Methodologies of positive economics - The measurement-without-theory debate

8 important questions on Methodologies of positive economics - The measurement-without-theory debate

Between who was the measurement-without-theory debate?

Cowles Committee (Koopmans) and the NBER (Vining)

What is the Cowles Commission?

Emperical research based on theory:
1. A priori mathematical model
2. Estimated with equations
3. Hypothesis testing

What is the empirical statement of the NBER?


"Theory is the result of observation"
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What was the statement of Koopmans?

"Theory is the input and outcome of observation".
Koopmans believed that economists are unable to conduct experiments to an economic system as a whole (so it is hard to seperate causes from consequences) (simultaneous equation bias)
- There are so many emperical relations and these relations can only be identified if the set of variables involved in each equation is specified in economic theory.
- Invariance problem; does it hold in the future?

What critique did Vining (NBER) give to Koopmans?

Descriptive statistics are useful to discover hypotheses
Cowles Comission's model is too simple to capture the diversity which the data seems to explain.

What were the three arguments to explain the implicatoins and limitations of the NBER's empiricist position?

1. For the purposte of systematic and large-scale observation of a many-sided phenomenon (such as the business cycle), theoretical assumptions about its nature cannot be disregarded, and the authors do so only at the expense of the analysis

2. The NBER provides no evidene for the assumption that the empirical relationships found would be eternal over time.

3. Statistical analysis of the data require additional assumptions about their probabilistic characteristics that cannot be subject to statistical testing from the same data. 

What was Cowles Comission's solution to the problem of incompleteness?

Problem of incompleteness: some variables excluded for empirical reasons may later prove to be importannt.

Solution: build ever more extensive models, which should include as many potential influences as possible.

What were the problems with econometrics?

1.  There are no laboratories to apply science to
2. One will need to use more variables to get more certainty and to have less chance of omitted variable bias, but this leads to a more complex model. (The NBER and Tinbergen)
3. Koopmans and Haavelmo: start with a small model based on theory and expand the model if necessary.

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo