Growing economic disparities between the industrialized West and underdeveloped countries and decreased costs of transportation have caused a spurt in immigration over the past few decades. Countries such as Canada, the U.S., Sweden, and Switzerland with more liberal immigration regimes have seen foreign-born populations expand into the double digits as a percentage of their total populations (OECD 2001 and author‘s calculations for Sweden). With this growth, however, have come calls from some quarters that countries re-examine their abilities to absorb large streams of immigrants (Briggs 1996).
Immigration affects host countries in a variety of ways–socioeconomically, demographically, spatially and culturally (Briggs 1996; Greenwood 1994; Simon 1989). These effects have been studied exhaustively through the past few decades using a variety of qualitative and quantitative approaches. However, most of these studies address historical conditions and patterns, while many of the questions being raised are of a —what if“ nature. Simulation, which can be used to examine the consequences of alternative immigration policies, is potentially one of the most useful tools for answering such questions but unfortunately is rarely used. The absence of large-scale economic-demographic models with detailed economic-population linkages has meant that few immigration policy handles are available for the social scientist to reliably simulate immigration.
One simulation method with the potential to shed considerable light on the effects of different immigration processes on recipient countries is microsimulation. Microsimulation was introduced over four decades ago by Orcutt (1957) and has experienced a revival in the social sciences over the past decade (Merz 1991; Clarke 1996; Isard et al. 1998; Williamson 1999). It has been used in national-level population projection studies (Fredriksen 1998), to investigate social security/pension contributions and benefits (Favreault and Caldwell 2000; Nelissen 1994, 1996, 1998; Andreassen, Fredriksen, and Ljones 1996; Zedlewski 1990), to examine the effect of various tax regimes on fiscal budgeting and inequality (Klevmarken and Olovsson 1996), to analyze support networks and retirement care needs as the population ages (Williamson 1996; Galler 1997a; Hancock 2000), to examine educational and health issues (Caldwell 1996; Harding 2000), to study wealth distribution (Caldwell, et al. 1998) and to assess housing policy (Oskamp 1995). Recent microsimulation modeling efforts have been made to examine spatial processes (Ballas and Clarke 1999; Caldwell, et al. 1998; Clarke 1996; Vencatasawmy et al. 1999).
The spatial dynamic microsimulation model (called SVERIGE or System for Visualising Economic and Regional Influences Governing the Environment) built at the Spatial Modelling Centre in Kiruna, Sweden, is one such model. It is the first national-level spatial microsimulation model available and permits analysts to study the spatial consequences of various national, regional, and local-level public policies. Assisting the model building effort is a unique database comprising longitudinal socio-economic information on every resident of Sweden for the years 1985 to 1995. The locations of the individuals in this database are given in co-ordinates accurate to the level of 100 meters. It is, therefore, possible to estimate behavioral equations on various geographical scales and to describe complex dynamic spatial relationships.
This paper describes some of the key components of the model and simulates the effect of several immigration policies in the Swedish context. The paper is divided into four sections. The first section describes the main features of microsimulation models. The second section outlines the SVERIGE microsimulation model structure, components, and unique characteristics. The third section explains how immigration is treated within the SVERIGE microsimulation model. The fourth section describes current Swedish immigration trends and issues and outlines the policy simulations that will be conducted. The fifth section presents empirical results and their implications. The paper ends with a summary and conclusion.
