Assessing Retail Assessing Retail CompetitionCompetition

David LoomisDavid LoomisIllinois State UniversityIllinois State University

309-438-7979309-438-7979dloomis@ilstu.edudloomis@ilstu.edu

OverviewOverview

Market CharacteristicsMarket Characteristics Standard Measures of Market Standard Measures of Market

CompetitivenessCompetitiveness A Model of Customer SwitchingA Model of Customer Switching Alternative Market Share Measures Alternative Market Share Measures

- AMS and NMMS- AMS and NMMS Lessons from TelecommunicationsLessons from Telecommunications

Characteristics of a Perfectly Characteristics of a Perfectly Competitive MarketCompetitive Market

Many buyers and sellers who are Many buyers and sellers who are price takersprice takers

No barriers to entry or exitNo barriers to entry or exit Mobility of factors of productionMobility of factors of production Perfect informationPerfect information Profit-maximizing firmsProfit-maximizing firms

Characteristics of Network Characteristics of Network IndustriesIndustries

Customer is ‘attached’ to a fixed Customer is ‘attached’ to a fixed networknetwork

Network maintained by regulated Network maintained by regulated entity (may also be default provider)entity (may also be default provider)

Customer must be linked with a Customer must be linked with a provider during ‘new connect’ provider during ‘new connect’

Purchases automatically recur Purchases automatically recur unless the customer switchesunless the customer switches

Characteristics that Characteristics that Impact Market Share Impact Market Share

MeasurementMeasurement

Passive DemandPassive Demand Presence of a Default ProviderPresence of a Default Provider Imperfect InformationImperfect Information

Unique Characteristics of Unique Characteristics of Electric MarketsElectric Markets

Inelastic DemandInelastic Demand Peak CharacteristicsPeak Characteristics Transmission ConstraintsTransmission Constraints Properties of ElectricityProperties of Electricity

Standard Measures of Market Standard Measures of Market CompetitivenessCompetitiveness

StructureStructure– Size and number of competitorsSize and number of competitors

ConductConduct– Pricing and marketing strategies and Pricing and marketing strategies and

behaviorbehavior PerformancePerformance

– ProfitabilityProfitability

Structure MeasuresStructure Measures

Regular Market Share (unit/revenue)Regular Market Share (unit/revenue) Alcoa (1945) - Judge Learned Hand stated Alcoa (1945) - Judge Learned Hand stated

that 90% of a market "is enough to that 90% of a market "is enough to constitute a monopoly; it is doubtful constitute a monopoly; it is doubtful whether 60 or 64 % would be enough; and whether 60 or 64 % would be enough; and certainly 33 percent is not."certainly 33 percent is not."

Herfindahl Index - Merger GuidelinesHerfindahl Index - Merger Guidelines If H is over 1800, "highly concentrated." If If H is over 1800, "highly concentrated." If

1000-1800, "moderately concentrated." 1000-1800, "moderately concentrated." Under 1000,"unconcentrated."Under 1000,"unconcentrated."

Passive vs. Active DemandPassive vs. Active Demand

Network Industries - provider Network Industries - provider decision once, continues until decision once, continues until switchswitch

Other Markets, active choice of Other Markets, active choice of brand for each purchasebrand for each purchase

Implications for Market Implications for Market ShareShare

Traditional analysis implicitly Traditional analysis implicitly assumes active demandassumes active demand

Total Market Share Total Market Share overemphasizes past decisionsoveremphasizes past decisions

Extremely affected by how Extremely affected by how decision is introduceddecision is introduced

ModelModel

Two stage - searching and Two stage - searching and switchingswitching

H - shopping cost H - shopping cost W - incumbent brand equityW - incumbent brand equity E - effort expended to switchE - effort expended to switch

First Stage - ShoppingFirst Stage - Shopping

Customer will shop if [PI-E(PC)]qi > H + W + E

So customers shop with higher usage than

qi > (H+W+E)/(PI-PI) = qx

Second Stage - SwitchSecond Stage - Switch

Customer will switch if [PI-PC]qi > W + E

So customers switch with higher usage than

qi > (W+E)/(PI-PC) = qs

Usage M.S. is less than # of customer M.S.

Active MS and New Move Active MS and New Move MSMS

AMS describes real-time decisions, but is AMS describes real-time decisions, but is difficult to measure.difficult to measure.

NMMS approximates AMS when customers NMMS approximates AMS when customers can choose during new-connectcan choose during new-connect

Entire Customer Base

Movers Shoppers

switchers switchers

New Move Market ShareNew Move Market Share

Easy to measure: supplier choices of Easy to measure: supplier choices of new connects are easily observablenew connects are easily observable

Reflects decisions during an ‘active Reflects decisions during an ‘active choice’ timechoice’ time

‘‘Real Time’ measure showing current Real Time’ measure showing current activityactivity

Highlights the importance of Highlights the importance of operational details (sign-up procedures)operational details (sign-up procedures)

Lessons From Lessons From TelecommunicationsTelecommunications

““Success” Case – Long DistanceSuccess” Case – Long Distance ““Failure” Case - LocalFailure” Case - Local

CHART A1.2 - INDICATORS OF AT&T MARKET SHARE

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100%

84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99

Year

Minutes Revenues (FCC) Revenues (SEC)

Lines International Revenues

Lessons from Long Lessons from Long DistanceDistance

Market share for the incumbent may fall Market share for the incumbent may fall slowly and take a long time to decline to slowly and take a long time to decline to competitive levelcompetitive level

Access charges were discounted to Access charges were discounted to competitorscompetitors

Competitive state may mean only a few Competitive state may mean only a few large providerslarge providers

Short run excess profits may earned by Short run excess profits may earned by incumbent – attracts entryincumbent – attracts entry

Local Market Share GraphLocal Market Share GraphILEC Market Share by Time Period

90.00%

92.00%

94.00%

96.00%

98.00%

100.00%

1992

1993

1994

1995

1996

1997

Dec

-99

Jun-

00

Dec

-00

Jun-

01Time

Mar

ket S

hare

%-a

ge

Revenue Share

End Line Share

Lessons from Local Lessons from Local TelecommunicationsTelecommunications

Barriers to entry is the single most Barriers to entry is the single most important factor in determining important factor in determining competitioncompetition

““All competition is local”All competition is local” Mere presence of competitors is Mere presence of competitors is

insufficientinsufficient

ConclusionsConclusions Traditional Market Share measures do Traditional Market Share measures do

not accurately reflect current decisionsnot accurately reflect current decisions Active Market Share is a better Active Market Share is a better

approach, but is hard to measureapproach, but is hard to measure New Mover Market Share approximates New Mover Market Share approximates

AMS and is easy to measureAMS and is easy to measure Barriers to entry must be monitoredBarriers to entry must be monitored

ReferencesReferences

Loomis, David G. and Eric Malm, “Active Loomis, David G. and Eric Malm, “Active Market Share: Measuring Competitiveness Market Share: Measuring Competitiveness in Retail Energy Markets,” in Retail Energy Markets,” Utilities PolicyUtilities Policy, , Vol. 8 No. 4, 1999.Vol. 8 No. 4, 1999.

Loomis, David G. and Eric Malm, “The Loomis, David G. and Eric Malm, “The Devil in the Details: An Analysis of Default Devil in the Details: An Analysis of Default Service and Switching,” 20th Annual Service and Switching,” 20th Annual Advanced Workshop on Regulation and Advanced Workshop on Regulation and Competition, May 24, 2001. Competition, May 24, 2001.