City objectives and monopoly franchising. An empirical analysis of - PowerPoint PPT Presentation

Infraday 2010 – 8/9 October City objectives and monopoly franchising. An empirical analysis of calls for tenders in Italian gas distribution Riccardo Marzano

Outline 2 • Introduction • Gas distribution in Italy • Background • Data and variables • The model • Results • Conclusions

Introduction 3 • Aims of the paper  Objectives pursued by Italian municipalities in franchising the gas distribution service  Testing the taxation by regulation effect (Posner, 1971-1972) • Empirical analysis on a sample of 174 calls for tenders (2001-2008 period) • Empirical methodology:  Linear simultaneous equations model  3SLS estimation

Gas distribution in Italy (1/2) 4 • Legislative Decree n. 164/2000 (Letta’s Decree)  compulsory competitive bidding procedures in selecting utility management units, designed by local governments, which are entrusted with functions of programming and control  definition of an upper bound to the franchise duration (12 years)  price (tariffs) regulation entrusted to the AEEG, the Italian energy regulator  property of infrastructures going back to the local authority at the end of the franchising term

Gas distribution in Italy (2/2) 5 • “Most economically advantageous tender” approach  Upper bounds on the scores • Scoring rules  Formulas to compute each score • Example Three dimensional scoring (Scores A B C) Upper bound on score A = 50 A i = [bid i (A)/bid max (A)]*50 Upper bound on score B = 30 B i = [bid i (B)/bid max (B)]*30 Upper bound on score C = 20 C i = [bid i (C)/bid max (C)]*20 Total Score i = A i + B i + C i

Background 6 • Monopoly franchising  Seminal idea: Demsetz (1968) • Critics on competitive franchising arrangement  Transaction costs: Williamson (1976), Goldberg (1976)  Taxation by regulation: Posner (1771-1972), Prager (1989), Beutel (1990), Otsuka & Braun (2002) • Scoring Auctions  Theoretical analyses: Asker & Cantillon (2008-2010)

Data and variables (1/2) 7 • Sample  174 calls for tenders spanning from 2001 to 2008 Region 2001 2002 2003 2004 2005 2006 2007 2008 Tot Abruzzi 3 6 1 1 11 Aosta V. 1 1 Apulia 2 2 Basilicata 3 2 1 2 8 Calabria 1 1 1 3 Campani 1 5 2 3 1 1 2 15 a Emilia R. 1 1 Friuli V.G. 1 1 Lazio 1 4 2 1 1 1 10 Liguria 2 2 Lombardy 4 6 6 15 7 7 11 56 Marche 1 1 2 Molise 2 1 1 4 Piedmont 1 3 1 1 1 1 8 Sardinia 1 2 13 16 Sicily 2 2 2 1 7 Tuscany 1 1 2 Veneto 1 4 4 5 6 5 25 Total 1 6 26 26 37 20 22 36 174

Data and variables (2/2) 8 • Score variables  Fee (upper bound on franchise fee score)  Trans (upper bound on terms for infrastructure transfer score)  Prices (upper bound on prices score)  Infra (upper bound on new infrastructure asset score)  Serv (upper bound on service quality and organization score) • City characteristics variables  FinAut (Financial Autonomy)  Debt (Level of obligations)  Liq (Liquidity indicator)  qProceeds (Quality of proceeds)  Turnover (Political turnover indicator)  Exp (Experience of bureaucrats)  Size (size of the city)  S (Localization dummy)  Poverty (Poverty indicator)  PubNet (Public network)  Constr (Construction dummy)

The model 9 • Linear simultaneous equations model: 5                     ln Sc ln Sc FinAut Debt Liq q Pr oceeds Turnover Exp Size u 1 10 1 i i 11 12 13 14 15 16 17 1  i 1  i 1 5                     ln Sc ln Sc Constr Debt Liq PubNet Turnover Exp Size u 2 20 2 i i 21 22 23 24 25 26 27 2  i 1  i 2 5               ln Sc ln Sc Poverty Turnover Exp Size u 3 30 3 i i 31 35 36 37 3  i 1  i 3 5                 ln Sc ln Sc Constr S Turnover Exp Size u 4 40 4 i i 41 42 45 46 47 4  i 1  i 4 5             ln Sc ln Sc Turnover Exp Size u 5 50 5 i i 55 56 57 5  i 1  i 5

Results – First equation only (Franchise fee) 10 (eq by eq IV) (3SLS) -77.6800 1654.10** Debt (-0.06) (2.00) -4.70100 100.400** Liq (-0.06) (2.00) -0.64200 13.630** FinAut (-0.06) (1.98) 0.00130 -0.03140* qProceeds (0.03) (-1.87) 0.84200 -17.2500* Exp (0.06) (-1.95) -1.49600 31.1200** Turnover (-0.06) (2.08) -0.00024 0.00495** Size (-0.06) (2.04) Note : standard errors in parentheses. ***, ** and * indicate, respectively, significance levels of <1%, <5% and <10%.

Conclusions 11 • Weak evidence for taxation by regulation effect • No evidence about some other aspects influencing competitive procedure designing • Difficulties in capturing other relevant aspects because of no availability of data

12 Thank you for your attention!