On the accuracy of gravity-RAS approaches used for inter-regional trade estimation: evidence using the 2005 inter-regional input–output table of Japan

Abstract

In contrast to international trade, it is still difficult to find regional trade statistics within a nation. Given that the gravity model continues to be very popular, we test two gravity-RAS approaches used for interregional trade estimation: a standard one and an extended version, which additionally estimates intra-regional flows. We assess the accuracy with the help of two measures and for different sectoral aggregation levels. For that, we use the survey-based 2005 interregional input-output table of Japan as a benchmark. Results show high overall accuracy levels for the standard approach, better than when using international data, albeit with heterogeneous errors for sectors and regions. We further find that the results of a multiregional input-output model are highly sensitive to the trade estimation approach and that errors slightly increase for increasing sectoral disaggregation levels. Results from an experiment based on a random number generator show how RAS influences results.

Doubly constrained gravity models for interregional trade estimation

“…Among these, there are studies explicitly inspired by the DCGMs of transportation planners—such as Lindall et al (2006) and Johansen et al (2018)—but there are also studies that do not put themselves in that tradition. The latter include Fournier Gabela (2020), Riddington et al (2006), Sargento et al (2012) and Yamada (2015). In these analyses, initial estimates of bilateral trade are first predicted using an empirical gravity equation (e.g., one of an econometric nature) and then improved upon using the RAS algorithm.…”

“…So does the way the values of , and are determined. Typically, parameter values are selected on the basis of theoretical considerations (Lindall et al, 2006), econometric analyses (Riddington et al, 2006; Yamada, 2015) or educated guesses (Fournier Gabela, 2020; Johansen et al, 2018; Sargento et al, 2012). In some instances, is calibrated to freight transport data using an iterative procedure (Lindall et al, 2006; Riddington et al, 2006).…”

“…In some instances, is calibrated to freight transport data using an iterative procedure (Lindall et al, 2006; Riddington et al, 2006). Occasionally Equation is augmented with a scaling factor (Fournier Gabela, 2020; Sargento et al, 2012). Sometimes the model includes additional explanatory variables (Johansen et al, 2018; Sargento et al, 2012) or is assumed not to apply to intra‐regional flows (Fournier Gabela, 2020; Sargento et al, 2012; Yamada, 2015).…”

“…Occasionally Equation is augmented with a scaling factor (Fournier Gabela, 2020; Sargento et al, 2012). Sometimes the model includes additional explanatory variables (Johansen et al, 2018; Sargento et al, 2012) or is assumed not to apply to intra‐regional flows (Fournier Gabela, 2020; Sargento et al, 2012; Yamada, 2015). We can dispense with such idiosyncrasies here, as they do not affect our main argument.…”

“…As a result, it is not always clear how to generalize beyond the specific data availability conditions under which they were developed, or how to assess the relative strengths and weaknesses of alternative approaches. It adds to the difficulty of drawing lessons from past experiences that—again, due to a lack of suitable data—very few attempts have been made to evaluate the accuracy of the trade estimates (Distefano, Tuninetti, Laio, & Ridolfi, 2019; Fournier Gabela, 2020).…”