Author: Chris Smaje
In 2009, political scientist Elinor Ostrom (1933-2012) won the Nobel Prize for Economics for her analysis of common pool resources (CPRs). A pithy summation of this work has been sardonically labelled ‘Ostrom’s Law’ – “a resource arrangement that works in practice can work in theory”. On the face of it, such a ‘law’ seems blatantly obvious. Yet it’s surprising how often analytical work in the social sciences trips up on this point. In this article, I look briefly at what Ostrom achieved in her work and then draw out some more general lessons about the way social scientists model the human world.
One context for Ostrom’s important book Governing the Commons (1) was a hugely influential article by ecologist Garrett Hardin, ‘The tragedy of the commons’ (2). Hardin’s thesis was that when a natural resource such as a pasture is openly available to all, then each user is incentivised to maximise their own returns from this resource. However, this is ultimately to their individual and collective detriment as this resource rapidly becomes exhausted from overuse. In Hardin’s words, “Ruin is the destination toward which all men rush, each pursuing his own best interest in a society that believes in the freedom of the commons” (1968: 1244).
If this thesis is accepted, it implies the need to curtail the freedom of the commons in one of two possible ways: either breaking it up into smaller private holdings, where the owner is better incentivised to preserve the long-term viability of the resource for their personal long-term benefit; or else, according governance rights, to some larger body which is able to rationally control use of the resource and enforce sanctions against overuse. This then yields the familiar market versus state duality associated respectively with the political right and left.
But the fact is that there are many resources which have been managed by their users, often for long historical periods, with neither state control nor purely private ownership and notably without obvious degradation. These include pastures, as in Hardin’s example, as well as fisheries, irrigation infrastructures, water catchments and even less tangible resources such as intellectual property. A signal feature of Ostrom’s approach was that she examined in detail how these different kinds of commons functioned – or, in some cases, malfunctioned – in their practical workaday details, and then tried to formulate some general principles about commons and their governance. Unlike Hardin, she did not assume that common pool resources led to ruin on a priori theoretical grounds.
I can’t precis Ostrom’s detailed analyses here, but I’d like to draw from them some wider issues about the way social scientists and their audiences move backwards and forwards between the real world and their models of it. Typically, the social scientist posits some kind of simplified model of how relationships in the real world operate – an abstraction from the world based on a theory of some sort about how people interact. Then they collect real-world data to test the model. Analysis of these data may lead them to confirm or reject the validity of the model, or to consider refinements of it. There is usually plenty of room for others to dispute the conclusions they reach, perhaps by questioning the formal logic of the model-reality relationship: Do the variables chosen fully capture the relationships conceived in the model? Do the data collected fully capture the quantities conceived in the model variables, or might they be biased in some way? Does the analysis of the variables admit to other possible interpretations than the one favoured by the investigator? And so on. Alternatively, it may be possible to question the validity of the model in the first place, because social science models about the nature of social life are inevitably recursively related to social life itself – high level products of the very social world we’re trying to analyse and which are therefore contestable.
In the case of Hardin’s tragedy of the commons thesis, the abstraction was that individuals will maximise their personal returns from a resource in the short-run regardless of longer-term consequences unless they are prevented by a restraining force. This is an economic arrangement that confronts the resource-user directly with the consequences of their actions – which in the tragedy of the commons literature has usually been understood as either private property or the controlling hand of the state. The close relationship involved here between the abstract model and commonplace economic and political assumptions about human nature and human society is fairly apparent. The surprising thing is that Hardin didn’t test his model empirically in his original paper, but merely asserted it as a truism. This rather basic methodological failing didn’t prevent the paper from achieving enormous influence – perhaps because of that recursive relationship between scientific modelling and the commonplace socioeconomic narratives of everyday life. Hardin’s thesis seemed to capture in quasi-scientific – and therefore, in our modern culture, putatively objective – terms (it was published in the journal Science, after all) a story that many of us already believed.
The gap between Hardin’s model of the dysfunctional commons and the real-world functional commons that Ostrom investigated can be explored with reference to economic concepts such as transaction costs and informational symmetry, and with sociological concepts such as social status. Thus, Hardin’s model assumes that graziers are incapable of realising that the sum of their individual actions may threaten the livelihood of all and develop protocols of self-government accordingly. In fact, collective agreements of this kind are widespread and usually feature in successful and enduring common pool resource systems, which specify limits around the resource in question – the community that can use it, and how it can be used. Those who infer support from Hardin’s model for private property rights without further empirical investigation generally neglect the question of whether pooled resource use can deliver greater net benefits than individual use (an argument that has raged at least since the pamphleteers of 18th century England debated the enclosure of the common fields, woods and pastures (3). Those, on the other hand, who infer support from his model for state control of resources generally neglect the question of whether the transaction costs involved in appointing state functionaries to police the commons exceed the costs of self-organisation by the community of actual users.
Ostrom’s contribution was to notice that some of the characteristic positions adopted by researchers concerning common pool resources were over-generalisations that amounted to setting a parameter equal to a constant, and that what was required instead was empirical enquiry into these parameters (hence ‘Ostrom’s Law’)…
Ostrom’s work suggests that for exactly these reasons commons can sometimes be an optimal solution to resource use. But only sometimes. The kind of circumstances in which commoning solutions seem to work best are when there’s a relatively small number of people of fairly equivalent social standing who have a long-term interest in using a resource, particularly when that resource has a low value per unit area, or is erratically productive, or hard to intensify, or is hard to exclude people from. In such circumstances, communication between users is easy, lowering the transaction costs that might otherwise favour more purely public or private solutions. The costs of the kind of boundary demarcations that would be necessary in order to parcel up the resource into private units are also inordinately high. And it is relatively cheap and easy for the users themselves to police the commoning agreement, often by relying on little more than the cost to personal reputation and social standing of the infringer, who has long-term relationships with the other users that cannot be jeopardised lightly. In these circumstances, the policing costs incurred by an external (state) agency would be much higher.
Ostrom provides various examples of commons that fail when some of these conditions don’t hold – for example, allocation of fishing rights in a Sri Lankan village where a wealthy individual with wider political connections was able to subvert local control of the resource, water-rights allocations where information and transaction costs are high, and – a very common situation in contemporary societies – schemes where the number of users is too large and too heterogeneous in their interests for effective sanctions against misuse to be easily applied.
Ostrom’s work leads to an important conclusion – one that can be framed as a point of good methodological practice for social statisticians and other social scientists, concerning the political implications of analytical work in the social sciences and as a philosophical point about the nature of enquiry in the social sciences. I’ll conclude with a few comments on each of these three dimensions.
In terms of methodology, I can hardly phrase the issue better than Ostrom herself in the following passage:
‘Among many academics there are strong preferences for tight analytical models that will yield clear predictions. To make a model tractable, theorists must make simplifying assumptions. Many of these assumptions are equivalent to setting a parameter (e.g. the amount of information available to participants, or the extent of communication) equal to a constant (e.g. complete information, or no communication). Because the resulting model appears to be relatively simple, with only a few “moving parts”, it may be considered by some to be general, rather than the special model that it is. Apparent simplicity and generality are not, however, equivalent. Setting a variable equal to a constant usually narrows, rather than broadens, the range of applicability of a model’ (4).
Ostrom’s contribution was to notice that some of the characteristic positions adopted by researchers concerning common pool resources were over-generalisations that amounted to setting a parameter equal to a constant, and that what was required instead was empirical enquiry into these parameters (hence ‘Ostrom’s Law’ mentioned above). Whenever one attempts to create and test a model of the social world it’s worth bearing this insight in mind.
Turning to the political implications of this point, researchers (and research interpreters) tend to have preconceived biases about the constant values they’re inclined to assign to various model parameters. In the examples we’ve been considering, and as Ostrom hints in the passage cited above, enthusiasts for private property rights tend to assume there is no communication between users in common pool resource situations, hence making such situations ‘tragic’. Conversely, enthusiasts for state control tend to assume that the state’s agents will have complete information and will therefore be able to secure optimum resource use costlessly. Neither of these assumptions is necessarily well-grounded, but their plausibility is only revealed by empirical research, not a priori assumption. This point also applies to the large contemporary literature on commons in alternative economic and activist circles. Repelled by the failures of both capitalist economies grounded in private property rights and socialist economies grounded in dirigiste state control, many people have identified the commons as a kind of ‘third way’ for our economic future, and have found their justification for this in Ostrom’s work and her refutation of Hardin’s anti-commons position (5). A theory of human nature is implicit in these positions: the right-wing and left-wing solutions to the ‘tragedy of the commons’ assume that people are inherently selfish, individually or collectively, and therefore require what Hardin, echoing the traditions of social contract theory, famously called “mutual coercion mutually agreed upon” (whether in the form of private property relations or the collective power of the state) in order to overcome the negative consequences of their selfishness. By contrast, contemporary pro-commons writers, often invoking Ostrom, detect human altruism in the spontaneous self-organising nature of commons. But I find it hard to read Ostrom as ‘for’ the commons in any straightforward way. Rather, her work shows when commons can function well and when they can’t, the problems they can overcome and the problems they encounter. She writes,
Institutions are rarely either private or public – “the market” or “the state”. Many successful CPR institutions are rich mixtures of “private-like” and “public-like” institutions defying classification in a sterile dichotomy (6).
I would argue that scenarios for future economic betterment might focus more profitably on drawing out what some of these “rich mixtures” might be, rather than simply lighting on “the commons” as inevitably the best way forward. The theory of human nature that emerges from Ostrom’s analysis seems closer to Hardin’s than the commons-as-altruism theorists: people can be selfish (all enduring CPRs involve sanctions against free-riders), so commons in fact involve mutual coercion, mutually agreed upon. Perhaps what’s most interesting about them is that the mutual coercion is self-organising, contingent to the tasks of the everyday economy, and sustained by the practical consent of actual face-to-face communities, rather than the disinterested operation of more abstract social norms.
Finally, and most broadly, as illuminating as Ostrom’s work is, she offers no escape from the basic philosophical dilemma faced by the social scientist, for indeed no escape is possible. This is the circular or recursive nature of social science enquiry mentioned above. The social scientist is, like everyone else, a social creature who, in a famous phrase coined by anthropologist Clifford Geertz, is “suspended in webs of significance he himself has spun” (7). We can to some extent stand back from these webs and, by making them partially explicit, abstract and test models of social life from them which can help illuminate political decisions and processes. But I’d argue that these models are only second order derivatives of our grounding assumptions, and so, as Geertz puts it, the analysis of culture is “not an experimental science in search of law but an interpretive one in search of meaning” (8). Perhaps that impossibility of law-like generalization in the social sciences is, paradoxically, another implication of the ‘law’ to which Ostrom’s name is attached.
References
(1). Ostrom, E. 1990. Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press.
(2). Hardin, G. 1968. ‘The tragedy of the commons’, Science, 162: 1243-8.
(3). Neeson, J. 1993. Commoners: Common Right, Enclosure and Social Change in England, 1700-1820, Cambridge University Press.
(4). Ostrom, op cit p.184
(5). See, for example, Tarinski, Y. (2016) ‘The commons is the future’ http://www.resilience.org/stories/2016-01-29/the-commons-is-the-future
(6). Ostrom, op cit p.14
(7). Geertz, C. The Interpretation of Culture, p.5
(8). Ibid