The CC Reports on Supermarket 'Power'

In February 2006 I had an entertaining email exchange with an "Information & Records Officer" at the Competition Commission. I was asking how they defined "competition" and unfortunately the matter was left unresolved after I'd followed three separate links that failed to answer my question. You'd think that the Competition Commission would have a decent understanding of the economic concept "competition", but even if that view was faulty, you'd hope they'd be able to tell you what it was. Having glanced through the report into the Groceries Market that came out last April, it's depressing to conclude that their reasoning is circa 1970s GSCE textbook. I can't find a single mention of the term 'contestibility', as instead they deem market concentration to be relevant. Isn't it a shame that after all this time, money and effort to create a 269 page report, they didn't feel it necessary to study competition theory?

In my lecture on competition policy I make the following bold claim:

• Regulators use the term “competition” as an adjective that labels the conditions of various market states. This is consistent with textbook economics, but is a faulty (and retreating) type of economic reasoning.
• Increasingly the discipline is returning to the pre-c20th attention to processes, evolution and spontaneous order that used “competition” in the way the general public does - as a verb.

Bottom line

• The argument in favour of competition does not rest on the conditions that would exist if it were perfect

Student's have responded by suggesting that I'm creating a straw man, but this report seems to support my view.

The main conclusions:

First, we found that several grocery retailers have strong positions in a number of local markets. Barriers faced by competing grocery retailers that could otherwise enter these markets mean that consumers get a poorer retail offer in terms of prices, quality and service than would otherwise be the case, while those grocery retailers with strong local market positions earn additional profits due to weak competition in those markets.

Second, we found that the transfer of excessive risk and unexpected costs by grocery retailers to their suppliers through various supply chain practices if un-checked will have an adverse effect on investment and innovation in the supply chain, and ultimately on consumers.

A section on "Measuring market concentration" (p.109)

There are a number of ways to measure market concentration. These include measuring the number of competitors in a market, the relative size of competitors as measured by sales shares (either by revenue or volume) and indices, such as concentration ratios or the Herfindahl-Hirschman Index, which reflect both the number of firms in a market and their relative size 

6.13 In summary, to assess the extent of local market concentration, we focused on those local markets with three or fewer fascias in total where one of those fascias had a share of local grocery sales area that is greater than 60 per cent within a 10- or 15-minute drive-time.174 We defined these markets as ‘highly-concentrated local markets’.

We are taking a number of steps to address the problems that we identified. We are recommending to Government and the devolved administrations that a competition test be applied, as part of the planning process, to proposed new stores (and proposed extensions to existing stores). The competition test will favour new entrants and grocery retailers other than those which already have a significant local market share. We will also require grocery retailers to relinquish control over landsites in highly-concentrated markets that we have identified as inhibiting entry by competing retailers. Further, we will be limiting the ability of grocery retailers to prevent land being used by their competitors in the future. p.9

Applications would pass the [competition] test if within a 10-minute drive-time of the developed store (as calcu-lated according to readily available software):
(i) the grocery retailer that would operate the store was a new entrant in the local area; or
(ii) the total number of fascias (including any of the full-range national or regional grocery retailers and symbol groups) operating larger grocery retail stores in the local area were four or more; or (iii) the total number of fascias were three or fewer and the grocery retailer operating the developed store would operate less than 60 per cent of groceries sales area (including the new or extended store). p.17

My understanding is that this all comes down to regulatory issues regarding local planning. There is nothing wrong (from an economic perspective) with creating land banks, the problem comes from a highly distorted market for land. The greater the costs of buying and building a supermarket, the greater the advantage of scale. The way to facilitate competition for the benefit of consumers is to allow shops reach their optimum size (whatever that may be) through free competition.

Note as well, this report isn't independent. Take a look through some of the submissions, and realise that vested interestes lie at the heart of the CC argument.

Comments

[Jim]

"The greater the costs of buying and building a supermarket, the greater the advantage of scale. The way to facilitate competition for the benefit of consumers is to allow shops reach their optimum size (whatever that may be) through free competition."

Taking this as read for the moment, does that mean you agree that this market is much less competitive than it could / should be?

"Note as well, this report isn't independent. Take a look through some of the submissions, and realise that vested interestes lie at the heart of the CC argument."

Those are the submissions by third parties, not the argument of the CC itself. I'm not sure what you would have preferred the CC to do - rule out accepting any submissions from certain third parties and listen only to impartial experts like yourself?

[Gabriel]

"Here we go again!" (tm)
{Tune in Benny Hill theme song.}

First of all, how can a concept be "faulty"? Internally inconsistent? You have indicated no internal inconsistency in any of the leading models (be them more G.E.-like or game theoretic). -- You can say that a concept is a poor choice for a framework of analysis for particular phenomena, but even if this were the case, the concept would not be "faulty".

The most you can ask of a bureaucracy such as this is for it to employ, to the best of their ability, the best economic theory available, as exposed by leading departments and I.O. scholars, which means that, given the quasi-consensus in the profession, that we should be looking for signs of them having read Tirole & Aghion, not Kirzner.

Anyway, it seems to me that in their analysis and in much mainstream theory, competition can be "more" or "less", in which case your attack on lack of "perfection" is a strawman. Perfect competition is *the most* competition but regulation can aim more modestly at at least increasing competition.

Lastly, it's very much worth emphasizing that most regulatory action either fails in its stated goals (whenever stated clearly enough ;-) or is just misuse of public powers for private means (corruption, etc.) but this suggests not a failure of theory but rather that they're using theory as a source of useful, superficial excuses to go ahead and do whatever they want to do.

Investigations into supermarkets are under way both in France and in Romania. Actually, in Romania they ended with everyone (supermarkets, suppliers, smaller vendors, government) agreeing to drafting a "code of good conduct".

The main issue is this... supermarkets create a surplus, pure profit, because of various forms of market power, some of it by politically-imposed scarcity in terms of location and so on. Other people, naturally, want a piece, and supermarkets will bargain with them to shut them up.

[Alex]

"First of all, how can a concept be "faulty"?"

I agree with this, in the sense that there is no language or concept police which to forbid someone to define "perfect competition."

"The most you can ask of a bureaucracy such as this is for it to employ, to the best of their ability, the best economic theory available, as exposed by leading departments and I.O. scholars,[...]"

This is actually not true. If most economists are deluded and use unrealistic models, one can do better by ignoring their advice, or by using models alternative to the mainstream -- if such good models exist.

Alex

[Gabriel]

Alex, it boggles the mind to think, either practically or in theory, how a bureaucrat in charge with antitrust would go about judging the merits of theory... Really!

For laymen, there's little choice but to trust what we can loosely define the quasi-consensus in any science. And it's a good thing (tm).

[Alex]

"Alex, it boggles the mind to think, either practically or in theory, how a bureaucrat in charge with antitrust would go about judging the merits of theory... Really!"

Then perhaps they should not be in charge of making decisions which depend on judging the value of theories.

Unless there is a strong consensus in the scientific community about something, doing nothing may be the best option. This especially the case when the supposed benefits of the action are minuscule -- as they would be in this case.

Alex

[Gabriel]

Only that the benefits are miniscule based on a theory-employing judgement.

Choosing not to act is just another choice and whether it's good or bad one can only be judged in relation to some theory.

Your beliefs inform your actions and your beliefs are based on (explicit or implicit) theories you hold. So you might as well make it explicit, to check for consistencies and so on.

I think it makes sense to rely on the judgment of the medical profession when it comes to health decision or to rely on the engineering profession when it comes to bridge building or place design. Similarly, some reliance could be placed on the economic profession. In which case AJE's debate partners should be his coleagues at top universities, not the (most likely) clueless bureaucrats that try to give an air of scientific respectability to their decisions (which most of the time have nothing to do with the stated goals anyway).

[Alex]

"Only that the benefits are miniscule based on a theory-employing judgement."

This is getting too abstract. Anyway, I don't claim that you should use zero theory for guiding action --that's impossible-- but that you can use very rough, coarse, simple but true theory to evaluate when the supposed benefit is large enough to likely offset the downside of government acting while not knowing what they're doing.

The requirements for such a theory are much more modest, so claiming that "You need a theory anyway" is a silly argument, even though it is true that I need some theory.

I don't argue that theories shouldn't be made explicit (and who said that arguments should not be directed towards the economic profession too?).

As I said though, this is getting too abstract.

As far as competition goes, some similar bureaucratic entity thought that in fact Walmart competes too hard -- so they decided to ban it from opening new stores.
So is harsh competition good or bad?

They don't really know what they're doing, but they like having the power to make such decisions all the same.

Alex

[aje]

Jim:

Taking this as read for the moment, does that mean you agree that this market is much less competitive than it could / should be?

I agree that any market with regulatory intervention (beyond what's necessary to provide a market) is less competitive than it could be. (Whether it "should" is obviously a normative issue which I won't get into.)

Those are the submissions by third parties, not the argument of the CC itself. I'm not sure what you would have preferred the CC to do - rule out accepting any submissions from certain third parties and listen only to impartial experts like yourself?

Slight misunderstanding, I wasn't claiming that they are the views of the CC, but that they are the arguments that are framing the debate. Contra Gabriel this is not a level playing field between various groups - this legislation is being driven by vested interests. The "third parties" are typically rival companies that can't compete in the current market. They are not arguing for a level playing field, but an active hand in their favour. It is fraudulent to report this as a "consumer investigation". I don't mind the CC posting submissions from crackpots like myself, my point is that the general public shouldn't be hoodwinked into thinking that they're best interests are dictating policy.

Gabriel:

First of all, how can a concept be "faulty"? Internally inconsistent? You have indicated no internal inconsistency in any of the leading models (be them more G.E.-like or game theoretic). -- You can say that a concept is a poor choice for a framework of analysis for particular phenomena, but even if this were the case, the concept would not be "faulty".

I think I understand your criticism, but perhaps I should clarify. The intent of that claim was that they're relying on a framework that isn't up to the task - i.e. unsound as opposed to invalid. I wasn't saying neoclassical theory was internally inconsistent, just that it is a faulty choice because it doesn't adequately address the problem that the CC are investigating. Namely, the neoclassical framework does not adequately deal with creative discovery, process, etc. It defines competition in terms of concentration rather than contestibility. Its charm is its abstract completeness, but employing it as a real world benchmark is misusing it.

However, I would also argue that there are internal contradictions. But I guess that deserves a separate post.

[Jim]

"I agree that any market with regulatory intervention (beyond what's necessary to provide a market) is less competitive than it could be."

You seem to be assuming that any differences in land values must be due to regulatory intervention. But land values will be higher in some places than in others without regulation. The fundamental tent of land economics is that land values are higher in urban centres because that's where demand is highest (due to transport costs, economies of agglomeration etc). Since, as you said, "The greater the costs of buying and building a supermarket, the greater the advantage of scale", then city centre location will naturally imply an advantage to scale, will it not?

[Alex]

"[The neoclassical framework] defines competition in terms of concentration rather than contestibility."

I should point out that if the main measure of market power used is concentration rather than, say, barriers to entry, then the bureaucrats and their economist advisers are simply incompetent instead of following some established neoclassical doctrine.

For instance, you can have a single firm in the entire market without having anything close to a monopoly -- indeed, while having something close to perfect competition -- if the barriers to entry are virtually non-existent.

What constitutes barriers to entry is also a subject where one may find incompetent opinions among economists. (Reputation for instance, is not a barrier to entry, neither is a large capital requirement for entry.)

I was fortunate to take classes with Franklin Fisher, who was IBM's economic adviser in their anti-trust trial and the government's adviser in the Microsoft anti-trust trial.
Through him I found out about incompetence horror stories among economists -- for instance some economist who claimed that IBM was a monopoly because it had profits. (That's nonsense, both because accounting profits are different from economic profits and because the economy may be out of equilibrium, in which case profits do not imply the presence of monopoly power.)

Alex

[aje]

Yes, obviously. I'm not saying that there's no advantage of scale in a free market. Just that this will be based on superior efficiency, which is a good thing for consumers.

[aje]

Alex - I agree. I make a point of distinguishing between barriers to entry and costs of entry. The latter include capital requirements, reputation effects, asset specific expertise, etc and all serve the purpose of creating a market test for potential entrants. I also stress Frank Knight's definition of profit, and contrast this to rents. Most neoclassicist's deem imperfect competition (i.e. any market activity in the real world) to generate rents. But rents only stem from barriers to entry, and these are under the control of government.

[Jim]

So if I understand you correctly you're simply defining 'cost of entry' as 'efficient' because they stem only from the market, while anything the government does represents a 'barrier to entry' and is therefore bad, whether or not the phenomenon in question (e.g. a particular market structure caused by significant scale economies) looks the same in either scenario (notwithstanding the fact that the government-less scenario is entirely conceptual).

If we're agreed that property development can involve negative externalities, then does a Pigouvian tax on those externalities represent a cost or a barrier to entry? What about a regulation that achieves a similar effect due to cases where taxes are inappropriate?

[Alex]

Barriers to entry are not always under the control of the government.

Suppose you have a very small market and the production process involves a large initial investment in a factory. If the extent of the market is large enough that it can allow the recuperation of the investment (plus some monopoly profits) for one factory, but small enough that the initial investment in two factories can not be recuperated (all this from a present value perspective) then you do have genuine, non-governmental barriers to entry.

Other examples exist.

I'm not so sure that the report described genuine barriers to entry. What do they claim those barriers to be?

Judging the degree of competition by market concentration is nonsense all the same.

Alex

[Alex]

It is ridiculous how correct the libertarian simplified story actually is.

First, they don't find significant barriers to entry intrinsic to the market:

"In conclusion, we do not find that convenience store operators currently face a barrier to entry or expansion that arises from any cost disadvantage relative to other grocery retailers."

Similar conclusions hold for mid-size retailers and there is a very qualified claim that there "may" be such barriers for large retailers.

"Regional grocery retailers and, to a certain extent, some large grocery retailers may, however, face barriers to entry or expansion in new regions due to the economies of density that existing retailers in those areas are able to derive from their distribution systems. We have not, however, identified any particular region of the UK where this might give rise to a finding that such a barrier was having an AEC."

So they don't have much evidence for such barriers to entry.

When it comes to city planning rules, however, barriers to entry exist:

"An inevitable consequence of a plan-led system that seeks to meet the broad range of objectives set out in paragraph 7.35 is that grocery retailers may not always be able to open a new larger grocery store in the location of their choice."

So the world of grocery markets works just like in a libertarian caricature: lack of competition is induced by the "benevolent" and incompetent planners.

This is too funny.

Alex

[aje]

Just to clarify - a barrier to entry is something that obstructs a more efficient competitor from entering the market. I think it's important to make a categorical distinction between obstructions that stem from legally enforced monopoly privileges, and obstructions that are part and parcel of a competitive market economy. If I am an entrepreneur and wish to sell a new product I must pass a market test of assembling the resources I need to produce it, and then sell it. Those resources are scarce, and have to be bid away from competing uses.

I completely reject the notion that a "barrier" is just anything that is costly to overcome. The only thing stopping Virgin from competing with Tesco is their assumption that they can't provide a more efficient business model. It's true that "barriers" might be small in reality, and "costs" might be very high. But they're a market test. Most of the time when people complain about "barriers" it boils down to superior efficiency.

[new jordans]

A friend is, as it were, a second self.

[Air Force Ones]

Let us recommend for the human being in crisis that he stop concentrating so upon the risks concerned as well as the difficultie¡ê?and focus rather upon the opptunity---for there exists normally possibility in crisis.

[Cheap Jordan 1]

Thanks for sharing. This website is to I too have to help. Very good.

[moncler down jackets]

Don't know what is wrong what is rite but i know that every one has there own point of view and same goes to this one