Pages

Showing posts with label Evolution. Show all posts
Showing posts with label Evolution. Show all posts

Friday, 4 July 2014

Balloon Kenya, or ‘doing’ Complexity Economics

This post originally appeared on the Balloon Kenya blog which you can check out here

When you live in a big city, you often get to know the geography of different ‘islands’ within it without being able to connect them in your head. When this eventually happens, and West and Central London lock together, it’s a good feeling. When this same thing happens with two ideas, the feeling is even better. When this happens with two ‘spheres’ of your life, say intellectual and professional, it feels bloody awesome.

I’ve recently begun a project with Balloon Kenya, an organisation which trains UK based students (a good mix of international and British) in entrepreneurship and uses their enthusiasm and skills to support aspiring Kenyan entrepreneurs. The approach to entrepreneurship they promote is based on the ideas of Sara Sarasvathy, who differentiates between ‘causal’ and ‘effectual’ reasoning. ‘Causal’ reasoning starts out with a “pre-determined goal and a given set of means, and seeks to identify the optimal, fastest, cheapest most efficient alternative to achieve the given goal”. It’s a planning mentality, where you break the problem down into its constituent parts, come up with some logical answers, build your product, and either sink or swim.

‘Effectual’ reasoning starts with the means and “allows goals to emerge contingently over time,” in an iterative method with a heavily action-oriented, trial-and-error-based mindset. Testing – and learning along the way – is key. It’s an exciting, and potentially revolutionary approach to unlocking the entrepreneurial instincts buried inside all of us. And it locks in perfectly to an evolutionary view of the economy as a complex adaptive system, my pet intellectual concept at the moment. Cue mental fist pump. In fact, complexity can explain why this ‘effectual’ approach to entrepreneurship is the most effective.



The Marshmallow Challenge demonstrates the difference between causal and effectual thinking. The aim is to build the tallest structure out of spaghetti, tape and a marshmallow – which has to go on the top.

As I’ve been reading recently in Eric Beinhocker’s ‘The Origin of Wealth’ the economy, in the evolutionary complexity-based view, is a system of millions of interacting, boundedly-rational agents; who are constantly learning and adapting their behaviour. This produces an economy that is dynamic and has ‘emergent’ properties which are more than the sum of its constituent parts. Its behaviour is non-linear, meaning that small changes can have huge effects (think the financial crisis or mobile money transfer cascading across the Kenyan economy) All this makes it mighty hard to predict or plan, beyond human intelligence in fact. Alain Lewis, for example, shows that the perfect rationality required to process all the information in the economy is beyond a ‘Turing Machine,’ an imaginary, all-purpose, all-powerful computer. ‘Causal’ thinking for entrepreneurship here takes a big hit.

“Evolution is cleverer than you” says Orgel’s Second Law. And in the economy, wealth is created not by individual human rationality but by the operation of a simple evolutionary algorithm that differentiates, selects and amplifies successful models for conducting business, with the market as its ‘weapon of choice’. Essentially this algorithm ensures that those ’fit’ models best suited to the environment survive and therefore proliferate, whilst the unsuited die out.

Evolution basically says I will try lots of things and see what works and do more of what works and less of what doesn’t. The effectiveness of ‘effectual’ thinking makes total sense in light of the above. A customer-oriented approach – which takes the burden of selecting successful and innovative ideas from individual human brains and places it with ‘customers’ (Balloon Kenya’s terminology) - aggregated into ‘the market’ (evolutionary economics-speak) is a formula for success.

For Eric Beinhocker, complexity economics pioneer and author of 'The Origin of Wealth', “Wealth is knowledge and its origin is evolution”. For Eric Ries, Balloon Kenya pin-up and author of the ‘The Lean Startup,’ “Value in a startup is….validated learning about how to build a sustainable business”.

West and Central London just locked together.

Wednesday, 25 June 2014

Left wing? Right wing? It’s complex.

In ‘The Origin of Wealth’ (see initial review here), Beinhocker has a long section where he applies his ideas about complexity and evolutionary economics to more practical contexts. He’s a McKinsey man, so one is obviously business strategy and organisation. Another is politics, where his ambitions are humble: “the Complexity approach to economics has the potential to make the historical framing of politics obsolete”. And we aren’t just talking “mushy” (his word) Third Way, New Labour pragmatic populism here.

"Nail them, even at personal cost to yourself" credit: SenorGif
He characterises the dominant Left-Right distinction as based on two disagreements: 1) how to view human nature (Left: humans are essentially altruistic and bad things are society’s fault vs. Right: humans are essentially selfish and pursuing self-interest is an inalienable right). And 2) the role of states vs. markets. According to complexity economics in both cases they are both wrong. On human nature, it turns out we are “conditional co-operators and altruistic punishers,” that is to say we operate according to a (hilariously) modified Golden Rule of: “do unto others as you would have them do unto you... but if others don’t do unto you, then nail them, even at personal cost to yourself”. Which has an intuitive resonance given my childhood Monopoly-playing experiences. Anyway, that nugget should be the foundation of policymaking (and Beinhocker shows how in welfare debates, for example), not outmoded conceptions based on Left/Right distinctions.

More interesting are his ideas about states and markets. I’m summarising horribly but basically, strong government intervention in the economy as advocated by the Left is misguided because, 1) faced with the vast complexity of the economy, human rationality is hopelessly inadequate - as Orgel’s Second Rule states “evolution is cleverer than you are,” so don’t try to be a market. And 2) if the arbiter of economic fitness is some politician or party at the centre (in Beinhocker's terms, 'Big Man') then the economy is likely to start to orientate itself towards promoting the interest of that politician or party. On the other hand, the Right's fantasy of unadulterated markets existing in pristine isolation is equally misguided, because the market as part of the economic evolutionary system is constructed out of a vast array of Social Technologies, comprising institutions, norms etc., many of which rely on government. The complexity solution? States help shape the ‘economic fitness function’ (= what we value and want the economy to work towards) whilst markets carry out the job of differentiating, selecting and amplifying Business plans according to this fitness function.

So what, expressing things in terms of complexity doesn't remove the niggly problems and trade-offs that are inevitable in politics, right? This is my initial feeling - taking the example above, to be honest at this point, pretty much everyone in the political mainstream agrees that states and markets both have an important role to play, so the caricatured socialist Left and free-market Right that Beinhocker slays aren't the issue. But complexity's basic insight that micro-behaviour matters for the operation of the system as a whole could be key. In fact Beinhocker suggests that it's here that we can find the answer to persistent poverty: in the norms guiding individuals’ behaviour (i.e. culture), which means that the usual prescriptions of the Left (redistribution) and the Right (laissez-faire, individual incentives) won't solve the problem. Some of the stuff he says about national cultures I am immediately wary of (anyone citing Samuel Huntington’s ‘clash of civilisations’ gets my alarm bells ringing), but otherwise it seems plausible.

I had a little rummage around Google Scholar and apparently the above paragraph sums things up more or less: according to Paul Cairney at the University of Aberdeen (gated, apologies), some people think that Complexity Theory in politics is a passing fad, whilst others gibber unintelligibly about a ‘paradigm shift’ (*ahem*). Getting concrete remains the issue: ‘mapping the landscape’, ‘modelling the struggle’ and ‘encouraging systemic emergence’ all sound wonderful but are basically unintelligible. Add in a greater use of trial and error, learning from pilot projects, and accepting a degree of ‘error’ when designing policies instead of seeing error as ‘failure’ and we’re getting closer to 'Monday morning' realities, but still vague.

Great, but what does it mean in practice? credit: http://thewildpeak.wordpress.com/


Specifics are in short supply, so as a conclusion I’ll veer hazily back to fun, broad, sweeping ideas. If evolution is super clever at sorting through designs to get the best one, we need to bring evolutionary processes inside political decision-making somehow. Duncan Green spells out the implications: “we need to find a way of designing and strengthening institutions to make non-market forms of selection and amplification as effective as possible”. And this means nibbling away at the margins at the goals of individual agents, the norms that guide behaviour and the feedback loops in the system.

This all amounts to saying that the answers remain very much unclear. But at least we have a new question: no longer “Left versus Right” but rather “how best to evolve?”