Sep 22, 2014
CPI ICN Column edited by Maria Coppola (U.S. Federal Trade Commission)
The ICN’s Recommended Practices on Competition Assessment: Reflections on Measuring Competitive Harm – Alden F. Abbott (Edwin Meese III Center for Legal and Judicial Studies, Heritage Foundation)1
The International Competition Network’s (ICN) adoption of 13 “Recommended Practices on Competition Assessment”2 (CARPs) at its 2014 Annual Conference in Marrakech presents competition agencies (CAs) worldwide with both an opportunity and a challenge. The opportunity stems from being able to rely upon new broad-based international consensus support for reining in pernicious government restraints on competition. The challenge comes from the practical difficulties inherent in applying the principles embodied in the CARPs to factually complex and often politically sensitive government programs. This column attempts to offer guidance in areas that the CARPs have not addressed, in particular, how to measure the competitive harm of a law, regulation, or policy.
The CARPs delineate in orderly fashion a multi-dimensional framework CAs may consult in developing and applying their competition advocacy program. Significantly, in addition to describing the sorts of regulatory restrictions that harm welfare and competition, the CARPs provide helpful insights on the sort of policymaking environment that can advance competition principles; on building alliances and relationships that can advance CAs’ initiatives; on the prioritization and selection of targets for advocacy; and on the evaluation of the competitive effects stemming from a public restraint under investigation. The latter point, while not elaborated in the RPs, is particularly important in building support for a competition advocacy program. By producing credible measures of the actual welfare losses to society stemming from a government restriction (and, in particular, the harm to consumers), a CA may enhance the credibility of its advocacy program and strengthen its hand in arguing against government measures having significant negative welfare effects. In short, a robust set of techniques for estimating welfare losses – a “metric”— can assist CAs in meeting the dual challenges of “operationalizing” advocacy initiatives and building the political support needed to “institutionalize” advocacy as a core part of their mission. This article seeks to complement the RPs by providing additional guidance for agencies that want to engage in measuring these effects.
Conceptualizing the Metric
A metric may be used to estimate the reduction in consumer welfare stemming from a restrictive government policy – embedded in laws, regulations, or guidance – that distorts competition. Consumer welfare is often defined in terms of “consumer’s surplus” (the difference between the maximum amount consumers would be willing to pay for a good or service and what they actually pay), although an alternative measure is the sum of producer’s surplus (the difference between the price producers receive and what they would be willing to accept for a good or service) and consumer’s surplus. “Deadweight loss,” the inefficient reduction in welfare caused by a restriction that is captured neither by consumers nor consumers, may also be estimated by the metric.3
Whatever measure of welfare a CA uses, it should apply it consistently to all of the different types of restraints that it scrutinizes. A CA should also be fully transparent about the welfare measure it has adopted and the techniques it employs in applying a metric (see following discussion), in order to build public trust.
Applying the Metric
A CA could apply the metric by viewing the existing (or proposed) restriction under consideration as an economic shock, and estimating the shock-induced reduction in consumer welfare. This could be measured in “present value” terms, looking at the value in today’s currency of the sum of the annual reductions in consumer welfare over time (discounted to reflect the time value of money). Alternatively, a CA might opt for a measure of “annual consumer welfare lost” due to a restriction, measured in today’s currency. The welfare loss figure could be phrased as a rough approximation, plus or minus a certain percentage, to take into account a potential range of likely error. Such an approach could add credibility by recognizing imperfections in estimation and limitations on knowledge, while at the same time highlighting the real harm to domestic interests flowing from a restriction.
A variety of specific techniques (with respect to which there exist publicly available sources of guidance) can be deployed in carrying out estimates. These include, among others, cost-benefit analysis,4 cartel analysis,5 merger analysis,6 regression analysis,7 and various forms of “before and after” effects that are observed in the real world when a restriction is imposed and/or lifted – “natural experiments.”8
CAs should show flexibility in applying the metric, adapting it to their resources and the constraints they face as one among many government actors. CAs differ widely in terms of their current resources, experience, and personnel capabilities. A small relatively new CA with few or no economists and limited data cannot be expected to carry out a complex, data intensive evaluation, for example. Thus, the metric methodology employed by a CA should reflect its capacities and resource constraints – metric estimation may fall within a spectrum from very basic to highly detailed. As an NCA grows in experience and resources, it may be able to deploy increasingly sophisticated tools.
When a CA lacks the resources or the data to carry out studies employing sophisticated techniques, it may nevertheless be able to derive rough approximations from simple calculations, assuming other economic conditions have not changed markedly. Moreover, a CA may be able to derive a preliminary and tentative estimate of the likely magnitude of potential harm, based on some simple market information, before it carries out more sophisticated analyses. (The CA may find this particularly useful when it wants to highlight a restriction’s negative effect at an early stage, before it becomes too entrenched.) For example, if prices swiftly doubled and consumer sales fell 25% when a new government restraint abruptly required most firms to exit a particular industry – and other economic conditions remained roughly the same – the CA might use those figures in calculating an initial rough estimate of restraint-specific harm.
CAs may face criticism for using a metric by organized special interests that benefit from a public restraint, and their champions in government ministries. The criticism may be that estimates of harm ignore the broader social benefits of a particular restrictive policy, such as the promotion of domestic industries or the protection of indigenous jobs. CAs could respond by pointing out that those benefits can generally be achieved at lower social cost through direct assistance to the favored groups – for example, tax credits (or other direct payments) for industries and job retraining and unemployment assistance for workers. CAs may also wish to highlight less restrictive means that could advance the purported goals.
Also worth publicizing is research that suggests government restrictions on domestic competition undermine growth, productivity, and poverty reduction in the domestic economy.9 In particular, studies indicate that the removal of regulatory distortions benefits consumers and drives sustainable economic growth, especially in developing countries.10 Given public concern about reducing poverty, estimates of harm to poorer consumers associated with a particular government restraint may prove particularly valuable in strengthening the hand of CAs.
Finally, CAs could explain that competition policy should not be distorted to support social goals unrelated to consumer welfare. If government wishes nonetheless to advance those goals in a welfare and competition-inimical manner, it should do so openly using its other legal policy instruments, subject to public scrutiny.
Employing Readily Available Resources of the International Community
In building their competition advocacy programs, CAs can rely on valuable sources of advice and support from the ICN community, including informal advice from experienced ICN member agencies and non-governmental advisors, as well as the burgeoning materials on the ICN’s website.11 Other established international organizations have valuable competition policy-specific resources, such as the OECD (whose competition assessment toolkit bears directly on governmental restraints),12 UNCTAD,13 and the World Bank.14
The World Bank’s research illustrates the feasibility of obtaining specific quantitative estimates of the welfare costs of particular restraints. The World Bank released a variety of studies documenting the effects of specific anticompetitive government restrictions at its 2013 and 2014 Pre-ICN Forums15 preceding the 12th and 13th Annual ICN Conferences. To cite just one example, a 2013 Pre-ICN Forum analysis presented by the Mexican Federal Competition Commission found that more efficient Mexican airline slot allocations would raise consumer welfare by $3 billion and reduce airline prices by 60 percent. The Pre-ICN Forums also included written commentaries by various developing country CAs on how they had sought to implement their advocacy programs. These and other future studies may yield valuable insights to CAs as they seek to develop their own metrics.
Being able to produce specific credible estimates of the welfare costs of governmental restraints would substantially enhance the effectiveness of CA competition advocacy efforts. Deployment of a neutral yet flexible metric may be key to deriving such estimates. CAs may wish to keep this in mind (and to draw upon the resources of international competition policy institutions) as they develop their competition advocacy initiatives. The ICN, through its Advocacy Working Group, could lend valuable support to such CA initiatives by assigning high priority to metric-related work.
1 Alden F. Abbott is the John, Barbara, and Victoria Rumpel Senior Legal Fellow and Deputy Director of the Edwin Meese III Center for Legal and Judicial Studies at the Heritage Foundation. He serves as a Non-Governmental Advisor to the International Competition Network. The views expressed herein are his alone and should not be attributed to his employer.
2 See http://www.internationalcompetitionnetwork.org/uploads/library/doc978.pdf for the text of the RPCAs.
3 For very basis definitions of consumer’s surplus, producer’s surplus, and deadweight loss, see, for example, “Consumer and Producer Surplus,” at https://www.khanacademy.org/economics-finance-domain/microeconomics/consumer-producer-surplus
4 For example, the U.S. Office of Management and Budget has prepared “Regulatory Impact Analysis: A Primer,” which provides U.S. Executive Branch agencies with guidance in carrying out economic cost-benefit analyses, see http://www.whitehouse.gov/sites/default/files/omb/inforeg/regpol/circular-a-4_regulatory-impact-analysis-a-primer.pdf
5 Cartel analysis is described, for example, in Michael A. Utton, The Effects of Cartels: Markups and Welfare Losses, in Michael A. Utton, CARTELS AND ECONOMIC COLLUSION, ch. 4 (2011), see http://www.elgaronline.com/view/9781849807708.00008.xml. These techniques may be most usefully applied to government restrictions that have cartel-like effects.
6 See, for example, International Competition Network, “The Role of Economists and Economic Evidence in Merger Analysis” (2013) (updated chapter 4 of the ICN Investigative Techniques Handbook for Merger Review),
http://internationalcompetitionnetwork.org/uploads/library/doc903.pdf. Such techniques might be most useful when a new government restriction abruptly reduces the number of competitors in an industry.
7 For a basic overview of econometric techniques, see, for example, ABA Section of Antitrust Law, ECONOMETRICS (2nd ed. 2014).
8 See, for example, Mark R. Rosenzweig and Kenneth I. Wolpin, Natural “Natural Experiments” in Economics, 38 J. Econ. Lit. 827 (2000), at http://www.uh.edu/~adkugler/Rosenzweig&Wolpin.pdf. An ideal “natural experiment” is one in which it is possible to find a “control group” that has not been subject to a harmful restriction and another control group that has been subject to the restriction, and then compare the differences in economic effects across groups before and after the restriction was implemented. (For a rare study that employed this technique fruitfully, based on differences among U.S. states in the adoption of a particular regulatory restriction affecting alcoholic beverage pricing, see James C. Cooper and Joshua D. Wright, Alcohol, Antitrust, and the 21st Amendment: An Empirical Examination of Post and Hold Laws, 32 Int’l Rev. Law & Econ. 379 (2012).) Unfortunately, ideal natural experiments seldom occur.
9 See, for example, Nick Godfrey, “Why is Competition Policy Important for Growth and Poverty Reduction?,” OECD Global Forum on Investment (March 2008), at http://www.oecd.org/investment/globalforum/40315399.pdf.
10 See Marcus Kitzmuller and Martha Martinez Licetti, “Competition Policy: Encouraging Thriving Markets for Development,” World Bank Forum (2013), at file:///C:/Users/abbotta/Downloads/SSRN-id2310917.pdf.
11 See http://www.internationalcompetitionnetwork.org/.
12 See OECD, Competition Assessment Toolkit, Volume II: Guidance (version 2.0 2011), http://www.oecd.org/daf/competition/45544507.pdf.
13 See http://r0.unctad.org/en/subsites/cpolicy/.
14 See https://www.wbginvestmentclimate.org/advisory-services/cross-cutting-issues/competition-policy/.
15 See World Bank, Pre-ICN Forum, Making Markets Work for Development: A Reform Agenda on Competition (Apr. 23, 2013), at https://www.wbginvestmentclimate.org/advisory-services/cross-cutting-issues/competition-policy/pre-icn-forum-making-markets-work-for-development.cfm; World Bank, Pre-ICN Forum, Competitive Domestic Markets: The Cornerstone to Boosting Trade and Competitiveness (Apr. 22, 2014), at https://www.wbginvestmentclimate.org/advisory-services/cross-cutting-issues/competition-policy/preicn-2014-agenda-and-presentations.cfm