In some cases there is a clear answer. For civil aircraft there is only room for two companies in the world, Boeing and Airbus. The scale of investment in a new product and the commercial risks involved mean that only two companies can make a living. In fact only one can really make a living, for were it not for subsidies from European taxpayers Airbus almost certainly could not be a credible challenge to Boeing.
That is probably the most extreme example of corporate concentration in the world, though some might argue that the position of Microsoft (by coincidence, also based in Seattle, headquarters of Boeing) was equally extreme.
At the other extreme there is - well, a colleague volunteered hairdressing, which was a better example than anything I could come up with. ,In between come the vast mass of other activities from motor manufacturing, to accountancy, to banking, to airlines and so on.
In this great in-between area comes pharmaceuticals. Ten years ago there were British companies called Glaxo and Wellcome and Beecham and now these three are to be stirred into the giant pot with the US SmithKline. Result: on some measures the world's third largest company. Yet despite its size, the merged group would only have about 7 per cent of the world market in prescription drugs. The pharmaceutical companies industry is not particularly concentrated compared with many others; the companies are large because the industry is large.
To see this, compare the top ten pharmaceutical companies with an industry where the structure is more settled, the motor manufacturers (see graphs). True, the basis is slightly different because one is in money terms and the other in units, but in the case of pharmaceuticals the largest player at present is only twice the size of number 10 and the merged giant would only be three times, whereas in cars the ratio is five to one.
Will the motor manufacturing model become the standard for most industries? Certainly the urge to merge is so powerful at the moment among accountancy firms, airlines, banks - to start at the beginning of the alphabet - that there is the slightly alarming prospect that many industries which still have sufficient players to ensure reasonable competition between them will be sucked into becoming oligopolies.
How far this trend runs will depend partly on regulation - it is possible, at a price, to keep companies apart - but more on the fundamental economics of different businesses. To what extent will all world business become more like civil aircraft and to what extent will it remain more like hairdressing?
There are at least five forces for concentration, but there are also at least five pushing in the opposite direction. The five in favour of concentration are:
the growth of cross-border trade, for as world trade is increasing faster than world growth, countries are becoming more specialised in their output;
the growing importance of global brands, a function of the globalisation of media and information;
the growing size of commercial risks;
the costs imposed by government regulation (it is easier for large firms to cope with these than small);
investor preferences, as funds tend to seek to place assets in large lumps.
That might seem a pretty overwhelming broadside, but the game is not entirely one-sided, for ranking against these are:
the surge of information as a result of the internet, which cuts the entry cost for many small businesses by allowing them to sell to a global market, and allows purchasers to track them down;
the need for greater specialisation of function, which forces large groups to sub-contract more and more of their activities;
the growth of services (including hairdressing), many of which cannot be traded across national borders, in relation to manufacturing (including aircraft), which will inevitably be traded across borders;
growing consumer preferences for exclusivity in both goods and services, which will enable specialist providers to sustain much higher margins than mass producers;
the growing importance of performers (sports and film stars, investment bankers, celebrity lawyers, and so on) in relation to executives.
The second list may not at the moment be as convincing as the first. I suspect that the forces for concentration have still quite a long way to run, but the game is not entirely one-sided. For those of us who are concerned about the concentration of power that seems to be taking place, here are three simple ways we can, individually and collectively, counter it.
One is to use our own purchasing power to favour smaller enterprises. For example, we should shop, where practical, at local stores rather than the supermarket chains. We should try and travel on smaller airlines. If our bank or building society is taken over by a larger one, we should move the account. We should avoid going to chain restaurants - and so on.
Second, we should encourage government not to load regulations on companies, as these almost invariably benefit the large ones. Large groups can afford the specialists to tackle the bureaucrats; small cannot.
Third, if we are working for a large group we should not use the financial power of that group to put pressure on small suppliers - for example, by that common trick of delaying payment of invoices.
These limited measures may seem poor defence against the power of concentration. But all economic change is at the margin: quite small changes in the pattern of spending or the way consumers make their choices have cumulatively large impact. What is happening in pharmaceuticals is one large industry catching up with the typical global structure of other industries. There are other forces, less visible, encouraging people to spend more of their income on goods and services which the multinationals will never provide - including hairdressing.