Regional innovation, growth and convergence–divergence patterns in the EU: George Petrakos and George Anastasiou

It is obvious that regions that are not sharing these pos it ive charac ter istics are facing ser ious obs tacles to competing and growing in the new eco nomic environ ment. The differential impacts of these factors, along with successful or less successful pol icy agendas, are the main drivers that increase spatial in equal ity and allow for a non-linear pattern of growth at the EU regional level (Petrakos, 2008;­Petrakos­&­Artelaris,­2009). The im port ant relev ance given to the regional capa city for know ledge and in nova tion as growth factors stimulated a qualit at ive change in the regional convergence liter at ure, incorporating the new para meters. The theor et ical and empirical studies that examined the issue are neither conclusive re gard ing the convergence-divergence­trends­in­the­EU­regions,­nor­give­a­clear­pic­ture­of­the­ location patterns of innov at ive activity. Furthermore, most of the econo metric studies­re­gard­ing­the­convergence-divergence­ana­lysis­tend­to­overlook­the­rel­ at­ive­ im­port­ance­or­size­of­each­region­(Doloreux­&­Bitard,­2005;­Hollanders,­ 2007). This shortcoming could produce unrealistic or misleading results. The aim of this chapter is to con trib ute to the better understanding of the formation­of­ the­new­Euro­pean­scenery­under­ the­ influence­of­ the­know­ledge­eco­ nomy. In the present study we use OLS, Panel and WLS estimators to ana lyse RRSII­data­within­208­EU25­regions­over­a­five­­year­period,­in­order­to­identi­fy­ patterns and trends of regional convergence or divergence. We argue that the Weighted Least Squares (WLS) methodology is more appropriate for models

using regional data. The comparison between the methods investigates the as sump tion that when WLS methodology is used in order to account for the signific­ant­dif­fer­ences­among­Euro­pean­regions­with­respect­to­size,­a­U­­shaped­pro­ cess­could­provide­a­better­fit­to­regional­data.­The­findings,­even­if­not­conclusive,­ raise im port ant research questions that are related to the spatial footprint of the in nova tion dy namics in Europe, with sharply differing pol icy im plica tions. Do less innov at ive regions in the EU tend to catch up with more innov at ive ones? Is convergence or divergence the prevailing pattern of spatial change with respect to in nova tion? Does the mix of market forces and pol icy dy namics that shapes the innov at ive charac ter istics of regions result in convergence or a new division? The chapter is or gan ized as follows. The next section presents a brief review of recent approaches and studies re gard ing regional convergence or divergence and­ their­ relation­ to­ in­nova­tion­ ac­tiv­ities.­ Section­ 12.3­ presents­ the­ empirical­ ana­lysis­with­the­estim­ated­regressions­and­discusses­the­results,­while­the­final­ section presents conclusions.