In the 1930s, economics was at a crossroads. While physicists were refining their theories through rigorous experimentation, economists were building elaborate theoretical systems that seemed disconnected from the real world. Then came a young British economist named Terence Hutchison, who dared to ask a provocative question: Can economics truly be called a science if its theories can’t be tested?
Table of Contents
The radical call for testability
When Terence Hutchison published The Significance and Basic Postulates of Economic Theory in 1938, he launched what would become one of the most influential critiques of economic methodology in the twentieth century. Hutchison attempted to bring logical positivism into economics, arguing that economic theory had become little more than a collection of empty statements without empirical content.
His central thesis was straightforward but controversial: for economics to qualify as a genuine science, its propositions must be empirically testable and, crucially, conceivably falsifiable. This meant that economic theories needed to make predictions that could potentially be proven wrong through observation and evidence. Without this quality of testability, Hutchison argued, economics rested on a hypothetical-deductive edifice that remained a body of empty tautologies.
Think of it this way: If I claim “it will either rain tomorrow or it won’t rain tomorrow,” I’ve made a statement that’s technically true but tells you absolutely nothing useful. Hutchison worried that much of pure economic theory had become like this-technically consistent but empirically meaningless.
The ceteris paribus problem
At the heart of Hutchison’s critique was his challenge to how economists used the phrase “ceteris paribus”-Latin for “all other things being equal.” This assumption is ubiquitous in economics. When we say “an increase in price leads to a decrease in quantity demanded, ceteris paribus,” we’re assuming that nothing else changes: consumer income stays constant, preferences don’t shift, competitor prices remain stable, and so on.
The problem, as Hutchison saw it, was that economists had become irresponsible in their use of this clause. He contended that the overuse of the ceteris paribus clause rendered microeconomic theories effectively unfalsifiable, stripping them of meaningful empirical content. If a theory’s predictions failed to materialize, economists could simply claim that some “other thing” had changed, saving their theory from refutation.
Why this matters in practice
Consider a real-world example. Suppose an economist predicts that raising the minimum wage will reduce employment, ceteris paribus. When employment actually rises after a minimum wage increase, the economist might say: “Ah, but consumer demand also increased,” or “immigrant workers entered the market,” or “technological changes occurred.” The ceteris paribus clause becomes a convenient escape hatch, making the original prediction impossible to falsify.
Hutchison argued in his 1938 work that generalizations in economics are qualified by a ceteris paribus clause, yet its interpretation remained controversial. The clause had transformed from a useful simplifying assumption into a shield against empirical scrutiny.
The Hutchison-Machlup showdown
Hutchison’s insistence on testing both a theory’s predictions and its basic assumptions sparked one of the most famous debates in economic methodology. In 1956, he published a critique of Fritz Machlup’s approach to verification in economics, igniting an intellectual confrontation that would define methodological discussions for decades.
What should we test?
The core disagreement centered on a deceptively simple question: Do we need to test the assumptions that underlie economic theories, or is it sufficient to test only the predictions those theories generate?
Fritz Machlup, following a tradition that would later be championed by Milton Friedman, argued that assumptions didn’t need to be realistic or directly testable. What mattered was whether the theory’s predictions matched observed outcomes. If the predictions worked, the assumptions could be treated as useful fictions or “as if” statements. From this perspective, it didn’t matter if the assumption that firms always maximize profits was literally true, as long as theories built on that assumption generated accurate predictions about firm behavior.
Hutchison vehemently disagreed. He maintained that all statements in a scientific theory should be testable in principle. Hutchison advocated for assumptions based on observed behavior, not on convenient mathematical simplifications or introspective hunches about human nature. For Hutchison, testing only predictions while ignoring unrealistic assumptions was methodologically unsound-it allowed bad theory to persist as long as it happened to generate correct predictions by accident or coincidence.
The deeper philosophical divide
This wasn’t just a technical disagreement about research methods. It reflected a fundamental philosophical split about what economics should be. Machlup represented the view that economics was primarily a deductive science, built from general principles through logical reasoning. Testing was important, but it applied mainly to the conclusions drawn from those principles.
Hutchison, influenced by logical empiricism and the scientific method as practiced in natural sciences, believed economics should be an empirical-historical discipline. Economic propositions needed to be grounded in observable phenomena from the start, not just at the level of final predictions. This approach demanded that economists pay closer attention to how businesses actually make decisions, how consumers actually behave, and how markets actually function-rather than building theories from idealized assumptions.
Why this debate still matters
You might wonder: Why should we care about an academic dispute from the 1950s? The answer is that these questions remain strikingly relevant today. Modern economics continues to grapple with the tension between elegant mathematical models and messy empirical reality.
The 2008 financial crisis, for instance, exposed how economic models built on unrealistic assumptions-perfectly rational agents, efficient markets, normally distributed risks-failed catastrophically when those assumptions didn’t hold. Critics pointed out that economists had known these assumptions were unrealistic but had proceeded anyway, much like Machlup suggested they should, focusing on whether predictions worked in normal times while ignoring the flawed foundations.
Behavioral economics has partially vindicated Hutchison’s approach by systematically testing the psychological assumptions underlying economic theory. When researchers discovered that people don’t actually calculate utility functions in their heads or exhibit perfect self-control, they didn’t just note that predictions sometimes failed-they built new theories on more realistic assumptions about human behavior.
The legacy of logical empiricism in economics
Hutchison’s contribution extended beyond this single debate. His 1938 book made the case for economics as an empirical science, drawing on logical empiricist authors while also incorporating Karl Popper’s falsifiability criterion. This synthesis helped introduce standards of scientific rigor to a discipline that had often relied on pure reasoning and introspection.
Though he earned the label of “ultra-empiricist” from critics like Machlup, Hutchison’s position was more nuanced than this caricature suggests. He wasn’t opposed to abstract reasoning or theoretical work. Rather, he insisted that such work needed to maintain contact with empirical reality at every stage-not just in the final predictions but in the fundamental assumptions as well.
Today’s econometric revolution, with its emphasis on randomized controlled trials, natural experiments, and careful identification strategies, owes an intellectual debt to Hutchison’s insistence on testability. When modern economists design field experiments to test whether development aid improves outcomes, or when they use regression discontinuity designs to evaluate policy effects, they’re following the empirical path Hutchison advocated: making economics testable, falsifiable, and scientifically credible.
What do you think? Should economists focus primarily on whether their predictions are accurate, even if built on unrealistic assumptions? Or does scientific credibility require that we also test and validate the fundamental assumptions underlying our theories? How might the choice between these approaches affect economic policy recommendations?
Leave a Reply