More on entry

No sooner had I posted about the competitive impact of new entrants on incumbents than the Commerce Commission published its fascinating analysis of what actually happened after the event to the mergers it cleared.

It's a tricky thing to do, because you can't see the (dreadful-word alert) counterfactual - what would have happened in the merger markets but for the merger. And even if you could, you'd still be left with the problem of sorting out what was down to the merger and what was exogenous, and you're very often not going to have the granular data you'd need to answer that question. So if you're going to try and figure out whether you made the right call, you're in all likelihood going to have to use some blunt instrument.

That said, blunt instruments can be surprisingly effective, and this latest exercise from Lilla Csorgo (chief economist in the Commission's competition branch) and Harshal Chitale (assistant economist, ditto) is well worthwhile: I can see other competition agencies following their lead. What the Commission did was this:

the Commission’s new approach to ex post reviews pulls back from the question of whether a particular decision was right or wrong. Instead it focuses on particular aspects of its original analysis to see which ones held true. In particular, the Commission reviews whether certain market conditions developed as predicted. This is done with the aim of refining and improving the tools and techniques used in making those predictions

The paper singles out 18 mergers involving 40 markets, where the decision hinged (broadly) on One Big Thing - entry/expansion, existing competition, divestment, and an aspect of countervailing power (buyers of the merged entity's product sponsoring new entry). Here are the results.

I mentioned yesterday that a competition authority doesn't want to be either congenitally upbeat or congenitally downbeat on the prospect of entry/expansion, and ought to try and steer some informed middle path. On these facts, maybe you could argue that the Commission was a tad too optimistic about entry/expansion turning up, with five instances where, in the end, it didn't - though as the paper notes, and I mentioned yesterday, in those five cases you couldn't "preclude the possibility that the threat of entry, or some other source of competitive discipline, helped assure pre-merger competitive outcomes" (my italics).

The analysis throws some light on why, in New Zealand circumstances, entry doesn't always happen when you might have expected it. Exchange rates can be wrong; general market conditions in the potential source of new imports may not be conducive; overseas companies may not give a monkey's about New Zealand (the paper puts this in proper 'opportunity cost' language). And it sensibly recommends adding those factors to the checklist of things to consider when you're weighing up the likelihood of new entry.

I liked this paper: if you haven't come across it before (it was presented at this year's NZ Association of Economists' conference), give it a go. As well as being intrinsically interesting, and a good example of identifying and institutionalising improved practice, it's a good accountability exercise: as it notes

Competition agencies have for years been under pressure to show their effectiveness by demonstrating that their decisions were “really the most appropriate ones that could have been taken.

It also raised something I hadn't thought about before. Competition authorities have generally moved on from mechanical analysis of mergers: there was a time when allowing five competitors to merge down to four, or four to three, or (especially) three to two, or (even more so) two to one, would have been completely out of the question. These days, things like the number of players and the concentration ratios take a backseat to competitive conditions, and rightly so.

But, as the paper says, there can be an underappreciated risk in this:

While there [are] often reasons to be concerned by mergers that result in duopolies, another is that the competitive consequence of withdrawal of one player, which can happen for any number of exogenous reasons, is monopoly. Given that not all exogenous changes are unforeseeable, greater consideration should be given to the possibility of such eventualities in more concentrated markets. 

That's good advice.