Comments of the week: Wilde, Radford and Davies on complexity theory and building a narrative

This is a series of connected comments on Geoff Davies’ essay “The Nature of the Beast”, which considers the question of how do we build a narrative linking the various heterodoxies.

Keith Wilde  —  August 3, 2010 at 11:33 am

It seems to me that the editor has already provided the key to building a narrative that links the various heterodoxies. That is, the element that unites them is their common disdain for neoclassical orthodoxy. I recently made a requested conference presentation on “appropriate economics” (in which I made abundant use of quotations from this site) which concluded that the function of neoclassical orthodoxy is “propaganda for the ruling class”.

Each brand of heterodoxy has as its favorite whipping boy a particular aspect of reality that is missing from the neoclassical model of perfection. This defect becomes especially acute when orthodoxy is invoked to support an issue of policy. That is, each of the missing elements of reality is linked to an implicit policy orientation and favorite prescriptions. It is a situation captured in verse about the six blind men of Hindustan who encountered an elephant and each described the full beast according to the particular part that he had happened to touch. With the consequence that each was partly in the right but all were in the wrong. 

I fantasize that if, during two decades as a dissident policy analyst in government agencies, I had had a handbook consisting entirely of authoritative refutations of orthodoxy from each of many heterodox disciplines, my life would have been much simpler and more satisfying. When policy proposals were shown to be defective by first one and then another of the heterodox arguments, policy development would be forced to become sane and focused on relevant aspects of reality as they bear on frankly acknowledged value priorities.

Could not the construction of such a handbook provide the desired “narrative”?

 

Peter Radford  —  August 3, 2010 at 4:57 pm |

Exactly Keith.

The key observation to make is that it is entirely within the power of heterodox economists to create such an integration. No one else can.

The question becomes how?

But this is not new. It has been the case for decades. I see several problems:

One, is that heterodox economists are not as well acquainted with each other as they seem to be with the problems of neoclassicism. To extend your metaphor we have many more than six alternative versions in our particular Hindustan.

Second, there seems to be resistance to achieving coherence in heterodoxy. I think this stems from a misunderstanding, and possibly a fear, that each particular heterodox strand might lose its identity within a larger, more comprehensive, ‘meta heterodoxy’. I see it differently: each strand could become a vital part of an organic whole. This should provide vibrancy and an expanded context for development and research. After all if the economy is a giant complex system then economics is likely complex as well.

Third, economics is highly resistant to the importation of new ideas from outside disciplines. We all know this is especially true of neoclassicism, but I suspect it is also true of heterodoxy as well. Geoff’s essay above is a good case in point. One of the most well worn books on my bookshelf is a compendium of essays from the Santa Fe Institute entitled “The Economy As An Evolving Complex System”. It is a mine of new thinking and gives economists insight into the power and use of complex systems analysis. That work was published in 1997. Another example is the work of Georgescu-Roegen on entropy and economics. His work was first published in 1971. We all have plenty of similar examples of well established by still largely ignored useful modes of thinking yet to make a central impact on economics. The opportunity is huge if we can venture outside our own box.

Fourth, following from the above, another form of resistance comes from the reaction many heterodox economists have to ‘scientism’. They appear to conflate science with the elimination of humanity from analysis. That is a legacy of neoclassical analysis, but should not necessarily preclude the importation of analytical techniques from outside to provide our real world efforts with both a rigor and a human face. In my opinion one of the reasons for the failure of economics – as a whole – is the atrophy of its analytical toolkit.

Fifth, heterodoxy is rich with traditions studying the impact of institutions, geography, gender, class, knowledge and so on, all of which were discarded as irrelevant when neoclassicism disappeared down its constrained allocation hole. But those traditions are not stitched together any more than neoclassicism is. Which is why your desired handbook is still missing.

My larger point speaks to your issue of policy impact. We are trying to combat a prevailing worldview that is far too well entrenched in the policy making elites, whether they be in government, business, or academia. Free market dogma is dangerous, hegemonic, and downright false. It needs to be overturned. Not by ‘six blind men’, but by one determined group.

The satisfaction of your desire for an alternative narrative depends on our ability to set aside our differences and become that group.

 

Keith Wilde  —  August 4, 2010 at 1:21 pm |

Peter, my suggestion is much less ambitious than the integration you propose, but I think it would be a useful beginning. And there does seem to be pretty widespread agreement with the editor’s observation that the one common element among the varieties of heterodoxy is their disagreement with policy implications of the neoclassical model.

The handbook I have in mind would not knit them all together into a new whole, but it could, I think, be a step toward uniting the various groups into one that is united against the free market dogma that claims economic theory as a foundation. They don’t have to be integrated beyond that minimal degree in order to be useful in a resistance movement.

Think of it as an encyclopedia rather than a textbook. It would consist of articles that reduce each brand of heterodoxy to the essence of its disagreement with the “adversary”. Not an impossible job, I think. Your mention of Georgescu-Roegen reminds me of my collaboration with others to bring him to Ottawa in the late seventies. He did a pretty good job of presenting the essence (minus the physics) in a couple of lectures.

The product is not a textbook. It is a tool for policy analysts and advisers; no more than a reading list item for teachers. When the government analyst is asked to evaluate a proposal, he can leaf through its contents for the most relevant and authoritative arguments that refute whatever it is in the text on his desk that turns his stomach.

This kind of application does not mean that the analyst/adviser then turns around to endorse the full program of the heterodox variety that was most useful to him in a critical evaluation of the proposal. He may have used more than one of them. His evaluation will (may) mean that the issue is thrown back into the pot for further committee-built policy development, with the prospect that it might emerge in a more benign version, i.e. less anchored to the free market dogma.

Keith

 

Geoff Davies  —  August 4, 2010 at 2:16 am |

Peter Radford makes some useful points, though my perspective may be somewhat different. It does make sense to me that heterodox economists refer more to neoclassicism than to each other, and that they have some resistance to ideas from other heterodoxies or other disciplines.

I would expand point Four – other disciplines can supply not only analytical toolkits and methodologies but different conceptions.

This gets to the core of our topic. The requirement is not just for a collection of heterodoxies, with or without handbook, but for coherence. It has to be, because economies will only be usefully understood through a framework large enough to accommodate everything about economies, rather than just an aspect of economies.

There are many crucial things missing from neoclassicism, like social interaction, the way real people make choices, the way real firms are managed, our inability to foretell the future (!), economies of scale, that the economy is a subset of the biosphere and subject to physical and biological constraints, money (amazingly), marketing, people’s need to belong, and so on. We won’t get a useful replacement by focussing on only one of these aspects. A useful framework has to accommodate them all.

My contention is that complexity does provide such an overarching framework. There is no reason why it can’t accommodate each of the factors I just mentioned. I don’t know how to include them all – that is the program for the future – though there are likely to be some conclusions that emerge fairly quickly. As more aspects of real economies and real people are included, so the character of the beast will come into sharper focus.

On the other hand “complexity economics” will not accommodate other conceptions that aspire to frame the whole subject (neoclassicism, Marxism?). So there will be some incompatibilities among heterodoxies.

You may agree or disagree that complexity provides a general framework, but the need is there for a general framework, not for a collection of heterodoxies. Of course my summary is very brief, and it is not too surprising if the central conception doesn’t immediately take hold of readers (as evidenced by the comments so far). But is there a curiosity to explore this path further … ?

Peter mentions a book from the Santa Fe Institute, and I referred to Beinhocker’s The Origin of Wealth, both of which provide many stimulating analyses and ideas. My own introduction to complexity was through Mitchell Waldrop’s Complexity (Touchstone, NY 1992) which was also based around the Santa Fe Institute. I highly recommend it as a good read with many insights.

Another word on science (scientism?). What excited me when reading about complexity is that it can accommodate living things. It becomes apparent that old reductionist science is brilliantly successful for studying the inanimate world, but hopeless for studying life. Because of the phenomenon of emergence, the reductionist approach of breaking a system into pieces (e.g. asocial agents with static and identical preferences) fails. A holistic view is (also) required if the full behaviour of the system is to be understood.

The notion of complexity has emerged only in the past few decades. You can still do good science with complex systems, but you must be holistic and humble, as we must also be when studying people. So reactions against old-style ‘scientism’ may not be relevant.