Counterfactual Spatial Distributions

Abstract

Recent contributions provide researchers with a useful toolbox to estimate counterfactual distributions of scalar random variables. These techniques have been widely applied in the literature. Typically, the dependent variable of interest has been a scalar and little consideration has been given to spatial factors. In this paper we propose a simple method to construct the counterfactual distribution of the location of a variable across space. We apply the spatial counterfactual technique to assess how much changes in individual characteristics of Hispanics in the Washington DC area account for changes in the distribution of their residential location choices.

Accounting for central neighborhood change, 1980–2010

“…First, using a procedure akin to that proposed by DiNardo, Fortin & Lemieux (1996) for decomposing wage distributions, we decompose the sources of changes in demand for central neighborhoods into those due to secular demographic shifts, holding neighborhood choices constant, and those due to changes in neighborhood choices of various demographic groups, holding demographic shares constant. While our focus is on central neighborhoods, this method is broadly applicable for decomposing drivers of demographic change for any type of neighborhood, as is also discussed in Carrillo & Rothbaum (2016). In this portion of the analysis, we highlight the importance of the growth in minority population shares to counteract negative group-speci…c demand shifts for downtown neighborhoods, mitigating downtown population declines driven by changing neighborhood choices among all demographic groups.…”

The contribution of the spatial dimension to inequality: A counterfactual analysis for OECD countries

“…DFL decomposition is a semi‐parametric approach that enables us to extend the results to the whole distribution of income, since it works with the entire population; not only with the mean. Although this is not the first time that the DFL methodology has been applied to the analysis of income inequality focusing on the spatial dimension (Carrillo & Rothbaum, ; Dickey, ), as far as we know, an analysis similar to the one we describe below has not yet been conducted. We focus on the particular contribution of territory to income inequality not only from a semi‐parametric perspective, but linking it to the traditional and more theoretical proposals.…”

A Spatial Perspective on the Econometrics of Program Evaluation

“…The term “counterfactual” here specifies counterfactual scenarios or experiments that attempt to define how treatments would differ based on differing hypothesized counterfactual distributions of some factor (often measured as a scalar value) impacting outcomes. Carrillo and Rothbaum (2016) provide an overview of counterfactual spatial distributions as a necessary extension in the context of decomposition economics, building on work by Fortin, Lemieux, and Firpo (2011) and others. Our discussion extends more broadly beyond the counterfactual context of decomposition strategies, focusing instead on the distillation of the research design required in counterfactual research across multiple methods.…”