Organizational sociology after Nikefication
[Ed note: This is the sixth of 14 posts in a virtual panel on The Future of Organizational Sociology.]
Organizational sociology may have reached its high water mark 25 years ago, when Chick Perrow penned “A society of organizations.” Perrow argued that organizations had absorbed society, which implied that organizational sociology was now the master key for making sense of society. He stated, “I argue that the appearance of large organizations in the United States makes organizations the key phenomenon of our time, and thus politics, social class, economics, technology, religion, the family, and even social psychology take on the character of dependent variables.” Stratification happened through organizational practices of hiring and promotion. Work went on inside organizations, structured by organizational rules. Social movements increasingly constituted themselves as formal organizations. In a society of organizations, organizational sociology should be the sun around which the other subfields in sociology orbit. Instead, organizational scholars are scarce on the ground in most departments today, as if the Rapture had come and left behind only the demographers and criminologists.
Many or most of the disappeared wound up in business schools. It’s not hard to see why: the money is better, and the jobs were more plentiful. Yet it would be a mistake to imagine that b-schools are crammed full of organizational sociologists, at least in North America. While fancy schools like Stanford, Northwestern, and MIT are strong outposts for organization theory, most schools are not. Hiring is typically driven by teaching needs, and there is surprisingly little demand among MBA students for courses on organization design (much less institutional logics or categorization). Most organization theorists in business schools wind up teaching strategy and, if they want to get tenure, publishing work that can pass for strategy. All of this bodes ill for organization theory, wherever it is done.
How did things go so wrong? The short answer is that the “society of organizations” described by Perrow has de-composed back into a primordial soup. The ever-expanding, enveloping organizations of the post-War era have reversed course; now, they want to be as “focused” and streamlined as possible, with as few assets and employees as is feasible. Every week brings news of corporations either splitting up into constituent elements (Hewlett-Packard, Symantec, GE, Time-Warner, Procter & Gamble, Abbott Labs, Sony), going private or bankrupt to radically restructure (Dell, GM), or evaporating entirely (Lehman, Countrywide, arguably Kodak). The number of public corporations in the US has dropped by 55% since 1997, and those that remain are generally a lot smaller than they used to be: GM has as many employees today as it did in 1928. Organizations are increasingly hollow shells.
I have argued elsewhere that this shift is largely due to low-cost information and communication technologies that make it possible to rent (rather than buy) components of a supply chain and scale up or down as needed without bearing the cost of maintaining an ongoing organization. (Call it “Nikefication.”) It’s expensive to be Sony, with 150,000 employees. It’s cheap to be Vizio, with 200 employees and flat-screen TVs that are much less expensive and more popular than Sony’s. It’s expensive to be Blockbuster, with 83,000 employees and 9000 stores. It’s relatively cheap to be Netflix, with 2000 employees and server capacity rented from Amazon. (Amazon, in turn, rents temporary employees in bulk from staffing services. The fim plans to temporarily expand by 80,000 workers just for the holiday season this year.) The “society of organizations” increasingly resembles Disney’s Main Street USA, a series of facades with little substance behind them.
Perhaps the reason there are fewer organization-level questions of interest is an ontological one: organizations qua organizations have largely evaporated. Going forward, we are likely to see more pop-ups (like Vizio) and “platforms” (like Uber) and fewer countable, bounded, goal-oriented, hierarchical, employing organizations. The six characteristics of bureaucracy outlined by Weber in Chapter 11 of Economy and Society have little obvious application in this world. (As a reminder, they were: rule-governed jurisdictional areas; hierarchically-organized offices; official written files; specialized training for office-holders; office-holding as a full-time job; and management according to stable, general rules.) Enterprise resource planning software enables “flat” hierarchies by automating employee schedules and performance evaluations. Retail is en route to being even more Taylorized than manufacturing was, with human management replaced by algorithms. Those forced into the TaskRabbit economy (the “precariat”) have an even more marginal connection to regular employment than those in retail. The new book by Dan Clawson and Naomi Gerstel, Unequal Time, highlights some of the difficulties imposed on the working class by unpredictable scheduling and uncertain employment.
Venerating the elders, as sociologists love to do, may be a path to irrelevance. Aiming to document the timeless truths of Weber’s writings is unlikely to lead to a deep understanding of the contemporary organizational landscape and the new world of precarious employment. We may be weighed down by a devotion to the past. (Sociologists also don’t help themselves by endless professional mitosis. 52 sections, srsly? When is Economic Sociology going to spawn Economic Astro-Sociology as a new section?)
Where to next? When Dick Scott and his collaborator updated the classic text in organizational sociology in 2007, the new name was Organizations and Organizing to highlight the verb over the noun. And although network analysis provides some tools for thinking about organizing, theory has not caught up yet with our provisional world. Sociology is in a unique position to contribute to the study of organizing. Mustafa Emirbayer has called for a relational sociology, and there have been various efforts at field theories. These point in the right direction. Perhaps abandoning the countless divisions within sociology among those who study organizations, collective action and social movements, stratification, economic sociology, and other related domains would be a step forward toward understanding how coordinated human action happens today.
Jerry Davis is Wilbur K. Pierpont Collegiate Professor of Management, Professor of Sociology and Co-Director, ICOS (Interdisciplinary Committee on Organization Studies) at the University of Michigan.
Reblogged this on Notas desde la Orilla del Camino and commented:
Really interesting point about sociology of organizations.
” Most organization theorists in business schools wind up teaching strategy and, if they want to get tenure, publishing work that can pass for strategy. All of this bodes ill for organization theory, wherever it is done.” Simply brutal!!!
Jerry, recognition among organizational sociologists of the developments Chick described had happened long before 1991. In 1979, an author wrote “Organizations — producing goods, delivering services, maintaining order, challenging the established order — are the fundamental building blocks of modern societies, and must stand at the center of the analysis of social change.Focusing on populations rather than isolated organizations forces us to come to terms with the complexity of industrial societies, directing our attention to what goes on outside as well as within organizations.” Organizations and Environments, 1991, p. 350.
I agree with you that business schools are the key to understanding where organizational sociology went off the tracks, but I think it is because, until recently, they’ve been totally focused on turning out MBAs who’ll work for a Fortune 500 firm or another big company. B-school professors studied BIG firms, consulted for BIG firms, and didn’t stop to ask what was actually happening re firm dynamics in the economy. As you’re fond of pointing out, very few people in b-schools to this day realize that the # of publicly held firms in the USA has dropped by about half in the past few decades. Why did they not see that?
If “entrepreneurship” as a field of study had achieved legitimacy sooner (but there are complex reasons for that, as we’ve discussed), maybe a few more organizational sociologists might have seen the light. The strategy people were quicker off the mark, leaving OT/OB people in the dust. Hats off to the b-schools that have departments of “strategy and entrepreneurship.”
I meant to write “Organizations and Environments, 1979.”
Howard, some might argue that Presthus predates everybody with “The organizational society” (1962), which says some of the same stuff about organizations being essential building blocks of society. But I was citing Perrow for a claim about sociology, not about society: the other sub-fields (e.g., social psychology) were now merely subsidiaries of organizational sociology because we could explain their dependent variables. At least that’s the way it looked in 1991.
Entrepreneurship is certainly an apt topic of study, and it’s unfortunate that it often gets absorbed within the field of strategy (which tends not to be very sociological). But I would advocate for something more than just orienting more scholars toward entrepreneurship (and would hate to see yet another new ASA section). Let’s merge some of those 52 sections of ASA back into a single uber-section that looks at origins, forms, and changes in systems of coordinated action (organizations, pop-ups, flashmobs, markets, social movements).
The various trends you talk about are definitely very interesting, and I imagine there are more emerging trends you might have included (the ‘sharing economy’, crowdfunding, etc.). But it’s not clear to me that these trends are of a piece (e.g., i’m not sure what the decline in the # of public companies has to do with the others; I mean, Michael Jensen predicted this in 1987, albeit for the wrong reasons), I don’t see how you end up concluding that “organizations qua organizations” have largely evaporated. Even in industries where there is disintegration of the supply chain, that actually creates more “countable” organizations, not less. (Whether this means that organizations are still the key variable on the right hand side of every equation, I really have no idea because I don’t know what that means.) Is this a testable hypothesis? Have you tested it?
(And even if it’s true, I’m not sure what it has to do with demand for organizational sociology or organization theory. I’m skeptical there has been much change, and if there has, that it has anything to do with the prevalence or size of organizations out there in the world)
Thanks for the comments. The various trends I describe are of a piece because the transaction costs of organizing have declined drastically and made it much cheaper to create short-term enterprises that can be large in scale but relatively temporary in outlook (e.g., Vizio, Flip, etc.), and these don’t look much like the organizations that organizational sociologists theorize about. Perrow’s essay contemplated a world of integrated enterprises with long time horizons and long-term employment. These enterprises (e.g., GM, GE, AT&T, Sears) were almost inevitably public corporations because they required long-lived, large-scale capital investments, and their employment practices favored long attachments, which is why they were seen as “enveloping” in their effects.
You say you’re skeptical there has been much change. But some things are unambiguous: the largest US employers for most of the post-War era were manufacturers with long-term employment attachments; since 2000 or so, they have almost all been in retail, where wages are low and attachments short. (Walmart allegedly has annual turnover over 50%.) It’s hard to be “enveloped” in a job that, on average, lasts just a few months. Jensen’s prediction in 1989 about the eclipse of the public corporation was based on the idea that there was an irreconcilable conflict between owners and managers, but during the next 8 years, the number of listed corporations in the US increased by 50%. (This seems like a pretty good case for falsification.) The substantial decline since 1997 is not due to agency costs, but because in many cases public corporations are not worth the cost: there are cheaper and more flexible alternative formats.
What does this have to do with organizational sociology? An organizational sociology that’s really good at making sense of GM and AT&T may not be well-suited to an Amazon staffed by temps, or a Twitter staffed by almost nobody. Moreover, the factors shaping the kinds of enterprises we have today are the same ones that shape other forms of coordinated action, from social movements to the actions of states. Maybe it makes sense to re-combine some of the endlessly dividing sections of ASA to make common theoretical cause.
Let’s say I grant you that capital intensity is decreasing, thus reducing the need to raise capital on public markets. I still am not sure how this leads to your claim that there are fewer “countable” formal organizations today. That is the claim I am most skeptical about. As I see it, there are three problems with it:
(a) There is no evidence for it, at least as far as I know (and how do we measure it—relative to the size of economic activity maybe?);
(b) If the size of the average organization is decreasing, this would– in the first instance- *increase* the number of [smaller] organizations rather than reduce it ;
(c) From what i can see, the environmental factors that privilege formal organizations– i.e., the demand for reliability and accountability (see Hannan & Freeman 84; cf., Kreps 1990)– are still very salient. The best paper on this that I know of is the Mollick & Kuppuswamy piece I linked to in my piece in this forum. (I can’t seem to figure out how to link to it, so here it is again: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2377020).
There is another way to put this point (as Bob Freeland and I do in a working paper; see here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2523245)– given the various literatures that have predicted the demise of firm hierarchy over the past 30 years or so [for different reasons than you do; they usually emphasized the problems hierarchies have at eliciting the kind of high-commitment orientation necessary in the knowledge economy) , it is amazing how little has changed– how much of our “knowledge economy” is dominated by (intrinsically hierarchical) firms.
P.S. To clarify—I didn’t say I was skeptical that any change had occurred. Rather, I said I was skeptical they the various changes you noted were of a piece; and more to the point, there had been much change in demand for org soc or org theory. And finally, of course I agree re Jensen. As I said, his prediction was right [in the long run] but not for the reasons he gave.
It seems that we are stuck on the word “countable.” Here was my original sentence: “Going forward, we are likely to see more pop-ups (like Vizio) and ‘platforms’ (like Uber) and fewer countable, bounded, goal-oriented, hierarchical, employing organizations.” Countable was only the first of several features of what we think of as formal organizations. It is possible to be countable without being bounded, goal-oriented, hierarchical, or employing. If you have a credit card, you can incorporate online right now here: http://www.liscr.com/liscr/. Depending on your credit limit, you might be able to incorporate dozens of new entities, and count them (and run regressions on them). But that’s not what Weber had in mind when he described bureaucracy, and that’s not what Perrow had in mind when he wrote about the society of organizations.
I allude to Uber not merely because it’s trendy, but because its model (and variants arising in other areas such as AirBnB) explicitly aims to avoid being bounded, hierarchical, and employing. They are doing their utmost to convey the impression that Uber drivers are independent entrepreneurs, and that Uber does not control their actions to a degree that counts as “employment.” Uber wants to be a platform more than a formal organization. If you want to claim that each driver should be regarded as a separate, “countable” organization, feel free, but that would be crazy. Amazon’s all-temp practices also seek to avoid being regarded as an “employer.” The people staffing their warehouses are outside the boundaries of Amazon, by design, and the term of their engagement is typically brief. This applies a fortiori to TaskRabbit, Mturk, Odesk, and so on.
The Mollick & Kuppuswamy piece you cite is informatively titled “Why firms are Potemkin Villages” (cf. my claim about Main Street USA). The survey of Kickstarter projects they drew on defined a formal organization as “an incorporated company, partnership, or nonprofit.” Of course, the whole point of a Potemkin Village is that a façade stands in for the real thing, and it is utterly trivial to create a “formal organization” in the sense they define it. In Michigan, creating an LLC costs $25 and has almost no obligations, but provides the benefits of limited liability. It’s also easy to rent a fake office in Silicon Valley, print up business cards, set up a Square account to accept credit cards, and so on. You can do all these things without being bounded, hierarchical, or employing anybody. (Goal-oriented would take another post to unpack.) Finding that Kickstarter funders prefer “formal” organizations in the minimalist sense that they measure it is not that informative about the prevalence of formal organizations in the sense that organization theorists mean it.
I skimmed your paper with Bob. At many points you refer to corporate law, but as you surely know, LLCs and other non-corporations are en route to replacing the corporation in prevalence in the US (see Ribstein’s book The Rise of the Uncorporation: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1463684). You say firms (by which you mean corporations) are “inherently hierarchical” and then propose to analyze “why this hierarchical form dominates the capitalist economy” (p. 14). First, you use “intrinsically hierarchical” to mean that the legal personality of the firm retains ultimate control rights, which is a fairly specialized sense of “hierarchical.” (Readers might mistakenly think you mean that firms must always have a pyramid-shaped system of authority, with bosses and so on.) Second, what is “the capitalist economy”? Do you mean the US, or the Netherlands (with about 100 listed corporations), or Vietnam (with 2), or every economy in the world except North Korea? And what does it mean to say that they “dominate” the economy? The biggest corporate employers in the US are a lot smaller than they used to be (other than Walmart), and the duration of employment in the biggest firms is a lot shorter. See http://webuser.bus.umich.edu/gfdavis/davis_09_AMP.pdf for facts and figures. And the number of public corporations keeps dropping as delistings, liquidations, and going-private transactions outstrip IPOs. (Firms that go private and adopt a non-corporate form may no longer be “intrinsically hierarchical” in the sense that you mean it.) See http://pas.sagepub.com/content/41/2/283 for details. I believe this evidence is consistent with my claim about the declining importance of formal (bounded, employing, etc.) organizations. But to say that the knowledge economy is dominated by intrinsically hierarchical firms seems to require empirical documentation.
I think the key difference between us is that you define firms as corporations, and you are focused on public companies in particular and use them as emblematic of what is going on with “countable, bounded, goal-oriented, hierarchical, employing organizations.” That is not a leap that I am prepared to make.
(Note that while you say that we equate firms with corporations, that is not correct; accordingly, your riff about the number of listed firms in Vietnam etc is beside the point).
Another area where you are lumping and I would be inclined to split pertains to the trends in employment. If we saw a substantial shift from employees to contractors, I would tend to agree that something important is going on—from a world where work is largely coordinated via authority to one where it is largely coordinated by price. Though it would also depend on the terms of those contracts, which can be quite hierarchical (recall Stinchcombe’s classic piece on this; still quite relevant). And my sense is that this is the case for the contracts between Uber and its contractors and Amazon and its contractors, etc, but I’m no expert. But I’m not sure that a decline in the *size* of the largest firms means that hierarchy is any less dominant, and I am especially unsure why employment tenure is relevant. If we are looking for indicators of movement away from the mid-20th century “Detroit consensus” model, then sure—decreased length in tenure would be an indicator of that. But I don’t know why low average tenure is an indicator of a the absence of hierarchy. You could argue the opposite (job security reflects the firm’s making commitments to refrain from using its right to terminate).
Finally, I think it can be misleading to focus on what’s novel and then jump to the conclusion that it is taking over. Odesk and mturk etc. are very interesting (and they have been great for my research!). Are they taking over as *the* model of how work will be organized? Color me skeptical.
I think another difference between us is that from our standpoint, the death of hierarchy was actually proclaimed a long time ago—by org theorists and industrial relations folks who demonstrated persuasively that hierarchical control is alienating and that the highest performing firms tend to limit use of hierarchical controls. So we might actually have expected the death of hierarchy a while back and celebrated it when it arrived! But when we look at the issue from that standpoint, the question becomes why hierarchy has so much staying power rather than why it (purportedly) has declined. So, I guess it’s a matter of perspective whether you think the glass is half full or half empty.
As for Mollick & Kuppuswamy’s piece, your interpretation is a reasonable one. Their evidence suggests that funders are looking for signs of formal organization, but you’re right that in the first instance, the indicators are myth and ceremony. Hmm… this can also be read that what Meyer and Rowan taught a generation ago is still highly relevant!
Finally, I’m not sure what it is that you don’t like about our definition of a (rights) hierarchy, I’d be curious what yours is. Hierarchy is an easy word to throw around, but it is a hard word to define. The interested reader will see that we do not take the issue lightly.
Thanks for the clarifying comments. It seems that we agree on a couple of things, most importantly that our quarry is “the organizational structure of the (US) economy,” that is, how and where economic activities are organized. We both see the employment relation as a central feature of the firm. You argue that inherently hierarchical firms (still) dominate the (knowledge) economy. This is puzzling because hierarchy is alienating, which may be especially bad for knowledge workers. In this sense, things have not changed that much from when Coase initially puzzled over the prevalence of firms rather than market-based transactions (back in the days when firms actually made things). You see Uber, Odesk, etc. as novelties and not necessarily signs of how work will be organized in the future.
I am claiming that there has been a regime shift in the transaction costs of organizing that broadly favor small and dispersed firms and contingent use of labor. The web, enterprise resource planning, mobile telephony, low-cost CNC tools—the usual ICT suspects—are a big part of this; new forms of legal organization are another. Widespread Nikefication is one result: Vizio (with 200 employees) sells more TVs than Sony, X5 (with 45 employees) is the world’s second-largest classical music distributor, and so on. In various studies I have documented that big firms in the US keep disappearing or getting smaller. I’ve also found that new firms going public since 2000 create few jobs, and the jobs they do create tend to be part-time and/or low-wage. (The biggest IPO job creator by far is GameStop, a strip-mall video game retailer with a large workforce of part-time teenage sales clerks: http://webuser.bus.umich.edu/gfdavis/Papers/Davis%20PoW%202013a.pdf). Entire sectors that one might expect to be wellsprings of employment are the opposite: the Computer & Electronics Products industry has shed 750,000 jobs in the US since 2000; the Information Services industry has lots more than one million jobs.
Big corporations keep getting smaller; new firms avoid growing big. Why? Employees are expensive, and corporations have found many ways to avoid hiring them, particularly for full-time permanent jobs. The reason Amazon’s use of 80,000 rented workers for the holidays, or Uber’s “contractor not employee” models are interesting, is not just their novelty, but because they are consistent with this broad trend, and if their models work (as they seem to), they will spread further. ICTs make it a lot cheaper to use contingent, non-employee labor. In a capitalist economy, cheaper formats often win out, and these days the cheaper formats are often employee-free (or at least employee-lite).
The legal form of organization is relevant here: tiny firms can organize as partnerships, LLCs, worker-owned coops, mutuals, etc. If you want to be giant, you probably want to be a public corporation (unless you are the Koch brothers). If you want to be a node in a supply chain, you can be a contractor, a sole proprietorship, an LLC, a Liberian corporation, or whatever. Most likely the least costly format will win.
You say that I am incorrect in stating that your paper with Bob conflates firm and corporation. But when you define the “legal foundations of the firm” early on, you use the standard features of the corporation (legal personality, property rights, and the nature of the employment relation), which you argue are inherently hierarchical, and you seem to draw on legal literature that is specific to the corporate form. Is it your understanding that these apply to LLCs, partnerships, sole proprietorships, and all other “legally recognized economic organizations consisting of two or more people”? Is it the case that these same features apply worldwide (e.g., to German AGs and GMBHs, or to Chinese corporations)? If LLCs do not have these same “inherently hierarchical” features, and LLCs are vastly outpacing corporations in prevalence in the US, would that count as evidence against your argument?
Thanks for the thoughtful response. I agree that there are interesting things going on, and I appreciate your taking the lead in teaching us about them. I think we probably are going to need to agree to disagree about a few things though. One of them is whether dis-integrating a supply chain into a set of hierarchies is of a piece with reducing the size of a hierarchy by substituting contractors for employees. Also, I’d stress again that not all contractors are of a piece either. But insofar as coordination via authority is being replaced by coordination via price, that would certainly be a an important trend to follow and understand. The only other thing I’d add is that I wonder if you’ve read the legal section of our paper carefully because the last paragraph of your post just does not summarize it accurately– nor is it accurate to say (and I hesitate here because Bob is the lead on this section of the paper and knows the law much better than I do) that the employment relation is a “standard feature of the corporation.” On the one hand, the legal personality at the top of the firm hierarchy is often not a corporation (we say explicitly that it could be a sole proprietorship or LLC—really anything that can assume legal personhood); and on the other hand, many corporations are not firms because they do not have employees or direct human agents more generally. But your feedback is useful. Perhaps we have to make these points even clearer.
P.S. Thanks Howard– I certainly have appreciated Jerry’s enagement on this.
Jerry & Ezra: congrats on maintaining this debate at a very high level of scholarship & professionalism! I’m learning a lot & I’ll bet that 00″s (000’s?…but who’s counting) of readers are, as well.
Pingback: Why before How: “Distinctive and indispensable” beats “sophisticated but superfluous” | Work in Progress
Pingback: The end of “organizational sociology” as we know it? | Work in Progress
Pingback: Panel – The future of organizational sociology? | Work in Progress
Pingback: No political economy in the future of organizational sociology? | Work in Progress