What is the UK doing in the field of digital-market regulation, and what do you think it is achieving?
There are probably four areas to consider.
The first is that the UK’s jurisdiction on mergers increased with Brexit. The UK is not subject to the same turnover threshold as under European law, and this enables it to call in a wider range of deals. It has also been able to look at different theories of harm in digital markets. It has done that in probably more than 10 cases where it examined issues like potential competition, vertical exclusion, etc.
Although it may sometimes be costly for firms, I think it’s quite good that these questions are being asked and explored. Potential competition conglomerate effects could give rise to harm even though they are often presumptively pro-competitive. And I think the Competition and Markets Authority (CMA) has done a good job in asking those questions.
The second area is enforcement against single-firm conduct. The CMA has done a small number of cases, one against Google. But I don’t think that’s been so far the mainstay of what it’s been doing on digital.
The third area is the digital-market regime, where the UK is pretty unique. Under this regime, the CMA can conduct market studies or market-investigation references. The CMA has conducted market-investigation references on digital advertising and cloud services. Concerning its activities in cloud services, the CMA faced some legal setbacks but won, overall, in court.
Fourth, there is legislation in Parliament, the Digital Markets, Competition and Consumers Bill (DMCC), that will strengthen the enforcement arm by improving interim measures.
What are your thoughts on the DMCC?
The legislation will notably make the remedial process with regard to market-investigation references more agile. Currently, if the CMA does a market-investigation reference and puts in place a remedy—which you can do, including a structural remedy—the review process for that, if the market conditions change, is unbelievably cumbersome. Making that more flexible is probably appropriate.
The DMCC also establishes an ex-ante and open-ended regulatory regime. The legislation has been amended a bit, but I think it still lacks clarity. While I don’t object to ex-ante regulation that follows the same substantive rules as abuse of dominance, I am concerned that the Digital Markets Unit (DMU) doesn’t provide the CMA with a limiting principle. This will lead to a lot of complaints. There will be complainant companies whose interests do not particularly align with the consumer, and the CMA may struggle to hold the line at the standard of “Is this going to enhance productivity growth? Is it going to enhance consumer welfare?” And there could be tension between the Competition Appeal Tribunal (CAT) and the CMA on the threshold for intervention.
Also, a big movement at the moment argues that the competitive process is a good thing in itself. In economic theory, a less-efficient competitor can improve corporate welfare. There are scenarios in which that can happen. However, the legislation also does not provide such a standard.
There is also the possibility of large transfers, particularly to news publishers, like in Australia. I worry that it will make the UK a slightly more rent-seeking economy, rather than a productivity-enhancing economy; or, at least, a competition regime that contributes more to rent seeking than competition. So, that’s my concern with it.
All in all, I think a lot of what the CMA is doing is sensible and good. But I worry it’s going to get a lot of power. And when you have a lot of power, there’s a risk that it gets used by others.
You mentioned that one of the problems under the DMCC/DMU would be the lack of limiting principles. What would be appropriate in this respect?
I think something about consumer welfare, and I think something around similar standards, using a different instrument to get the same outcome you would get under existing substantive competition law. In other words, this intervention is going to improve competition in the market for the benefit of consumers. The proxy for that might be that it benefits smaller competitors if that improves outcomes for consumers.
But I can think of so many examples, like retail grocery, for one, where over the years, there’s been a huge pressure on competition authorities to intervene to protect small shops. Even if consumers are not going to small shops and high streets are dying, your government might decide it wants less-efficient, out-of-mode supply to continue in the market for social reasons.
If, when electricity came in, the government had an objective to keep the candle or paraffin industry from suffering, it would have been a tough job. And so, with technological change, you’re always going to have winners and losers, and we know from productivity economics that actually, the biggest productivity effect comes from entry and exit, not from horizontal competition between existing players.
The big risk is that we tilt away from standard competition-policy rules toward protecting smaller firms.
It seems you are in favor of protecting less-efficient competitors so long as this actually benefits consumer welfare?
When I was in government, I always thought about consumer welfare as long-term consumer welfare, and I talked about it in a dynamic context.
If you have a market with big learning effects, scale effects, and so on, it may very well be the case that the entrant will not have the same level of efficiency as the incumbent for quite some time. But if the incumbents’ prices are up here, and the entrant is going to bring their prices down and cause them (the incumbents) to be more innovative in response to that entry, then the entrant can get more efficient over time. I think this is the sort of example where less-efficient competitor tests may be useful and where the as-efficient competitor test is not exactly the right proxy for consumer welfare.
Growth is also essential. When the UK government reformed its competition law at the beginning of the 20th century, the founding document was a white paper on productivity growth. The fundamental rationale that Gordon Brown had for the big reforms of UK competition law—including our market regime, our mergers regime, and our prosecution of criminal cartels—was getting competition to increase productivity growth.
Remedies have so far proved largely unsuccessful in digital markets. Do you think this will be different under the DMCC?
In abuse-of-dominance cases, I think it can be difficult to design remedies. This is particularly so in markets where the firms with high market shares have been successful in an innovative virgin space. A lot of the thinking on remedies in competition law in Europe resulted from lazy and inefficient state-owned monopolies. Remedies were very easy to design in that space; you just needed to change the law to allow new entrants.
Ireland, for instance, had lots of statutory monopolies. And so the European Commission came along and said, “No, you get rid of these.” And Ireland got rid of them. And we got the second cement company, competition in electricity and telephones and other things. And in some cases, technology came along, like mobile telephony, which eroded some of the market power of fixed-line telephony. For example, when my parents got married, they had to wait four years for a telephone because that was the gentle pace at which the telephone monopoly thought it appropriate to roll out telephones to people. They had a lot of power from being lazy.
Remedies have also been more successful in pharma cases. The European Commission has brought many of those. These cases have also been more amenable to remedies, because they stem from the way in which intellectual-property protection operates. A little bit like a statutory monopoly; it’s coming from legislation.
Problems arise when competition issues stem directly from a strategic entry barrier or an entry barrier that’s inherent in the market process—where you’re having to make judgments on whether an entrant could achieve the scale to effectively compete with an incumbent that is operating close to the efficiency frontier. It’s quite difficult to design a remedy that’s going to really work in these cases.
Does that mean you’re somewhat skeptical of the remedies the DMCC will be able to achieve, or have enforcers learned the lessons from cases like Microsoft and Google?
I’m skeptical. And the other reason why I’m skeptical is that, if I look at other regulated markets, like telecoms, energy, water, and so on, what I’ve seen over the last 20 years with regulatory systems that are bespoke and sector-specific, is that the regulator just gets more and more involved in resolving business-to-business disputes within the industry. And the industry builds up bigger regulatory functions, so that a huge amount of what a company does is mediated by the regulator.
At the end of the day, the whole industry becomes a bit slower, a bit less competitive, and nobody’s any better off. It’s like everybody’s standing up at a football match, and nobody sees any better, but everybody’s less comfortable. The consumer is the one who suffers in that, because what will happen is that the incumbent firms will be very very good at anticipating what the regulator might do—gaming it, persuading it. The regulator is going to rely on information from them, and the remedies are not going to work quite as intended. And you have the best of intentions, people doing a good job, companies that are trying hard to comply and everything else. But at the end of the day, the experience of other regulated industries is not enormously reassuring.
Therefore, I worry on two fronts. I worry that the people who think that the Digital Markets Act (DMA) or DMU is going to be the panacea to all the problems are going to be very disappointed. And in five or 10 years’ time, we’re going to be having a debate about needing new regulatory instruments, because this does not work. I also worry that we’ve created an industry of lobbying and protectionism that will be very difficult to dismantle, because these things can’t easily be taken away.
For 20 years, I have watched competition authorities look at sector-specific economic regulators and think of them as slightly captured. It’s a little ironic they now all want to become the thing that they didn’t particularly like.
You mentioned elsewhere that self-preferencing can sometimes have redeeming benefits for consumers. Could you expand on that a little bit?
I think we should look at self-preferencing through a standard competition-policy lens, because it has ambiguous effects.
For example, when laptop computers were first in use, you needed a modem slot at the side of the computer in the laptop to connect to the internet. Suddenly, computer manufacturers like Dell started to make their own modems and integrated them into the computer. I think a self-preferencing case in that context could have delayed the (innovative) process.
Another example is grocery retailing, where there are a lot of strong brands. Strong brands charge high prices, and consumers know there are benefits associated with that. But ultimately, poorer consumers benefit hugely from the ability of supermarkets to have their own label products. That’s why 50% of what’s on the shelves in the UK are the supermarket’s own label.
But self-preferencing could also be used to raise rivals’ costs, or exclude rivals, etc.
So outright bans on self-preferencing could harm consumers as much as they could benefit them. This is why I think we should apply competition-law standards to this behavior, and why I think a blanket ban on self-preferencing doesn’t make sense.
If you were head of the DMU, or in that sort of position tomorrow, what would be an enforcement priority? How should someone in that position think about the new powers they have and how they should use them?
I think if I was in the UK CMA doing that, I would be very very keen to make the injunctive-relief process work better. It’s something I’ve been enthusiastic about since I ran the agency, and we amended the law 10 years ago, but it didn’t seem to have had the desired effect. And I think I would focus on that as a way of getting these things done.
I would also develop guidance that establishes limiting principles on what the agency can do and try to prioritize cases that have clear and tangible long-term consumer benefits. In addition to the direct impact on consumer welfare, this would disincentive rent-seeking business-to-business disputes coming forward.
Finally, I would follow what enforcers are doing with competition law. I think some of the competition cases that are being brought around the world are quite good, because many practices of tech companies can be treated under competition law. And so, consequently, I would be trying to use the DMU to go after those and to try and get quicker resolution on some of these issues.
