Dear Terry and Gunnar,

Terry’s reply brought new dimensions to this thread. These comments are useful, and one deserves further development.

Ronald Coase’s theory of the firm and the concept of transaction costs adds dimension to the thread that are not a necessary outcome of Gunnar’s earlier query.

The reason I focused on Ricardo’s theory of comparative advantage is that Gunnar’s query raised questions about the possible relation between David Ricardo’s theory of comparative advantage and Clement Greenberg’s theory of modernist art. Gunnar posed both of these in terms of specialization. I see neither as a case of specialization, and I don’t see Greenberg’s approach to modernism in relation to Ricardian comparative advantage.

Terry’s description of how digital information processing leads to changes within firms and industries was very helpful in several slightly different ways.

This shed light on the earlier thread about Watson, the computer that won playing Jeopardy. Computers and algorithms that enable computer to take over some human tasks have a range of things they do very well. These typically involve activities governed by tight rules that do not require preference, judgment, or emotional sensibility, or human engagement.

Human activities in nearly every field involve repetitious or mechanical tasks that may not require judgment. Many such activities involve a sequence of judgment-neutral tasks, repeated processes that are in principle always the same. Others involve a series of repeated mechanical motions. A computer or robots can replace human beings when the activities are essentially mechanical and algorithmic rather than judgment-based. This includes the activity of computing – the title “computer” was first applied to human beings whose job it was to undertake computations for other human beings. In general, computers worked for other people whose work required computing but who were either focused on another main task, too busy, or too highly paid to do their own computing. These human computers were among the first human beings to lose their jobs at the dawn of the computing era now known as the information age.

Computers can also manage processes that seem to require judgment. This occurs when brute force computing can be used to process large numbers of possible decision paths and outcomes at a speed that far outstrips human capacity. This is not a substitute for human intelligence, but a series of programmed pathways that allow a computer to replace the human judgment of likely moves and outcomes on such a rapid time scale that no human can manage to surpass the brute force approach in a rule-bound game. This is how Deep Blue beat Gary Kasparov, and this is why new chess software is even better. It is also how Watson won Jeopardy.

This explains two issues.

The first issue is that computers can replace much human labor because many tasks that human workers perform involve repetitious or rule-bound activities. For better – and for worse – this describes a great deal of human work. The fact that this is so was the inspiration for Frederick W. Taylor and his “one best way” vision of “scientific management” for planning work tasks. While this leads to productivity gains under certain circumstances, it also leads to problems. When human beings are required to carry out repeated tasks on a continuous basis, they grow tired and bored, their attention flags, performance and quality decrease. In the absence of defects or disruptions, computers and robots carry out programmed tasks exactly as planned without change or let-up.

The second issue is that computers cannot replace any human activity that requires judgment, empathy, intuition, or the capacities that lead to human preference.

Great amounts of human work involve activities that computers can replace. A significant but lesser amount of work is of the second kind. In some cases, the two aspects of work are closely linked, requiring an engaged human decision-maker who takes responsibility for the full process, the mechanical parts and the judgmental parts together. In other cases, it is possible to disaggregate different kinds of tasks. In those cases, computers can replace those aspects of a human job involving the first kind of activity, while those portions of the same job involving the second kind of activity are unbundled and aggregated with other tasks. Terry described some of the changes to banking and design firms that flow from the ability to unbundle parts of jobs while restructuring the remaining work so that fewer human workers can do it.

This is how mechanization and computerization in typesetting replaced a great many people, including typesetters, press men, and many junior designers. Computers have replaced nearly all the designers and typographers who were called “mechanicals” when I first worked in publishing in the 1960s. Most human mechanicals carried out the instructions of senior designers in the same way that human computers carried out the instructions of mathematicians, statisticians, physicists, and actuaries.

In this respect, Terry’s explanation of how computing has transformed industries such as banking is useful and elegant.

Terry’s description of transaction costs requires understanding Coase’s theory of the firm, an issue that did not appear in the post.

Ronald Coase won the Nobel Prize in Economics in 1991 for “his discovery and clarification of the significance of transaction costs and property rights for the institutional structure and functioning of the economy” (The Royal Swedish Academy of Sciences 1991.)

Coase’s (1937) article on “The Nature of the Firm” essentially created the field of transaction cost economics. In essence, Coase asked a simple yet profound question: “Why do business firms exist?”

In one kind of ideal world, economic theory proposes that we should be able to buy specialist services from one another in an open market. If this were the case, then – in theory – we could each focus on the areas in which we have comparative advantage or some preferential position. For anything else, we would buy goods and services from others.

But this is an ideal world of a certain kind. This world does not exist. Human beings create organizations and firms rather than simply trading for services in an open market. Coase examined this issue in a profound new way. When he asked, “why do firms exist?” he came to an answer that he labeled transaction costs.

Transaction costs are those costs that make it expensive in one way or another to buy goods or services in an open market as distinct from the cost or value of the goods and services we buy. The concept of transaction costs involves many issues – secure knowledge of the quality of goods and services; availability at the right time; true pricing; trust; and many other factors. This is a very short explanation lacking in detail and nuance.

In technical terms, transaction costs are “the costs of resources utilized for the creation, maintenance, use, and change of institutions and organizations. They include the costs of defining and measuring resources or claims, the costs of utilizing and enforcing the rights specified, and the costs of information, negotiation, and enforcement” (see: Furubotn and Richter 1997: 40). Firms exist because it is often less expensive and safer for an organization to provide these services within the boundaries of the organization than to buy them in an open market.

Transactions in open markets – and within the closed boundaries of the firm – can be described in the same way in economic theory. A transaction “occurs when a good or service is transferred across a technologically separable interface” (see: Williamson 1996: 379). In other words, what happens when organizations can provide their own services and goods across intra-organizational boundaries in the form of a transaction within the organization. But the key insight in Coase’s (1937) article is that firms exist to reduce all the many kinds of transaction costs that may exist.

Terry’s description seems to me a bit confusing for two reasons. One is that some organizations provide services and information as their direct product. The description in Terry’s post placed material and wages under direct costs and information services and decision-making under transaction costs. This is not entirely the case in some organizations. The other confusion involves the fact that many of the transaction costs that exist in open markets diminish or possibly disappear within the boundaries of some firms. The description did not mention these or note that one reason that transaction costs are described as transaction costs arises from the fact that these costs were costs incurred when individuals or group bought goods or services outside the group in a trade transaction. Indeed, the notion of trade transactions under the theory of comparative advantage would be one reason to think of Coase’s theory of the firm as an advance against Ricardian economics. While in one sense this is the case, it is useful to remember that David Ricardo was born in 1772, dying in 1823, nearly 90 years before the birth of Ronald Coase. Like Newton before him, Coase saw farther because he stood on the shoulders of giants.

In a short and not very detailed explanation, the major reason for creating firms is to ensure that human beings can manage the goods and services vital to a business. If market failures can impede business activities, it is best to internalize them within the firm. It therefore makes sense to create organizations that can take on and allow human beings to control the processes, goods, and services central to the core business even though they might otherwise purchase them in an open market.

To drill down into the details of Terry’s explanation of transaction costs would require a careful and nuanced post that would be far longer. I’m up to nearly 1,600 words now, and I don’t have time to write at the length of a journal article to explain more.

Since I don’t have time to develop this issue properly, I have posted a copy of the Coase (1937) article, “The Nature of the Firm” to my Academia page. I will leave it up until September 27. It is accessible under teaching documents at URL:

http://swinburne.academia.edu/KenFriedman

Those who wish to learn more about Ronald Coase and the work that won the Nobel Prize will find useful information and linked resources at the Nobel Prize web site:

http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/1991/coase-facts.html

The 1991 press release is especially helpful.

There is also a wealth of materials at the web site of the Ronald Coase Institute:

http://www.coase.org/index.htm

The obituary on this web site provides a nice summary of Coase’s work and its influence.

Yours,

Ken

Ken Friedman, PhD, DSc (hc), FDRS | University Distinguished Professor | Swinburne University of Technology | Melbourne, Australia | [log in to unmask]<mailto:[log in to unmask]> | Mobile +61 404 830 462 | Home Page http://www.swinburne.edu.au/design/people/Professor-Ken-Friedman-ID22.html<http://www.swinburne.edu.au/design> Academia Page http://swinburne.academia.edu/KenFriedman About Me Page http://about.me/ken_friedman

Guest Professor | College of Design and Innovation | Tongji University | Shanghai, China

--

References

Coase, R.H. 1937. The Nature of the Firm. Economica, New Series, Vol. 4, No. 16 (November 1937), pp. 386-405.

Stable URL:
http://links.jstor.org/sici?sici=OO13-0427%28193711%292%3A4%3A16%3C386%3ATNOTF%3E2.0.C0%3B2

Furubotn, Eirik G., and Rudolf Richter. 1997. Institutions and Economic Theory: The Contribution of the New Institutional Economics. Ann Arbor: The University of Michigan Press.

The Royal Swedish Academy of Sciences. 1991. “The Prize in Economics 1991 - Press Release.” Nobelprize.org. Nobel Media AB 2013. URL: http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/1991/press.html
Accessed 2013 September 12.

Williamson, Oliver E. 1996. The Mechanisms of Governance. Oxford: Oxford University Press.

--

Terry Love wrote:

—snip—

After reading your post, I wondered about the benefit of looking beyond Ricardo. The advent of electronic data processing has changed many of the assumptions on which Ricardo’s analyses and theories were based. Perhaps more importantly, currently we are in a transition situation between the before and after effects of electronic data exchange and knowledge
automation.

In what follows, the focus is on the effects of electronic data exchange, the transitional effects of knowledge automation are even more significant but left to another time.

There are radical changes afoot due to electronic data processing practices that have been obvious since around 2000 and are now in play and these radically (as in ‘by its roots’) change the ‘specialization picture’ and appear to turn some of Ricardo’s theories on their head, although you might regard what the issues that follow in terms of a redefinition of Ricardo’s ‘rent’.

Six issues relating to the effects of electronic data processing on the future benefits or otherwise of specialisation are:

1. Dis-Economy of Scale (the opposite of ‘economy of scale’)
2. Role of Coasian Transaction costs in defining the dis-Economies of Scale and ‘economies of Scale’
3. Effects of electronic systems and improved algorithmic/neural net systems) (Computers, Software, wide and local area networks, Internet, Artificial Intelligence, computers such as Watson, etc.) in reducing the transaction costs that result in dis-economies of scale.
4. Consequenct significant increases in economic efficiency (profits) of organisations from operating at massively higher levels of scale than currently found.
5. Subsequent pressures towards acquisitions and mergers and extinction of ecosystem niches for very small specialist organisations.
6. Consequent tendency towards reduced system stability and increased possibility of global system collapse due to reduced number of stabilisaiton loops in the systems and high positive feedback.

The cost of producing an item or providing a service can be separated into two components: a) the direct costs including materials, wages etc; and, b) the transaction costs (taken more broadly as the costs of providing the organisations infrastructure, documentation, decision making etc. associated with and providing support for the activities that generate the direct costs). Note: this is an extension of the concept of Coase’s theories about transaction costs that emerged in the MIS field around 2000.

As organisations increase in size, the direct costs reduce (what is often called the economies of scale). Problematically, however, the transaction costs increase as organisations increase in size because there is more internal communication and the like (this is called in MIS, the ‘dis-economy of scale’). The maximum size of organisations occurs when the effects of the dis-economies of scale due to transaction costs mean that there is no more advantage in increasing size to gain economies of scale.

Electronic systems significantly reduce transaction costs. For example, around 2001, to process a cheque cost a bank around $20 per cheque using conventional banking methods. To processes the same money as an electronic payment via internet banking cost around 1/10 of a cent. The transaction cost reduced enormously. Banks had been limited in scale by the increases in internal transaction costs of managing a larger banking organisations. The reduction in transaction costs offered by electronic banking enabled banks to jettison their previous transaction processes (e.g staff, and branches) and offered scale benefits leading to mergers and transition to a much smaller number of larger banks. Incidentally, as transaction cost reductions were uneven, there were increased advantages in banks moving towards retail banking (where transaction cost reductions were highest) and with a tendency to specialise (as a retail bank, or commercial bank etc) at a much higher organisational size and then gain the benefits of market segmentation by offering boutique services under different brands as if they were smaller organisations. This is pseudo-specialization.

The same is found in any organisational ecosystem. When I interviewed CEOs of large design companies around 2003-2005, the above process was playing out with them laying staff off because ot the reduction in ‘transaction costs’ (and improved productivity) offered by Apple computers. Typically, it seemed that they moved to laying off 3 staff in 4. Profitability of the larger firms increased. Smaller firms operated in the eco-system niches. Many designers moved into design education. One of the aspects of ‘transaction costs’ in design activity is the cost of the ‘transactions’ with the body of knowledge and expertise (fonts, research, best practices, etc). These aspects of the transaction can now be accessed directly using computer software and hardware(Adobe and Apple spring to mind again) at much lower ‘transaction costs’ than training and using individual human designers. In essence, you can see the scale and profit changes of the design industry being split between ‘design businesses’ and ‘automated design service software businesses’ (Adobe and Apple) . The benefits have almost exclusively gone to the latter, who have specialised, reduced transaction costs, and massively incr4ased in size and profit via transaction cost reduction. In contrast, small design firms are on a much slower track caught between specialisation, reducing prices, and merger and acquisition to reduce transaction costs. Worse, the reduction in transaction costs offered via Adobe, Apple and the like, have resulted in a large body of competitors as diy desktop publishing. The latter resucing the scale of the market and creating its own pressure on prices. In theory, this leaves design businesses caught between multiple profit reducing factors, some pressuring towards specialisation to increase profit by cost reduction and some pressurising towards ‘generalisation’ to have sufficient work at smaller organisational scale. All are shaped by the effects of changes in transaction costs due to electronic systems.

Similar developments are obvious in university systems.

Mostly, this goes a bit beyond Ricardo. Which is I guess part of the reason Ronald Coase got a Nobel prize for it.

—snip—



-----------------------------------------------------------------
PhD-Design mailing list  <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------