Theory and Evidence: How Economics Tests Ideas

How economists move from theories about behaviour to evidence about the real world, and why finding causes is the hardest part.

Economics has two main activities. One is building theories: simplified explanations of how people, firms and governments behave. The other is gathering empirical evidence: data about what actually happens. Good economics needs both, and the relationship between them is the subject of this module.

Why theory is not enough

Theories can point in different directions. Standard theory suggests a higher minimum wage might reduce jobs, because labour becomes more expensive. But another theory, about employers with power to set wages, suggests a modest increase might not reduce jobs at all. Only evidence can tell us which effect is larger in a particular place and time.

The central problem: causes

Most important questions in economics are about causal effects. Does a training programme raise earnings? Does a tax cut increase investment? Does more schooling make people richer?

To answer, we need to compare what happened with what would have happened without the programme or policy. That second, unseen situation is called the counterfactual. The fundamental difficulty is that we can never observe it directly. A person either went through the training programme or did not; we cannot see both versions of their life.

The training programme puzzle

People who complete a job training programme earn more afterwards than people who did not. Did the programme cause this? Maybe. But perhaps the people who signed up were more motivated to begin with, and would have earned more anyway. Simply comparing the two groups cannot separate the effect of the programme from the effect of motivation.

The credibility revolution

Since the 1990s, economics has placed much more emphasis on research designs that can convincingly identify causes. Economists Joshua Angrist and Jörn-Steffen Pischke called this the credibility revolution. The methods in this module, including randomised trials, natural experiments and cutoff-based comparisons, are its main tools. Several Nobel prizes in recent years have recognised researchers who developed or applied them.

Thinking data speaks for itself

Data does not explain itself. The same numbers can support very different conclusions depending on how they are analysed. The key question for any study is how it deals with the missing counterfactual.

Key takeaways

Economics combines theory, which explains behaviour, with evidence, which tests it.
Theories often point in different directions, so evidence is needed to judge effects.
Finding causal effects is hard because the counterfactual can never be directly observed.
The credibility revolution brought research designs that identify causes more convincingly.