The Regrettable Loss of Mathematical Molding in Econometrics - Mechanism and Causality in Biology and Economics - page 65

. Thus, by assuming that
purchasing power is dependent on economic activity, Tinbergen was able to extend
the period compared with the production lag and thus arrived at a more realistic
business-cycle mechanism.
In his search for appropriate business-cycle mechanisms, Tinbergen ( 1931 )
found an even better example than the “pork cycle”: the shipbuilding cycle. This
mechanism, a combined lag and cumulative relation, showed how a lag of 2 years
could generate a cycle of 8 years:

For this scheme, the cycle period was equal to 2.7 θ

X

ð t Þ¼ aX ð t θÞ (4.6)

where X represents world tonnage and the average needed time to build a ship,

approximately 2 years. The parameter a was a constant value between ½ and 1. The
cycle generated by this mechanism has a period equal to 4

θ

8 years.
In a survey on “quantitative business-cycle theory,” Tinbergen ( 1935 ) outlined
systematically the criteria for an appropriate business-cycle theory: “The aim of
business cycle theory is to explain certain movements of economic variables.
Therefore, the basic question to be answered is in what ways movements of
variables may be generated” (p. 241). The core of the business-cycle theory was a
“mechanism” that he defined as a “system of relations existing between the
variables; at least one of these relations must be dynamic. This system of relations
defines the structure of the economic community to be considered in our theory.
Such a mechanism may perform certain kinds of swinging movements that are
characteristic of the system as such” (pp. 241-2). A mechanism, according to
Tinbergen, is a specific set of structural relations that together explain the business
cycle. Tinbergen emphasized the distinction between the mathematical form and
the economic meaning of the equations:
The mathematical form determines the nature of the possible movements, the economic
sense being of no importance here. Thus, two different economic systems obeying, how-
ever, the same types of equations may show exactly the same movements. But, it is evident
that for all other questions the economic significance of the equations is of first importance and
no theory can be accepted whose economic significance is not clear. (Tinbergen 1935 ,p.242)

θ ¼

Mathematical molding was an essential element of Tinbergen's business-cycle
research in the 1930s. Economic theories did not contain any guidelines that could
lead to an appropriate formalism. They either were verbal accounts or, if mathe-
matical, only gave descriptions of static systems. Mathematical molding was a trial-
and-error process that started with the assumption of a production lag. As Hanau
( 1928 ) showed empirically and Aftalion ( 1927 ) theoretically, lags generate endog-
enous fluctuations. However, basing dynamics on a production lag alone has several
disadvantages. In the first place, as discussed above, to explain a Juglar, the
assumed production time would have to be far too long. This was the reason why
Tinbergen introduced various “complications” into his dynamic schemes. In the
second place, the disadvantage of postulating lags is that they must be stated in
advance and have a fixed length. “This has been repeatedly felt as a too rigid