Locational Determinants of Foreign Investors in Istanbul

Abstract

The fact that relationships at global levels have increased necessitates redefinition of the roles of nations, regions, and metropolitan areas. In recent years, during the economic transportation process experienced on a global scale, the traditional role of cities has been changing, and preferences about location selections, which are related to the requirements of urban functions, have displayed changes as well. In this study, the objective is to ascertain the determining characteristics of decisions about location choice by foreign direct investors in Istanbul. The study is based on a sample of 100 companies that were surveyed in 2002. The ranking of location factors has been analyzed using not only variance of means but also principal components factor analysis. The result of the analysis verifies that the processes which influence location choice in Istanbul are similar to those found in other related studies. Since the beginning of the 1990s, Istanbul has become the hub of a vast hinterland extending from the Balkans to the Caucuses and represents a prime focus for foreign investment. For this reason, it is very significant to know the location choice criteria sought especially in the regions of Istanbul, where foreign capital is to be invested, in order to attract foreign-owned companies in the industrial and service sector. In this study, the criteria to which the foreign-owned investments in the industrial and service sector attach significance in location choice are set for each sector.

Full text

Locational Determinants of Foreign Investors in Istanbul
Lale Berköz
1
Abstract: The fact that relationships at global levels have increased necessitates redefinition of the roles of nations, regions, and
metropolitan areas. In recent years, during the economic transportation process experienced on a global scale, the traditional role of cities
has been changing, and preferences about location selections, which are related to the requirements of urban functions, have displayed
changes as well. In this study, the objective is to ascertain the determining characteristics of decisions about location choice by foreign
direct investors in Istanbul. The study is based on a sample of 100 companies that were surveyed in 2002. The ranking of location factors
has been analyzed using not only variance of means but also principal components factor analysis. The result of the analysis verifies that
the processes which influence location choice in Istanbul are similar to those found in other related studies. Since the beginning of the
1990s, Istanbul has become the hub of a vast hinterland extending from the Balkans to the Caucuses and represents a prime focus for
foreign investment. For this reason, it is very significant to know the location choice criteria sought especially in the regions of Istanbul,
where foreign capital is to be invested, in order to attract foreign-owned companies in the industrial and service sector. In this study, the
criteria to which the foreign-owned investments in the industrial and service sector attach significance in location choice are set for each
sector.
DOI: 10.1061/共ASCE兲0733-9488共2005兲131:3共140兲
CE Database subject headings: Turkey; Urban development; Investment; Foreign projects
.
Introduction
In the year 1991, the amount of direct foreign investment equaled
160 billion U.S. dollars. This figure increased to 1.5 trillion dol-
lars by 2000, then decreased to 755 billion dollars in 2001 and
650 billion dollars in 2002. The foreign direct investment 共FDI兲
increase in developing countries remained at equal levels,
whereas there was a great recession in investment in developed
countries, particularly in Europe.
Since 1983 the entire sum of foreign capital has been restruc-
tured in legal, financial, and institutional terms, which has led to
an increase in foreign investment in Turkey. For the past ten
years, the country has witnessed an influx of a significant amount
of credit owned by international institutions and private finance
companies. At the close of 1987, total foreign capital investments
totalled only 655.24 million U.S. dollars. By 1997, this had figure
increased to 1.645 billion dollars. In 1987, a total of 795 foreign
firms had investments in Turkey, while by 1997 the figure was
3,707 firms. It is the service sector in which the increase in for-
eign investments has been the most significant, whereas the influx
of foreign capital into the manufacturing sector has taken place at
a slower pace. Of the 3,707 foreign-owned companies in 1997,
963 were in the manufacturing sector, 98 were in agriculture, and
2,646 were in the service sector. The situation in 1987 was far
different, with figures of 255, 35, and 505, respectively. While
in 1983, the service subsectors of commercial enterprises, hotels,
financial institutions, and banking companies numbered 305,
57, and 24, by 1997 these same subsectors totaled 1,525, 269, and
60 关data provided by the Foreign Investors Association
共YASED兲兴.
The YASED figures illustrate that in 1997 there were 2,153
foreign capital investments in Istanbul. Of these investments,
0.9% were made in agriculture, 19.2% in industry, and 79.9%
in the service sector. 78.9% of the companies investing in the
agricultural sector and 75.6% investing in the industrial sector
were members of OECD countries. OECD countries also repre-
sented a majority of those investing in the service sector 共50.3%兲,
followed by Middle-Eastern and Islamic countries, with 25.3%
共Berköz 2001兲.
There are a great number of studies on analyses of the deter-
minants of the location decision influenced by national or regional
characteristics 共cheap labor, incentives, union strength, etc.兲 in
foreign direct investment literature 共Woodward and Glickman
1991; Hill and Munday 1992; Ondrich and Wasylenko 1993;
Dicken and Quevit 1994兲.
Little 共1978兲 regressed shares of U.S. inward investment
against state characteristics in order to present that foreign-owned
companies’ location decisions were more sensitive than domestic
firms’ decisions to wage differentials and the availability of a port
infrastructure. Using a similar methodology, Glickman and Wood-
ward 共1988兲 made a survey to show how the location of a foreign-
owned plant within the United States is explained by variables
such as energy cost, essential infrastructure, and the labor climate.
Bagchi-Sen and Wheeler 共1989兲 found a significant relation-
ship between foreign-owned service industry investment and
population size and growth rates, together with per capita retail
spending.
Hood and Young 共1983兲 suggested a range of variables that
may be important, but pointed specifically to survey evidence that
stressed market size and incentives. Dunning and Norman 共1979兲
1
Prof. Dr., Istanbul Technical Univ., Faculty of Architecture, Dept. of
Urban and Regional Planning, Taskısla, Taksim, 34437, Istanbul, Turkey.
Note. Discussion open until February 1, 2006. Separate discussions
must be submitted for individual papers. To extend the closing date by
one month, a written request must be filed with the ASCE Managing
Editor. The manuscript for this paper was submitted for review and pos-
sible publication on January 5, 2004; approved on August 19, 2004. This
paper is part of the Journal of Urban Planning and Development, Vol.
131, No. 3, September 1, 2005. ©ASCE, ISSN 0733-9488/2005/3-140–
146/$25.00.
140 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005

examined foreign-owned office location in the United Kingdom
and found that infrastructure variables such as access to airports,
telephone/telex links, and accommodation availability were sig-
nificant determinants of location density.
The current empirical work in both the United States and the
United Kingdom suggests a great number of factors that explain
why some specific regions are attractive. Some of these factors
include infrastructure availability, labor variables, access to adja-
cent markets, and capital incentives 共Little 1978; Hill and Mun-
day 1991; Glickman and Woodward 1988; Arpan 1981兲. What
underlies these common factors is the concern of the foreign in-
vestor for minimizing costs while having access to key markets.
Questionnaire-based approaches examining locational motivation
often assess such influences as low labor costs and the availability
of financial assistance undisclosed for reasons of corporation
image 共Munday 1990兲.
Preferential assistance and spending on new infrastructure and
improvements also affect the decisions governing FDI location.
Such location factors are more likely to vary at the regional level.
Hill and Munday 共1995兲 studied FDI location in the United King-
dom and France and found that financial assistance, manufactur-
ing wages, and regional well-being play an important role in de-
cision making.
According to econometric studies, regional assistance and
infrastructure spending highly affect the regional distribution
of FDI.
The major factors underlying location choice have in recent
years been infrastructure, access to markets, and financial assis-
tance and, with a likely increase in service sector investment,
especially financial and business services. On the other hand,
it is likely that other factors are assuming greater significance,
such as the presence of a highly educated workforce 共Collis and
Noon 1994兲.
There are a number of studies on this subject, including
Tokatlı and Erkip 共1998兲, Balasubramanyam 共1996兲, Balkır
共1993, 1996兲, Togan 共1994兲, and Berköz 共2001兲.
Research Area
The data in this study have been acquired through questionnaires
filled out during personal interviews in 100 leading foreign-
owned companies registered to the Foreign Investors Association
in Istanbul. Fifty companies operate in the industrial sector, and
50 are in the service sector. The companies are marked by their
high ranks in the capital amount they possess. The survey was
carried out through questionnaires filled out during personal
interviews with the company managers between November and
December 2002.
Respondents were asked to identify the most important char-
acteristics of their existing business location and the most critical
factors that influence the choice of location. The characteristics of
the respondents are shown in Tables 1 and 2.
More than half of the industrial companies within the scope
of the study 共52%兲 preferred towns that form the inner zone of
Istanbul as hubs. The data on the number of employees indicate
that 46% of the companies have less than 250, whereas 36% of
them have more than 500 employees. 46% of the companies were
established after 1980, and it is certain that this is due to the
economic, legal, and organizational arrangements made in Turkey
in 1983.
54% of the industrial companies have research and develop-
ment 共R&D兲 units. 20% of these receive R&D service from outer
sources. 44% of them stated that they would need R&D services
in the future. Such companies are planning to meet this re-
quirement either by setting up an R&D unit within the scope of
the company or developing an existing one 共30%兲 or by getting
R&D services from outer sources 共12%兲. The data about the in-
tensity of export facilities show that 46% of the companies export
1–25% whereas 34% of them export 26–50% of the products they
produce.
It has been found that 70% of the service sector companies
within the scope of the study operate in the inner zone of Istanbul,
whereas 14% of them operate in the core zone. It is seen that 40%
of the companies have focused on subsectors including finance,
insurance, and real estate 共Table 2兲. These findings show that the
service sector companies within the scope of this study are rela-
tively new firms, 82% of which were established after 1980 and
58% of which established after 1990. 32% of the companies have
less than 50 employees, whereas 20% of them are firms with
more than 500 employees.
The 100 firms were asked to rate factors on a four-point
Likert–type scale. In studies within Turkey, usually 1–5 or 1–4
scales are used, because the Turkish society is apt to respond to
Table 1. Characteristics of Industrial Sector Firms
Characteristic Frequency Percent
Distribution of industrial firms
Core districts 3 6
Inner districts 26 52
Outer districts 21 42
Distribution of industrial firms by subsector
Food manufacturing 8 16
Ready-made garments 4 8
Medicine 4 8
Chemical industry 3 6
Paper 3 6
Transport equipment 11 22
Other 17 34
Export intensity
1–25% 23 46
26–50% 17 34
51–75% 8 16
76–100% 2 4
Date of Establishment
Before 1980 27 54
1980–1989 11 22
1990–1999 12 24
2000+ — —
Employment range
1–50 8 16
51–100 6 12
101–250 9 18
251–500 9 18
501+ 18 36
Firms undertaking research and development
(R&D)
27 54
Firms not undertaking R&D 23 46
Receiving R&D services from other sources 10 20
Companies planning to receive R&D 22 44
By setting up R&D unit within scope of company
or developing existing one by getting
15 30
R&D services from other sources 612
JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005 / 141

this scale more easily. There are difficulties in feeling slight dif-
ferences due to the intellectual level of the respondents in 1–10
scales; thus, in Turkey a 1–4 scale is preferable.
The significance of a specific factor with respect to location
decision has been measured both by the mean index score and the
percentage of firms which indicated that a factor was either “de-
cisive” or “of major importance” in their decision. The mean
index score has been calculated using the following procedure:
4⫽decisive; 3⫽of major importance; 2⫽of some importance; and
1⫽unimportant.
The factors considered “crucial” in the choice of location are
those with a mean index score of 2.0 or greater and which 50% or
more of the respondent firms indicated to be decisive or of major
importance. Factors “of secondary importance” include all those
with solely a mean index score of 2.0 or greater.
Table 3 presents the arithmetic means and standard deviation
values of the industrial companies. As seen in the table, “reliable
electrical power” is the variable with the highest arithmetic mean
value 共3.58兲; thus, it is the criterion which the company managers
consider to be the most important in location choice. On the other
hand, the criterion considered to be the least important is the
“quality of local basic schools.” Moreover, the criteria with low
arithmetic mean values and high standard deviation values,
namely, “ownership,” “attractive living environment for
managers/administrative staff,” and “existence of related indus-
tries in the environment,” were found to be the least important
criteria.
Table 4 shows that the “customer potential” criterion, with the
highest arithmetic mean value 共3.62兲 and the lowest standard de-
viation value 共0.72兲, is the one considered to be the most impor-
tant by the service sector companies, and this is followed by
“quality of the communications infrastructure,” ”physical condi-
tion of the office,” and “quality of the building.” Conversely, the
ones thought to be the least important are “ownership,” “personal
or family reasons of the owners,” and “relatively cheap real estate
property,” respectively.
The variables with the lowest arithmetic mean values and the
highest standard deviation values were not included in the factor
analysis.
First, using the SPSS package program, a factor analysis of the
industrial sector data was made. Factor analysis is also called a
data reduction method, so instead of multiple variables as a means
of describing a subject, factor analysis can be used, which makes
it easier to describe the subject with fewer variables.
The statistical values presented in Table 5 indicate that 30
variables in the first factor group had a total variability of 17.29%.
The figure for the second factor group was 11.42%; the third
factor group was 9.45%; the fourth factor group was 7.14%; the
fifth factor group was 6.95%; and the sixth factor group was
Table 2. Characteristics of Service Sector Firms
Characteristic Frequency Percent
Distribution of service sector firms
Core districts 7 14
Inner districts 35 70
Outer districts 8 16
Distribution of service sector firms by subsector
FIRE 共finance, insurance, real estate兲 20 40
Administrative 3 6
Telecommunication/communication 5 10
Transportation 1 2
Real trade 5 10
Export-import 10 20
Tourism 6 12
Date of establishment
Before 1980 9 18
1981–1990 12 24
1991–2000 26 52
2001+ 3 6
Employment range
1–50 16 32
51–100 9 18
101–250 10 20
251–500 5 10
501+ 10 20
Table 3. Reasons for Choice of Location by Industry
Variables Mean
Standard
deviation
Reliable electric power 3.58 0.70
Quality of communications infrastructure 3.50 0.84
Public water supply and infrastructure 3.48 0.76
Infrastructure especially designed for industry 3.42 0.76
Labor with required skills 3.40 0.78
Presence of sector subsidiary firms 3.24 0.80
Suitable plot of land 3.24 0.72
Suitability of transportation cost 3.20 0.64
Disposal of waste 3.20 0.90
Financial incentives 3.14 0.76
Availability of parking space 3.10 0.65
Access to developed rail network 3.08 0.75
Technical/maintenance services 3.08 0.85
Relatively cheap real estate property 3.06 0.74
Government persuasion 3.06 0.68
Financial pressure on alternative locations 3.02 0.80
Pleasant surrounding environment 3.02 0.80
Plentiful and cheap labor 3.02 0.65
Ownership 3.02 1.04
Close to major suppliers 2.96 0.90
Cost of public services 2.94 0.77
Suitable building 2.92 0.92
Close to major customers 2.92 0.90
Presence of related industries 2.88 0.96
Limited local competition for products 2.88 0.96
Close to health facilities 2.82 0.72
Distance from specialist services and facilities 2.78 0.62
Property availability for lease 2.72 0.76
Space for expansion 2.70 0.84
Public transportation to plant site 2.64 0.80
Close to international maritime port 2.62 0.81
Distance from international airport 2.48 0.81
Attractive living environment for managers and
administrative staff
2.48 0.99
Employment agencies 2.46 0.95
Image of location 2.44 0.95
Access to developed rail network 2.30 0.76
Personal or family reasons of owners 2.30 0.91
Locally available technical training 2.08 0.80
Quality of local basic schools 1.98 0.87
142 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005

6.16%. The total variation of the sixth factor group cited was
58.44%. The “quartimax” transformation grouped the criteria by
importance in location choice, as reported by company officials,
into six main groups.
Application of Kaiser-Meyer-Olkin 共KMO兲, the measure of
sampling adequacy, and the Bartlett test of sphericity indicated
that the factor analysis results were reliable.
The KMO values of the industrial and service sector samples
suggest that the factor analysis results may be accepted with
confidence 共Norusis 1992兲. The Kaiser-Meyer-Olkin 共KMO兲
measure, which tests sampling adequacy, indicated that a factor
analysis was adequate, because correlations between pairs of vari-
ables can be explained by the other variables.
Consequently, as a result of applying this procedure, the origi-
nal conceptual six factor groups are considered as operational
representatives underlying the complete set of variables.
The first significant factor was the “government support and
urban public services” factor. This factor had a total variance of
17.29%. It was determined that, in the selection of site, industrial
firms with foreign capital attached great importance to criteria
related to management support and the presence of urban public
services. Four variables from this group exceeded the weight of
0.7%, these being “financial pressure on alternative locations,”
“financial incentives,” “quality of communications infrastruc-
ture,” and “cost of public services.”
We named the second factor group “infrastructure and produc-
tion factors.” Its total weight was 11.42%. The second factor
group consisted of two variables with a factor weight exceeding
0.7%, these being “public water supply and infrastructure” and
“disposal of waste.”
The third factor group was named “localization factors.” It had
a total weight of 9.45%. In this group, three variables had a
Table 4. Importance of Location Factors for Service Sector Firms
Variables Mean
Standard
deviation
Customer potential 3.62 0.72
Quality of communications ınfrastructure 3.58 0.70
Physical condition of office 3.48 0.74
Quality of building 3.48 0.65
Visible location 3.46 0.65
Proximity to business center 3.42 0.67
Pleasant surrounding environment 3.32 0.68
Prestigious location 3.28 0.88
Suitable building 3.26 0.78
Labor with required skills 3.22 0.86
Availability of parking space 3.18 0.87
Quality of public services 3.10 0.81
Suitability of type of operations to setting 3.10 0.76
Government persuasion 3.04 0.78
Access to developed road network 3.04 0.73
Quality of ınfrastructure 3.04 0.86
Public transportation to firm site 3.00 0.88
Suitability of transportation cost 2.96 0.83
Distance from specialist services and facilities 2.94 0.68
Cost of public services 2.84 0.89
Proximity to complementary sector 2.80 0.73
Access to international airport 2.76 0.89
Property available for lease 2.72 0.95
Proximity of financial center 2.66 0.77
Lower municipal taxes 2.64 0.85
Technical/maintenance services 2.64 0.85
Proximity to firms in same sector 2.64 0.75
Low cost of land and rents 2.60 0.83
Ownership 2.58 1.03
Employment agencies 2.48 0.84
Personal or family reasons of owners 2.22 1.02
Relatively cheap real estate property 2.00 0.99
Table 5. Principal Components Factor Analysis of Location Determi-
nants by Industry
Factors
Factor
loading
Eigen-
values
Percentage
of variance
1. Government support
and urban public services
6.93 17.29
Financial pressure on
alternative locations
0.816
Financial incentives 0.725
Quality of communications
infrastructure
0.711
Cost of public services 0.710
Close to health facilities 0.626
Presence of sector subsidiary firms 0.600
Presence of industries in same sector 0.566
Technical/maintenance services 0.542
Pleasant surrounding environment 0.538
Space for expansion 0.508
Labor with required skills 0.453
2. Infrastructure and
production factors
2.64 11.42
Public water supply
and infrastructure
0.793
Disposal of waste 0.774
Reliable electric power 0.646
Infrastructure especially
designed for industry
0.627
Close to major suppliers 0.356
3. Localization factors 2.27 9.45
Distance from specialist services
and facilities
0.635
Public transportation to plant site 0.632
Suitable building 0.626
Suitable plot of land 0.596
Availability of parking space 0.453
4. Government persuasion
and accessibility factors
2.15 7.14
Government persuasion 0.723
Access to developed road network 0.616
Access to developed rail network 0.430
Suitability of transportation cost 0.397
Close to international maritime port 0.220
5. Ownership affordability 1.85 6.95
Relatively cheap real estate property −0.610
Property available for lease −0.556
6. Production cost 1.67 6.16
Close to major customers −0.694
Plentiful and cheap labor 0.658
Cumulative percentage 0.58
Note: KMO: 0.613; quartimax rotation. Bartlett: 719.50.
JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005 / 143

weight that exceeded 0.6%, these being “distance from specialist
services and facilities,” “public transportation to plant site,” “suit-
able building,” and “suitable plot of land.”
The fourth factor group was related to “government persuasion
and accessibility.” One variable from this group, “government
persuasion,” had a weight exceeding 0.7%.
We addressed the fifth factor group as “ownership affordabil-
ity.” It gave a total weight of 6.95%.
The sixth factor group was related to cost of production and
turned out to be the least significant group. This factor had a total
weight of 6.16%.
Table 6 shows the results from the factor analysis of the ser-
vice sector data. In this solution, all factors were readily interpret-
able. We labeled factor 1 as “government support and urban pub-
lic services.” Factor loading in this factor consisted of “cost of
public services,” “lower municipal taxes,” “quality of communi-
cations infrastructure,” “cost of public services,” “government
persuasion,” “technical/maintenance services,” “quality of infra-
structure,” and “availability of parking space.”
When it comes to choosing a location, it can be seen that the
managers of service sector firms attach special importance to the
cost of public services, lower municipal taxes, quality of the com-
munications infrastructure, government persuasion, technical/
maintenance services, and quality of the infrastructure.
“Physical condition of building and accessibility,” along
with “quality of building,” “physical condition of office,” and
“access to developed road network,” formed the factors covered
by factor 2.
Factor 3 dealt with the “localization factor,” including state-
ments about “image of location,” “distance from specialist ser-
vices and facilities,” “visible location,” and “pleasant surrounding
environment.”
The “desire for centralized location” formed the focal point of
factor 4. Factor loading on this factor contained the “suitability of
the type of operations to setting,” “customer potential,” “proxim-
ity to complementary sectors,” “proximity to business center,”
“proximity to financial center,” and “suitable building.”
Factor 5 was related to “transportation and cost factors” such
as “public transportation to firm site” and “suitability of transpor-
tation cost.” The variables in this group are those that were at-
tached the least importance by the management of the firms.
The objective of the factor analysis is to explain numerous
variables by a couple of common factors. The cluster analysis
aids in dividing the variables of which natural groups are not
accurately known into subclusters that are similar to one another.
In this study, K-means cluster analysis has been applied in order
to determine how the groups of factors that the firms attach sig-
nificance to in location choice form clusters 共in other words, in
what way such factors are divided into subgroups兲.
The application of K-means cluster analysis on the factor
groups of the industrial firms yields the result that the industrial
firms are divided into two subgroups, as shown in Table 7. It has
been found out that the 41 firms which form the first group attach
the greatest importance to “government support and urban public
services” in location choice, followed by “infrastructure and pro-
duction factors” and “government persuasion and accessibility
factors,” respectively. The firms in this group do not find “pro-
duction cost,” “ownership affordability,” and “localization fac-
tors” as important.
On the other hand, the nine firms in the second group find
“production cost,” “ownership affordability,” and “localization
factors” important, because the firms in this group are in the sub-
sectors of textiles, industrial chemicals, petroleum and coal prod-
ucts, plastics, and tobacco products, and as such they produce
mostly raw materials. This difference is why second group firms
find the variables that decrease production costs significant,
Table 6. Principal Components Factor Analysis of Location Determi-
nants by Service Sector Firms
Factors
Factor
loading
Eigen-
values
Percentage
of variance
1. Government support
and urban public services
4.66 17.55
Quality of public services 0.738
Lower municipal taxes 0.735
Quality of communications
infrastructure
0.709
Cost of public services 0.700
Government persuasion 0.694
Technical/maintenance services 0.602
Quality of infrastructure 0.565
Availability of parking space 0.560
Relatively cheap real estate property 0.550
Proximity of financial center 0.394
2. Physical condition of building
and accessibility
3.10 10.99
Quality of building 0.856
Physical condition of office 0.823
Access to developed road network 0.369
3. Localization factors 2.72 10.65
Image of location 0.852
Distance from specialist
services and facilities
−0.582
Visible location 0.503
4. Desire for centralized location 2.10 10.17
Suitability of type of operations
to setting
0.743
Customer potential 0.649
Proximity to complementary sectors 0.648
Labor with required skills 0.570
Proximity to business center 0.545
Pleasant surrounding environment −0.435
Suitable building 0.106
5. Transportation and cost factors 1.63 7.58
Public transportation to firm site 0.550
Suitability of transportation cost 0.513
Cumulative percentage 56.90
Note: KMO: 0.526; varimax rotation. Bartlett: 565,74.
Table 7. K-Means Cluster Analysis of Industrial Firms
Factor scores of factor groups Cluster 1 Cluster 2
1. Government support and urban
public services
0.20805 −0.94778
2. Infrastructure and production factors 0.17866 −0.81389
3. Localization factors − 0.03650 0.16630
4. Government persuasion
and accessibility factors
0.17345 −0.79014
5. Ownership affordability −0.10722 0.48845
6. Production cost −0.16516 0.75239
Note: Number of cases in each cluster: cluster 1⫽41; cluster 2⫽9;
valid⫽50; mission⫽0.000.
144 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005

whereas “government support and urban public services,” “infra-
structure and production factors,” and “government persuasion
and accessibility factors” are insignificant.
The application of K-means cluster analysis on the factor
groups of the firms of the service sector has revealed the fact that
the service firms are clustered into two subgroups 共Table 8兲.
Forty-one of the 50 firms are in the first group. The subsectors
gathered in this group are finance, insurance, and real estate
共FIRE兲; telecommunications; tourism services; and administration
services. The firms in this group find “government support and
urban public services,” “localization factors,” and “physical con-
dition of building and accessibility” significant, whereas “trans-
portation and cost factors” are insignificant. This is because it is
necessary for the firms of the aforesaid subsectors to be located in
the urban cores, due to their business scopes.
The trade, transportation, and export-import firms attach great
importance to “transportation and cost factors” in location choice,
because it is the factor group required in such business sectors.
Conclusion
Turkey has been undergoing a process of restructuring its rela-
tionship within the world economies since 1980. In the 1980s,
with the purpose of alluring foreign capital investment into the
country, some economic, legal, and organizational changes were
made. Istanbul, due to its propinquity to regional markets and its
variety of natural resources, has an important place in the eco-
nomic system of the world in the globalizing process. To put it in
another way, Istanbul holds the potential both of controlling a
huge interregional network and of acting as the center for com-
mercial activities.
This study aimed at determining the factors that were attached
importance by firms with foreign capital in determining the loca-
tion of their site, aided by data from a questionnaire covering 50
industrial and 50 service sector companies that invested in Istan-
bul, with the purpose of establishing the preferences of location of
firms with foreign capital in the metropolitan area of Istanbul.
The findings of this study show that FDI firms tend to be
located in the inner zone of Istanbul.
The mean results’ variance shows that the most important lo-
cation determinants for the industry were reliable electric power,
quality of the communications infrastructure, the public water
supply and infrastructure, and infrastructure especially designed
for industries.
The variance of mean results confirms that reliable customer
potential, quality of communications infrastructure, physical con-
dition of office, and quality of building were the highest important
determinants of location choice by service sector firms of foreign
investors.
As per the results of factor analysis of the industrial sector of
FDI locations, six factor categories were formed, each showing
the variables rated as important by company officials. The “gov-
ernment support and urban public services” factor was found to
be the factor group of greatest importance. Second in importance
was “infrastructure provisions and production,” while the third
most important factor was “localization factors,” the fourth con-
tained “accessibility and cost factors,” the fifth was “ownership
affordability,” and the sixth and least important factor was the one
concerning “production cost.”
The study showed that service sector of FDI found “govern-
ment support and urban public services” critical in their selection
of a location. “Physical condition of building and accessibility”
turned out to be the most important factor in the choice of
location. The third most important factor was the “localization
factor,” the fourth factor was “desire for centralized location,” and
“transportation and cost factor” was the fifth and least important
factor.
The results of this study show that government persuasion
and financial incentives do have some effect on the location of
foreign investments. Similar results were obtained by Hill and
Munday 共1992兲.
Factor analyses were applied, respectively, to both sets of cri-
teria, as attached importance by the industry and service sectors in
their selection of location. Upon factor analysis and grouping of
important criteria by both sectors, it was found that “government
support and urban public services” was the factor that was at-
tached greatest priority and importance.
Applying K-means cluster analysis on the factor groups ob-
tained as a result of factor analysis yielded two subgroups in not
only the industrial but also the service sector. In both of the sec-
tors, 82% of the total firms are collected in the first groups.
Within the industrial and service sectors, the sectors gathered in
the first subgroup attach the utmost importance to “government
support and urban services” in location choice.
This finding supports the results of the factor analysis. In both
of the sectors 共industrial and service兲, there are fewer firms gath-
ered in the second subgroup 共12% of the total firms were in this
group兲 and they find “cost reducing factors” significant, as this is
something necessitated by their sector structure.
This research shows that the similarity of the factors that were
attached greatest importance in site selection by both industrial
and service sector firms with foreign capital, in spite of their
different structure, was due to their primary need for government
support and reliability, unlike the needs of domestic companies.
Thus, the results of this survey and those of earlier studies about
the locational determinants of FIRE 共Berköz 2000兲 and producer
services 共Berköz 1998兲 in the Istanbul metropolitan area confirm
that the criteria which were attached greatest importance by do-
mestic and foreign companies in determining location differ from
each other.
This study further showed that, in developing countries
like Turkey, reliability and support by the government and the
presence and quality of urban public services were the loca-
tional incentives that most attracted the foreign investor. In
addition, whereas industrial firms with foreign capital tend to pre-
fer areas with first-rate infrastructural conditions, with a fine lo-
cation and high accessibility, service sector firms prefer areas
with a fine location, high accessibility, and buildings in good
physical condition.
The results of this research have important implications for
Turkey. Governments must make necessary legal arrangements
that attract foreign capital and reduce long-lasting procedures.
Table 8. K-Means Cluster Analysis of Service Firms
Factor scores of factor groups Cluster 1 Cluster 2
1. Government support and urban
public services
0.34207 −1.55832
2. Physical condition of building
and accessibility
0.11185 −0.50955
3. Localization factors 0.17119 −0.77985
4. Desire for centralized location 0.03227 −0.14703
5. Transportation and cost factors −0.04171 0.18999
Note: Number of cases in each cluster: cluster 1⫽41; cluster 2⫽9;
valid⫽50; missing⫽0.000.
JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005 / 145

However, the government must also ensure economic stability,
which will make it reliable both within the country and abroad.
Moreover, it will be more possible for foreign investors to invest
in Turkey if municipalities apply the location-determining criteria
in this research in industrial and central areas, thus rendering
conditions more attractive.
References
Arpan, J. 共1981兲. “The impact of state incentives on foreign investors’ site
selections.” Econ. Rev.,66共8兲, 36–42.
Bagchi-Sen, S., and Wheeler, J. D. 共1989兲. “A spatial and temporal model
of foreign investment in the US.” Econ. Geol.,65共2兲, 113–129.
Balasubramanyam, V. N. 共1996兲. “Foreign direct investment in Turkey.”
The economy of Turkey since liberalisation, S. Togan and V. N. Bala-
subramanyam, eds., Macmillan, London, 112–130.
Balkır, C. 共1993兲. “Turkey and the European Community: Foreign trade
and direct foreign investment in the 1980s.” Turkey and Europe,C.
Balkır and A. M. Williams, eds., Pinter, London.
Balkir, C. 共1996兲. “Türkiye ve Avrupa Topluluğu: 1980’lerde Dıș Ticaret
ve Doğrudan Dıș Yatırımlar.” Türkiye ve Avrupa likileri, C. Balkır and
A. M. Williams, eds., Sarmal Yayınevi, Istanbul, Turkey, 137–188 共in
Turkish兲.
Berköz, L. 共1998兲. “Locational preferences of producer service firms in
Istanbul.” European Planning Studies,6共3兲, 333–349.
Berköz, L. 共2000兲. “Location of financial, insurance, and real estate firms
in Istanbul.” J. Urban Plann. Dev., 126共2兲, 75–88.
Berköz, L. 共2001兲. “The interregional location of foreign investors in
Turkey.” European Planning Studies,9共8兲, 979–994.
Collis, C., and Noon, D. 共
1994兲. “Foreign direct investment in the UK
regions: Recent trends and policy issues.” Regional Studies,28共8兲,
843–848.
Dicken, P., and Quevit, M., eds. 共1994兲. “Transnational corporations and
European regional restructuring.” Geographical Studies 181, Royal
Dutch Geographical Society, Utrecht, The Netherlands.
Dunning, J. H., and Norman, G. 共1979兲. Factors influencing the location
of offices’ muitinational enterprises, Location of Offices Bureau and
Economists Advisory Group, London.
Glickman, N., and Woodward, D. 共1988兲. “The location of foreign direct
investment in the US: Patterns and determinants.” Int. Region. Sci.
Rev.,11共2兲, 137–154.
Hill, S., and Munday, M. 共1991兲. “The determinants of inward invest-
ment: A Welsh analysis.” Appl. Econom.,23共1兲, 761–769.
Hill, S., and Munday, M. 共1992兲. “The UK regional distribution of for-
eign direct investment: Analysis and determinants.” Reg. Studies, 26,
535–544.
Hill, S., and Munday, M. 共1995兲. “Foreign manufacturing investment in
France and the UK: A regional analysis of locational determinants.”
Tijdschfift voor Economische en Sociale Geografie, 86, 311–327.
Hood, N., and Young, S. 共1983兲. Multinational investment strategies in
the British Isles, Her Majesty’s Stationery Office, London.
Little, J. S. 共1978兲. “Locational decisions of foreign direct investors in the
US, New England.” Econ. Rev., July/August, 43–63.
Munday, M. 共1990兲. Japanese manufacturing investment in Wales, Uni-
versity of Wales, Cardiff, Wales,
Norusis, M. J. 共1992兲. SPSS for Windows professional statistics
, SPSS
Inc., Chicago.
Ondrich, J., and Wasylenko, M. 共1993兲. Foreign direct investment in the
United States, Upjohn Institute for Employment Research, Kalmazoo,
Mich.
Togan, S. 共1994兲. Foreign trade regime and trade liberalization in Turkey
during the 1980s, Avebury, Aldershot, U.K.
Tokath, N., and Erkip, F. 共1998兲. “Foreign investment in producer ser-
vices.” TWPR,20共1兲, 87–106.
Woodward, D. P., and Glickman, N. J. 共1991兲. “Regional and local deter-
minants of foreign firm location in the United States.” Industry loca-
tion and public policy, H. W. Herzog and A. Schlottmann, eds., Uni-
versity of Tennessee Press, Knoxville, Tenn. 190–220.
146 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / SEPTEMBER 2005