minimum it seems that you should have discussed the disconnect between your 2006 view and what turned out to have been happening.

How can you guys even discuss this book without acknowledging how far off the mark it was?

Have I missed something?

In addition to regulatory capture I guess we need to also catalog think tank capture.

Have you since written something that deals with the market/government failure that currently has 1/6 of the population out of work and millions of homes in bankruptcy?

I’ve reposted since the system told me it was rejecting? I am on sbcglobal and thus I get a new url each night. As a result your system may have me confused with some one who has used this url before?

[There was a temporary hold on your comment unrelated to your being on sbcglobal. If you post using the same email address again, your comments will probably not be temporarily held up. Sorry for the temporary delay.–Econlib Ed.]

Russ Roberts
Dec 29 2009 at 3:58pm

Richard Knox,

There are two sentences in Winston’s book on the Fed. Yes, he suggested it was well-run. That turned out in 2006 to be overly optimistic. Winston’s assessment of the Fed, an assessment made in passing, is not exactly a focus of the book.

There are many podcasts at the EconTalk archives on the Fed’s recent performance. You might enjoy the conversation with John Taylor:

Dec 29 2009 at 7:41pm

The discussion of OSHA reminded me of a friend some years back who had a construction company that was building a new Walmart in town. He said OSHA forced them to put fire extinguishers on the concrete and steel infrastructure before the walls were up…so after the first round of fire extinguishers were all stolen they put them back up to please OSHA but locked them to their respective steel girders and only a foreman had the key…so in the unlikely event that the concrete and steel caught fire nobody on the scene was likely to benefit from the fire extinguishers at least until the guy with key arrived. I wonder if the workers felt safer.

Jim Labbe
Dec 30 2009 at 6:02am

Russ and Econtalk,

Apart from some reference to municipal transportation policies (which I would be very interested in learning about), Winston’s analysis appears to focus almost entirely on the efficacy of federal regulatory policies. In my area of work- local environmental policy making at the state and local level- I see what I believe are real benefits and progress in environmental improvement driven in part by local and regional government policies that employ a mix of market-based and more regulatory mechanisms. In many cases these are a direct response to federal laws and an effort to avoid federal regulatory standards which are recognized as blunter and less adapted to regional circumstances. From my vantage, this appears to be a very positive benefit of federal environmental laws that are implemented locally.

On a related note, I found it interesting that Winston’s solution to what he identified as the core problem, corruption and regulatory capture, is categorically less government regulation and more privatization (with few caveats about being smart about implementation of any policy). The notion that this will somehow free us from corruption is very naïve and perhaps even dangerous because the design of market mechanisms and the private contracting of public service is itself subject to corruption and capture.

We might term this “deregulatory capture” where-in special interests control the elimination of public interest laws to the benefit their interests.

I am not an expert in this area but I have read about and witnessed plenty of cases of deregulatory capture at the state and municipal level wherein privatization and deregulation schemes- poorly overseen by the public- have cost huge amounts, yielded little or no benefit, or undermined positive government regulation protecting real public interests (especially in environmental protection). At least part of the current financial crises is probably related to de-regulatory capture.

In my view and experience, there is really only one way to solve the problem of corruption and regulatory or deregulatory capture: more democracy and citizen oversight in the making and implementation of policy at all levels of government. I strongly suspect that, more than anything, this has produced the positive results in the local and regional efforts to improve water and air quality in my community. As Aristotle observed centuries ago: “If liberty and equality, as is thought by some, are chiefly to be found in democracy, they will be best attained when all persons alike share in the government to the utmost.”

But Winston’s solution to the core problem of government failure- regulatory and deregulatory capture- is really the opposite: government can’t solve anything so people should just not put their faith in government solutions and instead choose privatization and exclusively market oriented solutions. This strikes me as a radical oversimplification which in practice will only rationalize apathy. It doesn’t encourage people to take an active part in solutions through the process of self-governance. That is probably not his intent, but in the political world that is often the effect- and likely the intent- of politics that demonize government. All this is to say that anti-government politics and policies- absent an engaged public- have an equally poor record of advancing freedom and the public interest in efficiency, equity, and quality of life.

One final and perhaps related question for Ecotalk: When are you going to apply your very useful and engaging perspectives to questions of immigration policy? (Or have you already and I need to explore some earlier podcasts?). I can’t think of an area where government regulation and policy more impacts basic human freedom: that basic natural right to move around on this great green planet of ours.

Jim Labbe

Dec 30 2009 at 2:32pm

Did anyone else have trouble with the link to the book? I couldn’t download it, and I could only find a paid version on the Brookings site.

Dec 30 2009 at 10:10pm


When I clicked the link it downloaded as a .php file. I changed the extension to .pdf and it opened fine.

Bill Gates
Dec 31 2009 at 6:48am

Russ, the briefest search of your bio shows you to be a libertarian, austrian school free market fundy.

Yet in my own research I’ve yet to find a successful libertarian society any time, any where, in the history of the Earth.

Could you cite one for me?

Thanks in advance for not deleting this comment like the others.

If you want extra credit, how do you reconcile your beliefs with working at a 501(c)3?

[This comment has not been deleted, but if you want your future comments to not be deleted, please email the We do not countenance repeated use of false names and false email addresses.–Econlib Ed.]

Russ Roberts
Dec 31 2009 at 2:18pm

Bill Gates,

Your challenges don’t have anything to do with this podcast, so I will not respond here. But next week, I will respond to your comment at my blog

Mort Dubois
Dec 31 2009 at 7:31pm

Re: safety regulations. I run a small factory stocked with dangerous tools, and have done so for 25 years now. In that time I have never seen an OSHA inspector, and have never even looked at the regulations covering my industry, as it’s not necessary for me to know the details of the regs in order to see that keeping my workers safe is a good idea. So I should be in favor of getting rid of them, no? No. I like the fact that there is a minimum standard of safety required of all of my competitors – we are all obligated to spend some time and money on keeping our workers safe. We avoid the race to the bottom of safety costs. When workplace safety laws are blithely dismissed, I can’t help but wonder whether the human costs of industrial accidents have been considered. The market is good at driving down costs, but not all costs should be competed away – sure, the customers might save a little money, but the human cost is real. Regulations keep that externality from appearing, or at least minimize it.


Ak Mike
Dec 31 2009 at 8:46pm

Mort – respectfully disagree. The race to the bottom for safety won’t happen because, e.g., you will be killed by worker’s compensation insurance costs if you have too many accidents, you will wind up getting sued (not by a worker, but by somebody), you will be dropped by your liability insurance carrier, and the good workers will avoid working for you.

You know, Marx predicted that the free market would necessarily result in workers’ wages dropping to below subsistance levels. The opposite has happened, and not because of government regulation.

In other words, “the human cost” is not real. There is little if any evidence that the absence of safety regulation would increase injuries.

Jan 1 2010 at 9:31pm

First- I love this podcast. My occupation and education are far from the world of economics, so I consider this podcast to be a fantastic contribution to my education on the topic.
The discussion with Mr. Winston concerning the FAA and the design and production of airliners seem to miss something I observed within the auto industry-regulations are insurance. There ARE choices engineers can make, and sometimes the science of safety is not clear. Regulations provide some security in knowing that a decision will now haunt the company later. The approval process that Boeing will go through with the FAA provides Boeing with back-up when the unforeseen or unpredictable happens far in the future (and it WILL happen with any man-made complex system). The regulations evolve as the industry, and regulating body, learn together. But until there is firm data on the design of a particular component with regard to safety, following regulations allows the product’s development to continue, and allow progress to be made without the fear of retribution as long as developers act in good faith.
I also think that regulations like seatbelts or airbags may have ‘trickled down’ to every price level eventually through market-forces. But the regulations that forced these developments into every vehicle saved countless lives by speeding up the process. Regulations like these that are announced in advance, and coordinated with the industry also serve to make the market for safety innovations predictable, and therefore justify investment in developing more efficiencies in their design and production well-before the regulations take hold.
I agree with the general point of the discussion, however, that government intervention should be judged on results, not ideology or theory. However, ALL of the results should be taken into account.

Jan 5 2010 at 1:12pm

I find it interesting that “government” is the problem, yet as stated in this podcast, the market solution relies on threat of legal action and prison sentences if business causes harm. But aren’t the court and prison systems part of government? And if we didn’t have government regulation, on what consistent basis would businesses/people be prosecuted?

My family runs a manufacturing organization, and personally, I am grateful for the regulation in our industry. It puts our competitors and us on an even playing field (in an effort to gain a cost advantage, we can’t cut corners on materials or worker safety), and when something *does* go wrong, we are less likely to be successfully sued if we are in compliance and acting in good faith. Without this, we would be at the mercy of forces beyond our control. Regulation allows us to reduce the error bars in our risk forecasts, which in turn allows us to charge lower prices.

Jan 6 2010 at 9:29pm


Early in the podcast, examples were given of types of market failure. You and Clifford sounded satisfied that a complete list was provided. Adverse selection (insurance, used cars, buffet restaurants, etc.) seemed conspicuously absent to me. Was this an oversight, or do I have an incorrect understanding of market failure?

Thanks for another great podcast.

Jan 8 2010 at 11:04am

I have very little faith in the capacity of governments to effectively regulate, but the suggestion by Roberts that Boeing can be trusted to regulate itself because safety failures are bad for business is oddly out of sync with the repeated discussions of agency problems that have taken place here over the last year or so. E.g., Bear Stearns did a lot of things that were bad for Bear Stearns. As did ValueJet. And if “too big to fail” creates a moral hazard (I’m not convinced) Boeing must have that problem in spades. We can surely let CitiBank go down the Crapper with more equanimity than would greet the same fate for Boeing.

Jan 8 2010 at 11:44am

vonb: Adverse selection in insurance is a regulatory failure, not a market failure. Insuring those expected to be well is more profitable than insuring those expected to be sick only when insurance companies are prevented from properly pricing known risk.

Or maybe I simply don’t take your meaning. How do buffet restaurants engage in adverse selection? Or used car dealers? Do the former have some way of driving away hearty eaters?

Marc D
Jan 8 2010 at 2:27pm

I agree with much of what has been said to date, as it relates to the OSHA / FAA aspect of the podcast, but would add the following:

I found that the examples provided to illustrate that the market will adjust for unsafe workplaces to be somewhat disingenuous. Not every work related injury is both horrific in scope and instantly recognizable, such as an employee losing a hand.

In many cases, injuries take a great deal of time to manifest, hearing loss being a good example. Neither employees nor well intentioned factory owners would neccesarily have any idea that harm were being done, but for the fact that a regulation exists which states what level of sound exposure is acceptable before ear protection is required.

Even if their ability to enforce them is in some way not “up to par”, OSHA’s role in establishing and maintaining safety standards is a valuable service.

Jan 8 2010 at 5:15pm

Gandydancer, thanks for the response.

Hearty eaters select themselves to eat at buffet restaurants. Serving them increases the buffet’s cost so that the buffet has to raise their price to the point where light eaters are priced out of the market. Light eaters would be glad to eat there at a price where the buffet would be glad to serve them, but since the buffet charges everyone the same price this doesn’t happen. A mutually advantageous trade is possible but it is not made. This is the definition of market failure.

It is similar for the other examples: less healthy people proportionately buy more health insurance until healthy people are priced out of the market; unreliable cars more often appear in the used car market driving down the price so that it is not worth it to sell your reliable car. So go the standard arguments, at least.

Jan 9 2010 at 8:05am

vonb: But information failure WAS specified as a market failure type. And in each of your examples the problem is inability to price properly due to missing or hidden information. Or, as I pointed out, information use can be forbidden.

Anne B.
Jan 10 2010 at 1:17pm

Thanks for the podcasts. In world w/ so much disinformation, it’s useful to have an honest perspective – whether or not I always agree.

One area to explore on privatization of government functions is in building codes. In the US, the most common model is for equipment manufacturers to develop the codes and then for local government to adopt and enforce them. Arguably, this is a long-standing example of privatization of regulation.

While I suspect that building codes have increased construction quality, there’s a series of inefficiencies, overengineering and sometimes impractical solutions imbedded in them. I don’t know of anyone who has studied this, but it would interesting to hear the extent to which the private sector has done this well vs. the degree that this function has been captured by special interests.

Jan 10 2010 at 9:11pm

Gandydancer: it was brief so I missed it… but information failure was there. Thanks for pointing that out.

Jan 12 2010 at 11:35pm

1. Somewhat disappointed that Russ mentioned “skepticism” about empirical work in economics, but didn’t really get into how that impacts the conclusions being drawn in this discussion — would have liked to hear how exactly you can tell if an OSHA or an FAA is (or isn’t) having an effect.

2. One benefit of regulation not mentioned in the conversation (or did I miss it?) is a signaling benefit. That is, if as a potential employee I know a factory is regulated by OSHA, I’ve got some assurance from an outside source that the employment won’t be particularly hazardous to my health and won’t demand a wage level to compensate for the potential risk that isn’t actually there, etc. Would you buy beef from a producer that wasn’t regulated and stamped as safe by the USDA? Maybe that falls under regulatory capture?

Gavin Andresen
Jan 23 2010 at 9:51am


1. Skeptics get that all the time. The basic answer is we’re less skeptical of effects that are large and easily shown. We’re more skeptical of effects that are small and that require sophisticated statistical techniques to “tease out” an effect from noisy data.

2. Those signaling benefits do not require a government regulator; industries can (and do) establish private certification authorities (like the Fair Trade certification) to signal safety (or wholesomeness or whatever other characteristics the market demands).

And I’m fairly sure my local farm share’s meat (which I eat with gusto) doesn’t get stamped by the USDA…

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Articles:

Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. [Recording date: December 18, 2009.] Most people hear the word "market failure" and think it means markets didn't do what I wanted them to do. They failed me. That's not what economists mean. Narrow definition. Textbook definition of market failure. Precise notion, focus on efficiency issues--the allocation of resources. Looking at situations where we get an equilibrium where it would be possible to make one person better off without making anybody worse off. Technical term for that--Pareto optimality. Can reallocate resources. Late 19th century economy Vilfredo Pareto. One measure: how would we know the world A is better than world B? Really we are talking about potential Pareto-improvements, because you usually have to actually compensate people to assure that they are no worse off, and somebody is better off. Notion of actual improvement is very rare--no policy maker would allow that to remain. Usually it involves a transfer of resources where you could at least compensate people and still have something left. Public finance and public welfare theory--increasingly uneasy with it. Policy changes that simply make the world more efficient are usually just making the pie bigger; but not going to worry about whether some of the people might actually be worse off so long as the gains to others are sufficiently great that they could compensate those who are worse off. Increasing the size of the pie has implications for distribution that are often presented as conflicts, but they need not be. If you are inside the frontier you can often get ways of expanding the pie and cut it up to redistribute. Want to know that we are getting the most out of the economy; if there are laws in place that are hampering that, or laws that could be put in place to improve it, might want to go in that direction. Major justification for government. Government is supposed to come up and institute policies, expanding the pie. Efficiency objective of government policy. Textbook claim; the way we teach students--markets don't always work perfectly, and government's supposed to step in in those cases on efficiency grounds.
6:17Examples of market failure: Monopoly, abuse of market power leading to prices above marginal cost, leading to a deadweight loss. If the behavior that underlies it is anti-competitive, this is what Antitrust policy is intended to correct. Define deadweight loss: occurs in a situation where there is actually a loss of resources to society, in the course of maximizing their profits, the monopoly would reduce output. Even though there are cases where on the margin people would like to pay the cost of the output that is produced, they are not able to. The output is basically lose. There is a potential exchange that would make the monopolist and the buyer better off, but the monopolist would have to lower the cost of all the other units to do that, which would discourage the monopolist from doing that; so there is a foregone net benefit. Technological example, natural monopoly where the costs are minimized with one producer. Declining average cost could do it. Marginal cost below average cost and with competitive pricing people would be losing money; or one supplier. This is the justification for one utility regulation: allow one person to produce the output at least cost, but we'll do something about putting a cap on prices so you don't get exploited and have excessive prices. We have regulation, though, in cases where there is not a natural monopoly. Big example is externalities: consumption externalities, for a consumer, auto congestion, driving in peak period you delay other people, but you don't have to account for that in your decisions; impose a social cost and there should be something in the market that makes you take account of your decision but nothing makes the market do that. No one owns the roads. Well, the government owns the roads: is there something they could do that would make people account for the social costs in their decision. Production externalities: pollution, firm produces something and dumps something in the water; air pollution; climate change. What's the market doing explicitly to force firms to take account of what they are doing to the broader section of society? Information failures: people are not acting with full information or are even being deceived so the utility they expect to get from their goods is not what they wind up getting, so we have a variety of information policies. Austrian perspective: it's unrealistic to have perfect information be a goal; we need to get rid of friction. Straw man critique of economic activity. Public production: you have certain services that are socially desirable but they may not be privately profitable. Natural monopoly similar; or you could have capital requirements that make it hard for one firm to produce the interstate highway system. Orphan drugs. Argument is that the government has to get involved and be the provider of these goods or services even though they may even lose money.
12:13Textbook says government needs to set the right tax, set the right subsidy, regulate, produce it themselves. Book: instead of asking whether the textbook is right, ask to see how government actually behaves. Theory; but also need to look at the evidence. Easy to theoretically point out these problems with markets; but who actually performs better? Do markets make efforts to correct these problems? Does government step in and actually do a better job? Intervention justified? Or do they come in and make things worse? So many of these issues come up time and time again, but the public policy debate always starts from square one. Start to assemble what we know and start accumulating evidence; don't just treat these things as though they've never happened before. Book is available free of charge online. What do we know? Government interventions have turned out to be remarkably disappointing across the board. It is certainly the case that markets do fail--we do have pollution, congestion; but there are other cases where it's hard to find evidence of the kind of market failure that would justify government intervention. More troubling is the lack of evidence of whether government interventions--justified or not--significantly turning things around. Not even saying we should use the benchmark of optimal government policy--just clearly producing benefits. Hard to find evidence that led to that conclusion. Focused on academic research, not on government reports--not many of those. Scientific assessment. Secondly, often public approaches these things as ideological--you can predict based on one's political affiliation where these things are going to come out. Had a representative sample across the board in terms of political persuasions; didn't limit search to particular scholars. Big qualification: there's a lot we don't know, room for more work. At this stage, a disturbing picture and big warning sign for expecting positive results for big government. Russ--love conclusion, but skeptical with empirical approaches. Law of large numbers--not all colluding. As n gets large, unless only talking about publication bias. Group think. Russ: will press on the nature of the empirical work. Sophisticated techniques with lots of assumptions along the way different from averages and facts, relatively transparent. Reality checks, simple descriptive data. Certainly economists are opportunistic; if they have a chance to come up with an article that says all the previous research is wrong, they're going to do it. Surprised by consensus that appears to emerge. Exchange on antitrust issues.
19:20Turn to the evidence. A few broad categories. Broad strokes overview, then philosophical issues. Antitrust: What was interesting was the absence of positive evidence. Didn't find things that said that government investigations of collusion led to price increases per se; just didn't find evidence of significant price decreases. Anti-monopoly cases, Microsoft recent example, no cases that led to benefits for consumers. People now saying: What was the point of the Microsoft case? Could have said the same thing about the IBM case, the cereal case? Now we see the government is investigating Intel. Have accumulated evidence of lack of consumer benefit from prosecution of anti-monopoly behavior; should give one pause. Some evidence there is consumer harm. Wouldn't the interventionist claim that that's just because those are the cases that you observe? Could be that without this occasional foray into threatening action against collusion and mergers--there's a hidden benefit you don't see. Defense: You don't see the main point of anti-trust, which is deterrence; without this prices would break loose, collusion would be the rule of the day, mergers till we get monopoly. There have been efforts to address that, controls, international comparisons; strengths of antitrust laws between the United States and Canada, or between the United States and the United Kingdom. Even those studies don't seem to find anything productive in producing a more competitive economy with antitrust laws. Still an open question. Not a deep empirical understanding--gaps in knowledge. Lack of engagement with economists on this issue; people have very strong positions despite the evidence; a lot written either pro or con about the Bush Administration's position on antitrust, but no reconciliation with the evidence. Pervaded by religion. Another area where that's true--macroeconomics, strong set of prior beliefs. At least in macro there is a desire to generate evidence; with antitrust, not much concern. Hasn't struck a chord. Difficult to get data; but agencies who have economists could make things easier. Disappointing for both policy and the economics profession. Economists have a large personal stake financially in the antitrust wars--they consult; very lucrative. But they compete. Open question: micro/macro discussion--lack of broad overview of the social deadweight loss to the United States. Arnold Harberger worked on this, a long time ago, very crude, lots of assumptions; suggested a small one. Overview of how industries are performing. Subsequent investigations have not suggested that a big problem in the United States is high markups because of monopoly power. Reality check that would be powerful to pursue. Weakness of academic field of Industrial Organization (IO) is it's not given us enough hands on overview.
27:38Contrarian take: Defense of the interventionist approach would be okay, the United States looks very competitive--Schumpetarian, competition everywhere, monopolists see competitors come out of the blue--sure there's not much monopoly, and what would be the big gain from throwing away all this apparatus? Let's get rid of all this antitrust stuff. Contrary view would say it's not that expensive and maybe the deterrence effect is important. Policy implications: first, one should focus on the most egregious cases. XM and Sirius. Ford and GM. Second: Don't want to encourage collusion. The harm is the policy is gamed. Microsoft's competitors had a lot to do with bringing this case; huge transactions costs, attention by management, considerable time and resources. Government isn't spending billions of dollars. Can create a culture where your managers spend more time gaming the government than producing better products. Regulated, even deregulated industries, grew up with the government and hard to cut the cord. Not the climate one wants: I'm having a problem with my competitor so it's time for me to start lobbying my Congressman and turn to the antitrust authorities. Destructiveness of rent-seeking; Munger and Don Boudreaux podcasts. Pervasive problem in all areas where government is involved; political. Any measures of the magnitude of that cost? Not really. Fun to do a case study at a corporate level--how many hours spent on these issues, how big the legal staff.
32:03Safety reforms, social regulation. Deals with the externalities, safety being one. Workplace safety. Descriptive evidence of workplace accidents going down; but when you try to isolate the effects of the Occupational Safety and Health Administration (OSHA), supposed to be inspecting the safety of plants, people tend not to find statistically significant effect of the OSHA regulation. In the world, seems that things getting better; value of regression analysis, isolate the counterfactual--what can we attribute to the policy as opposed to other factors. More detailed study of how these regulations actually work in practice. Attraction of an eloquent speaker, like the President, is that one gets the idea that someone of his intelligence is carrying out and performing these regulations; but the President doesn't go to plants and inspect toys. Agency not particularly well-staffed, spread out--one toy tester for China. OSHA, similar kind of thing. General morale, difficulty of persisting in these investigations. Not to say that problems don't exist. Market provides huge incentives for providing safe workplaces--courts, liability system; and can be killed in the market if things are unsafe or even in the labor market--have to offer a compensating differential in wages. Imperfect information--workers don't know that, get exploited? Area of safety clear when an airplane crashes or a mine caves in or when people repeatedly have hand chopped off from a machine that doesn't work well. The word gets out. Economics profession--job market blogs, potential employers. Hard these days to hide the fact that working for a particular firm is safe when it's unsafe. Regulatory agencies, constraints on people. Regression analysis: always tricky to "hold other things constant." What's interesting about the safety issue--data pretty clear, overwhelmingly clear, that the overall long-term trend across every dimension of our lives, auto safety, workplace safety, has been getting better and better; and it's been going on for a hundred years, fifty years, as long as we have the data. Clear that safety is a "normal good"--correlated with income and prosperity. As we get more prosperous, we are willing to devote more resources to these issues; firms compete. Higher value of life, higher quality of life. When you look at the data and see that nothing is change because of OSHA, it doesn't prove that OSHA didn't make a difference. It's possible, could have been worse without them. But it puts the burden of proof is on OSHA defenders. Law says improvement, but data say no improvement. Laws versus the bottom-up emergence that have improved safety. Mechanism: deeper micro understanding abstracting away from the data. What exactly are these inspectors doing that's telling the plant how to become safer over time? Airlines: what exactly is it the Federal Aviation Administration (FAA) is going to tell manufacturers and airlines that they don't already know? If anything, they've taught the government something. Dreamliner just tested by Boeing, still has to be certified. The FAA is going to tell Boeing if they think the plane is safe enough to fly? Imagine the quality of the people who work for Boeing on that plane, their training and knowledge; going to ask the FAA if they know what they are doing? There can be mechanisms where one can point something out. Possible critical in the area of financial regulations; but tell me the mechanism. One has to go beyond the abstraction. What exactly it is they can do and what exactly they are going to do. Boeing example great because really highlights distinction between government failure and market failure. It's true that Boeing is not really eager to crash its planes. Not good for Boeing when plane crashes. Could argue that they'll cut a corner here and there because they want to make a profit. Economist's response is they care about both today and tomorrow; organization is ongoing with repeated to interactions; incentives will probably be aligned with desires of consumers. Okay, but not perfectly. So, what the government can do is just to make sure that they haven't cut a corner here and there just to make small profit. But then: fair enough; now tell me what the structure of the FAA would be that would in practice reduce that corner-cutting instead of just making life more difficult for Boeing or its competitors. Boeing itself could be cutting a corner that it doesn't know it is doing or isn't aware of the tradeoffs. Clear in some investigations of the FAA that the airlines realize they have not changed their configuration of putting a screw on to some part of the plane; still say they are going to fly these things, in line with procedure. But there's no way that this is going to compromise the flight of the plane. When they cut corners that have significant repercussions, liability, negligence issues that can also have criminal implications. Doesn't mean some of them won't take a chance; life-changing things going beyond just making money--people go to prison for this kind of stuff. People self-select to go into this line of work. Observed data: remarkably safe. Stress point by Sam Peltzman, podcast--market robustness. How people and markets themselves try to respond to externalities. Nice example is congestion: market response is the congestion you experience is under your control, endogenous. Where you choose to live, where you choose to work. If someone really dislikes driving in congested traffic, he will make choice to try to find place to live to reduce that congestion cost, albeit may be paying a little more for housing. Data in estimates of time--people who have the highest value of time have shorter commutes; people with low value of time have longer commutes--market's way of responding to an externality. Area of noise--people who don't like to live near runways or airports. In Boston, they've located the School of the Deaf right near the runway--the land is cheap and people not annoyed as much by the noise. Financial area--don't expect same types of mistakes, going to be learning.
46:13Turn to classic case where government regulation may have improved matters: pollution. Private incentives to reduce pollution; pollution is foregone efficiency, natural long-term trend toward less waste because waste is costly. But always a temptation to dump that waste into somebody else's air, somebody else's water. Net effect of government pollution regulation? There are alternative ways we address pollution or externalities. Can use the price mechanism, in which you try to put a charge for somebody for doing this, or a quantity mechanism, saying there are certain technologies you have to use to abate the pollution. Government's preferred approach is usually what is called "command and control." Air pollution and automobile pollution have gone down; heavy hand by the government in pushing things too far back. At each point you can ask is the social cost exceeding the social benefit? What the evidence has shown has been yes, we've had benefit, but at very high cost. In case of the automobile increasing the cost of production. Most of the evidence: fairly balanced, what gains we have made in terms of pollution have been balanced by the higher production costs of doing so. Lessons and concerns with current climate debate: are we going to go ahead with a policy that is too heavy-handed, sacrifice too much of GDP? Goal to reduce pollution and not going to worry about the cost. We have limited resources. If at the same time we want to spend money on subsidizing somebody else in the social area we are not going to have the resources to do that. Goal, want to achieve particular level, pay little attention to the cost. What mechanisms could we have used to achieve those levels of air pollution reduction at lower cost? Pricing emissions. Could actually monitor the social cost of pollution; ways of doing this without intruding on drivers; could add actual price. Tax emissions. Similar example with airplane noise. Have command and control regulations--have to make plane engines that have certain decibel levels. Could have a noise tax to airplanes--if you go over this, here is what you have to pay. Decibel exposure down dramatically; but cost to plane is increased capital cost, planes obsolete, reduced value of capital stock; reflected in prices. Tradeoffs we often make. General engineering mentality that comes about with a lot of these policies, want to achieve the goal, ignore costs. Spills over into production. Want to build the extension out to Dulles Airport on the Metro line [in Washington, D.C.]; when all done, will be excited. But billions of dollars and cost overruns. Those kinds of cost/benefit tradeoffs are not central in public policy formation. General distribution of costs by the taxpayers enables it to move forward--individual taxpayer doesn't pay very much, benefits may be localized. We all benefit, but we also are all paying. Quibble with tax idea: seems that it's better to have a tax than the command and control mandate of particular technology--catalytic converter--for every unit of pollution put out above a certain amount pay a tax versus this is the technology you have to use. At the same time, Honda met the same standard of pollution that the catalytic converter produced in other cars, but they were required to have one anyway. Distortion for political reasons. Quibble: we could put a tax on the social cost. What we can measure is the pollution. Still a challenge to figure out the social cost, the harm those pollutions cause; and problem to set the level of tax. Role of the economist. Congestion tolls--use estimated value of time. As a practical matter, these things can become politicized quickly. Seeing that already--highway example. Intercounty connector toll schedule has been announced; seems very high to users or observers. Will be political pressure.
55:52Philosophical questions: Accept findings, summary of what many people have studied, gloomy picture: government intervention often doesn't achieve what it was intended to do, or servers counter goals, or certainly not what textbooks describe. Two broad reasons for that. One would be incompetence: incentives, morale, technical challenges. Second more sinister: idea that that economists and political scientists have been writing about, regulatory capture--once a regulatory idea gets set up, the people who are regulated have the biggest incentive to make sure it serves them rather than a so-called public interest. Any evidence? Latter. Political economy driving a lot of these things, interest groups in general. Often not capture per se. America getting increasingly better at lobbying; returns from it are really high. Highway: not that people are being regulated but that they want the expenditures to benefit them. Answer to the question extremely important: where do we go from here? Policy conclusion has to derive from: What is the underlying problem? Interest groups. Calls for not the wishful thinking of more government; at a time when markets are suspect, more reliance on markets. Didn't talk much on natural monopoly; major examples are ones where we've had deregulation or partial deregulation. Learned about gains. Some of this stuff is on the horizon. Skeptic: Enron story in California shows the danger of deregulation. It was the market that outed Enron. Energy part of it: attempts of California. Something called deregulation that was so mismanaged and went in the face of what was trying to be accomplished. Mismanaged terribly. Chances for privatization: airport, metro system, highways; very important to run very carefully designed experiments to avoid the California problem. What kind of experiments? We have some on the book. We have examples of airports trying to become private airports. Chicago Midway first major one. Unfortunately during the crisis their funder wasn't able to come up with the money. Imagine a bus service contracted out, private provider taking it. Munger Chilean bus podcast. Thoughts on the feasibility of the politics? Easy to say we should move to more privatization; people worried now about private incentives, but coalitions big threat. Yandle, Bootlegger and Baptist. When actual safety regulation gets written, it's written by people who have a stake in it. How might the special interest be quieted? Ties in with the micro/macro question. With public policy reform, need some entrepreneur to sell it to the public. Broader vision: trucking regulation; inflation during the 1970s, think regulation will help fight inflation. Crisis provides us with an opportunity. Concerns: budgetary, we will have to pay a lot of this back, governments at all levels running huge deficits; want also to spur growth. Privatization: can help budgetary considerations. Also turn private sector loose and they may come up with innovations. Think more broadly about where it can help many people in the economy.

More EconTalk Episodes