Darren Bush, Mark Glick, and Gabriel A. Lozada argue that the Consumer Welfare Standard  is inconsistent with modern welfare economics and that a modern approach to antitrust could integrate traditional Congressional goals as advocated by the Neo-Brandeisians. Such an approach could be the basis for an alliance between the post-Chicago economists and the Neo-Brandeisians.


Economists evaluate policy goals using normative economic theory. Welfare economics is the subspecialty in economics that studies normative economic theory. Most modern welfare economists have eschewed surplus theories of welfare, such as the predominant consumer welfare standard (CWS) or post-Chicago trading partner surplus, which define welfare as the difference between what a consumer is willing to pay for a good and what they actually pay. For example, Nobel Laureate Angus Deaton has said, “there is no valid theoretical or practical reason for ever integrating under a Marshallian demand curve” (i.e., calculating consumer surplus). This is true for both normative reasons (surplus theory is ethically questionable) and for positive reasons (surplus theory is riddled with problematic and persistent inconsistencies). 

Some scholars have advocated instead for a “protecting competition” or “protecting the competitive process” goal to replace a surplus-based antitrust standard. However, such a goal itself requires a standard to help define “legitimate competition” or competition on the merits as opposed to “illegitimate competition” or exclusionary conduct. 

We argue that antitrust policy should observe contemporary welfare analysis and reintegrate  traditional Congressional goals, such as income equality, as articulated by the Neo-Brandeisians. Although competition cannot form the core of a new antitrust standard, it can serve as a bridge between the Neo-Brandeisians and post-Chicago economists. 

Why modern welfare economics abandoned surplus models 

All surplus approaches to antitrust limit welfare to economic surpluses. By contrast, most modern welfare economics evaluates policies on the basis of their impact on social welfare, as determined by a social welfare function. Such a social welfare function depends on, but is almost never a simple sum of, the individual welfares of the relevant population. Social welfare functions possess at least three aspects: the relevant population with standing (usually all impacted individuals but in antitrust participants in the relevant market), the measurement of individual welfare or well-being, and the aggregation of individual welfare. 

Few, if any, welfare economists would limit the measurement of welfare to economic output. For example, welfare economists would include quality of life indicators or the capabilities approach along with economic growth when measuring the welfare of a particular nation. Notably, these approaches typically list the traditional Congressional goals of income equality, democracy, and small entrepreneurship as critical factors that impact the quality of life, and these goals are backed by solid research that show that they are important for human well-being.

CWS cannot be reformed

Some CWS advocates, including post-Chicago theorists, argue the CWS is broad and flexible enough to incorporate numerous antitrust goals. This is untrue. The CWS is simply Alfred Marshall and Arthur Pigou’s consumer surplus approach (or consumer and producer surplus and rent) rebranded by Judge Robert Bork, one of the most prominent representatives of the Chicago School. Surplus is defined by the difference between demand and price, so only factors impacting these variables such as output (and, potentially, innovation or quality) are relevant, but nothing else. Labor rents are potentially relevant, though not in the models created by the Chicago School. 

The CWS also a priori deems irrelevant other significant considerations impacting human welfare that do not directly influence demand, like democracy, even if they can be affected by competition policy. Bork introduced the CWS into antitrust for precisely this reason. He sought to rid antitrust of its traditional goals of political democracy, protection of small business, and preventing income transfers. 

Post-Chicago economists have contributed to positive economic theory but cling to CWS

Along with its normative economic standard, the CWS, the Chicago School also offered a positive economic theory to justify curtailing antitrust enforcement. Post-Chicago economists have largely refuted the Chicago School’s flawed positive economic conclusions. Neo-Brandeisians applaud those advances, and they are particularly important in the current judicial environment in which six Supreme Court cases have endorsed the CWS to some degree. 

However, most post-Chicago School advocates have not abandoned the normative economic approach of the Chicago School, maintaining that “welfare” is based solely on economic surplus (whether consumer, consumer and producer, or trading partner). Even Steve Salop, who does endorse goals such as democracy as part of his Reasonable Competitive Conduct standard, reverts back to a narrower welfare definition when he urges the Neo-Brandeisians to “[balance] economic welfare effects with other goals.” In modern welfare economics, there is no “economic welfare” (the primary goal) and “other” (secondary) goals. There is only welfare. A welfare-based antitrust standard should include Congressional goals, both because statutes are a legitimate way society’s social welfare goals are established, and because social science research has empirically demonstrated that Congress’s antitrust goals align with many people’s perception of their own welfare. 

That being said, economists value competition because it may improve welfare over its alternative. A modern welfare theory can help define legitimate competition far better than the CWS. This approach could form the foundation for an alliance between the post-Chicago economists and the Neo-Brandeisians sought by Jonathan Baker and Steve Salop.

However, “protecting Competition” as a polestar cannot by itself lead to any tangible policy prescriptions. Under Arrow-Debreu assumptions, “perfect competition” is Pareto Optimal. Absent the assumptions of universal perfect competition (and considering the theory of the Second Best) it becomes unclear how to determine when perfect competition advances human welfare, let alone when “competition” that is not perfect is legitimate and when it is potentially exclusionary. The role of an antitrust standard is to help resolve the ambiguities in what antitrust means by legitimate competition or competition on the merits. These are difficult issues to resolve even within a revised antitrust welfare standard, but a revised welfare standard will enable economists to integrate Neo-Brandeisian proposed reforms within an economic framework. 

Retaining The Surplus Approach to Welfare is Not Viable Option

A. The Surplus Approach’s Normative Problems: Bias and Ignoring Distribution

Under the surplus approach, only the sum of the surplus matters. (This is also true of cost/benefit analysis using Equivalent Variation and Compensating Variation, as described below). Asserting that only the sum of surplus matters requires the additional assumption (using cardinal utility) that the marginal utility of money is constant for each person and equal for everyone, meaning an additional dollar has the same utility for the rich hedge fund director as for the lower-income worker. Under this assumption—one criticized unrelentingly and convincingly by welfare economists on ethical grounds—transfers of income between stakeholders within the firm have no welfare implications for antitrust policy to consider. No economist we are aware of supports this notion.

Worse, the surplus approaches weight the preferences of the rich (including large corporations) more heavily than the poor because the rich have higher effective demand than the poor for normal goods. Higher effective demand implies greater surplus. As such, economic policy and court rulings observant of the CWS disproportionately prioritizes the affluent over the poor. 

While some economists admit this, they contend tax policy can remedy such bias. This argument is pretextual and unserious. Tax policy is not geared to individually compensate policy losers. Indeed, when the rich are benefitted by antitrust policy, they can better influence tax policy in their favor. Moreover, saying society should adopt “Policy X” because “Policy X plus tax changes” would be good, is, as Jules Coleman pointed out in 1980, the non sequitur that people would consent to a policy that harms them “in virtue of its potential to be something other than it is.” In 1978, welfare economists John Chipman and James Moore said that advocating `Policy X’ because `Policy X plus tax changes would be good’ “is to wash one’s hands of the responsibility for one’s own actions.”

Another flawed justification for ignoring distribution is doing so improves productivity. This “trickle down” theory is completely contrary to the stylized fact that periods of high income equality in the U.S., such as the 1950s and 1960s, have higher productivity, growth, and innovation than periods of high income inequality, such as the post-1980 period. 

B. The Surplus Approach’s Inconsistency was Never Resolved

In addition to the failure of surplus approaches to account for unequal distribution of surplus, they are plagued by logical inconsistencies. Marshall used surplus to assess the effect of commodity prices and quantities on welfare, but he was aware that doing so is incorrect unless the amount consumed of the good in question was unaffected by income changes. Because the amount consumed of goods usually is affected by income changes, this “quasilinear utility” assumption is a serious positive shortcoming of using surplus to assess the effect of price and quantity on welfare. 

In 1941, Sir John Hicks, building on work by Nicholas Kaldor, developed better measures to assess the effect of price and quantity on welfare. They are called Compensating Variation (CV) and Equivalent Variation (EV), or, more intuitively, willingness (and ability) to pay, and willingness to accept. For policies resulting in gains, CV = WATP and EV=WTA; for policies resulting in losses, CV= –WTA and EV= –WATP. 

But CV and EV are not the same. Tibor Scitovsky showed in 1941 how this can lead to policy inconsistencies, and Paul Samuelson showed how the dual nature of value can give rise to intransitivity in the social rankings. If society has two options, A and B, the CV/EV approach sometimes recommends that if society is at A, society should move to B, but if society is at B, society should move to A. Gorman showed the only way out of this problem was to make the untenable “quasi-homothetic utility” assumption that everyone consumes goods in the same proportion regardless of their income, an obviously false assumption. We detail these issues further here.

Antitrust economists perpetuate a myth that Robert Willig’s 1976 paper solved the inconsistency problem by setting limits for the deviations between CV and EV (and consumer surplus). In an earlier paper, we refute this myth. Willig actually shows that the non-approximated absolute value of the proportional difference between CV and EV and consumer surplus can be as large as 97%. Accordingly, another subfield, environmental economics, always teaches in its textbooks that CV and EV are often far apart, with comments such as “experimental findings both in environmental economics and in other microeconomic studies have found large differences.”

Even worse, Willig’s theory only applies to aggregate data if a representative consumer exists, which requires unlikely and unrealistic assumptions: quasi-homothetic utility for the average consumer to be representative, and almost as restrictive assumptions for any consumer to be representative. Antitrust econometricians lacking comprehensive household-level data typically have to make these same assumptions when estimating consumer surplus.

Over the years, numerous other inconsistencies have surfaced, including Boadway’s 1974 work showing that a policy with a positive sum of CV’s or EV’s did not necessarily imply winners could compensate losers, and Brekke’s 1977 work showing that conclusions using CV and EV can reverse depending on one’s choice of numeraire. There are many other problems and difficulties as well

Conclusion

Replacing the CWS with a modern welfare economics approach incorporating evidence-based, objective research on the determinants of human welfare, with deference to the original concerns of Congress, will allow economists to make important contributions to the debates about the future of antitrust. Economists will no longer be required to make policy arguments tethered to a narrow and biased welfare approach, nor will they be forced to defend unrealistic assumptions such as those required to establish consistency or adopt ethically unattractive positions dismissing the importance of distribution. Rather, economists will be able to study and, when appropriate, incorporate other important welfare-promoting factors into antitrust policy. Modern welfare economics can be the basis of a truly united front between Post-Chicago economists and the Neo-Brandeisians.

Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.