Measurement of the Effects of Regional Policy Instruments by Means of Linear Structural Equation Models and Panel Data
Two kinds of problems have generally been encountered in measuring the effects of instruments of regional policy. The first concerns the construction of an adequate method, which should (a) take explicitly into account the multidimensional character of a regional profile, (b) disentangle policy effects from autonomous developments, (c) make it possible to compare the effects of different policy instruments, and (d) be capable of dealing with theoretical constructs and measurement errors. The second problem relates to the scarcity of regional data, both concerning time-series and cross-sectional data. A linear structural equation model is presented which meets the first set of conditions. This model is then modified in such a way that it can be estimated by pooling cross-sectional and time-series data, thus increasing the number of observations. In addition, as an application the effects of regional industrialization policy on investments in the Netherlands during the period 1973–1976 are estimated.