Fejlettség és egyenlőtlenségek: Magyarország megyéinek és kistérségeinek esete

According to the hypothesis of our study there is a close relationship between the standard of economic development and the scale of internal spatial disparities. Central in our applied theory is that the first phase of capitalist economic development is characterised with an initially fast and later more modest pace of polarisation affecting the formerly balanced spatial structure. In the more advanced economies it is followed by the period of diminishing regional segmentation. The scale of internal regional income-inequalities plotted against national development standard results a characteristic reversed U-shaped diagram. The hypothesis examined and supported by many researchers may be interpreted in two ways. The first interpretation is justified by the examination of time series: a country or a region covers this path in the course of its long term development. The second interpretation considers at a particular time the magnitude of regional disparity uniformly calculated for regions and countries on different levels of development. In this study, using basically the second approach we are looking for the connection between regional development and disparity for Hungarian counties and micro-regions. The period examined is the 15 years between 1990 and 2004, the data are for settlements, and the development standard is expressed by the taxable income of the population. The first part of the study reveals the internal income segmentation of the examined spatial units via two disparity indicators. Our findings suggest that there are considerable spatial differences in this respect both on the level of counties and micro-regions. The second part the paper focuses on the reasons generating these disparities by applying multi-variable regression models but only on the level of micro-regions. We have worked out three model-types. In the first model we are testing the effect of income levels on internal disparities without involving any other control-variables. In the second model-type the circle of independent variables is expanded by involving some basic geographical characteristics of the micro-regions. In the third-type empty control-variables of technical nature influencing the circumstances of calculation enter the model. Our results refer to the fact that the relationship between the income level and the scale of disparities - suggested by our initial hypothesis - can be detected on the level of micro-regions as well. This relationship has been strengthening since 1990, however its impact on spatial segmentation is more modest on the level of micro-regions than in the case of the larger spatial units such as the counties or the countries. One of the reasons for this phenomenon is the wide variety of "measuring technique variables" (ex. number of settlements, character of settlement-structure) expressed in the values of disparity variables, distorting or hiding the real correspondences. Most probably the relatively poor explanatory force of the regression models on the micro-regional level has another set of reasons: namely that on this level the local factor markedly appears.