Bronk’s Romantic Economist
R. Bronk The Romantic Economist Cambridge University Press 2009
This book was brought to my attention by some of Gordon Brown’s advisers, who said that he (Brown) had read it in the summer of 2009 and greatly appreciated it. (Economist as hero?) It had resulted in a greater appreciation of the work of Shackle and Keynes.
My interest is in what it has to say of relevance to uncertainty and rationality. The book is not written in a technical or mathematical language, so I have interpreted it.
Narrow rationalism, as in mainstream economics before 2008, is inadequate. It can only cover what happens when the situation is stable. Such stability is typically upset by some innovation, which may result in a battle to construct the next paradigm and hence creative destruction of the previous certainties. Bronk holds that this aspect is more important than efficiency within a period of stability.
The approach commended is pragmatic, in that one works within the current socially constructed reality but looks out for changes and reacts to them. Practically, one cannot hope to have a complete, consistent, theory or paradigm, so one draws on the widest range of relevant viewpoints, seeking to identify which differences make a difference in the actual situation at hand, and hence to use common-sense reasoning to construct a tractable composite theory.
Anything else is regarded as “yearning for an impossible to achieve unified but systematic vision.”
The book is also studded with lots of insights into economies that might have benefited from being more widely appreciated prior to 2008. It is appreciative of the work of Mill and Keynes, but not how they came to be interpreted. Bronk holds up Shackle and Brian Arthur (from Santa Fe) as providing the best insights.
Bronk sees the romantics as taking their cue from Kant. He highlights the following themes:
- the importance of organic rather than mechanistic metaphors
- the absence of any single scale of value
- the need for a fuller psychology of human motivation
- the key roles played by language, perspective, metaphor and imaginative intuition.
The main lessons from romanticism cited are:
- the importance of national or local, rather than universal, approaches
- the limitations of utilitarianism
- the necessity of self-creation and the creation of preferences at the individual and social level
- the crucial role of imagination as well as reason in ‘reading’ what is going on in our world and in forming expectations or strategies
- the use of organic rather than mechanistic metaphors (“the whole is more than the sum of the parts”).
His recommended methodology is inspired by these ‘romantic’ ideals. He also suggests that an approach to real economies ought to draw more on the social and human sciences. His own thinking along these lines seems to lead him to recommend nation-based approaches.
The above findings and methodology seems not to be specific to economics, but to apply to any endeavour that can’t be confined to a single academic discipline.
As one might expect, the account is based around Kuhn, with Russell and Popper being marginalised.
Bronk states that “Keynes did not theorize extensively about the ultimate cause of economic uncertainty”. This is to overlook Keynes’ Treatise and does not reflect Keynes’ famous ‘Economic Consequences ..‘ or his work with Whitehead or Smuts . Bronk also underrepresents Keynes’ understanding of the limitations of complexity, evolution, rational choice and the extent to which rational choice theory might be fixed rather than discarded completely.
Whereas Bronk rightly criticises mainstream economists for look at Keynes through too narrow a lens, he seems to have considered too narrow a range of Keynes’ work. In fact, if the essence of romanticism is as above, then Keynes, judged as a whole person, would seem to be a romantic.
Shackle is given as the source first to note the drawbacks of equilibrium models, yet Keynes seems more forensic. According to Keynes, equilibria are only ever temporary and depend on people’s expectations: people acting as if things will be stable tends to encourage stability, but does not ensure it.
Russell and Popper have, independently, noted that Plato’s concept of rationality was intended to hood-wink people into believing in stability and so into acting rationally, thus making it easier for their rulers to achieve stability and so sustain their rule. (Which all may prefer to chaos.) From this viewpoint the economic problem of the noughties could be seen as one in which people continued acting rationally while the rulers neglected their oversight role of maintaining stability.
The 1920s theory of stable economic regimes is that they are purely conventional in the sense that multiple regimes are possible. One may get established almost by accident, but once it does it will tend to continue because people expect it to. For any particular regime of actual behaviours (e.g., share process) there may be many stories that could become accepted as explanations. When the regime ends (either through external events or internal inconsistencies) the old story will be seen to be false. A regime will tend to co-emerge with a new story, which may arise ‘naturally’ or which may be being told by some charismatic leader. Typically, the new story will be a variant some existing story, with a minimal modification. This may involve the equivalent of epicycles, or may involve rationalisations. This is a social process.
Bronk regards a holistic view as necessarily being a snapshot of particulars, in distinction to general laws. This usage is quite different from Smuts and Whitehead, who see the Holistic view as encompassing everything relevant, with a structure. As in Holism, it seems vital to distinguish between different levels of model, to aid comprehension. Bronk does not seem to do this.
In the 1920s there was cross-fertilization between quantum physics, evolutionary theory and economics. With this in mind, much of Bronk’s assumptions look doubtful. In quantum mechanics one has complimentary views which can only be reconciled using mathematics, not common-sense. One also has the notorious uncertainty principle. Bronk notes a variant: that the more one focusses on the current situation, the more one is likely to be taken by surprise by new phenomena. But there are more: the more everyone is thinking alike within a complex system and the more the situation appears stable, the more risky things actually are.
In evolution it is hard for minority organisms to establish themselves, just as it is hard for new ideas to get established. In biological evolution a minority may thrive in some relatively isolated niche and then emerge to establish themselves more broadly. This suggests that Kuhn’s view of paradigm change may be reflective of a particular, inappropriate, culture. In any case the nation is no longer the clear unit of thought.
Keynes showed how it is necessary to distinguish between the short-run and the long-run. The short run is an extrapolation of what has been happening, until things change. Beyond that is the long-run. Thus in an economy the short-run is until the next crisis or – if we are in a recession – until the economy next ‘takes off’. Uncertainties tend to arise in the long-run. A crisis is typically preceded by a tightening of timescales if influence, as the ‘short-run’ gets very short.
In many places Bronk makes critical remarks that seem obvious but which conflate short and long-runs. In essence his arguments seem to suggest that conventional methods are appropriate in the short run, and do not need to think about the long-run until it is upon us. This might be true, depending on the influence one has on what happens longer term.
Bronk’s approach to methodology seems straightforward and familiar enough, and might usefully be regarded as a form of pragmatism. But it is clearly very different from that advocated by or used by Russell, Whitehead and Keynes. This issue strikes at the core of what we may regard as ‘good reasoning’ or science when faced with complexity and / or uncertainty. Bronk seems to take it for granted that each paradigm will be expressible in ordinary language, with the concepts being expressible in terms of objects, attributes and relations. Indeed, it is only by expressing them in a common language that one can have to reason about them, so this seems an essential assumption of the approach.
Mathematics, as portrayed by Bronk, would seem to deal with precise, numeric, formulae, and hence be even less expressive than ordinary language. But Russell and Whitehead overturned these old assumptions, and Keynes has shown how mathematical models can be used to capture complexity and uncertainty in ways that other approaches cannot. Thus one has the option of using mathematics as the common language. At the least, this can be used to ensure that the resultant model is credible.
If Bronk is right, then one ought to be able to show circumstances in which non-rational behaviour is definitely better than rational. Binmore gives an example, which takes conventional decision theory and relaxes some of the assumptions. It is quite wrong to suppose that the conventional theory is uniquely mathematical or formal.
Like many, Bronk suggests that one acts rationally but looks out for changes and then reacts to them. In fact, as Keynes shows, one may be able to theorise and spot precursors to change, and react to them. Bronk also suggests that the new concepts will necessarily have been created through a process of bricolage. Instead one can proceed by comparing and contrasting assumptions, seeking to relax them while still being able to draw on experience to establish useful theories and identify critical new experiments. Bronk is also dismissive of game theory. But he only considers single-level games (with fixed rules). The theory of multi-level games with learning provides an established methodological framework for thinking formally beyond the conventional.
Bronk comments on the crisis of 2008. This is significantly more tangible, and hence useful to economists, than the book. He credits key insights into uncertainty to Knight (following Bergson) rather than Keynes, continuing to ignore Keynes’ Treatise. He also recommends taking a modelling approach similar to that used for modelling epidemics.
A 2000 paper, written at the height of the tech bubble, has much of the same agenda in terms of recommending national approaches but overlooks the vulnerabilities in the UK’s financial system:
Over the last decades, there has been considerable focus on the degree of flexibility shown by the different models in response to economic shocks, technological changes and the internationalisation of capital markets. ….
By contrast, flexibility is built into the heart of the Anglo-Saxon model, in particular by the rigours of the capital markets on which it relies for finance. …
This may explain why the book was not written earlier.