Factor Price Equalization in Finland

The Heckscher-Ohlin trade model leads to clear conclusions on the absolute and relative factor prices in a two-commodity specification of the model where both commodities are produced and factors can move freely within the economy. Even though the tests on factor price equalization fail in international comparisons, regional approach offers a prospective way of characterizing how factors of production, e.g. skilled and unskilled labor are utilized in optimally behaving markets and which potential reasons are behind deviating from the ideal situation. Within one economy, Finland, where regions are well integrated nationally by goods trade and factor mobility, we test factor price equality between regions. Robust tests derived from Heckscher-Ohlin trade model developed by Bernard and Schott (2001) and further by Bernard et al. (2002) are used for testing the absolute and relative equalization between factor prices. As a distinction to absolute equalization, the relative equalization allows regional differences in the total factor productivity. The tests are robust to unobserved regional productivity differences, unobserved regional factor quality differences and variations in production technology across industries. Regional relative factor prices can also be used for studying regional industry mix (different cones of diversification) that can vary between regions but are in accordance with the theory. These different industry mixes on the other hand imply different skill-premium to prevail in different regions depending on the availability of endowments. Large supply of skilled labor will result in lower relative wages for skilled labor. Tests are applicable to any pairs of factors and do not necessarily need information e.g. on capital inputs. We compare the results with similar regional studies conducted in the United States and the United Kingdom. Compared to flexible labor markets in the U.S. and small but densely populated United Kingdom, the differences are expected to be even larger in Finland, since the country is a sparsely populated but relatively large country. Markets are well integrated in Finland, though, and common regulations prevail throughout the country. The data used for Finland deviates from the reference studies. We utilize Population Census of employed labor force and concentrate the analysis on the manufacturing sector between 1987 and 1999. We start with one cross-section and then extend the analysis to several years. The data includes 135 000 observations based on employment relations. Education level is used for dividing the labor force into skilled and unskilled labor force. The data is connected with the information on provincial identification as well as the industry coding of the employer. The country is divided to 15 employment regions and the employer coding follows 3-digit SITC classification. Both in the U.S. and the UK the hypothesis on the same relative factor prices are widely rejected in the literature. As an explanation there may be varying industry mixes, region-industry technology differences, agglomeration and increasing returns to scale. For Finland we aim at estimating possible regionally varying effects of joining to European Union and are regions differentially exposed to pressures from increased international trade.