Search

As science becomes increasingly cross‐disciplinary and scientific models become increasingly cross‐coupled, standardized practices of model evaluation are more important than ever. For normally distributed data, mean squared error (MSE) is ideal as an objective measure of model performance, but it gives little insight into what aspects of model performance are "good" or "bad." This apparent weakness has led to a myriad of specialized error metrics, which are sometimes aggregated to form a composite score. Such scores are inherently subjective, however, and while their components may be interpretable, the composite itself is not. We contend that, a better approach to model benchmarking and interpretation is to decompose MSE into interpretable components. To demonstrate the versatility of this approach, we outline some fundamental types of decomposition and apply them to predictions at 1,021 streamgages across the conterminous United States from three streamflow models. Through this demonstration, we hope to show that each component in a decomposition represents a distinct concept, like "season" or "variability," and that simple decompositions can be combined to represent more complex concepts, like "seasonal variability," creating an expressive language through which to interrogate models and data. Plain Language Summary: Models are essential scientific tools for explaining and predicting phenomena ranging from weather and climate, to health outcomes, to economic development, to the origins of the universe, and testing competing models is one of the most basic scientific activites. Yet, how scientists evaluate and justify their models can be inconsistent or even arbitrary. Traditionally, one performance metric—such as mean squared error—is used to identify the best model, but one metric provides little insight into what aspects of a model are "good" or "bad." This paper proposes a basic language for expressing different aspects of a model's performance. On one hand, this is useful for determining which aspects of model may require revision, but it also allows the modeler to separate out the best elements among several models and combine them to form an ensemble, analogous to how an audio engineer mixes together multiple tracks to form the best rendition of a musical piece. Key Points: Mean squared error (MSE) is an objective but somewhat enigmatic measure of model performanceMSE can be decomposed into components that quantify specific aspects of model performance, such as bias and varianceMixing components among models yields a form of ensemble prediction [ABSTRACT FROM AUTHOR]

Natural resource dependence, although commonly invoked in natural resource sociology, has often been ambiguously defined. Communities are frequently described as dependent on natural resource development, but limited attention has been paid to what that means. In the literature, resource dependence is often treated as over‐specialization in, or over‐reliance upon, the natural resource sectors. However, the logic of over‐specialization conceptually grounds dependence in poor economic outcomes. Thus, a one‐dimensional typology of dependence based on a threshold of the share of development in the natural resource sector—as commonly used—does not fully capture the concept and risks tautology. In this paper, I address this ambiguity by formally defining natural resource dependence as over‐specialization in the natural resource sectors. I then present an ideal typology, known as the Natural Resource Community Typology, and a corresponding classification scheme for rural communities in the United States. The typology integrates both extractive and non‐extractive natural resource activity and has two dimensions—the level of development and the level of economic prosperity—and six mutually exclusive categories—extractive specialized, extractive dependent, non‐extractive specialized, non‐extractive dependent, hybrid specialized, and hybrid dependent. I classify counties from 2000 to 2015 and find that while extractive dependence decreased over the study period, non‐extractive dependence increased. [ABSTRACT FROM AUTHOR]

Policymakers across the Southern U.S. have directed considerable attention to regional economic development plans, with mixed results. Questions remain over what types of industries, firms, and/or policies can drive long-range growth. Over the past two decades, researchers have started to focus on high-growth firms (HGFs) due to their potential innovation spillovers and job creation prospects, both of which drive further economic development. Employing databases from the Inc. 500, this study explores the dynamics of HGFs within Southern and comparative contexts. The first goal of this paper is to explore Southern HGF activity in comparison with the U.S. at-large. The second is to examine the geographical characteristics of HGFs within the South, which will include an exploration of locations, industrial mix, and the factors that could encourage HGF development and agglomeration. This paper will also analyze the longer-term ownership and acquisition patterns of HGFs across the South. Are such firms remaining embedded in the South or moving elsewhere through mergers or acquisitions? The paper concludes with a discussion of the impacts of HGFs on regional economic development and moreover, whether existing policies are conducive to these firms. [ABSTRACT FROM AUTHOR]

Between 1957 and 1990, nearly 100 nuclear reactors were constructed throughout the U.S., and nuclear power currently accounts for 20 percent of electricity production nationwide. Nuclear plants are often constructed in small communities for which they constitute a large source of employment and income. To date, 24 nuclear reactors have undergone decommissioning, and more are expected in the future, particularly as nuclear reactors age and face increasingly strict regulations. This paper examines the effects of nuclear decommissioning over time at the county-level on measures of employment, income, and population using difference-in-differences regression and propensity score matching. Panel data are obtained from the U.S. Census Bureau, and cover the years 1975-2014. The analysis finds that nuclear decommissioning is associated with positive and statistically significant increases in employment and per capita income over time. Results suggest that nuclear decommissioning may actually be a positive force in regional economic development, and concludes with limitations of the approach and implications for future research. As an emerging area of research, this paper is meant to build on previous work, as well as to provide a basis for further discussion and debate on the economic future of nuclear host communities and regional economies. [ABSTRACT FROM AUTHOR]

In the last year, a central issue in regional economic growth debate has been represented by the empirical analysis of Verdoorn's law related to the long-term dynamic relationship between the rate of growth in output and the productivity growth due to increasing returns. Several papers have tested Verdoorn's law on European countries as well as many other world economies. Recently, attempts have been made to provide foundations for a spatial version of the original law specification. The main contributions were dedicated to the inclusion of spatial dependence in the economic model. Surprisingly, in the literature on Verdoom's law the analysis of the spatial heterogeneity is not often considered. The aim of this paper is the regional analysis of the spatial dependence and heterogeneity in Verdoorn's law, identifying spatial regimes that can be interpreted as clusters of productivity growth in European regions at NUTS 2 level. To pursue this objective, an optimization algorithm for the identification of groups is used. This constitutes a modified version of Simulated Annealing. [ABSTRACT FROM AUTHOR]

This paper outlines the emergence of a New Washington Consensus associated with leading philanthropies of the new millennium. This emergent development paradigm by no means represents a historic break with the market rationalities of neoliberalism, nor does it represent a radical departure from older models of early 20th century philanthropy. Rather, it is new in its global ambition to foster resilient market subjects for a globalized world; and new in its employment of micro-market transformations to compensate for macro-market failures. Focusing on reforms pioneered by the new philanthropic partnerships in education and global health, the paper indicates how the targets of intervention are identified as communities that have been failed by both governments and markets. The resulting interventions are commonly justified in terms of 'return on investment'. But the problems they target keep returning because the underlying causes of failure are left unaddressed. [ABSTRACT FROM AUTHOR]

Abstract: Innovation in local economic development strategy is necessary to support struggling regions and foster growth across municipalities with varying resources and needs. This paper presents a case of innovation in local economic development strategy to illustrate how institutional entrepreneurship can be used as a tool to help local governments innovatively approach economic development strategy. Through the creation of the Wireless Research Center of North Carolina, institutional entrepreneurs transformed the institutionalized role for local government in economic development, while simultaneously creating a new institutional form that would support local development goals. [ABSTRACT FROM AUTHOR]

This paper utilizes a novel dynamic propensity score matching approach for multiple cohorts of U.S. counties between 1989 and 1999 to examine local economy resilience to rare natural disasters. Affected counties are sorted based on disaster intensity and are carefully matched to similar counties that did not experience a disaster. A difference-in-difference estimator compares trends of affected counties' postdisaster business establishments, employment, and payroll to counterfactual trends in the matched counties. All affected counties experienced short-run drops in economic activity that was particularly noticeable in higher-intensity disasters. In the longer run, less distressed counties returned to their estimated counterfactual trends, but countieswith lower predisaster socioeconomic conditions still lagged in growth, particularly in cases of lowerintensity disasters. Policymakers can use this information to better prepare responses to future disasters. [ABSTRACT FROM AUTHOR]

This paper examines the relationship between immigrants' genetic diversity and economic development in the United States during the late nineteenth and the beginning of the twentieth centuries, a period commonly referred to as the age of mass migration from Europe to the New World. Our panel model estimates show that during this period, immigrants' genetic diversity is significantly positively correlated with measures of U.S. counties' economic development. There exists also a significant positive relationship between immigrants' genetic diversity in 1870 and contemporaneous measures of U.S. counties' average income. (JEL J11, O51, Z13) [ABSTRACT FROM AUTHOR]

Abstract: Using data on 48 contiguous U.S. states and a spatial econometric approach, this paper examines short‐ and long‐run effects of productive higher education and highway infrastructure spending financed by different revenue sources on state economic growth. Following the Lagrange Multiplier, Wald, and Likelihood Ratio tests, the data are found to be characterized by both spatial lag and spatial error processes, leading to the estimation of a dynamic spatial Durbin model. By decomposing results of the dynamic spatial Durbin model into short‐ and long‐run direct as well as indirect (spillover) effects, we show that accounting for spillover effects provides a more comprehensive approach to uncovering the effects of productive government spending on growth. We find that, regardless of the financing source, productive higher education and highway spending have statistically significant short‐ and long‐run direct as well as spillover effects on state income growth. [ABSTRACT FROM AUTHOR]

Despite economic growth since the recession, the gap between the richest and the poorest segments of the population remains one of the most pressing concerns of contemporary America. This paper uses IR-95/20, IR-80/20, and IR-65/35 ratios to measure the income divides between the richest and the poorest segments in the mid-to-large-sized metropolises of the U.S. Southeast, their variation across ethnicities, and their association with metropolitan level attributes such as diversity, segregation, socio-economic, and other built-environment, and labor characteristics. The income divide ratios serve as the dependent variables whereas principal components along with state-dummy variables serve as the explanatory variables in regressions analyses. The metropolises that are large, diverse, and better educated are the most income-divided whereas those with lower educated people are less divided. Metropolises with larger shares of their labor engaged in primary sectors of economy have higher income divides; this observation also holds true for African Americans and Hispanics. Metropolises that gained in intermixing during 2000-2014 are associated with a lower income divide and vice versa. [ABSTRACT FROM AUTHOR]

This paper addresses change in oil and gas policies pushed by the U.S. Bureau of Land Management (BLM) during the administrations of Presidents George W. Bush and Barack Obama. Administrative policy changes occurred not only because of election outcomes and the emergence of new governing coalitions but BLM's selective utilization of policy tools such as rulemaking, planning, environmental impact analyses, and the use of discretionary authority to increase or relax enforcement decisions. The data reveal that BLM put more emphasis on the use of discretionary authority to limit environmental inspections and to limit environmental reviews of proposed drilling projects under Bush while agency officials gave priority to adopting new planning procedures to allow greater stakeholder input under Obama as well as increasing the number of environmental inspections. [ABSTRACT FROM AUTHOR]

ABSTRACT This paper investigates the role of the business size distribution on income and employment growth in U.S. counties from 1990 to 2000. We measure the business size distribution as the share of employees across nine establishment size categories that range from microfirms (one to four employees) to large firms (1,000+ employees) and using three indices similar to a Gini coefficient. Results show that the business size distribution has a significant impact on county-level growth patterns. Employment shares in small firms increase employment growth, but decrease income growth. One possible conclusion suggests policies emphasizing small firms and entrepreneurship during times of high unemployment. [ABSTRACT FROM AUTHOR]