Macroeconomic Implications of Changes in the Composition of the Labor Force

The United States has experienced several sectoral shifts over the last hundred years, a shift from agriculture to manufacturing, manufacturing to services, and within these sectors the shift from the home sector to the market sector. These shifts have been attributed in the literature to different sectoral productivity growth rates. It is likely that these sectoral re-allocations have far reaching impacts on economic indicators we measure. My dissertation concerns the effect of the difference in productivity growth rates across sectors on the market to home hours ratio, sectoral employment trends for men and women, and the consumption to output ratio as measured in the National Income and Product Accounts.;The first chapter of my dissertation focuses broadly on the ten percentage point increase in the consumption to output ratio since the 1950's. There has been much speculation from the popular press that this increase implies that Americans are consuming beyond their means and that this path of consumption is unsustainable. Economists who have researched this trend can account for part of the increase through an increase in the depreciation rate and decline in the population growth rate. Both imply that a necessary condition for an increasing consumption to output ratio is the economy must be in transition to a new balanced growth path. I find that balanced growth is an unnecessary condition, and the important aspect of the economy to model to capture this trend is the home sector.;The home sector is not included in the National Income and Product Accounts, but it produces consumption goods that have many close substitutes with the market sector. After 1950, the growth rate of market total factor productivity was greater than the growth rate of home total factor productivity and hours of work shifted from the home sector to the market sector. Unmeasured home goods were replaced by measured market substitutes and since home consumption and production is not included in NIPA, the consumption to output ratio increased. The increase in the consumption to output ratio is therefore a natural outcome of structural transformation and can occur when the economy is on an aggregate balanced growth path.;Most of the increase in the market hours to home hours ratio post-1950 can be attributed to the increase in female labor supply, and a majority of the increase in the consumption to output ratio is a result of the increase in market services. Not all market goods are close substitutes with home produced goods. Home produced services such as childcare, care for elderly parents, and meal preparation have substitutes that can be purchased only in the market services sector, such as daycare services, nursing home services and restaurant meals. It has also been documented that post-1950 the manufacturing sector grew at a faster rate than the market services sector, which grew at a faster rate than the home services sector. Taking into account all of these facts, the second chapter of my dissertation asks if differences in total factor productivity growth rates in the home, market services, and goods sectors can generate the shift in employment shares that occurred during this time period for both men and women, while remaining consistent with an increase in the consumption to output ratio during the transformation process.;As sectors grow at different rates, employment shifts to the more productive sector if the goods are close substitutes and away from the more productive sector if the goods are complements. Male employment moves from the manufacturing sector to the market services sector and female employment moves out of the home sector to market services. Working women replace home production with market purchased services. Since home production and consumption are not included in the national accounts, a shift from home to market consumption increases the NIPA consumption to output ratio, consistent with the previous study.;The first two chapters created theoretical models that could account for aggregate trends of interest, but did not explore the micro-foundations in great detail. The third chapter delves into the Panel Study of Income Dynamics and the American Time Use Survey to explore how households respond to female labor supply on both the extensive and intensive margins. The goal is to determine the extent to which expenditure and time use patterns differ between married households in which both the male and female are employed and married households in which only the male is employed.;The constraint in this analysis is the ability to overcome selection issues with the available data. Person fixed-effects will be used in the regression analysis performed with the PSID data to mitigate the selection issues. The same strategy cannot be implored with the ATUS because it is a repeated cross-section. Therefore, we control for as many variables as possible that may affect expenditures (eg. age, education, number of children, etc.). The results are consistent with the story proposed in the first two chapters. Women spend less time taking care of their children, less time cleaning the house, and less time preparing meals for their family. Households in which the women are employed also purchase more daycare services and meals away from the home.