What Covid-19 teaches us about barriers to entry and expansion
When I started my career in the 1990s, the standard mode of merger analysis was to define the market, calculate market shares, and make a preliminary assumption that market shares equate to market power – so material increments to shares, particularly in concentrated markets, led to prima facie concerns. One then explored whether there were any barriers to entry or expansion that would imply that equating high shares with market power was a safe assumption. If no such barriers existed, competition authorities were often relaxed about the likely effects of mergers, on the basis that any attempt to raise prices would lead to entry or expansion post-merger, and so to prices returning to their pre-merger state, with a timeframe of 1-2 years typically considered. This approach led to a strong focus on two areas: first, on defining the market; second, on exploring barriers to entry and expansion.
Over time, thinking and practice has evolved. Merger analysis in many jurisdictions now has a much more sophisticated approach to assessing whether there is a loss of static competition. The focus is on the closeness of competition between merging parties, rather than on their shares on a particular market definition. This approach recognises that market shares are often not a good guide to the competitive interaction between firms and market definition in a situation where firms’ offerings are differentiated often creates an arbitrary division between those firms that are “in” and “out” of the market. Rather, measures of closeness such as the diversion ratio are typically employed, together with estimates of upwards pricing pressure (such as the GUPPI). These approaches give a better assessment of the level of any static loss of competition.
There is much to be welcomed in this evolution. In many circumstances market definition and market shares are blunt tools that give unhelpful results from the perspective of genuinely trying to assess the likely impact on a merger. However, the flip side of a more sophisticated approach to assessing the static impact of competition has been an almost total abandonment of the assessment of barriers to entry and expansion, and so a failure properly to assess whether the future dynamic evolution of the market is likely to offset any static incentive to raise prices. Models such as GUPPI assume that market structure is fixed and there is no possibility of any firm entering or expanding in response to a price increase by the merged firm.
To take a UK example, an area where barriers to entry and expansion seem to be very low is high street retail. Since the recession of 2008-2009, and the increased shift to online retailing over the last decade, occupancy rates on High Streets have been well below 100% in almost all areas of the UK. This suggests that if there were profitable opportunities on the High Street, then firms would enter or expand their operations in response. But merger assessment practice typically focuses on the static impact of mergers only, and entirely ignores the dynamic effects. For instance, in CMA’s 2015 assessment of the acquisition of 99p Stores by Poundland, two high street discount retailers, the Final Report contains no assessment of barriers to entry or expansion at all.
Is this sensible? Perhaps the faith of the 1990s in competitive markets identifying profitable opportunities and responding to those was misplaced, and it has been a sensible approach to cease to give arguments about barriers to entry and expansion much credence. On the other hand, if markets do identify profitable opportunities quickly, then the current static approach is likely to lead to errors of excessive enforcement.
It would therefore be helpful to try to test the speed of response to profitable opportunities. This is challenging in normal circumstances, as it is hard to identify profitable opportunities and see responses to those in real time. But these are not normal times. The Covid-19 pandemic has led to a dramatic change in the structure of the economy, given the lockdowns that have been imposed on economies. One can therefore see how firms have identified profitable opportunities and adjusted their businesses to seek this out, pretty much as it happens. Examples of which I am personally aware include:
- almost immediately, videoconferencing services such as Zoom and Houseparty saw dramatic increases in usage;
- within a week of lockdown, fruit and vegetable wholesalers such as those at New Covent Garden market in London were setting up delivery services to sell their produce to residential customers (much of which would normally have been sold to takeaway food outlets and restaurants);
- cafes were selling fresh yeast in response to the dramatic increase in home bread making and the decline in the demand for restaurant-consumed items;
- flour suppliers were readjusting their customer base, having initially struggled (i.e. in the first few weeks) to adjust to greater demand for flour in 1.5kg bags rather than catering sacks; and
- children’s bookshops were adding local delivery services.
Other examples I have heard about include garage forecourts expanding the range of services they sell as they are classed as essential retailers, substantial expansion of sales of hair clippers, a dramatic expansion on online delivery capabilities in supermarkets, and many new suppliers of facemasks. No doubt others have their own examples that they can point to depending on their own experiences.
These examples suggest that the market mechanism has led to an extremely rapid response to new profitable opportunities and the abandonment of now unprofitable activities, often within a matter of days or weeks. This is Adam Smith’s invisible hand, visibly at work.
One cannot rule out, of course, that barriers to entry and expansion affect some markets. But a lesson of Covid-19 would seem to be that given a profitable opportunity, firms will seek it out, even if this involves adjusting their business processes, finding new delivery mechanisms, or alternative sources of supply. An approach to merger enforcement that looks only at static effects, and does not consider the rational responses of rivals to profitable opportunities that it predicts to emerge, is likely to lead to over-enforcement. This suggests that we should return to a situation where the assessment of barriers to entry and expansion plays an important part in the merger assessment process.