Strategic Pricing and Competition in g g p Retail Electricity - - PowerPoint PPT Presentation

Strategic Pricing and Competition in g g p Retail Electricity Markets in Germany

Vigen Nikogosian Tobias Veith

German Electricity Market German Electricity Market

• production and transmission level:

production and transmission level:

– E.On, EnBW, RWE and Vattenfall Europe operate the transmission network and – keep about 85 percent of production capacities

di ib i

• distribution:

– more than 850 regional separated markets (distribution networks) for household customers in 2008, – have been mainly integrated with energy providers

• retail:

–

• n average more than 40 additionally providers active in submarkets

– according to regulatory office 6 37 percent of households switched to alternative according to regulatory office 6,37 percent of households switched to alternative provider – two types of contracts: standard contract (universal service obligation): fall-back option if competitor or contract drops out of the market, higher price p p p , g p competitive contracts (main differences to standard contract):

• ffered at lower prices

either by incumbent or by new entrants y y

Contracts and Household Customers Contracts and Household Customers

7 98 TWh 7,98 TWh 6,37 % 43,89 TWh 35,03 % 73,43 TWh 58,60 % , %

Customers with standard contract Customers with competitive contract offered by incumbent Customers served by alternative supplier

Source: Bundesnetzagentur Monitoringbericht 2008

German Electricity Market

Relevant geographic market for household customers Providers offering contracts below standard contract

Usage 4000 kWh

Data provided by e‘net and verivox

German Electricity Market

Distribution Charge Standard Contract Price

728,51 - 750,00 750,01 - 800,00 800,01 - 850,00 850,01 - 900,00 900,01 - 950,00 , , 950,01 - 999,61

Data provided by e‘net and verivox

Agenda Agenda

• Literature overview
• Theoretical model
• Data and empirical model
• Estimation results
• Conclusion

R l t d Lit t Related Literature

• There is little literature on analyzing competition in retail electricity

markets

• Salies and Waddams-Price (2004)

– analyze how competition in the UK retail energy markets is affected by the market power of single providers

• prices are determined both by customer characteristics and cost factors but also

p y by incumbency

• market power in particular by former regional incumbents
• Salies (2008)

Salies (2008)

– analyzes what affects competition and thus contracts offered to private customers

• in more densely populated areas prices are significantly lower
• transmission charges both on higher voltage but also on distributional level also
• transmission charges both on higher voltage but also on distributional level also

positively affect prices

• significantly higher prices for incumbents’ contracts

Th ti l M d l Theoretical Model

• sequential price competition under the threat of potential market entry

q p p p y in differentiated product markets

• electricity supply contracts offered by a dominant provider/incumbent

and entrants and entrants

• contracts:

– standard contract at price p t (incumbent) standard contract at price pst (incumbent) – competitive contract at price p1 (incumbent) – contracts at prices p2 (newcomer) and p3 (potential entrant)

t

:

• customers:

– choose between staying with current contract (standard contract if not switched yet) and selecting alternative contract  select exactly one contract

• additional assumptions:

 constant per unit costs ci and constant distribution costs cnet  prices are strategic complements prices are strategic complements

Th ti l M d l Theoretical Model

N p N N    

 

0, 0, ,- 1,2,3, ,

i i i i i i i i

N p N N i i st i i p p p p

  

             

• structure of the game:
• 1. Incumbent determines the price for the standard contract
• 2. Newcomer and potential entrant observe the standard contract price and potential
• 2. Newcomer and potential entrant observe the standard contract price and potential

entrant decides to enter the market or not

• 3. Simultaneous (price) competition for competitive contracts between incumbent,

newcomer and potential entrant if market entry occurs

• profit functions:

1 1 1 1 1 1

( )(1 ) ( )

st

p dc c N p dc c N F         

2 2 3 3 3 3 2 2 2 3

( ) ( ) p dc c N F p dc c N F          3

3 3 3 3 1 2 3

with N N N N   

Intermediate results from theory Intermediate results from theory

entrants set prices below the standard contract price to attract

• entrants set prices below the standard contract price to attract

customers

• incumbent can affect choice space for competitors’ contracts => if

incumbent can affect choice space for competitors contracts if standard contract price reduction is sufficiently strong potential market entry can be deterred St t i f t d d t t l b d if t

• Strategic use of standard contract can only be used if customers are

sufficiently price-sensitive

• if “switchers” affected sufficiently strong relative to the (total
• if switchers affected sufficiently strong relative to the (total

incumbent’s) revenue entry deterrence might be profitable

I t di t lt f th Intermediate results from theory

• Entry deterrence is only profitable if

1 1 * A A

   

y y p => As long as demand-driven revenue effect is sufficiently positive (exceeds price reduction effect), the incumbent is better off choosing a lower standard contract price

1 1

st

p

ˆst p

 st

p

 A

p

A



st

p

1



1 



st

N



A st

N

st

N

A st

N ˆ

st

N



Data Survey Data Survey

• Cross-sectional aggregated data for all German regional electricity

submarkets and customers on zip code level (regions where two or submarkets and customers on zip code level (regions where two or more distribution grids abut are excluded)

• Data sources:

– Acxiom, customer information – E’net, distribution network information (date of collection: 08/2008) – Verivox, contract information (date of collection: 08/2008)

• selected information for three user groups with usages

1500 kWh, 2800 kWh, 4000 kWh

• Prices are total per year prices as for customers not technical

composition of prices but total price per contract and year of relevance

Descriptive Statistics

2800 kWh # Obs Mean Std Dev Min Max 2800 kWh # Obs. Mean

• Std. Dev.

Min Max # contr. bel. std. contr. 7912 92.5 29.8 5 196 # providers 7912 46.1 6.165 13 62 std contr price 7912 644.4 28.8 542.2 807.8

• std. contr. price

7912 644.4 28.8 542.2 807.8 distribution charge 7906 163.2 19.4 113.6 230.6 price diff. 7906 481.2 26.7 389.5 604.9 low contr price 7912 605.2 30.5 525.0 696.4

• purch. power/single hh

7907 69.4 67.3 581.6

• fluct. rate

7904 0.106 0.050 0.002 1.738 low contr. price lowest price 7912 524.7 7.984 393.6 574 share apartment buildings 6505 0.365 0.169 0.051 0.939

• distrib. area (lv)

7323 20.9 9.810 0.801 40.6 distrib area (lv)/area 6363 0.459 0.170 0.161 1.009

• distrib. area (lv)/area

share cable 6216 0.693 0.102 0.300 0.975 # tapping points lv 7518 0.761 0.076 0.173 0.837

Econometric Model Econometric Model

• cross sectional model for German retail markets
• cross-sectional model for German retail markets

( , , )

i i i i

y f pricediff grid customers 

st

pricediff price dc   ( , )

st i i i

price g customers grid  ( ) d h d

i i i

pricediff price dc  

• model structure: Econometrical implementation

( )

i i

dc h grid 

1

log( ) log( ) log( )

i pricediff i dc i

y pricediff dc grid customers            

1 2 2

log( ) log( )

grid i cust i st i grid i cust i

grid customers price grid customers grid dc                   

3 3

log( )

gr i i idgrid

dc      

Estimation Results – Competition Equation

log(# contracts below standard contract) log(# providers) log(# contracts below standard contract/providers) kWh 1500 2800 4000 1500 2800 4000 1500 2800 4000 kWh 1500 2800 4000 1500 2800 4000 1500 2800 4000 log(price diff) 2 581*** 2 400*** 2 276*** 0 170*** 0 012 0 072*** 2 409*** 2 412*** 2 348*** g(p ) 2.581*** 2.400*** 2.276*** 0.170***

• 0.012
• 0.072*** 2.409*** 2.412*** 2.348***

(0.022) (0.026) (0.025) (0.013) (0.015) (0.014) (0.016) (0.020) (0.019) log(distributio 0 696*** 0 326*** 0 490*** 0 186*** 0 297*** 0 315*** 0 882*** 0 623*** 0 805*** g( n charge) 0.696*** 0.326*** 0.490*** -0.186*** -0.297*** -0.315*** 0.882*** 0.623*** 0.805*** (0.034) (0.035) (0.034) (0.021) (0.020) (0.020) (0.026) (0.026) (0.027)

Estimation Results – Competition Equation

2800 kWh log(# contr. bel. std. ) log(# providers) log(contr. bel. std. / d ) 2800 kWh contr.) log(# providers) contr./providers) Competition Equation log(price diff) 2.400*** ‐0.012 2.412*** (0.026) (0.015) (0.020) log(distribution charge) 0.326*** ‐0.297*** 0.623*** (0.035) (0.020) (0.026)

• distrib. area (lv)/area

‐0.106*** ‐0.032*** ‐0.074*** (0 013) (0 008) (0 010) (0.013) (0.008) (0.010) log(purch. power/ hh size ‐0.118*** ‐0.111*** ‐0.007 (0.007) (0.004) (0.005) share multi‐apartm. houses 0.013*** 0.009*** 0.004 (0.004) (0.002) (0.003) 0 009 ‐0 005 0 014*** fluctuation rate 0.009 0.005 0.014 (0.005) (0.003) (0.004) constant ‐30.6*** 6.800*** ‐37.4*** (0.509) (0.294) (0.385)

Estimation Results – Standard Contract Price

Standard Contract Price Equation (continued) log(purch. power/ ‐0.010*** ‐0.010*** ‐0.010*** hh size) (0.001) (0.001) (0.001) ‐0 026*** ‐0 027*** ‐0 025*** share multi‐apartm. houses 0.026 0.027 0.025 (0.003) (0.003) (0.003)

• fluct. rate

0.008*** 0.008*** 0.008*** (0.001) (0.001) (0.001) log(distrib. area (lv)) 0.003** 0.001 0.004*** (0.001) (0.001) (0.001)

• distrib. area (lv)/area

‐0.009*** ‐0.010*** ‐0.009*** (0 002) (0 002) (0 002) (0.002) (0.002) (0.002) share cable ‐0.067*** ‐0.066*** ‐0.069*** (0.005) (0.005) (0.005) ‐0.055*** ‐0.051*** ‐0.055*** log(# tapping points lv) 0.055 0.051 0.055 (0.005) (0.005) (0.005) constant 6.461*** 6.465*** 6.461*** (0.009) (0.009) (0.009)

Estimation Results – Distribution Charge g

Distribution Charge Equation (continued) 0 005* 0 000 0 011*** log(distrib. area (lv)) 0.005* ‐0.000 0.011*** (0.002) (0.002) (0.002)

• distrib. area (lv)/area

‐0.053*** ‐0.055*** ‐0.049*** (0 004) (0 004) (0 004) (0.004) (0.004) (0.004) share multi‐apartm. houses ‐0.056*** ‐0.060*** ‐0.051*** (0.007) (0.007) (0.007) ‐0 372*** ‐0 371*** ‐0 386*** share cable 0.372 0.371 0.386 (0.013) (0.013) (0.013) log(# tapping points lv) ‐0.291*** ‐0.275*** ‐0.298*** (0.012) (0.012) (0.012) (0.012) (0.012) (0.012) (0.020) (0.020) (0.020)

Estimation Results – HV Zones

Standard Contract Price (4000kWh)

log(# contr. bel. std. contr.) log(# providers) log(contr. bel. std. contr./providers)

0 045*** 0 046*** 0 045***

hv zone Vattenfall

0.045*** 0.046*** 0.045*** (0.005) (0.005) (0.005)

hv zone E.On

• 0.019***
• 0.021***
• 0.020***

(0.005) (0.005) (0.005)

hv zone RWE

0.025*** 0.027*** 0.024***

hv zone RWE

(0.005) (0.005) (0.005)

hv zone EnBW

0.062*** 0.064*** 0.061*** (0.005) (0.005) (0.005) Distribution charge *** *** ***

hv zone Vattenfall

0.090*** 0.087*** 0.085*** (0.013) (0.013) (0.013)

hv zone E.On

• 0.074***
• 0.081***
• 0.077***

(0.012) (0.012) (0.012)

h RWE

0.001 0.002

• 0.003

Competition Equation

hv zone RWE

(0.013) (0.013) (0.013)

hv zone EnBW

• 0.090***
• 0.086***
• 0.094***

(0.013) (0.013) (0.013)

hv zone Vattenfall

0.843*** 0.569*** 0.274*** (0.022) (0.011) (0.019)

hv zone E.On

1.001*** 0.480*** 0.520*** (0.020) (0.010) (0.018) 1 012*** 0 512*** 0 499*** p q

hv zone RWE

1.012 0.512 0.499 (0.020) (0.010) (0.018)

hv zone EnBW

1.245*** 0.718*** 0.525*** (0.021) (0.011) (0.018)

Conclusion Conclusion

• distribution charges seem to provide potential for a barrier to entry for

id d tl i t t f f l providers and consequently an instrument for foreclosure

• difference between the standard contract price and the distribution

charge has a significantly positive effect on the number of contracts

• in contrast, the effect on the number of providers is smaller

co t ast, t e e ect o t e u be o p o de s s s a e => the standard contract price is an instrument for adjusting short-run competition Contracts can be established and withdrawn a lot easier than a Contracts can be established and withdrawn a lot easier than a provider could enter or exit a regional market due to fix/sunk set-up costs.

• future extension: impact of vertical integration on contract

features and prices

Thank you for your attention attention.