Library Header Image
LSE Theses Online London School of Economics web site

Stabilising liberal societies in a world of radical innovation: committed actors, adaptive rules, and the origins of social order

Finighan, Reuben (2023) Stabilising liberal societies in a world of radical innovation: committed actors, adaptive rules, and the origins of social order. PhD thesis, London School of Economics and Political Science.

[img] Text - Submitted Version
Download (3MB)
Identification Number: 10.21953/lse.00004598


Long-standing questions about social order, and about liberal democratic capitalist orders in particular, remain unsettled. They are of renewed importance in our age of crisis and democratic backsliding. Adam Smith addressed two such questions at the founding of political economy: First, what are the forces that sustain all societies, and liberal societies in particular? Second, what combination of market and state makes such societies prosperous and powerful? A third question, addressed by Hayek, Polanyi, and Keynes in their own period of crisis and backsliding, pertains to interactions between the two: how does the combination of market and state affect the stability of liberal democracy? If we are to answer these questions, I argue we need a realistic theory of innovation. Real-world innovation is Schumpeterian: it is uncertain and often radical, so the future may unexpectedly break with the past. Real-world innovation is Baumolian: it is socially ambiguous, and may be productive or extractive. Consequently, the innovations of political and economic entrepreneurs bring the rise, but also the fall, of societies. Given the last two decades, we may be more open to the idea that Fukuyama’s “End of History” never arrives. Our task is to stabilise and optimise cooperation in both politics and the market. “Cooperation” is defined as the alignment of private returns with social returns; it is exemplified by Smith’s “invisible hand”, and is the precondition for growth. The usual formal methods for identifying cooperative equilibria fail in a world of Schumpeterian and Baumolian innovation. Beyond the short-run, there are no lasting Nash equilibria. Game forms are destroyed and remade. The institutional forces that we hope will restore cooperative equilibria are themselves subject to innovative attack. How, in this unstable world, is it possible to sustain cooperation over long periods of time? And how can we model and predict cooperation? This thesis adopts an analytic strategy that makes this problem tractable. I borrow concepts and formal models from evolutionary sociobiology, a field that deals with cooperation under radical and ambiguous innovation. As in Acemoglu and Robinson’s Narrow Corridor, the core concept is the adversarial innovation race (the “Red Queen’s race”). Most important in this thesis is the race between 4 innovating cooperators and defectors. Social order becomes the probabilistic outcome of a dynamic process—of whether cooperator or defector innovations are superior in a given period. Under the right circumstances, outcomes are predictable. All complex social orders, anthropic and biological, combine “commitment” and “rules” (which, in the definitions of this thesis, includes institutions) into a self-sustaining system. Commitments are essential. They are motives that are exogenous to the innovation race; while all else changes, they continue to draw the system towards a cooperative equilibrium. They come in two forms: one is an intrinsic interest in others’ payoffs, and one is an extrinsic dependence on others’ payoffs. However, commitments are impotent, and indeed are destroyed, if there are no rules or institutions that can control defectors—or if committed actors fail to invest sufficiently in adapting rules so that they keep up in the race against defectors. In short, social order depends on (A) commitments (i.e. motives to run the race that are innovation-proof) that (B) are channelled into the adaptation of rules, to run the race against defectors. Accordingly, the outcomes of innovation races are predictable under two circumstances: when (A) there is no source of commitment to group payoffs, or (B) when committed actors perversely disinvest from running the race, so play the “sleeping Hare” of Aesop’s fable. In either case, loss of the race and collapse of cooperation is guaranteed. On the first question raised by Smith, I present an impossibility theorem for any society built from rules—from institutions, incentives, and so on—alone. Both liberal and authoritarian orders rest on commitment. Smith’s Theory of Moral Sentiments is supported: the “very existence” of liberal orders rests on other-regarding preferences (which, I show, is a product of trust). It is the only innovation-proof force available to them. Authoritarian orders can be explained via the ruler’s extrinsic commitments alone, though other-regarding preferences sometimes play an important role. On the second question, every regime of economic regulation is within the innovation race and vulnerable to unanticipated counter-innovations. I show that every regulatory regime can be described as a particular “division of regulatory labour” between institutional actors and market actors. Institutional actors and market actors are essential complements, with distinct comparative advantages. A 5 principal task for the institutional regulator is to structurally simplify complex markets; otherwise, those defectors that have advantages in the innovation race (of which there are many) will predictably exploit both regulator and market actor. Central planners and Hayekian liberals (and libertarians) endorse extreme divisions of labour between regulator and market actor. They are mirror images and fail in predictable ways. Central planners refuse to use market actors, so allocate hyper-complex (and impossible) regulatory tasks to the state. This produces broad inefficiencies and blocks productive innovation. Hayekian liberals refuse to adapt institutions, so allocate hyper-complex (and impossible) tasks to market actors. This produces crises specifically in complex markets—finance, healthcare, insurance, education, and so on—and soaring rents. Its end point is anarchy. Hayekian liberals suppose advance knowledge of the consequences of basic market institutions. But the unforeseeability of innovation, and distributed nature of knowledge, are double-edged swords: markets produce both productive and extractive innovations that the theorist cannot foresee. To block institutional adaptation is to play the sleeping Hare, and guarantees loss of the innovation race. On the third question, central planning and Hayek’s classical liberalism ultimately lead to authoritarianism. In the case of central planning, Hayek’s argument is supported: to attempt the impossible tasks allocated to it, the state must concentrate power, and voters cannot win the political innovation race to control such a state. In the case of Hayekian liberalism, the state cannot run the market innovation race. Market anarchy and crisis erode the commitments on which liberal orders depend, fuelling distrust and parochiality. As Smith observes, “faction” and “fanaticism” are the greatest threats to the liberal order. To use Hayek’s terms, central planning and his own classical liberalism are “fatal conceits”: they suppose access to distributed and future knowledge that no one possesses. They are both “roads to serfdom”: one via excessive control, the other via anarchy. I describe the “middle of the road”, where commitments are channelled into the adaptive, mixed economic strategy advocated by Keynes. As after the Great Depression, this in turn can create economic outcomes that sustain other-regarding commitments. There, the liberal order can make its home.

Item Type: Thesis (PhD)
Additional Information: © 2023 Reuben Finighan
Library of Congress subject classification: H Social Sciences > HB Economic Theory
J Political Science > JA Political science (General)
J Political Science > JC Political theory
Sets: Departments > Government
Supervisor: Soskice, David

Actions (login required)

Record administration - authorised staff only Record administration - authorised staff only


Downloads per month over past year

View more statistics