Amighini

Concentration, diversification or decline?How do Italian local economic systems perform in international markets?: 

Concentration, diversification or decline? How do Italian local economic systems perform in international markets? Alessia Amighini, Marinella Leone and Roberta Rabellotti Università del Piemonte Orientale III Workshop PRIN - Novara

Motivation : 

Motivation Contrasting evidence on the dynamics of comparative advantages at local level, especially for IDs provinces: District effect: persistence of RCA positively correlated with the presence of IDs (De Benedictis, 2005; De Benedictis and Padoan, 1999; Epifani, 1999) Growing evidence on IDs changing specialisation both across and within sectors (De Arcangelis and Ferri, 2005; Rabellotti, 2005)

The theoretical background: 

The theoretical background Trade integration among economies at different levels of technological development within imperfectly competitive markets (Grossman and Helpman, 1991): Two sectors; Differentiated final goods: F Capital goods to produce final goods: K (De Arcangelis and Ferri, 2005)

Two opposite predictions: 

Two opposite predictions More advanced economies tend to specialise in K and despecialise in F(Ricardo effect or Heckscher-Ohlin effect) More advanced economies tend to despecialise in K because of higher competition in K

Research questions: 

Research questions Are there changes in specialisation across and within sectors? with a special focus on provinces with IDs. Different scenarios: concentration on core competencies diversification out of core competencies overall decline.

Data: 

Data Export flows from 103 Italian provinces to the world in 1995 and 2005 in CPAteco 2002 13 manufacturing sectors at 5 digits (95 sub-sectors) (ISTAT); Sector reaggregation according to end-use categories: intermediate, final and capital goods following RPI classification (ISTAT).

Methodology: 

Methodology Symmetric Balassa index of RCA for each province on n sectors by end-use relative to the world (Comtrade); Analyisis of the dynamics of overall distribution of RCA for each province by comparing the median of RCA and a Spearman rank correlation coefficient between 1995 and 2005; Analysis of the dynamics of across and within sector specialisation; Scenarios.

Slide8: 

Overall RCA distribution

Slide9: 

Overall RCA dbu: provinces with IDs by area

Slide10: 

Spearman’s rank correlation coefficients

Slide11: 

Dynamics of RSCA: Textile

Slide12: 

Dynamics of RSCA: Clothing

Slide13: 

Dynamics of RSCA: Footwear

Slide14: 

Dynamics of RSCA: Machinery

Slide15: 

Dynamics of RSCA: Textile (IDs vs non-IDs)

Slide16: 

Dynamics of RSCA: Clothing (IDs vs non-IDs)

Slide17: 

Dynamics of RSCA: Footwear (IDs vs non-IDs)

Slide18: 

Dynamics of RSCA: Machinery (IDs vs non-IDs)

Slide19: 

dRSCA dMedian STRENGTHENING DIVERSIFICATION WEAKENING CONCENTRATION Four different scenarios

Slide20: 

IDs textile provinces

Slide21: 

IDs clothing provinces

Slide22: 

IDs footwear provinces

Slide23: 

IDs machinery provinces

Slide24: 

Scenarios: Machinery (provinces without IDs)

Slide25: 

Provinces with IDs: Clothing vs Textiles

Slide26: 

Provinces with IDs : Clothing vs Machinery

Slide27: 

Provinces with IDs: Textile vs Machinery

Slide28: 

Provinces with IDs: Footwear vs Machinery

Slide29: 

Main preliminary findings Provinces with IDs overall are more persistent but several changes across and within sectors; Tendences towards increasing CA in machinery and intermediate goods and decreasing CA in final goods in some IDs provinces specialised in clothing and footwear; Prima facie evidence on linkages between delocalisation and verticalisation.