July 5 2009
I am most often irritated by those who attack the bishop but somehow fall for the securities analyst — those who exercise their skepticism against religion but not against economists, social scientists, and phony statisticians.Nassim Nicholas Taleb
In the wake of the market meltdown of 2008, it’s fascinating to see mainstream opinion wising up to the reality of just how badly economics has failed as a predictive discipline. This is nothing new — it reflects a long term structural failure of economics to become the science it wished it was. Despite its questionable success, over the past century, economics has been afforded a privileged position at the best universities and distinguished accolades such as Nobel Prizes. Even after all this time, economics still seems to be missing basic scientific consensus on fundamental issues. This emperor is almost certainly wearing no clothes, and may well have reached the end of his parade.
Criticisms of economics establishing itself as a science date back to the 19th century, but are mostly moral in character. A fair explanation of how economics has misused science requires a background excursion into 20th century philosophy. For those not indoctrinated in the cult of the academy, the philosophy of science is a distinct area of scholarship within philosophy, and is mostly concerned with epistemology (trying to explain what knowledge is). After the influence of Karl Popper, one of the central elements of this subject became how to separate non-science from science.
The most important contribution of Popper was to generalize this apparatus, to show, not just how science works, but how science improves. Popper's move was to emphasize that theories must be falsifiable. A workable theory must be able to be logically disproved - in other words, you can never confirm that a theory is true, but you can prove that it is false. Popper firmly believed that continuous experimental testing and falsification would yield better and better theories.
In the 1930s, scientists unanimously understood this view of progress (it was how they worked), yet philosophers at the time were apt to describe science as a complete mosaic of knowledge with new blocks of knowledge being added to enlarge the picture. Despite the caning from Popper, it took until the 1950s and ‘60s for scholars to agree that scientific reality is not such a comfortable process of knowledge accumulation. One of the most famous books of its time, Thomas Kuhn's The Structure of Scientific Revolutions lined up an array of historical evidence to show that scientific progress is wild and messy, with a tendency to dissolve in a fog of confusion with radical new theories exploding on the scene every so often, tearing down previous traditions that he called ‘normal science’, until eventually the radical new theory settles into the mainstream and itself becomes normal. Most literate people have been touched by the book, even if they are unaware of Kuhn himself - his usage of the phrase “paradigm shift” to explain the onrush of a new scientific era has become deeply embedded in popular culture.
The most interesting thing about The Structure of Scientific Revolutions is that when it came to defining a clear outline of how science operates, the actual prose of the book was as foggy and turbulent as the science it described. Critics of Kuhn later came to view the book’s reception as an open letter to humanities disciplines, encouraging them to view themselves in a more scientific light. Everyone wanted a paradigm of their own.
It is much easier in hindsight to pick apart the bizarre contradictions this trend implied. Sociologists, psychologists and anthropologists who desired the edifice of science had gained a great pillar they could throw down in support, and the book rose in popularity, perhaps not-coincidentally, at the same time as certain post-structuralist tendencies were emerging in the humanities disciplines (leading to endless arguments for philosophers over how Kuhn was misinterpreted). The value of a research field being recognized as science was more than just a matter of prestige — it opened the door to vast amounts of research funding, so it is certainly understandable why so many sought this status.
Which brings us to economics, also considered social science, and exerting a huge influence over this time period. It is interesting that the impact of Kuhn was nowhere near as great for economics at this time, as it was for other social sciences. Economists — perhaps because of relentless mathematization of their work — appeared to be operating in a bubble of security away from endless doubting of their scientific integrity. Few economists would question that economics was a science at all, an apparent attitude reinforced by what seems to be remarkable absence of writing on the history of economics during the mid-20th Century. Unlike sociology and other social sciences, the 20th century economics establishment never faced serious resistance from insiders as to whether it actually should strive to be a science.
The work that seeks to challenge this lack of history in the most depth is Phillip Mirowski’s More Heat Than Light, a complex and deeply challenging historiography that is not as widely known as it deserves to be. The subtitle is revealing — economics as nature's physics; physics as nature's economics. Essentially, Mirowski argues that late 19th century economists were able to capitalize on the success of the energy concept in mainstream physics. This led to a distinct new discipline of economics breaking away from its traditional foundations of moral philosophy to become mired in a very specific kind of algebraic mayhem, obsessed with building equilibrium models. In pop speak, we would describe economics as having “physics envy”. To quote the Cambridge synopsis of the book:
Misuse of conservation principles led to serious blunders in the mathematical formalisation of economic theory.
Such mixed metaphors and confused calculus actually led to the explosive growth of economics, but it was not all one way — Mirowski also documents instances where physics borrowed ideas from economics. By World War 2, economics had built up a vast archive of mathematical literature and a distinctly separate community. By this point economists were no longer studying the economy per-se, they were studying a complicated series of algebraic puzzles - games that only insiders with the requisite mathematical knowledge could play. Modeling techniques became more baroque, and attentions drifted towards researchers with a grasp of complex linear algebra, leaving psychologists and sociologists who still looked at the behavior of ordinary human beings in the dust.
Enjoying major institutional and military support, the discipline of economics prevailed in its own hubris, and was never forced to justify itself. This led to a stability that other social sciences could not maintain, in their perpetual crises of identity and had far reaching consequences.
By constructing increasingly abstract models, economists became distanced from the actual economic behavior they were claiming to represent. Normative assertions about social reality were treated as mere parameters to equations, and thus vital questions were never asked about these fundamental assumptions; a situation that has generated an endless stream of criticism of economics over the years.
The biggest point of contention was the basic list of assumptions that formed the very bedrock of neoclassical economics:
Instead of trying to establish a testable, empirical foundation for these assertions (which would have come up short, due to peculiar aspects of the human condition), generation after generation of economists accepted them as rote, in order to get on with the task of mathematical model building.
It’s interesting to note here, that Popper himself rejected these assumptions of rationalism, yet he praised economics as being the best social science, and weakly defended it on the basis that the use of rationality was the way forward for all social sciences.
If classical economics was improving like a science should, it would have continued to generate concrete testable predictions about the future of the economy, of which, there would be widespread agreement. Instead, as outsiders would view it, the academic community spent most of the 20th century bickering and engaging in pissing contests and displays of mathematical one-upmanship. And so unsurprisingly, today we find state-of-the art economics being wholly useless in real world practice and mainstream economics degenerating into a complete and total mess:
What arises in my mind is the strong suspicion that economic theory, as it is practiced and taught at the world’s leading institutions, is so far from consensus on certain fundamental questions that it is basically useless for adjudicating many profoundly important debates about economic policy. One implication of this is that it is wrong to extend to economists who advise policymakers, or become policymakes themselves, the respect we rightly extend to the practitioners of mature sciences. There is a reason extremely smart economists are out there playing reputation games instead of trying to settle the matter by doing better science. The reason is that, on the questions that are provoking intramural trashtalk, there is no science.
Much of the work that we now think of as economics, political science and other social sciences will likely be displaced by some hybrid of biology, experimental economics, psychology and other fields that can evaluate hypotheses for the quantified prediction of human behavior through structured falsification tests (or, sometimes, true “natural experiments” in which non-intentional random assignment has occurred).
As it turns out, there is a fundamental obstacle to the predictive potential of economics, one that has been glaringly apparent in recent months as the global financial sector collapses under its own greed and derangement. This show-stopper is the plain fact that economic agents are themselves knowledgeable about economics. They can assess a situation and change the situation to benefit their own interests. This reflexive effect is a contradiction that economics cannot cope with:
Any algorithm that can reliably predict the behavior of a financial market will be used by participants in that market to earn money, altering the system in a way that leaves you right back where you started. In this sense our ability to model economics will always be worse than our understanding of the weather, since we don't have to worry about a raindrop anticipating that it will hit the ground before it even forms, and taking steps to change the outcome.
An important lesson to learn here is that the deployment of rigorous mathematical models is not enough to make a theory scientific. Falsifiability, testing of assumptions, learning how to be wrong, is what science depends on, and the problems with applying economic theories to some of the most complex scenarios in human development are clearly evident. Compartmentalization of knowledge is not complimentary to the goal of developing a comprehensive social science.
The hubris, arrogance, and anti-social disunity of 19th and 20th century academia needs to be better recognized. Economics in its traditionally enshrined format is an awkward relic of this twisted history and should be accepted as such. The sooner, the better. In fact, it seems that this is already happening.