Competition, Quality Choices and Vertical Differentiation: Applications to the Nursing Home IndustryPublic Deposited
This dissertation contains three essays that investigate various factors affecting firms' choice of quality in the context of nursing homes. The first essay examines how strategic interactions with competitors affect quality levels selected by nursing homes. I explore nursing homes' responses to minimum nurse staffing standards imposed in two large states: California and Ohio. I compare the response to the standards of nursing homes with different initial levels of quality, and the response by high-end nursing homes in a given market to increases in quality by low-end homes in the same market. I find that minimum staffing standards increase nursing homes' total nurse staffing hours per patient day by 5% (0.14 hours) on average. This increase in quality largely comes from nursing homes initially positioned close to the minimum standards level. I also find that nursing homes tend to substitute cheaper labor inputs for more expensive ones to meet the minimum standards. Finally, consistent with vertical differentiation theory, if high-end nursing homes do increase quality, it occurs in the markets where they have the strongest incentive to vertically differentiate from their low-end competitors. The second essay considers how market structure affects the choice of product levels. I analyze whether nursing home residents in more concentrated markets enjoy higher levels of quality as measured by total nursing hours per patient day. In order to address the potential endogeneity of market structure, I exploit shocks to the Herfindahl-Hirschman Index (HHI) caused by statewide exists of chain-affiliated nursing homes. Using these exists as an instrument for changes in HHI, I find that a one standard deviation increase in HHI increases total nursing inputs by 80% of total nurse staffing Hours Per Patient Day (HPPD). Among markets with positive change of HHI, total HPPD increase 40%. By comparison, the estimates are insignificant at conventional levels using fixed effects OLS regression, implying that changes in market structure are endogenous to quality choices. The last essay investigates the role of market competition in shaping nursing home ownership choice and the dynamic effects of ownership conversions on nursing home performances. I find that competition leads to more conversion to for-profit from non-profit or government status but less conversion to non-profit from other status. I also find that facilities converting from for-profit to nonprofit exhibited significant increase in total nurse staffing inputs starting from two years after the conversions. There is no significant change in other dimensions of performance such as size, capacity and patient composition for both pre and after conversion periods. Neither do I find significant changes in performance from nursing homes converting from nonprofit to for-profit status.