Catching-up and regional disparities: a resource-allocation approach

ABSTRACT

While the allocation of regional assets has been theoretically and empirically analysed in regional growth theories, the subsequent logical step on how such allocation affects regional inequalities has not found direct quantitative validation. This paper tries to fill this gap by presenting an original simulation able to capture the relative weight of each economic resource in the process of decreasing regional disparities. Results witness that not all resources play the same role and that inequalities are better overcome when growth assets fit the regional productive specialization. The former result represents a guide for raising policymakers’ awareness on resource priority investments, while the latter proves once again that the local productive vocation is dramatically relevant, and that Cohesion Policy cannot be designed without considering local specificities, in line with the theoretical arguments of the Smart Specialization Strategy.

KEYWORDS:

• Regional disparities
• growth assets
• convergence
• Theil index

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 See, among others, Lopez-Bazo et al. (1999), Giannetti (2002), Cappellen et al. (2003), Fischer and Stirböck (2006), Dall’Erba and Le Gallo (2008), Geppert and Stephan (2008), Artelaris, Kallioras, and Petrakos (2010), Del Bo, Florio, and Manzi (2010), Petrakos, Kallioras, and Anagnostou (2011), Smetkowski and Wójcik (2012), Borsi and Metiu (2015), Von Lyncker and Thoennensen (2017), Ayala, Martín-Román, and Vicente (2020), and for a meta-analysis Abreu et al. (2005).

2 We considered Less Developed Regions (LDRs) as those identified for the 2014–2020 programming period, according to the idea that those are the regions on which the European Cohesion Policy mainly focuses at the moment. For a complete list, the reader can refer to the official Commission Implementing Decision, accessed 21 August 2020 at https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:32014D0099.

3 Source: ESPON, TRACC Project. The indicator was chosen because it combines several modal accessibility indicators (road, rail, inland waterways and air). The per capita indicator allows us to have a density measure, capturing the infrastructure supply relative to the potential demand.

4 The 2013 NUTS classification has been considered within the present work.

5 The two periods were chosen considering that 2012 was the first year of recovery from the crisis at the European level (Capello and Caragliu 2020).

6 Metropolitan regions are NUTS3 regions or a combination of NUTS3 regions which represent all agglomerations of at least 250,000 inhabitants (for further details: https://ec.europa.eu/eurostat/web/metropolitan-regions/background, accessed 13 December 2019).

7 See Table 2 for a few exceptions due to data availability.

8 Our intention is not to enter agricultural matters, rather to characterize some LDRs according to their stage of development/specialization pattern. Therefore, the aim is not to suggest agricultural policies, but cohesion policies that work better in regions characterized by an agricultural specialization.

9 ANOVAs showed that in both cases, the statistical difference between the two categories is extremely significant (1%).

10 As we will see later, the number of regions affects the order of magnitude of the simulation results. For this reason, comparison among simulations run on different regional categories must be made with cautiousness.

11 In more detail, after 2008, there was an overall slowing down in the convergence process due to a deceleration in the reduction of between-country disparities and to an increase in within-country disparities (see OECD 2018; Camagni et al. 2020).

Additional information

Funding

A first version of the empirical analysis was supported by Directorate-General for Regional Policy: [Grant Number 2017CE16BAT125].