Why Economics Cannot Be Physics
A recent essay titled "How to Make Economics a Science" argues that economics fails to meet the criteria we expect from genuine sciences. The author proposes six attributes that define scientific status and measures economics against them. His conclusion is that economics could become a science if it adopted better methods, abandoned failed theories, and embraced heterodox approaches like Modern Monetary Theory (MMT).
The essay is ambitious and raises legitimate concerns about the discipline. But it suffers from a fundamental problem: the author never asks whether economics could meet his criteria given what it actually studies. He assumes economics can function like physics or chemistry if it simply tries harder. This assumption deserves scrutiny.
The Nature of the Subject Matter
Physics studies objects that do not choose. When a physicist drops a ball, the ball does not decide to fall. It does not weigh alternatives, consider its preferences, or change its mind halfway down. The regularities physics discovers are genuine laws. The ball will fall the same way tomorrow, next year, in another country, under identical conditions.
Economics studies something categorically different: human action. When someone buys a house, they are making a choice based on preferences, expectations, available information, and countless contextual factors. Another person facing identical external circumstances might choose differently. The same person might choose differently tomorrow if their preferences shift or their expectations change.
This distinction matters enormously for what kind of knowledge is possible. Ludwig von Mises made this argument decades ago, and it remains underappreciated. The methods appropriate for studying physical phenomena may not be appropriate for studying purposeful human behavior. The essay never engages with this possibility.
The Falsifiability Problem
The essay repeatedly invokes Karl Popper's falsifiability criterion. Scientific theories must make predictions that can be tested. If the predictions fail, the theory should be abandoned or revised. He criticizes mainstream economics for clinging to theories despite failed predictions.
But consider what falsifiability means in physics versus economics.
In physics, falsification is relatively clean. If I predict that a ball dropped from a certain height will hit the ground in a specific time, and it does not, something is wrong with my theory or my measurement. The prediction is precise, the test is clear, the outcome is unambiguous.
In economics, falsification is far messier. Suppose I predict that raising the minimum wage will increase unemployment among low-skilled workers. The wage is raised, and unemployment does not increase. Have I falsified the theory?
Not necessarily. Perhaps demand for labor increased simultaneously due to economic growth. Perhaps employers reduced hours instead of jobs. Perhaps they cut non-wage benefits. Perhaps they raised prices and consumers absorbed the cost. Perhaps the wage increase was small relative to existing wages in the affected markets.
The problem is not that economists are stubborn or unscientific. Human choices depend on subjective valuations, expectations, and contextual factors that cannot be held constant or directly observed. The ceteris paribus assumption that underlies economic reasoning can never be fully satisfied in the real world because we cannot control for everything that influences human decisions.
This is not a technical limitation that better methods will overcome. It reflects the fundamental nature of the subject matter.
The Prediction Problem
The essay criticizes mainstream economists for failing to predict the 2008 financial crisis. He praises heterodox economists like Steve Keen and Hyman Minsky who, he claims, saw it coming.
But this framing accepts a premise worth questioning: that economics should be in the prediction business the way astronomy or meteorology is.
An astronomer can predict eclipses to the minute because celestial mechanics do not involve choice. The moon does not decide whether to pass between the earth and sun. Given initial conditions and physical laws, the outcome is determined.
Economic phenomena are different. A financial crisis occurs when enough market participants simultaneously revise their expectations and change their behavior. When will that happen? It depends on future human choices that have not been made yet. No model can predict when people will collectively decide that asset prices are unsustainable, because that decision depends on information, beliefs, and psychological states that do not yet exist.
Steve Keen began warning about debt-fueled instability in the early 2000s. He was eventually vindicated in 2008. But if you predict crisis every year, you will eventually be right. The question is whether your framework generates consistently accurate predictions about timing and magnitude. By that standard, heterodox approaches have not demonstrated clear superiority.
This is not a defense of mainstream economics. It is a challenge to the assumption that any economic framework can reliably predict complex phenomena that depend on millions of individual choices.
The Scientism Trap
Ironically, the mainstream economics the essay criticizes became what it is precisely because economists tried to make their discipline scientific in the way he demands.
The desire for rigor pushed economics toward mathematical formalism. Equations look scientific. They can be manipulated, tested, published in prestigious journals. But the mathematics often obscures rather than illuminates because the assumptions required for tractability bear little resemblance to how people actually behave.
Rational expectations, representative agents, and equilibrium assumptions did not emerge from empirical observation. They emerged from the need to make models solvable. Economists chose assumptions that permitted elegant mathematics over assumptions that captured messy reality.
Dynamic Stochastic General Equilibrium models, which the essay rightly criticizes, exist because economists wanted rigorous, testable frameworks comparable to those in physics. The result is sophisticated machinery that excludes banks, debt dynamics, and financial fragility because including them makes the mathematics intractable.
The pursuit of scientific status, measured by the standards of physics, led economics away from understanding the phenomena it studies. The essay's prescription is more of the same: better models, clearer falsification conditions, convergence on settled questions. But this may be exactly wrong. The solution to physics envy is not better physics imitation.
What Economics Can Actually Do
If economics cannot be a predictive science like physics, what can it be? This question deserves a serious answer.
Economics can identify logical relationships. Minimum wages set above market-clearing levels create surplus labor, all else equal. Inflation transfers wealth from creditors to debtors. Price ceilings create shortages. Price floors create surpluses. These are not predictions about specific magnitudes at specific times. They are logical deductions about the consequences of actions given the nature of human choice.
Economics can trace causal mechanisms. How do interest rate changes propagate through an economy? What happens when governments finance spending through money creation versus taxation? How do trade restrictions affect domestic industries and consumers? Economics can map these processes even when it cannot predict their precise outcomes.
Economics can identify tradeoffs. This may be its most important function. Every choice involves giving up something to get something else. Spending resources on one project means not spending them on another. Protecting one industry imposes costs on consumers and other industries. Economics illuminates what those tradeoffs are, even when it cannot tell you which choice to make.
Economics can expose fallacies. The broken window fallacy reveals how destruction does not create prosperity despite appearing to create jobs. The lump of labor fallacy shows why employment is not a fixed quantity to be divided. Mercantilist fallacies about trade persist despite centuries of refutation. Economics has genuine insights about how intuitive reasoning goes wrong.
These contributions are valuable even though they do not resemble the predictive power of physics. The question is whether we judge economics by appropriate standards or hold it to standards it cannot meet.
The Tradeoff Problem the Essay Ignores
The essay is remarkably un-economic in its reasoning. He argues that mainstream economics misunderstands monetary constraints, that governments can spend more freely than orthodoxy claims, that deficits do not impose the limits commonly assumed.
Set aside whether Modern Monetary Theory's operational description is correct. Even if governments do spend first and tax later, even if sovereign currency issuers cannot "run out" of money in a technical sense, this does not eliminate tradeoffs. It relocates them.
The constraint becomes inflation rather than nominal budget balance. If government creates money and spends it while productive capacity remains constant, prices rise. The real resources used for government projects are not available for private uses. Workers employed in public programs are not available for private employment.
The essay acknowledges this in passing: "The real question is not whether we can create the money but whether we have the real resources to accomplish the goal without causing inflation." But this is the entire question, and it is treated as an afterthought.
Economics is fundamentally about scarcity and choice. Human wants exceed available means. Every choice to do one thing is a choice not to do another. This remains true regardless of your monetary theory. Sophisticated arguments about how money is created do not change the reality that resources are finite and uses compete.
The essay's policy implications may or may not be correct, but they cannot be derived from monetary mechanics alone. Whether we should spend more on healthcare, infrastructure, or anything else requires showing that the benefits exceed the costs, including the opportunity costs of resources diverted from other uses. That is the economic question. Monetary theory, however sophisticated, does not answer it.
The Persistence of Schools
The essay treats the persistence of competing schools of economic thought as evidence that economics is not scientific. Physics resolved its foundational disputes. Economics has not. Therefore economics is not a science.
But another interpretation is available. The schools persist because they emphasize different aspects of an irreducibly complex phenomenon.
Austrian economics emphasizes individual choice, subjective value, and the impossibility of central planners possessing dispersed knowledge.
Keynesian economics emphasizes aggregate demand, coordination failures, and how individual rationality can produce collective irrationality.
Monetarism emphasizes the relationship between money supply and prices.
Institutional economics emphasizes how rules, norms, and organizations shape behavior.
These are not necessarily competing theories about the same thing in the way that Newtonian and Einsteinian physics compete. They may be different lenses on a phenomenon too complex for any single framework to capture.
Physics can unify because particles do not have preferences. They do not form expectations about the future. They do not respond strategically to the predictions of physicists. Human beings do all of these things. A complete economic theory may be impossible. The subject matter resists the kind of unification physics achieves, and no amount of rigor can overcome this.
The Heterodox Double Standard
The essay applies rigorous standards to mainstream economics while giving heterodox approaches a pass.
He criticizes mainstream models for failing to predict the 2008 crisis. But Minsky's Financial Instability Hypothesis does not predict when crises will occur or how severe they will be. It predicts that stability breeds instability in general. By this standard, any crisis confirms the theory and periods without crisis simply represent the stability phase preceding future instability. This is not more falsifiable than the mainstream theories the essay criticizes.
He praises Modern Monetary Theory for making "scientific claims that permit empirical assessment." But when critics point to currency crises in countries like Turkey or Argentina, MMT proponents respond that these countries do not truly meet the criteria for sovereign currency issuers. This is the same immunizing strategy the essay attributes to mainstream economics.
A genuinely scientific assessment would apply identical standards to all frameworks. The essay's criteria are constructed to favor its preferred approaches while disadvantaging mainstream economics. This is advocacy dressed as methodology.
What Genuine Humility Looks Like
The honest conclusion is not that economics should try harder to be like physics. It is that economics studies something fundamentally different from physics and must accept appropriate limitations.
Economic reasoning can tell us that actions have consequences, that tradeoffs exist, that incentives matter, that prices convey information. It can help us think clearly about complex phenomena. It can identify logical errors in popular reasoning.
What it cannot do is predict with precision, settle all disputes empirically, or provide the kind of certainty we expect from physical sciences. This is not a failure of the discipline. It is a reflection of subject matter that involves choosing beings whose choices depend on factors that cannot be fully observed or controlled.
The essay wants economics to converge on settled answers the way physics does. But physics studies a world without choice. Economics studies a world constituted by choice. The methods appropriate for one may not be appropriate for the other.
The solution is not better models or more rigorous falsification. The solution is appropriate humility about what economic reasoning can and cannot accomplish. Economics is valuable despite its limitations. Pretending those limitations can be overcome through methodological improvement sets the discipline up for perpetual failure measured against inappropriate standards.
What Economics Can Be
"How to Make Economics a Science" raises legitimate concerns about the state of economic knowledge. Mainstream models failed conspicuously in 2008. Replication rates are embarrassingly low. Foundational disputes persist after more than a century. Economists sometimes present ideological commitments as scientific findings.
But the essay's proposed solution misunderstands the problem. Economics cannot become physics because it does not study physical phenomena. It studies human action, which involves choice, preference, expectation, and interpretation. These features of the subject matter impose inherent limits on what economic knowledge can look like.
The question is not how economics can become a science like physics. The question is what kind of knowledge we can have about human action and how we acquire it. That is a question the Austrian tradition has engaged seriously. The essay, despite its length and apparent sophistication, never gets there.
Economics is messy because people are messy. The solution is not better models. The solution is appropriate humility about what economic reasoning can and cannot tell us.