The following online article has been derived mechanically from an MS produced on the way towards conventional print publication. Many details are likely to deviate from the print version; figures and footnotes may even be missing altogether, and where negotiation with journal editors has led to improvements in the published wording, these will not be reflected in this online version. Shortage of time makes it impossible for me to offer a more careful rendering. I hope that placing this imperfect version online may be useful to some readers, but they should note that the print version is definitive. I shall not let myself be held to the precise wording of an online version, where this differs from the print version.
The Myth of Diminishing Firms
Communications of the ACM, vol. 46 no. 11, Nov 2003, pp. 25–8.
Ronald Coase in the spotlight
Attempts to grasp what the internet will mean for the future shape of business, now the dotcom bubble has inflated and subsequently burst, have led to an explosion of interest in the ideas of the 93-year-old economist Ronald Coase. Coase won the economics Nobel in 1991, but he did the work for which he is chiefly known as long ago as the 1930s, and this is the work which has acquired a new audience with the rise of e-commerce. The standard textbook e‑Business 2.0  has a foreword by business-strategy guru Don Tapscott saying that the digital economy is “destroying the old model of the firm” and explaining that Coase’s early writings are the key to this. The Economist magazine’s publication E-trends notes that “parts of established companies are vulnerable to being ‘blown to bits’” [5, p. 193], and quotes as its reason for this judgement Coase’s classic 1937 article on “The Nature of the Firm” , based on a lecture given in the Scottish city of Dundee in 1932. References to Coase abound in writings about e-business. His name has become much better known in recent years than at any earlier point in his long career.
There is indeed a link between Ronald Coase’s theory of the firm, and changes in the business environment brought about by information technology. But the implications of Coase’s ideas have been misunderstood. Coase’s theory does not justify the conclusions which are being drawn about “diminishing firms”.
The essence of Coase’s theory of the firm is an attempt to answer the questions why firms exist at all, and why they are the size they are, not larger or smaller. Before Coase, no-one had addressed the former question, let alone the latter.
In a market economy, we see numerous transactions carried out through contracts freely agreed between economic agents. But within this ocean of market relationships, we find “islands” within which decisions on deployment of resources are made via hierarchical, managerial mechanisms: the islands are what we call firms or companies. Why, Coase asked, are all economic interactions not market interactions? – why is economic life not carried on entirely by individuals contracting with one another, with no islands of managerial relationships?
Coase’s answer, which has been found convincing by economists for more than sixty years, was expressed in terms of transaction costs. There are costs involved in entering into a transaction on the open market: these include the cost of searching for suitable trading partners, acquiring information about what is on offer from a seller or what terms would be appropriate to offer to a buyer, negotiating a contract and then monitoring its fulfilment, and so on. Transaction costs are estimated to represent large fractions of overall economic activity in a modern society. Within a firm, the organizational costs of achieving corresponding arrangements are often much lower. A manager who needs a worker to execute a given task does not need to search throughout society to find one: he knows which employees he is entitled to give work to and what they are qualified to do. Little negotiation is needed; if the task falls within an employee’s terms of employment, he is bound to carry it out, and the payment forms part of a regular wage or salary rather than having to be bargained over. Likewise, a manager has ongoing control over a particular range of non-human resources, and deploys them at his discretion to achieve the functions for which he is responsible.
That is Coase’s explanation for why firms exist: they are a device for replacing high transaction costs by lower organizational costs. But, according to Coase, as a firm grows larger its costs for achieving arrangements managerially tend to rise. The larger the firm, the more complex and hence expensive its management becomes, until further growth would make the cost of managing the newly-internalized operations higher than the cost of transacting them on the market. That is how Coase explains company size: firms grow, until conversion of further transaction costs into internal organizational costs ceases to represent a net saving.
The frictionless economy
Why should this idea find a special resonance in the new world of e-business? The reason is that one of the most salient effects of internet commerce is to diminish various transaction costs. Search costs, in particular, are reduced dramatically when company catalogues and the like can be found in seconds on the Web. Negotiation costs may be greatly reduced by techniques such as electronic auctions. Some commentators have begun to use the term frictionless economy for the trading environment which the internet is bringing into existence, implying that transaction costs are becoming only a minor supplement to the prices paid for goods and services.
If transaction costs fall, Coase’s theory predicts that, other things being equal, the point at which organizational costs equal transaction costs should occur at smaller company sizes. This theoretical prediction chimes with observation of the current trend to outsourcing: many firms are focusing narrowly on their core expertise, and buying in goods and services that used to be produced in-house. Adam Wishart and Regula Bochsler have written about an “‘Age of the Pygmies’, dominated by herds of tiny fleet-footed firms” [10, p. 120].
The link between Coase’s theory of the firm, information technology, and outsourcing was first expressed (to my knowledge) in a far-sighted paper by Thomas Malone, JoAnne Yates, and Robert Benjamin , who predicted that “information technology will lead to an overall shift toward proportionately more use of markets – rather than hierarchies – to coordinate economic activity”. But the point did not continue to be stated in such measured words. The idea that Coase implies shrinking companies really took off with Larry Downes and Chunka Mui’s Unleashing the Killer App , containing a foreword by the academic superstar Nicholas Negroponte. Downes and Mui used Coase’s ideas to promulgate a new Law of Diminishing Firms: “as transaction costs in the open market approach zero, so does the size of the firm”. Since then, this has been the orthodoxy. Information technology reduces transaction costs, so e-businesses will be small businesses.
Some commentators make obscurer claims about the relationship between information technology and Coase’s theory. Claudia Loebbecke writes “The traditional rationale for the existence of companies, as articulated by Coase and others, is the minimization of transaction costs … This analysis is no longer generally valid … [information technology] has dramatically reduced transaction costs” . Loebbecke’s suggestion seems to be that since Coase’s theory makes inferences from transaction-cost levels to size of firms, if the levels change Coase’s theory is refuted. That is illogical: Coase did not assume that transaction costs must always reach the same level, he said that their level at any particular era will affect firm size at that era. Most commentators read Coase that way, and say that since transaction costs are falling, firm sizes are shrinking.
An over-simple interpretation
We might suspect that this is an over-simple interpretation, if we look at what Coase wrote in 1937 about the communication technology of that time. In 1937 there were no computers, but for instance telephones had recently been integrated into business operations. Coase felt clear about the implications: “Changes like the telephone and the telegraph which tend to reduce the cost of organizing spatially will tend to increase the size of the firm” (my emphasis).
The point is that Coase’s theory does not say that firm size is determined by one factor, the cost of carrying out transactions on the market. It says that what counts is a balance between two factors: the cost of achieving a transaction on the market, versus the organizational cost of achieving the same result through a managerial mechanism. Firms will shrink if transaction costs fall relative to organizational costs. If organizational costs fall relative to transaction costs, firms will expand.
Coase evidently saw the telephone and telegraph as impacting on organizational costs, rather than (or, at least, more than) on the costs of transacting business between companies. I am not sure why; in the era of operator-connected calls, perhaps the public telephone service simply was not a very convenient means for separate companies to do business, but company-internal phone networks were doing a good job of keeping branches and head offices in touch. I do not know whether that is the correct interpretation, but the question is now of historical interest only.
The point relevant today is that, if we want to know how information technology will affect company sizes, it is not enough to look at its impact on transaction costs. We also have to look at its effect on company-internal organizational costs. If organizational costs were unaffected by information technology, then lower transaction costs would be predicted to yield smaller firms. But organizational costs are heavily affected by information technology.
Enterprise resource planning
Probably the most significant single category of business software application is so-called enterprise resource planning (ERP) systems, supplied by vendors such as SAP and PeopleSoft. ERP systems are about integrating the internal operations of a company, so that the implications of a business action at any one point ripple automatically to all other points which are affected. An incoming order might trigger scheduling of production, outgoing orders to raw-materials suppliers, raising an invoice, and so forth, all with minimal human intervention. ERP reduces organizational costs – if it did not, it would not be used. Large companies do also sometimes link supplier companies into their ERP systems via extranets, so the cost savings associated with ERP may affect inter-company transactions as well as company-internal operations; but that is a secondary use of ERP. Its main impact is on individual firms’ organizational costs.
Information technology reduces internal organizational costs in many other ways too. Automation of functions such as payroll dates back long before the arrival of the internet, which has triggered the sudden surge of interest in Coase’s theory. But ERP, as well as being a particularly large-scale business application, is one which is spreading at the present time, so that organizational costs are changing simultaneously with changes in transaction costs arising from e-commerce.
If transaction costs alone were relevant to firm size and they were changing in one direction only, then the theory would predict that firms should be getting smaller. But if both transaction and organizational costs are falling, then no simple prediction is possible. Instead of merely looking at the direction of change, we would need to study detailed figures on how the potential costs of achieving particular arrangements in an internet-mediated market compare with the costs of achieving similar arrangements managerially within an ERP-based company. Obtaining such figures is difficult or impossible. Apart from the problem of factoring out expenses attributable to individual operations within the overall costs of running large-scale systems, many figures will be purely hypothetical. For a particular company at a particular time, a given transaction will be achieved either on the market or as a managerial operation, but not both.
An inaudible correction
Ronald Coase is well aware that his theory makes no unambiguous prediction about the impact of information technology on company size. He was interviewed for the “millennium edition” of the Wall Street Journal dated 1st January 2000 . Responding to the interviewer’s suggestion that reductions in transaction costs would lead to a society of individual entrepreneurs, Coase explained “The question is whether the costs [sic, for cost] of transacting decreases as fast as the costs of organizing. My guess is that sometimes it does and sometimes it doesn’t.” But, by then, the myth of diminishing firms had taken such hold that this correction was inaudible. Ten months later, a New York Times article by Bob Tedeschi  celebrated the anniversary of Ronald Coase’s original lecture by discussing how plunging transaction costs, thanks to the internet, are enabling companies to focus on narrow product slivers or business activities. Tedeschi quoted Coase as having little interest in e-commerce: “So much is wrong with economics that I’m trying to correct some other things”. But the bandwagon belief that Coase implies shrinking firms rolled on.
The truth is that transaction costs would not yield an unambiguous prediction, even if organizational costs could be ignored. Apart from Coase himself, the one author to have queried the myth of diminishing firms, to my knowledge, is Simon Avenell of Murdoch University, Western Australia, in a November 2001 conference presentation . Avenell pointed out that only certain classes of transaction costs, notably search costs, are reduced through internet commerce: some, such as protection of intellectual property rights, may well be increased.
Dazzled by the immediate
The community of business analysts seems to have been dazzled by what they can see, to the point of completely forgetting what they cannot see (but ought to know about). Internet trading is an exciting new fact of life which is available for us all to experience immediately in our homes. As consumers we get a direct sense of how the time and effort needed to compare prices and availability is shrunk by the internet, and it is easy for us to extrapolate this from e-tailing to business-to-business dealings. Consumers are not concerned with other transaction costs, such as ipr protection, which may be increased by the internet. And ERP is a back-office function that is normally invisible to people who interact with companies that use it – the average consumer has surely not even heard of ERP. So it is understandable how a consumer’s-eye view of business has fostered the myth of diminishing firms. However, the viability of a company depends on the costs of all its activities – not just the face it shows to trading partners.
No forecast possible
Unquestionably, Ronald Coase is an intellectual giant. It is remarkable to reflect that, before him, no-one had systematically asked the question why firms should exist. Coase provided an answer which continues to convince (as well as making many other profound contributions to economic understanding). But it is just a mistake to imagine that his theory tells us that information technology must shrink company size. It is not a mistake which Coase himself makes.
Many firms are smaller than their counterparts were twenty years ago: outsourcing is a fact of life. Outsourcing is encouraged by factors apart from information technology – often, it is a response to low wage levels in the Third World; but, in many cases, outsourcing exploits information technology to achieve collaboration across company boundaries over activities which, previously, were necessarily conducted within a single company.
At the same time, though, we are seeing many mergers producing larger companies. Hewlett-Packard/Compaq, and IBM/PricewaterhouseCoopers Consulting, are two recent examples in the information technology sector.
Coase’s theory of the firm gives us no reason to believe that either of these trends must predominate over the other. As information technology is more fully exploited by business in the years to come, the spectrum of firm sizes may shift towards more smaller companies; it may shift towards more larger companies; or the size spectrum may remain much as it is. Transaction cost theory cannot tell us which it will be.
1. Avenell, S. E-commerce and transaction costs. Online at fizgighs.notlong.com
2. Coase, R.H. The nature of the firm. Economica n.s. 4.386–405 (Nov 1937).
3. Coase, R.H. Talking about tomorrow. Wall Street Journal (1 Jan 2000); online at petruism.notlong.com
4. Downes, L., and Mui, C. Unleashing the Killer App. Harvard Business School Press, Boston, Mass., 1998.
5. The Economist. E-trends. Profile Books, London, 2001.
6. Kalakota, R., and Robinson, M. e-Business 2.0. Addison-Wesley, Boston, Mass., 2001.
7. Loebbecke, C. Online delivered content. In S. Barnes and B. Hunt, eds, E-Commerce and V-Busines, Butterworth-Heinemann, Oxford, 2001.
8. Malone, T.W., Yates, J., and Benjamin, R.I. Electronic markets and electronic hierarchies. Communications of the ACM 30, 6 (Jun 1987), 484–97.
9. Tedeschi, R. Coase’s ideas flourish in the internet economy. New York Times (2 Oct 2000); online at unpuzzie.notlong.com
10. Wishart, A., and Bochsler, R. Leaving Reality Behind. Fourth Estate, London, 2002.