Use current concentration thresholds cautiously

An HHI number is a screen within a broader competitive analysis.

The 2023 U.S. Merger Guidelines describe a market with HHI above 1,800 as highly concentrated and treat an increase above 100 points as significant in the structural presumption described there. For economics exam purposes, however, the durable skill is calculating HHI and interpreting a rise as increased concentration. A question should supply any enforcement threshold it expects you to apply because policy guidance can change.

Even a correct HHI can mislead if the market is defined too narrowly or too broadly. A merger of two local sellers looks more concentrated when the relevant market is one town than when customers readily buy from a national online market. Evidence about entry, substitution, capacity, coordination, and buyer power completes the analysis. Concentration raises a question. It does not by itself prove monopoly conduct.

Concentration change

Two firms with shares 15 and 10 merge. The HHI increase attributable to combining them is 2(15)(10)=300, the same result obtained by recomputing the squared shares.

Market share is not market power by itself

Ask whether customers have alternatives, rivals can enter, and the firm can profitably keep price above marginal cost. Share is evidence within that analysis.

Government concentration thresholds are administrative screens, not timeless economic constants. Agencies revise guidelines as economic evidence, law, and enforcement priorities change. A test of durable economics should either supply the threshold it wants applied or ask for a qualitative conclusion such as “the merger raises HHI and concentration.” Memorizing one year’s number is less valuable than understanding the index and its limitations.

Even when a threshold is supplied, the order of analysis matters. Define the relevant product and geographic market, calculate premerger HHI, calculate the postmerger level and change, and then consider competitive conditions. An arithmetic presumption can focus investigation. It does not replace evidence about substitution, entry, capacity, coordination, or efficiencies.

Same firms, different market definition

Two clinics have shares of 30 and 20 percent among providers within a small county. If patients readily travel to neighboring counties or use telehealth, those shares may overstate constraint lost by a merger. If emergency patients cannot travel and insurers need both clinics, the local definition may be economically meaningful. The correct boundary follows customer substitution, not political lines alone.

Entry must be timely, likely, and sufficient to discipline a price increase. The mere possibility that someone could open a business is not enough if licenses take years, customers face switching costs, essential inputs are locked up, or minimum efficient scale is large. Conversely, firms currently outside the market may provide strong potential competition if they can redirect capacity quickly.

Market shares can also change rapidly in innovative industries. A snapshot may miss platform effects, data advantages, ecosystem lock-in, or a disruptive entrant. Strong buyers can sometimes resist price increases, but buyer power protecting a few large purchasers may not protect smaller customers. Imported products matter only to the extent they are credible substitutes and supply can expand.

Efficiency claims require discipline. Lower marginal cost can lead to lower prices or improved quality, especially if competition remains, but a firm with greater power may retain some gains. Fixed-cost savings affect profitability yet do not automatically change the profit-maximizing marginal decision. An exam distractor may cite “economies” without connecting them to consumer outcomes.

Keep legal and economic language separate. Higher concentration means the market structure became more concentrated. It does not logically prove an antitrust violation, collusion, or consumer harm. Likewise, a threshold below a presumption does not prove the merger harmless. Use the wording supplied in the stem and do not claim more certainty than the evidence supports.