The Dynamics Of Regional Population And Employment Growth

The authors investigate what Muth labels as the “chicken and egg†or what others label it as “jobs follow people†versus “people follow jobs†debate by studying the dynamic relationships of changes in population and employment in the snowbelt, the sunbelt, and the nine census regions of the United States. The vector autoregressive (VAR) modeling method is used for the annual time series data (1955–1988) on population and employment. The Granger causality tests show that, in the snowbelt region, population tends to precede employment, while in the sunbelt region, the opposite is true. The impulse response functions generated for the forecasting decade suggest that “jobs follow people†in the snowbelt and “people follow jobs†in the sunbelt. Similar findings in general hold for the census regions belonging to the snowbelt and sunbelt. These findings raise doubts about the proposition of others that the growth process is either demand or supply†driven in all phases of regional growth.