How do you measure competition policy’s performance?

Some time ago, I wrote a post about the impact of competition policy on economic growth. I argued that it was an important question since competition authorities in developing countries have to struggle with where to assign financial resources and that they should do so based on the the potential of policies in solving the population’s more urgent problems (extreme poverty, for example). The conclusion was that, so far, there is no consensus on the effects of the policy on GDP growth, in part because differences in quality and performance of antitrust agencies is hard to measure.

In addition to the arguments in the mentioned post, there is an important issue that I did not address but that I came upon in a lecture given at the Munich’s Antitrust Law Forum (Münchner Kartellrechtsforum) by Prof. Richard S. Markovitz of the University of Texas. One of his points was that it is a mistake to focus competition policy on consumer welfare and his arguments were, surprisingly, very much grounded on neoclassical economics. His point, I find, is quite compelling. Market failures come not only in the form of market power but also externalities, information assymetries, and underprovision of public goods. Prof. Markovitz explained that there is no theoretical nor empirical support for assuming that a state intervention that reduces market power will be neutral in terms of the other sources of inefficiencies and that the effects on these can very well be negative.

The point can be illustrated with an example. If an antitrust agency uncovers a cartel in the munfacturing of cigarettes and manages to make the firms compete more aggressively in price, the negative externality that smokers impose on other people will increase and the net effect on efficiency will be ambiguous.

Another example with information assymetries can be the following. Higher margins may allow firms to invest more in advertising that, among other things, increases consumer awareness of different product traits. An antitrust intervention that reduces the market power of firms (say, by blocking a merger) will not necessarilly enhance consumer welfare since search costs will increase if firms start spendig less money on advertising.

The result is that even if competition policy in a given country succeeds in curtailing market power, its effects on efficiency and economic growth will not necessarily be positive. Since the net effects are in theory ambiguous, the matter is an empirical one. However, we go back to our first problem, which is how does one measure differences in the policy’s performance.

Last week, I read an interesting post based on research regarding the measurment of the deterrence effects of antitrust law. My first impression when reading the title was of wonder. One has to get creative in order to measure something that you can not see. However, the research mentioned in the post found a way by exploiting data on 500 legal and illegal cartels and their overcharges. This information allowed them to run simulations and provide conservative and upper-bound estimates. For more information on the research you can check out the post in question. The point I wanted to make is that if you can capture differences in deterrence effects across countries or through time, the data could serve to have a more appropriate measure to plot against other variables such as investment rate and GDP growth.

The authors themselves advise for further research on the topic. However, it might be that we are finally approaching a satisfactory measure of competition policy’s performance.


Tagged ,

3 thoughts on “How do you measure competition policy’s performance?

  1. Leonard Ugbajah says:

    Dear Paco,

    Thank you for the great work you are doing through this blog. I’ve been an avid follower.

    I have just read the above post and I want to make a few comments on your points about tobacco and advertising as it relates to the welfare or efficiency effects of competition law enforcement.

    On the tobacco example, it is true that competition law enforcement could lead to lower prices and more consumption, with the attendant negative externality as you have well argued. However, my take is that this is not a good enough reason to allow the operation of the tobacco cartel. This is the kind of scenario that requires the efficient application of two or more regulatory regimes to address. In this case, the solution to the negative externality should be sought in the tobacco control regulations (possibly more tax on tobacco), not in a lax enforcement of competition law.

    On the second point about consumer welfare gains from advertising, I would rather argue that: one, the assumption that higher margins leads to more aggressive advertising needs to be tested. It seems more plausible to me that higher margins driven by market concentration leads to reduced spending on advertising. Two, it would interesting to have some empirical evidence to measure the cost of information asymmetry to the consumers in this kind of cases in relation to the cost of market concentration to the consumer. This would provide a guide to measure the net welfare gain or loss. Three, if we even accept the assumption about the relationship between higher margins and advertising, then I would argue that the solution to the problem of information asymmetry should be found in another body of regulation, namely, consumer protection regulations.

    So this brings us back to question of measuring performance of competition policy. The first point I want to make here is that the gains from competition policy enforcement depends heavily on the effectiveness of other related policies and regulations. This much has been shown in the arguments I have used above. I also find the model you have cited very useful and worthy of more attention by researchers. In most developing countries, cartels and monopolies ensure that prices remain excessively high. I imagine that doing a price comparison across countries or regions with similar economic characteristics may be able to show whether those countries enforcing competition law have better prices than those that are not. Of course, this kind of study would have to address the causality problem. One example that comes to my ind is that price of cement in Nigeria and other countries with similar economic characteristics.

    In all, let me say that it may be more useful to focus on the (potential) welfare gains from competition reforms of particular sectors or markets. This is basically the model we have used in advocating for a competition law in Nigeria.

    Let me end here and to once again thank you for the useful posts on this blog.

    Leonard Ugbajah.

    Liked by 1 person

    • Paco Beneke says:

      Dear Leonard,

      Thank you very much for your kind words, I really appreciate them.

      Regarding your comments, I think you bring excellent points to the table. I agree that a tax on cigarettes would be a better policy option in the long run. However, I do think that until such a policy change comes, the question remains of whether it is better to let the cartel charge higher prices or force them to compete.

      With your other point on the relationship between advertising and economic profits, I agree that it is mostly an empirical question. In the post I tried to sketch a possibility that could in theory be true. However, that should be tested, if it already hasn’t. I must confess I have not looked on whether such studies exist.

      Again, thank you very much for your comment, which I think enriches the discussion.

      Best regards,


      Liked by 1 person

  2. […] I explained in a previous post, an intervention aimed to curtail market power can have detrimental/positive effects on other […]


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 )

Facebook photo

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

Connecting to %s

%d bloggers like this: