One of the primary purposes of economics, perhaps the purpose, is reliable prognostication. Economics fails as a social science if it is not capable of adequately predicting and prescribing effective solutions for economic crises. Absent this capacity, economics is no different from history-- a short-sighted description and perpetual re-interpretation of the past, with no intention to extrapolate beyond the data. Consider it another way, assume that economics has no predictive powers, what then is its purpose?
If forecasting is a central responsibility of economics, then I believe its function is distorted by two primary forces-- the insistence that economics functions like a real science and that arrogant belief that economic progress can be achieved without ever consulting a historian.
Regarding the first problem, the most persistent delusion among economists is the belief that their discipline has no limitations and that, like in science, everything can be reduced, quantified, and calculated. The sooner that this hubris is abandoned, the better.
There is an exactitude and consistency to scientific observations that gives rise to robust predictive power. The laws of motion or gravity, for example, are not capricious--exactly 100% of people who jump out of windows will fall to the ground. Tides goes in, tides goes out; there’s never a miscommunication. And contrary to what some political commentators would have you believe, these phenomena can easily be explained with a third-grade understanding of science. The observations of economics, however, are dictated by the infinitely more complex and inconsistent whims of human behavior, which makes predictions impossibly complicated.
Nevertheless, there is a bias in economics to believe that human behavior must also be subject to these immutable laws and that the tides of human interaction can be perfectly communicated through a simple equation. The ‘dismal science’, perhaps named as such because its scientific precision is disappointingly overstated, consistently makes assumptions that are inconsistent with reality. From the perspective of classical economics, humans are Homo Economicus-- cold, calculating, rational creatures seeking utility-maximization everywhere. But this simplified caricature of reality ignores revelations from behavioral science which shows that human behavior is predictably irrational-- we disregard information that conflicts with our beliefs; we abandon intuition and follow the herd; we overestimate our ability to control and interpret events; we discount the usefulness of past experience when dealing with the present; the list goes on. What utility do models and theories have, even from an academic perspective, if they don’t resemble reality?
This obsession with economics-as-science has destroyed its reputation in the wake of the current financial crisis, and unless this arrogance is tamed, little progress will be achieved within the discipline.
This is not to say, however, that economics shouldn’t strive for better assumptions, but that it should reserve judgment on issues until it has established theories that are actual reflections of reality. Scientists are okay with temporarily confessing ignorance until all the evidence is in, but economists operate on a different principle: over-simplify reality, make data conform to textbook theories, and fervently argue the conclusions as if they’re scientific law. This might not be a dangerous practice if it was limited to the realm of entertainment, but using it as the basis of academic progress or regulatory policy is frighteningly naïve.
Similar to how torture produces false confessions that can lead to the misapplication of government resources, so to can economic predictions grounded in false assumptions lead to catastrophic overreactions.
To give an example, consider the Efficient Market Hypothesis which humiliates the complexity that underpins human behavior and reduces it to trite buzzwords-- “rational actors”, “perfect competition”, “efficient markets.”
The theory asserts that a product reflects all information available to it, which implies that asset prices will always reflect their true value. Market strategist Jeremy Graham argued that belief in this theory encouraged regulators to dismiss accusations of speculation in the housing industry during the financial crisis. Housing prices, according to the theory, had to reflect the true value of the housing market and therefore government intervention could only lead to inefficiencies. This faith-based devotion to macroeconomic theory contributed to the greatest economic downturn since the Great Depression.
Long before the crash, studies by behavioral economists convincingly demonstrated that the value of stock is much more a reflection of informational biases, herd behavior, and irrational exuberance than efficient pricing. Despite this, economists refused to abandon the theory or negotiate a new assumption, and thereby indirectly manufactured our current economic crisis.
Long-term Capital Management (LTCM), a hedge-fund management firm, provides another example of how faith in economic models is misguided. The company used complex mathematical equations to take advantage of the minuscule differences in returns between U.S., Japanese, and European government bonds, making profit based on a concept known as ‘arbitrage’. Over time, the company ran out of good deals, so it had to start using massive amounts of money to exploit incredibly small differences between the diminishing pool of remaining bonds. Eventually, a crisis in Russia disturbed the entire financial landscape and forced investors to abandon Japanese and European bonds for ones in the United States. The models that LTCM relied on failed to take into account this possibility, and so the company lost $1.85 billion in capital and required a $3.625 billion bailout. Rather than relying on obvious common sense that cautions against risky, highly-leveraged bets involving minuscule returns, the hedge fund managers instead put their faith in inaccurate models at the expense of American welfare.
I assumed econometrics and complex models would imbue the discipline with much needed objectivity, at least from an academic standpoint, yet economists have failed to answer Larry Summers’ challenge to name a single economic theory definitively disproved by econometrics. If economics behaved more like a science, you’d expect to see far more progress than has been observed. We should have closed the door on a lot of theories while proving others by now. Perhaps I’m being too harsh, given the ambitious agenda of economics, but there needs to be expectations of a discipline called the “dismal science.”
In physics, scientists do not invoke the geocentric conception of the galaxy or Lamarkian evolution in the formation of their theories, and you would expect to see this general progression in any discipline. Such behavior, however, is less readily observable in economics, in which the popularity of century-old theories fluctuate in concordance with the business cycle.
Frederick Hayek’s criticism of Keynsianism, for example, which suggests that central government planning is the main cause of the business cycle, was thoroughly repudiated by Reagan’s failed economic policy for which Hayek was a main advisor, but his theories are now being re-evaluated in light of the recent crash. Keynesianism was discredited after stagflation in the 70s, but is now hailed for its robust explanatory powers. Keeping an open mind is needed in any discipline, but I fear that ideology has so infected economics, given its normative nature, that consensus-building is almost impossible.
Continuing with the science analogy, I will concede that there are, of course, rare exceptions to scientific laws--Newtonian Gravity, for example, breaks down in the event horizon of a black hole. But these exceptions are few and far between and hardly encumber day-to-day scientific predictions. Economics, however, seems to encounter a black hole every business cycle. Shocks and recessions necessitate the radical restructuring of the entire discipline; models are thrown away, assumptions retooled, and data re-examined, and yet economists are not getting better at making economic forecasts.
On October 17, 1929 the economist Irving Fisher declared, ”stocks have reached what looks like a permanently high plateau.” In the 1990s, there was a general economic consensus that Japanese growth was so rapid that its economic prowess would soon surpass that of the United States. Four months into the 2007 financial crisis, Henry Paulson, then Secretary of the Treasury, said that there was zero risk of a housing bubble—it hadn’t ever happened in history, after all. All of these sanguine forecasts proved to be categorically and almost immediately incorrect. So why are economists so bad at making predictions?
John Montier, in his letter, “In Defense of Old Always”, produced a graph that brutally detailed the failure of economic consensus over the last four decades. It demonstrated that at no point in time was consensus forecasting anywhere close to predicting actual inflation, bond yields, real GDP, or the likelihood of a recession. Despite this, economists and laypeople continue to cling to predictions as if they contain any merit at all. The phenomenon of “base rate neglect” explains why this reliance is so tenacious. Humans have a psychological need to make and receive forecasts because they cannot grapple with the idea of a chaotic world unaffected by human inquiry and understanding. As a result, when evaluating our own belief in the reliability of forecasts, we cling to information close at hand, such as a recent prediction or intuition, but we discount the base rate of the relevant event, such as the percentage of failed predictions. On the aggregate, this means that most humans will accept economic predictions without scrutiny, purely because they are discomforted by the admission of ignorance. Economic forecasters, then, are frighteningly similar to snake-oil salesmen; they exploit their desperate consumers with unverifiable quackery to make a quick buck. But unlike psuedo-scientific hacks, if these economic salesmen are exposed, they are instead given a job as a Treasury Secretary or Fed Chairman.
Financial economics is no better at making predictions. A study in the journal Economics and Portfolio Strategy tracked 452 managed funds and found that, from 1990 to 2009, managed funds beat the market only 13 times. Monkeys throwing darts to select stocks at random performed better than people who get paid six-figures to manage rich people's money. My intuition is that there is a certain obsession with complexity that economic and financial experts are infected with -- the more nuanced and sophisticated the predictions forwarded, the more ingenious their discovery is observed to be-- and the greater their perceived reputation. The more extensive and complex the chain of causality navigated to come to a conclusion the better, for its complexity insulates it from scrutiny while soliciting celebration of genius. And when the predictions turn out to be utterly incorrect, experts are not demonized and dethroned from their position, they’re instead called back to their pet news program to espouse even more absurd predictions.
A similar finding was observed by Philip Tetlock, a Berkeley professor of Psychology, who found that 'so-called' experts in a variety of subjects performed worse at forecasting events than average people. Tetlock found that the more specialized an individual becomes, the worse their predictions are; there is a point at which hyperspecialization yields worse predictive accuracy than general familiarization. The logic behind the argument should be familiar: imagine you're searching a massive online database to diagnose your own minor medical ailments (I can't be the only one, right?). Undoubtedly, the exhausting list of results that appear will be petrifying-- cancer, lupus, maybe even boneitis (what a funny name for a horrible disease!). In the same way, experts with robust conceptual knowledge have a difficult time delineating between random symptoms and real, systemic problems-- to them, everything seems systemic. Applying theoretical knowledge to empirical reality becomes increasingly difficult as a function of one's expertise; the more knowledge one has, the more chains of logic one can hastily pursue, and the further down the rabbit hole one gets.
I believe these biases can be largely attenuated with a robust understanding of economic history. As of current, economists are being educated with models and theories, instead of empiricism and history. At the very least, mandatory economic history curriculum will tame the overconfidence to which economists are so prone today.
Consider that the 1920s and the Great Moderation of the early 2000s shared an immense amount in common-- pernicious levels of inflation, rising levels of indebtedness, and high capital inflows from abroad. History suggests that booms are followed by speculation and risky behavior which invariably culminates in recession. Yet few economists foresaw the dissolution of the Great Moderation-- Ben Barnake even argued that financial innovation and rising housing prices insulated the United States from a shock.
In the book, “This Time It’s Different”, two famous economists detail with excruciating clarity and persuasion that, over the course of eight centuries, sixty-six countries and five continents, nearly every substantial economic crisis was handled with similar historical myopia. The authors demonstrate that economic crises follow the following storyline: 1) a human bias towards optimism ensures that the reality of economic problems are ignored until it is too late; 2) when the reality of the crisis comes into full view, experts declare that history is an irrelevant guide to policy-making, that “this time it’s different”; 3) solutions are prescribed based on abbreviated data-sets, political expediency, and obsession with theory.
If economists had a robust understanding of history and awareness of their own cognitive biases, I believe that forecasting would be imbued with much needed clarity and the problems we have today would be substantially minimized.
No matter the social science, forecasting is going to be hard: chaos theory dictates that any dynamic system with a large number of initial conditions will yield results that, though completely deterministic, are realistically unpredictable. Nevertheless, there are steps that should be taken to limit the amount of chaos introduced into the system.
First, economists need to refuse the impulse to treat their discipline as if its science. Human behavior is unbelievably complex and oversimplifying the macro-economy so that it can be modeled does more harm then good.
Second, economic history needs to be taught and reinforced in every school around the world. Too many economists are trying to change lanes without using a rear-view mirror, and such hubristic behavior has resulted in one too many crashes.
Third, the public should take economic forecasting with a grain of salt. The dangerous implications of forecasting will be minimized if society treated them like fortune cookies or horoscopes—harmless forms of entertainment.
Perhaps we are just big-brained apes, incapacitated by the limitations of our reptilian brains, but hopefully by being aware of our biases and prejudices, we can help economics move forward as a discipline.
— Evan DeFilippis, political science and economics and political science junior
The Oklahoma Daily is pleased to provide you the opportunity to share your thoughts about this article. We encourage lively debate on the issues of the day, but we ask you refrain from using profanity or other offensive speech, engaging in personal attacks or name-calling, posting advertising, or straying from the topic at hand. To comment, you must be a registered user of OUDaily.com. Thanks for taking the time to offer your thoughts.
You must be logged in to leave a comment. Log in | Register
Orpheus_Fiam 1 year, 1 month ago
I agree with this article, especially the second point the author makes. Perhaps if the public had a greater knowledge of economic history there would be more accountability, and economists would add a touch of realism to their theories.
demetrius 1 year, 1 month ago
Young Evan, This is a dangerous article for you to write in such a left wing student paper. Keynes did just what you argue against in his General Theory, and is accepted by the lefties. What you describe is an cough Austrian cough business cycle, accepted mostly by the Libertarians.
JKC 1 year, 1 month ago
The biggest point that, IMHO, should be taken from this column is the second suggestion to teach economic history. This will do wonders to shatter the narrow, unrealistic assumptions of economic theory. Also, it will help expose the infancy of the discipline. Adam Smith wasn't that long ago, and we haven't made many gains since. Learning the evolution of the theory from Smith to Keynes to Friedman will help in the last suggestion of taking forecasts with a grain of salt.
That being said, another tangential point is the idea that we should abolish forecasting all together. Trying to predict the beginning and ends of recessions with any degree of certainty or without doing so ex post facto is a pointless exercise. Maybe economics should limit itself to explaining things as they happen and/or as they happened in the past?
Orpheus_Fiam 1 year, 1 month ago
@JKC
I don't know that abolishing forecasting altogether is very practical. The market tends to be jittery enough, and forecasts whether good or bad give the markets more short term stability. If we took forecasts out altogether there is a much higher danger for people to overreact to both good and bad news. I realize forecasts are often incorrect, but if they help us avoid minor panics and bubbles which result in the hasty decisions of insecurity then I think we can afford to let the economists have their fun. That being said, the forecasts do need to be taken with a grain of salt if we are to avoid more serious economic disasters.
Mjmiller 1 year, 1 month ago
I appreciate the points made in this article. I definitely agree that we, in general, rely far too much on economic forecasts and predictions. Cut and dry formulas are much more comforting than the admittance that we cannot ever know precisely what will happen economically. Not only is it more comforting to base our economic decisions on an abstract formula - it is also easier: just pull out your money from this investment when formula XYZ reaches code blue and falls under -6.66%. What seems to be scientific and sophisticated decision making is really just blind adherence to the belief (hope) that people will act the way they are expected to act. Assuming that these massively complex formulas created by incredibly intelligent people have the abilities to predict human decision making that are just beyond our ken is like assuming the king is wearing clothes that we just cannot see. Sure, it would be nice to have a definitive formula that tells us our economic future, but as history has shown us, that magical formula is more like an alchemist's gold: misleading at its best; fraud at its worst.
Michael 1 year ago
Demetrius I think your interpretation is inaccurate. If the efficient markets hypothesis is wrong and cognitive biases promote shortsighted behavior that is irrational in the long run then government regulations are critical safe guards and the libertarian argument for laissez-faire policies is merely a recipe for future crises.