The Corporate Hierarchy Dies, Again

22 August 2010 at 11:14 pm Peter G. Klein 16 comments

| Peter Klein |

Ronald Coase described his 1937 paper on the firm as “much cited, but little used.” He was referring to the academic literature, but these days it seems to apply to the popular press as well. Almost every week brings a new article on the death of the corporate hierarchy: you know, firms only exist to deal with transaction costs, and the Internet has reduced them to almost zero, so who needs firms? This argument shows up again and again. But it’s wrong. Of course there are transaction costs between firms (search, bargaining, enforcement). But there are also transaction costs inside firms (agency and information costs, the Misesian calculation problem). The firm straddles these margins. Both sets of transaction costs matter, and both can be reduced through technological change. Coase was not as clear on this point as he could have been, but Williamson has been explaining it for decades, in terms of “comparative contracting costs.” You have to compare both sets of costs, not just look at one. Why is it so hard to see?

Saturday’s WSJ gives us the latest version of the bogus argument, this time from Alan Murray. Same old story: Internet, transaction costs, Tapscott and Williams, wikipedia, yada yada yada. “Mr. Coase received his Nobel Prize in 1991 — the very dawn of the Internet age. Since then, the ability of human beings on different continents and with vastly different skills and interests to work together and coordinate complex tasks has taken quantum leaps. Complicated enterprises, like maintaining Wikipedia or building a Linux operating system, now can be accomplished with little or no corporate management structure at all.” Yawn. “[T]he trends here are big and undeniable. Change is rapidly accelerating. Transaction costs are rapidly diminishing. And as a result, everything we learned in the last century about managing large corporations is in need of a serious rethink.” Zzzzzzzzzzzzzzz. Mr. Murray, please read The Victorian Internet three times fast and have a report on my desk first thing in the morning. “The new model will have to be more like the marketplace, and less like corporations of the past. It will need to be flexible, agile, able to quickly adjust to market developments, and ruthless in reallocating resources to new opportunities.” Right, no corporations of the past ever tried to do this.

Entry filed under: - Klein -, Myths and Realities, Strategic Management, Theory of the Firm.

Note on Anonymous Comments Austrian Economics PhD Course

16 Comments Add your own

1. Gregory Rader | 23 August 2010 at 12:25 am

Peter, I am curious…are you arguing just that this isn’t any sort of conceptual revolution or are you also arguing that Murray’s article mis-characterizes trends in corporate management (regardless of conceptual justification).

It seems to me that the trends are real even if the theoretical justification is wrong. I think I would want to say anecdotally that though both sets of transaction costs matter, incumbent firms have not adopted new practices as readily as have new entrant firms. As such the larger incumbent firms are not capitalizing on the potential for internal reductions in transaction costs and the new entrant firms, being generally smaller, are concerned less with internal costs than external costs. In other words the technology for reducing internal costs is maturing more slowly…that is just me thinking out loud though.
2. Peter Klein | 23 August 2010 at 12:50 am

Gregory, yes on both counts. There are lots of interesting anecdotes about smaller firms and flatter hierarchies and network production and the like, but very little systematic empirical evidence. I just don’t see the trends in the data. And there are many counterexamples of IT leading to larger, more integrated firms — Walmart being the most obvious example.

Your thesis sounds reasonable to me. Let’s take it to the data!
3. Marcus Linder | 23 August 2010 at 2:37 am

Regarding the role of hierarchies versus markets in a world with minimal transaction costs, I think Nickerson & Zenger has some interesting thoughts in their 2004 Org. science paper about the “problem-solving perspective”.

Their focus is on the economics and strategy of knowledge sharing and creation, and how knowledge owners can deal with the problem of appropriating from valuable knowledge. (The problem being that it is difficult for the buyer to estimate the value of knowledge before it is shared, but once it is shared it is difficult for the seller to appropriate from it.)

A common solution is to productize the knowledge – to sell applications of the knowledge, packaged and ready for use. Of course, decreased transaction costs will help that process along.

But the authors also make an interesting argument for why this solution wont be suitable for a large set of valuable problems. In particular, so called non-decomposable problems – problems with solutions corresponding to the optimization of ‘complex’ systems. Because the systems to be optimized are complex, you can’t solve the problem by optimizing one component at a time as that would lead to a sub-optimization. Consequently, you need to deal with the problem of knowledge sharing in a different manner. Thus, the existence of various forms of hierarchies – firms etc.

It seems unlikely that improved ICT will remove the basic problem of opportunistic behavior in settings requiring intense knowledge sharing. If so, ICT won’t change the need for firms for solving at least the some types of problems.
4. srp | 23 August 2010 at 4:09 am

Also, practices inside corporations have changed since, say, the 1970s. Strategic planning has been pushed down more to the SBU level, much more white-collar work is project based, lean manufacturing methods have transformed inventory management, and so on. Combined with his total lack of attention to opportunism and conflicts of interest (do you really want to have brand names shared across firms, for instance) Murray manages to give superficiality a bad name.
5. pj | 23 August 2010 at 8:53 am

It is true that as transaction costs approach zero the corporate hierarchy will die … and your co-blogger Richard Langlois has done some work confirming an evolution in that direction (http://web2.uconn.edu/langlois/Vanishing.html).

So maybe you should call the argument “hyperbolic” instead of “bogus.”
6. Michael Webster | 23 August 2010 at 10:21 am

I read the article, but not the book you have cited. But it seems to me that the major benefit of the internet has been the rapid ability to form groups which coordinate very quickly and complete transactions which would have taken months or years prior to the internet.

(The network effects that appear to be described in the Victorian Internet do not seem to be of this type.)
7. Peter Klein | 23 August 2010 at 10:55 am

pj, who is this Langlois guy you’re talking about?!

Dick and I agree that the phenomena he describes in Vanishing Hand are real, but we (may) disagree about the magnitude and duration. See his recent exchange with Sako and Helper:

More on Managerial Coordination and the Vanishing Hand
8. pj | 23 August 2010 at 2:53 pm

Peter, I think there are some places you can look for evidence. I am working on a history of the evolution of social structure which will provide instances, should be ready next year.

But just as a theoretical matter, the firm as a hierarchical structure is incompatible with the highly sophisticated interactions that occur under zero transaction costs for the same reason a socialist hierarchical organizaiton of all society is incompatible with Misesian economic calculation during contemporary (finite) transaction costs. So as transaction costs decline, it’s true that both firm and market transaction costs decline, but the equilibrium shifts toward a greater reliance on market/network structures rather than hierarchical structures.

To give one data point that’s consistent with the evolution story, following upon the telecommunications and Internet revolutions there’s been a huge rise in international trade, and a larger fraction of that trade is inter-firm than the more slowly rising domestic trade.

Of course zero transaction costs is an eschatological concept, and changes on decade time scales could be driven primarily by debt and business and demographic cycles. So there’s room to doubt how much the margin has been shifted by recent technological advances. But I think when you delve into the evidence there’s sufficient ground for saying a real evolution has occurred.
9. Peter Klein | 23 August 2010 at 2:59 pm

PJ, I take your meaning, but am still not convinced. We have to be careful, first, to define “transaction costs” precisely. If we mean “anything that impedes economic relationships,” then under zero transaction costs, organizational form is indeterminate — “market” and “hierarchy” are equivalent. If you mean the usual sorts of search, negotiation, motivation, monitoring, and calculation costs, then I still don’t think the answer is as clear ex ante as you imply. Check out Nicolai’s paper, “Misesian ownership and coasian authority in hayekian settings: The case of the knowledge economy” (http://www.springerlink.com/content/g3v10720741420gr/). Even under the environmental conditions you describe, authority sometimes outperforms market/network modes of organization.
10. pj | 23 August 2010 at 3:35 pm

I partially agree. I do think that the “usual sorts of search, negotiation, motivation, monitoring, and calculation costs” are only a small part of the important transaction costs, and that there is a need to reconceptualize transaction costs in a way that makes the natural evolution in the economy more easy to describe. Under zero transaction costs, everyone would have a relationship with every other living person, and each relationship would be customized with its own optimized terms. I think that in conventional understandings of “market” and “hierarchy” this zero transaction cost economy better fits our mental maps of “market” than of “hierarchy.” However, I am not fond of the market/hierarchy/network lingo and don’t use it in my own work.

But these issues are too big for a comment section. Thanks for the link to Nicolai’s paper.
11. Peter Klein | 23 August 2010 at 3:36 pm

Thanks for the excellent comments!
12. Gregory Rader | 24 August 2010 at 4:32 pm

Peter,
I wonder if we aren’t conflating size of the firm with the extensiveness of the bureaucracy…and I understand the article conflates these two also. Walmart is an irrefutable example of the scope of the firm increasing but I would guess that it is reductions in bureaucracy that allows that expansion of scope. As an example, Walmart recently initiated a supply deal with Li & Fung, which is noted for its expansive scope and relative lack of bureaucracy and hierarchy. (http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aSDzLcEyoNas)

To your comment to PJ: “If we mean “anything that impedes economic relationships,” then under zero transaction costs, organizational form is indeterminate — “market” and “hierarchy” are equivalent.”

Isn’t that what we are seeing to some degree? As transactions cost decrease we see both spontaneous organization and collaboration in the “market” and corporations (the successful ones anyway) acting more like collaborative networks/markets than rigid bureaucratic institutions.
13. Peter Klein | 24 August 2010 at 10:38 pm

Gregory, I guess I want to be careful in how we define hierarchy. There certainly is evidence, mostly anecdotal, about firms becoming “flatter,” which could be consistent with some kind of IT improvements. I take it you’re suggesting that reductions in external transaction costs (due to IT) either a) make firms smaller or b) make them flatter (though possibly larger). But here things can get really confusing if we’re not careful! Walmart, Toyota, and other firms that rely on various “lean” manufacturing and distribution practices are still examples of managerial control, compared to loose networks of independent firms, and it’s not clear how to reconcile that with the standard IT-reduces-TCs story. But your’e raising some really interesting points about organizational structure.

Again, there are many exceptions, e.g. work by Hubbard and Nickerson and others on long-haul trucking, where mobile telephony and GPS were used to increase managerial oversight of driver behavior. One could imagine these kinds of technological changing that lead to more company-owned trucks driven by employees and fewer driver-owned trucks.
14. Lesetipps für den 27. August | Netzpiloten.de - das Beste aus Blogs, Videos, Musik und Web 2.0 | 27 August 2010 at 2:17 am

[…] 2.0 The Corporate Hierarchy Dies, Again: Ronald Coase described his 1937 paper on the firm as “much cited, but little used.” He […]
15. Gregory Rader | 29 August 2010 at 9:10 pm

Thanks for the response Peter. I can see that what you mean about the semantics getting very fuzzy. It seems we have a number of notions that would need to be defined and distinguished (hierarchy, bureaucracy, managerial control, overall scale, etc) and that in any given case the direction of change from a given reduction in transaction costs might vary across each value. For example, the changes you note in the trucking industry might lead to more managerial control but less bureaucracy insofar as that managerial control is applied less ambiguously, arbitrarily, or is less mediated by human managers.
16. Tom | 3 September 2010 at 3:41 pm

The Victorian Internet is a complete misnomer for the telegraph. The internet and the telegraph are wildly different forms of networks. Their effects on society are wildly different. Their uses by firms are wildly different. Alex Field has some good work about which industries the telegraph had a large influence. To point to the telegraph and say that the internet is just a more sophisticated iteration of an old technology is incorrect. That being said, I agree completely that Murray’s article was way off base.