Residential Mobility in the United States and the Federal Republic of Germany

  • Wolfgang Schneider
  • Konrad Stahl
  • Raymond J. Struyk


Casual evidence of the kind reviewed in the introduction indicates that the residential mobility rates of West Germans are sharply lower than those of Americans. 1 Such an outcome is expected for a number of reasons, which, for ease of exposition, can be divided broadly among transactions costs, search costs, and factors outside the housing market.


Housing Market Residential Mobility Household Type Mobility Rate Tenure Status 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    West German mobility rates also appear to be lower than those of several other industrial countries, including England and Japan. See L. H. Long and C. Boertlein, “The Geographic Mobility of Americans: An International Comparison” (Washington, D.C.: U.S. Bureau of the Census, Current Population Reports, series P-23, no. 64, 1976 ).Google Scholar
  2. 2.
    However, lower turnover implies fewer units on the market only if the length of time that units remain on the market is roughly equivalent. Some very rough data suggest, in fact, that units in the United States experience lower vacancy durations than those in West Germany. Thus, the broad statement in the text seems correct.Google Scholar
  3. 3.
    P. Rossi, Why Families Move (New York: The Free Press, 1955 ), and A. Speare, Jr., S. Goldstein, and W. H. Frey, Residential Mobility, Migration, and Metropolitan Change ( Cambridge, Mass.: Ballinger Books, 1974 ).Google Scholar
  4. 4.
    For a review see J. Quigley and D. Weinberg, “Intra-Metropolitan Residential Mobility: Review and Synthesis,” International Regional Science Review, vol. 2 (1977), pp. 41–66.CrossRefGoogle Scholar
  5. 5.
    J. MacMillan, Mobility in the Housing Allowance Demand Experiment (Cambridge, Mass.: Abt Associates, Inc., 1980 ).Google Scholar
  6. 6.
    For a formulation of this type in analyzing the moving behavior of renters in the Housing Allowance Demand Experiment, see E. A. Hanushek and J. Quigley, “An Explicit Market of Intra-Metropolitan Mobility,” Land Economics, vol. 54 (1978), pp. 411–29.CrossRefGoogle Scholar
  7. 7.
    F. Cronin, “Search and Residential Mobility: Part I, Economic Models of the Decisions to Search and to Move among Low-Income Households” ( Washington, D.C.: The Urban Institute, 1980 ).Google Scholar
  8. 8.
    Alternatively speaking, the higher the transactions costs of moving, the longer the expected stay in a unit, and the more we can expect postmove consumption patterns to reflect future events, such as increases in household size.Google Scholar
  9. 9.
    J. L. Goodman, Jr., “Linking Local Mobility Rates to Migration Rates: Repeat Movers and Place Effects,” in W.A.V. Clark, ed., Modelling Housing Market Search ( London: Croom Helm, 1983 ).Google Scholar
  10. 10.
    Income has other obvious effects on mobility as well. Higher-income households will find the cost of search greater because of their higher opportunity cost; on the other hand, they will also have greater resources with which to pay for the out-of-pocket expenses associated with moving.Google Scholar
  11. 11.
    For details about the sampling procedures see M. Nourney, Stichprobenplanudes Mikrozensus ab 1972, vol. 11 ( Bonn: Wirtschaft and Statistik, 1973 ), pp. 631–38.Google Scholar
  12. 12.
    The questionnaire and other details are in Statistisches Bundesamt Wiesbaden, 1%—Wohnungsstichprobe, vols. 1–5, Bautaetigkeit and Wohenungen, ( Wiesbaden, 1981 ).Google Scholar
  13. 13.
    For details see, for example, U.S. Bureau of the Census, Annual Housing Survey: 1978, Part A, General Housing Characteristics for the United States and Regions ( Washington, D.C.: U.S. Government Printing Office, 1980 ).Google Scholar
  14. 14.
    In a mobility study it would be very helpful in analyzing the causes for relocating to know household characteristics and dwelling unit and neighborhood attributes before and after a move. The West German sample does not contain all this information. Nearly all characteristics sampled are for conditions at the time of the interview, with a few retrospective questions asked. Because there is no comparable data for the Federal Republic of Germany, the information on previous residence available in the Annual Housing Survey file was not used.Google Scholar
  15. 15.
    In the Annual Housing Survey, missing or inconsistent data were corrected or added from a previous similar observation. Such allocations are marked, so we deleted those cases.Google Scholar
  16. 16.
    In addition, other difficulties would arise when comparing the incomes as reported, which are monthly net income (tax and social insurance fees deducted) for West Germany and yearly gross income for the United States. Boersch-Supan has nevertheless done some calculations of this type, and they are reported in chapter 3 of this volume.Google Scholar
  17. 17.
    In fact, average population density in West Germany is about ten times as great as that in the United States.Google Scholar
  18. 18.
    Bundesforschungsanstalt fuer Landeskunde und Raumordnung, Informationen zur Raumentwicklung,vols. 11–12 (Bonn, 1981).Google Scholar
  19. 19.
    We emphasize that we are comparing homeownership rates in the “free market.” Had we included occupants of subsidized rental housing units, the homeownership rate of West Germans would fall to 36.8 percent and that of the Americans to 64.1 percent, and Americans would be 74 percent more likely to be homeowners.Google Scholar
  20. 20.
    The technical aspects are summarized in the appendix to this chapter.Google Scholar
  21. 21.
    Since each household is either a mover (homeowner) or not, the “to-be-a-mover” (homeowner) distribution is binomial. Its variance varies with the expected value of being a mover (homeowner), which itself is estimated by the observed mobility (homeownership) rate. These variances and the number of original (unweighted) observations constitute the basis for the estimation of the standard errors in the estimation of log odds of the mobility (homeownership) rates for each stratum. These standard errors differ substantially across strata. They are (due to higher precision) smaller, the greater the number of households in the stratum and they are greater with rates near zero percent and those near 100 percent. For further information on this standard error see H. Theil, Principles of Econometrics (New York: John Wiley and Sons, 1971), p. 636. On regression procedures under heteroscedasticity see ibid., p. 244; and A. S. Goldberger, Econometric Theory (New York: John Wiley and Sons, 1965), p. 235. See also the appendix to this chapter.Google Scholar
  22. 22.
    There are eight dummy variables for household type in the owner and renter mcidels and seven in the pooled model. It was not possible to “force” the final variable into the estimated model because of its lack of statistical significance.Google Scholar
  23. 23.
    To assure reliability we use only strata with enough original household observations (i.e. ten in the separate models for homeowners and fifteen in the pooled models). To avoid infinity-problems with the “log odds” we eliminated all strata in which all households belong to only one category (of renters or owners, and movers or non-movers, respectively).Google Scholar
  24. 24.
    Compare: H. Theil, Principles of Econometrics, page 636, problem 5. 1.Google Scholar
  25. 25.
    This is mainly necessary for technical reasons to get nearly correct numbers of degree of freedom and connected statistics in the SPSS’-package procedures.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1985

Authors and Affiliations

  • Wolfgang Schneider
  • Konrad Stahl
  • Raymond J. Struyk

There are no affiliations available

Personalised recommendations