When you think of economics, traditional concepts such as of supply demand, interest rates, inflation and GDP all come to mind. These are relatively abstract notions that do not seem very important on a gloomy, cold and rainy morning in March. Nor do they provide a natural gust of excitement in your body the way tabloid pop culture news might. Arguably, the problem with economics is that we have lost sight of what it really is: A common denominator of human interaction.
You might say that my definition appears far removed from economics. Well, the opposite is actually true. It's nothing more than common sense re-rationalized. Economics is really about human interaction; more specifically about exchange and the conditions of exchange. Human interaction, be it on a micro or macro level, is really a behavioral science. So, in fact, when we are studying markets we are really studying aggregate psychological behavior: the sum of all interactions.
Currently mathematics is used to study economics and finance by constraining the "human" elements prevalent in the exchange. Common assumptions such as “risk neutral”, “risk averse”, “homogenous expectations” and “rational actors” are all foundations upon which many economic and financial models are based. To give credit where credit is due, it has to be said that by constraining human behavior, we are able to examine how perfect markets would work and this has contributed greatly to our understanding of economics and financial markets.
However, humans are far more than merely rational, utility maximizing robots. We have feelings, emotions, memories, a conscience, and are often absorbed by greed. So, in fact, human rationality is a biased rationality, if not a flawed rationality. Human rationality is different from the machine-like rationality upon which clever mathematicians and econometrists build their assumptions models. This begs the question of what this means for markets?
Behavioral finance is a relatively new field that draws on heuristics, cognitive biases and bounded rationality. The basic premise is that behavioral biases play an important role in markets. Even more interesting is the study of neuroeconomics, which studies how the brain makes choices in combination with psychology, economics and neuroscience.
Imagine modeling the neuroeconomic behavior of macro market movers such as hedge fund managers. A computer model with A.I. (artificial intelligence) capabilities would be able to predict and model the market scenarios and move against them accordingly. Since economics as a social science destroys the concept of ‘perfectly perfect’ markets anyways, the neuroeconomic models could create tremendous arbitrage opportunities. There are some limitations that lie at the core of not only this idea, but general economic modeling that need to be considered first.
Read The Social Science of Economics Part 2 as we enter the twilight zone of economics!
You might say that my definition appears far removed from economics. Well, the opposite is actually true. It's nothing more than common sense re-rationalized. Economics is really about human interaction; more specifically about exchange and the conditions of exchange. Human interaction, be it on a micro or macro level, is really a behavioral science. So, in fact, when we are studying markets we are really studying aggregate psychological behavior: the sum of all interactions.
Currently mathematics is used to study economics and finance by constraining the "human" elements prevalent in the exchange. Common assumptions such as “risk neutral”, “risk averse”, “homogenous expectations” and “rational actors” are all foundations upon which many economic and financial models are based. To give credit where credit is due, it has to be said that by constraining human behavior, we are able to examine how perfect markets would work and this has contributed greatly to our understanding of economics and financial markets.
However, humans are far more than merely rational, utility maximizing robots. We have feelings, emotions, memories, a conscience, and are often absorbed by greed. So, in fact, human rationality is a biased rationality, if not a flawed rationality. Human rationality is different from the machine-like rationality upon which clever mathematicians and econometrists build their assumptions models. This begs the question of what this means for markets?
Behavioral finance is a relatively new field that draws on heuristics, cognitive biases and bounded rationality. The basic premise is that behavioral biases play an important role in markets. Even more interesting is the study of neuroeconomics, which studies how the brain makes choices in combination with psychology, economics and neuroscience.
Imagine modeling the neuroeconomic behavior of macro market movers such as hedge fund managers. A computer model with A.I. (artificial intelligence) capabilities would be able to predict and model the market scenarios and move against them accordingly. Since economics as a social science destroys the concept of ‘perfectly perfect’ markets anyways, the neuroeconomic models could create tremendous arbitrage opportunities. There are some limitations that lie at the core of not only this idea, but general economic modeling that need to be considered first.
Read The Social Science of Economics Part 2 as we enter the twilight zone of economics!
2 comments:
Excellent points made here. I have long believed that the perfect markets theory is flawed when applied to the stock market, due to a number of factors such as psychological based autocorrelation on the (time - n) scale, the concept of strategic ignorance, and unequal information flow that disrupts proper game theory scenarios. An AI that could emparically parse the effect of these factors could very well have a heavy advantage over the humans that can't help be affected by them.
-The Cottage Economist
TheCottageEconomist.com
Interesting comments, I wanted to add that in defense of the Chicago school, that markets need not be perfect as long as they are efficient. When we think in terms of efficiency it is much easier to include some of those factors that you mentioned and look at where sources of inefficiency are and how they affect the pricing spread in markets.
I am very curious as to the effect of AI in exchange markets. What will happen to the trading profession when these systems become operational?
How do you beat an unemotional rational mean variance AI optimizer system on the market with a supercomputer and algorythm arsenal?
Post a Comment