Elite constitution formation and public choice theory

A short, superficial post on what seems like a mild irony in the intellectual history of public choice theory – specifically, concerning the theory of constitution formation.

In 1913 Charles Beard published ‘An economic interpretation of the constitution of the United States’ – a book which I haven’t read, though I probably should.  In that work, Beard (I gather) argues that the US constitution was centrally shaped by the class interests of its drafters.  Obviously I’m in no position to have a view on Beard’s historical claims.  Nevertheless, the very broad category of claim it makes – i.e. that class interests can and do play a major role in constitution formation – is in my view clearly correct.  And this category of claim was a target of early public choice theory, in particular Buchanan and Tullock’s ‘The Calculus of Consent’.  In that work, Buchanan and Tullock discuss the Beard thesis in two locations.  On page 14 they make the startling claim that “a nation of small freeholders, perhaps roughly similar to the United States of 1787” would fit well their assumed theoretical model in which most of the inhabitants of the model society “are, in fact, essentially equivalent in all external characteristics”.  Buchanan and Tullock footnote this remark with a citation to Robert E. Brown’s ‘Charles Beard and the Constitution’, which, they claim, “establishes the fact that economic differences, at least in terms of class, were not important in 1787.”  (This seems like a difficult claim to support as stated, to put it mildly.)

Then on pages 25-27 Buchanan and Tullock discuss the Beard thesis at greater length.  Here they are concerned to distinguish their own form of economic explanation – which takes utility-maximising individuals as its fundamental unit of analysis – from Beard’s form of economic explanation – which, they argue, takes classes and class interests as its fundamental unit of analysis.  Buchanan and Tullock are keen to emphasise that these two analytic approaches – both of which can be called ‘economic’ – are radically divergent.  Moreover, their core argument against Beard is methodological. For Buchanan and Tullock, it is simply misguided for social scientists to use class as a central category: doing so obscures the individual interests that are essential to the dynamics of political-economic life.

This methodological argument for rejecting class analysis seems to me to be mistaken.  I am, for what it’s worth, sympathetic to methodological individualism – I think there are indeed important features of political-economic life that can only be understood if we drill down to individual-level motivation.  Similarly, I don’t have any specific objection to rational choice theory, which is in my view a capacious and fruitful analytic framework.  But it seems to me that Buchanan and Tullock are here attempting to make a substantive argument on methodological grounds, where the substantive conclusions simply can’t be derived from the methodological premises.  Even if we begin with the individual as our unit of analysis, there is no reason to believe that individual preferences and actions cannot generate macro-level phenomena that are well-captured using the vocabulary of class.  Indeed, it seems obvious (to me) that they can and do.  Moreover, assuming we grant that class phenomena are compatible with methodological individualism, we should not have to start from scratch with individual-level analysis every time we wish to speak about those macro-level phenomena – the intellectual division of labour makes possible a methodological pluralism which we can all embrace, without having to rederive any given analytic framework in every instance of its application.

In any case, Buchanan and Tullock raise two objections to the Beard thesis, one empirical and one methodological: empirically, they argue that it’s simply wrong as history; methodologically, they argue that it should be rejected because its class categories are not grounded in methodological individualism.  

But there is a third, broader argument about elite constitution-formation, which in some sense animates the entire project of ‘The Calculus of Consent’.  This argument is that the classical liberal institutional ideal of small government bound by substantial constitutional constraints was and is good.  Where Beard and his allies seek to present the early US constitutional settlement as an instrument of class power and class domination, Buchanan and Tullock seek to present – at least an idealised version of – this classical liberal institutionalist project as broadly normatively desirable.  Beard is, for Buchanan and Tullock, not only wrong empirically and methodologically, he is also wrong normatively or politically – and I think it is reasonable to see this argument as the core motivation of the dispute around constitution-formation, even if this dimension of the debate is at times somewhat tacit.

Now, there is an irony here – in that there are two elements of the public choice project operative in this discussion, which stand in some tension.  On the one hand, there is the element of public choice that focuses on institutional capture, rent-seeking, ‘politics without romance’, the analysis of political actors as self-interested seekers of advantage, etc.  This element of public choice has as its target the administrative state and the redistributive state – the elements of state apparatus that, public choice theorists argue, seek to extend state power in the service of special interests and the tyranny of the majority.  On the other hand, there is the element of public choice that focuses on constitution-formation, where a well-configured classical liberal constitution can, in principle, serve as a constraint on these unfortunate elements of political-economic self-interest.  These two elements of the public choice project have always gone together – but the relation between them also raises a question: why should we not see the process of constitution-formation as itself determined by the kinds of institutional capture, self-interested motives, rent-seeking, etc. etc., that characterise the administrative and redistributive states?  Is it not possible that part of the threat of the Beard thesis and similar arguments to the public choice project lies in the risk of a symmetrical application of the idea of ‘politics without romance’ to the process of constitution-formation itself?  And might this be part of the reason why Buchanan and Tullock’s objections to the Beard thesis are centred on empirical and methodological issues, rather than the normative and political issues that are at least as central to the debate?

In this connection, I was interested to read some of Randall Holcombe’s recent work on ‘political capitalism’.  Holcombe is working within the public choice and Austrian political-economic traditions, but unlike Buchanan and Tullock, Holcombe does not take methodological individualism to be an insurmountable barrier to class analysis.  Rather, Holcombe is part of what I would characterise as a turn within recent conservative discourse to place class categories closer to the centre of its analysis and rhetoric.  (Another example of this trend within Austrian economics is Peter Boettke’s recent ‘The Struggle for a Better World’, which aims to direct the political energy of critiques of “the 1%” towards classical liberalism rather than socialism.)

Holcombe argues that good liberal individualist institutionalist analysis can give us all the resources we need to use categories like ‘economic elites’.  Specifically, he argues that Coasean arguments around the transaction costs associated with establishing coalitions of sufficient economic power to wield political influence will tend to favour large corporate interests over the interests of ‘the masses’.  It is hard to build a coalition of (even a significant proportion of) ‘the 99%’ with the goal of wielding political influence.  On the other hand, it is relatively easy to build a coalition of a small number of corporate decision-makers to form a cartel or lobbying group.  We would therefore expect, Holcombe argues, for a small number of people each with a large amount of economic power to wield more political influence than a large number of people each with a small amount of economic power.  This observation (as Holcombe notes) is of course a commonplace of critical left literature – as is the conclusion that the state under the influence of such interest groups can be understood as “a committee for managing the common affairs of the whole bourgeoisie.” Holcombe’s argument is that this analysis has not received enough attention in public choice and Austrian institutional political economy.  That tradition has typically understood the largest-scale problem of political-economic institution design in terms of the conflict between state and market societies – the left wishes to expand the role of the state; the right wants to restrict the role of the state.  Holcombe’s ‘political capitalism’ thesis presents the possibility: what if there is an equilibrium state of capitalist society in which the state is, rather, the organ of corporate interests, and this constitutes its own form of political-economic organisation? This is, as it were, ‘regulatory capture’ as a mode of production.

There are, of course, a number of ways in which one could object to Holcombe’s analysis. But to keep within this framework, Holcombe’s normative conclusion is the one you might expect from a scholar working within the Austrian tradition. Left and right may agree about the problem of interlocking state and corporate power, but for the socialist left (broadly speaking) the solution is the expansion of state power; for the libertarian right the solution is the paring back of state power. (There are, of course, a number of important ways in which the ideological terrain doesn’t line up with this ‘spectrum’, some of which I’ve discussed on the blog before – but still.) For Holcombe – coming from a libertarian, Austrian perspective – we need a pared back, minimalist, classically liberal state which can hold the line against the abuse of state powers in the service of both socialism and ‘political capitalism’. 

But if we have rejected the methodological argument against class analysis – as we must, if we are to adopt ‘political capitalism’ as an analytic category – have we not also rejected the basis on which the public choice tradition – at least in one of its foundational works – dismisses the idea that class interests can also be in play in the process of constitution-formation?  Is not the next logical step to adopt a Coasean, public choice version of ‘the Beard framework’ – to consider how the balance of political power is at work in the formation of the core institutions of classical liberalism itself? And if we do this, doesn’t this raise the question of the extent to which capitalism was ever anything other than ‘political capitalism’, even in its ‘classical liberal’ variants? I think questions like these present a challenge for the public choice tradition – but also present an alternative public choice research programme of sorts: the analysis of classical liberal institutions ‘without romance’.