ABSTRACT
In the early 1930s there was a lively academic and policy debate in the Netherlands on exchange rate policy. Although the UK had left the gold standard in 1931, the Dutch government was greatly in favour of having the guilder remain fixed to the gold standard. Those opposed argued that leaving the gold standard would mean devaluation which would enhance the competitive position of the country and would therefore be beneficial to employment. Tinbergen’s model (1936) was especially built to investigate these policy questions. So even at the dawn of model-building there was interaction from policy questions to model formulation and usage. Since then, one of the most obvious ways in which model and policy interaction goes on is in the use of models to analyse the effects of possible policy actions and thus act as a substitute for natural experiments (or even the controlled experiments available in some natural sciences).
