Migration in the Israeli Urban System.

Abstract

Previous studies have investigated the factors associated with different types of urban structure. This study takes the initial structure of a particular urban system as given, and then examines the effect on interurban migration of changing that structure. The setting is Israel, where a unique experiment has been carried out to change the structure of the urban hierarchy from primacy to rank-size distribution, by creating new towns all over the country. The data come from the 1961 census of Israel, pertaining to net migration between 1956 and 1961, and from surveys of internal migration for each year from 1965 to 1968. The analysis is in two parts. First is an examination of the matrices in which the migrants in each period are classified according to their previous and new settlements. The gravity model is used to generate expected migration streams, which are taken to embody that part of the overall pattern of interurban migration that is due to the current size-distance configuration of settlements. These expectations are factored out by division, and attention is focused on the residuals. The second part of the analysis considers migration in its role as a stimulus to the growth of the receiving settlements, which are viewed as being exposed to the "risk" of experiencing growth under the impetus of in-migration. A rate of in-migration is calculated for each settlement, based on its own population and the number of in-migrants from all other settlements. From the point of view of each settlement, this rate indexes the overall pattern of interurban migration. A model is formulated in which this rate is expressed as a function of (a) the effect of current gravitational attraction between places, and (b) the carry-over effect of earlier government action to change the structure of the urban hierarchy. Gravitational attraction is indexed by the population potential of each place in the urban field as a whole. Prior government action is indexed by a ratio indicating how much the size of each settlement deviated from the rank-size expectation at an earlier date, since the goal of policy was to eradicate such deviations. The argument draws on the inertia of systems. The model is estimated by means of multiple linear regression. Main conclusion: after taking account of the effect of the current size-distance configuration of settlements, the pattern of migration in the Israeli urban system during the mid- to late-1960's showed a carry-over effect from earlier government decisions concerning urban development. Also, there was an independent effect attributable to current policy, as expressed through the settlement of new immigrants from overseas. In sum, migration within the system did serve as an instrument for adjusting the structure of the urban system to fit the rank-size model to which the Dispersion of Population policy was committed. Evidently, government initiatives effectively structured information and transaction costs for migrants in such a way as to induce movement to target settlements. There is no evidence of any deliberate attempt to direct newly arriving Orientals disproportionately to settlements with already high concentrations of Orientals. However, the policy of only providing assistance to those immigrants who were willing to make their first settlement in the new towns, combined with the greater initial dependence of Orientals on government assistance, worked to concentrate the Orientals even further in settlements with high concentrations of similar persons. They then tended to circulate among such settlements at high levels. Thus they tended to be segmented away from the rest of the urban system, and perhaps thereby from the mainstream of Israeli society.