This dissertation explores two economic phenomena involving forest-dependent areas: wage distribution and migration pattern of individuals. Are forest-dependent rural areas less desirable for workers from the standpoint of labor market returns? Are different skills (e.g. education, experience) rewarded differently in these areas? If there are interregional wage differences, would that influence a working-age individual's migration decision and residential location choice? These questions are examined here with data from Public Use Microdata Survey (PUMS) of the United States 2000 decennial census. I focus my inquiry on the socio-economic characteristics of forest and non forest-dependent areas, the interregional difference in labor market outcomes, and the influence of wage differential and other factors on the migration decisions and residential location choice of working-age individuals.
Reduced form log wage equations are estimated for individuals, which incorporate explanatory variables related to both skill factors (such as years of schooling and potential experience) and to non-skill factors (such as minority status) potentially influencing wage. Variations in the interregional wage distribution resulting from differences in skill mix of workers are removed by using a standardized skill distribution. The results indicate that average wages and variation in wage distribution are lower in the forest-dependent areas than other areas.
For migration analysis, a Partially Degenerate Nested Logit (PDNL) model is used to model migration decisions and location choice of individuals in the Northwest between 1995 and 2000. Migration decision is modelled as a two stage process: first an individual decides whether to move or stay in his/her original location. Conditional on migration, the individual next chooses his/her destination location from a set of alternative areas. Empirical results indicate that educational attainment and labor market outcomes influence individual's migration decision. Conditional on moving, non-forest dependent areas are more attractive to individuals as residential destinations.