Benchmarking With An Application to Electricity Distribution GAP Workshop 14 December 2005, Berlin Astrid Cullmann , DIW Berlin - PowerPoint PPT Presentation

1 / 16
About This Presentation
Title:

Benchmarking With An Application to Electricity Distribution GAP Workshop 14 December 2005, Berlin Astrid Cullmann , DIW Berlin

Description:

Data Envelopment Analysis (II) Advantages: ... Schiffer, H-W. ( 2002) Energiemarkt Deutschland, 8. Auflage, K ln, T V-Verlag GmbH. ... – PowerPoint PPT presentation

Number of Views:42
Avg rating:3.0/5.0
Slides: 17
Provided by: wip96
Category:

less

Transcript and Presenter's Notes

Title: Benchmarking With An Application to Electricity Distribution GAP Workshop 14 December 2005, Berlin Astrid Cullmann , DIW Berlin


1
BenchmarkingWith An Application to Electricity
DistributionGAP Workshop14 December 2005,
BerlinAstrid Cullmann , DIW Berlin
2
Agenda
  • 1. Overview - Benchmarking Methodologies
  • 2. Application in the Electricity Sector
  • 3. Transfer to the Airports
  • Literature

3
Overview of Benchmarking Techniques
Benchmarking
PartialApproaches(one-dimensional)
Multi-dimensional Approaches
Average Approaches
Frontier Approaches
InducedApproach
Parametric
Parametric
Non-Parametric
DataEnvelopmentAnalysis(DEA)
StochasticFrontierAnalysis(SFA)
ModifiedOrdinaryLeast Squares(MOLS)
OrdinarayLeast Squares(OLS)
CorrectedOrdinaryLeast Squares(COLS)
Total FactorProductivity(TFP)
Stochastic DEA(SDEA)
PerformanceIndicators
4
Data Envelopment Analysis (DEA) (I)
5
Data Envelopment Analysis (II)
  • Advantages
  • - Identifies a set of peer firms (efficient
    firms with similar input and output mixes) for
    each inefficient firm.
  • - Can easily handle multiple output.
  • Does not assume a functional form for the
    frontier or a distributional form for the
    inefficiency error term.
  • Drawbacks
  • - May be influenced by noise.
  • - Traditional hypothesis tests are not
    possible.
  • Requires large sample size for robust estimates,
    which may not be available early on in the life
    of a regulator.
  • ? Sensitivity Analysis by Bootstrapping

6
Stochastic Frontier Analysis (SFA) (I)
  • SFA Assumption about the residuals
  • vi are random variables
  • assumed to be iid, independent of the
  • ui usually assumed to be half normal distributed
    (truncated)
  • accounting for technical inefficiency

X
7
Stochastic Frontier Analysis (SFA) (I)
  • Specify production (or cost) function
  • Cobb Douglas
  • 2) Translog Functional Form
  • Shortcoming
  • Can handle only one output
  • ? Aggregation
  • ? Distance Functions
  • - The decomposition of the error term into noise
    and efficiency component may be affected by the
    particular distributional forms specified.

8
Agenda
  • 1. Overview - Benchmarking Methodologies
  • 2. Application in the Electricity Sector
  • 2. Transfer to the Airports
  • Literatur

9
Efficiency Analysis in the Electricity
Distribution
  • 1) Efficiency Analysis of German Local
    Distribution Utilities
  • 2) Efficiency Analysis of East European
    Distribution Companies (Poland, Hungary, Czech
    Republic, Slovakia) in Comparison to Germany
  • The Issue
  • - Increased use of efficiency analysis in the
    regulation of network industries
  • - Reform of the electricity sector Incentive
    based regulation
  • - EU Directive 2003/54/EC and German Energy Law
    (July 2005)

10
Choice of Variables
  • Inputs
  • LABOR number of employees
  • NETWORK LENGTH approximation for capital input
    (factored high-, medium- and low-voltage lines
    51.61)

Outputs UNITS SOLD (in MWh) NUMBER OF
CUSTOMERS (residential) INVERSE DENSITY INDEX
(supplied area in square kilometres per
inhabitants)
  • Number of customers is determined by industry
    and households within the supply area can be
    considered as a given date
  • Demand of the end users is quite inelastic and
    must be satisfied

Output is fix, input has to be minimized
11
Our Empirical Application
  • I) We analyze technical efficiency (no cost data
    is available, VDEW data 2001)
  • DEA is applied as main productivity analysis
    technique
  • Constant Returns to Scale (Variable Returns to
    Scale for verification)
  • Input-orientated approach
  • Input distance function approach with SFA for
    verification

II) Specify a translog functional form, general
unrestricted form Truncated normal distribution
for the technical inefficiency random
variables Specification of Battese and Coelli,
1995 Maximum likelihood method to estimate the
parameters (Frontier Version 2.1, Coelli)
12
Selected Results
  • German local distribution
  • East German Utilities more efficient
  • East European regional Distribution
  • Poland features by far the lowest efficiency
    scores
  • Scale inefficient

13
Measurement of Scale Efficiency
  • Difference Model 2, DEA VRS CRS
  • Economies of Scale seem to be limited, big is
    not necessarily beautiful
  • Evidence for economies of scale in Poland (area
    of increasing returns to scale)
  • Slovakia scale inefficiency due to decreasing
    returns to scale

14
Agenda
  • 1. Overview - Benchmarking Methodologies
  • 2. Application in the Electricity Sector
  • 3. Transfer to the Airports
  • Literatur

15
Transfer to Airport Benchmarking
  • Decide which methodologies to use
  • Stochastic Frontier Analysis not widely used.
    Integrate SFA, at least for verification and
    validation method
  • Focus on technical efficiency or allocative
    efficiency?
  • Dynamic analysis with panel data?
  • Special Issue ? technical change
  • Panel Data Models
  • Choose appropriate input and output factors
  • Difficult task ? many activities,
    heterogeneous

16
Literature
  • Aigner, Dennis J., Lovell Ashley C., Schmidt
    Peter, 1977. Formulation and Estimation of
    stochastic Frontier Production Function Models.
    Journal of Econometrics 6/1, 21-37.
  • Christensen, L.R., Jorgensen, D.W. and Lau, L.J.
    1971. Conjugate Duality and the Transcendental
    Logarithmic Production Function. Econometrica 39,
    225-256
  • Coelli, Tim, Prasada Rao, Dodla S., Battese,
    George E., 1998. An Introduction to Efficiency
    and Productivity Analysis. Kluwer Academic
    Publishers, Bostron/Dordrecht/London,
  • Coelli, Tim, 1996. A Guide to Frontier Version
    4.1 A Computer Program for Stochastic Frontier
    Production and Cost Function Estimation. CEPA
    Working Paper 96/7, Department of Econometrics,
    University of New England, Armidale NSW
    Australia.
  • Estache, Antonio, Rossi Martin A., Ruzzier
    Christian A., 2004. The Case for International
    Coordination of Electricity Regulation Evidence
    from the Measurement of Efficiency in South
    America. Journal of Regulatory Economics 25/3,
    271-295.
  • EBRD, Transition Report 2004, London.
  • Filippini, Massimo, Hrovatin, Nevenka, Zoric,
    Jelena, 2004. Regulation of the Slovenian
    Electricity Distribution Companies. Energy Policy
    32, 335-344.
  • Jamasb, Tooraj, Pollitt, Michael, 2003.
    International Benchmarking and Yardstick
    Regulation An Application to European
    Electricity Distribution Utilities. Energy Policy
    31, 1609-1622.
  • Kocenda, Evzen, Cabelka, Stepan, 1999.
    Liberalization in the Energy Sector in the
    CEE-Countries Transition and Growth.
    Osteuropa-Wirtschaft 44/1, 196-225.
  • Shephard, Ronald W., 1970. Theory of Cost and
    Production Functions. Princeton University Press,
    Princeton.
  • Frontier Economics, and Consentec (2003)
    Netzpreisaufsicht in der Praxis, Abschlussbericht
    für VIK und BDI, London.
  • Riechmann, C. (2000) Kostensenkungsbedarf bei
    Deutschen Stromverteilern, Wirtschaftswelt
    Energie, 55, 6-8.
  • Schiffer, H-W. (2002) Energiemarkt Deutschland,
    8. Auflage, Köln, TÜV-Verlag GmbH.
Write a Comment
User Comments (0)
About PowerShow.com