Whether antitrust policy should pursue a goal of "general welfare" or "consumer welfare" has been debated for decades. The academic debate is much more varied than the case law, however, which has consistently adopted consumer welfare as a goal, almost never condemning a practice found to produce an actual output reduction or price increase simply because productive efficiency gains accruing to producers exceeded consumer losses.
While some practices such as mergers might produce greater gains in productive efficiency than losses in consumer welfare, identifying such situations would be extraordinarily difficult. First, these efficiencies would have to be "transaction specific," meaning that they could not be attained by other means. Given the widespread availability of alternatives such as nonexclusive licensing, transaction specific efficiencies are the exception rather than the rule. Second, these would necessarily be gains that accrue at lower output levels than previous to the practice; otherwise there would be no consumer harm to balance. But most efficiency gains accrue at higher rather than lower output levels. Third, collusion facilitating practices spread welfare losses across an entire industry, while production gains typically accrue only to the participants in a merger or similar practice. Fourth, the reigning tradeoff models generally assume a market that was competitive prior to the practice and became monopolized after. Most practices that facilitate the exercise of market power occur in markets that were noncompetitive to begin with. In these situations consumer losses are relatively larger and producer gains smaller. Finally, the administrative problems of computing actual efficiency gains and netting them against actual consumer losses would be heroic. It would require estimates about loss of consumers' or producers' surplus in an area of the demand curve where there is currently no production. Computing deadweight loss is far more difficult than computing a simple overcharge on already made sales. As a result, antitrust analysis without significant exception proceeds by condemning practices that result in lower output and higher prices. Efficiencies change outcomes only when they are sufficient to keep output at or above pre-practice levels, which means that there is no consumer harm at all and nothing to trade off.
In contrast to the voluminous literature on welfare tests for antitrust policy, relatively little has been written about consumer welfare and intellectual property law. A well functioning IP system would increase consumer welfare in both the short and long run. That is, no tradeoff need be made between lost consumer surplus and productive efficiency. In the imperfect system that we have, however, consumer losses do occur, mainly when exclusive IP rights are given in excess of what optimal innovation policy requires, or when the rights are given to things that would have been developed (or have already been developed) anyway by ordinary market processes. Unfortunately, producer capture in the IP law making process has largely excluded consumer interests.
Hovenkamp, Herbert J., "Consumer Welfare in Competition and Intellectual Property Law" (2014). Faculty Scholarship at Penn Law. 1830.