Over the last couple of weeks, this blog has focused on new thinking about macroeconomic policymaking. It has looked at the area of macroprudential policy making, both in terms of what it is and who should do it. The core reason for this re-thinking of macroeconomics is the need to bring financial market imperfections into policy considerations. Whilst this work in the policy area is absolutely necessary, more work is also needed in the theoretical area that will ultimately underpin the policy sphere. My own attempt at this has been the Cubic IS curve, but a more fundamental theory has been developed by Andrew Lo over the past decade. Lo’s Adaptive Market Hypothesis (AMH) is a very intriguing theory that could prove an excellent basis for bringing financial market imperfections into broader macroeconomic models.
What is AMH?
AMH is a direct challenge to the Efficient Market Hypothesis (EMH) that has been so dominant in economic theory. EMH is based on the notion that markets fully, accurately and instantaneously incorporate all information into market prices. Market participants are said to be fully rational and make optimal decisions about economic trade-offs. The attacks on EMH have become ever more frequent since the financial crash, but Lo began developing AMH in the early 2000s. Unlike EMH, AMH is based on principles of evolutionary biology, such as competition, mutation, natural selection and reproduction. For Lo, these forces come in waves and determine a fluctuating level of efficiency in the market.
The motivation for developing this new theory has come from behavioural sciences like psychology, which is something that I am very open to. Lo cites behavioural studies like that of Kahneman & Tversky where individuals assign probabilities according to ‘representative’ ideals rather than basic probability theory. For behaviouralists these and many other studies – studies that show loss aversion, over-confidence, over-reaction, and regret – show that market theory based on rational agents are likely to be misplaced.
Lo’s Neuroscience perspective & the Triune Brain Model
In Lo (2005), Lo begins this section with a challenge to the mainstream view of economists that emotion moves individuals away from rationality. From studies in neuroscience, Lo shows that emotion is central to our rational behaviour. For example, emotion is the basis for reward and punishment situations that facilitates the choice of advantageous outcomes. Moreover, fear and greed are the products of evolutionary behaviour that help individuals adapt to increase the probability of survival.
If emotions are not to blame for irrational behaviour, what is? Lo sheds light on this question with the help of the Triune Brain model – which rests on the basic point that the brain is not a homogenous mass of cells, but rather a complex collection of specialised components.
Lo summarises this view of the brain as the outcome of the evolutionary process of brain development, where basic survival functions appeared first (the reptilian brain), more advanced social features came next (the mammalian brain), and finally the more complex cognitive abilities emerged most recently in the last 100, 000 years (the hominid brain). While the interactions of these three areas are complex, this categorisation helps us to understand how seemingly irrational behaviour can occur. In this post I described Asch’s conformity studies, where participants made obviously false claims just to go along with the group. In this case, we could think of this as the social areas of the brain overpowering the cognitive parts. Similarly, a bar fight would represent an extreme situation where the survival areas of the brain clearly engage to overrule the cognitive functions – also known as the flight or fight response.
What does this mean for economic theory and models? Endogenising human behaviour!
For Lo, the increased evolutionary specialisation of the brain is a sign of increased fitness and adaption to the environment to encourage survival. As environments change, we have the ability to adapt and learn to implement more advantageous behaviour. For economic theory, this adaption of human behaviour implies that ‘preferences’ are not stable through time, as assumed by EMH. In economic jargon, therefore, human behaviour needs to be endogenous in the model – in other words, mechanisms need to exist in a model whereby the human element of the model can change and adapt as the environment around them changes.
This idea lends itself to fit in with Minsky’s theory where the economy is inherently unstable , but it also reminds me of De Grauwe’s work and my own Cubic IS curve. In my own attempt to do this, individuals are heterogeneous in their propensity to take on debt, but as house price expectations increase, more and more individuals take on debt. This debt then fuels consumption and provides the instability in the system. In De Grauwe’s work, humans also react to the environment around them. Agents in his model view the economy through one of two heuristics: a fundamental view, where agents forecast a steady state output gap, compared to the other ‘extrapolative’ view, where agents forecast the current output gap into the future. From a positive output shock, the extrapolative view is correct and this attracts others from the fundamental heuristic to this view. This process becomes self-fulfilling, as more and more people participate in the asset boom, more and more people are drawn to it, and so on.
The next steps are the challenge
AMH makes a lot of sense to me, but formalising this in economic models is the challenge going forward. De Grauewe’s model, although producing non-normal output gaps, is a dynamic model, but it is in equilibrium. The Cubic IS curve does touch on the inherent instability of an economy, but it is static. Economics will probably push forward in both respects: equilibrium models that exhibit more volatility whilst remaining in equilibrium, and another class of models where instability is in-built and disequilibrium is a possible outcome.
 Lo, A, W (2005). Reconciling efficient markets with behavioural finance: The adaptive market hypothesis