New developments in the use of location quotients to estimate regional input-output coefficients and multipliers

This study compares the survey-based regional input-output coefficients and production multipliers published by STATISTICS FINLAND, 2000, with estimates obtained by applying location quotients (LQs) to national data. The consequences of using alternative adjustment formulae, the 'SLQ, CILQ and FLQ' are illustrated by an input-output model constructed for the Keski-Pohjanmaa (K-P) region. The results indicate that the SLQ and CILQ both produce highly misleading regional input-output coefficients and multipliers. These adjustment formulae are clearly not good enough for the purposes of making local policy and regional planning. The F L Q formula (beta = 1) yields much better regional input-output coefficients and multipliers than the SLQ and CILQ. The FLQ gives very good estimates for regional multipliers in nearly all industries. The difference between the multipliers generated by the FLQ and the survey-based K-P regional multipliers is on average about -0.3%. The multipliers for the K-P region are typically much lower than for Finland as a whole, indicating that the economic structure of the K-P region is dependent on that of other regions. Hence there is a need to make proper allowance for interregional trade. In the case of the K-P region, the FLQ with beta = 1 was able to offset the tendency of the CILQ to generate excessively large regional multipliers.