Search

I. INTRODUCTION Clean indoor air laws .are hotly debated in state and local jurisdictions across the country. The increasing popularity of these laws raises questions about unintended consequences of regulation. The laws are designed to protect worker and patron health from the adverse effects of second-hand smoke. However, it is common for entrepreneurs to claim that the laws will harm business activity, as smoking customers alter their behavior. Disagreements between health advocates, regulating authorities, and the business community have spawned a large empirical literature devoted to isolating the economic impacts of smoking bans. Scollo et al. (2003) provide a comprehensive review of 97 studies conducted prior to August 31, 2002. They find that most of the "high-quality" studies conclude that there is no negative impact of smoking bans on a range of economic performance measures. (1) However, most of the research to date has focused on data aggregated to the county or city level. These data provide important evidence on the aggregate impacts of regulatory policy, but ignore the possibility that smoking regulations may have very different impacts on individual firms. Thus, even if a smoking law has no aggregate impact, it may still distort outcomes across firms. This possible heterogeneity in policy -impact should be taken into account by policymakers when weighing policy decisions. In addition, many studies on smoking bans have focused on the impact on sales and related measures of economic performance. The interest in sales-related data makes sense because it is .an important determinant of firm profits and also an important part of the sales tax base. However, sales data are not the only measure of economic performance that matter for policymakers and the business community. Employment is a key variable input into the production of restaurant services and it will provide information on the economic effect of regulation. Employment is also a critical benchmark for policymakers and recent research has begun to identify possible employment impacts of smoking bans. (2) Here the results have been mixed. Thompson et al. (2008), Hyland, Vena, and Cummings (2000), and Hyland and Cummings (1999) find that smoking bans had no significant adverse impacts on employment or turnover in restaurants located in Arizona and New York. In contrast, Adams and Cotti (2007) find that smoking bans did affect bar employment using county-level data for the United States (but not restaurant employment). Further, Dunham and Marlow (2003) find that many bar and restaurant owners report that they have reduced worker benefits and/or raised worker responsibilities in response to a smoking ban. One possible explanation for the contrasting aggregate employment results so far may be differences in smoking prevalence across regions. Much of the research on employment impacts (and sales impacts as well) has focused on states with relatively low rates of adult smoking. For instance, in 2006, Arizona (18.1%) and New York (18.3%) both had rates of adult smoking prevalence that were well below the national median of 20.2%, according to data for the Centers for Disease Control. Thus, previous research may have found that smoking bans do not affect restaurant employment because there were relatively few smokers in the market area. This implies that the impacts of smoking laws might be highly heterogeneous across regions. Indeed. Adams and Cotti (2007) find that the impact of smoking laws varies with county smoking prevalence for both bars and restaurants. Further, Dunham and Marlow (2000b, 2004) find that smoking prevalence has a significant impact on -nonsmoking seat allocations. Using a probabilistic theoretical model, we expect smoking regulations to affect employment through the demand side of the labor market. Demand-side considerations are driven by changes in the desirability of restaurant visits when smoking in the establishment is restricted. …

We analyze the impact of fiscal decentralization on U.S. county population, employment, and real income growth. Our findings suggest that government organization matters for local economic growth, but that the impacts vary by government unit and by economic indicator. We find that single-purpose governments per square mile have a positive impact on metropolitan population and employment growth, but no significant impact on nonmetropolitan counties. In contrast, the fragmentation of general-purpose governments per capita has a negative impact on employment and population growth in nonmetropolitan counties. Our results suggest that local government decentralization matters differently for metropolitan and nonmetropolitan counties.

This research analyzes manufacturing growth and decline across metropolitan and nonmetropolitan regions during the 1972–2002 period. We decompose real value added growth across local labor market areas in the lower 48 U.S. states into contributions from labor, capital, and total factor productivity. We then estimate a model describing the long-run growth of labor, capital, and productivity and find that increased productivity increases the growth of labor and capital, as well as a positive correlation between labor and capital stock growth. We also find evidence that human capital investment and agglomeration economies encourage productivity growth, while unionization discourages it.

This article analyzes determinants of growth across labor markets in the United States, using a production function approach based on four inputs: labor, manufacturing investment, human capital investment, and public capital investment. We find little role for public capital investment in growth, but that manufacturing investment spurred growth in nonmetropolitan areas, in contrast to metropolitan areas. We also find that human capital investment mattered more for metropolitan areas than for nonmetropolitan areas. Further, the presence of more colleges and universities, more household amenities, and lower tax rates are all found to have encouraged human capital accumulation in U.S. labor markets. Copyright 2008, Oxford University Press.

A dynamic labor market model is used to motivate the inclusion of population characteristics and industrial structure as determinants of regional employment instability. We examine how these factors influence regional employment instability using data from both metropolitan and nonmetropolitan regions in the United States. We find that population characteristics are important determinants of employment volatility and that increased industrial specialization (reduced diversification) increases employment volatility, but the magnitude of that influence drops substantially once population characteristics are considered. We also find that the influence of population characteristics and industrial specialization varies significantly across metropolitan and nonmetropolitan regions.

This article examines spatial aspects of distributional dynamics and finds that the distribution of US metropolitan incomes relative to their neighbours has diverged during the 1969-1999 period. Use of a spatial Markov approach shows that non-metropolitan neighbours of metropolitan regions have tended to converge during the period, with roughly equal rates of upward and downward mobility within the distribution. Non-metropolitan regions, not neighbouring metropolitan regions, show much less tendency to converge and reveal higher rates of downward rather than upward mobility. Results highlight regional differences in mobility coherence, with metropolitan areas in the West tending to outpace their non-metropolitan neighbours.