Number 1 — 2009	 71
Competition Policy Newsletter
MERGERCONTROL
Fuel for thought — StatoilHydro/ConocoPhillips (Jet) 
Jérôme Cloarec, Dag Johansson, Philippe Redondo, Daniel Donath, Elzbieta Glowicka
and Cyril Hariton (1
)
Introduction(1
When faced with a proposed merger, antitrust authorities
have to assess the likelihood and the magnitude
of anticompetitive effects that may occur
following the removal of one of the merging parties
as an independent force in the various markets
affected by the transaction. These possible anticompetitive
effects must then be weighed against potential
efficiency gains. To enable it to complete this
task in the limited amount of time provided by the
legislators, the Commission collects and contrasts
information from different sources. The notifying
party’s compulsory notification (Form CO) is the
initial source of such information and contains a
description of the industry along with more specific
details regarding the affected markets. In addition,
to ensure that it has a complete understanding of
the competitive landscape in each of the affected
markets, the Commission supplements this information
with the views of other market participants
such as the merging parties’ suppliers, competitors
and customers. (2
)
Competitors are usually well informed about the
market conditions in the affected markets and the
competitive pressure that each merging party exerts
on its counterpart in the transaction. However, competitors’
views may be biased by their own interests.
For example, a competitor can welcome a transaction
that removes a very competitive market player.
In such cases, competitors’ replies to the Commission’s
market investigation may support their strategic
views of the deal rather than provide an objective
assessment of the transaction. Customers, on
the other hand, are less likely to have strategic interests
in the transaction and are therefore less likely
to provide biased responses. This is particularly the
case when there are numerous customers of modest
size. Unfortunately, as a result, such customers
are also less likely to have the necessary resources
or access to the requisite information to respond
meaningfully to the Commission’s questionnaires.
An extreme case occurs when customers are indi-
1
( )	 The content of this article does not necessarily reflect the
official position of the European Commission. Responsibility
for the information and views expressed lies entirely
with the authors.
2
( )	 From time to time, experts are directly appointed by the
Commission to examine certain (usually technical or economic)
aspects of a transaction. The Directorate-General
for Competition also often benefits from the technical
­expertise of other Directorates-General.
viduals, as it is virtually impossible to access such
customers with standard questionnaires, although
each individual may have a private (and sometimes
informed and documented) view of the likely effects
of a merger.
This was the case for the Norwegian oil company
StatoilHydro’s acquisition of Jet petrol stations in
Scandinavia (67 in Denmark, 40 in Norway and 163
in Sweden), owned by ConocoPhillips of the US
(case COMP/M.4919 — StatoilHydro/ConocoPhillips),
which was subject to an in-depth investigation
by the Commission. (3
) Therefore, in addition to
the standard market investigation that also included
gathering evidence from internal documents, the Commission
structured its market investigation around
two pillars. First, several econometric studies that were
based on an extensive request for data relating to
the daily running of the fuel retail businesses were
carried out to gauge the extent to which the two
merging parties exerted competitive constraint on
each other. Second, a customer survey was conducted
in selected countries to obtain insights into customers’
views regarding the main questions posed during
the assessment of the transaction. (4
)
This article is divided into six sections. The next two
sections describe the Swedish and Norwegian retail
fuel markets and discuss the role that Jet played in
those markets. (5
) The econometric and customer
survey analyses are described in the fourth and fifth
sections. The last section offers some general conclusions
and, in addition, makes some important
points as to the data demands that are associated
with analysing mergers where the merging firms’
customers are widely dispersed.
Sweden
StatoilHydro was the largest retail supplier of motor
fuels in Sweden, with a total of more than 1 000 fuel
stations at the end of 2007 and accounting for more
3
( )	 Relevant documents issued by the European Commission
regarding this case, including the non-confidential
version of the final decision, can be downloaded at
http://ec.europa.eu/competition/mergers/cases/index/
m98.html#m_4919
4
( )	 The opinions of some customers’ associations were also
solicited.
5
( )	 The Commission’s final decision found that the transaction
would not lead to competition concerns in Den-
mark.
72	 Number 1 — 2009
Merger control
than 30% of total sales. (6
) Jet was the country’s sixth
largest supplier of retail motor fuels: it accounted
for more than 10% of total sales in 2007 and had by
far the highest average throughput of all fuel station
networks in Sweden. Following the transaction, the
merged firm’s largest competitor, OK-Q8, would
have had a market share less than half that of the
combined entity. The second and third largest competitors,
Shell and Preem respectively, would be even
smaller with market shares of [10-20]% each. (7
) By
acquiring Jet, StatoilHydro would thus further consolidate
its position as the leading service-station
chain in the Swedish market, in which substantial
“green-field” entry of a new competitor is unlikely
due to considerable entry barriers. (8
)
The Jet business model is specific. First, it exclusively
operates company-owned and company-operated
stations only in densely populated areas under the
unique Jet brand. (9
) Second, Jet stations are all unmanned,
with a limited range of services compared
to traditional manned (or full-service) stations. Third,
contrary to most of its competitors, which offer a
wide range of fidelity and corporate cards that allow
customers to obtain price reductions such as volume
rebates, Jet has built a strong brand position based
on a transparent net pricing policy that advertises net
prices directly on the pump. The company furthermore
applied a price differential with respect to competing
full-service station networks of SEK 0.25 per
litre. The combination of these elements resulted in
Jet becoming the most efficient retail fuel supplier in
Sweden, which allowed Jet to defend the SEK 0.25
6
( )	 StatoilHydro operated full-service fuel stations under
the Statoil and Hydro brands and automated fuel stations
under the 1-2-3, Hydro and Uno-X brands. Statoil’s own
network consists of 562 fuel stations: 478 full-service stations
and 84 automated fuel stations branded “Statoil” or
“1-2-3”. StatoilHydro also owns Hydro’s fuel station network,
which comprised 426 stations at the end of 2007:
376 automated fuel stations branded “Hydro” or “UnoX”,
45 full-service stations under the Hydro brand and
5 unbranded stations (“white pumps”).
7
( )	 In addition, there are a number of smaller competitors operating
brands such as Tanka, Din-X, Pump, Gulf, Q-star,
St1 and ICA-Tapp.
8
( )	 The availability of suitable sites that can be used for building
fuel stations as well as the need to obtain permits are
crucial factors for assessing barriers to entry (as well as
barriers to expansion) in the fuel retail markets. In addition,
green-field entry involves building forecourts, installing
pumps and tanks, negotiating fuel supplies and
establishing a brand by investing in promotion and advertising.
The costs and the time required for such a strategy
are substantial.
9
( )	 For example, as stated in footnote 6, StatoilHydro operated
full-service fuel stations under the Statoil and Hydro
brands and automated fuel stations under the 1-2-3, Hydro
and Uno-X brands. In addition, stations may be operated
by a third party for a given company, or even be only associated
with a company through a marketing venture (essentially
using the company’s name along with its supply
of fuel).
per litre differential if challenged by competitors. (10
)
Jet’s strategy has been highly successful: while it operates
only 4.5% of all fuel stations in Sweden, its
market share amounts to [10-20]%.
The success of Jet’s strategy has also been recognised
by its competitors. In particular, the Swedish
retail motor fuel market has witnessed an expansion
of the number of automated stations that today represent
more than half of all fuel stations, and certain
market players have shifted their pricing strategy
from high fuel prices at full-service stations and
volume-based rebate schemes towards “net pricing”
schemes without rebates. (11
) Related to such strategic
repositioning, Shell initiated a nation-wide price war
in April 2005 in order to establish its Shell Express
brand of automated stations by applying the same
price differential for petrol in relation to full-service
stations and other automated stations that Jet (and
some other smaller automated networks) applied.
While this price war ended in August 2006 when
all automated chains started to apply the same SEK
0.25 differential to full-service stations as Jet, it was
soon followed by a second price war that started in
April 2007 for both petrol and diesel. (12
) The second
price war was initiated by StatoilHydro, which sought
to reduce the differential between full-service stations
and unmanned stations from SEK 0.25 per litre
to SEK 0.15 per litre. Jet (and other competitors)
however resisted StatoilHydro’s initiative and sought
to maintain the SEK 0.25 per litre differential.
The important role of Jet as an independent competitor
was also confirmed by the respondents to
the market investigation and StatoilHydro’s internal
documents. Several respondents to the market investigation
emphasised the fact that fierce competition
between the Statoil network, Jet and Shell had
decreased the margins and eroded the profitability
of smaller and weaker networks that were viewed
as too small to influence price levels in a manner
similar to Jet. StatoilHydro’s internal documents
suggested that (i) Jet has a very strong brand image,
and consumers perceive Jet as the cheapest supplier,
and (ii) StatoilHydro views Jet as its most efficient
competitor.
10
( )	 It is, however, important to note that Jet was not a mere
“price follower”, as Jet’s chosen differential varied over
time, and competing networks had to consider the possibility
that Jet might seek to change this differential over
time.
11
( )	 In addition, operators are also restructuring their networks
to improve efficiency, for example by closing stations
with low throughputs.
12
( )	 Diesel sales have been rapidly gaining ground at the expense
of petrol in Sweden. Although there were relatively
few diesel cars in Sweden, they accounted for 35% of new
cars sold in Sweden in 2007 (compared to 20% in 2006).
Following the introduction of a more favourable tax regime
for diesel cars in 2007, diesel sales are expected to
grow further.
Number 1 — 2009	 73
Competition Policy Newsletter
MERGERCONTROL
This evidence thus indicated that Jet is an important
competitive force in the Swedish retail fuel market
and that the transaction would remove an important
and well-established low-cost operator.
Norway
The Norwegian market is highly concentrated. The
four main players account for more than 90% of
the total sales. StatoilHydro is the largest competitor,
with a [30-40]% market share. Jet is the fifth largest
fuel station network present in Norway, with a market
share of [0-5]%. Jet’s retail outlets are located
only in the south-east of Norway, where the company’s
market share is higher, at [5-10]%. (13
) The
confinement to the south-eastern region of Norway
is related to logistics hurdles that Jet faces. (14
) Unlike
the other four competitors, Jet has no storage
depots or terminals in Norway and instead operates
a depot in Strömstad, Sweden, situated close to the
Norwegian border. It is noteworthy that while the
four main market participants have granted mutual
access to each other’s depots to limit fuel logistics
costs throughout the country, Jet has been unable to
conclude similar agreements.
Despite these logistics constraints, Jet’s purely automated
fuel network enabled the company to compete
aggressively on price, in particular compared to
the full-service fuel stations of its competitors. Jet
promoted a net pricing policy that was associated
with a price differential in relation to rivals’ manned
stations and was the most efficient competitor in
terms of average throughput, with an average fuel
volume per site twice as high as the national average.
Jet was therefore an important competitive force in
the Norwegian market, as was also confirmed by
the market investigation and in particular by internal
StatoilHydro documents. These internal documents
(i) confirmed that Jet was viewed by consumers
as the cheapest of the automated networks and
(ii) showed Jet’s capability to react to price changes
introduced by competitors. The other Norwegian
competitors to the merging parties also confirmed
that Jet exerted a strong competitive constraint
in the Norwegian market and in particular in the
south-east of Norway. Jet’s role in the Norwegian
market was also confirmed by PFC Energy’s report
for Norway (October 2007, page 50): “Jet is still the
main driver of gasoline price wars, due to its ‘low-cost’ strategy,
and has been marketing diesel at its Norwegian outlets,
which rapidly brought down prices in the diesel segment. Jet’s
13
( )	 There are some counties in south-east Norway in which
Jet’s market share is as high as [10-20]%.
14
( )	 Jet’s logistical constraints, in particular access to depots,
indicate that barriers to entry are high in Norway. No fuel
retailer entered the Norwegian market after Jet’s entry in
1992.
presence being limited to the southeast of Norway confines
much of the fierce competition to this area.”
These elements indicated that the other market
players had different incentives from those of Jet,
potentially even including a joint interest to prevent
Jet from expanding, and that Jet was an important
competitive constraint, in particular in south-east
Norway. Therefore, the transaction would remove
an independent and well-established low-cost operator
competing in Norway.
Econometric analysis
An interesting feature of the retail fuel market in
this case was the way Statoil fuel station manag-
ers (15
) set pump prices. On the one hand, Statoil’s
headquarters would set national recommended prices,
issue payment cards and undertake advertising
and promotional campaigns at national level in each
country (Denmark, Norway and Sweden). On the
other hand, individual fuel station managers could
deviate from these national recommended prices after
monitoring the prices of their local competitors
located in each station’s primary catchment area (the
so-called cluster). There is therefore a tension between
national and local aspects of the geographic
market definition. (16
) Accordingly, as in its previous
decisions, the Commission defined the geographic
market as at most national although it left open the
possibility of having smaller local markets. (17
)
The local characteristics of the market were, however,
fully taken into account in the competitive assessment
of the merger, as they lend themselves particularly
well to examining the extent of the competitive
constraint that Jet placed on StatoilHydro’s fuel
stations. If Jet’s fuel stations put competitive pressure
on StatoilHydro, one would expect to find that
15
( )	 At the time the transaction was notified, Statoil and Hydro
had just themselves merged (see case COMP/M.4545 —
Statoil/Hydro, documents available at http://ec.europa.eu/
competition/mergers/cases/index/m90.html#m_4545)
and their networks were still run separately to some ex-
tent.
16
( )	 It is often argued that with each station having a different
catchment area, eventually the overlapping of these
catchment areas would result in a chain of substitution
that would spread change of price in one place to another
that is indirectly connected through a chain of overlapping
catchment areas.
17
( )	 The definition of the relevant product markets for motor
fuel retail has also attracted some attention in recent years.
However, this transaction does not allow these issues to be
meaningfully discussed. For example, it was not assessed
whether on-motorway and off-motorway stations belong
to the same relevant market, as Jet hardly operates any onmotorway
stations. Neither does this transaction make it
possible to assess whether products such as LPG should
be considered as belonging to the same product market as
petrol and diesel for reasons related to supply-side common
distribution, as Jet does not distribute LPG.
74	 Number 1 — 2009
Merger control
StatoilHydro’s prices would be systematically lower
whenever Jet is in the vicinity of StatoilHydro’s fuel
stations. Alternatively, if StatoilHydro’s prices do not
systematically differ with Jet’s presence, this would
imply that Jet does not pose a significant competitive
constraint on StatoilHydro.
To test this empirically, it was thus necessary to at
least collect data on pump prices at StatoilHydro’s
fuel stations and information on the location of Jet’s
fuel stations relative to StatoilHydro’s fuel stations.
However, other factors may influence pump prices.
For example, full-service stations may charge higher
prices than unmanned stations to cover labour costs.
On the other hand, fuel stations located in densely
populated urban areas may face more competition
due to the presence of other fuel providers than fuel
stations in isolated urban areas and thus are more
likely to charge lower prices. If all of these factors
are not correctly accounted for, it is impossible to
correctly model the way that the managers at StatoilHydro’s
fuel stations set the pump prices and thus to
capture the “Jet effect” on StatoilHydro’s pricing.
The Commission therefore undertook an extensive
data gathering exercise. Daily prices of diesel and
95 octane petrol at StatoilHydro stations were collected
for the period from 1 June 2005 to the end of
May 2008. In addition, it was possible to collect information
on the total number of fuel stations and
to identify each competitor situated in the vicinity
of each StatoilHydro station. (18
) StatoilHydro also
submitted an extensive data set with station characteristics
(e.g. whether the stations were manned or
unmanned or whether they had a convenience store
or a car wash on their premises).
The Commission used pooled cross-sectional multiple
regression analysis to model the relationship
between pump prices at StatoilHydro’s fuel stations,
Jet’s presence and any other factors that could have
had an effect on the pump prices. (19
) In particular,
the log of monthly pump prices at each StatoilHydro
fuel station (i.e. the so-called dependent variable)
was modelled as being dependent on (i) the
presence of Jet, (ii) the presence of other competitors
such as OK-Q8, Preem and Shell in Sweden or
Shell, Esso and YX in Norway, (iii) transport costs
from StatoilHydro fuel depots to StatoilHydro fuel
18
( )	 To aid fuel station managers’ monitoring of local competitive
conditions, StatoilHydro defined clusters, which
contained a list of stations in the surrounding area of each
StatoilHydro station.
19
( )	 Ordinary cross-sectional regression analysis compares
prices at Statoil’s stations that faced Jet as a competitor
with prices at Statoil’s stations that did not face Jet as a
competitor at a given point in time. Pooled cross-sectional
regression analysis pools across different points in time.
In this instance, as the daily price data was aggregated to
monthly levels, the regressions were pooled across the different
months.
stations to correct for differences in operating costs
across the different clusters, (iv) an indicator variable
that captured whether the station is manned or unmanned,
(v) the total number of stations in the cluster
to account for the general level of competition,
and (vi) the Rotterdam price index for diesel and 95
octane petrol to account for the general movement
of fuel prices over time. (20
)
The regression models were estimated for both
Norway and Sweden for diesel and 95 octane petrol
separately, as both products were subject to different
market dynamics. For example in Sweden, the
95 octane petrol market was the subject of two price
wars, while the diesel market was only subject to a
single price war. On average, the results suggested
that both StatoilHydro’s petrol and diesel prices in
Sweden were systematically lower by [0-5]% whenever
a Jet fuel station was located in the vicinity of
a StatoilHydro fuel station. The same was true for
Norway. It is important to note that, in total, three
different regression specifications were estimated for
each country and each product to ensure that the regression
results were sufficiently robust. The econometric
analysis therefore indicated that Jet appeared
to place an important competitive constraint on StatoilHydro’s
pricing in both Sweden and Norway.
Customer survey
Another method of gauging the role that Jet played
in the Norwegian and Swedish retail fuel markets
was a survey, which is particularly useful when customers
are individuals such as in cases of retail and
service markets. A customer survey takes into account
the views of individual customers by carefully
choosing a sample of customers that is representative
of the whole population and asking them questions
regarding a specific issue. The replies of the
respondents can then be aggregated to approximate
the views of the whole population of customers.
The use of customer surveys in merger analysis
started only recently: Ryanair/Aer Lingus was the
first merger case where a customer survey of this
type was conducted. (21
) In this case, the resulting
answers from the survey helped to assess the degree
of substitutability between the two airlines.
20
( )	 The presence of Jet was captured by a dummy variable
that equalled one whenever Jet was present in a cluster
and zero otherwise. The same technique was also used for
the other competitors such as Shell. Similarly, the indicator
variable that captured the manned status of the fuel
station equalled one when the station was manned and
zero otherwise.
21
( )	 See case COMP/M.4439 — Ryanair/Aer Lingus, documents
available at http://ec.europa.eu/competition/mergers/cases/index/m88.html#m_4439,
and in particular Annex I
to the final decision, which describes the survey, provides
an analysis of the answers and addresses methodological
issues.
Number 1 — 2009	 75
Competition Policy Newsletter
MERGERCONTROL
Two key issues need to be addressed when designing
a customer survey. (22
) First, it is important to carefully
select the pool of respondents to ensure that
the sample of respondents is representative of the
whole population in order to be able to substitute
the views of the respondents from the sample for
the views of the population. This is usually accomplished
by surveying a sufficiently large group of
random respondents, as the larger the surveyed group
the more likely it is that it will contain all the different
views that are present in the population. For
example, for the purposes of this case, 1 250 Swedish
and 1 001 Norwegian motor fuels consumers
were interviewed by telephone or over the internet
in June and July 2008.
The second key issue are the actual questions to be
asked in the survey. One of the main difficulties was
to identify a limited number of core issues to be
tested in a single one-shot survey. Thus, after drafting
an initial set of questions, the Commission discussed
them with the notifying party to ensure that
the questions were sufficiently clear and effective in
assessing the effects of this merger. The questions
focused on key issues raised during the assessment
of the case, including the reasons why consumers
fill up their cars at a particular fuel station, customers’
specific brand preferences and their perceptions
of Jet vis-à-vis its competitors. For example, to understand
consumers’ fuel purchasing behaviour, respondents
were asked to rank the following three
factors in deciding whether to fill up at a particular
fuel station: (i) price of fuel, (ii) distance from home
or work, and (iii) the availability of convenience
shopping for food, tobacco and other products.
They were also asked what price differential per litre
would prompt them to change stations and were
given six choices. To gauge the effect of Jet as well
as the effect of the merger, respondents were asked
(i) how they rated Jet’s fuel prices compared to competitors’
prices, and (ii) what impact the acquisition
of Jet by StatoilHydro would have on competition
between the remaining fuel station chains in Norway
and Sweden.
Given that different consumer groups can be expected
to behave differently, it is important to ask
an additional set of control questions to identify
possible customer segmentation. While all respondents
were members of households with a car at their
disposal, the control questions allowed the Commission
to differentiate between those driving a lot and
those driving little, between different types of fuel
22
( )	 It is also important to note that the survey must be outsourced
to professional research companies that have the
know-how and experience to access a representative group
of respondents in a timely manner to ensure that the survey
is completed within the short time span of merger
proceedings.
used (petrol, diesel, ethanol, other) or the type of
fuel card used, if any. For Norway, given that Jet
was only located in the south-eastern region of the
country, it was also important to differentiate the
respondents depending on the area that they live
in. Of the 1 001 customers in the sample, 506 (or
50.5%) were from south-east Norway, which roughly
corresponds to the proportion of the total Norwegian
population living in that part of the country.
The customer survey showed that in both countries
Jet was perceived as a low-price supplier of motor
fuel. In Sweden, 40% of respondents who expressed
an opinion stated that Jet always charged lower prices
and 29% indicated that Jet sometimes charged lower
prices. (23
) The survey also revealed that consumers
were price-sensitive: 54% in Sweden and 78% in
Norway ranked price as the most important factor
when deciding where to fill up. (24
) It confirmed that
customers found it easier to compare prices when
net pricing schemes instead of rebate cards were
applied by fuel selling stations: 61% of Norwegian
respondents found that net pricing schemes facilitated
comparison of prices “to a very large extent” or
“to a relatively large extent”. (25
) Finally, as many as
54% of Norwegian customers and 50% of Swedish
customers believed that the proposed merger, along
with the disappearance of the Jet brand, was likely to
reduce competition in the retail fuels market, while
only 10% of Norwegian customers and 14% of
Swedish customers thought that that merger would
have a positive impact on competition. (26
) This is
consistent with customers’ view that Jet played a specific
role in driving price competition.
It should be noted that the answers differed between
the total population in Norway and the population
in south-east Norway. For example regarding
Jet’s prices compared to competitors’ pump prices,
while 40% of the overall sample considered that Jet
was always or sometimes offering lower prices and
46% had no opinion, the perception of Jet being
always or sometimes cheaper increased to 64% and
the number of respondents without an opinion decreased
to 22% when only respondents located in
south-east Norway were considered. This divergence
in views provides further evidence that regional aspects
of retail fuel markets must be taken into account
in addition to national market considerations.
23
( )	 32% of the respondents did not know how Jet’s fuel prices
compared with competitors’ pump prices. Unless otherwise
specified, the remainder of this section displays percentages
of respondents among those who expressed an
opinion.
24
( )	 1% of the Norwegian respondents and 8% of the Swedish
respondents had no opinion.
25
( )	 11% of the Norwegian respondents had no opinion.
26
( )	 16% of the Norwegian respondents and 10% of the
­Swedish respondents had no opinion.
76	 Number 1 — 2009
Merger control
Conclusion
The acquisition by StatoilHydro of ConocoPhillips’
Scandinavian retail fuel businesses (Jet) required the
Commission to assess the effects of the transaction
without having access to the customers of the
merging parties. This is because the customers in
this instance are individuals, and it is thus virtually
impossible to reach them using standard questionnaires.
In addition to the standard market investigation
that also included gathering evidence from internal
documents, the Commission therefore based
its market investigation on econometric studies and
customer surveys. By doing so, the Commission was
in a position to assess the relevance and likelihood
of alternative possible theories of harm, as well as
their likely effects, and came to the conclusion that,
in the Swedish and Norwegian markets for retail
sales of motor fuels, the transaction would raise serious
doubts as to its compatibility with the common
market and the EEA agreement.
To alleviate the Commission’s concerns in Sweden,
StatoilHydro proposed to divest a network consisting
of 158 unmanned stations, including 118 of the
most efficient stations currently operated by Norsk
Hydro and 40 Jet stations. This “remedy fuel station
network” would have the third highest average
throughput among all competitors in Sweden. The
remedy network would have a geographic coverage
throughout Sweden, with a strong presence in the
south. This network also covered the vast majority
of the clusters affected by the transaction. In addition,
StatoilHydro offered to divest the entire Jet
network in Norway to remove the Commission’s
concerns in that country.
The investigation of this transaction nicely demonstrates
that the absence of sizeable customers
does not prevent the Commission from collecting
relevant information necessary to reach an appropriate
and carefully balanced outcome. Customer
surveys and econometric studies are useful tools
that can effectively supplement the Commission’s
standard market investigation. Performing customer
surveys in the course of merger proceedings is inherently
constrained, among others things by (i) the
limited number of questions that can be addressed
by a survey, (ii) the methodological hurdle related
to avoiding insidious questions or questions subject
to broad interpretation, and (iii) the procedural
time constraints. The econometric analysis usually
requires highly reliable data and proper care must be
taken when modelling the competitive interactions
between the merging parties and their competitors
to ensure that the resulting econometric results are
sufficiently robust. To ensure that the results from
both the survey as well as the econometric analysis
can be considered meaningful, it is thus important
that the Commission and the notifying party engage
in a dialogue as early as possible in the notification
process (preferably at the pre-notification stage), to
discuss data and timing issues as well as the analyses
that can be undertaken.