catallaxy files

catallaxy in technical exile

Government failure vs market failure

with 17 comments

A heads up to readers – I haven’t studied this in full yet but Brookings economist Cliff Winston has published a compendium and review of economic studies called ‘Government failure vs market failure’ which is freely available for download here. This is the abstract:

When should government intervene in market activity? When is it best to let market forces simply take their natural course? How does existing empirical evidence about government performance inform those decisions? Brookings economist Clifford Winston uses these questions to frame a frank empirical assessment of government economic intervention in Government Failure vs. Market Failure: Microeconomics Policy Research and Government Performance.
Markets “fail” when it is possible to make one person better off without making someone else worse off, thus indicating some degree of inefficiency. In economics parlance, Pareto optimality has not been achieved. On the other hand, governments “fail” when an economic intervention proves to be unwarranted, either because markets are performing adequately or public policy does not correct a market failure efficiently. In such cases, government intervention may actually exacerbate a problem or produce unintended negative results. Winston concludes that the cost of government failure may actually be considerably greater than the cost of market failure: “My search of the evidence is not limited to policy failures. I will report success stories, but few of them emerged.” Government failure may result in missed opportunities, wasted resources, and waning public support.


Written by Admin

October 19, 2006 at 4:47 pm

Posted in Uncategorized

17 Responses

Subscribe to comments with RSS.

  1. “Markets “fail” when it is possible to make one person better off without making someone else worse off, thus indicating some degree of inefficiency.”

    I REALLY REALLY hope Mr. Winston means IMpossible here.


    October 19, 2006 at 5:32 pm

  2. Yes I have to admit I did a double take as I read that bit FDB. I hope he did mean what you hoped he meant.


    October 19, 2006 at 5:36 pm

  3. No, on further research, he’s right. Surely every new marketable good or service (or one for which the market is growing) constitutes a situation where it is possible to make one person better off without making anyone else worse off.

    Isn’t that why we need economic growth?


    October 19, 2006 at 5:38 pm

  4. Pareto optimality (if wiki has it right) is achieved when it is impossible to make someone better off without making someone else worse off.

    Like when there are no resources left to distribute, and no technological advances left to be made.

    What a bizarre, suboptimal situation.


    October 19, 2006 at 5:40 pm

  5. Efficient not optimal in a broader sense. I think the point is about trade generally is there some way two people could exchange something to improve both (or at least one) of their situations. It has nothing to do with fair, as it can exist in some sense when say one person owns everything and everyone else nothing (so long as the one person doesn’t value the other people’s wellbeing).

    Steve Edney

    October 19, 2006 at 6:11 pm

  6. I think that was meant to be ‘impossible’. Someone on the AEI-Brookings site hasn’t been proofreading.

    Jason Soon

    October 19, 2006 at 6:38 pm

  7. Well, the sentence makes perfect sense in Paretonian (?) terms. Just that it doesn’t really help understand anything about making things good.



    October 19, 2006 at 7:03 pm

  8. That is the strict technical definition of Pareto optimality FDB. That is the Platonic benchmark for addressing market failure. Of course it doesn’t make sense in the real world. And the point of the report is to show that even taken on its own terms, this criteria for market failure has its own problems because it ignores government failure.

    Jason Soon

    October 19, 2006 at 7:07 pm

  9. Markets “fail” when it is possible to make one person better off without making someone else worse off, thus indicating some degree of inefficiency. In economics parlance,

    How so? How can two willing adult people not be better off when they do a voluntary trade? Pareto distribution or pareto effect applies to income distribution is a society not if two people doing a deal are means there is a loser.


    October 19, 2006 at 9:58 pm


    “When should governments intervene?” asks Winston, who then proceeds to tell us its all about MF vs GF. In doing so, he misses equity as a rationale for intervention. Maybe he covers it somewhere in a footnote, but the synopsis doesn’t give one much confidence.

    Tom N.

    October 20, 2006 at 11:41 am

  11. rather persnicketty comment, TomN. firstly he didn’t write the synopsis which is primarily about empirical studies of government regulation anyway, which renders this whole ultra-didacticism about definitions a moot point since govt policy in reality is not motivated by Paretianism.
    secondly if equity is an issue it would be an issue under market failure. what do you mean by equity? in a lot of standard discussions of policy, it is about comparing MF vs GF and then distributional considerations are discussed like an add-on because ethics is not the economists’ province. so I don’t know what the import of your comment is.

    Jason Soon

    October 20, 2006 at 12:02 pm

  12. look here is the table of contents of the book. I rather think most of the comments so far have just flew by what the purpose of the book is.
    we all know Paretianism is basically a useless criterion in real life. This adds to rather than undermines the case for reliance solely on the existence of market failure as a rationale for policy and therefore supports the thesis of the book. If by equity you mean ‘redistribution’ then reference what I said above but it’s not really particularly relevant to assessing the rationales for market failure-based intervention he discusses below except insofar as it involves some sort of ex post redistribution which is usually a background assumption to such discussions. ‘equity’ isn’t an economic concept in any case. it can mean whatever people want it to mean and is therefore conceptually empty.

    Foreword vii
    Acknowledgments ix
    1 Introduction 1
    2 Methodological Perspective 7
    3 Market Power: Antitrust Policy and Economic Regulation 13
    4 Social Regulation: Imperfect Information and Externalities 27
    5 Public Production 61
    6 Policies to Correct Market Failures: Synthesis and Assessment 73
    7 Market Failure and Social Goals Policies:
    Common Failures and Conflicts 87
    8 Policy Recommendations Motivated by Policymakers? Learning 93
    9 Microeconomics Policy Research and the Policy Community 103

    Jason Soon

    October 20, 2006 at 12:07 pm

  13. I wasn’t being deliberately picky Jason. I just wasn’t sure whther there was an ‘im’ missing or there was some (basically completely inverted, i.e. wrong) definition of Pareto optimality being used. Or both.

    It seems that what I wanted the “possible/impossible” sentence to say is what it was intended to say, so I’m pretty cool with it. If the economics of a market amount to a zero-sum game, then the market is failing to increase total well-being.

    The next sentence about Pareto seems to be simply wrong-headed.


    October 20, 2006 at 12:21 pm

  14. In his reponse to my point about the omission of equity, Jason (comment 11) claims that “ethics is not the economists’ province” and that “if equity is an issue it would be an issue under market failure.”

    Three points: first, I think it useful to not conflate ethics and equity. People can have moral views on things quite separate from ‘fairness’ issues: it is morally wrong to work on Sundays, for instance.

    Second, ethics may not be within some economists’ province, but the Productivity Commission at least appears to see them as being relevant to their work. For instance, in its 1999 gambling report, the PC took into account inter alia the effects of the proliferation of pokies on people’s ethics, which it argued would then feed into their behaviours. it also recognised the psychic costs that people with Libertarian ideals might incur from government action to ban or otherwise restrict gambling.

    Third, in relation to equity, my understanding (though I’m happy to be corrected) is that weflare economics defines social welfare as a function of efficiency and equity and recognises that efficiency gains alone do not exhaust the (legitimate) reasons for intervention. That is, inequity may persist after all market failures are corrected.

    Tom N.

    October 20, 2006 at 7:01 pm

  15. Fair enough Tom but I still don’t see how that is relevant to a survey of empirical studies of the impacts of regulation when the proponents of the regulation themselves have not bothered to quantify the ‘equity benefits’ assuming this can even be done. I don’t see what it adds to the analysis, frankly.

    If the empirical studies discussed by Winston show that the regulations in question have costs in terms of higher prices and few benefits in terms of reduced accident rates or whatever the relevant measure of success of the objective is, then what exactly is wrong with noting this? And how does then worrying about equity change your conclusion?

    If you were inclined to support the regulation anyway, you can use equity to fluff your conclusion by saying ‘nonetheless I believe the equity benefits exceed these net costs’ If you weren’t inclined to support the conclusion you fluff it by saying ‘whatever equity benefits are associated with this regulation are insufficient to overrule the net costs we have found’. So where does this lead you? It’s an empty concept except in tautological algebra. It’s a bit like invoking the ‘vibes’.

    If you aren’t going to go by the results of cost-benefit analysis then why bother going to the trouble of preparing it in the first place (given it is such an information intensive exercise) if it can be overruled by a little verbal trick?

    Jason Soon

    October 20, 2006 at 7:18 pm

  16. On the specific point, the reason one does a BCA is so that one can make an informed judgement about whether the efficiency costs entailed in looking after the beneficiaries of an existing intervention, such as a regulatory restriction on completition, are worth it. Of course, a good policy adviser would also examine other things, such as whether any distirubtional ovjectives currently being met through the intervention can be achieved at less cost in other ways, such as a direct transfer.

    On broader matters, I agree Jason that my original point is not a major one in the context of this thread. However, I think that economists sell themselves short, and also open themselves up to unjustified criticisms, when they take short-cuts of the kind Winston did. Economics is not concerned only with efficiency issues: equity is also a part of economics – at least the welfare economics area, but also some voodoo … errr … macro-economics too. Moreover, while economics has little or nothing to say of the origins of different ethics, such issues also have a place in economic analysis.

    Tom N.

    October 20, 2006 at 11:41 pm

  17. Right, thanks Tom. That makes the reason for your comment clearer and I can accept that.

    Jason Soon

    October 21, 2006 at 7:54 am

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: