How to break up oligopolists

The most discussed problem with my proposals on how to fix capitalism is how one decides which companies to break up, and how one goes about doing it. Needless to say, I do have some answers


One thing that I do not think I can emphasise too much is that current competition (anti-trust in the US) law is also full of problems and grey areas. I am suggesting shifting the grey areas, not creating them. Many of the problems remain exactly the same, especially the key one of defining the relevant market in each case.

Another is that the aim of this is to make the real world more like text-book perfect competition markets. We can prove that perfect competition delivers optimal results, apart from the problems of inequality and externalities (how we can minimise those is another problem).

Current laws are problematic too

Current laws are also heavily subject to political influence and policy changes depending on the influence of the target (Murdoch in the UK) and how “business friendly”, i.e. mercantilist, current policies and attitudes are. The neo-mercantilist consensus of the time is at least partly to blame for the failure to impose effective remedies in the Microsoft case.

Breaking up a company does not lead to confiscation of wealth from shareholders. The same people end up owning the same assets (with some exceptions). The major effect is a loss of the centralised power that the management of the broken up company possessed. The shareholders may even be better off as a result. It has been argued that break-ups of giants such as AT & T and Standard Oil actually made shareholders richer, as well as making the market more competitive, although I have not been able to find clear evidence.

As I propose a reversal of the burden of proof, so that the ologopolists must prove the public benefit of their existence, it will be much easier for the regulators to take action. I am not concerned that there will be a failure to make a good case when one exists: these are big, rich companies, and if there is a case to be made, they are well able to make it. Nor do I think it is unfair: competition works, and to operating in a less than competitive market is a privilege, not a right.

The easy bits

The simplest businesses to deal with are those that are clearly not natural monopolies, nor have clear evidence of economies of scale that out-weigh the harm done by the lack of competition. There is clearly a need for someone to judge whether claimed economies of scale are strong enough, or whether they are big enough. I do not see that this is much more of a problem than the current need to prove that an alleged monopoly actually is one and that the monopoly is being abused.

This covers the majority of businesses: retailers, banks, most manufacturers, agriculture, food and beverages, etc. There are clearly economies of scale in large plants, but the case for companies that large own numbers of factories (or other units of production) is harder to make. There are certainly some benefits, but are they sufficient to undermine the basic functioning of a free market economy?

Many people will argue that there are no economies of scale in production, but claim that they exist in R & D, and that therefore the big companies are necessary for progress. The biggest problem is that it generally appears that small companies are more innovative. Where do most of the big new ideas come from: start-ups and academic research. When big companies do fund really innovative research they often fail to actually commercialise it successfully (Xerox Parc, for example). In most cases it is more likely that R & D is subject to diseconomies of scale.

With the huge global markets that now exist for most industries, a company can be big enough to gain almost every imaginable economy of scale while still having just a small proportion of the market.

Another argument is that big companies can take big risks with new products that small companies cannot take. The obvious problem with this is the existence of small business that do take big risks: all those internet start-ups and bio-tech companies commercialising one person’s academic research. Venture capitalists can fund them, and there are plenty of founders prepared to take the risk.

Suggesting that only huge companies can absorb risks also implies that there is a huge failure in either stock markets or corporate governance. These are diversifiable risks that shareholders should be a happy to take. If the companies they own are not, we need to fix the reason for that: the management are focusing on their own job security rather than their duty to the shareholders.

Natural monopolies

Natural monopolies cannot be dealt with perfectly. There are a number of methods, all of which have been tried extensively at various times and places — so at least we have the benefit of knowing a bit more about what works.

There are two methods I would favour in most cases. One is minimising the monopoly by proving competitors with as much access to the competitors infrastructure as possible. This makes the remaining minimal business easy to regulate (because it is simple) and provides competition where possible. Examples of this include local loop unbundling in telecoms, or multiple electricity suppliers sharing the same infrastructure.

An alternative is mutual or non-profit ownership. For example, a water company is owned by everyone who lives in the area it supplies. Unlike state ownership, the owners have a direct stake in its success and have little temptation to use it to achieve other goals (such as creating employment). This approach is best suited to stable industries.

The final alternative is state ownership, but I do not feel that this is necessary unless there is a fatal objection to both the other alternatives.

What cannot be broken up

There are some businesses that are hard to break up, and are not clearly natural monopolies in the traditional sense either. Google search is a good example. Even is the company were to be broken up, there is no practical way to share the customer base that uses and its national equivalents between several companies. There are also clear disadvantages to having lots of companies each of which would have to develop its own web crawling infrastructure: the economies of scale are fairly clear.

Strong brands are another problem. There can only be one Coca-cola, for example.

These need to be treated on a case by case basis. In some cases excessive vertical or horizontal integration can be prevented by forcing the spin off of separable businesses. For example, Microsoft could be broken up into separate operating systems, office software, server software and gaming companies. In that particular case, I think it is big enough to be broken up into two or more of at least the first two.

Network effects can also be minimised. Abolishing patents (or at least most types of patents) would help a lot. Much more could be done: forcing software companies to publish full specifications of their APIs and file formats, promoting any type of affordable inter-operability, etc.

Of course none of the above is easy to implement, and I have still not presented solution to many practical problems. However, being unable to figure out how to do this in all industries is not a reason not to do it at all. If we cannot work out what to do about Google, lets leave it alone for now and start by tackling Tesco, Glaxsmithkline and Vodafone.

Give shareholders an incentive to do it

One of my original proposals was to introduce progressive tax on companies, with the tax rates that increase with group value added (in the UK, small companies currently get a lower rate of tax based on group profits). This gives them a strong incentive to break themselves up to save tax.

It is a simple way of ensuring that scale would at least be questioned and management pushed to demonstrate its advantages — unless corporate governance is even more broken that I think.

2 thoughts on “How to break up oligopolists

  1. Interesting ideas. While I’m not sure that all of them are the right thing to do, you have done a lot of work on this, and at the least these ideas could form a foundation for future action.


  2. Any specific criticisms or ideas? Its a difficult area and there are a lot of things I am note sure of either.

    I have thought of one thing I think should be state run since I wrote this (postal services).

Comments are closed.