PDF《Consolidating炉he燱ater营ndustry》

Michael?Zschille International?Conference? Contracts,?Procurement?and?Public\Private?Arrangements ? Paris,?May?30,?2012 Consolidating?the?Water?Industry An?Analysis?of?the?Potential?Gains?from?Horizontal? Integration?in?a?Conditional?Efficiency?Framework Michael?Zschille

1 Introduction?and?Motivation

2 Methodology

3 Data?and?Model?Specification

4 Results

5 Conclusions Agenda 2012\05\30

2 2012\05\30 ? High?fragmentation?of?German?water? supply:?6211?water?utilities? (Umweltbundesamt,?2010,?p.

?75) ? Consolidation?e.g.?in?England?&

?Wales? and?the?Netherlands,?fragmentation?e.g.? in?Portugal?and?Japan ? Formerly?consolidated?water?industry?in? East?Germany?(16?utilities);

?more?than?550? utilities?after?1990 ? Evidence?for?scale?economies?in?water? supply?across?different?countries? (Saal et?al.,?2011) ? German?Monopolies?Commission?claims? for?consolidation?(Monopolkommission,? 2010) ? Evaluation?of?current?market? structure ? Determination?of?returns?to?scale? characteristics?in?German?water? supply ? Analyze?the?efficiency?impact?of? horizontal?integration?in?water? supply

3 Motivation

1 2012\05\30

4 Methodology?I?C Data?Envelopment?Analysis

2 ? Data?Envelopment?Analysis?(DEA) ? Nonparametric?approach?based?on?linear?programming ? Assumption?of?input\orientation ? Different?returns?to?scale?(RTS)?technologies?possible:?constant?(CRS),?variable?(VRS),? non\increasing?(NIRS)?and?non\decreasing?(NDRS) ? Simple?production?model ? Inputs:?number?of?employees,?network?lengths ? Outputs:?water?delivered?to?final?customers,?water?delivered?to?other?water?utilities,?number?of? connections ? Conditional?efficiency?framework?(Daraio and?Simar,?2005,?2007)? ? Necessity?to?account?for?operating?environment?in?efficiency?analysis?(e.g.?different? densities?of?supply) ? Aim?to?compare?like?with?likes ? Estimation?of?a?Kernel?function?for?the?environmental?variables ? Only?observations?with?similar?operating?environment?are?used?to?construct?the?frontier ? Consideration?of:? ? Output?density ? Share?of?water?losses ? Share?of?groundwater?input 2012\05\30

5 Methodology?II?CReturns?to?Scale

2 ? Scale?efficiency ? Scale?efficiency:?SE?=?TECRS/TEVRS ? Qualitative?analysis?of?returns?to?scale ? F?re and?Grosskopf (1985) ? Comparison?of?frontiers?obtained?under? different?returns?to?scale?assumptions ? Classification?into?firms?operating? under?IRS,?CRS?or?DRS ? Quantitative?analysis?of?scale?elasticity?in?DEA?(F?rsund and?Hjalmarsson,?2004) ? Scale?elasticity:?maximum?proportional?expansion?in?outputs?relative?to?a?proportional? increase?in?inputs?(F?rsund,?1996) ? IRS?for?scale?elasticity?greater?than?one ? Based?on?dual?formulation?of?DEA?program?(with?VRS) ? Shadow?price?on?RTS?restriction?used?to?calculate?scale?elasticity ? Not?uniquely?defined?for?corner?points,?but:?estimation?of?bounds 2012\05\30 ? Consideration?of?J firms?to?be?merged?out?of?the?full?set?of?observations?k={1,…,?K}

6 ? Merged?firm?DMUJ obtained?by?direct?pooling?of? all?inputs?and?outputs ? Potential?overall?gains?from?horizontal?integration? calculated?as: ? denotes?the?pre\merger,?conditional? technology?set ? Decomposition?into?learning,?harmony?and?scale?effects: Source:?adapted?from?Bogetoft and?Otto?(2011) Methodology?III?C Horizontal?Integration?(Bogetoft and?Wang,?2005)

2 Abbr. Min. Median Mean Max. Std.?dev. Inputs Employees [number] labor 2.00 10.00 29.85 2326.00 139.97 Network?length [km] net 21.00 164.50 329.68 7858.00 600.80 Outputs Final?water?supplies [1000m3] wdel 89.00 1154.50 3144.41 197?900.00 12?720.40 Bulk?water supplies?[1000m3] bws 0.00 1.00 264.21