Dr Justine Hastings

April 7, 2004

Prepared Statement
of
Justine S. Hastings
Before
The United States Senate Committee on the Judiciary
Subcommittee on Antitrust, Competition Policy and Consumer Rights
April 7, 2004
2
Introduction
Mr. Chairman and members of the committee, my name is Justine Hastings. I am an
Assistant Professor of Economics at Yale University and a Faculty Research Fellow at
the National Bureau of Economic Research, Program on Industrial Organization. I have a
Ph.D. in Economics from the University of California at Berkeley. Firm conduct,
competition and consumer preferences are the focus of much of my research. In
particular, my empirical research in these areas has been applied to the gasoline industry.
I have analyzed extensive data on retail and wholesale gasoline market structure and
prices for a diverse group of US metropolitan areas covering the 1990's. I have used this
data to conduct independent, academic research into the relationships between vertical
market structure and competition in gasoline refining and marketing. Two of my current
research projects include an empirical assessment of the impacts of wholesale price
discrimination on retail and wholesale gasoline prices and an empirical analysis of the
affects of gasoline content regulation on market concentration, conduct and arbitrage.
Through this research, I have gained a wealth of knowledge about the market structure of
gasoline refining and marketing. My independent research and my acquired knowledge
of the gasoline industry form the basis of my comments before this committee.
The theme of these hearings is to identify the factors that lead to increased gasoline
prices, including but not limited to increases in crude oil prices, environmental
regulation, and changes in market structure. I would like to make the following points to
the Subcommittee.
The Contribution of Crude Oil Prices to Retail Gasoline Prices
Rises in crude oil prices certainly contribute to increases in gasoline prices.
Although crude oil prices are an important determinant of gasoline prices, they
are not the only determinant. For a given crude oil price, the wholesale and
retail prices of gasoline can be significantly different across regions of the
country.
Crude Oil is the major input to gasoline production, and therefore it is not surprising that
variation in crude oil prices explains a significant amount of the variation in gasoline
prices. The following simple table illustrates this point. Although a significant amount of
variation in monthly average retail prices (excluding taxes) is explained by monthly
average crude oil prices over time within a state, it is clear that crude oil price variation
explains much more of the variation in retail prices in South Carolina, than it does in
California.
3
Table I: Fraction of Variation in Average Retail Prices Attributable to Variation in
Average Crude Oil Price 1
State
Fraction of Retail Price Variation
Explained by Variation in Crude Oil Price
Alabama 0.892
Arizona 0.714
California 0.696
Delaware 0.878
Georgia 0.907
Idaho 0.826
Iowa 0.849
Illinois 0.787
Massachusetts 0.875
North Carolina 0.906
New York 0.877
Ohio 0.852
Pennsylvania 0.900
South Carolina 0.910
Texas 0.886
The following sections discuss other factors that contribute to gasoline prices levels and
volatility.
The Effects of Market Concentration on Gasoline Prices
Fewer competitors may lead to increased market power
Higher concentration in most industries leads to the concern over an increase or
enhancement market power, which leads to higher-than-competitive price levels. Markets
across the country vary considerably in the number of companies supplying or producing
wholesale gasoline at the distribution rack. The number of competitors also has changed
over the past decade as refiners and marketers have merged, and as various forms of
environmental regulation over gasoline formulation and content have come into effect.2
Antitrust merger policy is based on the principle that price-cost margins are increasing in
the market concentration level, and market concentration is and has been a primary
consideration in merger analysis - shaping the challenges to and divestiture requirements
for mergers in the petroleum industry.
1 Data were taken from the Energy Information Administration's (EIA) website. The retail data are the
EIA's monthly average retail prices, excluding taxes, for each state from 1998-2003. The wholesale prices
were averaged from the Daily WTI spot price for crude oil posted to the EIA's website.
2 See Gilbert, Richard J. and Hastings, Justine, "Vertical Integration in Gasoline Supply: An Empirical Test
of Raising Rivals' Costs" (June 2001). UC Berkeley Competition Policy Center Working Paper No.
CPC01-21.
4
Not only number of competitors, but the identity of competitors is important for
competition. Independent refiners are uniquely important for competition.
Independent refiners do not have an incentive to raise rival's input cost to
increase retail profits.
Independent wholesalers compete intensely on price in a homogeneous
goods market with highly elastic demand.3
Because of these factors, unbranded refiners ensure sufficient unbranded
gasoline supply at competitive prices - this is necessary for the entry and
survival of independent retailers, including new chains such as KMart,
Walmart, Costco, and RaceTrac.
Unbranded wholesale markets are truly competitive. They are the only market where
gasoline is gasoline, and retailers are free to purchase from lowest price supplier.
Unintegrated refiners compete on price, and unlike integrated refiners, have no integrated
retail component that might benefit from increases in unbranded wholesale prices.
Purchasers of unbranded wholesale gasoline are free to purchase from any supplier -
enforcing intense price competition at the wholesale level.
In addition, a thick and competitive unbranded wholesale market leads to lower branded
wholesale prices in markets with many dealer-owned stations. Branded retailers who own
their own stations can choose to switch to the unbranded market (and drop their retail
brand) if their branded refiner's wholesale price is excessively higher than the unbranded
wholesale price. In this way, dealer-owned stations link competition in unbranded
markets to competition in branded markets. Vertically integrated stations (whether lesseedealer
or company-operated) do not provide this competitive link, since the stations
cannot switch between refiners over any period of time.
Independent retailers are important for competition:
They increase competition at retail level.
They allow entry into concentrated wholesale markets.
In addition, because independent retailers typically do not sell brand-differentiated
gasoline, they tend to increase local retail competition, lowering retail prices.4 In fact, in
vertically concentrated markets where refiners are able to price discriminate in wholesale
prices (charge different wholesale prices to their lessee or contract dealers), station level
wholesale prices as well as retail prices are significantly lower in the presence of
unbranded competitors.5
3 Demand is elastic since demanders (retailers) have zero switching costs only in the unbranded market.
Branded stations have positive switching costs, lowering demand elasticity of branded refiners.
4 See for example, Margaret Slade (International Journal of Industrial Organization, December 1986),
Janet Netz and Beck Taylor (Review of Economics and Statistics, February 2002) and Justine Hastings
(American Economic Review March 2004).
5 From preliminary analysis of wholesale price discrimination in gasoline market. National Science
Foundation Grant for 2003-2005 "Estimating Demand with Consumer Heterogeneity: an Application to
Wholesale Price Regulation in Retail Gasoline Markets"
5
In addition, in markets with concentrated refining capacity, producers can increase prices
above competitive levels only if there are barriers to entry. Market power at refinery level
depends on the number of refiners - but it also depends on the ability of outside
wholesalers to enter market when prices rise. Outside gasoline producers can only enter a
market if they have access to transportation, terminal and storage facilities, and a
significant number of non-captive, independent retail stations through which to sell their
product.6 It is important to note that large volume independent chains, such as RaceTrac,
amplify the ability for outside entry into wholesale markets. Because they purchase to
supply many stations (instead of a single station), they increase the ability for outside
refiners to enter the market and supply their stations.7 Antitrust and merger policy
currently considers the effects of vertical structure on market conduct and the exercise of
market power. Many of the aforementioned issues relating vertical market structure to
wholesale and retail market conduct and performance were considered in the evaluation
of the potential anticompetitive effects of recent petroleum industry mergers and in the
design of divestiture requirements associated with those mergers.
Even when market concentration as measured by firm market share is low,
each firm may exercise market power if demand is very inelastic, and overall
industry supply is at capacity.
As demand increases to consume available supply at existing refining capacity, each firm
may find itself with market power, even when the market is fairly unconcentrated. When
demand is very inelastic and supply is also very inelastic, as would be the case in
electricity markets and gasoline markets when demand is near the total production
capacity of all firms, a small decrease in supply can result in a substantial increase in
price. Hence every firm is able to affect market prices, even if every firm constitutes a
relatively small fraction of total possible output. Factors such as market segmentation due
to the environmental regulation of gasoline content may exacerbate the tightness of
supply in many regulated markets.8
The Effects of "Boutique Fuels" on Market Performance
Boutique Fuels segment markets, decreasing effective number of competitors
Reformulated Fuels Requirements change the identities of competitors and the
competitive structure of the market place.
Boutique fuels segment markets and increase refiner concentration in the following two
ways. First, if there is a supply disruption, supply cannot be imported from other regions
of the country to meet demand if other refiners in other regions of the country do not
6 See also Statement of R. Preston McAfee before the U.S. Senate, April 25, 2001, Committee on
Commerce, Science and Transportation, Subcommittee on Consumer Affairs, Foreign Commerce, and
Tourism.
7 See also Statement of R Preston McAfee before the U.S. Senate, May 2, 2002, Committee on
Governmental Affairs, Permanent Subcommittee on Investigations.
8 See the Federal Trade Commission Report on Midwest Gasoline Price Investigation, March 2001.
6
produce fuel that meets local emissions requirements. In addition, reformulated gasoline
requirements may change the identity of competitors supplying the market by decreasing
the number of unintegrated refiners competing in the unbranded gasoline market. Often
large integrated refiners choose to upgrade to supply reformulated gasoline markets, but
unintegrated refiners choose not to upgrade but to supply only conventional gasoline
markets, effectively changing the composition of competitors in the reformulated
gasoline market. The boutique fuels market will have a few, large integrated suppliers,
and few to no unintegrated suppliers. This may lead to less wholesale market competition
and higher wholesale prices for independent retailers.
Given the fact environmental regulation of gasoline content and formulation often causes
unintegrated refiners to exit the market, it is not at all clear that bringing the whole nation
under the most stringent gasoline standards will reduce price volatility. The secondary
impact of such a regulation on market structure could substantially adversely affect
market performance. It is possible that the adverse affects to market concentration could
outweigh the gains from geographic integration.9
Comments on Various Regulatory Proposals in Retail Gasoline
Wholesale price regulations such as "Fair Wholesale Pricing", "Branded-
Open-Supply", and "Zone Price Elimination" will not increase competition.
They may lead to higher average wholesale and retail prices as well.
There are several proposals that require refiners to charge the same wholesale price to
their Lessee-Dealer stations. Common names of these proposals are "Fair Wholesale
Pricing", "Branded Open Supply" or "Zone Price Elimination." I will refer to these
legislations as "Fair Wholesale Pricing" (FWP). FWP legislation would effectively force
integrated refiners to charge the same wholesale price to all of their stations.
Currently refiners charge different wholesale prices to different franchised station.
Preliminary statistical analysis suggests that refiners price discriminate based on factors
that affect local demand elasticity. Economic theory suggests that competition between
refiners is softened in markets where retailers have a small degree of market power if
refiners can price through retailers instead of directly at the pump. FWP does not change
this fact, since refiners will still set a wholesale price, which is transmitted through
retailer's pricing decisions to the final customer, thereby muting competition between
refiners. FWP will only change the profit maximizing price the refiner charges to its
stations. Economic theory predicts that wholesale prices could actually increase if
refiners are forced to charge one wholesale price. The profit maximizing single price to
all stations may actually be higher than the average of the wholesale prices under price
9 This is the topic of current research that I am conducting with colleagues at Yale and UC Berkeley into
the price effects and market structure effects of gasoline content regulation.
7
discrimination. FWP may actually raise gasoline prices - making consumers worse off
than they were before.10
Divorcement will not lead to lower prices, and may increase inefficiency.
Divorcement prohibits refiners from directly operating the stations they own, forcing
them to have the station operated instead by a dealer. Several academic studies have
presented evidence that divorcement will not lead to lower gasoline prices.11
Divorcement does not decrease entry barriers into wholesale markets, and it does not
increase competition in retail markets. Stations owned by a refiner are still integrated -
regardless of whether a refiner or a lessee-dealer sets the retail price. In addition, if
refiners have chosen company-operation at certain stations in order to minimize costs,
forcing them to convert these stations to lessee dealers may lead to higher, less efficient,
operating costs. In general, to maximize the benefit to consumers, we want to encourage
firms to lower costs and lower prices - divorcement will accomplish neither of these
goals.
Minimum Mark-up laws do not increase competition in the short-run or the
long-run. Minimum mark-up laws increase the price of retail gasoline without
increasing competition. They may also lead to inefficiencies in gasoline
retailing - they encourage an over supply of gasoline stations.
Minimum mark-up laws (or sales-below-costs laws) are currently law in several states.12
These laws typically require that retailers charge a 6 percent mark-up over cost. In the
case of gasoline, this is supposed to lead to lower prices. Requiring a minimum mark-up
will lead to higher prices in the short term if required mark-up is higher than the freemarket
mark-up. However, the goal of the legislation is to foster competition. Proponents
of this law claim that major refiners will act to predatory-price (charge price below cost)
independent retailers, forcing them out of the market. The refiners will then be able to
raise prices and increase profits. So, in the long run, prices will be lower in states with
minimum mark-up laws, because independent retailers will still be in the market,
preserving competition. So even though there is a mandated mark-up, this mark-up
prevents predatory pricing by oil companies, and preserves competition in the long run.
Empirical evidence rejects the hypothesis that these laws have acted to preserve
independent marketers. For example, Utah has had a minimum mark-up law in place
since 1987. New Mexico has never adopted this law. If the law accomplished its goal, we
would expect to see independents exiting in Albuquerque, for example, while remaining
(or even entering) in Salt Lake City. Examining the market share of independents in
10 From preliminary analysis of wholesale price discrimination in gasoline market. National Science
Foundation Grant for 2003-2005 "Estimating Demand with Consumer Heterogeneity: an Application to
Wholesale Price Regulation in Retail Gasoline Markets"
11 See for example John Barron and J. Umbeck (Journal of Law and Economics, October 1984) Justine
Hastings (American Economic Review, March 2004)
12 New York has just passed such a regulation under the New York State Motor Fuels Marketing Practices
Act, which will take effect at the end of this month.
8
Albuquerque and Salt Lake City refutes this claim. Both Salt Lake City and Albuquerque
have seen an almost identical decline in the market share of independents over the 1990s
- both by about 15 percentage points.
Not only is there empirical evidence showing that minimum mark-up laws do not
preserve competition in the manner they claim, but they may induce inefficiency in the
market. These laws benefit both independent and integrated stations. All stations,
regardless of affiliation, are guaranteed a minimum profit. This may lead to an excessive
number of gasoline stations - integrated or unintegrated. Consumers are worse off under
this legislation. It is also important to note that it is illegal for a company to require a
minimum mark-up on its own - that would be resale price maintenance.
A Final Suggestion
I would like to take this opportunity to impress upon you the following two facts: i) there
is a need for independent academic research into factors that affect petroleum pricing in
all markets and at all levels of the production chain, and ii) it is extremely difficult to
acquire data to conduct such research. Private industry data is very expensive, and there
is no single federal agency that funds economic research into energy policy, like the
National Institute for Health (NIH) does for economic research into health-related policy
questions. Perhaps we should introduce such grant programs for economists at the
Department of Energy.
In addition, the Energy Information Administration collects data, but does not have a
mechanism that allows it to be accessed by carefully screened academics at any
meaningful level of aggregation. In comparison, the Census Bureau has worked hard at
disseminating data in a range of aggregation levels, with corresponding levels of security
to protect confidentiality. They have a model program of data organization and high
security research centers that has significantly contributed to the production of high
quality research, informing a large range of public policy decisions. The adoption of this
program lead to a wealth of academic research into issues related to Labor Economics
that have been tremendously informative to policy makers. We should encourage the
development of similar programs at the government energy agencies, to increase
independent research into industries as important to our economy as petroleum and
electricity.
Summary and Policy Recommendations:
1. Crude oil price increases explain a considerable amount of increases in gasoline
prices, however, differences in market structure due to horizontal and vertical
concentration, as well as environmental regulations also contribute to increased
gasoline prices.
2. Inelasticity of demand for gasoline leads to large price increases in response to small
supply decreases. Increasing the number of refineries will expand supply and ease
9
tension in tight markets, lowering price volatility. Increasing the number of refineries
may also decrease market concentration if the new refinery means a new competitor
enters the market.
3. Optimal environmental policies should incorporate secondary impacts on market
structure and competition as well the impacts on pollution abatement in order to
maximize consumer welfare.
4. Wholesale price regulations such as "Fair Wholesale Pricing" or "Zone Price
Elimination" do not increase competition in the market place, since retail outlets
cannot switch between refiners when their refiner's wholesale price exceeds the price
of another refiner. This type of legislation may actually increase gasoline prices
5. Below Cost Pricing legislation is typically aimed at preventing unbranded retailers
such as Race-Trac, Costco, or Wawa from entering the market and/or increasing
competition. They only serve to dampen price competition, and act to maintain or
raise gasoline prices.