Input‐Output Modelling of the Urban and Regional Economy: The Importance of External Trade

Methods for regionalizing input-output tables

“…Lahr (1993) emphasized that most hybrid methods (in their initial step) are based on non-survey techniques, which are then later enhanced by additional regional data. One of the earliest methods (Morrison and Smith 1974) comparing different methods suggests the use of hybrid techniques, although, since then a lot of other studies have also reinforced these findings: (Brand (2012) in Finland, Kowalewski (2012) and Lindberg (2011) in Sweden, Patriquin et al (2002) in Western-Middle Alberta, Ralston and Hastings (1986) in Delaware, Harris and Liu (1997) in Porthmouth, Bonfiglio (2005) in Marche region, Oosterhaven et al (2003) in the Netherland, Jiang et al (2012) in China and Round (1983)). By examining several methods (LQ, , PLQ, RLQ, FLQ, AFLQ, SDP, RSP, RAS), the authors concluded that even if improved methods (like FLQ) can generate relatively accurate results, the assumptions of non-survey methods are generally too strict to reflect true input-output relations.…”

“…RAS can be applied alone, but most hybrid methods use RAS as an intermediate step in the procedure. In general, RAS produced satisfactory results (Miller and Blair 2009, Morrison and Smith 1974, Jalili 2005, Riddington et al 2006, Round 1983, Harris and Liu 1997. Flegg and Tohmo (2013) suggested the combined use of FLQ and RAS to improve the accuracy of estimation.…”

Beyond Intermediates: The Role of Consumption and Commuting in the Construction of Local Input–Output Tables

“…In practice, these assumptions are unlikely to hold. Therefore, several authors have analysed the extent to which LQ-based estimates of regional input coefficients are biased (Schaffer and Chu, 1969;Smith and Morrison, 1974;Round, 1978;Harrigan et al, 1981;Willis, 1987;Harris and Liu, 1998;Tohmo, 2004;Stoeckl, 2012). Typically, the primary emphasis is on the influence of cross-hauling.…”

“…Typically, the primary emphasis is on the influence of cross-hauling. This is not captured in LQ methods, which results in an underestimation of imports and exports and an overestimation of local intermediate transactions (see Harris and Liu, 1998, for a detailed discussion). An exception is Stoeckl (2012), who explores the role of differences in productivity.…”

“…Many authors are critical of employing non-survey methods for this task (e.g. Harris and Liu, 1998). Hybrid methods are favoured, as they retain significant accuracy while requiring less primary data collection than would full surveying .…”