TÜSİAD-KOÇ UNIVERSITY ECONOMIC RESEARCH FORUM
WORKING PAPER SERIES
INTEGRATION WITH THE GLOBAL ECONOMY
The Case of Turkish Automobile and Consumer Electronics Industries
Erol Taymaz
Kamil Yılmaz
Working Paper 0801
February 2008
________________________________________________________________
TÜSİAD-KOÇ UNIVERSITY ECONOMIC RESEARCH FORUM
Rumeli Feneri Yolu 34450 Sarıyer/Istanbul
Integration with the Global Economy
The Case of Turkish Automobile and Consumer
Electronics Industries*
Erol Taymaz**
Kamil Yılmaz***
METU Koç University
Abstract
This paper aims to contribute to the extensive study of the World Bank Commission on Growth and
Development by a case study of the Turkish automotive and the consumer electronics industries. Despite a
macroeconomic environment that inhibits investment and growth, both industries have achieved remarkable
output and productivity growth since the early 1990s and played a critical role in generating employment and
fostering growth. Although there are similarities between the performances of automobile and consumer
electronic industries, there seems to be significant differences between their structures, links with domestic
suppliers, technological orientation and modes of integration with the global economy. The automobile industry
is dominated by multinational companies, has a strong domestic supplier base, and has seized the opportunities
opened up by the Customs Union by investing in new product and process technology and learning. The
consumer electronics industry is dominated by a few, large domestic firms, and has become competitive in the
European market thanks to its geographical proximity, productive domestic labor, and focus on a protected and
technologically mature CRT color television receivers segment of the marker, which also helps explain the
recent decline in industry’s fortunes. It is without doubt that these industries could have performed even better
had governments in Turkey adopted more responsive macroeconomic policies. It is certain that governments
could be more responsive only if far-reaching political/institutional reforms are undertaken by changing the
Constitution, and current political party and election laws in order to establish public control over the political
elites.
JEL Classification: F14, L60, P16.
Keywords: Macroeconomic policies; automobile industry; consumer electronics industry; political
elites; political reforms.
* This is a revised version of the paper prepared for the Country Case Studies Workshop organized by the
Commission on Growth and Development (Washington DC, April 12-14, 2007). We would like to thank our
discussants, Homi Kharas and Martin Rama, and Zafer Mustafaoğlu and other workshop participants for their
very helpful comments and suggestions.
** Erol Taymaz, Department of Economics, Middle East Technical University, Ankara 06531 Turkey. e-mail:
etaymaz@metu.edu.tr
*** Kamil Yılmaz, Department of Economics, Koc University, Istanbul 34450 Turkey. e-mail:
kyilmaz@ku.edu.tr
1
1. Introduction
Why do growth rates differ? This is one of the fundamental questions that have raised intense debate
among economists, policy makers and ordinary citizens for decades if not centuries. Although there is
no consensus among economists on explaining growth rate differentials across
regions/nations/countries, recent theoretical and empirical studies emphasize the critical role of
institutions, the macroeconomic environment and (macroeconomic) policies.
This paper aims to contribute to the extensive study of the World Bank Commission on Growth and
Development by providing a case study of the automotive and the consumer electronics industries in
Turkey. These industries are selected from Turkey because they have achieved a remarkable growth in
output, productivity and exports in the last decade against all odds. We attempt to uncover how these
industries have performed so well, especially in international markets, and what role macroeconomic
policies have played in their success.
The paper is organized as follows. Section 2 summarizes macroeconomic problems, policies and
performance in the long term. After a brief look at the four periods of growth and decline since 1923,
macroeconomic policies adopted since 1980 are discussed in detail to provide the background
information. Section 3 focuses on the quantitative measures of industrial performance, and provides
the data on production, productivity, exports and vertical integration, and profitability for Turkish
automobile and consumer electronics industries. The factors behind the performance of these two
industries and the role of macroeconomic policies, with a special emphasis on the effects of the
Customs Union with the European Union in 1996, are discussed in Section 4. Section 5 introduces the
issue of sustainability and its preconditions, namely the need for political reform. Section 6
summarizes the main findings of the study.
2. Background: Macroeconomic policies and performance
2.1. Long term growth patterns in the Turkish economy
Even a cursory look at the patterns of economic growth and economic policies in Turkey reveals that
there are four distinct periods of economic growth since the establishment of the Turkish Republic in
1923 (see Figure 1).1
The first period from 1923 until the Second World War is marked with high but
widely fluctuating growth rates that reflect economic conditions of that era (the dominance of
agricultural production, the lack of any industry, and the recession in the world economy in the early
1930s). The thrust of economic policy was to establish main industries (textile, food, chemicals and
light engineering) by the state. Because of the initial low level of income, the economy achieved
substantial growth rates in this period.
The second period from the Second World War to 1960 is the period of transition towards a multiparty
political system. The first multi-party elections were held in 1946, and the second multi-party
elections in 1950 put an end to three decade long rule of the Republican People’s Party. The new
government achieved rapid growth in the first half of 1950s, but this period was ended by a military
coup in May 1960, following the widespread discontent due to the corruption allegations, the
government’s tendency towards authoritarianism and slow economic growth in the second half of
1950s.
The third period from 1960 to 1980 is characterized by a new (more democratic) constitution and a
new set of institutions. The new 1961 constitution envisaged (indicative) planning as an important tool
for economic development and instigated the State Planning Organization in charge of preparing five-
1
Figure 1 depicts 5-year moving average values for annual rates of real GDP growth in Turkey. The forecasted
value for 2007 is used in calculating the average for 2005.
2
year development plans covering all aspects of economic development. Turkey adopted import
substitution industrialization (ISI) strategy in the 1960s and 1970s, and achieved remarkably high and
relatively stable growth rates in these decades. The state played an active role in developing a number
of industries that produce intermediate products and (especially in the 1970s) machinery and
equipment through state-owned enterprises (SOEs). The share of SOEs in manufacturing employment
and value added exceeded 35 and 40 percent, respectively, in the late 1970s.2
The SOEs had a
dominant position in resource and scale intensive industries (paper, chemicals, cement, iron and steel,
certain types of machinery and equipment) and produced more than half of the value added of these
sectors.
As in many other countries that adopted ISI, Turkey faced with a serious balance of payment crisis in
the late 1970s as a result of rapid increase in the cost of oil imports and increasing import needs during
the process of capital deepening. The balance of payment problems slowed down economic growth,
and caused even a decline in GDP in 1979 and 1980. The period of ISI was ended by another military
coup in 1980.
Turkey adopted export-oriented “industrialization” (EOI) policies after 1980. A wide range of
liberalization programs were implemented in the 1980s: The foreign trade regime was liberalized to a
large extent in the 1980s that was followed by full capital account liberalization in 1989. Exports were
encouraged by various means (more than 100 percent devaluation of the Turkish Lira in 1980,
generous export subsidies throughout the 1980s, and decreasing real wages until 1988). Labor and
resource intensive sectors were the first to respond to these measures, and total volume of exports (in
US$) increased more than 4-fold from 1980 to 1989. Economic growth rates increased in the 1980s
thanks to the export boom and easing balance of payments problems, but macroeconomic environment
deteriorated sharply in the 1990s and the economy was trapped into rising boom and bust cycles that
finally ended with a devastating economic crisis in 2001 (for an assessment of the industrialization
experience since 1960, see Şenses and Taymaz, 2003).
The Turkish economy once again achieved historically high growth rates since 2001. GDP grew at
high rates for 5-years in a row since 2001 (the average annual growth rate of GDP reached 7 percent)
without any sign of reversal in the growth rate in 2007. Exports increased at phenomenal rates during
and after the crisis in 2001: the average annual growth rate of the value of exports (in US$) was about
25 percent in the period 2001-2006. Two sectors, automobiles and consumer electronics, are among
the leading sectors behind the export boom. Automobile exports (including parts and components)
increased at 45 percent per year whereas the growth rate of consumer electronics exports exceeded 30
percent per year since 2001. In the following sections of the paper, macroeconomic policies since the
1980s and the evolution of automobile and consumer electronics industries will be examined in detail
to shed light on the factors behind their export performance.
2.2. Macroeconomic policies and performance, 1980-2001
Turkey followed an inward-looking, import-substituting development strategy during the 1960s and
1970s. While this strategy worked well throughout the 1960s, it proved impossible to continue with it
in the wake of oil price hikes in 1974 and 1979. After the first oil price shock the government decided
not to reflect the international oil price hikes to the domestic market. As a result, the demand for
gasoline continued to be strong and the trade deficit soared. The budget deficit also soared thanks to
the government decision to pay most of the oil import bill itself. At the same time, Turkey had gone
through the worst political instability in its history. These developments led to the balance of payments
crisis of 1979 and a jump in inflation rate to 64 percent.
The first serious attempt to take inflation under control was announced on January 24, 1980. The
stabilization package, which contained a 70 percent devaluation of the Turkish Lira, cut in government
2
The data refers to all SOEs and private establishments employing 10 or more workers.
3
expenditures and substantial increase in the government controlled prices, including that of gasoline.
The result was a jump in inflation close to 100 percent (see Figure 2). On January 24, 1980, a date
generally considered as the beginning of a new economic era, a comprehensive IMF-backed
stabilization program was launched following a major currency crisis. The immediate objective of the
program was to stabilize the economy by improving the balance of payments and containing inflation.
The long-term goal, however, was much more ambitious: to change the structure of the economy
fundamentally.
Another important objective of the program was to remove the dominance of the state in key industries
and in banking, and to minimize the state’s intervention with the pricing and resource allocation
processes of the market economy. In order to accomplish the external adjustment, export-oriented
growth became the key policy objective and the government used export subsidies to promote exports.
Starting from 1984, export oriented policies had later been coupled with a gradual reduction in tariffs
and non-tariff barriers, and finally culminated in the Customs Union (CU) with the European Union
(EU) in 1996. As part of the efforts to increase exports, from 1980 to 1988 the government allowed
Turkish Lira to depreciate in real terms by 36 percent (Figure 3).
Although a lot had been achieved until 1986, one thing that was conspicuously lacking was the control
over the fiscal deficit. Despite favorable domestic and international circumstances, neither the military
government of 1980-83 nor the Özal government of 1983-87 was able to cut the budget deficit on a
permanent basis. Actually, there were some positive signs in the first few years of the decade. As a
result of the pressure on real wages and cuts in politically motivated government expenditures, public
sector borrowing requirement was reduced from 10 percent of GNP in 1980 to 5.4 percent in 1981.
The inflation brought down, from 107 percent in 1980, to 37 percent in 1981 and 28 percent in 1982.
However, 28 percent has since proven to be the lower limit of annual inflation in Turkey until recently
(Figure 2)
The reorientation of the Turkish economy had its effects in the 1980s. From 1981 to 1987 the
economy attained annual growth rates above 4 percent per annum. However, this initial successful
phase of reforms could not be sustained. While the external adjustment of the economy could be
sustained over the long-run, there were real problems emerging in the internal adjustment process.
With the free general elections of 1987, the full competition in politics restarted. In the wake of
intensified political competition, populist policies were discovered. During the 1987-2001 era,
economic performance was very erratic. A strong growth performance in a year or two was followed
by a rather sharp decline in the growth rate to negative territory (Figure 4). As a result, Turkey
achieved only an average of 2 percent growth rate in per capita GDP over this period.
In the second half of 1980s, the government issued decrees to remove some of the barriers to foreign
direct investment (FDI). As a result, the FDI inflows increased above the dismally low levels of the
period before. While the average annual FDI inflows between 1981 and 1987 was only US$ 110
million, after the Government decree to promote FDI was put into effect in 1986, the FDI inflows
increased to US$ 354 million in 1989, to US$ 662 million and US$ 684 million in 1990.
Trade and financial liberalization efforts of the decade finally culminated in the opening of the capital
account in August 1989. While the decision to open the capital account can be viewed as a step in the
right direction, at the time it was criticized on the grounds that it was an immature decision. While the
reforms to liberalize financial markets had already been undertaken at the time, an effective regulation
system for the financial sector was not put in place. Another criticism for this decision stems from the
fact that the government’s decision to liberalize the capital account in 1989 was mostly driven by its
desperate need to finance increasing budget deficits. As the borrowing through the domestic market
reached its limits the interest rates started to increase. At a time of intense political competition, the
government ruled out any attempts to tighten the fiscal policy. The capital account liberalization was
an immediate solution for the government at the time.
4
With the decision to liberalize capital account, government decided to move away from an exchange
rate policy that was geared to promote exports, towards one that views the exchange rate as an implicit
anchor for domestic prices. This new function for the exchange rate is also enforced by the portfolio
capital inflows that led to real appreciation in TL. From 1988 to 1993 Turkish Lira appreciated by 18
percent, before a maxi devaluation during the 1994 crisis (Figure 3).
While Turkey successfully liberalized its foreign trade regime, removed price ceilings on goods and
services and other distortions in product markets, and deregulated its financial sector long before many
Latin American and East European countries, she was not able to attain macroeconomic stability.
The inflation varied widely: From close to 100 percent in 1980 it declined to 34.6 percent in 1986; it
increased over time and reached 140 percent during the economic crisis of 1994. As can be seen in
Figures 1 and 4, output growth has also been volatile with periods of rapid growth followed by sharp
contractions. As the Turkish Lira lost its store of value function, foreign exchange deposits became
increasingly important at times representing almost 50 percent more than the domestic money supply
measure M2 (during the 1994 economic crisis and after September 11, 2001).
The reason for the volatile macroeconomic environment and high inflation has been the inability for
successive governments to deal with the underlying causes of poor public finances. During 1980’s the
government’s reform efforts focused on external adjustment. The internal fiscal adjustment that was
necessary to reduce inflation permanently was never achieved nor seriously attempted. Despite early
fiscal successes following a stabilization program, public sector deficits soon resumed its secular
climb. Fiscal dynamics drove monetary and exchange rate policy mix and set the key parameters for
evolution of the economy and the functioning of financial markets.
The capital account was opened untimely in 1989 under unstable macroeconomic conditions. Under
the open capital account, the Central Bank adopted a managed exchange rate policy to control
inflation. When there is no adjustment of fiscal balances, a managed float can be instrumental in
keeping the inflation under control only for a while. Eventually, the pegged exchange rate policy was
doomed to fail. Appreciated domestic currency increases the vulnerability of the economy to external
shocks and domestic policy mistakes. This process explains the crises of 1994 and 2001 in Turkey.
The economic crisis of 1994 did not alter the course of fiscal policies. Domestic and foreign investors
were willing to finance the fiscal deficits as long as the Treasury was willing to offer higher and higher
yields. In the mean time, the government made full use of the state-owned banks to finance the fiscal
deficit through short-term loans to the Treasury. It seemed at the time that the system would have
worked fine as long as the short-term capital inflows continued. But that was not the case. Chronic
fiscal deficits coupled with high real interest rates led to the unsustainable debt dynamics. Public
sector borrowing requirements increased from 5 percent of GNP in 1995 to as high as 15.6 percent in
1999 and the ex-post real interest rates on bonds and T-bills were above 20 percent, reaching as high
as 36 percent (see Table 1).
Both domestic and external shocks in the final year of the 1990s exposed the impossibility of
sustaining macroeconomic policies. Turkish economy was directly affected from the repercussions of
the Russian debt crisis of 1998. As Turkey also suffered from chronic fiscal imbalances, it experienced
an outflow of short-term portfolio investments following the outburst of the crisis in August 1998.
Starting in September 1998, the economy slowed down gradually. Economic downturn intensified
throughout 1999, especially after the devastating earthquake in the Marmara region on August 17,
1999.
Following the earthquake, the government was convinced for the need of a stabilization program. At
the end of 1999, the government and the IMF agreed upon an exchange rate-based disinflation
program that went into effect in 2000. The program entailed tight fiscal policy along with a crawling
peg exchange rate regime under which the exchange rate was announced in advance in order to anchor
inflation expectations. The exchange rate based stabilization was expected to bring real interest rates
5
down and ensure the sustainability of the already high public debt. Another important component of
the program was the banking sector reform, entailing the establishment of an effective banking sector
regulatory body.
Initially, the program had the desired impact on interest rates. From above 100 percent in 1999, the Tbill
rate fell below 40 percent in the first half of 2000 and declined further below to 35 percent in the
second half of 2000. Inflation, on the other hand, declined gradually from 63 percent in December
1999 to 33 percent in December 2000. However, not unlike other countries that had implemented the
exchange-rate-based programs in the past, in February 2001 Turkey witnessed a dramatic collapse of
the disinflation program and experienced the worst economic crisis in its history with a quasimeltdown
in financial markets.
Even though the crisis was triggered by a confrontation between the President and the Prime Minister
in a top level meeting, there were several factors that jointly contributed to the outburst of the crisis.
To start with, there were problems with the initial design of the program. Reached as a compromise
between the government and the IMF, the crawling peg regime could not bring about a rapid decline in
inflationary expectations and hence the actual inflation rate. The ensuing rapid overvaluation of the
lira proved to be fatal in the presence of a financial system with many loss-making public and private
banks. In the weak and unregulated Turkish financial system, the crawling peg regime further
strengthened the local banks’ appetite for external borrowing and increased their vulnerability to a
rapid depreciation of the Lira. The inability of the government to implement the structural measures
entailed in the disinflation program was also another key factor that led to the crisis. At the time there
was a coalition government in power with little coordination and effective governance. In the first
three quarters of 2000, the government gave in to the pressure from bank owners and delayed the
establishment of the Banking Regulation and Supervisory Agency. This delay already alarmed foreign
investors and convinced some to move their portfolio investments outside the country.
An important milestone for Turkey during the 1990s was the formation of the Customs Union (CU)
between Turkey and the EU in 1996. Experience of the Turkish economy after joining the CU
provides insights for its future performance as it integrates more with the EU economies. That is why,
the effect of the CU on Turkish economy has to be analyzed briefly even though it’s not part of the
macroeconomic policy framework.
Traditionally Turkey had strong trade relations with Europe. Between 1999 and 2003, trade with the
EU-25 accounted for 53 percent of Turkey’s exports and 51 percent of imports. Defining Europe to
include countries that have become members in 2004, and Bulgaria and Romania that have become
members in 2007, the share of Europe in Turkish exports and imports in 2004 rises to 59 percent and
53 percent, respectively. Even though, as a result of the rapid increase in imports from China in the
2000s its share in imports declined to 49 percent by the end of 2005, the EU continues to be the most
important trade partner for Turkey.
The overall impact of the CU on the Turkish economy has been positive, but less than its full potential.
First, there was a small decline in import tax revenues as Turkey lowered tariffs for imports from the
EU3
. Import tariff revenues fell from 2.8 percent of total tax revenues in 1995 to 1.1 percent average
over the last five years (2001-2005). The decline in tax revenues, however, is too small to be blamed
for the large budget deficits.
Increased competition in the form of higher imports from the EU forced productivity improvements in
the manufacturing industry. Some of these sectors such as automotive, durable home appliances,
electrical machinery and equipment had continued to receive protection behind high tariff barriers
despite the import liberalization process that started a decade ago. Compared to other sectors,
productivity growth was higher in import-competing sectors.
3
According to calculations reported in Togan (1997), un-weighted average tariff rate for the manufacturing
industry decreased from 13.5 percent in 1995 to 3.6 percent in 1996.
6
The track record of the Turkish manufacturing industry in response to the CU has been better than
expected, especially when one considers that Turkey received very little financial support from the EU
to help ease the adjustment burden. In Taymaz and Yılmaz (2007), we showed that even though the
total factor productivity in the manufacturing industry did not increase much after the CU, the
productivity in those sectors that experienced significant increases in important penetration rates
increased substantially.
The CU agreement with the EU did not have much impact on Turkish exports in the first five years
(for growth rates of exports and imports, see Figure 5). The EU had already removed tariffs on imports
from Turkey long before the CU went into effect. In addition, despite the CU, the EU continued to
reserve the right to impose anti-dumping duties on Turkish exports to the EU as well as keeping
technical (regulation) barriers. Coupled with the appreciation of the Lira, it is therefore not surprising
that Turkish exports did not surge to the EU countries immediately after the CU.
The impact of the CU on Turkish exports was realized with a long delay, only after the 2001 crisis.
The depreciation of the Lira and the contraction in domestic demand that followed the economic crisis
of February 2001 forced domestic producers to search for export markets. Export revenues increased
by only 12.6 percent in 2001. Exports continued to grow even after the domestic market resumed
growth in 2002 and 2003 at a rate higher than the period prior to the crisis. Exports grew by 15 percent
in 2002, 31 and 34 percents in 2003 and 2004 and by 16 percent in 2005 and 2006. Better-thanexpected
export performance in 2002 and 2003 was achieved despite a 25 percent real appreciation,
and even nominal appreciation, of the Turkish Lira during this period. This remarkable export
performance is in part due to the newly acquired competitiveness of the Turkish manufacturing
industries that was forced by the increased competition after Turkey joined the CU.
The major disappointment with the CU has been the lack of any change in FDI flows to Turkey. It was
expected that FDI inflows (especially those from Europe) to Turkey would increase substantially
following the CU agreement. Turkish governments had hoped that multinational companies would
have taken advantage of the country’s geographical location, low-cost and disciplined labor, and long
experience with a market economy. Unfortunately, this expectation did not come true. Four years after
the CU, in 2000 Turkey attracted US$ 15 of FDI per capita. The same year Poland attracted US$ 105,
Romania US$ 50, Brazil US$ 96, Algeria US$ 32, and Ireland US$ 4,750.
Unlike the CU, the EU’s decision at the end of 2004 to start accession talks with Turkey had
substantial impact on FDI inflows to Turkey. As a result of the structural reforms implemented since
2001, this time around the macroeconomic policies and structural reforms have created an investment
climate that is much more suitable for FDI inflows. Experience with the CU so far demonstrates the
resiliency and adaptability of Turkish manufacturing industries to rapidly changing conditions during a
period with almost no net FDI.
2.3. Macroeconomic policies and performance since 2001
The 2001 crisis was the worst economic crisis Turkey had ever experienced since its foundation in
1923. As a result of the crisis, GNP dropped by 9.5 percent in 2001. A short period of turmoil in
financial markets had a severe toll on public finances. The government debt has increased by more
than 40 percent of GDP by the end of the year. As the Lira depreciated by 30 percent in six months,
inflation picked up very rapidly to reach 70 percent by the end of the year.
A huge IMF bail-out package was prepared with the conditionality of the implementation of a
stabilization program. Kemal Derviş of the World Bank was invited by the Prime Minister Ecevit to be
in charge of the economic recovery program. Derviş played a pivotal role to ensure the success of the
post-crisis recovery efforts. The program was eventually put in place to address the two bottlenecks of
the economy: namely, to ensure the sustainability of the government debt burden through a primary
7
surplus of 6.5 percent of GNP over a couple of years, and to rescue the banking system. The banking
sector reform was implemented without any further delay.
More than a year after the crisis, the coalition government collapsed in July 2002 and the country was
set to go for yet another early general election in November 2002. In the November 2002 elections, a
recently founded party, AKP (Justice and Development Party) received 34 percent of the votes and
formed a single party government. Initially, the government sent populist messages to the public
hoping to obtain a large financial aid from the United States in exchange for the opening of the
northern front in Iraq through Turkish border. However, as the Turkish parliament rejected the motion
to open a corridor for the US forces to the Northern Iraq, the government had to take drastic measures.
Under these conditions, the government declared that it would stick to the primary surplus of 6.5
percent of GNP target of the IMF-supported program. Since then, the AKP government succeeded in
achieving the 6.5 percent of GNP target over the last four years (Figure 6). As foreseen in the
“Transition to a Strong Economy” program prepared by Kemal Derviş and his team, the persistent
primary surplus over the last five years led to a gradual decline in the overall fiscal deficit and the
public sector borrowing requirement (see Figure 6 and Table 1). As the public sector’s demand for
loanable funds declined over time so did the nominal and real interest rates. The nominal interest rates
declined from 60 percent in 2002 to less than 20 percent in early 2005. The real interest rates, on the
other hand, declined from 25 percent per annum during 2002, to close to 10 percent in mid-2003
(Figure 7).
The last five years’ economic performance was not only a result of the successful implementation of
macroeconomic reforms. In addition, the prevalence of a very benign international environment for
emerging markets since 2003 helped Turkey grow faster than the domestic savings would have
allowed. Due to the easy monetary policy and the resulting low interest rates in the US and other
industrial countries, financial capital has been consistently flowing to emerging markets. In addition,
high and robust growth rates attained by China and India along with the ever-growing US helped pull
the world economy along. While the international portfolio flows to Turkey over the 2002-2005 period
and the ensuing overvaluation of the Lira helped the inflation stabilization efforts, a reversal would
lead to a jump in the expected and actual inflation. Developments in international financial markets in
May-June 2006 showed how closely is the success of the inflation stabilization program and Turkey’s
overall macroeconomic balances linked to the direction of international portfolio capital flows. Turkey
will continue to feel the positive and negative impact of international financial flows and the
authorities should be better prepared to manage the short-lived turbulences in international financial
markets. It remains to be seen, however, whether Turkish economy could sustain its growth by largely
relying on capital flows from abroad.
Following the 2001 crisis, Turkey adopted the flexible exchange rate regime. Since then the Central
Bank of Turkey (CBT) implicitly followed inflation targeting as the main monetary policy framework.
At the beginning of 2006, the Central Bank switched to explicit inflation targeting. Since 2002, the
flow of portfolio capital to emerging markets helped policymakers in the implementation of fiscal and
monetary policy measures. The CPI-based reel effective exchange rate index has increased from 116
at the end of 2001 to 171 at the end of 2005. However, following the Federal Reserve’s decision to
raise the Federal Funds rate further in May 2006, there was a huge flow of international portfolio
capital from emerging markets to industrial countries. The reversal in the direction of portfolio flows
had substantial impact on Turkish financial markets. In real terms, the Turkish Lira lost about 20
percent against the US dollar and the Euro in May and June 2006. As a result, the reel effective
exchange rate dropped to 142. The situation could only be stabilized after a 4.25 percent hike in the
overnight interest rate by the Central Bank in two consecutive meetings to 17.5 percent. In the
meantime, the deprecation of new TL against the Dollar and the Euro fed into inflation. At the end of
2006 inflation rate was 9.7 percent, much higher than the 5 percent target for the year.
Banking sector reforms, the switch to a flexible exchange rate regime and fiscal discipline started
paying off. After six years of tight fiscal policy, the gross public sector debt has been brought down
from its peak of 107 percent of GDP at the end of 2001 to 63 percent at the end of 2006 (Table 1).
8
When one looks at the net public sector debt to GDP ratio, this figure declined to 45 percent by the
end of 2006. Moreover, the banking system went through a period of recovery after injection of US$
40 billion dollars to recapitalize private and public banks. Five years after the 2001 crisis, Turkish
banking system is in much better shape. As a result, following the EU’s December 2004 decision to
initiate membership negotiations with Turkey, multinational banks’ interest in Turkish banks increased
substantially. As of June 2007, the share of foreign capital in the banking system reached 42 percent.
In addition to the banking sector, foreign investors showed keen interest in the privatization of large
public enterprises. The foreign interest in the financial sector and the privatized companies together
resulted in the sudden jump of FDI inflows from less than US$ 1 billion to US$ 9.7 billion in 2005 and
to US$ 20 billion in 2006. The total of FDI inflows continues to be robust in 2007 despite the political
uncertainty. It reached to US$ 10 billion in the first four months of 2007 and is expected to surpass
the US$ 20 billion mark by the end of the year. While the FDI inflows have been playing a critical role
in financing a significant portion of the current account deficit in 2005 through 2007, there are serious
signs that large FDI inflows may not continue beyond 2008. More than 90 percent of FDI inflows that
took place over the last three years have been in the form of mergers and acquisitions and mostly
targeted companies in the service sectors. Turkey is still unable to attract multinational corporations as
an alternative destination for large-scale export-oriented greenfield investments.
Six years after the 2001 economic crisis Turkey is still a vulnerable economy. The country still has
high public and external debt ratios compared to other countries. In addition, despite the flexible
exchange rate regime, the current account deficit increased rapidly over the years reaching 8 percent of
GNP at the end of 2006 (see Figure 8). One reason for the rapid increase in the current account deficit
was the oil price hikes since 2004. Once corrected for the increased crude oil prices the current
account deficit is not very large. If the oil prices stayed around US$ 25 per barrel since 2003, the
current account deficit would have stayed around 4 percent of GNP since 2004. Even if the imports
increase can be accounted for by the oil price hikes since 2003, the actual current account deficit could
have been under control if TL had not appreciated substantially over the period. One of the challenges
for the future will therefore be to adopt policies that would make the economy less vulnerable to
unexpected shocks.
Another important challenge for the future is to insure that the country grows close to its full potential.
Since 2002 Turkey grew by an average of 7 percent per annum (Figure 4). While this is definitely a
commendable achievement at par with East Asian countries, the problem lies with the downward trend
in the growth rates. The growth rate declined to 6 percent per annum in 2006 and is expected to be
below 6 percent in 2007. This downward trend in growth rates clearly shows that Turkey is falling
behind its desired growth rate. Turkey needs to grow above 7 percent per annum (against a population
growth of around 1.5 percent) to close the huge income gap with the EU. In 2005, GDP per capita in
Turkey at purchasing power parity was only 28 percent of the average GDP per capita in EU-25
countries. Without moving its GDP per capita close to 40 percent of that of the average for EU-25 it
will prove very difficult to convince EU public that Turkey’s eventual membership will not bring a big
burden on EU.
Perhaps the surge in unemployment was the only lasting impact of the 2001 crisis. The unemployment
rate increased from 6.5% in 2000 to 8.4% in 2001 before reaching to 10.5% in 2003 (see Figure 9).
Since then Turkish economy created a rather respectable 1.2 million new jobs, but it was not sufficient
to bring the unemployment rate down to pre-crisis levels. As of the end of 2006 unemployment rate
stood at 9.9%. The unemployment rate could not be brought down despite 400,000 new jobs created
every year, because the agriculture has shed 2 million jobs since 2001 and 1.1 million jobs since 2003.
When we include those discouraged workers as part of the unemployed pool, the expanded definition
of unemployment is close to 17% as of the end of 2006 (Figure 9). All these numbers related to the
labor market indicate the urgent need for micro reforms that will reduce the cost of creating new jobs.
9
Finally, inflation is another area where there is much left to be done. At the moment inflation seems
stuck at high single digits. Given the possibility of external as well as domestic economic and political
shocks, it will prove quite difficult to achieve the low inflation target of 4 percent and below.
Challenges facing the Turkish economy have to be confronted by further reforms. Reforms towards
openness and financial soundness have to be combined, with micro “supply-side” reforms. The
structural reforms that had started in earnest after the 2001 economic crisis should be deepened and
widened. Banking sector, social security, regulatory and fiscal reforms that have been implemented so
far need to be further deepened. In addition, those structural reforms that had been delayed so far have
to be implemented as soon as possible. The most important of those reforms is the tax reform and the
containment of shadow economy. Already high indirect and direct tax rates have to be brought down,
while the tax collection has to be enforced vigorously. Implementation of the tax reforms will require
a complete overhaul of the tax administration. Without tax reform it cannot be possible to lower the
cost of labor especially the tax burden on minimum wages. In addition, institutional and political
reforms that will ensure the smooth functioning of the market economy needs to put in place. We will
discuss political reforms in Section 5.
3. Automobile and consumer electronics sectors
3.1. Production
The first automobile assembly plant in Turkey was established in 1960, and a number of joint ventures
entered into the market in the late 1960s. Under the ISI policy, the automobile industry in Turkey
aimed at satisfying domestic demand, and there were almost no exports until the late 1980s. Domestic
production increased until 1976 (total production reached 110 thousands units), but declined until
1980 because of supply shortages. There had been an increase in the number of local suppliers in the
Marmara region during the 1970s. Production increased steadily after the early-1980s, and as a result
of increasing demand and the prospects of the CU with the EU in 1996, the industry expanded its
production capacity in the early 1990s.
The economic crisis in 1994 had a disastrous impact on production that declined by almost 50 percent.
In spite of the 1994 crisis, the industry attracted FDI, and new companies (for example, Toyota in
1994, Honda and Hyundai Assan in 1997) or new plants by existing companies (for example, Ford in
2001) expanded the production capacity. Total production gradually increased after the 1994 crisis and
exceeded its 1993 level in 2000 partly as a result of increasing exports to the EU countries. The 2001
crisis was a step backward in production (up to the 1994 crisis level), but the industry bounced back
very strongly after the crisis, and achieved almost 4-fold increase in output from 2001 to 2006 (or 2fold
increase from 2000, the peak year, to 2006, see Figure 10). Exports played an important role in
increasing automobile production in the post-crisis period (for a detailed history of the automobile
industry in Turkey, see State Planning Organization, 2005).
The Information and Communication Technology (ICT) industries4
in Turkey have been relatively
underdeveloped, and have had a small share in total manufacturing output. Turkey initiated a
comprehensive program to modernize the information and communication infrastructure in the late
1980s. This program led to substantial investment in infrastructure and encouraged a number of
foreign firms to invest in Turkey. Other segments of the industry (computers, components and
consumer electronics) had moderate growth rates in the 1980s. After years of stagnation and even a
slight decline in real output, the consumer electronics sector started to grow very rapidly in the mid-
1990s, and achieved almost 10-fold increase in real output in a decade (from 1995 to 2005, see Figure
4
The ICT industries refer to office, accounting and computing machinery (ISIC 3000), electronic valves, tubes,
etc. (ISIC 3210), TV/radio transmitters; line communication apparatus (ISIC 3220) and TV/radio receivers and
associated goods (ISIC 3230). ISIC codes refer to International Standard Industry Classification, Revision 3.
10
10). Most of the increase in the output of consumer electronics sector in Turkey is accounted by rapid
increase in the production and exports of color television receivers (see Çakır, 2004). While the
industry produced 1.8 million color TV receivers in 1995, production increased to 6.9 million in 1999,
to 12.3 million 2002 and to 20.4 million in 2004. However, the production of color TV receivers
stagnated (20.7 million units) in 2005 before declining to 18 million in 2006.
The decline in the exports and production of color TV receivers in recent years are directly reflected in
the decline of the industry’s production index by 16 percent in 2006. The industry’s fortunes are
further worsened by the fact that it is losing market share not only in the EU market but also in the
domestic market due to rapid switch from tube TV receivers to plasma and LCD TV receivers. In the
first four months of 2007 the industry’s production index further slipped by another 33 percent
compared to the 2006 level.
In the late 1990s5
, the ICT and automobile6
industries accounted for 7.7 percent of manufacturing
value added and 5.4 percent of manufacturing employment (see Table 2). Motor vehicles and
automobile components sectors have the highest share in value added and employment, followed by
TV/radio transmitters and TV/radio receivers (consumer electronics) sectors.
The automobile industry employed about 50000 people in 2000, and there is 60 percent increase in the
number of production workers until 2006.7
When micro- and informal firms and suppliers in other
sectors are taken into account, the automobile industry is undoubtedly one of the leading sectors (in
terms of employment generation and creation of value added) in Turkey.
As shown in Table 2, the ICT industries are unevenly developed in Turkey. Computing machinery and
electronic components sectors are extremely small. These two industries together account for less than
0.2 percent of manufacturing employment and value added. TV/radio transmitters and TV/radio
receivers sectors have a larger share (slightly more than 1 percent of employment and 2 percent of
value added). As will be shown in Section 3.4, TV/radio transmitters sector is dominated by large,
foreign firms that serve only the domestic market. However, the TV/radio receivers sector is wellintegrated
with the world markets and export a substantial part of its output. Therefore, one needs to
focus on consumer electronics (or, more specifically, cathode ray tube color television receivers)
sector in analyzing Turkey’s integration with the world economy in ICT industries.
3.2. Productivity
The pattern of productivity growth in consumer electronics and automobile sectors is very similar to
the pattern of output growth. An analysis of the indices of labor productivity8
reveals that labor
productivity growth rates in consumer electronics and automobiles were at par with that of
manufacturing industry. The automobile industry did not increase its labor productivity to a large
extent from the early 1990s until the early 2000s, a period dominated by the boom and bust cycles.
The negative and detrimental effects of economic crisis on labor productivity in the automobile
5
As of writing the final version of this report, the annual data for manufacturing industries for 2002 were not
available. The data for manufacturing industries exclude private establishments employing less than 10 people,
and those operating in the informal sector.
6
Hereafter, “automobile industries” refer to motor vehicles (ISIC 3410), automobile bodies, trailers and semitrailers
(ISIC 3420), and parts/accessories for automobiles (ISIC 3430).
7
The Annual Survey of Manufacturing Industries conducted by the Turkstat provides basic data for
employment, output and value added at the sectoral level. Unfortunately, the latest year for which the data are
available is 2001. For the post-2001 period, the Short Term Statistics (STS) collected quarterly by the Turkstat
are used to estimate employment and output growth rates. The STS covers only large establishments producing
about 90 percent of sectoral value added.
8
Labor productivity is defined by valued added per employee.
11
industry are seen in 1994, 1999 and 2001. However, labor productivity has recovered rapidly after the
2001 crisis and increased two-fold from 2001 to 2006 (Figure 11).9
The consumer electronic sector had a steady but slow labor productivity growth in the late 1980s and
early 1990s. The increase in total factor productivity (TFP) growth was even lower (cf. figures 11 and
12). Productivity, especially TFP, started to grow rapidly and continuously after 1995, and average
annual growth rate of labor productivity (TFP) reached 15 percent (10 percent) for a decade.
To sum up, the consumer electronics industries achieved above average growth rates in productivity
since the mid-1990s, whereas the automobile industry’s productivity growth performance is almost
equal to the manufacturing average. However, the level of productivity is also important is assessing
industrial performance. The data on labor productivity reveal that there are substantial productivity
differentials between ICT and automobile final producers and other manufacturing industries. Motor
vehicles industry is 86 percent more productive than the manufacturing industry average, whereas
productivity differential reaches 125 percent for TV/radio transmitters and 52 percent for TV/radio
receivers. Components producers for ICT and automobiles (electronic components, automobile bodies
and automobile components) are much less productive (about 53, 59 and 78 percent of the average,
respectively). As a whole, an average worker in the ICT and automobile industries produce 42 percent
more value added than an average worker in the manufacturing industry.
These industries pay higher wages as well. Wages in TV/radio transmitters, TV/radio receivers and
motor vehicles industries are, on average, 181 percent, 47 percent and 83 percent higher than those in
the manufacturing industry. Although automobile components industry is less productive than the
average, its workers are paid more than the average wage rate. As a whole, the ICT and automobile
industries pay 59 percent higher wages. These figures indicate that if these industries continue to grow
faster than the manufacturing average, they would have a positive impact on average productivity and
wages by increasing their share in total manufacturing.
3.3. Exports and vertical integration
The recent performance of consumer electronics and automobile industries is seen most strikingly in
export figures. As a result of the export boom achieved in the early 1980s, the value of manufactured
exports jumped from US$ 2.2 billion in 1980 to US$ 10.3 billion in 1988, and continued to increase at
a lower rate until the mid-1990s (US$ 14.4 billion in 1993). Export growth gained a momentum after
the 1994 and especially 2001 crisis, and the value of manufactured exports reached US$ 30 billion in
2001 and US$ 80 billion in 2006.
The share of consumer electronics and automobiles in total manufactured exports remained low until
the early-1990s (each about 2 percent, see Figure 13). Automobile exports started to increase their
share after the 1994 crisis. Although the value of exports increased at a very rapid rate (average annual
growth rate was about 10 percent from 1993 to 2001, and 20 percent from 2001 to 2006), automobile
exports grew so fast that the share of automobiles in manufactured exports increased from 2.5 percent
in 1993 to 16 percent in 2006. Consumer electronics also achieved a remarkable export performance.
The share of consumer electronics in manufactured exports remained almost constant from 1989 to
1997 (2 percent), but afterwards it continuously increased to 4.6 percent in 2005 before declining to
lower than 4 percent in 2006. ICT exports declined by 26 percent in the first quarter of 2007.10
9
Total factor productivity growth (TFPG) in the automobile industry had a similar trend (see Figure 12). Total
factor productivity for the 1989-2001 period was calculated à la Olley-Pakes. TFPG series were imputed for the
2001-2006 period by using the index for labor productivity.
10
UK’s CTV imports (GTP 8528) increased by 40 percent in 2006 compared to 2005 and reached 3.1 billion
pounds, market share of Turkish CTVs declined from 19.2 percent in 2004 to 14.4 percent in 2006.
12
Export performance of other ICT sub-sectors shows that Turkey could not achieve across the board
growth in the ICT industry. Consumer electronics, or, to be more specific, color television receivers
(CTVs), was the only sub-sector of the ICT industry that achieved considerable level of exports.
Exports of computing machinery, electronic components and TV/radio transmitters remained at very
low levels (US$ 69, 36 and 40 million, respectively, in 2005) whereas exports of TV/radio receivers
were US$ 3074 million in 2005 (it was only US$ 223 million in 1990, see Table 3). The increase in
exports of TV/radio receivers was accompanied by a sharp increase in the value of electronic
components imports (from US$ 429 million in 1990 to US$ 1695 million in 2005). Imports of
TV/radio receivers increased at a lower rate in the same period and Turkey imported US$ 1145
millions of TV/radio receivers in 2005.
Motor vehicle exports reached US$ 7802 million in 2005 and automobile components US$ 2308
million. Imports of motor vehicles and automobile components also increased as a result of integrating
in the global production chains. Turkey imported US$ 8271 million of motor vehicles and US$ 3948
million of automobile components in 2005. Although Turkey is still a net importer of motor vehicles
and components, the ratio of exports to imports increased substantially in the last couple of decades
(from 14 percent in 1990 to 83 percent in 2005). Turkey is expected to be a net exporter in
automobiles in a short time period.
Changes in the direction of foreign trade in ICT products and automobiles provide useful evidence on
the mode of integration into the global economy (see Table 4).11
One third of Turkey’s imports of ICT
components originated from the developing countries in 1990. However, the share of developing
countries in Turkey’s imports of ICT components increased gradually and reached 47 percent in 2005.
A similar trend is observed in ICT products as well. The share of developing countries in Turkey’s
imports of ICT products increased rapidly, from 20 percent in 1990 to 50 percent in 2005. However,
the developed countries have become the main destination for Turkish exports of ICT components and
products. The developed countries received 43 percent of Turkish exports of ICT components in 1990,
and their share increased to 67 percent in 2005.12
They received the vast proportion of Turkey’s
exports of ICT products (that is, TV receivers) in 1990 (90 percent), and they even increased their
share in the last decade (95 percent in 2005). In other words, Turkey increasingly imports electronic
components (and some ICT products such as computing machinery) from the developing countries,
and sells final products made of these components to the developed countries (mainly the EU
countries).
The direction of foreign trade in the case of automobiles is completely different. Turkey imports a
large part of automobile components and automobiles (final products) from developed, mainly EU,
countries (97 percent in 1990 and 90 percent in 2005). A large proportion of Turkey’s exports of
automobile components goes to developed countries (81 percent in 1990 and 74 percent in 2005).
Moreover, the developed countries have increased their share in Turkey’s exports of automobiles,
from 58 percent in 1990 to 87 percent in 2005. In other words, intra-industry trade has become more
important between Turkey and the EU in automobiles and automobile components.13
Turkey both
imports and exports these products at an increasing level to/from the EU, i.e. the Turkish automobile
industry has fully integrated with the European production chains.
A comparative analysis is necessary in order to shed light on the determinants of modes of integration
with the global economy. Table 5 presents the data on relative labor productivity, relative wages and
unit labor cost in Turkey and a selected group of new and former members of the EU, and East
11
Kaminski and Ng (2006) defines automotive and ICT networks as “producer-driven” networks.
12
Note however that the value of Turkish exports of ICT components was very low, only US$ 3 million in 1990,
and US$ 36 million in 2005.
13
Since most of automobile companies in Turkey are owned by multinational companies operating various
manufacturing plants in European countries, a large part of intra-industry trade is indeed intra-firm trade. For
example, Ford Otomotiv imported YTL 2.4 billion worth of inputs (40 percent of sales revenue) from its parent
company in 2005 (the Independent Audit Report for Year 2005 submitted to the Istanbul Stock Exchange).
13
European, Latin American and Asian countries. “Labor productivity” is measured as value added per
employee (measured at current prices and exchange rate) relative to the level in the US. Relative
wages are calculated in the same way. “Unit labor cost” is simply the ratio between wage bill and
value added (divided by the US ratio), and shows the wage cost of producing one unit of value added.
In order to reduce the effects of annual changes, the average values for the last five years for which the
data are available for most of the countries in the sample are calculated.
Relative labor productivity data for TV/radio receivers (ISIC 3230) sector show that Turkey is among
the most productive countries in the sample. The level of productivity in Turkey exceeds all East
European countries and Spain to a significant extent, somewhat higher than or equal to the level in
developed EU countries (the UK and Germany), and higher than those of rapidly industrializing
countries (Korea and Singapore). These results reflect outstanding productivity growth performance of
the Turkish consumer electronics industries in the second half of 1990s. Relative wages in Turkish
TV/radio receivers sector is higher than those in East European and less developed countries, but
substantially lower than those observed in former EU countries (Spain, the UK and Germany) and
rapidly industrializing countries (Korea and Singapore). As a result, the (relative) unit labor cost in
Turkey is lower than all European countries in the sample. There are only a few less developed
countries (Mexico, India and Indonesia) that have lower unit labor cost thanks to their very low wage
levels. Thus, it is not surprising that Turkish CTV producers have rapidly increased their exports to the
EU in the second half of 1990s and early 2000s.
Motor vehicle (ISIC 3410) and automobile components (ISIC 3430) producers in Turkey are
comparably less productive. Their productivity is about 24-27 percent of the US level whereas relative
labor productivity of Turkish TV/radio receiver producers is 51 percent. However, European
producers are also poorly productive: Slovakia, Poland and Spain are less productive than Turkish
producers, whereas the productivity differential between Turkish producers on the one hand, and
British and German producers on the other hand is not substantial. Wages in Turkish automobile
industry seem to be higher than those in Hungary, Slovakia and Poland, and much lower than those in
Spain, the UK and Germany. As a result, the unit labor cost is lower in Turkey than in European
countries (with the exception of Hungary in our sample). Turkey has a cost disadvantage against most
of less developed and rapidly industrializing countries (Mexico, Indonesia, Korea, Malaysia and
Philippines). Apparently, Turkey has been increasing its automobile exports to the EU thanks to its
low unit labor costs relative to European producers, and geographical proximity to main markets (and
suppliers) that provides cost and delivery advantages vis-à-vis distant low cost producers. Although
automobile component producers in Turkey seem to be slightly more competitive than automobile
producers, the patterns of productivity and wages is these two sectors are quite similar.
The data on the degree of vertical integration and product quality would provide additional
information on the structure of production and the mode of integration to the global economy. The
ratio between value added and output value can be used as a first order approximation for the degree
of vertical integration in a given industry. If an industry does not purchase any input from other
industries, i.e. if the firms in that industry are fully vertically integrated, then the value added/output
ratio will be equal to one. In the other extreme case where the industry purchases almost all inputs
from others, than it will not produce much value added, and the ratio will get closer to zero.
The data on the value added/output show that Turkish industries are more vertically integrated
especially relative to their counterparts in East European countries (see Table 6). For example, in the
case of TV/radio receivers sector, the US has the highest ratio of value added/output ratio (0.35),
followed by Germany, Japan, Korea, Brazil and Turkey (all have value added/output ratio in the range
of 0.29-0.32. It seems that the degree of vertical integration among Spanish and East European
TV/radio receiver producers is quite low (the corresponding rates are 0.18 for Poland, 0.16 for Spain,
0.15 for Slovakia, 0.14 for Czech Republic and 0.10 for Hungary). There is a similar pattern in
automobile and automobile components industries where Turkish producers have relatively higher
value added/output ratios whereas East European producers have quite low shares of value added.
There seems to be a significant difference in modes of integration to the global economy for Turkish
14
and East European producers. It is likely that Turkish producers undertake more manufacturing
operations whereas East European producers tend to rely on outsourcing components, or they
themselves operate as subcontractors.
The average unit price of exported and imported passenger cars could be used as a measure of product
“quality”. The average (fob) unit price of passenger car exports was about 10-20 percent lower than
the average (cif) unit price of imports in 1999 and 2000. The economic crisis 2001 and the devaluation
of Turkish Lira in the same year led to a decline in export price and an increase in import price
(denominated in US$). The average unit price of exports tended to increase gradually, from US$ 7600
in 2002 to US$ 12800 in 2005, whereas the average unit price of imports remained almost the same
(US$ 13200 in 2002 and 2005, see State Planning Organization, 2005: 24-27). Thus, the price
differential between export and import prices declined up to 3-5 percent. Considering the cost of
insurance and freight, one may conclude that there is not substantial quality difference, on average,
between imported and exported passenger cars in Turkey.14
The improvement in the quality of products necessitates substantial investment in process renewal and
new model development. According to Automotive Manufacturers Association, the industry has
invested on average about US$ 400 million per year in the last decade. The level of investment
declined considerably after the crises in 1994 and 2001, and reached the peak in 2006 (about US$ 1
billion). The structure of investment has also changed in this process. Although the share of new
model development investment was about 20 percent of total investment in the 1990s, it increased to
40-50 percent since 2000 to satisfy the demand for higher quality and more diversity.
3.4. Firm-level performance
The automobile and consumer electronics industries have undergone through a process of
transformation in the last decade, and the outcome of this process has been observed in recent year.
Since the official data at the sectoral for recent years are not available, firm-level data could provide
additional information about this transformation process.
The Automotive Manufacturers’ Association (OSD in Turkish acronyms) is the main umbrella
organization for the automobile producers in Turkey. All major producers are members of the
organization. There are six passenger car producers that have all foreign participation (four of them are
majority foreign-owned).15
There are 12 other companies (5 of them foreign owned) that produce
trucks, pickups, buses, minibuses and road and farm tractors. Thus, the automobile industry is
dominated by foreign companies.
There are a large number of suppliers located mainly in the Marmara region. The Association of
Automotive Parts and Components Manufacturers (TAYSAD) has 236 members.16
There are 40
foreign companies (most of them are majority owned) among TAYSAD members.17
There are three major producers in the consumer electronics sector: Arçelik, Beko and Vestel (Arçelik
and Beko are subsidiaries of Koç Holding). Beko and Vestel are specialized in consumer electronics
whereas Arçelik produces other consumer durables (the so-called “white-goods” and household
appliances) as well. None of these companies have any foreign participation (for more information,
see Ozcivelek and Zontul, 2004).
14
The average unit price is determined by a large extent the composition of imports/exports. We implicitly
assume that there is not much change in the composition of imports/export during the period under investigation.
15
Home countries: EU (2), Japan (2), US (1) and Korea (1).
16
The member companies of OSD and TAYSAD employed 42700 and 66000 people, respectively, in 2006 (see
the organizations web sites, www.osd.org.tr and www.taysad.org.tr).
17
Home countries: EU (25), Japan (8), US (5), Korea (2) and Switzerland (1).
15
A number of automobile and consumer electronic companies are listed on the Istanbul Stock Exchange
(see Table 7). The financial statements of these companies are audited by independent auditors and are
publicly available. Thus, the financial data on listed companies could be used to shed light on recent
changes in these sectors.
The data on net real sales of these companies show that especially large final producers (Ford and
Tofaş in automobiles and Arçelik, Vestel and Beko in consumer electronics) achieved rapid growth in
sales after the 2001 crisis, and reached the peak sales level in 2004-2005 (see Figure 14).18
The
increase in sales is striking given almost a decade long stable level of sales of those companies until
2001.
The export data at the firm level reveal that automobile and consumer durables producers reacted
swiftly to the 2001 crisis and the devaluation of the Turkish Lira, and increased export rates
substantially in 2001 and 2002 (see Figure 15). The Turkish Lira appreciated rather rapidly after the
crisis until 2006, so much that it was (in real terms) 25 percent overvalued in 2005 than its level in
2000. In spite of the appreciation of the Turkish Lira, the automobile and consumer electronic
producers were successful in keeping their export rates at a level higher than the pre-crisis level.
Apparently, their export intensity has reached and remained at a higher plateau after the crisis.
The automobile and consumer electronic producers experienced, on average, declining profitability19
in the second half of 1990s that hit the bottom during the 2001 crisis. They gradually recovered after
the crisis and the profit margin increased to positive numbers in 2002 and 2003. However, the profit
margin remained at low levels (3-4 percent, weighted average) in the last few years without any sign
of recovery (see Figure 16).
The profitability data show that they operate on thin margins, and their sales revenues are only slightly
more than their expenses. In other words, price competition seems to be very important especially for
consumer electronics producers.
4. Macroeconomic policies and sectoral performance
4.1. Automobiles
The automobile industry in Turkey has proved to be a vibrant and growing sector, and achieved an
outstanding export performance in the last decade in spite of the macroeconomic problems that
plagued the country. What are the main factors behind its performance?
The automobile industry is well integrated within international production chains. From its inception
in the 1960s and 1970s until the late 1990s, foreign firms, either through joint ventures with major
domestic business groups, or through wholly owned subsidiaries, have been dominant in the industry.
Although these companies were oriented towards the domestic market until the early 1990s, they were
able to seize new market opportunities opened by the CU with the EU in 1996. New foreign
companies entering the Turkish market in the second half of 1990s have targeted the EU market as
well. These companies have strong links with their subsidiaries in the EU, and intra-firm trade has
apparently played an important role in producing automobiles in Turkey and marketing them in the
EU countries.
Although the automobile industry is well integrated within international, or, more specifically,
European production chains, it has also benefited to a large extent from the existence of a strong
domestic industrial and supplier base. The automobile parts and components sector has developed to
18
Sales values are deflated by GDP deflator.
19
Profit margin is measured as new profit after taxes/net sales ratio.
16
some extent in the 1970s and 1980s, and has attracted foreign investment in the 1990s. Strong and
responsive supplier-producers links have enabled automobile producers to expand their capacity and
output rapidly after the 2001 crisis (for a comprehensive analysis of supplier-buyer links, see Wasti,
Kozan and Kuman, 2006).
The automobile industry in Turkey would not be successful had it failed to adopt itself on time to new
conditions imposed by the CU. The Automotive Manufacturers’ Association (OSD) played an
instrumental role in anticipating new challenges, and orchestrating a common course of action to face
these challenges. The OSD regarded the CU as an inescapable fact, and considered it as an opportunity
in the early 1990s.20
The first challenge was to adopt massive EU rules and regulations affecting the
industry. The process of discovering, understanding and transposing EU rules and regulations proved
to be useful in enhancing the competence of technical personnel employed by automobile producers
(and government officials). After achieving a certain level of technological sophistication necessary to
satisfy the EU rules, the technical personnel pushed forward to improve quality and to introduce new
designs (especially in the commercial vehicles segment) to be more competitive in the EU market.
“Research and development” has become a catchword in the late 1990s. Current plans drawn by
industry representatives are targeting to make Turkey the third largest producer in Europe by 2013 as
well as making it a center for design and research and development.21
4.2. Consumer electronics
The structure of consumer electronics industry in Turkey is rather different than that of the automobile
industry. The consumer electronics industry is dominated by a few large domestic firms that are not
supported by any domestic supplier base. Consumer electronics producers use mostly commercial-onthe-shelf
components and import them from the East Asian and the European countries. Domestic
electronics components sector is almost non-existent, and consumer electronics producers could
outsource only some non-electronic parts to local suppliers.
The components used by consumer electronic producers (the most important being cathode ray tube
(CRT) for color television receivers) and their final products are commodity-like products sold in
almost perfectly competitive markets where profit margins are razor-thin. Moreover, CRT televisions
are perceived as “old technology” and are being displaced by new technology like LCD and plasma
televisions in developed country markets.
Under those adverse conditions, the Turkish consumer electronics producers have become quite
successful in the EU market by producing the right product at the right time in the right place.
In the 1990s, the CRT television technology was a mature technology and new LCD and plasma
technologies were expected to displace rapidly CRTs especially in the large-screen television market.
This means that technological entry barriers were low in this market segment in which the European
producers would be forced to exit. In other words, CRT color television segment, which was still the
largest television segment during the 1990s in the EU, was the only segment where new producers
would enter. Turkish TV producers have become competitive in this market thanks to their flexibility
and low-cost manufacturing.
When Turkish producers entered into the EU market en masse in the mid 1990s (right after the CU in
1996), Asian producers had low production costs and were competitive against Turkish producers that
rapidly increased their productivity over time. However, that was the right time to enter the EU market
20
During the negotiation stage of the CU agreement, the automobile industry was expected to be one of the
industries that would face the toughest competition from the EU-origin imports. Thanks to the forceful lobbying
by the industry, the full liberalization of the automotive imports was phased out over a period of five years.
21
Interview with Ercan Tezer, the Secretary General of the Automotive Manufacturers’ Association, 4 April
2007.
17
because the European Commission imposed provisional (1994) and definitive (1995 and 2002) antidumping
duties on color television receivers originated in China, Malaysia, Korea, Thailand and
Singapore.22
The European Commission initiated anti-dumping proceedings concerning imports of
color television receivers originated in or exported from Turkey in 1992 and 2000, and decided not to
impose any duty.23
Since Vestel imported CRTs from China and Beko from Malaysia and Korea, they were subject to
anti-dumping duties for their products originating from these countries, but the level of duties were
much lower. For example, the rate of duty, set in the Council Regulation 1531/2002 that is still in
force, is 44.6 percent for Chinese producers whereas it is only 24.5 percent for Vestel for CTVs
assembled in Turkey that incorporate CRTs originated from China.24
The time was right because, as a
member of Vestel’s Executive Committee admits, “[i]n 2000, if the additional taxes to Chinese
producers were lifted, it would have been devastating for Vestel” (Cengiz Ultav, quoted by Karabati
and Tan, 2005: 7). Vestel has become one of the largest and competitive producers in Europe thanks to
its investment in flexible automated process technologies, effective global sourcing and quick
response and point delivery capabilities. In other words, the EU anti-dumping measures imposed on
low cost Asian producers provided Turkish producers a kind of “infant industry protection” at a
crucial time.
Finally, Turkish consumer electronics producers were located in the right place. They were located
very close to a major market (the EU), and employed skilled labor at low cost. Geographical proximity
provides cost advantages over Asian producers but more importantly it helps to reduce the delivery
time. According to Karabatı and Tan (2005), the delivery time to Europe is 1-2 weeks for Vestel, but
1.5-2 months for Chinese producers. Thus, Turkish producers can operate on a “made-to-order” basis,
thanks to their short delivery time and manufacturing flexibility.
4.3. Lessons learned
The history of Turkish economy over the last quarter century leaves no ground to doubt the failure of
successive governments in putting the country on a sustained growth path. Despite policies that
ascertain the external adjustment, governments failed to undertake the right mix of policies to insure
internal adjustment and create an economic environment that is conducive for long-term investment.
The performances of the automotive and consumer electronics industries, on the other hand, showed
that despite the macroeconomic policies that inhibit investment and growth, it is possible that certain
sectors perform well enough to integrate with the world markets.
22
The anti-dumping measures in place regarding imports originating in Singapore remained in force until 2002.
For anti-dumping measures, see the Council Regulations 2376/94, 710/95 (amended by 2584/98) and 1531/2002
(amended by 511/2006).
23
See the European Commission Decisions 95/92 and 2001/725. The Commission’s decision is based on the
following argument: “… the origin of the cathode-ray colour television picture tube (CPT) virtually determined
the origin of the CTVs, since the ex-works price of the CPT represented, in all cases, at least 35 percent of the
ex-works price of the CTV. It should be noted that there is no production of CPTs in Turkey and, therefore, all
CPTs are imported. Accordingly, the origin examination concluded that the CTVs exported from Turkey during
the [investigation period] were of the origins of the CPTs used in their assembly.” (the Commission Decision of
28 September 2001, 2002/725).
24
The rates are 25.1 percent for Malaysia and 15.0 percent for Korea whereas Beko is subject to 18.2 percent
and 12.3 duty for its CTVs originating from Malaysia and Korea, respectively. The European Commission
accepted a joint undertaking with seven Chinese TV producers and the China Chamber of Commerce for Import
and Export of Machinery and Electronics Products and waived the anti-dumping duty on these companies on the
condition that they would respect minimum import price levels and quantitative ceilings established by the
Commission. The European Commission has re-imposed a 44.6 percent anti-dumping duty on imports of
Chinese color television receivers after one of the companies refused to allow the EU to inspect its premises in
2006 (see the Commission Decision 2006/258 and the Council Regulation 511/2006).
18
The successes of the automotive and consumer electronics industries have much to do with the
increased competition after the CU rather than the macroeconomic policies. Had Turkey adopted the
correct mix of macroeconomic and sectoral policies in the 1990s, the majority of the other
manufacturing sectors would have realized structural transformations similar to that of the two sectors.
As a result, Turkey could have undertaken the transformation from a low-middle income economy
with competitive advantage in labor-intensive sectors, into a high-middle income country with
increased focus on technology intensive sectors.
The statistics on FDI inflows during the 1990s provide additional evidence about the role of the
macroeconomic environment. Throughout the 1990s, a period during which the global capital flows
increased exponentially, Turkey was not able to increase the FDI inflows. The average annual FDI
inflows during the 1990-2004 period was below US$ 1 billion. Not only had the multinational
corporations (MNCs) decided not to invest in new projects in Turkey, during this period of uncertainty
some had decided to close their factories in the country. While Turkey was tied up with its domestic
political and economic uncertainty, its competitors in the developing world and among the transition
economies adopted right mix of policies that enabled them to attract billions of dollars in foreign direct
investment every year. To give an example, throughout the late 1990s and early 2000s, Czech
Republic, Hungary and Poland were able to attract close to US$ 5 billion a year, whereas Turkey
attracted less than US$ 1 billion a year.
While both sectors have done quite well after the CU, the impact of the CU on the two industries
differed substantially. In the case of consumer electronics, the CU enabled Turkish consumer
electronics companies multiply their exports to the EU market in a very short period, confronting with
very little barriers to entry. Instead, their ability to capture a large share of the EU market for the
traditional CRT TVs owe a lot to the trade barriers that EU raised towards imports from the East Asia
and China. In order to achieve this, companies in the consumer electronics did not have to undertake
large investment expenditures on new product development and innovation. Instead, they needed to
create huge capacities to take advantage of scale economies in the production of especially CRT color
TVs.
This rather easy transformation of the sector into an export sector with close to 50 percent of the EU
market for CRT TVs also prepared the ground for future threats. Without investing in new technology
the sector was excessively dependent on component imports, from East Asian countries. The valueadded
at the firm level was quite small even in CRT TVs and further declined with the widespread
commercialization of plasma and LCD televisions, and profitability eroded tremendously.
After years of upward trend, the sector’s output declined by 16 percent in 2006 and by 33 percent in
the first four months of 2007. Without developing its own technology and increasing its value-added,
the industry will be facing more serious threats in the coming years. Turkey’s geographical proximity
to Europe used to be viewed as an important advantage for Turkish ICT producers against the East
Asian and Chinese producers. In order to overcome this advantage, in recent years the East Asian and
Chinese producers founded LCD and plasma TV factories in the Central and Eastern European
countries.
In the early 1990s, there was no reason for the automotive industry to change the way it ran the
business. However, in the last 10 years following the formation of the CU between Turkey and the
EU, Turkish auto industry went through a serious transformation. The seeds of change came about in
the early 1990s. The trade liberalization throughout late 1980s and early 1990s reduced the protection
against the motor vehicle imports. However, as the domestic demand was booming, the industry was
performing quite well in terms of profits. Import penetration was rather small thanks to the quite stillhigh
protection rates.
The most significant step in the transformation was the formation of the CU between Turkey and the
EU. During the negotiation stage of the CU agreement, Turkish auto industry was expected to be one
of the industries that would have faced the toughest competition from the EU-origin imports. The
19
industry did not hide this fact and forcefully lobbied the government to gain as much time as possible
in order to postpone the full impact of the CU to take place. As a result, auto industry was one of the
sensitive sectors that Turkish side wanted to include in the agreement. The full liberalization in the
auto imports were phased out over a period of five years. Imports of used cars are still prohibited and
likely to continue that way in the near future.
Immediately before the formation of the CU in 1996 there was no serious change in the policies of the
industry to prepare for competition. However, despite the clauses that were incorporated in the CU
agreement to slow down the penetration of imports, imports surged very rapidly, accounting half of
the market by 1998. As the domestic producers observe the realization of the threat from imports they
started to make investments that would keep them competitive. However, these efforts could not bear
fruits until late 1999 and 2000. Until then Turkish auto industry continued to rely mostly on the
domestic market. As a result of the investment projects the producers started to develop models that
can compete with foreign built cars domestically and internationally.
A natural implication of Turkey’s large domestic market is the presence of FDI directed largely
towards the domestic market. However, the automotive sector is a good example of how an initially
protected home market can be transformed into a competitive and increasingly export-oriented
industry through FDI inflows coupled with the availability of low cost highly qualified labor force.
During the debate on the CU, the automotive sector was expected to be the worst affected from
lowering protection on EU imports. However, that prediction was proven wrong. Over the last couple
of years, the automotive sector has become the second largest export sector.
By the mid-1990s, there were four MNCs with a sizeable market share and more than 20 years of
experience in the Turkish automotive industry (Fiat, Ford, Mercedes Benz and Renault).25
In the mid-
1990s with the increasing prospects of a CU agreement with the EU, Japanese and Korean companies
(Honda, Hyundai and Toyota) started investing in Turkey in joint-ventures with Turkish industrialists
or, as in the case of Isuzu, expanded already existing joint venture.26
Perhaps because of the uncertain
business environment in Turkey, these companies did not make substantial investments initially and
built plants with small production (in the vicinity of 10-20,000 units per year) capacities. Once the CU
with the EU went into effect in 1996, domestic market had gradually opened up to competition from
the EU. Actually, in the first couple of years of the CU, the sector struggled with wild fluctuations in
domestic demand as well as competition from imports. The contagion from the Russian crisis of 1998
and the Marmara earthquake of 1999 effectively hit the demand in the auto market.
However, there was a lot at stake. There was already a substantial production capacity coupled with a
competitive parts and accessories industry. In addition, domestic business establishments with years of
experience in the automotive industry and cheap but good quality labor induced MNCs in the
automotive sector to increase their investments in Turkey and built new capacity to produce motor
vehicles for the European market. None of the MNCs mentioned above decided to close down their
plants in Turkey.27
In the meantime, the auto parts and accessories industry has also been attracting foreign investors.
Most of the world leaders of the sector have joint ventures with Turkish partners. Some of them are
big suppliers like Robert Bosch, Valeo, Delphi Packard and Mannesmann Sachs. Altogether, between
1992 and 2000, the automotive industry realised a total of US$ 3.4 billion worth of investment. Of this
25
Obviously there were other producers active in the domestic market. The listed four had the largest market
shares in the automotive industry.
26
Of these four MNCs, Toyota and Honda decided to become the sole owners of their production units (and
Hyundai increased its shares to 70%) once they decided to target their production towards the European market
rather than the domestic market. This fact can be taken as an example of the difficulty that foreign investors
would face while entering the domestic market without an insider on board.
27
Only Opel closed down its small plant near Izmir that used to undertake the montage of some of its car
models.
20
amount, US$ 750 million was used for capacity development, US$ 976 million for new model
development, US$ 497 million for modernization, US$ 300 million for localisation and US$ 195
million for quality improvement. Moreover, due to new investment projects towards the production of
new models, in recent years this investment amount increased by close to US$ 1 billion.
5. Sustainability, institutional change and political economy of reform
Even though Turkey had traveled a long distance since 2001 crisis, Turkish economy still faces
challenges ahead. These challenges, mostly driven by domestic institutional factors, cannot be
resolved on a permanent basis without reforming the country’s political system, shaped by the current
legal and institutional framework. The 1982 Constitution, the election law and, most importantly, the
political party law passed by the military regime created an environment that gives the political elites
autonomy from the scrutiny of the rank-and-file of the political parties and the society at large.
Political parties are the cornerstones of the western-style democracies, through which political
competition is realized. Unfortunately in the Turkish political system that is not the case. The current
political party and election laws were not designed to promote the fundamental principles of political
democratic competition.
Since its foundation in 1923, Turkish Republic can be described as a strong state which was founded
and governed by military/civil bureaucratic elites. With the 1946 elections the country made its
transition to democracy. In the 1950 elections, the power was transferred away from the Republican
People’s Party (RPP) that represented the single-handed rule of the bureaucratic elites. As the
democracy has taken a hold in the country, political elites started to share power with the bureaucratic
elites. Since then, three military interventions and the ensuing changes in the constitutional and
institutional framework have laid the foundations for the current political system.
The current political party and election laws raise significant barriers to entry and exit, and hence
curtail democratic competition within the political parties. Under the current political system each
political party is captured by a group of political elites who are involved in patronage, populism, and
rent seeking activities at different stages. Once a group of political elites capture the party leadership,
they would never leave it to the opposition in the party.
As the current political party law gives all the power to the party headquarters, the party leadership
controls the channels through which the party rank and file can replace the leaders/cadres that are
proven to be unsuccessful. As a result, the rank and file has very little chance of removing the
leadership. The party headquarters can annul the local chambers of the party and appoint new
members. Consequently, in the Turkish political system, political parties have become nothing more
than the fiefdoms of the party leaders and their immediate supporters. Those members who oppose the
leadership, on the other hand, stand no chance of survival within the party. As the political party law
does not allow the democratic channels to function properly, those members that are thrown out of
their parties can continue their political activities in another party or form a new party themselves.
There are several implications of this political system. First is the emergence of a plethora of political
parties in the political landscape of the country. Another is the short life-span of the political parties.
A third implication is the turnover of the parties in the government. All these phenomena could be
observed in the Turkish case over the last two decades.
The election performances of the two center-right parties, the Motherland Party (MP) and the True
Path Party (TPP), over the last 20 years provide excellent examples of what the hegemony of political
elites imply for the parties in the long-run. The two parties together received 55 percent of the popular
vote in the 1987 general elections, which was the first democratic election after the 1980 coup d’etat
(see Figure 17). However, both parties were controlled by the political elites so much that until the
21
November 2002 general elections they were unable to replace their unsuccessful leaders Tansu Çiller
and Mesut Yılmaz. Yılmaz stayed as the Chairman of the MP for more than 11 years, whereas Çiller
was at the helm of the TPP for more than 9 years. Yılmaz lost five elections in a row, including the
local elections, whereas Çiller lost four elections. With Yılmaz at the helm, MP’s votes declined from
24 percent in 1991 elections to 5.1 percent in 2002 elections. Similarly, TPP’s votes declined from 27
percent in 1991 to 9.6 percent in 2002 under Çiller’s leadership (see Figure 17)
Despite their proven track records, neither of the parties was able to replace their respective leaders.
The political party law enabled the leaders of both parties to exert complete control over the whole
party organization. Offering no real solution to major economic problems and unable to change the
unsuccessful and unpopular leaders, popular support for the center-right and center-left parties eroded
over time; so much that in the November 2002 general elections, the two parties could gather a total of
only 14 percent of the popular vote. Thanks to the undemocratic threshold imposed by the election
law, neither party was able to gain seats in the Parliament. Leaders were finally pushed out, but only
after this major defeat. It was rather too late. The two center-right parties that once appealed to more
than half of the voters are currently facing extinction.
Another important consequence of the existing barriers of entry and exit in party politics is the
proliferation of political parties and fragmentation of politics. While only 7 political parties
participated in the 1987 general elections, the number of parties that participated in the nationwide
campaign increased to 21 in the April 1999 elections and 18 in the November 2002 elections.
While the popular support base of the center parties had eroded, there was the struggle to capture
center right and left votes. Tired of the past failures and increased burden of living in every next
election, voters used their only democratic right to try a party which was not tested before. In the 1999
and 2002 general elections, marginal parties received the bulk of the votes and form the government.
Once they became coalition partners, these parties declared repeatedly that they were actually parties
closer to the center. Since the constraints and incentives faced by the political elites controlling these
parties have been no different from the ones that ruled the country before, they had no reason to stop
the process of the transfer of government resources and start delivering solutions to the age-old
problems of the country.
Going into the November 2002 elections, all major political parties on the right and the left had taken
part in government before. The current ruling Justice and Development Party (JDP), an offspring of
the Welfare Party, which was founded only fifteen months before the elections, was the only untested
party that completed its country-wide grass-roots organization. Being the only untested party with a
potential appeal to the electorate JDP made good use of this golden opportunity.
Before the 2002 elections, JDP leaders were proudly announcing how democratic their party by-laws
were; ensuring the active participation of the party rank-and -file in the decision making process.
However, just three months into the power, the Founders’ Council of JDP decided to change several
articles in the party by-laws to make sure that the chairman and the party headquarters are in complete
control of the party. JDP now is no different from other parties that it had so vehemently opposed and
criticized before the elections for lacking the within-party democracy. They are also not much
different from the political elites in other parties in terms of making good use of the government
resources to support the businesses close to the party. There has been numerous news reports over the
last four and a half years that document the JDP’s decisions to grant major local and central
government investment projects to the businesspeople linked with the party.
The incentives and constraints the political elites are facing have to be rewritten in order to ensure that
both political and economic stability will be sustained in the medium to long-term. Unless the current
political party and election laws are scrapped and new laws that promote and guarantee the withinparty
political competition are adopted, the political elites will behave no different from their
predecessors. Without these major political reforms, the 2007 general elections will not insure that
micro “supply-side” economic reforms that have been delayed for so long will be implemented in the
22
near future and the growth performance of the last five years will be sustained in the long-run. To the
contrary, as the political opposition in the parliament will become stronger after the July 22 elections,
the JDP may turn to more populist policies; just as the “reformer” Turgut Özal favored populist
policies when his party faced strong opposition in the parliament after the 1987 general elections.
The upshot of this short political economy analysis is that Turkey will not be able escape the populist
cycles of the 1990s and the ensuing economic crises unless it reforms the current political party law,
the Constitution, and the election law to establish public control over the political elites.
Unfortunately, it is not in the interests of political elites to undertake these reforms.
23
6. Conclusions
The development of the Turkish economy since the early 1990s shows that despite the macroeconomic
policies and conditions that inhibit investment and growth, it is possible that certain industries could
perform very well, and play a very important role in generating employment and fostering growth.
Automobile and consumer electronics industries are certainly among the most successful industries in
Turkey in the last decade. These two industries have achieved remarkable output and productivity
growth rates, and have been very competitive in international markets.
Although there are similarities between the performance of automobile and consumer electronic
industries, there seems to be significant differences between their structures, links with domestic
suppliers, technological orientation and modes of integration with the global economy. The
automobile industry is dominated by multinational companies, has a strong domestic supplier base,
and has seized the opportunities opened up by the Customs Union by investing in new product and
process technology and learning. Both the final product and supplier segments of the industry are wellorganized,
and have established a shared vision of the future through organized dialogue within the
industry and with the public sector as well. Industrial leadership, coordination and cooperation have
been vitally important for the success of the automobile industry.
The consumer electronics industry is dominated by a few, large domestic firms, and has become
competitive in the European market thanks to its geographical proximity, productive domestic labor,
and focus on a protected and technologically mature niche market, namely CRT color television
receivers. The market is almost characterized by “perfect competition” with very low profit rates, and
is being replaced by technologically advanced substitutes, large-screen LCD and plasma televisions.
Therefore, the Turkish consumer electronics industry is under severe competitive pressures, and could
revitalize itself only if it could achieve a breakthrough in its technological orientation.
Our analysis suggests that neither the automobile nor the consumer electronics industries have directly
benefited from macroeconomic policies in the 1990s. On the contrary, they have been successful under
adverse macroeconomic conditions, and have increased their exports after crises, especially after the
worst one in 2001. However, this does not mean that macroeconomic policies and environment are
inconsequential for industrial performance and economic growth. It is without doubt that these
industries, together with others that did not perform so well in the last decade, could perform even
better had governments in Turkey adopted more responsive macroeconomic policies. However,
governments would be more responsive only if far-reaching political reforms are undertaken by
changing the Constitution, and current political party and election laws in order to establish public
control over the political elites.
24
References
Çakır, E. (2004), Televizyon Sektör Raporu [Report on the Television Industry], Istanbul:
Istanbul Chamber of Commerce (ITO).
Kaminski, B. and Ng, F. (2006), Turkey’s Evolving Trade Integration into Pan-European
Markets, World Bank Policy Research Working Paper 3908, Washington DC.
Karabati, S. and Tan, B. (2005), Vestel Electronics: Transition to the Leading European TV
Manufacturer, Graduate School of Business Case Study No. 605-015-1, Istanbul: Koc
University.
Ozcivelek, R. and Zontul, H. (2004), Insights into the ICT Industry in Turkey, Technical
Report EUR 21392 EN, Seville: IPTS.
State Planning Organization (2005), Dokuzuncu Kalkınma Planı (2007-2013) Otomotiv
Sanayii Özel İhtisas Komisyonu Raporu, Ankara: SPO.
Şenses, F. and Taymaz, E. (2003), “Unutulan Bir Toplumsal Amaç: Sanayileşme Ne Oluyor?
Ne Olmalı?” [A Social Objective Forgotten: Industrialization], in A.H. Köse, F. Şenses
and E. Yeldan (eds.), İktisadi Kalkınma Kriz ve İstikrar, İstanbul: İletişim.
Taymaz, E. and Yılmaz, K. (2007), “Productivity and Trade Orientation: Turkish
Manufacturing Industry Before and After the Customs Union”, Journal of International
Trade and Diplomacy (1): 127-154.
Togan, S. (1997), “Opening up the Turkish Economy in the Context of the Customs Union
with EU”, Journal of Economic Integration (12):157-179.
Wasti, S.N., Kozan, M.K. and Kuman, A. (2006), “Buyer-Supplier Relationships in the
Turkish Automotive Industry”, International Journal of Operations and Production
Management (26): 947-970.
Yılmaz, K. (2007), Türkiye için Doğrudan Yabancı Yatırım Stratejisi’ne Doğru [Towards a
Foreign Direct Investment Strategy for Turkey], Istanbul: International Investors
Association of Turkey (YASED).
25
Figures and Tables28
Figure 1. GDP per capita growth rates in Turkey, 1926-2005
(5-year moving averages)
Figure 2. Consumer price inflation - 1980-2006 (end of year)
0
20
40
60
80
100
120
140
1980 1985 1990 1995 2000 2005
%
28
Unless otherwise stated, Turkstat is the source of all data presented in tables and figures.
-5
0
5
10
1925 1935 1945 1955 1965 1975 1985 1995 2005
Annualgrowthrate(%)
26
Figure 3. Real Effective Exchange Rate (CPI-based, annual average)
80
90
100
110
120
130
140
150
160
170
1980 1985 1990 1995 2000 2005
indexvalue(1995=100)
Figure 4. GDP Growth rates (1980-2006, percent per annum)
-8
-6
-4
-2
0
2
4
6
8
10
1980 1985 1990 1995 2000 2005
27
Figure 5. Growth rates of manufacturing exports and imports (1981-2006)
Figure 6. Budget balance in Turkey (1990-2006, percent of GDP)
Source: Undersecretariat of Treasury
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1980 1985 1990 1995 2000 2005
Annualgrowthrates(%)
Exports Imports
-20
-15
-10
-5
0
5
10
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
%ofGDP
Overall balance Primary Balance
28
Figure 7. Nominal and real rates of interest (January 2002-May 2007)
0
10
20
30
40
50
60
70
80
01-02 07-02 01-03 07-03 01-04 07-04 01-05 07-05 01-06 07-06 01-07
percent
Nominal rate Real rate
Source: Undersecretariat of Treasury
Figure 8. Current account deficit (CAD) and the impact of the recent hike in oil prices
-4
-2
0
2
4
6
8
10
1990 1993 1996 1999 2002 2005
%ofGNP
CAD - actual oil prices CAD - oil price at $25pb
29
Figure 9. Labor force and employment (2000-2006)
(in thousands) 2000 2001 2002 2003 2004 2005 2006
Non-institutional population 66,187 67,296 68,393 69,479 70,556 71,611 72,606
Population 15 years of age and over 46,211 47,158 48,041 48,913 49,906 50,826 51,668
Labor Force 23,078 23,491 23,818 23,641 24,290 24,565 24,776
Employed 21,581 21,524 21,354 21,148 21,792 22,046 22,330
Agriculture 7,769 8,089 7,458 7,165 7,400 6,493 6,088
Industry 3,810 3,774 3,954 3,846 3,987 4,284 4,407
Construction 1,364 1,110 958 965 1,030 1,173 1,267
Service 8,637 8,551 8,984 9,171 9,374 10,096 10,569
Unemployed 1,497 1,967 2,464 2,493 2,498 2,520 2,446
Outside the labor force 23,133 23,667 24,223 25,272 25,616 26,261 26,892
Readyfor work, not looking for job (1) 1,139 1,060 1,020 945 1,223 1,714 2,087
Labor force participation rate (%) 49.9 49.8 49.6 48.3 48.7 48.3 48.0
Unemployment rate (%) 6.5 8.4 10.3 10.5 10.3 10.3 9.9
Non-farm unemployment rate (%) 9.3 12.4 14.5 13.8 14.3 13.6 12.6
Expanded unemployment rate (%) 10.9 12.3 14.0 14.0 14.6 16.1 16.9
Source: TURKSTAT
1) Including those looked for a job but couldn't find in the last 3 months.
5
6
7
8
9
10
11
12
13
2000q1 2001q1 2002q1 2003q1 2004q1 2005q1 2006q1 2007q1
unemployment rate 4-quarter moving average
Figure 10. Industrial production index (1980-2006, 1997=100)
0
50
100
150
200
250
300
350
400
450
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006
Manufacturing Cons. electronics A utomotive
30
Figure 11. Industrial labor productivity index (1980-2006, 1997=100)
Figure 12. Total factor productivity index (1989-2006, 1989=1, 2001-2006 imputed)
Source: Authors’ calculations based on Olley and Pakes method.
0
50
100
150
200
250
300
1988 1990 1992 1994 1996 1998 2000 2002 2004 2006
Manufacturing Cons . electronics A utomotiv e
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
totalfactorproductivity(1989=1)
Automotive Cons. Electronics Series2 Series4
31
Figure 13. Manufacturing exports and shares of automobiles and consumer durables (1989-
2006)
0
2
4
6
8
10
12
14
16
18
1989 1991 1993 1995 1997 1999 2001 2003 2005
%shareinmanufacturingexports
0
10
20
30
40
50
60
70
80
90
BillionUSD
Automotive (%) Cons. elect. (%) Manufacturing exports (bil.USD)
Figure 14. Net sales of automobile and consumer electronic companies in Turkey, million
YTL (1993 prices)
Source: Istanbul Stock Exchange
0
5
10
15
20
25
30
1992 1994 1996 1998 2000 2002 2004 2006 2008
Netsales
Isuzu Ford Otosan Karsan Otokar Tofaş Arçelik Beko
Vestel Bosch Ditaş Ege Endüstri F-M Mutlu Parsan
32
Figure 15. Export-output ratio of automobile and consumer electronic companies in Turkey,
1998-2005
0.00
0.20
0.40
0.60
0.80
1.00
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Exportrate
Isuzu Ford Otosan Karsan Otokar Tofaş Arçelik Beko
Vestel Bosch Ditaş Ege Endüstri F-M Mutlu Parsan
Source: Istanbul Stock Exchange
Figure 16. Net profit margin of automobile and consumer electronic companies in Turkey,
1993-2006 (net profit after taxes/net sales)
-0.30
-0.20
-0.10
0.00
0.10
0.20
0.30
0.40
1992 1994 1996 1998 2000 2002 2004 2006 2008
Netprofitmargin
Isuzu Ford Otosan Karsan Otokar Tofaş Arçelik
Beko Vestel Bosch Ditaş Ege Endüstri F-M
Mutlu Parsan Average
Source: Istanbul Stock Exchange
33
Figure 17. General Election results, 1987, 1991, 1995, 1999 and 2002
1987
RP
7.2%
MHP
2.9%
DYP
19.1%
DSP
8.5%
SHP
24.7%
Bağ.
0.4%Diğer
0.8%
ANAP
36.3%
1991
RP
16.9%
MHP
0.0%
DYP
27.0%
ANAP
24.0%
DSP
10.8%
SHP
20.8%
Diğer
0.4%
Bağ.
0.1%
1995
RP
21.4%
MHP
8.2%
DYP
19.1%
ANAP
19.7%
DSP
14.6%
CHP
10.7%
Bağ.
0.5%
Diğer
1.6%
HADEP
4.2%
1999
FP
15.4%
MHP
18.0%
DYP
12.0%ANAP
13.2%
DSP
22.2%
CHP
8.7%
Bağ.
0.9%
Diğer
4.9%
HADEP
4.7%
2002
AKP
34.3%
MHP
8.3%DYP
9.6%
ANAP
5.1%
DEHAP
6.2%
Genc
7.2%
Diğer
5.1%
Bağ.
1.1%
SP
2.5%
CHP
19.4%
DSP
1.2%
34
Table 1. Macroeconomic indicators, 1995-2006
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Percent of GNP
Public sector borrowing requirement 5.0 8.6 7.7 9.4 15.5 11.8 16.4 12.7 90.3 40.7 -0.4 -3.0
Primary surplus 3.3 1.7 0.1 4.3 1.8 5.3 5.8 4.0 5.2 6.1 7.7 7.7
Interest expenditures (Consol. budget) 7.3 10.0 7.7 11.5 13.7 16.3 23.3 18.9 16.4 13.2 9.0 8.0
Public sector debt stock (gross) 37.6 40.3 40.5 41.3 51.8 68.1 107.3 93.4 83.2 77.3 71.6 63.3
Domestic 14.6 18.5 20.2 21.7 29.3 43.1 71.1 56.3 56.4 54.5 52.9 46.4
External 23.0 21.8 20.3 19.6 22.5 25.0 36.2 37.1 26.8 22.8 18.7 16.9
Percent per annum
Interest rate on bonds and T-bills 124.2 132.2 107.4 115.5 104.6 38.2 99.6 62.7 46.0 24.7 16.2 18.0
Inflation (CPI, end of year) 76.0 79.8 99.1 69.7 68.8 39.0 68.5 29.7 18.4 9.0 7.1 9.1
GNP growth rate 8.0 7.1 8.3 3.9 -6.1 6.3 -8.5 7.8 50.9 7.1 7.1 6.0
Real exchange rate (CPI) 103 102 116 121 127 148 116 125 141 143 171 160
Notes: Real Effective Exchange Rate in December.
Sources: TURKSTAT, Undersecretary of Treasury, and the Central Bank of Turkey
Table 2. Descriptive statistics on ICT and automobile industries in Turkey
(1995-2000 averages)
Number of Employment Value added Relative labor Relative
employees share share productivity wages
(2000) (percent) (percent)
Computing machinery 1186 0.06 0.08 1.11 0.92
Electronic components 1353 0.11 0.06 0.53 0.83
TV/radio transmitters 5535 0.53 1.11 2.25 2.81
TV and radio receivers 7375 0.59 0.87 1.52 1.47
Motor vehicles 28060 2.20 4.11 1.86 1.83
Automobile bodies 2762 0.28 0.17 0.59 0.81
Automobile components 18042 1.65 1.30 0.78 1.11
Total (ICT and automobiles) 64313 5.42 7.69 1.42 1.59
Source: Calculated from UNIDO, Industrial Statictics Database, 2007
Table 3. Export and import values on ICT and automobile industries in Turkey
(million US$)
Exports Imports
1990 2005 1990 2005
Computing machinery 20 69 467 2465
Electronic components 3 36 429 1695
TV/radio transmitters 14 40 164 1765
TV and radio receivers 223 3074 226 1145
Motor vehicles 81 7802 794 8271
Automobile bodies 3 115 8 112
Automobile components 90 2308 480 3948
Total (ICT and automobiles) 433 13445 2568 19401
Total manufacturing 10400 67900 16200 93400
Source: UNIDO, Industrial Demand-Supply Balance Database, 2007
35
Table 4. Direction of foreign trade in ICT and automobiles, 1990 and 2005
(million US$)
Developing % Developed % Developing % Developed %
1990
ICT components 143 33 285 67 2 57 1 43
ICT products 169 20 689 80 26 10 230 90
Automobile components 8 2 480 98 18 19 75 81
Automobiles 25 3 769 97 34 42 47 58
2005
ICT components 795 47 899 53 12 33 24 67
ICT products 2698 50 2676 50 147 5 3037 95
Automobile components 345 8 3715 92 625 26 1799 74
Automobiles 888 11 7383 89 1044 13 6758 87
Note: ICT components: ISIC 3210; ICT products: ISIC 3000, 3220 and 3230
Automobile components: ISIC 3420 and 3430; Automobiles: ISIC 3410
Imports Exports
Source: UNIDO, Industrial Demand-Supply Balance Database, 2007
Table 5. Relative productivity and wages, selected countries
(1998-2002 average values)
3230 3410 3430 3230 3410 3430 3230 3410 3430
Turkey 0.51 0.27 0.24 0.30 0.23 0.17 0.60 0.87 0.70
Czech Republic 0.10 0.14 0.12 0.12 1.26 0.87
Hungary 0.14 0.33 0.15 0.14 0.16 0.12 1.02 0.47 0.79
Slovakia 0.04 0.10 0.08 0.08 0.10 0.08 2.14 0.95 1.00
Poland 0.23 0.13 0.13 0.22 0.15 0.14 0.96 1.15 1.05
Spain 0.38 0.26 0.42 0.63 0.49 0.49 1.68 1.87 1.18
Brazil 0.35 0.24 0.24 0.25 0.31 0.19 0.70 1.32 0.81
Mexico 0.40 0.42 0.25 0.14 0.19 0.16 0.36 0.46 0.65
India 0.10 0.06 0.05 0.05 0.07 0.04 0.52 1.22 0.76
Indonesia 0.08 0.40 0.09 0.04 0.06 0.04 0.48 0.14 0.44
Korea 0.42 0.64 0.46 0.37 0.45 0.32 0.87 0.71 0.71
Malaysia 0.12 0.22 0.15 0.13 0.12 0.10 1.14 0.53 0.66
Philippines 0.12 0.09 0.08 0.07 0.69 0.76
Singapore 0.38 0.18 0.35 0.62 0.47 0.52 1.63 2.60 1.47
UK 0.44 0.30 0.46 0.83 0.82 0.71 1.90 2.72 1.55
Germany 0.53 0.30 0.53 1.08 0.87 0.85 2.02 2.85 1.60
Japan 0.83 1.23 0.84 0.96 1.17 0.88 1.15 0.95 1.05
USA 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Note: Poland, 2002; Malaysia, 2000-2002; Mexico, 1998-2000; Philippines, 1998, 1999 and 2001; Turkey, 1998-2001
All other countries, 1998-2002 averages.
Source: Calculated from UNIDO, Industrial Statictics Database, 2007
Relative labor productivity Relative wages Unit labor cost
36
Table 6. Degree of vertical integration, selected countries
(1998-2002 average values for value added/output ratio)
3230 3410 3430
Turkey 0.293 0.330 0.470
Czech Republic 0.140 0.254
Hungary 0.096 0.180 0.298
Slovakia 0.154 0.116 0.292
Poland 0.177 0.168 0.314
Spain 0.155 0.154 0.298
Brazil 0.308 0.318 0.453
Mexico 0.276 0.262 0.317
India 0.154 0.177 0.265
Indonesia 0.270 0.604 0.322
Korea 0.309 0.384 0.387
Malaysia 0.133 0.222 0.352
Philippines 0.283 0.260
Singapore 0.175 0.311 0.407
UK 0.263 0.183 0.352
Germany 0.317 0.222 0.351
Japan 0.311 0.253 0.307
USA 0.346 0.322 0.438
Note: Czech Republic, 1998 and 2002; Poland, 1998-2000 and 2002; Malaysia, 2000-200
Mexico, 1998-2000; Philippines, 1998, 1999 and 2001; Turkey, 1998-2001
All other countries, 1998-2002 averages.
Source: Calculated from UNIDO, Industrial Statictics Database, 2007
Table 7. Major automobile and consumer electronic producers listed on
the Istanbul Stock Exchange in Turkey
Employees Export rate Import rate F/D
Automotive
Anadolu Isuzu 741 15.0 34.0 F
Ford Otosan 7722 42.7 63.0 F
Karsan 957 3.5 61.8 D
Otokar 988 37.0 43.0 D
Tofaş 4379 48.7 46.1 F
Parts and components
Bosch Fren Sistemleri 246 81.2 57.0 F
Ditaş Doğan 576 41.0 24.0 D
Ege Endüstri 474 59.0 25.0 D
F-M İzmit Piston 24 22.3 4.1 D
Mutlu Akü 561 36.0 55.0 D
Parsan 565 66.0 33.0 D
Consumer electronics
Arçelik 9203 27.5 47.0 D
Beko 3907 76.3 65.2 D
Vestel 11286 75.4 50.0 D
Notes: 2005 data for the number of employees, export rate and import rate.
Export rate is the share of exports in sales revenue. Import rate is proportion
of imported inputs to sales revenue.
Source: Istanbul Stock Exchange