Exploring Brexit with dynamic spatial panel models: some possible outcomes for employment across the EU regions

Abstract

Starting with a reduced form derived from standard urban economics theory, this paper estimates the possible job-shortfall across UK and EU regions using a timespace dynamic panel data model with a spatial moving average random effects structure of the disturbances. The paper provides a logical rational for the presence of spatial and temporal dependencies involving the endogenous variable, leading to estimates based on a state-of-the-art dynamic spatial generalized moments estimator proposed by Baltagi et al. (Reg Sci Urban Econ, 2018.

Merging macroeconomic and territorial determinants of regional growth: the MASST4 model

“…Several works have been recently trying to capture the likely long-run effects of Brexit, but any such attempt clashes with the fact that as of today the way this process will actually take place (hard vs. soft Brexit) is still unclear (Fingleton 2018;Dhingra et al 2017). In the (for many, worst case) scenario whereby Brexit would happen with a complete cancelation of all free trade and movement agreements (6) Pr(CHANGE OF INNOVATION PATTERN r ) = f(R&D r , HC r , LQ HIGHTECHr , POPULATION DENSITY r , TECHNOLOGICAL DIVERSIFICATION r , IIA r , SCDA r , STAA r , )…”

Italexit, is it another Brexit?

“…The assumption is that the parameter estimates remain at their estimated levels into the future, and that the Italexit impact on employment is captured by changes to the trade flows between Italian regions and the regions in the rest of the EU. Attention is focussed on 2020 and later, so as to allow comparison with the Brexit effect estimated in the companion paper (Fingleton 2019), given that the UK's formal exit from the EU is scheduled for the first half of 2019, so 2020 will the first full year outside the EU. Given assumptions regarding future levels of and , for = 2020 onwards there are two scenarios: one based on the trade flows assuming no-Italexit effect and the other assuming an Italexit effect on trade flows, and the difference between them is…”

“…There are similarities and contrasts with Brexit. In a companion paper to this one (Fingleton 2019) which uses precisely the same methodology, predictions of jobshortfall are also made as a consequence of the UK's exit from the EU. As with my Brexit predictions, I emphasize that the Italexit predictions are treated with caution, to repeat my earlier warning, 'the numbers SHOULD NOT be taken too seriously.…”

“…Although Switzerland and Norway are formally outside the EU, they are an integral part of the analysis. Their regions are referred to as EU regions for convenience.2 Data for 2011 are not used in estimation but held for one-step ahead prediction.3 The companion (Brexit) paper,Fingleton (2019) gives details of theory leading to this specification.…”

Entrepreneurship and Income Inequality: A Dynamic Spatial Panel Data Analysis

“…Compared with non-spatial dynamic panel data models where serial correlation occurs in the time dimension, the DSPD models may have a correlation in the time dimension as well as spatial correlation across spatial units. As argued by Baltagi, Fingleton, & Pirotte, (2019), Elhorst (2012Elhorst ( , 2014a, Elhorst, Zandberg, & de Haan, (2013), Fingleton (2017Fingleton ( , 2019, Fingleton and Szumilo (2019), Yu (2010a b, 2014), Yesilyurt and Elhorst (2017), among others, DSPD models are able to deal with unobservable spatial, individual and/or time-period specific effects. They also tackle more efficiently endogeneity problems, such as the potential bias in the coefficient of the spatial lag of the dependent variable.…”