i

Fast Moving Consumer Goods

Competitive Conditions and Policies
Financial support for this paper has been provided by the Economic Policy
Research Institute (TEPAV/EPRI) within the framework of the the
FIAS/Competition Authority/TOBB project on “Competition Policy and the
Improvement of Investment Environment in Turkey: Sectoral/Institutional and
Legal Framework”.

Aydın Çelen
Competition Authority
Department III
Ankara Turkey
[email protected]

Tarkan Erdoğan
Competition Authority
Department III
Ankara Turkey
[email protected]

Erol Taymaz
Department of Economics
Middle East Technical University
Ankara Turkey
[email protected]

TEPAV | EPRI
Economic Policy Research Institute
December, 2005

ISBN 9944-927-02-3

Matsa Basımevi
Ankara, Turkey +90 (312) 3952054

Ugur Mumcu Street No: 80/3 G.O.P. Ankara, Turkey

Phone: +90 312 446 0409 Fax: +90 312 446 3006
www.epri.org.tr [email protected]

ii
iii
CONTENTS

Executive Summary ............................................................... vi

1. Introduction..........................................................................1
2. The market ..........................................................................3
3. Market dynamics .................................................................9
4. Retailers’ conduct..............................................................17
5. The future ..........................................................................33
6. Issues for competition policy .............................................37
7. Conclusions.......................................................................41
References ............................................................................43
Tables and Figures ...............................................................45

iv
v
Executive Summary

Fast moving consumer goods (FMCGs) constitute a large part of consumers’

budget in all countries. Retail trade in these products, that is, their supply to
households, has attracted considerable interest from consumers and policy-
makers because a well-functioning retail sector is essential for daily provision of
these essential products at high quality and low cost.

The retail sector for FMCGs in Turkey is in the process of a drastic transformation.
New, “modern” retail formats, like chain stores and hyper/supermarkets, have
rapidly diffused in almost all major urban areas, and increased their market share
at the expense of traditional formats (grocery shops, green groceries, etc.) in the
last couple of decades. Moreover, alternative consumption forms, like restaurants,
catering services and hotels, rapidly increase their share in the consumption of
FMCGs. These processes change the structure and competitive conditions both in
FMCGs retail trade and other vertically related sectors.

The first immediate impact of the transformation of the retail trade of FMCGs in
Turkey is observed in new competitive conditions. Traditional formats are rapidly
losing their market share, and new formats (chain stores, hyper markets, and
supermarkets) tend to set the rules of competition. The level of concentration is
increasing slightly but it is still very low compared to European countries. Local
supermarkets and discount chains operating small supermarkets (BIM is the best
example) have been quite successful in gaining market shares at the expense of
traditional formats in recent years. Since most of the traditional retailers tend to
operate informally, the size of informal sector tend to shrink by the growth of
organized retailers. These trends are likely to have a positive impact on product
diversity and the quality of products/services offered by retail stores.

Some of the marketing practices applied by retail companies (price flexing, listing
fees, slotting fees, etc.) are potentially anti-competitive. However, these practices
do not yet seriously distort competition in the retail market, because retailers seem
to lack a significant degree of market power even at the local level. The raise of
large retailers could even be a counter-balance against the market power of some
suppliers who could have significant market power in certain segments of FMCGs.
The retail market in Turkey seems to be quite competitive. There are no legal
restrictions on entry, and no discrimination against foreign companies.

vi
The raise of large retailers has led to the proliferation of private labels. The share
of private label products is expected to increase in the future. Private label
products can help some medium-sized suppliers to be more competitive against
large suppliers who have established brands.

The transformation of the retail market is likely to have a long-lasting impact on

wholesale trade and the distribution of FMCGs as well. Traditional wholesalers are
the most likely losers, because large retailers tend to buy directly from suppliers.
Logistics companies that provide a wide range of complementary services will play
an increasingly more important role in the distribution of FMCGs.

Turkey has almost no regulatory restrictions that impose unduly barriers against
the establishment and operation of retail companies by domestic and foreign
investors. Foreign entry in the retail market was observed since the early 1990s,
and some of the large multinational companies established their braches in Turkey
either through majority-owned subsidiaries or joint ventures with large domestic
companies. New foreign entry in the market is expected. However, the market is
quite competitive and firms operate on a very thin profit margin. Therefore, a
shake up/consolidation in the market is likely. After the first version of this report
was written Carrefour acquired Gima and Endi from Fiba Holding (the decision
was announced on May 3, 2005). Subsequently, Migros of Koc Holding, that failed
to acquire Gima and Endi, announced on August 22, 2005 that it would buy
Tansaş from Doğuş Group. Further consolidation through mergers and/or exits in
the market could be expected. The process of consolidation is likely to effect local
supermarkets to a significant extent. The lack of land near city centers and the
exclusion of “hypermarkets, shopping centers and car parks” from investment
incentive may restrict the establishment of new large hypermarkets, and
encourage growth through medium-sized and small supermarkets, as it is the case
in the last five years.

Although no retailer seems to establish a dominant position in the national market,

it is possible that some retailers may gain a dominant position in certain local
markets, because the relevant market in the retail sector should be defined locally
rather than nationally. Local market power could be a problem especially following
a merger activity and/or exits. Therefore, the Competition Authority needs to pay a
close attention to changes in the retail market.

vii
1. Introduction
Fast moving consumer goods (FMCGs) constitute a large part of consumers’
budget in all countries. Retail trade in these products, that is, their supply to
households, has attracted considerable interest from consumers and policy-
makers because a well-functioning retail sector is essential for daily provision of
these essential products at high quality and low cost.

The retail sector for FMCGs in Turkey is in the process of a drastic transformation.
New, “modern” retail formats, like chain stores and hyper/supermarkets, have
rapidly diffused in almost all major urban areas, and increased their market share
at the expense of traditional formats (grocery shops, green groceries, etc.) in the
last couple of decades. This rapid transformation has raised concerns about
competitive conditions in the sector.1

This study is aimed at to shed light on competitive conditions prevailing in the

FMCGs retail trade sector in Turkey. We analyze how the structure of the market
is being transformed in recent years by new retail formats. The study is focused on
the analysis of competitive dynamics (inter-firm rivalry, pricing and non-price
policies, barriers to entry, regulatory conditions, etc.) within the sector, and draws
lessons for competition policy. Since the FMCG retail sector is closely related to
suppliers (FMCG producing industries), other services (most importantly,
wholesale trade), and users of FMCGs (hotels and restaurants), the backward and
forward industry linkages are also taken into account.

The study is based on four sources of information. First, we extensively use official
statistics collected by the State Institute of Statistics (SIS). Although the SIS
provides comprehensive data on the retail trade sector and supplier industries
(number of firms, employees, production, foreign trade, etc.), the data are not up-

1
Competitive conditions in the retail trade for fast moving consumer goods received considerable interest in
many developed countries as well. As a result of concern s raised by the public and consumer organizations,
competition authorities in developed countries conducted specific studies on competitive conditions and anti-
competitive practices in this sector. One the most comprehensive studies was conducted for the UK
Competition Commission (2000). For competition issues in the retail sector, see Mazzarotto (2001) and
Dobson et al. (2001).

1
to-date (most of the data are not available beyond 2002). Second, we conducted a
series of interviews with the main observers and actors both in the private
(FMCGS retailers as well as suppliers) and public sectors. Interviews provided
very valuable information on various business practices and competitive dynamics
in the sector. Third, we conducted a comprehensive survey, partly to get
quantitative evidence on the issues raised by the interviewees. A list of 100 main
retailers and about 200 suppliers was collected. Two questionnaires, one for
retailers and the other one for suppliers, were prepared and the survey was
conducted in the fourth quarter of 2004. The response rate was about 50 percent
for retailers and 40 percent for suppliers. Finally, we used the HTP Household
Consumption Panel data to analyze market share dynamics and pricing behavior.

The study is organized as follows: Section 2 presents the data on the structure of
the FMCG retail market. Sector 3 summarizes recent changes in the markets
(market dynamics). Section 4, drawing on the survey and HTP data, describes the
conduct of retailers and suppliers, and analyzes the implications for competitive
conditions. Section 5 discusses likely changes that can be observed in the future.
After a brief discussion on competition policy issues in Section 6, the last section
summarizes main findings of the study.

2
2. The Market

The retail market for fast moving consumer goods (FMCGs) consists of various
retail channels. The International Standard Industry Classification (ISIC, Revision
3) classifies retail channels into seven categories at the 4-digit level: ISIC 5211
retail sale in non-specialized stores, ISIC 5219 other retail sale in non-specialized
stores (department stores, etc), ISIC 5220 retail sale of food, beverages and
tobacco in specialized stores, ISIC 5231 retail sale of pharmaceutical and medical
goods, cosmetic and toilet articles, ISIC 5251 retail sale via mail order houses,
ISIC 5252 retail sale via stalls and markets, and ISIC 5259 other non-store retail
sale. Since there is no firm in categories ISIC 5251 and 5252 in Turkey, they are
excluded from our analysis.

Table 1 presents the summary data on the retail sector2 in Turkey for the period
1997-2001.3 The data on wholesale sectors (5121 wholesale trade in agricultural
raw materials and live animals, and ISIC 5122 wholesale trade in food, beverages
and tobacco) are also included in the table.

The retail sector in Turkey sold $ 29.8 billion worth of goods in 1999. Its
contribution to GDP amounted to $ 6.7 billion.4 The value of goods sold declined
sharply in 2001 (21.9 billion) because of the severe economic crisis in that year.
The Turkish lira depreciated almost by 100 percent whereas manufacturing prices
increased by 67 percent. The value of goods traded by the wholesale industry
experienced a similar decline from 1999 ($ 23.1 billion) to 2001 ($ 19.4 billion).
The retail sector employed 231 thousands people, and the number of people
engaged in the sector (paid workers plus owners, self-employed and unpaid family
workers) was 580 thousands in 2001. The retail sector, together with the

2
Unless otherwise stated, since our study is focused on FMCGs, the “retail sector” refers to only those
sectors that trade FMCGs (ISIC 521 and 522), and excludes other sectors such as retail trade of
pharmaceutical and medical goods, textiles, clothing, footwear, household appliances, hardware, paint and
glass (ISIC 523), retail sale of second-hand goods in stores (ISIC 524), retail trade not in stores (mail order
houses, etc., ISIC 525) and repair of personal and household goods (ISIC 526).
3
The State Institute of Statistics conducted the Census of Businesses in 2002, but the results were not
available as of January 2005.
4
The share of rent expenses and interest payments in total value added was around 4 percent and 1 percent,
respectively.

3
wholesale sector, provides employment for 690 thousands people. In other words,
it is one of the leading employment generation sectors in Turkey.

There were about 282 thousands retail and 20 thousand wholesale establishments
in 2001, i.e., in an average retail establishment, there are only two people working,
whereas an average wholesale trader operates with 5.5 people. The sector shrunk
more than 10 percent in real terms in 2001, but the number of establishments and
the number of employees/ engaged people increased slightly in the same year,
thanks to its flexibility.

The retail sector is closely related with agriculture and FMCG supplying
industries.5 Total value added created by the agriculture sector was $ 27.2 billion
in 2000 (Table 2). FMCG supplier industries added $ 7.2 billion. Agriculture
employs almost 35 percent of all working people in Turkey (about 7.5 million
people). Since a significant part of the population lives in rural areas and are
engaged in agricultural production, a large part of agricultural goods are consumed
there. The FMCG supplying industries employed 203 thousands people in 2000
(down from 216 thousands in 1998). These industries lost further 10,000 jobs
during the economic crisis in 2001.

The retail sector provides households essential consumption goods. However,

these same products are consumed by households as services provided by hotels
and camping sites (ISIC 5510) and restaurants, bars and canteens (ISIC 5520 that
also includes catering activities and take-out activities). These services purchase
FMCGs from wholesale and/or retail trade outlets and substitute for consumption
at home. These two sectors’ sales for private domestic consumption were about $
5 billion in 1998.6 Thus, hotels and restaurants demand a considerable amount of
FMCGs and they provide these goods embodied in their services to households as

5
The following industries are included in FMCG supplier industries (ISIC Revision 3): 1511 meat and meat
products, 1512 fish and fish products, 1513 fruit and vegetables, 1514 vegetable and animal oils and fats,
1520 dairy products, 1531 grain mill products, 1532 starches and starch products, 1533 animal feeds, 1541
bakery products, 1542 sugar, 1543 cocoa, chocolate and sugar confectionery, 1544 macaroni, noodles,
couscous, 1549 other food products, 1551 spirits; ethyl alcohol, 1552 wines, 1553 malt liquors and malt,
1554 soft drinks, mineral waters, 1600 tobacco products, 2101 pulp, paper and paperboard, 2102 corrugated
paper, containers, 2109 other articles of paper and paperboard, 2424 soap and detergents, cleaning
preparations, perfumes.
6
It is calculated from the 1998 Input-Output table.

4
substitutes. Moreover, the share of these sectors in total FMCG consumption
tends to increase. Total output of hotels and restaurants increased in dollar terms
73 percent in only three years, from 1997 to 2000, whereas the sales of the retail
sector grew 31 percent in the same period.7

Turkey imported, on average, $ 1.9 billion worth of agricultural products annually in

the period 1998-2003, and its average annual export revenue from agricultural
products was about $ 2.4 billion in the same period. It is a net exporter in food
products ($ 1.9 billion exports vs $ 1.3 billion imports), and a net importer of paper
and paper products. It exported somewhat more soap and detergents, cleaning
preparations, and perfumes than it imported in the last 6 years. The most
important imported food items are meat and meat products and vegetable and
animal oils and fats (total import value in 2003 was almost $ 1 billion).

An analysis of the market structure in supplier industries is necessary to

understand the performance of the retail sector. The data on concentration rates
(4-firm concentration ratios,8 CR4) are shown in Table 3. The 4-firm concentration
rates are higher than 50 percent in fish and fish products (53), dairy products
(696), starches and starch products (913), cocoa, chocolate and sugar
confectionery (933), macaroni, noodles and couscous (173), spirits and ethyl
alcohol (539), wines (31), malt liquors and malt (329), soft drinks and mineral
waters (763), tobacco and tobacco products (3143), and soap and detergents,
cleaning preparations and perfumes (1215).9 In order to determine the degree of
concentration in domestic supply, we need to check the level of concentration in
imports, and the share of imports in domestic supply as well. The SIS data on
concentration in imports show that 4-firm concentration ratio in imports, i.e., the
share of 4-largest importers in total imports, exceed 50 percent in only macaroni,
noodles and couscous, and beverages (all four sub-sectors). Since imports make
up less than 25 percent of domestic supply of all FMCG-related products (with one
exception, pulp, paper and paperboard), foreign trade does not likely to have a

7
Professional catering services is one of the fastest growing sectors in Turkey. Although there was no
catering firm among the largest 500 private firms in 2001 (listed by Capital journal), two catering firms,
Sofra and Sodexho were ranked 290th and 346th in 2003.
8
Since the SIS collected data at the establishment level, concentration ratios are calculated for
establishments. The data excludes private establishments employing less than 10 people.

5
major impact in reducing the market power of domestic suppliers.10 Thus, we
conclude that seller power could be a problem for retailers, especially for small
ones, for the aforementioned products.

The SIS does not calculate concentration rates for retail sectors. However, we
have collected sales data from major retailers for the period 2000-2003. We have
estimated 4-firm concentration rates for the non-specialized retail trade in stores
sector (ISIC 521) by assuming that the largest chain stores in our sample do not
compete with specialized retail sector (ISIC 522 and 523) and non-store retail
trade (ISIC 525). Our estimates suggest that 4-firm concentration rates in the non-
specialized retail trade in stores was 10.8 percent in 2000 and 11.5 percent in
2001.11 The level of concentration is much lower in Turkey than in many European
countries, but is expected to increase gradually as a result of the increasing
market share of and the wave of mergers between large retailers.12

Finally, we will look at the financial performance of the retail sector and FMCG-
related sectors. Table 4 presents the data on profit margin (operating
profits/turnover).13 As may be expected, the profit margin is very low in retail trade
(ISIC 521): it was around 1 percent in 1997 and 1998, but it became negative in
1999 and continues to be negative. Retail trade of new goods in specialized stores
(retail trade in pharmaceutical and medical goods, textiles, clothing, footwear,
household appliances, hardware, paint and glass) had a much higher profit

9
Sales values in 2001 are provided in parentheses to give an idea about the size of the market.
10
As a result of the customs union with the EU, tariff rates for imports of industrial products from the EU are
equal to zero percent. Tariff rates for agricultural products are rather high, about 50 percent for live animals
and animal products, 39 percent for vegetable products and 23 percent for edible oils (trade weighted
averages for 2003). Tariffs for imports from other countries are slightly higher than those from the EU
(Togan and Taymaz, 2005). High tariff rates for these products help large domestic suppliers to protect their
market power.
11
According to the HTP data, the share of four largest chains (BİM, Migros, Tansaş and Gima) in total
FMCG sales in 2003 was 8.8 percent.
12
Among the European countries in the late 1990s, the lowest 5-firm concentration rates for food retailing
are observed for Italy (30 percent) and Spain (38), whereas the highest rates are observed in Sweden (87
percent) and Finland (96 percent). The rates for the UK, Germany and France were 67 percent, 75 percent
and 67 percent, respectively. The level of concentration increased in the 1990s in almost all major European
markets (Dobson Consulting, 1999: 45).
13
The data are collected by the Central Bank of the Republic of Turkey. Unfortunately, there are no data for
ISIC 522. The main drawback of the CBRT data is its coverage. It includes only those establishments that
applied for a loan from the banking system. Profit margin can be calculated from the SIS data as well. Profit
margin as defined by value added minus wage payments (including imputed wages for unpaid family
workers and self-employed) to sales ratio was around 10-13 percent for the retail sector (ISIC 521) in the
period 1997-2001.

6
margin, around 10 percent in the period 1997-2002. Profits margin in wholesale
trade is somewhat higher than the margin in retail trade (5.8 percent in wholesale
of agricultural raw materials, and 1.9 percent in wholesale of food, beverages and
tobacco). Among the FMCG-supplier industries, other chemical products and
tobacco have the highest profit margins.

Profit rates (profits before tax/equity ratio) are highly correlated with profit margins.
Retail trade in FMCG experienced a sharp decline in profit rate in 2001 (-43
percent), and sustained substantial losses in 2002 as well. However, retail trade in
non-FMCG (ISIC 523) has had quite high profit rates throughout the period under
investigation. Wholesale trade has a high profit rate (on average, around 25
percent in the period 1998-2002), and tobacco and other chemicals are among the
most profitable FMCG-supplying industries.

Although there are a few observations, there is a discernible positive correlation

between profitability measures and 4-firm concentration ratios (average values for
1999-2001). Highly concentrated sectors, like tobacco and other chemicals, score
well in profitability measures. Incidentally, retail trade in FMCG (ISIC 521) has the
lowest concentration rate and it is one of the least profitable sectors.

7
8
3. Market Dynamics

We have seen that the level of concentration in retail trade in FMCG is quite low
compared to the European countries. However, the concentration data provides a
snapshot of the sector without much information on the underlying dynamics.
Therefore, in this section, we will analyze market dynamics, i.e., entry and exit
processes with a special emphasis on entry by foreign firms, and changes in the
composition of the industry by retail type. (The process of internationalization of
retailers in Turkey is extensively studied by Tokatli and her colleagues, see Tokatlı
and Boyacı, 1997; Tokatlı and Özcan, 1998; Tokatlı and Eldener, 2002. For a
comparison between the retail market in Turkey and other emerging markets of
Europe, see Tokatli, 1999).

The market dynamics is to a large extent determined by the regulatory framework.

The Australian Productivity Commission (APC), in collaboration with the Australian
National University, has measured restrictions on trade in services for a number of
countries in the world.14 The OECD has also compiled a large database, the
OECD International Regulation Database, for various sectors, including wholesale
and retail trade (see, for example, Boylaud, 2000; Boylaud and Nicoletti, 2001). In
this study, we use the APC database to compare Turkey with various categories of
economies because it covers a large number of countries and summarizes
regulations in index form.

The trade restrictiveness index is calculated for two types of supply (domestic and
foreign) and two types of activities (ongoing operations and establishment of new
businesses). It covers all distribution services, i.e., wholesale and retail trade (ISIC
51 and 52). Table 5 presents the data on restrictiveness index scores for Turkey
and average values for four country categories (developed countries, EU-15, Latin
America and Asia). The domestic index scores for all country groups are quite low.
In other words, there are not many restrictions on establishment of domestic retail
firms and their ongoing operations. The domestic index score for restrictions on

14
The database was downloaded in December 2004 from the Australian Productivity Commission website:
http://www.pc.gov.au/research/rm/servicesrestriction

9
establishment is zero for Turkey, i.e., Turkey does not impose any serious
restriction that may impede the establishment of domestic retail firms. The
domestic index score for restrictions on ongoing operations is slightly higher than
the average of other countries, mostly because of insufficient protection of
intellectual property rights (IPRs).

Turkey seems to have minor restrictions on the establishment of foreign retail

firms: the index score is only 0.031 (much lower than the EU-15 average, 0.153;
other developed countries, 0.094; Latin American countries, 0.080; Asian
countries, 0.176). In other words, contrary to other countries, Turkey does not
discriminate against foreign firms in the retail sector. The index score for
restrictions on ongoing operation of foreign firms in Turkey is comparable to those
observed in other countries (0.096 for Turkey vs 0.086 for EU-15, 0.071 for other
developed countries, 0.074 for Latin America, and 0.105 for Asian countries).
“Insufficient protection of intellectual property rights” is again the main factor
contributing to the foreign index in Turkey. Turkey seems to restrict “movement of
people” that imposes additional restrictions for foreign firms.

Turkey has introduces a number of changes in protecting IPRs in recent years.

Turkey introduced a number of laws on the protection of patent rights, industrial
designs, geographical indications, and trademarks, and ratified the Patent Co-
operation Treaty, and Nice, Vienna and Strasbourg Agreements, and specialized
courts on IPRs were established. Turkey has become a member of the European
Patent Convention in 2000. The new law on foreign direct investment (No 4875,
enacted on June 5, 2003) guarantees national treatment for foreign firms
established in Turkey and allows 100 percent foreign ownership in almost all
sectors.15 Moreover, the law on work permits for foreign nationals (No 4817,
enacted on February 27, 2003) has reduced the administrative burden on getting
work permit, and opened up a large number of occupations to foreign citizens. We
can conclude that, with the recent legislative changes, Turkey has lifted almost all
restrictions in retail sector for domestic and foreign investors.

15
There are some exceptions defined in sector-specific laws. The exceptions are: (i) broadcasting, where
foreign shareholders’ equity participation is restricted to 25 percent; and (ii) aviation, maritime

10
The regulatory framework has changed so as to create an environment favorable
to entry of new companies, and efficient operation and growth of existing ones.
Turkey has a number of policy tools to promote investment in various sectors,
activities and/or regions. The Decree of the Council of Ministers on investment
incentives (No 2002/4367, June 10, 2002) provides the legal basis for state
support schemes. The Regulation on the implementation of the decree (No
2002/1, published in the Official Journal on July 3 2002), defines the administrative
procedures, and clarifies the types of investment activities that can benefit from
state support. The Regulation (Appendix 6, A.11) explicitly states that investment
in “hypermarkets, shopping centers and car parks” in any region will not benefit
from any investment incentive. Although the exclusion of hypermarkets and
shopping centers from the state support program is potentially an unfavorable
amendment for large chains companies, they did not raise much concern about
it.16

There is no reliable data on entry and exit in the retail sector. However, since the
average establishment size is very low (only 2 people per establishment), the
turnover rate is expected to be high. In spite of high turnover and the economic
crisis in 2001, the number of establishments in the retail sector has continuously
increased since 1997.17 For example, there were 260175 establishments in retail
trade18 in 1997, and it increased to 267370 in 1998, 273057 in 1999, 279329 in
2000, and 281911 in 2001. The number of establishments increased in other retail
types and wholesale trade as well.

Although there is a slight increase in the number of retail establishments, the

market has been transformed by the entry and diffusion of “organized” or “modern”
retailing (chain stores, hypermarkets and supermarkets). Migros-Turk, established

transportation, ports, fish processing and telecommunications services provided under concession
agreements, where foreign ownership is restricted to 49 percent.
16
An interviewee claimed that major chain stores have already invested in hypermarkets, and do not plan to
open many more hypermarkets in the future. Therefore, they were indeed in favor of excluding hypermarkets
from the coverage of investment incentives scheme because it may restrict entry into this segment of the
market. In other words, this amendment is favorable for incumbent chain stores.
17
However, AC Nielsen (2004) estimates that the number of retailers declined continuously from 176
thousands in 1996 to 143 thousands in 2003. The number of chains, hyper and supermarkets and specialists
(gas station markets, dry fruit vendors, etc.) increased in the same period. For the survival strategies of small
retailers, see Özcan (2000).
18
It includes sectors ISIC (Rev. 2) 5211, 5219, 5220, 5231 and 5259.

11
in 1954, was the first retail joint-venture between Municipality of Istanbul and the
Federation of Swiss Cooperatives in Turkey (Tokatlı and Boyacı, 1997: 105).
However foreign partner withdrew in 1975 and majority shares of the company
were transferred to Koç Holding. The first supermarket chain in Turkey, Gima, was
established in 1956 as a public undertaking. It was privatized in 1993, and sold to
the partnership of Bilfer and Dedeman. The majority shares of the company were
later sold to Fiba Holding in 1996 (Tokatlı and Özcan, 1998: 92). Another main
retailer, Tansaş, was set up in 1973 by Izmir Municipality. It was also privatized in
1996 and Doğuş Holding purchased the majority of its shares.

The 1990s witnessed entry by foreign firms. As the first foreign retailer, Metro
International entered into market in 1988. (Although Metro adopted the cash-and-
carry format, it is regarded as a “retail” store serving mostly small scale shops and
households.) Carrefour and Promodes, French retailers, entered into market in
1991 and 1992, respectively (Tokatlı and Boyacı, 1998: 6). Carrefour entered into
retailing by establishing partnership with Sabancı Holding (CarrefourSa) in 1996.
The merger by parent companies of Carrefour and Promodes affected the Turkish
retail market, and CarrefourSa acquired Continent in 2000. After operating ten
years in Turkey, Metro planned to set up a joint venture with Migros in 1998.
Although the Competition Board granted a conditional permission, the merger was
not realized. Metro established a new retailer, Real, in 1997, which was followed
by Dia in 1999 and BİM in 2000. The last major foreign entry occurred in 2003. A
leading retail chain store in the UK, Tesco, entered into the Turkish market in 2003
by merging with Kipa, a regional retailer. The industry analysts suggest that new
foreign retailers, for example, Wall Mart, are planning to enter into the Turkish
retailing market.19

Most of the large domestic chains that entered into the retailing sector in the
1990s, are members of business groups that operate in FMCG-supplier industries

19
The data on foreign firms in Turkey have been collected by the Undersecretariat of Treasury. Foreign firms
started to enter in large numbers in the mid-1980s. The number of foreign entrants in FMCG-related sectors
reached 60 firms per year in the 1993-1997 period. By the end of 2004, there were more than 1000 foreign
firms operating in FMCG-related sectors, and most of them (more than 80 percent) were majority-owned
foreign companies. Hotels and restaurants sector are leading in terms of foreign entry, and food and hotels in
terms of total capital. (The list of all foreign companies can be downloaded from the Treasury web site:
www.hazine.gov.tr.)

12
and service sectors (see Table 6). There seems to be a tendency towards both
vertical integration and horizontal integration. For example, Sabancı and Koç
groups are active in supplier industries and hotels and restaurants sector. In the
retail sector, they have different retail formats (hypermarkets, supermarkets and
discount stores) operated under different brands names. Other retailers also tend
to be active in various retail formats and own vertically related firms (Özcan,
2001).

As noted earlier, we have conducted a survey of large FMCG retailers and FMCG-
suppliers in Turkey, and received responses from 51 retailers and 79 from
suppliers. Table 7 presents the data on the time of establishment of these
companies that can be used as an indication of entry into the retail market by large
companies. Our survey data show that most of the largest retailers operating in
2004 were established in the 1990s. More than 60 percent of retailers were
established in the 1989-1998 period whereas those established before 1989
represent only 20 percent of retailers. 13 of 51 retailers who responded to our
survey belong to a (domestic) business group, and 5 are foreign-firms.20 Most of
companies belonging to business groups and multinational companies were also
established in the 1990s. There is no new large-scale entry since 1998.

The entry pattern of FMCG-suppliers is quite different than the one observed for
retailers. More than half of suppliers who responded to our survey were
established before 1984. This is valid for suppliers who belong to business groups
as well. The difference in entry patterns in retail and supplier sectors indicate that
FMCG retail sector, once considered to be dominated by traditional retailers, has
become quite appealing for large companies, and attracted significant amount of
large-scale entry in the 1990s. However, the FMCG-supplying industries seem to
be dominated by large, old (established) companies who do not face with entry
competition, as evident in high concentration rates in most of these industries.

The transformation of the retail market brought by entry of large (chain) companies
can be observed in FMCG purchasing patterns of consumers. We look for

20
Only one foreign firm operating in the retail sector did not respond to our survey. However, it is owned by
a business group, and its sister retailer company responded to our survey.

13
purchasing patterns of four groups of consumers, categorized by their socio-
economic status as AB, C1, C2 and D groups,21 because chain markets operating
mainly hypermarkets and large supermarkets may incline to serve well-to-do
consumers who can afford to travel these stores by their own cars.

Figures 1a-1d depict purchasing patterns of four categories of consumers,

respectively. AB group consumers used to purchase their FMCGs mostly from
traditional grocery shops. The share of grocery shops in total FMCG expenditures
of the AB group was more than 35 percent in 1999 but it declined steadily to 22
percent in 2004. The main winners are local supermarkets, and DVFV and kiosks,
gaining about 6 and 3 percentage points market share, respectively, in the last five
years. Thus, local supermarkets have been the largest retailer for AB group of
consumers since 2002. The single most important chain that achieved a
considerable increase in its market share is BİM, a hard discount store. Its market
share for AB group increased from 4 to 12 percent. Migros and Şok seem to be
the losers for the AB group. Their total share declined from 14.2 percent to 8.7
percent in the same time period.

Grocery shops have been the main retailer for the C1 group but its share declines
continuously (from 50 percent in 1999 to 35 percent in 2004). Local supermarkets
(from 24 percent to 31 percent) and BİM (from 4 percent to 12 percent) are the
main winners in this market as well. Migros has not been an important outlet, and
its share fluctuated around 4 percent.

The C2 group behaves as the C1 group. Grocery shops have a declining share
(from 53 percent to 42 percent), and local supermarkets (from 18 percent to 31
percent) and BİM (from 5 percent to 10 percent) increased their shares.

21
We use the HTP Household Consumption Panel data for this purpose. HTP runs the household panel since
1996 and it currently covers 4,900 households. The panel covered 12 provinces until, and expanded to 27
provinces in 2002 to represent all Turkey. We use the data on only 12 provinces because chain stores have
established mainly in large provinces. Since the FMGC market is mainly a local market, it is better to focus
on main provinces to analyze competition dynamics between retail types. We would like to thank Güntaç
Özler, Vural Çakır, Kerem Soğukpınar, Erdem Çiğdem and Ayşe Pancar of HTP and the Retailing Institute
for their generosity in sharing their data and knowledge with us.

14
The D group experienced possibly the most dramatic transformation. Grocery
shops had the dominant share for this group of consumers in 1999 (about 65
percent), but lost their market drastically (48 percent in 2004). Local supermarkets
(from 14 percent to 27 percent), BİM (from 4 percent to 10 percent) and DFV and
kiosks (from 2 percent to 7 percent) had increased their shares substantially.

Although the trends are similar for four categories of consumers (declining shares
of grocery shops, increasing shares of local supermarkets, BİM, and, to some
extent, DFV and kiosks), the levels of market shares of retail formats are still
significantly different across consumer groups. For example, the share of
traditional outlets (grocery shops and open bazaar) was only 22.4 percent for AB
group in 2002, it was 35.1 percent for C1 group, 43 percent for C2 group, and 48.6
percent for D group. In a similar way, chain stores have a larger share in FMCG
expenditures of well-to-do consumers. Total share of all chains (BİM, CarrefourSA,
Dia, Gima, Kipa-Tesco, Migros, Real, Şok and Tansaş) was 35.8 percent for AB
group and 28.1 percent, 18.9 percent and only 15.3 percent for C1, C2 and D
groups, respectively. However, chain stores excluding BİM have not been
successful in increasing their market shares in the last five years. The combined
market share of all chain stores excluding BİM increased only a few percentage
points for AB and C1 groups, and declined almost the same amount for C2 and D
groups. These findings indicate that large chain stores, as a group, has gained
some market share in the early- and mid-1990s, especially in those markets
serving AB and C1 groups, but their market share has been stabilized. However,
local supermarkets and BİM continue to increase their market shares at the
expense of traditional retail formats.

The informal sector is considered as one of the main obstacles for the
development of the “modern” retail sector because informal sector firms, by not
paying social security contributions and any taxes, gain an unfair competitive
advantage against firms obeying laws and regulations. Although it is almost
impossible to measure the exact size of the informal sector, its share in the market
can be estimated roughly.

15
According to the SIS 2003 Household Labor Force Survey, there were 4.1 million
people employed in wholesale and retail trade, hotels and restaurants, and 42
percent of these people were not covered by any social security system. Thus,
more than 40 percent of the people employed in “wholesale and retail trade, hotels
and restaurants” are employed “informally”.

We have asked in our survey the firms about the extent of the informal sector.
Almost three quarters of retail and suppliers firms indicated that they compete with
the informal sector firms. These firms estimated that the market share of the
informal sector was around 30 percent in retail and 20 percent in FMCG supplier
sectors. These estimates are somewhat lower than the estimates derived from the
HLFS22, but they indicate that informal sector continues to play and important role,
especially in the retail sector.

In spite of the rapid development of the modern retail formats in Turkey in the last
decade23, their market share is still very low compared to the European countries.
According to AC Nielsen data (AC Nielsen, 2004: 20), the markets shares of
hypermarkets and big supermarkets reached 36.0 percent and 23.3 percent,
respectively, in European countries in 2002 whereas their total share in Turkey in
the same year was only 22 percent (11 percent for hypermarkets and 11 percent
for big supermarkets). Among the European countries, the lowest total share of
hypermarkets and big supermarkets was observed in the Netherlands (25
percent), Austria (29 percent) and Greece (36 percent), and the highest shares in
Portugal (62 percent), the UK (76 percent) and France (77 percent). These
findings indicate that modern retail formats (hypermarkets and big supermarkets)
are likely to increase their market shares substantially in Turkey.

22
Informal sector firms are usually small firms with low turnover/employee ratio. Therefore, the size of the
informal sector estimated by survey respondents is consistent with the share of employees without any social
security as estimated by the Household Labor Force Survey.
23
High inflation rates in the 1990s were one of the main reasons behind the rapid increase in the market share
of modern retail formats (hypermarkets and supermarkets) which could make large financial profits due to
their superior financial management skills. We thank our referee for this comment.

16
4. Retailers’ Conduct

Large chain stores have entered into the FMCG retail sector in Turkey in the early-
and mid-1990s and have transformed its structure. Although their share in the
national market is still small, they may have significant market power in localities
they operate because the FMCG market is basically a local market. There are
different definitions for the geographical dimension of the market, but it is
commonly accepted that a customer is not likely to move more than 30 km for
shopping. Therefore, large retailers’ conduct should be analyzed to shed light on
possible abuses of market power. Since retailers sell FMCGs without any further
processing, we will focus on pricing behavior and supplier-retailer relations.

We have conducted interviews with about 20 large retailers, and on the basis of
our findings, designed two surveys, one for retailers and the other one for FMCG-
suppliers to get information about retailers’ conduct and retailer-supplier relations.
We received responses from 51 retailer and 79 from suppliers. The responses
rates were 50 percent and 40 percent, respectively. Table 8 presents the data
about the coverage of these surveys. According to the SIS statistics, total sales
value of the “non-specialized retail trade in stores” sector (ISIC 521) was $ 14.3
billion in 2001. Total sales of 44 firms who provided the sales data for 2001 for our
survey was $ 2.6 billion. In other words, the surveyed firms account for at least
18.3 percent of non-specialized retail trade.24 The SIS data are not available for
2003. However, HTP estimates total FMCG retail sales in 2003 as $14.4 billion.
Chains, discounters, hypermarkets and supermarkets sold 4.6 billion worth of
FMCG in the same year. Thus, according to HTP data, our sample of firms covers
30.3 percent of total sales and 94.6 percent of sales by chains and
supermarkets.25 These comparisons suggest that our sample firms provide a good
coverage of large retailers. The coverage ratio for FMCG-supplier industries is
also quite satisfactory (18.4 percent of sales in 2001). The retail firms’ turnover in
our sample was around $ 100 million, and they employed, on average, about 900

24
Since the surveyed firms are large firms, they account a smaller share of employment.
25
Some of the surveyed firms sell non-FMCGs as well. Therefore, the coverage rates could be slightly
overestimated.

17
people. Suppliers are slightly smaller ($ 78 million turnover and 750 employees
per firm in 2001).

In order to compare our results across firm size, we have classified firms into three
categories, small, medium-sized, and large. Since more retail firms provided their
employment data, we used the number of employees in 2003 for classification for
the retail sector (small means employing less than 150 people, medium-sized 150-
499 people, and large 500 and more people). However, for those firms we have
sales data, the classifications based on employees and turnover were almost the
same. For suppliers firms, we used turnover data for 2003 for classification (small
means turnover less than $20 million, medium-sized $ 20-50 million, and large
more than $ 50 million). The cut-off values are chosen such that firms are more-or-
less equally distributed in these three categories.

Table 9 shows the distribution of firms across size categories. All five foreign firms
in the retail sector are large firms. Moreover, all but one firm that belong to
business groups are medium-sized and large firms. There are 9 foreign firms in
our sample of suppliers (11 percent of all firms). There are two small, one medium-
sized and five large foreign suppliers. Suppliers belonging to business groups are,
on average, large firms.

There is a strong positive correlation between retailer size and store size (see
Table 10). Small retailers do not own any hypermarket, and concentrate on
supermarkets and small supermarkets (store area between 100-1000 m2). A few
medium-sized retailers operate some hypermarkets, but their preferred type is a
supermarket with 400-1000 m2 store area. Almost all hypermarkets are operated
by large retailers who have also a large number of small supermarkets. Foreign
retailers operate either hypermarkets or small supermarkets.

An analysis of large domestic and foreign retailers at the firm reveals clear
differences in firm strategies. More than 80 percent of small supermarkets of large
domestic firms are operated by only two firms, whereas there is only one foreign
firm that operates small supermarkets. In other words, most of large firms, either
domestic or foreign, concentrate on operating relatively small number of large

18
stores (hypermarkets, large supermarkets, etc.) whereas a few large firms have
been following the strategy of opening a large number of small supermarkets all
around the country.

Estimates on total store area indicate that large retailers have a dominant position.
11 large domestic retailers had 58 percent of total store area whereas 4 large
foreign retailers had 28 percent in 2003.

The number of products sold by a retailer changes positively by size as well. Small
and medium-sized retailers sell about 9000 and 12000 products, respectively.
Large retailers sell a large number of products, about 24000.

The survey questionnaire included questions that define the “relevant market” for
retailers. Three aspects of the market, consumers’ socio-economic status, retail
format, and geographical market, are used to define the “relevant market”. There
seems to be no difference between small and large retailers in terms of serving
different categories of consumers. Large retailers claim to serve all categories
more, but there is not any specialization towards serving any specific consumer
group. As may be expected, all retailers indicate that “supermarkets” constitute the
main competitive form. 75 percent of large retailers consider hypermarkets as a
part of their market. It is interesting to observe that discount markets and
cash&carry are closely related with large retailers’ markets.

Geographical aspect seems to be main aspect that defines the “relevant market”
for retailers. Almost all small and medium-sized retailers consider their market as
local (only one province), or regional whereas half of large retailers consider the
market as a national market. Moreover, the market is conceived as “international”
by one third of large retailers (5 firms). It is interesting that only one foreign retailer
considers its market as “international” whereas others (4 firms) consider it as
national and/or regional.

Firms were asked to provide data on the number of entrants into and exits from
the market in which they operate. Most of the firms (60 percent) could not respond
to this question. The average values for the number of entrants and exits were 3.6

19
and 2.5 for retailers, and 12.3 and 12.1 for supplies. Firms estimated their market
share in 2003. Interestingly, small retailers estimated larger market shares,
probably because of the fact that, as note earlier, they consider their market local.
In the case of FMCG-suppliers, large firms claimed to have larger shares, as one
would expect, because suppliers’ compete mainly at the national level.

FMCG retailers sell (or could sell) almost identical products. Therefore, price-
competition is likely to be very important. If they have local market power, they
would be able to raise their prices. Price comparisons could be helpful in
identifying if some retailers (mostly, chain stores) enjoy a certain degree of market
power. However, it is notoriously difficult to make a price comparison across
retailers and/or retail type because each retailer sells a different basket of products
and the number of products sold is very large. To mitigate the effects of
differences in product mix, we calculate relative prices for baskets of products sold
by each retailer/retail type as follows:

RPjt = ∑ pijt qijt / ∑ pit* qijt

i∈I j i∈I j

where RPjt is the relative price for retailer j at time (quarter) t, pijt the price of
product i sold by retailer j at time t, Ij the set of products sold by retailer j, and qijt
the quantity sold. p*it is the average price of product i at time t, and is calculated as
follows:
pit* = ∑ pijt qijt / ∑ qijt
j∈J j∈J

Thus, RP compares the cost of the basket of products purchased from retailer j to
the amount the consumer would pay for the same basket had s/he bought it at
(weighted) average of prices observed in the market, i.e., the amount a random
buyer would pay.

We calculated relative prices for various retail types and all national chain stores
for three groups of products, food, personal care products, and cleaning
products.26 Figure 2a presents the data on relative food prices for the period

26
We use the HTP data in calculating relative prices. The dataset includes price and quantity data for those
products that were sold by most of the retailers in almost all quarters under investigation (1999:1-2004:3).

20
1999:4-2004:3 (we use 4-quarter moving averages to mitigate the effects of
quarterly fluctuations). The data reveal consistent and persistent patterns in
relative food prices. For example, BİM, the hard discount store, had the lowest
relative food prices throughout the period. A BİM customer paid on average 8
percent less than what s/he would pay at average market prices. Other discount
stores, Dia and Kipa had also low prices. CarrefourSA, known for its aggressive
pricing strategies, was among the low cost chains. Gima and Migros had a
tendency to raise their prices relative to the market, and has become the most
expensive chains in 2004 (the price differentials were 6 percent for Gima, and 3
percent for Migros in 2004). Tansaş, Real and Şok chains increased their relative
prices especially in the period 2000-2003. It is remarkable to find that DFV and
kiosk and grocery stores have been 3-4 percent more expensive, and local
supermarkets 1-2 percent cheaper throughout the period. In other words, grocery
shops that are usually blamed for operating informally are not low-cost outlets for
consumers.27

The relative price data on personal care products reveals somewhat different
trends (Figure 2b). For these products, BİM chain and cash&carry and open
bazaar formats provide low cost alternatives. Real, Gima, Tansaş and Migros
chains are among the most expensive providers of personal care products in
recent years. Grocery shops are again relatively more expensive (1-2 percent).
Local supermarkets were relatively expensive in 1999 and 2000, but have reduced
their relative prices, and get close to the average in 2004. There seems to be an
upward trend in almost all relative prices. This is caused by the substantial shift
towards low-cost retailers.

Cleaning products have a different ranking. Open bazaars experienced a sharp

decline in relative prices of cleaning products after 2001 crisis. Carrefour and

27
A recent study by McKinsey Global Institute (2003) indicates that there are substantial productivity
differentials between “traditional” (for example, groceries) and “modern” (hypermarkets, supermarkets, etc.)
retail formats. It is estimated that the “modern” retailers are almost three times more productive than
“traditional” retailers (“productivity” is measured as value added per hour worked). Similarly, the SIS data
indicate that “large” retailers are 3-4 times more productive than “small” retailers. Although the productivity
differential between traditional and modern retailers is huge, the impact of productivity differentials on prices
is much smaller because the share of wage payments in turnover (sales revenue) is only about 2-5 percent,
and the share of value added in turnover is about 20 percent in the retail sector. Moreover, the traditional
retailers reduce their costs to some extent by avoiding tax and social security payments.

21
cash&carry were low-price leaders in 2004. Gima, Migros and Tansaş, as in the
case of other products, were relatively more expensive. Prices at local
supermarkets, grocery shops and BİM fluctuated around the average values.

There are two conclusions that can be drawn from price comparisons: first, price
differentials are not trivial. It was about 14-15 percent in all products (food,
personal care and cleaning products) in 2004. Considering quite thin profit margins
in the retail sector, these differences seem to be quite significant. Second,
consumers are fairly price-sensitive in food products:28 low-price retailers, most
importantly, local supermarkets and BİM have increased their market shares (for
all consumer categories), whereas grocery shops lost substantial market share in
the last five years.

Since there are quite significant price differentials between retail types and
customers have different shopping characteristics that depend on their socio-
economic status, prices paid by different groups of consumers may differ. This
issue is certainly important from welfare point of view.

We calculated relative prices paid by different groups of consumers as follows:

RPct = ∑ pict qict / ∑ pit* qict
i∈I c i∈I c

where RPct is the relative price paid by c group of consumers, and p*it is the
average price:
pit* = ∑ pict qict / ∑ qict
c∈C c∈C

There seems to be no difference in relative food prices paid by socio-economic

groups. The price differential was widened in 2001, after the economic crisis that
hit probably hardest the D category, but even in that year, the differential remained
less than 2 percent. Although well-to-do consumers, presumably the AB group,
could pay higher prices for food, and the D group is more sensitive to food prices,
they turn out to be paying almost the same price because the D group goes to

28
The share of food in total FMCG expenditures is about 85 percent.

22
grocery shops for shopping, but as noted earlier, grocery shops are relatively
expensive.

The behavior of relative prices for personal care and cleaning products is
completely different. The price differential between socio-economic groups was
quite small in 1999 and 2000 (only a few percentage points), but it widened up
rapidly after the 2001 economic crisis, and reached almost 15 percent in the case
of cleaning products. It is apparent that price sensitive poorer consumers (D
group) have substituted cheap, no-brand products after the economic crisis for
more expensive brands (recall the low prices in open bazaars), whereas well-to-do
consumers (the AB group) could afford more expensive brand-name products.

We have established that average prices differ consistently and considerably

across retail types. However, price differences do not directly imply market power.
Since the FMCG markets are local, chain stores face with different competitive
conditions in different local markets. Therefore, those companies that have market
power may differentiate their prices across local markets. This behavior, called
“price flexing”, can be an indicator for the exercise of market power.

We calculated weighted average of standard deviation of prices by retail type as

follows:

PF jt = ∑ wijtδ ijt
i∈I j

where PFjt is the degree of price flexing by retail j at time t, wijt the share of ith in
total sales of retailer j at time t, δijt the coefficient of variation of ith product prices
for retailer j during the period t. Thus, the PF variable shows the degree of price
dispersion (price differentiation) by retail types.

Figure 3a shows the data on price flexing in food products. All retail types
experienced the same trend: after a slight increase in the degree of price
differentiation until the middle of 2001, there was a sharp decline in the second
half of 2001 and throughout 2002. The decline in price differentiation continued at
a slower pace in 2003 and 2004. Since we calculated the PF index by using

23
quarterly observations, any change in prices within the quarter will increase the
index value. In other words, the general pattern of changes in the PF values
follows closely changes in inflation rates. When the rate of inflation is high, price
changes will be more frequent/ large, leading to a higher PF value.

A closer analysis of PF values reveals that traditional FMCG outlets, open bazaars
and grocery shops had higher price differentiation than all chain stores. Local
supermarkets and cash & carry stores had also high price differentiation values.
All chain stores had much lower values throughout the period. There is not any
permanent ranking of chain stores in terms of the PF values. Thus, it seems that
chain stores have either more stable prices or they do not differ much their prices
across their shops. Prices differ more across open bazaars, grocery shops and
local supermarkets. The lowest PF values are found among DVFs and kiosks that
mainly sell items whose prices are usually set at the national/regional level
(tobacco products, beverages, etc.).29 Thus, there is not much scope for price
differentiation for these shops.

A different pattern emerges in the case of personal care products. Grocery shops,
local supermarkets and cash & carry stores had higher PF scores for personal
care products, but some chains, most notably, the market leaders, Migros and
Şok, had also quite high PF values in 2004. There seems to be price flexing in
these chain stores as high as the one in traditional outlets.

Since there was not sufficient number of observations for a number of chain
stores, the PF values for cleaning products were calculated for a small set of

29
TEKEL, the state monopoly company, used to have a monopolist position by law in the markets for
alcoholic beverages and tobacco products. The market was gradually opened to competition, and a regulatory
agency was established in 2002. Moreover, the alcoholic beverages division of TEKEL was privatized in
2003. The law grants the Tobacco and Alcoholic Beverages Board the right to set prices for those firms
whose production, sales or imports are below a certain threshold level (the threshold level was 1 million liters
per year and could be gradually reduced by the Board in the next five years). Large firms whose capacity is
higher than the threshold level are free to set their own prices. There are similar restrictions in the tobacco
products market. Whereas large firms with an annual capacity to produce at least 2 billion cigarettes or
15,000 tons of other tobacco products per brand may freely import, price, distribute, and sell that brand, the
price and marketing principles for small importers are determined by the Board. The first attempt to privatize
the tobacco division of TEKEL in 2004 was unsuccessful. The tobacco products market is highly
concentrated. There are three multinational firms (Philip Morris Sabancı (PMSA), JTI and British American
Tobacco) that share half of the market (the other half is served by TEKEL). PMSA and JTI were 6th and 23rd
largest private firms (in terms of turnover), respectively, in Turkey in 2003.

24
chains. As in other cases, traditional retailers (open bazaar and grocery shops),
local supermarkets and cash & carry stores had higher price differentiation than
main chain stores.

The data on price flexing indicate that there are some differences in product prices
across different stores. These differences are higher among traditional outlets and
local supermarkets that do not have common ownership. Therefore, local
conditions seem to matter more for these retailers. In the case of chain store, there
seems to be some price flexing, especially in personal care products.

The data on prices give some information about the outcome of retailers’ conduct.
Therefore, in order to get more information about pricing behavior, the survey
included questions on what determines sale prices, and how firms apply price
flexing.

The retailers who participated in our survey suggest that input costs have a strong
impact on sales prices (Table 12). Almost all retailers consider the impact of input
costs on sales prices as “strong”. The second most important determinant of sales
price is other retailers’ prices (half of respondents consider that other retailers’
prices have a “strong” impact on their own prices). Large retailers take into
consideration the demand for their products, and small retailers their stocks in
pricing decisions. These findings suggest that retailers’ may have a certain degree
of flexibility in setting their prices, i.e., they may have some market power. FMCG-
suppliers prices, i.e., input prices for retailers, are determined largely by input
conditions, the label on the product (private label), and payment conditions.
Moreover, there is not any significant difference between small and large suppliers
in term of pricing decisions.

Since retailers (and, to some extent suppliers) are concerned about their
competitors’ prices, we asked them if they regularly monitor competitors’ prices. All
large retailers and almost all of small and medium-sized retailers (about 90
percent) said that they monitor other retailers’ prices regularly. Interestingly,
supplies are also likely to monitor retailers’ prices (85 percent).

25
Half of retailers acknowledge that they differentiate their prices across their stores
(Table 13).30 Moreover, there is a monotonic increase by size in the share of price
flexing firms. Although only 33 percent of small retailer said that they apply price
flexing, the ratio increases to 53 percent for medium-sized retailers, and 75
percent for large retailers. The proportion of price differentiating firms is much
higher in FMCG-supplying industries (68 percent). Thus, price flexing observed in
traditional retailers and local supermarkets may be caused by price differentiation
by suppliers that have some market power.

Other regional retailers’ prices seem to be the main determinant of price flexing.
85 percent of firms that acknowledge price flexing consider local competition the
main reason for price flexing. Prices set differently at new stores. Half of firms
suggest that regional demand is important. It is interesting to observe that
“regional cost differences” is found to be the least important reason among price
flexing. FMCG-supplying firms revert to price flexing because of regional cost
differences (65 percent of firms) and as a reaction to other suppliers’ prices (60
percent of firms). These findings indicate that there is a strong relationship
between price setting and the degree of local competition in the retail sector.

In addition to price flexing, price reduction through promotions is a widely-adopted

strategy in the retail sector. Half of all retail firms indicate that they always apply
promotions, and most of the remaining firms have promotions at least once a
month. Price changes are also made frequently (at least once a month). Small
retailers adjust prices either more frequently or adopt state-dependent pricing
policies (change prices as a response to changes in input costs). These findings
support our conclusion on the stability of prices in large chain stores.

Near- or below-cost selling is a practice used by some retailers to catch the

attention of consumers. Some retail companies in our interviews indicated that it is
used as an anti-competitive practice by some large retailers. In order to assess the
extent of near- or below-cost selling in the retail sector, we asked retailers about

30
In this table (and the following tables) the data on the proportions of firms are given. For example, in Table
13, the number in first row, first column indicates that 33 percent of small retailers responded to the question

26
the frequency of below-cost selling at their stores and in the sector at large. The
same question was asked to FMCG suppliers as well.

Below-cost selling seems to be quite common: more than half of retailers have
applied near- or below-cost selling at least once a month (Table 14). However,
contrary to our prior expectations, it is applied more frequently by small rather than
large retailers. Almost all retailers (about 90 percent) believe that below-cost
selling is applied by other retailers at least once a month. These findings indicate
that this practice is quite common but it is not used, by all retailers, to eliminate
their competitors. Interestingly, the proportion of suppliers who believe that
retailers’ apply below-cost selling is lower than the proportion of retailers who think
so.

Most of the firms (about 90 percent of retailers and suppliers) believe that below-
cost selling causes unfair competition, but a large majority of them (74 percent of
retailers and 84 percent of suppliers) suggest that it cannot be used systematically
(Table 15). A small group of firms (around 30 percent) claims that below-cost
pricing is used to push competitors out of the market and/or makes the market
more competitive. Below-cost selling is used as a marketing method according to
half of retailers. However, this practice seems to be harmful for suppliers (64
percent of retailers, 83 percent of suppliers), possibly because of the fact that
suppliers in some cases are required to accept lower prices.

Supplier-retailer relation is a contentious issue in studies on the dynamics of

competition in the retail sector. There could be two types of distortions in
supplier-retailer relations. First, if suppliers have market power, they can try to
blockade entry by other suppliers through various restrictions (for example,
exclusivity agreements) or providing discounts not related to production and
marketing costs (shelf space, product range, etc.). Second, if retailers have market
power, they can impose certain fees on suppliers that are not related their costs
(listing fees, slotting fees, etc.).

apply “price flexing” (i.e., differentiate prices across stores). “n” in the table indicates the number of firms
who responded to that particular question (item response rate).

27
In order to analyze supplier-retailer relations, we first get information about
procurement and distribution channels suppliers and retailers use. An average
retailer in our sample buys products from 500 suppliers (Table 16). There is a
positive relation between the size of the retailer and the number of suppliers:
although a small retailer buys from 180 suppliers, a medium-sized retailer is
served by 400 suppliers. The number of suppliers a large retailer deals with
reaches 1250 (median values for size categories). These data are consistent with
the number of products sold by retail size. Moreover, small retailers prefer to buy
from suppliers that sell a wide range of products. Small retailers buy about 50
products per supplier, medium-sized retailers 30 products per supplier, and large
retailers 20 product per supplier.

Most of large retailers (60 percent) buy products directly from suppliers thanks to
large quantities they purchase. However, small and medium-size retailers rely
mainly on distributors (about 55 percent). A small group of retailers (only 20
percent) buy their products through wholesalers. The share of wholesalers seems
to be small.31

On the supplier side, most of suppliers use distributors to supply their products.
Large suppliers tend to use more suppliers than small and medium-sized suppliers
do. The share of suppliers who sell their products through wholesalers is small (21
percent). The rest of suppliers (26 percent) sell their products directly to retailers.

There is a dual structure in supplier-retailer relationship. First, there is a

relationship between large retailer and large/suppliers. Second is the relationship
between small retailers and distributors. Large retailers can act as a counterveiling
power to large suppliers that may have market power in specific FMCGs industries
(Dobson et al., 2001). Moreover, they set supply prices at the national level, and
reduce regional/locational differences in prices, as observed in price flexing
(Figures 3a-3d). Small retailers may subject to restrictive vertical agreements by
distributors and are likely to be open to the effects of regional factors.

31
Recall that our sample of retailers is much larger than the sector average.

28
Table 17 summarizes the data on supplier-retailer relations. No significant
difference is found between practices of small and large retailers.32 Retailers claim
that suppliers share frequently to the costs of promotions, i.e., they pay for
promotions (insert fee, etc.). Slotting/position fees (gondola, “palet”, etc., fees that
depend on shelf position) are paid less frequently. Listing fee (fee paid for listing
the first time) is applied “sometimes”, and shelf fee (fee paid for the duration of
listing) is not common. However, all suppliers, irrespective of their size, claim that
promotion, slotting and listing fees are charged frequently by retailers. It seems
that charging these fees is commonly accepted as a conduct of business. These
practices could make entry by new suppliers more difficult.

In terms of supply conditions, suppliers are almost always are required to take
back returned items. Moreover, large retailers frequently ask suppliers to do
packing (small retailers usually do not have the power to impose such a condition),
and small suppliers face with such demands by retailers more frequently than
large suppliers do.

There are two potentially problematic practices that can be applied by retailers and
suppliers: exclusivity restrictions on suppliers (requiring suppliers not to sell their
products to other retailers), and prices fixing by suppliers. Retailers and suppliers
agree that exclusivity restrictions on suppliers are imposed rarely and price fixing
occurs “sometimes”.

Regarding discounts provided by suppliers, the most common types are quantity-
based (volume) discounts and advance payment discounts. These types of
discounts are closely related to the cost of supply. As noted earlier, suppliers
provide discounts to support promotions launched by retailers. Potentially anti-
competitive practices, discounts based on product rage, shelf area and exclusivity
restrictions (retailer required not to sell competitive suppliers’ products) are
observed less frequently. These restrictions, if imposed by dominant suppliers,
may create entry barriers for new suppliers.

32
There is only one exception. Packing of products by suppliers is observed more often among large retailers
than small ones.

29
In order to check the extent of price discrimination by suppliers (through, for
example, abovementioned discounts), we asked retailers if they pay the same
price as other retailers do. A majority of retailers (68 percent) claimed that there is
no difference in prices, 23 percent pay lower prices, and only 9 percent pay higher
prices. When asked about what they could do if they discover that a supplier sold
the same product to other retailers at a lower price, half of retailers stated that they
can charge retrospective discounts (to match the price differential) to the supplier.

There are 10 retailers (4 medium-sized and 6 large) in our sample that are a
member of a business group that also owns supplier firms. When asked about the
relations with suppliers in the same group, 6 retailers (4 of them are medium-
sized) said that they provide preferential access to shelf space for their sister
suppliers, and 5 (4 medium-sized) of them get lower prices and/or better payment
conditions. Among the sample of suppliers, 4 (2 medium-sized) have sister
retailers. 3 (2 medium-sized) suppliers stated that they get preferential access to
shelf space in sister retailers, and they offer lower prices and/or better payment
conditions. Although the number of vertically related retailers/suppliers is small,
these findings suggest that retailers (and suppliers) tend to favor their sister
companies. This practice could be a concern for competition policy if any one of
the vertically-related companies has a dominant position in the market. However,
in our sample, it seems that medium-sized companies, not the large ones, have a
stronger tendency to establish preferential relations with their sister companies.33
In other words, the relations between vertically-related suppliers and retailers is
not, at least for time being, likely to distort competitive conditions in the retail
sector.

Private label products have an increasing market share and changed the
competitive conditions in the market (for competitive effects of private label
products, see Cotterill, Putsis Jr. and Dhar, 2000; Morton and Zettelmeyer, 2004;
Sayman and Raju, 2004; Steiner, 2004). The share of private label products in
total sales is quite high (the arithmetic average across all product categories is

33
A sales manager of a large retailer that belongs to a business group claimed in the interview that they do
not have special arrangements with their sister supplier companies because they operate independently as
profit centers within the group.

30
around 30 percent). Cleaning products, packed food products, milk products and
paper products have higher private label shares. Large retailers have somewhat
higher private label sales in these categories. Moreover, retailers, especially large
ones, expect that the share of private label products will continue to increase in the
next three years in all product categories.

It seems that a large number of suppliers are involved in private label production.
Almost all firms producing cleaning products produce private label products for
retailers, whereas the proportion of suppliers that produce private label products is
about 40-50 percent in food and beverages categories. The share of private label
products in total turnover is around 16 percent: it is quite high in the case of
cleaning products (43 percent), and low in food products (6 percent). Private label
products have a higher share in total turnover of small firms (33 percent). They
contribute to 10 percent of turnover of medium-sized retailers and only 3 percent
of turnover of large retailers. Large suppliers that market their products under
national, well-known brand names are involved in private label production in small
quantities.

Retailers and suppliers believe that private label products are of inferior quality
(Table 18). Partly because of this reason, production cost is thought to be lower. A
majority of retailers and suppliers agree that these products are cheaper than
national brands. There are more suppliers who believe that private label products
are relatively cheaper than those who believe that their production cost is lower.
Thus, there seems to be a reputation premium on national brands.

31
32
5. The Future

The organized retail market (local supermarkets and chain stores) have grown
rapidly at the expense of traditional retail formats (grocery shops and open
bazaar). Of course, organized retailers face with certain problems in developing
their businesses. They claim that (high) tax rates restrict their growth (Table 19).
This is the most important obstacle cited by small and medium-sized retailers. In
addition to taxes, land/store availability and competition from informal firms are
also important barriers for further growth of the organized retail sector. It seems
that urban areas develop without a proper implementation of well-designed city
planning in Turkey. This has the most adverse effect on retailing sector. Since it is
rather difficult to find an estate suitable for a large-scale store in city centers, the
rents and prices for suitable places/areas in city centers may reach prohibitive
levels. In order to overcome this problem, the large-scale retailers prefer to
acquire supermarkets located in central areas. The lack of suitable locations acts
as a significant entry barriers for supermarkets and large chains, and prevents
their rapid diffusion.

Small and medium-sized retailers consider competition from large chain stores as
a noteworthy obstacle. Regulations, macroeconomic uncertainty, costs of
financing financial and transportation facilities are only partially important.
Interestingly, consumer demand is among the least important factor that inhibits
retailers’ growth. This is a striking finding because the SIS survey on capacity
utilization finds consistently that the most important reason for underutilization of
production capacity in manufacturing industries in Turkey is the lack of (domestic)
demand.

Retail characteristics that determine competitiveness provide the clue to

understand which retail formats are likely to grow in the future. All retailers, large
and small, said that the most important determinant of competitiveness in the retail
sector is stores’ location (Table 20). It is followed by quality-related aspects
(product quality, product range/ diversity, and retailer's brand/reputation), and
prices (promotions, proximity to consumers, and prices). Other services offered by

33
the retail (parking, packing, store’s appearance and loyalty cards) are slightly less
important than quality and prices. Product brand is also among the partially
important factors. Given those factors, retailers (and suppliers as well) believe that
hypermarkets, supermarkets and discounters will increase their market shares in
the next decade. Some suppliers (20 percent of all suppliers) predict that cash &
carry and gas station markets will also increase their markets shares. There is
almost no firm that believes that grocery shops, open bazaars and specialized
markets (butchers, green groceries, etc.) will be able to increase their market
shares. Consistent with these predictions, all large retailers who responded to the
survey plan to open new stores in the next three years. There are three out of 17
medium-sized and 4 out of 14 small retailers that do not envisage any increase in
the number of stores. No retailer predicts any contraction in the number of stores it
currently operates. One small retailer stated that it plans to exit from the market in
the next three years.

Almost all retailers and suppliers (about 90 percent of firms) expect that foreign
retailers’ market share will increase in the next decade (Table 21). The proportion
of small retailers that expect an increase in foreign presence is a little lower (75
percent). Those firms that predict foreign entry in the retail market believe that, as
a result of foreign entry, retail prices will decline somewhat, and product quality
and diversity will increase to a large extent. Retailers, especially small ones, are
skeptical on the impact of foreign entry on domestic suppliers’ production, but
suppliers, especially small ones, are hopeful that domestic suppliers’ production
may increase as a result of foreign entry.

As noted in the second section of this study, there is no specific law regulating the
retail market in Turkey. A draft law prepared last year initiated an intense debate
on a number of issues. It is obvious that almost all retailers and suppliers are in
favor of having a law regulating the retail market (Table 22). Majority of retailers
(67 percent) and almost all suppliers (90 percent) support the idea that the law
should impose restrictions on below-cost sales. The stance of suppliers on this
issue is consistent with their opinion that below-cost sales are harmful for
suppliers. Suppliers are also strongly in favor of restrictions on payment conditions
and exclusivity agreements whereas small and medium-sized retailers are

34
indifferent and large retailers are weakly against these restrictions. While retailers,
especially large ones, are against restrictions on promotions, suppliers are
somewhat in favor of these restrictions, too. Overall, suppliers seem to be worried
that retailers could pass on the costs of fierce competition in the market on their
shoulders.

The issue of imposing restrictions on private label sales by retailers (such as 20

percent ceiling) is a contested area where suppliers and retailers, and small and
large firms disagree each other. Large retailers who can capitalize on the
reputation they establish in the market by selling more private label products are
against restrictions on private label sales, whereas medium-sized and large
suppliers, who consider private label as a threat to their national brands, are in
favor of these restrictions. Small and medium-sized retailers, who may not benefit
much from private label products, are somewhat in favor of restrictions, and small
suppliers, whose position may not differ under private label production, are
indifferent. Private label products seem to be a tool that may shift the benefits of
brand name advantages in favor of large retailers.

35
36
6. Issues for Competition Policy
The conduct of firms in the retail and supplier sectors has received considerable
attention due to its direct impact on consumers. There have been 23 complaints
on retailers made to the Competition Authority in the period 1998-2003. Most of
these complaints are about below-cost selling and discriminatory practices.

The common feature of below-cost selling complaints is the claim that

hypermarkets sell their products at excessively low prices that may force small
retailers to exit from the market. According to the Competition Law, below-cost
selling or excessively low prices can be deemed as the violation of the law only if
the undertaking concerned has a dominant position in the relevant market. The
law defines dominant position as “any position enjoyed in a certain market by one
or more enterprises by virtue of which, those enterprises have the power to act
independently of their competitors and purchasers in determining economic
parameters such as the amount of production or distribution, price and supply”.
The Competition Board rejected all complaints about below-cost selling as out of
scope by arguing that the dominant position of any hypermarket in the relevant
market is unlikely because of low concentration ratios in the market, low entry
barriers and dynamic market conditions.

Other main complaints brought before the Competition Board are concerned with
discriminatory practices done by suppliers against small retailers in favor of large
retailers. Complaints were generally brought by small retailers or the Chamber of
Small Grocery Shops (Bakkallar Federasyonu Odası ). They claimed that suppliers
sell their products under more favorable conditions to large retailers. The
Competition Board deemed almost all these complaints as out of scope because it
decided that small retailers and large retailers are not in equivalent position
because of differences in their sizes, volumes of purchased products, product
diversity, etc.

There are two cases of infringement of the Competition Law. The Competition
Board decided in the case brought by the Istanbul Food Wholesale Traders

37
Association (IGTOD) that a number of large food suppliers (Benckiser, Sezginler,
Ülker, Besler, Eczacıbaşı Procter & Gamble, Marsa Kraft Jacobs Suchard,
Unilever, and LeverElida) violated the law by imposing sales restrictions on their
distributors. Likewise, another food supplier (Frito-Lay) was found to be abusing its
market power by imposing exclusivity restrictions on retailers. 34

There have been five merger cases in retail trade brought before the Competition
Board since 1998. In the first case, Metro and Migros joint venture (1998), the
Board granted a conditional permission, but the venture was not established later
on. In all other cases, Doğuş Holding-Tansaş (1999), Carrefour-Continent (2000),
Tesco-Kipa (2003), and Carrefour-Gima/Endi (2005) the Board permitted mergers
unconditionally.

As noted earlier, there is no special legislation regarding the establishment of large

retailers in Turkey. There have been several attempts to introduce a law for this
purpose, and three draft laws were brought before the Competition Board in recent
years. Although there were some differences between these laws, the common
aim was to help small retailers (groceries, green groceries, etc.) by forcing large
stores to be located “outside” the city.

The first draft law prepared in 2001 aimed at regulating the establishment of stores
having a sales area greater than 250 m2 subject to the permission obtained from a
Board composed of the Municipality, Chamber of Commerce, Competition Board
and consumer associations. The Board would give its decision by considering the
location (its distance to the city centre), demand and supply conditions in the city
concerned, and the competitiveness of small retailers. The same procedure would
apply to the stores that are larger than 1000 m2 that would be located 5 km away
from the city centre.

34
Frito-Lay was the dominant firm in the salty-snack market in Turkey (two largest firms controlled about 98
percent of the market). It was found that, Frito-Lay tried to establish an exclusive sales network at the final
sales points, especially in traditional retailers, in the period 1998-2003. The sales agreements explicitly stated
that the retailer could not sell any other brands. Frito-Lay applied also special promotions (awards, gifts,
discounts, bonuses, etc.) aimed at achieving exclusivity. Since Frito-Lay’s sales personnel visited retailers
once or twice a week, it was able to enforce the exclusivity agreements. After the investigation, the
Competition Board decided to withdraw the exemption for the non-competition (exclusivity) clause in
vertical agreements done between Frito-Lay and the retailers (final sale points), because it did not bring out
any economic benefit or efficiency gains, and acted as an barrier to entry.

38
The second draft law was prepared by Ministry of Trade and Industry in 2003. The
difference between the first and second laws was the fact the latter one did not
envisage any special Board. It assigned the authority to the governor or the
Ministry of Industry and Trade according to size of large stores. It also included
provisions that prohibited certain forms of conduct (predatory pricing tactics, etc.)
that could be addressed indirectly under the Competition Act.

The last draft law was put on the agenda in 2004. Those above-mentioned
prohibitions were excluded from draft law after the Competition Board’s objections.
Although there are some improvements in the new draft law, the Competition
Board opposed to two issues concerning restriction of private label sales by large
stores (the draft law envisaged 20 percent limit for private label sales) and
limitations on low-price sales promotions. The Competition Board states that these
restrictions harm consumers (by preventing price competition) and small and
medium-sized manufactures (who can gain competitive advantage by producing
private label products for large retailers). It seems that the law is not agenda of the
government, and is not likely to be enacted in recent future.

Although most of the retailers and suppliers who participated in our survey stated
that they welcome a law on regulating the retail market, restrictions on different
forms of competitive practices and on the location of large stores need to be
tackled with care. Since the competition law provides sufficient safeguards against
any anti-competitive behavior, there may not be any need to introduce additional
general restrictions. The idea of protecting small retailers by imposing a ban on the
establishment of new large stores around the city center is also questionable
because it basically helps the incumbent large retailers. The issue of land
provision for large stores and shopping centers can be better dealt within the
context of urban planning.

39
40
7. Conclusions

Major findings of our analysis can be summarized as follows:

™ The retail market in Turkey is competitive. There are no legal restrictions on
entry, and no discrimination against foreign companies.
™ Prices across retail formats differ substantially for a market operating on a very
thin profit margin. However, these differences are likely to stem from cost
differences.
™ There are some practices applied by retail companies that are potentially anti-
competitive (price flexing, listing fees, slotting fees, etc.). However, these
practices do not distort competition in the retail market seriously, because
these companies seem to lack a significant degree of market power. There are
some practices applied by supplier companies that are potentially anti-
competitive (discounts based on exclusivity agreements, shelf area, product
range, etc.). These practices could distort competition because of high level of
concentration in certain markets. These markets need to be scrutinized closely
by the Competition Authority to guarantee further development of the retail
market.
™ Supermarkets, chains stores and foreign firms are likely to increase their
market shares in the future. Any single retailer may not seem to establish a
dominant position in the national market. However, the relevant markets in the
retail sector should be defined locally rather than nationally. It is possible that
some retailers may establish a dominant position in certain local markets,
especially following a merger activity and/or exits.35

35
After the first draft of this study was written, Carrefour announced on May 3, 2005, that they reached an
agreement with Fiba Holding to acquire Gima and Endi for USD 132 million and it would become Turkey’s
biggest food retailer after the acquisition. (The merger was later approved by the Competition Board on June
17.) Koç Holding reacted to this announcement by stating that they had reached an agreement with Fiba
Holding on all matters including the price in their acquisition deal for Gima and Endi, but Fiba Holding
moved ahead with Carrefour instead without any prior notice. As a result, Koc Holding broke away with
Sabanci Holding in their planned joint bid for the privatization of Turk Telecom. A few months later, it was
announced on August 22 that Migros of Koç Holding agreed to buy 70.77 percent stake in organised retailer
Tansas for USD 387.million from Doğuş Group. Koç Holding declared that they would preserve the brands
Migros and Tansaş. After the acquisition, Migros will again be the largest FMCG retailer in Turkey.

41
™ The share of private label products is likely to increase in the future. The
increase in the share of private label is likely to lead to a more competitive
retail market.
™ These trends may have a slightly positive impact on retail prices, especially if
discount stores continue to increase their market shares.
™ These trends are likely to have a positive impact on product diversity and the
quality of products/services offered by retail stores.
™ Employment impact of these trends could be negative because
turnover/employee ratio is three times higher in chain stores than in traditional
retailers. However, modern retail formats can generate new jobs if they provide
additional services for their customers.
™ The transformation of the retail market is likely to have a long-lasting impact on
wholesale trade and the distribution of FMCGs as well. Traditional wholesalers
are the most likely losers, because large retailers tend to buy directly from
suppliers. Logistics companies that provide a wide range of complementary
services will play an increasingly more important role in the distribution of
FMCGs.

42
References
AC Nielsen (2004), Shopping Trends 2004, Unpublished report, Istanbul: AC Nielsen.
Boylaud, Olivier (2000), Regulatory Reform in Road Freight and Retail Distribution,
Economics Department Working Papers No. 255, Paris: OECD.
Boylaud, Olivier and Giuseppe Nicoletti (2001), “Regulatory Reform in Retail
Distribution”, OECD Economic Studies (32): 254-273.
Competition Commission (2000), Supermarkets: A Report on the Supply of
Groceries from Multiple Stores in the United Kingdom, 3 volumes, presented to
Parliament by the Secretary of State for Trade and Industry (www.Competition-
Commission.org.uk).
Cotterill, Ronald W., William P. Putsis Jr. and Ravi Dhar (2000), “Assessing the
Competitive Interaction between Private Labels and National Brands”, Journal
of Business (73): 109-137.
Dobson Consulting (1999), Buyer Power and its Impact on Competition in the
Food Retail Distribution Sector of the European Union, Report prepared for the
European Commission DG IV.
Dobson, Paul W., Roger Clarke, Stephen Davies and Michael Waterson (2001),
“Buyer Power and its Impact on Competition in the Food Retail Distribution
Sector of the European Union”, Journal of Industry, Competition and Industry
(1): 247-281.
McKinsey Global Institute (2003), Türkiye: Verimlilik ve Büyüme Atılımının
Gerçekleştirilmesi, Istanbul: McKinsey Company
(www.mckinsey.com/knowledge/cgi).
Mazzarotto, Nicola (2001), Competition Policy towards Retailers: Size, Seller
Market Power and Buyer Power, Centre for Competition and Regulation (CCR)
Working Paper No: 01-4, University of East Anglia.
Morton, Fiona Scott and Florian Zettelmeyer (2004), “The Strategic Positioning of
Store Brands in Retailer-Manufacturer Negotiations”, Review of Industrial
Organization (24): 161-194.
Özcan, Gül Berna (2000), “The Transformation of Turkish Retailing: Survival
Strategies of Small and Medium-sized Retailers”, Journal of Southern Europe
and the Balkans (2): 105-120.

43
Özcan, Gül Berna (2001), “Pattern of Vertical and Horizontal Integration in Turkish
Retailing”, European Regional Digest (32): 35-41.
Sayman, Serdar and Jagmohan S. Raju (2004), “Investigating the Cross-Category
Effects of Store Brands”, Review of Industrial Organization (24): 129-141.
Steiner, Robert L. (2004), “The Nature and Benefits of National Bran/Private Label
Competition”, Review of Industrial Organization (24): 105-127.
Togan, S. and Taymaz, E. (2005), Turkey: Trade Issues, unpublished paper.
Tokatli, Nebahat (1999), “A Comparative Report on the Profiles of Retailing in the
Emerging Markets of Europe: Turkey, Poland, Hungary, Portugal, and Greece”,
Journal of Euromarketing (8): 75-105.
Tokatlı, Nebahat and Yonca Boyacı (1997), “Internalization of Retailing in Turkey”,
New Perspective on Turkey (17): 97-128.
Tokatlı, Nebahat and Gül Berna Özcan (1998), “The State and the Corporate
Private Sector in the Recent Restructuring of Turkish Retailing”, New
Perspectives on Turkey (18): 79-111.
Tokatli, Nebahat and Yonca Boyaci Eldener (2002), “Globalization and the
Changing Political Economy of Distribution Channels in Turkey”, Environment
and Planning A (34): 217-238.

44
Tables and Figures
Table 1. FMCG wholesale and retail trade sectors, 1997, 1999 and2001
(million USD)

Number of Number of Number of Payments Input Output Value Total

estab. empl. engaged to empl. added sales
1997
51 - Wholesale trade (except of motor vehicles and motorcycles)
5121 - Agricultural raw materials and live animals 2 408 12 832 14 603 42 77 967 889 1 949
5122 - Food, beverages and tobacco 15 705 69 559 82 901 177 299 2 463 2 164 14 393
52 - Retail trade, except of motor vehicles and motorcycles; repair of personal and household goods
5211 - Retail sale in non-specialized stores 172 741 96 460 319 583 270 416 2 997 2 581 11 627
5219 - Other retail sale in non-specialized stores (departm 8 099 29 241 39 574 109 166 852 687 4 461
5220 - Retail sale of food, beverages and tobacco in spec 58 562 31 365 109 014 53 167 1 182 1 015 4 759
5231 - Retail sale of pharmaceutical and medical goods, c 20 717 23 940 46 906 45 113 712 600 2 503
5259 - Other non-store retail sale 56 546 563 1 3 9 6 14

1999
51 - Wholesale trade (except of motor vehicles and motorcycles)
5121 - Agricultural raw materials and live animals 2 474 12 824 14 573 64 109 1 256 1 147 3 398
5122 - Food, beverages and tobacco 16 664 77 359 90 754 246 569 4 456 3 888 19 676
52 - Retail trade, except of motor vehicles and motorcycles; repair of personal and household goods
5211 - Retail sale in non-specialized stores 180 934 110 803 338 897 399 520 3 950 3 431 15 247
5219 - Other retail sale in non-specialized stores (departm 8 629 43 197 52 212 205 273 1 353 1 080 5 477
5220 - Retail sale of food, beverages and tobacco in spec 61 937 36 013 119 373 72 245 1 631 1 385 6 124
5231 - Retail sale of pharmaceutical and medical goods, c 21 557 26 426 51 183 61 120 869 749 2 885
5259 - Other non-store retail sale 67 779 798 2 3 13 10 23

2001
51 - Wholesale trade (except of motor vehicles and motorcycles)
5121 - Agricultural raw materials and live animals 2 503 13 567 15 465 56 118 1 083 965 2 181
5122 - Food, beverages and tobacco 17 875 82 265 97 407 264 513 4 720 4 207 17 189
52 - Retail trade, except of motor vehicles and motorcycles; repair of personal and household goods
5211 - Retail sale in non-specialized stores 185 504 117 956 346 193 256 373 2 371 1 998 10 406
5219 - Other retail sale in non-specialized stores (departm 9 430 44 582 54 599 146 214 894 680 3 879
5220 - Retail sale of food, beverages and tobacco in spec 62 693 37 049 121 538 63 183 1 361 1 178 4 296
5231 - Retail sale of pharmaceutical and medical goods, c 24 201 30 753 57 050 66 95 941 846 3 213
5259 - Other non-store retail sale 83 581 617 1 2 8 6 69
Source: SIS.

45
Table 2. Value added, FMCG-related sectors, 1997-2001
(million USD)

1997 1998 1999 2000 2001

A - Agriculture, hunting and forestry 25815 33758 26665 27206 17890

D - Manufacturing
15 - Food products and beverages
151 - Meat, fish, fruit, vegetables, oils and fats
1511 - Meat and meat products 373 543 381 476 259
1512 - Fish and fish products 53 27 20 20 17
1513 - Fruit and vegetables 680 687 744 610 569
1514 - Vegetable and animal oils and fat 650 653 564 492 419
152 - Dairy products
1520 - Dairy products 235 305 384 412 224
153 - Grain mill products, starches, animal feeds
1531 - Grain mill products 191 225 196 209 136
1532 - Starches and starch products 31 42 41 33 26
1533 - Animal feeds 171 212 270 224 81
154 - Other food products
1541 - Bakery products 323 336 286 304 197
1542 - Sugar 203 414 401 410 396
1543 - Cocoa, chocolate and sugar conf 343 341 451 312 359
1544 - Macaroni, noodles, couscous 71 83 54 59 44
1549 - Other food products n.e.c. 336 479 373 461 307
155 - Beverages
1551 - Spirits; ethyl alcohol 177 438 523 444 453
1552 - Wines 17 20 15 14 16
1553 - Malt liquors and malt 231 225 215 208 191
1554 - Soft drinks, mineral waters 270 269 333 209 258
16 - Tobacco products
160 - Tobacco products
1600 - Tobacco products 552 817 1000 1063 2015
21 - Paper and paper products
210 - Paper and paper products
2101 - Pulp, paper and paperboard 181 159 161 260 162
2102 - Corrugated paper, containers 228 269 250 244 214
2109 - Other articles of paper and paper 242 474 170 193 144
24 - Chemicals and chemical products
242 - Other chemical products
2424 - Soap and detergents, cleaning pr 851 468 920 655 518
Total 6412 7487 7752 7312 7004

G - Wholesale and retail trade; repair of motor

51 - Wholesale trade (except of motor vehicles an
512 - Agricultural raw materials, live animals,
5121 - Agricultural raw materials and live 889 977 1147 1632 965
5122 - Food, beverages and tobacco 2164 2843 3888 3930 4207
52 - Retail trade, except of motor vehicles and mo
521 - Non-specialized retail trade in stores
5211 - Retail sale in non-specialized sto 2581 2879 3431 3668 1998
5219 - Other retail sale in non-specialize 687 1193 1080 1069 680
522 - Retail sale of food, beverages and tobac
5220 - Retail sale of food, beverages an 1015 1214 1385 1873 1178
523 - Other retail trade of new goods in specia
5231 - Retail sale of pharmaceutical and 600 706 749 811 846
525 - Retail trade not in stores
5259 - Other non-store retail sale 6 10 10 8 6
Total 7943 9822 11690 12993 9879

H - Hotels and restaurants

55 - Hotels and restaurants
551 - Hotels; camping sites and other provisio
5510 - Hotels; camping sites and other p 1797 1954 2449 2997 2489
552 - Restaurants, bars and canteens
5520 - Restaurants, bars and canteens ( 2385 2376 3491 4449 3250
Total 4182 4330 5940 7446 5739
Total (all sectors except agriculture) 18537 21639 25382 27750 22622
Note: The data on manufacturing industries exclude private establishments employing less than 10 people.
Source: SIS
46
Table 3. Concentration rates in domestic production, FMCG-related sectors, 1997-2001

1997 1998 1999 2000 2001

n CR4 n CR4 n CR4 n CR4 n CR4
D - Manufacturing
15 - Food products and beverages
151 - Meat, fish, fruit, vegetables, oils and fats
1511 - Meat and meat products 94 33.4 110 29.6 99 32.4 101 31.2 99 34.7
1512 - Fish and fish products 16 90.5 19 82.8 17 77.5 14 79.7 16 68.1
1513 - Fruit and vegetables 239 17.6 251 21.4 240 11.8 226 16.3 234 20.0
1514 - Vegetable and animal oils and fat 124 48.4 120 42.5 108 43.2 103 44.0 95 35.1
152 - Dairy products
1520 - Dairy products 113 51.1 118 50.0 112 51.4 110 49.5 114 51.8
153 - Grain mill products, starches, animal fee
1531 - Grain mill products 305 16.3 310 16.6 279 24.0 272 22.8 264 18.1
1532 - Starches and starch products 10 93.5 8 96.7 7 96.4 7 96.6 6 95.8
1533 - Animal feeds 141 23.9 149 26.3 144 27.9 134 29.1 130 33.0
154 - Other food products
1541 - Bakery products 431 36.6 442 38.6 390 39.0 365 43.8 372 35.5
1542 - Sugar 35 39.1 34 33.2 34 35.9 35 32.0 39 35.9
1543 - Cocoa, chocolate and sugar conf 86 64.1 91 59.3 84 59.1 85 59.6 85 61.4
1544 - Macaroni, noodles, couscous 15 79.2 16 68.0 13 75.7 16 69.1 19 61.6
1549 - Other food products n.e.c. 101 39.4 106 41.4 106 43.4 110 38.4 113 38.3
155 - Beverages
1551 - Spirits; ethyl alcohol 15 58.5 13 65.8 14 71.4 14 73.7 13 71.3
1552 - Wines 12 72.2 14 71.5 12 77.5 12 72.1 13 73.5
1553 - Malt liquors and malt 9 74.6 10 79.5 9 69.0 8 76.5 8 77.2
1554 - Soft drinks, mineral waters 54 65.7 60 67.0 60 63.5 55 67.9 54 75.0
16 - Tobacco products
160 - Tobacco products
1600 - Tobacco products 38 54.8 39 58.9 35 57.6 28 61.4 25 66.7
21 - Paper and paper products
210 - Paper and paper products
2101 - Pulp, paper and paperboard 40 46.7 46 37.7 40 35.9 43 31.5 46 36.6
2102 - Corrugated paper, containers 101 24.9 117 26.2 105 27.2 107 26.4 116 26.1
2109 - Other articles of paper and paper 50 59.9 68 60.1 52 47.7 60 46.3 56 42.5
24 - Chemicals and chemical products
242 - Other chemical products
2424 - Soap and detergents, cleaning pr 63 62.1 67 64.9 64 71.2 63 63.3 71 66.8

52 - Retail trade, except of motor vehicles and motorcycles; repair of personal and household goods
521 - Non-specialized retail trade in stores 194127 10.8 194934 11.5
Source: SIS, Concentration Rates in Manufacturing; Sector 52, authors' estimates.

47
Table 4. Profit margins in FMCG-related sectors, 1997-2002
(operating profits/turnover)

Sectors (ISIC Rev. 3) 1997 1998 1999 2000 2001 2002

011 Growing of crops 0.11 0.13 0.12 0.11 0.13 0.11
012 Farming of animals 0.08 0.12 0.10 0.06 0.07 0.10
014 Agricultural and animal husbandry service activities 0.04 0.04 0.04 0.03 0.03 0.02
050 Fishing, fish farms 0.12 0.10 0.11 0.08 0.14 0.08
151 Meat products 0.01 0.10 0.06 -0.04 0.00 0.04
152 Fish products 0.12 0.16 0.17 0.01 0.06 0.08
153 Fruit and vegetables 0.09 0.07 0.08 0.03 0.09 0.06
154 Vegetable and animal oils and fats 0.09 0.09 0.07 0.05 0.09 0.05
155 Dairy products 0.10 0.06 0.05 0.03 0.04 0.02
156 Grain mill products 0.05 0.04 0.02 0.03 0.04 0.05
157 Animal feeds 0.05 0.05 0.05 0.04 0.02 0.06
158 Other food products 0.08 0.10 0.06 0.06 0.11 0.12
159 Beverages 0.11 0.01 0.03 0.06 0.06 0.10
160 Tobacco products 0.16 0.13 0.09 0.14 0.08 0.13
211 Pulp, paper and paperboard 0.04 0.04 -0.02 -0.05 0.01 0.04
212 Containers of paper and paperboard 0.13 0.09 0.09 0.09 0.11 0.13
245 Other chemical products 0.14 0.18 0.17 0.11 0.13 0.12
512 Wholesale of agricultural raw materials 0.07 0.07 0.06 0.03 0.09 0.03
513 Wholesale of food, beverages and tobacco 0.03 0.01 0.02 0.02 0.02 0.02
521 Non-specialized retail trade in stores 0.01 0.01 -0.02 -0.02 -0.05 -0.02
523 Retail trade of new goods in specialized stores 0.06 0.11 0.15 0.08 0.11 0.06
551 Hotels; camping sites 0.10 0.08 -0.02 0.04 0.10 0.08
553 Restaurants, bars and canteens -0.02 0.00 0.01 0.02 0.00 0.03
Source: The Central Bank of the Republic of Turkey
Table 5. Restrictiveness index scores for distribution services (wholesale and retail trade)

Developed EU-15 Latin Asia Turkey

countries America
Domestic Index
Restrictions on commercial land 0.000 0.000 0.000 0.013 0.000
Direct investment in distribution firms 0.000 0.001 0.000 0.001 0.000
Restrictions on large-scale stores 0.011 0.012 0.000 0.009 0.000
Factors affecting investment 0.005 0.007 0.002 0.005 0.000
Local government requirements 0.017 0.005 0.013 0.011 0.000
Restrictions on establishment total 0.033 0.025 0.015 0.039 0.000
Wholesale import licensing 0.005 0.001 0.011 0.019 0.000
Limits on promotion of retail products 0.007 0.008 0.004 0.003 0.000
Statutory government monopolies 0.004 0.000 0.005 0.005 0.013
Protection of intellectual property rights 0.014 0.050 0.018 0.028 0.050
Restrictions on ongoing operations total 0.031 0.060 0.038 0.055 0.063
Domestic index total 0.064 0.085 0.053 0.094 0.063

Foreign Index
Restrictions on commercial land 0.043 0.100 0.000 0.038 0.000
Direct investment in distribution firms 0.000 0.005 0.031 0.085 0.000
Restrictions on large-scale stores 0.011 0.012 0.000 0.009 0.000
Factors affecting investment 0.010 0.012 0.004 0.018 0.000
Local government requirements 0.017 0.005 0.017 0.015 0.000
Permanent movement of people 0.013 0.020 0.027 0.011 0.031
Restrictions on establishment total 0.094 0.153 0.080 0.176 0.031
Wholesale import licensing 0.005 0.001 0.011 0.023 0.000
Limits on promotion of retail products 0.007 0.008 0.004 0.003 0.000
Statutory government monopolies 0.004 0.000 0.005 0.005 0.013
Protection of intellectual property rights 0.014 0.050 0.018 0.028 0.050
Licensing requirements on management 0.033 0.019 0.030 0.038 0.014
Temporary movement of people 0.007 0.007 0.007 0.008 0.019
Restrictions on ongoing operations total 0.071 0.086 0.074 0.105 0.096
Foreign index total 0.165 0.239 0.154 0.281 0.126
Note: Developed countries: Australia, Canada, Japan, New Zealand, South Africa, Switzerland and United States; EU-15: Austria,
Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden and United
Kingdom; Latin America: Argentina, Brazil, Chile, Colombia, Mexico, Uruguay and Venezuela; Asia: India, Hong Kong, Indonesia, Korea,
Malaysia, Philippines, Singapore and Thailand
Source: Productivity Commission Trade Restrictiveness Database, Australia

48
Table 6. Vertical integration in FMCGs sector in Turkey

Business group S e c t o r s
Retail Manufacturing Wholesale/Distribution Hotels and Restaurants
Store Brands Retail Formats Company Name Sector Company Name Company Name
Sabancı Holding Carrefoursa Hypermarket Marsa Kraft Oils and fats Philip Morrissa Tursa A.Ş.
Championsa Supermarket Gıdasa Macaroni, flour Ankara Enternasyonel.
Diasa Discount Store juice, biscuit, cake Otelcilik A:.Ş.
Philsa Tobacco
Koç Holding Migros Hypermarket Tat Meat products, dairy Düzey Pazarlama A.Ş. Divan A.Ş.
Supermarket products, macoroni Talya A.Ş.
Şok Discount Store canned food
bottled water
Doğuş Holding Tansaş Supermarket Antur Turizm
Macrocenter Supermarket Garanti Turizm
Voyager Turizm
Datmar Turizm
Göktrans Turizm
Fiba Holding Gima Supermarket
Endi Discount Store
Azizler Holding BİM Discount Store
Kombassan Afra Hypermarket Komas Gıda Macaroni, flour, Baykur A.Ş. Bera Turizm
Supermarket dry food Hotel Bera
Kardelen A.Ş. Bottled water
Yimpaş Holding Yimpaş Hypermarket Aytaç Meat products, dairy Yimpaş Otelcilik ve
Proma Supermarket products, oils and fats Restoranları
bottled water, juice
Canerler Group Canerler Hypermarket Anmar Mineral waters
Supermarket Beka Dry food
Keybi Supermarket Söğüt Meat products
İttifak Holding Adese Hypermarket Selva Gıda Macaroni
Source: Compiled by authors' from company web sites.

Table 7. Establishment year of survey firms

Retailers Suppliers
Total Foreign Holding Total Foreign Holding
1983 and before 6 3 38 3 11
1984-1988 5 1 12 3 5
1989-1993 19 3 4 11 2 1
1994-1998 13 2 2 7 2
1999-2003 7 1 5 1
Unknown 1 1 6 1 1
Total 51 5 12 79 9 22

49
 Figure 1b. C1 group purchasing patterns (4-q MA) 1999:4-2003:3

20 60

18
50
16

14
40
12

10 30

8
20
6

4
10
2

0 0
4

3
9q

0q

0q

0q

0q

1q

1q

1q

1q

2q

2q

2q

2q

3q

3q

3q

3q

4q

4q

4q
9

0
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

Open bazaar Cash and carry Other DFV and kiosk Bim
Carrefour-SA Dia Gima Migros Sok
Tansas Local supermarkets Grocery shops

50
 Figure 1c. C2 group purchasing patterns (4-q MA) 1999:4-2003:3

20 60

18

50
16

14
40

12

10 30

20
6

4
10

0 0
q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3
99

00

00

00

00

01

01

01

01

02

02

02

02

03

03

03

03

04

04

04
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
Open bazaar Cash and carry Other DFV and kiosk Bim Carrefour-SA
Gima Migros Sok Tansas Local supermarkets Grocery shops

Figure 1d. D group purchasing patterns (4-q MA) 1999:4-2003:3

20 60

18

50
16

14
40

12

10 30

20
6

4
10

0 0
4

3
9q

0q

0q

0q

0q

1q

1q

1q

1q

2q

2q

2q

2q

3q

3q

3q

3q

4q

4q

4q
9

0
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

Open bazaar Cash and carry Other DFV and kiosk Bim Migros Tansas Local supermarkets Grocery shops

51
Table 8. Survey coverage

Sales Number of Firm size

(million USD) employees (million USD)
SIS, 2001
Total retail (521 and 522) 18581 721917 0.093
521 - Non-specialized retail trade in stores 14285 563330 0.088
522 - Retail sale in specialized stores 4296 158587 0.116
a
Total suppliers 18602 194166 9.357

HTP, 2003
Total 14438
Chains/discounters/hyper and super 4617

b
Survey sample
c
Retail, 44 firms , 2001 2608 35063 59.3
c
Retail, 47 firms , 2003 4368 40025 92.9
Supplier, 69 firms, 2001 3418 53234 49.5

Coverage ratios
SIS total retail, 2001 0.140 0.049
SIS non-specialized retail trade, 2001 0.183 0.062
SIS suppliers, 2001 0.184 0.274
HTP total, 2003 0.303
HTP >supermarkets, 2003 0.946
a Supplier industries (ISIC Rev 3): 1511, 1512, 1513, 1514, 1520, 1531, 1532, 1533, 1541, 1542, 1543, 1544, 1549,
1551, 1552, 1553, 1554, 1600, 2101, 2102, 2109 and 2424.
b There are 51 and 79 firms who responded to the retail and supplier surveys, respectively.
c Turnover for 8 firms estimated as 0.78 and 0.11 USD turnover/employee in 2001 and 2003, respectively.

Table 9. Size distribution of survey firms

Retailers Suppliers
Total Foreign Holding Total Foreign Holding
Small 16 1 25 2 1
Medium 19 5 19 1 5
Large 16 5 6 26 5 13
Unknown 9 1 3
Total 51 5 12 79 9 22
For retailers, small: <150 employees (<10 million USD turnover); medium: 150-499 employees
(10-50 million USD turnover); large: 500+ employees (50+ million USD turnover).
For suppliers, small: <20 million USD turnover; medium: 20-50 million USD turnover; large:
50+ million USD turnover.

52
Table 10. Number of stores by size, 2000-2003

2000 2001 2002 2003

Small retailers (total) 18 31 39 71
Hypermarket (>2500 m2) 0 0 0 0
Large supermarket (1000-2500 m2) 4 6 7 10
Supermarket (400-1000 m2) 7 10 14 21
Small süpermarket (100-400 m2) 6 14 17 39
Market (50-100 m2) 1 1 1 1
Total area (000 m2) 13.5 21.1 26.4 42.0
n 10 12 12 14
Medium-sized retailers (total) 122 142 162 206
Hypermarket (>2500 m2) 3 3 4 4
Large supermarket (1000-2500 m2) 5 8 13 19
Supermarket (400-1000 m2) 63 71 73 96
Small süpermarket (100-400 m2) 50 59 71 86
Market (50-100 m2) 1 1 1 1
Total area (000 m2) 75.9 89.0 105.7 136.0
n 14 14 14 15
Large retailers - excl. foreign (total) 774 857 895 915
Hypermarket (>2500 m2) 51 62 67 69
Large supermarket (1000-2500 m2) 102 112 117 125
Supermarket (400-1000 m2) 162 179 192 221
Small süpermarket (100-400 m2) 456 500 516 497
Market (50-100 m2) 3 4 3 3
Total area (000 m2) 584.6 663.6 702.9 739.4
n 10 11 11 11
Foreign retailers (total) 581 684 801 958
Hypermarket (>2500 m2) 15 28 31 32
Large supermarket (1000-2500 m2) 0 3 3 5
Supermarket (400-1000 m2) 0 0 0 0
Small süpermarket (<400 m2) 566 653 767 921
Small süpermarket (100-400 m2) 0 0 0 0
Total area (000 m2) 194.0 266.5 305.5 351.0
n 3 4 4 4
Note: Data from 44 retailers
Total area is estimated by assuming 3500 m2 area for hypermarkets. For all other stores,
the mid-values are used.

53
Table 11. Relavant market for retailers
(per cent of retailers)

Small Medium-sized Large Total

retailers retailers retailers
Consumer group (socio-economic status)
A category 0.56 0.53 0.75 0.63
B category 0.56 0.73 1.00 0.80
C category 0.56 0.67 0.88 0.73
D category 0.11 0.27 0.44 0.30
n 40
Retail format
Hypermarket 0.36 0.50 0.75 0.54
Supermarket 0.79 0.83 0.88 0.83
Discount market 0.14 0.22 0.31 0.23
Cash & carry 0.00 0.11 0.25 0.13
Grocery 0.14 0.11 0.19 0.15
Bazaar 0.29 0.17 0.19 0.21
Gas station 0.00 0.06 0.13 0.06
Specialized 0.00 0.11 0.06 0.06
n 48
Geographical market
International 0.00 0.00 0.31 0.10
National 0.07 0.17 0.56 0.27
Regional 0.47 0.17 0.31 0.31
Local 0.53 0.72 0.31 0.53
n 49
Note: Colum totals exceed 1 because of multiple response.

54
 Figure 2a. Relative food prices (4-quarter moving averages), 1994:4-2004:3

1.10

1.05

1.00

0.95

0.90
4

3
9q

0q

0q

0q

0q

1q

1q

1q

1q

2q

2q

2q

2q

3q

3q

3q

3q

4q

4q

4q
9

0
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
Open bazaar Cash and carry Local supermarkets DFV and kiosk Grocery shops
Bim Carrefour-SA Dia Gima Kipa
Migros Real Sok Tansas

Figure 2b. Relative prices of personal care products (4-quarter moving averages), 1994:4-2004:3

1.15

1.10

1.05

1.00

0.95

0.90

0.85
q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3
99

00

00

00

00

01

01

01

01

02

02

02

02

03

03

03

03

04

04

04
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

Open bazaar Cash and carry Local supermarkets Grocery shops Bim
Carrefour-SA Dia Gima Kipa Migros
Real Sok Tansas

55
 Figure 2c. Relative prices of cleaning products (4-quarter moving averages), 1994:4-2004:3

1.15

1.10

1.05

1.00

0.95

0.90

0.85
q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3
99

00

00

00

00

01

01

01

01

02

02

02

02

03

03

03

03

04

04

04
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
Open bazaar Cash and carry Local supermarkets Grocery shops Bim
Carrefour-SA Dia Gima Kipa Migros
Real Sok Tansas

Figure 3a. Price flexing in food products by retail type, 1999:4-2004:3

0.38

0.33

0.28

0.23

0.18

0.13

0.08
q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3
99

00

00

00

00

01

01

01

01

02

02

02

02

03

03

03

03

04

04

04
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

Open bazaar Cash and carry Local supermarkets DFV and kiosk Grocery shops
Bim Carrefour-SA Dia Gima Kipa
Migros Real Sok Tansas

56
 Figure 3b. Price flexing in personal care products by retail type, 1999:4-2004:3

0.60

0.50

0.40

0.30

0.20

0.10

0.00
q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3
99

00

00

00

00

01

01

01

01

02

02

02

02

03

03

03

03

04

04

04
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
Open bazaar Cash and carry Local supermarkets DFV and kiosk Grocery shops
Bim Carrefour-SA Dia Gima Kipa
Migros Real Sok Tansas

Figure 3c. Price flexing in cleaning products by retail type, 1999:4-2004:3

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00
q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3

q4

q1

q2

q3
99

00

00

00

00

01

01

01

01

02

02

02

02

03

03

03

03

04

04

04
19

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

Open bazaar Local supermarkets Grocery shops Bim Carrefour-SA Cash and carry Tansas

57
Table 12. Determinants of product prices

Retailers Suppliers
Small Medium Large Total Small Medium Large Total
Input costs 2.9 3.0 2.9 2.9 3.0 2.8 2.9 2.9
Demand 2.3 2.4 2.6 2.4 2.0 2.1 2.0 2.0
Other retailers'/suppliers' prices 2.6 2.6 2.6 2.6 2.0 1.9 1.8 2.0
Level of stocks 2.4 2.2 2.0 2.2 1.8 2.0 1.9 1.9
Retailers' financial position 2.1 2.3 1.9 2.1
Payment conditions 2.4 2.6 2.2 2.4
"Private label" 2.5 2.6 2.5 2.5
n 50 74
Scale: 1 no impact, 2 some impact, 3 strong impact

Table 13. Price flexing

Retailers Suppliers
Small Medium Large Total Small Medium Large Total
Price flexing 0.33 0.53 0.75 0.54 0.63 0.61 0.79 0.68
n 50 74

Determinants of price flexing

Regional demand 0.80 0.50 0.33 0.48 0.33 0.43 0.80 0.48
Regional cost differentials 0.60 0.20 0.50 0.41 0.56 0.50 1.00 0.65
Other regional
retailers'/suppliers' prices 1.00 0.80 0.83 0.85 0.44 0.67 0.80 0.60
New stores 0.80 0.60 0.75 0.70
Entry by others 0.80 0.50 0.25 0.44 0.11 0.00 0.60 0.20
n 27 21

58
Table 14. Near- and below-cost sales

Never Once a Once a Once a Always

year month week
Near- or below-cost selling by yourself
Small 0.27 0.00 0.27 0.40 0.07
Medium 0.29 0.07 0.43 0.14 0.07
Large 0.36 0.45 0.09 0.00 0.09
Total 0.30 0.15 0.28 0.20 0.08
n 40

Near- or below-cost selling by other retailers

Small 0.08 0.08 0.31 0.15 0.38
Medium 0.00 0.00 0.53 0.24 0.24
Large 0.20 0.10 0.20 0.00 0.50
Total 0.08 0.05 0.38 0.15 0.35
n 40

Suppliers' assessment
Small 0.43 0.35 0.22 0.00 0.00
Medium 0.54 0.00 0.46 0.00 0.00
Large 0.33 0.33 0.19 0.14 0.00
Total 0.36 0.29 0.23 0.06 0.06
n 66

59
Table 15. Assessment on below-cost sales

Retailers
Small Medium Large Total
Used as a marketing method 0.53 0.50 0.53 0.52
Used to eliminate stocks 0.47 0.58 0.88 0.64
Used to force competitors to exit 0.33 0.32 0.19 0.28
Used to attract customers for other products 0.67 0.68 0.50 0.62
Cannot be used systematically 0.60 0.79 0.81 0.74
Harmul for suppliers 0.60 0.63 0.69 0.64
Makes markets more competitive 0.47 0.32 0.19 0.32
Causes unfair competition 0.93 0.89 0.81 0.88
n 48
Suppliers
Small Medium Large Total
Used as a marketing method 0.30 0.39 0.32 0.33
Used to eliminate stocks 0.70 0.72 0.72 0.71
Used to force competitors to exit 0.35 0.33 0.32 0.33
Used to attract customers for other products 0.43 0.71 0.60 0.57
Cannot be used systematically 0.86 0.89 0.79 0.84
Harmul for suppliers 0.74 0.88 0.88 0.83
Makes markets more competitive 0.35 0.28 0.12 0.24
Causes unfair competition 0.83 0.89 0.96 0.89
n 75

Table 16. Procurement and distribution channels

Retailers Suppliers
Number of Share of Number of Share of
suppliers producers distributors wholesalers retailers distributors wholesalers retailers
Small 178 0.20 0.60 0.19 10 0.47 0.23 0.30
Medium 400 0.27 0.52 0.21 10 0.44 0.29 0.28
Large 1250 0.60 0.31 0.09 28 0.66 0.14 0.20
Total 500 0.35 0.48 0.17 10 0.53 0.21 0.26
n 51 50 60 78
Note: Number of suppliers and number of retailers are median values.

60
Table 17. Retailer-supplier relations

Retailers Suppliers
Small Medium Large Total Small Medium Large Total
Retailers' charges on suppliers for shelf space
Listing (entry) fee 2.2 2.1 1.9 2.1 3.1 2.6 3.1 3.0
Shelf fee 1.4 1.3 1.2 1.3 1.7 1.9 1.8 1.8
Slotting/position fee (gondola, palet, etc) 2.4 2.4 2.3 2.4 3.0 2.6 3.1 2.9
Promotion (insert) fee 2.6 3.0 2.6 2.7 3.1 2.7 3.2 3.0
n 50 71

Supply conditions
Packing by suppliers 1.8 2.6 3.1 2.5 3.7 2.8 3.0 3.2
Returned items taken by suppliers 3.4 3.4 3.1 3.3 3.6 3.1 3.0 3.2
Exclusivity restrictions on suppliers 1.5 1.3 1.1 1.3 1.6 1.5 1.3 1.5
Prices set by suppliers 1.6 1.7 1.8 1.7 2.1 1.9 1.7 1.9
n 50 71

Discounts by suppliers
Quantity purchased 2.8 2.7 2.5 2.7 2.6 2.1 2.4 2.4
Payment period 3.2 3.0 2.9 3.0 2.4 2.7 2.5 2.5
Retailers' promotions 2.7 2.4 2.7 2.6 2.6 2.3 2.4 2.4
Product range 2.5 2.8 2.5 2.6 2.1 1.8 2.0 2.0
Exclusivity agreement 2.2 2.5 2.0 2.2 1.9 1.8 2.1 1.9
Shelf area 2.2 2.3 1.8 2.1 1.4 1.4 1.9 1.6
n 50 72
Scale: 1 never, 2 sometimes, 3 frequently, 4 everytime

Table 18. Characteristics of private label products

Retail Supplier Production Marketing Quality

price price cost cost
Retailers' assessment
Small -0.58 -0.58 -0.25
Medium -0.64 -0.71 -0.21
Large -0.87 -1.00 -0.27
Total -0.71 -0.79 -0.24
n 42

Suppliers' assessment
Small -0.59 -0.29 -0.53 -0.18
Medium -0.58 -0.17 -0.50 -0.25
Large -0.78 -0.35 -0.47 -0.35
Total -0.66 -0.28 -0.50 -0.26
n 54
Index values: -1 lower/worse, 0 same, +1 higher/better

61
Table 19. Obstacles for retailers' growth

Small Medium Large Total

Tax rates 3.3 3.4 2.9 3.2
Land/store availability 2.9 2.9 3.1 3.0
Competition from informal firms 3.1 2.7 3.1 3.0
Competition from large chain stores 3.1 3.1 2.4 2.9
Regulation on opening new markets/shops 2.7 2.3 3.1 2.7
Macroeconomic uncertainty (exchange rate, inflation, etc.) 2.7 2.9 2.5 2.7
Costs of financing (interest rate, etc.) 2.3 2.7 2.0 2.4
Conditions on financing (collateral, etc.) 2.2 2.3 2.1 2.2
Transportation 2.1 2.0 2.3 2.2
Customs and trade regulations 1.9 1.9 2.0 1.9
Consumer demand 2.0 1.8 1.7 1.9
n 48
Scale: 1 does not pose any problem, 2 partially important, 3 important, 4 the most important obstacle.

Table 20. Determinants of retailers' competitiveness

Small Medium Large Total

Stores' location 2.9 2.9 3.0 3.0
Product quality 2.9 2.9 2.9 2.9
Product range/diversity 2.9 2.8 3.0 2.9
Retailer's brand/reputation 2.8 2.8 3.0 2.9
Promotions 2.8 2.8 2.9 2.8
Proximity to consumers 2.9 2.8 2.8 2.8
Price 2.5 3.0 2.8 2.8
Other services provided (parking, etc.) 2.8 2.8 2.6 2.7
Sales services (packing, etc.) 2.9 2.6 2.6 2.7
Stores' appearance (lighting, open area, etc.) 2.9 2.6 2.5 2.7
Product brand 2.7 2.7 2.6 2.6
Loyalty card practices 2.4 2.1 2.1 2.2
n 50
Scale: 1 does not important, 2 partially important, 3 very important

62
Table 21. Expected change in foreign firms' market share in the next decade

Retailers
Small Medium Large Total
Foreign share will increase (% of retailers) 0.75 0.89 0.93 0.86
n 50

Impact on
Retail prices -0.36 -0.12 -0.43 -0.29
Product diversity 1.00 0.76 0.93 0.88
Product quality 0.73 0.59 0.64 0.64
Employment 0.55 0.47 0.79 0.60
Production of domestic suppliers -0.09 -0.06 0.21 0.02
n 42
Suppliers
Small Medium Large Total
Foreign share will increase (% of suppliers) 0.87 0.94 0.88 0.89
n 72

Impact on
Retail prices -0.35 -0.13 -0.22 -0.24
Product diversity 1.00 0.94 0.91 0.95
Product quality 0.80 0.25 0.68 0.60
Employment 0.65 0.44 0.68 0.60
Production of domestic suppliers 0.55 0.37 0.32 0.41
n 64
Index values: 1 increase, 0 no change, -1 decrease.

63
Table 22. Assessment of the draft law on retail sector

Retailers
Small Medium Large Total
A law regulating the retail sector 0.69 1.00 0.69 0.80
Restrictions on private label share (like 20%) 0.25 0.35 -0.25 0.12
Regulations on payment conditions 0.00 0.06 -0.19 -0.04
Restrictions on below-cost sales 0.62 0.84 0.50 0.67
Restrictions on exclusivity agreements 0.12 0.12 -0.13 0.04
Restrictions on promotions -0.31 -0.11 -0.75 -0.37
Regulations on work time (like Sunday holiday) 0.44 0.37 -0.06 0.25
Restrictions on hypermarket locations (outside
residential areas) 0.06 0.11 -0.63 -0.14
Requiring permission to open new markets 0.75 0.32 -0.06 0.33
n 50
Suppliers
Small Medium Large Total
A law regulating the retail sector 0.79 0.89 1.00 0.89
Restrictions on private label share (like 20%) 0.08 0.59 0.48 0.36
Regulations on payment conditions 0.78 0.89 0.72 0.79
Restrictions on below-cost sales 0.83 0.94 0.92 0.90
Restrictions on exclusivity agreements 0.47 0.41 0.41 0.43
Restrictions on promotions 0.08 0.39 -0.04 0.12
Regulations on work time (like Sunday holiday) -0.04 0.22 -0.13 0.00
Restrictions on hypermarket locations (outside
residential areas) 0.37 0.50 0.44 0.43
Requiring permission to open new markets 0.37 0.67 0.52 0.51
n 72
Index values: 1 in favor, 0 indifferent, -1 against.

64