The details of SkyTV/Vodafone

Just before the Easter weekend the Commerce Commission published its written reasons for turning down the SkyTV/Vodafone merger. Like other competition tragics, I read it over the holiday - all 566 paragraphs of it.

Like everything that comes in the Commission's door these days, the Sky/Vodafone fell in that twilight zone where you can make a decent case for both sides. In the event, the Commission said No: the merged entity would be able, and incentivised, to offer Sky Sport based bundles of pay TV and broadband/telco services that other Internet Service Providers (ISPs) and telcos wouldn't be able to match. There would consequently be a real chance of a substantial lessening of competition (SLC) as Sport-less rivals were less able to compete.

Sky/Vodafone was especially problematic because it fell into two of the more difficult areas for regulators to assess - the alleged use of market power in one market to gain advantage in another, and allegedly exclusionary bundling. These are always going to be line ball calls: the ACCC for example stopped the Aussie supermarkets from giving out large petrol discount vouchers ("shopping dockets") to buy petrol from them, because they thought the supermarkets were using their market power in the grocery trade to drive the petrol stations out of the petrol game (I thought the ACCC could be wrong about that).

And anti-competitive bundling is just as tricky, not least because the economics is not settled (you lawyers at the back, stop sniggering). As the abstract from one review article said

While the [economics] literature has demonstrated the possibility that bundling can generate anticompetitive harm, it does not provide a reliable way to gauge whether the potential for harm would outweigh any demonstrable benefits from the practice. As a result, the widespread application of the antitrust laws to bundling by firms can generate significant error costs by erroneously condemning or deterring efficient business practices. In the future, economists should seek to expand their understanding of both the anticompetitive and procompetitive reasons firms engage in bundling.

So given all the inherent uncertainties, I think the Commission came to a reasonable conclusion. And I think it fits well with what regulators should do in fast-moving sectors, which (as I argued here) is take a conservative view (though it is not always obvious whether doing something or not doing something is the conservative course), don't give leg-ups to anything, and generally try to let events unfold, but deal with any anti-competitive rorts that survive the Schumpeterian storms.

What struck me most about the whole thing was, why did Sky and Vodafone go for a clearance ("there is no SLC") rather than an authorisation ("there is an SLC, but there are compensating benefits")?

Maybe they genuinely believed there would be no SLC; maybe they didn't want the relative hassle of an authorisation (typically longer and more expensive, and the associated conference process isn't everyone's cup of tea, either). But having gone that route, they were stone cold dead if the Commission found an SLC (or, strictly speaking, couldn't be satisfied that there wouldn't be an SLC). That can prove a difficult barrier to get over when, as here, it's a dynamic sector where anything might happen, and it only takes one plausible enough ("real chance") scenario where there might be an SLC for the Commission to throw its rider at the fence.

But with an authorisation, you're still alive with the second leg of your argument: there are SLC downsides, but more than compensating benefits. And there were benefits: as the Commission said, for example at [198], "consumers may be better off in the short term as the merged entity offers better bundles (including better prices) and rivals react with their own bundles (and lower prices)", or again at [226], "it is likely that the bundles offered by the merged entity would be lower priced and/or higher quality, with more additional features than those available without the merger, or compared to those available on a standalone basis".

The Commission may still have jibbed at an authorisation, because it clearly worried that any short term consumer benefits would be taken back when the Sky/Vodafone entity had seen off some of its competitors, but it would have been a better route to go. An argument along the lines, "clear short term benefits in the (semi-predictable) near to medium term, but uncertain detriments in the (much less predictable) medium to long term" was a real runner, particularly as the Commission said at [358] that it is "difficult to predict with any certainty what effects might occur in the medium to long term". A bird in the hand today, versus more heavily time-discounted and more uncertain detriments sometime in the future, could have snuck through.

Especially (if Sky and Vodafone are minded to appeal) as the Commission has taken a somewhat debatable view on how hard it would be for ISPs to get into the post-merger game. The evidence shows that there are heaps of them today, all giving it a go despite the much heavier firepower of the big guys. This suggests easy conditions for entry.

Of course, the Commission would respond, that was then, what about the post-merger world?  But I for one would be inclined to think that (for example) a determinedly "cheap and cheerful" barebones ISP (or several of them) could still keep the merged entity honest. I wouldn't rule out the possibility of some other innovative bundling by competitors, or other attractive add-ons. And I also had quite a bit of sympathy for the point that NERA made, quoted at [454]: "NERA advised that if consumers were attracted to the merged entity’s discounts, then they would equally revert back to other TSPs [telecommunications service providers] if those discounts were reversed. If so, the merged entity would not be able to exercise market power".

If there was anything else I'd quibble with, it was the Commission's view that content and the means of delivering the content are becoming more the same thing in consumers' eyes. For example it said at [170] that

As services become more converged, consumers are likely to start associating them as part of the same purchasing decision – the content and the method of delivery become more closely linked in the minds of consumers

and again at [386.4] where it said that

broadband and mobile services become more closely aligned with the delivery of content

Personally, I'd agree that delivery services are indeed converging, but I'd argue that they're becoming commoditised, and that content is increasingly standing alone, which is what the rest of 386.4 says:

the importance of content to competition in relevant telecommunications markets is also likely to increase

Who provides Game of Thrones, and over what infrastructure, is of the utmost indifference to me. They're not - for this sample of one - becoming all the one deal, wrapped up with the series itself. Rather, they've become markedly more separate. Though if you follow that logic further, you start to think dark thoughts about entities monopolising a market for premium sports content (the Commission found at [259] that there is one). There's a lot happening in the commercial and policy space around telecoms and convergence: I wonder if some regulation of 'fair access' to premium sports is lurking in the mix.