Research and Monetary Policy Department

Working Paper No:06/07

Corporate Sector Financial Structure in
Turkey: A Descriptive Analysis
The Central Bank of the Republic of Turkey
December 2006
Halil . AYDIN
Cafer KAPLAN
Mehtap KESRYEL
Erdal ZMEN
Cihan YALIN
Serkan YT




Corporate Sector Financial Structure in Turkey:
A Descriptive Analysis
1



Halil . Aydn
a
, Cafer Kaplan
a
, Mehtap Kesriyeli
a

Erdal zmen
a, b
, Cihan Yaln
a
, Serkan Yiit
a



a
Central Bank of the Republic of Turkey, Research and Monetary Policy
Department, Ankara, Turkey.
b
Middle East Technical University, Department of Economics, Ankara, Turkey.

ABSTRACT
This paper presents and discusses some stylised facts of the corporate sector financial
structure in Turkey using the Company Sector Accounts compiled by the Central Bank of the
Republic of Turkey (CBRT). The findings of the paper suggest that non-financial firms in
Turkey have been heavily exposed almost all of the basic balance sheet risks. The corporate
sector appears to be excessively leveraged with relatively lower asset tangibility creating also
a credit risk for the lenders. The firms rely heavily on foreign currency denominated and
short-term debt instruments making them vulnerable to both exchange rate and interest rate
shocks through currency and maturity mismatches. The corporate sector can be characterised
as financially constrained as the deepening of the Turkish bank-based financial system is
rather low and the bank credits to the private sector tend to be crowded out by the mode of
domestic debt finance. The corporate sector vulnerabilities to maturity, interest rate and
currency risks are found to be improving with the firm size. With the relatively stable
macroeconomic environment and stricter prudential regulation on the financial system,
the corporate sector balance sheet risks, albeit still are at high levels, tend to be improving
after the financial crisis of 2001.

Keywords: Balance sheets, Capital structure, Corporate sector, Debt composition, Financial
crowding-out, Liability dollarisation, Turkey
JEL Classification: E22, F31, G32

1
The views expressed in the paper are those of the authors and should not be attributed to the Central
Bank of the Republic of Turkey (CBRT). We are much grateful to Fatih zatay, Erol Taymaz and
Gkhan Ylmaz for their invaluable contributions to the paper. We want to thank to Zerrin Grgenci,
Metin ner and Serap elen for their collaboration in having the data set from the Statistics
Department of the CBRT. The usual disclaimers apply. The fourth author gratefully acknowledges the
Research and Monetary Policy Department of the CBRT and Ahmet Kpc for providing an excellent
research environment.
Corresponding Author: Erdal zmen, Middle East Technical University, Department of Economics,
06531, Ankara, Turkey. e-mail: ozmen@metu.edu.tr.



1

I. Introduction
The structures of the balance sheets of the main sectors (public, commercial
banking, non-financial corporate and households) of an economy are crucially
important for financial stability, sustainable growth and the channels through which
monetary policy is transmitted. The recent balance sheet approach to financial crisis
literature such as Krugman (1999), Allen et al. (2002) and Aghion et al. (2004) stress
the roles of balance sheet vulnerabilities resulting from the level, currency composition
and maturity of debt of the main economic sectors in triggering and determining the
output cost of a crisis. In the same vein, credit channels of the monetary policy
transmission mechanisms focus on the balance sheets of lenders (bank lending channel)
and borrowers (balance sheet channel). Given the fact that the balance sheets of all the
main sectors of an economy are interrelated and a risk accumulated in one sector can
have serious repercussions on the others, the insights offered by the alternative
approaches may be interpreted as complementary rather than mutually exclusive.
In this paper we consider the balance sheets of the non-financial firms
2
in Turkey.
The conventional wisdom assuming perfect frictionless markets suggests that firms
real decisions are invariant to both their balance sheet structures and nominal changes
in the economy (Modigliani and Miller, 1958). The credit channel literature, however,
considers capital market imperfections (due to asymmetric information in form of
moral hazard and adverse selection, agency costs etc.) and provides two complementary
channels where the financial positions of non-financial firms matter for their real
economic activity (Bernanke et al., 1999 and Gertler et al., 2003). First channel which
is known as the bank lending, focuses on the asset side of the bank's balance sheets and
discusses the impact of changes in credit conditions on the investment/spending
decisions of borrowers (firms). The second channel that is usually named as the balance
sheet channel, focuses more on the balance sheets of borrowers rather than those of
lenders in case of change in monetary policy stance which matters for finance costs and
thus real activity of borrowers. In essence, the two channels may be expected to be
interrelated, amplifying the effect of a shock on the economy as suggested by the

2
We use non-financial firms and corporate sector (firms) interchangeably
2
financial accelerator mechanism (Gertler et al., 2003). The magnitude of the
amplification, in turn, depends on the sensitivities of firms balance sheets to monetary,
financial and real shocks.
The degree of the sensitivities of corporate sector to shocks is associated with
their capital structure, and the level, currency composition and maturity of their debt.
Indebtedness and poor collateral position make firms more vulnerable to interest rate
shocks and deepen the output cost of a financial crisis. Short-term debt, on the other
hand, can amplify the interest rate risk by creating a rollover risk. Financial positions of
firms are closely associated with the currency composition of their balance sheets and
sensitivity of their net worth to real exchange rate fluctuations. Liability dollarisation
can make firms earning mainly in domestic currency vulnerable to currency mismatch
risk and real exchange rate depreciations. The financial structures of firms in a country
may not be invariant to the prevailing institutional structure and the level of financial
intermediation. The corporate sectors of countries with relatively weak financial
intermediation can be expected to rely more on internal funds and bank lending than
those of the countries with developed securities markets allowing firms also bond
financing (Davis and Stone, 2004 and IMF, 2005). The firms in relatively less efficient
financial systems can be expected to use more trade credits due to an informational
asymmetry. The finance constraint, the magnitude of which can be more severe with a
lower level of financial intermediation, may be varying across firm groupings with
different size (Beck et al., 2005a). Beside the level of financial deepening, the public
sector domestic debt finance via the domestic banking system can lead a decrease in the
banking system preference to finance corporate sector (financial crowding-out) and
thus can make the non-financial firms to be severely bank-credit constraint.
In this paper we first present some stylised facts of the non-financial firms
liability structure in Turkey using the Company Sector Accounts compiled by the
Central Bank of the Republic of Turkey (CBRT)
3
. To this end, we focus on the basic

3
We restrict our sample to non-financial firms as the behaviour and capital structure of banks under
financial regulation are not comparable with those of non-financial firms. The CBRT Company Sector
database covers a wide range of information for around 8000 firms annually after 1990. The CBRT
website www.tcmb.gov.tr provides detailed information on the database and sectoral data for the years
after 1997.
3
sources of vulnerabilities including the level, maturity structure and currency
composition of the debt. The capital structure literature often focuses on the liability
side of corporate sector balance sheets and ignores the possibility that firms also can
hedge themselves. Furthermore, non-financial firms can even behave as financial
intermediaries under a low level of financial development, macroeconomic instability
and the consequent uncertainty. Examples for the financial intermediation of non-
financial firms may include their heavy reliance on trade credits and holdings of
financial assets including government debt instruments and foreign exchange
denominated assets to hedge against interest rate and exchange rate shocks,
respectively. Therefore, we consider also the firms assets and income statements and
discuss the stylised facts of their hedging behaviour using the relevant available
information from the corporate sector accounts.
The plan for the rest of the paper is as follows. In section II, we provide some key
corporate finance indicators in Turkey and compare them with those of groups of
industrial and developing countries. Section II.1 presents some evidence concerning
firms indebtedness and collateral base to assess their financial positions and
constraints. In II.2 we consider the composition of corporate sector liabilities and assess
the roles of trade credits and bank lending in the context of the level of financial
intermediation and the classification of financial systems as bank or market based.
Section II.3 focuses on bank loans and stresses on the low levels of financial deepening
and the banking system preference to finance corporate sector in leading the firms to be
severely bank-credit constraint in Turkey. The maturity, interest rate and currency risks
of the corporate sector are discussed in Section II.4. In II.4.1 we consider the maturity
structure of the debts and interest coverage ratios of the corporate sector firms. The
currency composition of the debt and the corporate sector liability dollarisation are
discussed in Section II.4.2. Section II.5 presents the evolution of the basic balance sheet
fragilities (debt level, maturity structure, liability dollarisation and interest coverage
risk) before, during and after the 2001 financial crisis. Section III is devoted to the
discussion of the financial asset holding behaviour of the non-financial firms using the
relevant available information from their assets and income statements. Government
domestic debt finance can alleviate or relax the firms financial constraints by draining
4
the available resources in the economy or by providing liquidity services, respectively.
In section III.1, we argue that these affects may not be invariant to both the degree of
financial deepening and the level and mode of government debt finance. In Section
III.2, we discuss the financial asset holdings of the firms not only in the context of the
conventional transactions/precautionary motives for liquidity but also considering
potential precautionary-cum-speculative motive under uncertainty which can lead the
allocation of assets into financial and real investments to be substitutes. Finally, Section
IV concludes.

II. Corporate Sector Financial Vulnerabilities in Turkey
Excessive debt and weak liability structure are often defined among the major
sources of corporate sector vulnerabilities. Table 1 reports some key corporate balance
sheet indicators for groups of developed, emerging market and small industrial
countries along with Turkey. The Turkish corporate sector appears to have the highest
leverage ratio (measured as debt over total assets) among all the country groups before
2004 as suggested by the lowest share of shareholders equity in total liabilities. The
leverage ratio (LR) appears to improve considerably and decreases to a level close to
those for small industrial countries in 2004. According to the figures, the Turkish firms
rely least on bond or bank finance and most on trade credits among all the country
groups suggesting a relatively low level of financial intermediation. The Turkish firms,
on the other hand, have the lowest asset tangibility and the highest debt ratio potentially
suggesting a corporate finance puzzle. Profitability of the Turkish firms, which declined
sharply with the financial crisis of 2001, appears to be somewhat between that of the
developed and emerging market countries. The liquidity ratio is closer to that of the
industrial countries and substantially lower than the group of emerging market
countries. Given the fact that the bulk of the Turkish corporate sector debt is short-term
(see below) in contrast to the industrial countries, such a liquidity ratio may be
interpreted as relatively low to offset the firms greater vulnerability to shocks. During
most of the period the opportunity cost of remaining liquid by holding Turkish lira (TL)
cash was extremely high under the sustained high inflation rates. However, as argued in
Section III.2, the salient features of the Turkish financial system allowed the firms to
5
remain relatively liquid by holding short-term and often foreign currency denominated
assets also for precautionary and transactions purposes. In the following subsections we
proceed with the discussion of the main balance sheet vulnerabilities of the Turkish
non-financial firms in more detail.

Table 1: Some Key Balance Sheet Indicators across Country Groups and Turkey

Shares of Corporate Liabilities
Bank
Loans

Bonds
Own Funds
and Equity
Trade
Credits
2

Asset
Tangibility
5

Profitability
(ROA)
4

Liquidity
Ratio
G-7 Countries
1
0.23 0.08 0.63 0.06 0.45
3
3.39
3
0.21
Germany
1
0.53 0.02 0.42 0.03 0.41 3.88 0.26
UK
1
0.23 0.08 0.64 0.06 0.27
Small Industrial
Countries
1

0.30 0.04 0.57 0.08 0.26
Emerging Markets
1
0.27 0.21
6
0.40 0.14 0.68 7.88 0.42
Turkey

(2000)
All NF Firms


0.26 000 0.34 0.40 (0.19) 0.25 2.67 0.26
5

Turkey

(2000)
Manuf. Industry
0.26 0.00 0.36 0.38 (0.20) 0.27 3.80 0.21
5

Turkey

(2001)
All NF Firms
0.30 0.00 0.29 0.41 (0.19) 0.24 2.20 0.24
5

Turkey

(2001)
Manuf. Industry
0.25 0.00 0.35 0.40 (0.21) 0.25 2.60 0.19
5

Turkey

(2004)
All NF Firms
0.17 0.00 0.51 0.32 (0.14) 0.28 4.00 0.22
5

Turkey

(2004)
Manuf. Industry
0.18 0.00 0.53 0.30 (0.18) 0.31 4.60 0.19
5

Notes: 1. 1999 or latest year, source Davis and Stone (2004). 2. The trade credits include also the other credits, the values in
parentheses give the ratio of pure trade credits for the Turkish corporate sector. 3. For the G-3 countries. 4. Source IMF
(2005a) for all the countries except Turkey. 5. Liquidity Ratio (Cash Ratio) = (Liquid Assets + Marketable Securities)/
Short-term Liabilities. 6. Davis and Stone (2004, p. 70) note that, this high share is due to the large share of bond financing
in Korea dominating the small sample. Sources: CBRT Company Sector Accounts, Davis and Stone (2004), IMF (2005a)
and our calculations.


II.1. Leverage Ratios
Glen and Singh (2003) compute the average leverage ratios (LRs) for developing
and developed countries in the period of 1994-2000 as 56.4% and 52.6%, respectively.
Consistent with the findings of Fan, Titman and Twite (2004), Davis and Stone (2004)
and IMF (2005a), developing country firms appear to have slightly higher LRs. This
may be a plausible result as financial development can lead firms to move away from
loan financing (bank and trade credits) towards market financing securities and internal
6
equity. However, such an interpretation ignores the fact that financial structures may
differ even between countries with compatible financial development levels. For an
international comparison, Figure 1 plots the LRs for the year 2000 provided by Glen
and Singh (2003). From the figure, it can be inferred that LRs vary substantially across
countries as well as within country groups. Although the overall picture from Table 1
may lend a support to the view that the share of corporate liabilities accounted for by
loans is decreasing in the level of economic development (Davis and Stone, 2004, p.
69), it also calls for a caution as bank based (Germany) and market based (USA and
UK) systems have substantially different firm debt levels albeit belonging to the same
developed country grouping
4
. These results, however, may imply that firms in
developed countries are expected generally to be less financially constrained and less
likely subject to informational problems and agency costs. These firms prefer mostly
retained earnings and direct finance that are cheaper than bank finance to satisfy a large
portion of their financial needs (Myers and Majluf, 1984).
The Turkish firms are amongst the most indebted as also suggested by Figure 1. A
similar picture arises when we consider the debtequity ratio (Table 1). The high level
of LRs can be a potential source of risk, as higher indebtedness increases the premium
that has to be paid on external finance, and, thus can potentially affect investment
adversely. High LRs can also be an indicator of the vulnerability of corporations to
macroeconomic shocks as recently suggested by the Asian crisis (Ratha et al., 2003 and
Davis and Stone, 2004).

4
See Allen, Chui and Maddaloni (2004), Byrne and Davis (2004) and IMF (2005) for recent
comparisons of bank-based and market-based financial systems. According to Byrne and Davis (2004)
there is some evidence of convergence towards a more market-oriented system in the major European
Union countries during the 1990s. According to IMF (2005, p. 125) there is no evidence that market-
based systems, or bank-based systems, are associated with better economic performance. IMF (2005,
p.95) further notes that despite the increasing importance of domestic and international bonds as a
source of corporate finance, bank lending remains the dominant source of corporate finance for all
emerging market regions.


7
Figure 1. Corporate Leverages in Developed and Developing
Countries, 2000
0
20
40
60
80
100
C
o
l
o
m
b
i
a
V
e
n
e
z
u
e
l
l
a
H
o
n
g

K
o
n
g
I
s
r
a
e
l
A
r
g
e
n
t
i
n
a
P
h
i
l
l
i
p
i
n
e
s
B
e
r
m
u
d
a
C
h
i
l
e
P
o
l
a
n
d
C
z
e
c
h
H
u
n
g
a
r
y
U
S
A
S
i
n
g
a
p
o
r
e
I
n
d
i
a
T
a
i
w
a
n
C
a
n
a
d
a
M
a
l
a
y
s
i
a
P
e
r
u
U
K
S
.

A
f
r
i
c
a
K
o
r
e
a
S
w
e
d
e
n
S
w
i
t
z
e
r
l
a
n
d
A
u
s
t
r
a
l
i
a
J
a
p
a
n
M
e
x
i
c
o
S
p
a
i
n
B
e
l
g
i
u
m
G
r
e
e
c
e
F
i
n
l
a
n
d
N
o
r
w
a
y
D
e
n
m
a
r
k
N
e
t
h
e
r
l
a
n
d
s
B
r
a
z
i
l
F
r
a
n
c
e
T
h
a
i
l
a
n
d
T
U
R
K
E
Y


A
u
s
t
r
i
a
P
a
k
i
s
t
a
n
G
e
r
m
a
n
y
I
t
a
l
y
I
r
e
l
a
n
d
I
n
d
o
n
e
s
i
a
LR(%)

Source: Glen and Singh (2003).

Figure 2 reports the LRs of the firms

during the 1990-2004 period. The
manufacturing firms classified as large appear to have smaller leverage ratios compared
to small and medium size firms during the period.
5
Large firms tend to finance their
activity through non-debt finance i.e. internal funds. For the manufacturing sector,
medium sized firms are generally more indebted compared to the small firms. Among
non-financial (NF) firms, manufacturing (Manuf) firms on average appear to be less
indebted. Large manufacturing firms have generally relied more on internal finance
compared to small and medium sized firms especially during expansionary periods as
these firms are less likely to be financially constrained and they prefer the cheapest
source of finance - retained earnings - to finance their investment in these times (Figure
3). However, the share of external finance, mainly debt, increased relatively during the
recessionary periods of 1999 and 2001 because of either squeezing of their retained
earnings or shifting to bank finance that is more suitable option for them in these times.
On the other hand, smaller firms are more likely to be financially constrained and they
rely more on external finance in general even though they have less access to bank

5
Consistent with the BACH (The Bank of Harmonised Data on Company Accounts) scheme, we
classified the firms as small if their sample means of net sales are not larger than EUR 7 millions. The
firms with sample means of net sales are larger than EUR 40 millions are classified as large whilst the
rest apparently constituting the medium sized firms.
8
finance compared to larger firms during recessions. In fact, banks are expected to ration
the credits and thus they do not want to provide funds to small firms with poor
collateral, instead they prefer collaterally rich large firms during the economic
slowdown (Gertler and Gilchrist, 1994). Accordingly, the LRs of small firms expected
to be more pro-cyclical compared to those of large firms. However, figures for the
small firms do not confirm this hypothesis in Turkey. The pro-cyclicality of the LRs
and thus investments arguments clearly do not consider the fact that potentially
constrained firms (especially small firms) had access to trade credits during recessions.
As will be discussed later in more details, substantially large share of trade credits
especially for small and medium sized firms may be interpreted as a buffer providing
them some chance of avoiding the negative impacts of the finance constraint, and thus
smoothing out the pro-cyclical behaviours of their real and financial activity during
recessions.












Figure 2. Corporate Sector Leverage Ratios, %
45
50
55
60
65
70
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
All NF Firms M_Small M_Medium
M_Large Manuf.
9


A striking result that can be inferred from Figures 2 and 3 is that the
leverage ratios for all the sectors tend to decline substantially after the financial
crisis of 2001. This may be consistent with a view that the Turkish corporate
firms, which were severely hit by the crisis due to their vulnerabilities
including very high debt levels, became more prudent about external finance
after the crisis. In addition, relatively stable macroeconomic environment after
the crisis improved firms investment and thus their internal funds.
Consequently, the LRs came down to the levels comparable to those of the
firms in industrial and small industrial countries by 2004 (Table 1). The decline
in the LRs and thus the increase in the use of own funds (Own Funds/Total
Assets = 1-LR) after 2001 appears to be more sharper for the large
manufacturing firms compared to other firm groups. Consistent with the fact
that they are the most constrained in terms of own funds, the post-crisis
adjustment towards lower LRs appear to be relatively sluggish for small
manufacturing firms.
Figure 3. Real GDP Growth and Manufacturing Sector LRs
45
50
55
60
65
70
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
L
R

(
%
)
-10
-8
-6
-4
-2
0
2
4
6
8
10
12
R
e
a
l

G
D
P

G
r
o
w
t
h
,

%
M_Small M_Medium M_Large GDP Growth
10
Another important indicator that provides information on the corporate finance is
the tangibility of firms assets (a proxy for the collateral level of firms). Figure 4
presents the collateral ratios measured as the ratio of the firms net tangible fixed assets
to total assets. Collateral ratios appear to be increasing with the size of the
manufacturing firms. The evidence that small manufacturing firms have the lowest
collateral ratios but not the lowest leverage ratios (Figures 2 and 4) is not consistent
with the argument that suggests positive association between leverage ratio and
tangibility and thus firm size in the Turkish case (Fan et al., 2004). As will be discussed
later, the share of bank loans in total loans tends to be monotonically increasing with
the firm size. It may be argued that asset tangibility is more related to bank loans rather
than the leverage ratio per se. This may not be implausible, since the LR contains other
kinds of debt including trade credits, which does not directly rely solely on the
collateral.










Compared to the figures in Table 1, the asset tangibility in Turkey is lower than
both the emerging market and the developed countries. According to IMF (2005, p.
103) asset tangibility in emerging markets is 50 percent larger than in the G-3
countries, supporting a higher level of corporate leverage. Informational asymmetries
between borrowers and lenders may decrease with higher asset tangibility. Therefore,
Figure 4. Tangible Fixed Assets (Net)/Total Assets, %
(Collateral)
20
25
30
35
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
All NF Firms M_Small M_Medium
M_Large Manuf.
11
higher asset tangibility may allow for higher leverage. The Turkish case, with lower
asset tangibility and higher leverage ratios compared internationally can be interpreted
as providing a puzzle. However, when we consider the maturity and currency
composition of the debt (see below), along with the already discussed fact that a
substantial part of the debt is short term trade credits, the Turkish evidence may become
much less puzzling. It is worth noting that the collateral positions of the Turkish
corporate sector firms tend to be gradually improving after the financial crisis of 2001.

II.2. The Composition of the Corporate Sector Liabilities
Figure 5 plots the composition of corporate sector liabilities in Turkey during
1992-2004. The share of bank loans in corporate sector liabilities appears to be close to
the advanced countries (G7) average (Table 1). However, such a comparison may be
misleading as the G7 group contains both market-based (e.g. USA and UK) and
bank-based (e.g. Germany) financial systems. Given the fact that corporate bond
issuance has virtually no role in corporate sector in Turkey, the Turkish financial
system may better be compared to bank-based systems rather than market-based
systems. Both the composition of corporate sector liabilities (Figure 5) and the share of
bank loans in total loans (Figure 6, below) show that bank loans are not the dominant
source of finance in Turkey. The share of bank loans in total corporate sector liabilities
fluctuates around 20 % during the period. Considering the ratios for a typical bank-
based (e.g. Germany, 53 %) and a market based country with active securities markets
(e.g. UK, 23%) presented in Table 1, it may be plausible to argue that bank lending to
private sector in Turkey is rather limited and close to market-based systems. However,
with the absence of an active and dominant corporate bond market, the Turkish firms
can be interpreted as severely finance-constrained.




12

The share of bank loans in total corporate sector liabilities tends to decline
considerably after the financial crisis of 2001. From figure 5, it may be inferred that
the increase in the use of own funds (hence the decline in the LRs) after 2001 is
basically due to the decline in the share of bank loans rather than the other forms of
the debt including trade credits. This might be due to stricter prudential regulation on
the banking system as well as firms behaving more prudent on excessive bank debt
after the crisis. However, given the fact that the corporate sector already being
severely bank credit constrained, such development may not be interpreted as growth
enhancing especially in the absence of a well developed corporate bond market.
A striking feature of the Turkish corporate sector is the high share of trade credits
in total liabilities. Demirguc-Kunt and Maksimovic (2002) claim that firms tend to use
more bank loans rather than trade credits in countries with more efficient financial
systems. The negative relationship between the development level and trade credit
share suggested by Table 1 is consistent with an argument that trade credit is an
important source of financing in economies with underdeveloped financial
intermediaries. However, this argument per se may not be sufficient to explain
substantially high share of trade credits in Turkey compared to both the emerging
market and small industrial countries by the financial development levels of countries.
Figure 5. The Composition of Corporate Sector Liabilities,
%
0
20
40
60
80
100
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Bank Loans Trade Credits Own Funds Others
13
The presence of trade credits is often explained by information/monitoring
advantage of suppliers over banks in financing constrained firms (Burkart and
Ellingsen, 2004 and Davis and Stone, 2004). This information asymmetry may be
expected to decrease with the level of financial development. Furthermore, suppliers
having access to privileged information about their customers creditworthiness and
ability to exert corporate control more readily than banks are often entitled to seize the
firms inputs and other assets in the case of a default. This can make inputs to be
considered also as collateral worthing more to the supplier who is in the same business
than to the bank (Frank and Maksimovic, 1998). In this context, trade credit can be a
substitute for bank credit even for firms with sufficient debt capacity.
Trade credits become more important when creditor protection is weaker
(Demirg-Kunt and Maksimovic, 2002 and Fisman and Love, 2003) and when firms
are undercapitalized (Burkart and Ellingsen, 2004). For the firms with credit constraints
and low asset tangibility (collaterally poor), trade credit becomes an important
complement to bank credits. Compared to the international evidence, the Turkish firms
appear to be highly leveraged with lower asset tangibility (see below). The high level of
debt owes much to the trade credits the share of which in total firm liabilities almost
equals to the share of bank credits during the sample period (see Figure 4). The lower
asset tangibility of the Turkish firms can thus be interpreted as amongst the causes of
trade credit complementary to bank credit under financial constraints. The
macroeconomic instability and persistent high inflation may also be interpreted among
the causes of short-maturity monetary contracts including trade credits in Turkey
6
.
Extensive use of trade credits can be also interpreted as evidence of informal nature of
financial structure of Turkish firms. Existing institutional framework does not prevail
market oriented relations in the financial system and this creates its second best solution
in the form of trade credits together with some vulnerabilities that prevent to see
underlying problems.

6
Trkan (2004) claim that high taxes on financial intermediation makes bank finance expensive and thus
lead to extensive use of trade credits in Turkey. Yaln et al. (2005) find that small firms have the highest
share of trade credits in total liabilities compared to other firm groups where large firms act as financial
intermediaries.
14
II.3. Bank Loans
In Turkey, both the levels of financial deepening and the banking system
preference to finance corporate sector are very low leading corporate sector to be
severely bank-credit constraint. Figure 6 plots the 2003 values of the ratios of the
private credit by deposit money banks (DMB) and other financial institutions
7
to GDP
(CREDIT/GDP), private credit by DMB to GDP (BANK_CREDIT/GDP) and DMB
assets to GDP (BANK_ASSET/GDP) for a large sample of developed and developing
countries. The figure also plots a Bank Credit Allocation Index (BCAI) measured as the
ratio of deposit money banks credits to private sector to deposit money banks assets.
According to BANK_ASSET/GDP, the level of bank/financial intermediation in
Turkey is relatively low
8
. The bank credit allocation index for Turkey computed as 0.35
appears to be among the least three (along with Algeria and Argentina) in a sample of
75 countries. Accordingly, only around 35% of deposit money banking system assets
are allocated to corporate sector in Turkey
9
. Note that, only ten countries have BCAI
lower than 0.6 and more than one-third of the countries have BCAI higher than 0.9. As
will be discussed later, the extremely low level of banking system preference to finance
corporate sector can be attributed to financial crowding out of government debt finance
via commercial banking system in Turkey after the mid 1980s. Given the relatively low
levels of financial development and banking system preference to finance corporate
sector, the ratio of private credit by financial institutions to GDP (CREDIT/GDP)
appears to be very low (amongst the lowest 4 in the cross country sample) in Turkey.
As can be observed from the differences of the CREDIT/GDP and
BANK_CREDIT/GDP values, non-bank financial institutions are also important

7
The other financial institutions comprise banklike and nonbank financial (insurance companies,
provident and pension funds, trust and custody accounts, pooled investment schemes, compulsory
savings schemes) institutions. These are institutions that serve as financial intermediaries, while not
incurring liabilities usable as means of payment. All the data for Figure 5 are from the World Bank
Financial Development and Structure 2005 Dataset, See, Beck, Demirg-Kunt and Levine (2000) for a
detailed description of the database.
8
Note that a similar picture arises when we consider the ratio of liquid liabilities (currency plus demand
and interest-bearing liabilities of banks and other financial intermediaries) to GDP which is the broadest
available indicator of financial intermediation and a typical measure of financial depth (Beck et al. 2000).
9
These findings are consistent with those reported in Sy (2005), Hauner (2006) and Hanson (2003)
suggesting that Turkey appears to have the highest ratio of government securities in total banking system
assets amongst a large number of developing and industrial countries. We discuss this issue in Section
III.1 in more detail.
15
sources of credit to corporate sector for a number of countries including the USA,
S:Arabia, Canada, Iran, Malaysia, Korea and Thailand
10
. In Turkey, the share of non-
bank financial institutions in credits is somewhat negligible (5.1 %). Consequently,
corporate sector in Turkey can be characterised as suffering from a general finance
constraint in Turkey.
Figure 7 reports the average share of bank loans in total loans for non- financial
firms. Given the fact that the corporate sector, in general, suffers from a bank credit
constraint, the figure may be helpful in assessing the evolution of the share of bank
finance over time and across firm groups. The share of bank loans appears to be
monotonically increasing with the firm size. The figure also suggests a tendency
through a general improvement in bank loans until 1997-1998. After 1998 the bank
loans tend to decline until a slight improvement after 2002 with macroeconomic
stability. It is worth noting that, unlike the recent 2001 financial crisis, we do not
observe a severe bank credit squeeze (except for small firms in 1994 to certain extend)
during and after the 1994 crisis. Compared to large firms, small and medium sized
manufacturing firms appear to be somewhat more negatively affected from the
worsening bank credit conditions. The overall picture suggests that the access of small
sized manufacturing firms to bank finance appears to be more sensitive to general
squeezes in bank credits. The observation that small firms use less bank finance is
consistent with the results of Beck et al. (2005) based on cross-country data.

10
Non-bank financial institutions provide 64% of private credits in the USA. Other notable
countries are S. Arabia (50%), Canada (31%), Iran (29%), Malaysia (29%), Korea (24%), Thailand (20
%), Norway (20 %), Bolivia (20 %) and Chile (19%).
F
i
g
u
r
e

6
.

F
i
n
a
n
c
i
a
l

D
e
e
p
e
n
i
n
g

a
n
d

C
o
r
p
o
r
a
t
e

F
i
n
a
n
c
e
,

I
n
t
e
r
n
a
t
i
o
n
a
l

E
v
i
d
e
n
c
e

(
2
0
0
3
)
0
0
.
2
0
.
4
0
.
6
0
.
8 1
1
.
2
1
.
4
1
.
6
1
.
8 2
Romania
Algeria
Argentina
Turkey
Jamaica
Nigeria
Paraguay
Mexico
Guatemala
Kazakhstan
Ukraine
Peru
Kenya
Indonesia
Bulgaria
Colombia
Ecuador
Pakistan
Sri Lanka
Poland
Costa Rica
Lat via
Czech Rep.
India
Iran
Brazil
Philippines
Slovakia
Hungary
Honduras
Slovenia
Nicaragua
Bolivia
Croatia
Uruguay
Morocco
S. Arabia
Mauritius
Egypt
Finland
Tunisia
Greece
Jordan
Kuwait
S.Africa
Chile
Belgium
Italy
France
Panama
Israel
Norway
Australia
Thailand
Canada
Sweden
Austria
Japan
Luxembourg
Spain
Ireland
Singapore
New Zealand
Germany
Cyprus
Korea, Rep.
Malaysia
China
United Kingdom
Portugal
Denmark
Netherlands
Hong Kong
Switzerland
United States
C
r
e
d
i
t
/
G
D
P
B
a
n
k
_
A
s
s
e
t
/
G
D
P
B
C
A
I
B
a
n
k
_
C
r
e
d
i
t
/
G
D
P




17














II.4. Maturity, Interest Rate and Currency Risks of the Corporate Sector
II.4.1. Maturity and Interest Rate Risks
High short-term debt may make firms be more vulnerable to insolvency and
rollover risk especially in the case of an interest rate shocks as cash flow must be
available for interest payments. Figure 8 presents the maturity structure of the corporate
sector debt measured as the ratio of short-term liabilities to total debt. The bulk of the
Turkish corporate sector debt appears to be short-term (with maturity less than a year).
The corporate debt maturity tends to increase with firm size. From the figure, it may be
inferred that there has been a tendency towards an improvement in the maturity structure
after the mid-1990s. However, as will be discussed later, this owes much to increased
liability dollarisation, which allowed firms to borrow in foreign currency with a relatively
longer maturity during the period. Consistent with an argument that the debt structure of
firms may not be invariant to their tangibility, firms with greater collateral can be
expected obtain longer-term debt. As we have shown in Figure 4, asset tangibility
increases with firm size, similarly as shown in Figure 8, debt maturity tend to increase
with firm size and thus with asset tangibility.
Figure 7. Bank Loans/Total Loans, %
20
25
30
35
40
45
50
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
All NF Firms M_Small M_Medium
M_Large Manuf.
18










Figure 9 plots the short-term debt (ST) ratios for East Asian, Latin American and
Eastern European and Turkish non-financial firms. The debt maturity of the Turkish
corporate sector appears to be relatively short compared to international evidence. Based
on a cross-country data for 5344 firms of 39 countries during 1991-2001, Fan et al.
(2004) compute that the median short-term debt ratios for developing and developed
economies are 65% and 45%, respectively. According to IMF (2005a, p.118), the ratio of
short-term debt to total debt in percentage for market participants (non-participants) is
35.7 (51.7) for emerging market countries, 29.4 (44.0) for Latin America, 39.1 (51.3) for
Asia and 38.5 (60.0) for Europe. In the IMF sample, Turkey has the highest short-term
debt ratio (55.3 for market participants and 65.5 for non-participants) among the
developing countries considered. Although the IMF (2005) figures are somewhat smaller
than those in Figure 8, potentially due to the very limited sample size of IMF (2005a), the
Turkish firms can be interpreted to rely more on short-term debt leading them to be more
vulnerable to shocks. According to Bleakley and Cowan (2004) and IMF (2005a), the
higher short-term debt ratios can lead firms to be more liquid due to the potential risks
stemming from maturity mismatches. The figures in Table 1 suggest that the Turkish
firms appear to have similar liquidity ratios with those of the industrial countries where
the corporate sectors in these countries can borrow more in longer maturities and in terms
Figure 8. Short Term Liabilities/Total Loans (%)
75
80
85
90
95
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
All NF Firms M_Small M_Medium
M_Large Manuf.
19
of their domestic currencies. Therefore, the liquidity ratio of the Turkish corporate sector
firms may be interpreted as yet to offset their greater vulnerability to shocks. This may be
plausible because of the high cost of remaining liquid in the face of the sustained high
inflation rates during the period.
11
In Section III.2, we discuss this issue in detail in the
context of non-financial firms demand for liquid financial assets under macroeconomic
uncertainty.








Sources: Turkey: CBRT, others: Ratha et al., (2003).


The relatively short maturity and high level of corporate debt in Turkey make firms
be vulnerable to insolvency as cash flow must be available for interest payments. Figure
10 reports interest coverage ratio (ICR) measured as the ratio of profit before interest and
tax to interest expenses. ICR can be taken as a proxy for firms ability to service debt as
debt service default risk tends to decrease with higher earnings relative to interest
payments.
12
Firms with an ICR below 100% can be interpreted as being not able to
generate sufficient cash flow to service their short-term debt. Note that this does not

11
Bleakley and Cowan (2004) suggest a positive relationship between short-term indebtedness and
liquidity and show that while East Asian firms tended to have more short-term debt than Latin American
companies, their short-term liabilities were generally matched with larger holdings of liquid assets.
However, the opportunity cost of liquidity, which is ignored by Bleakley and Cowan (2004), may not be
the same for the countries in the presence of substantially differing inflation rates.
12
We consider the ICR definition of the CBRT Company Sector Accounts. See Claessens, Djankov and
Nenova (2001) and Glen (2004) for alternative interest coverage ratio definitions and comparable data for a
broad number of countries including Turkey.
Figure 9. ST Liabilities (% of Total)
An International Comparison
0
10
20
30
40
50
60
70
80
90
100
East Asia L. America Eastern
Europe
Turkey
1995
1997
2001
2004
20
necessarily imply a debt default as firms can temporarily use alternative sources of cash
via asset sales or new security issues etc. (Glen, 2004). As a Ponzi type debt finance is
not sustainable for a firm, an ICR of at least 100%, however, is ultimately needed for the
solvency of a firm.
According to Figure 10 interest coverage ratios tend to improve with firm size. The
ICRs for small and medium sized manufacturing industry firms were at critical levels
during the financial crisis year of 2001. The small (and to a certain extend medium) sized
manufacturing firms, on average, can be interpreted as being highly vulnerable to
insolvency even years before the crisis potentially due to the high real interest rates during the
period. The value of the ICR for large firms just being around 100% may suggest that also
a sizeable number of large firms found themselves facing a debt service difficulty during
2001. All firm groups generally tend to improve their financial positions after the
financial crisis of 2001. The post crisis adjustment of the non-manufacturing and large
manufacturing firms appears to be remarkable. However, relatively sluggish post crisis
adjustments of the small and medium sized manufacturing firms make them still highly
vulnerable to interest rate shocks. With an ICR of just around 100%, the interest rate risk
exposure of the small firms can be interpreted as still extremely severe by 2004.





















Figure 10. Interest Coverage Ratio (ICR), %.
0
100
200
300
400
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
All NF Firms M_Small M_Medium
M_Large Manuf.
21
II.4.2. Currency Risk and Liability Dollarisation
The presence of liability dollarisation can make firms financially vulnerable to
external shocks through balance sheet currency/maturity mismatches and limit the scope
of macroeconomic policies especially by causing fear of floating as widely discussed in
the recent literature.
13
Furthermore, as noted by IMF (2005, p. 116) both currency and
maturity mismatches can exacerbate the impact of exogenous shocks in emerging
markets, increase the severity of crises, and slow down the post crisis adjustment.
14
The
financial fragility is expected to be more severe for low-exporting non-tradable firms
which are highly leveraged in foreign currency debt although their revenues are primarily
in domestic currency.
Figure 11 plots the ratio of foreign currency (FX) debt to total debt
(FX_Debt/T_Debt) as a measure of corporate sector liability dollarisation in Turkey. The
figure also presents the ratio of short-term FX debt to total FX debt as a proxy of FX debt
maturity mismatch. The corporate sector liability dollarisation, which was already high in
1992, sharply increased during 1992-1996 reaching a level of around 70% in 1996. After
1996, the dollarisation ratio fluctuated slightly and reached local peaks with the
implementation and collapse of the fixed exchange rate based stabilisation policy after
1999. The relative improvement of the macroeconomic conditions after 2001 appears to
be effective in reducing liability dollarisation around to a level of 1996, albeit which is
still a very high one. The data for 2005 are yet to be available, but the sustained and

13
For the recent accounts, see Cspedes, Chang and Velasco (2004), Calvo, Izquierdo and Meja (2004),
Cowan, Hansen and Herrera (2005), IMF (2005) and the references cited therein.
14
As acknowledged also by the IMF (2005, p.116) the presence of original sin (the inability of most
countries to borrow internationally in their own currencies) indeed prevents both emerging market
sovereigns and corporates from issuing domestic currency debt abroad. The inability of many countries to
borrow in domestic currency at long maturities and fixed rates even at home constitutes the domestic
dimension of the original sin (Eichengreen, Hausmann and Panizza, 2003). An important consequence of
the original sin is neatly summarized by IMF (2005, p. 116): an emerging market firm that is unable to
obtain long term funding locally faces a trade-off between financing long-term investments with short term
local currency liabilities, which creates a maturity mismatch, or borrowing long-term in foreign currency,
which creates a currency mismatch. The results by zmen and Arnsoy (2005, p. 599) suggest that
flexible exchange rates and strong macroeconomic policy stance with sound institutions are necessary but
not sufficient for redemption from original sin. Consequently, it may be argued that better governance and
macroeconomic stability with a flexible exchange rate regime can plausably have a role in decreasing
maturity and currency mismatches in developing countries.



22
improved macroeconomic stability during the period can plausibly be expected to yield a
relatively lower level of corporate sector liability dollarisation. The bulk of the FX debt
(more than 80%) appears to be short-term until 2000. The relative improvement of the
FX debt maturity with the stabilisation policy of 2000 seems not to be substantially
distorted even with the financial crash of 2001 potentially due to the credibility of the
post-crisis stabilisation programme. Although there have been some signs of
improvements, the maturity structure and especially the level of corporate sector liability
dollarisation can be interpreted still as a source of concern leading firms vulnerable to
external shocks.
Figure 11. Corporate Sector Liability Dollarization
in Turkey (%)
0
10
20
30
40
50
60
70
80
90
100
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
FX_Debt/T_ Debt (All Firms) FX_Debt/T_Debt (Man. Ind)
ST FXD/T_FXD (All firms) ST_FXD/T_FXD (Man. Ind)


The corporate sector liability dollarisation in Turkey can be interpreted as extremely
high when compared internationally. According to IMF (2005a, p. 118), the 1999-2003
averages of the corporate sector FX debt as percentage of total debt are 33.6 for Latin
America, 23.0 for Asia, 20.4 for Europe and 25.7 for all emerging market countries in the
sample. The Latin American countries appear to have the highest liability dollarisation
ratio. To have a better international comparison, we consider the Inter-American
Development Bank (IADB) database, which is used also a basic source for IMF (2005a).
The IADB database provides firm-level information for approximately 2,000 non-
23
financial firms from ten Latin American countries for the period of 19902002.
15
Figure
12 plots the liability dollarisation ratios (FX debt as a percent of total debt) for non-
financial firms in the Latin America and Turkey. Turkey (along with Uruguay) appears to
be amongst the most dollarised countries whilst the liability dollarisation for Colombia,
Chile and Brazil can be interpreted as relatively modest.
16
The liability dollarisation tends
to be relatively persistent for most of the countries during the period. Consistent with the
argument in which fixed exchange rate regimes encourage dollarisation, the countries
with hard pegs (Argentina) and de facto (Reinhart and Rogoff, 2004) crawling pegs
(Bolivia, Costa Rica, Peru, Uruguay and Venezuela) are generally more dollarised
compared to the countries with floating exchange rate regimes (Brazil, Chile, Colombia)
over the sample period. It is worth noting that the countries with lower dollarisation ratios
are also the countries enforced strict regulations on financial transactions in foreign
currency (Brazil, Chile, Colombia and to a certain extent Mexico, see Singh et al., 2005
Chapter VI for financial dollarisation and regulations in Latin America). Therefore, the
impact of exchange rate regime inflexibility on dollarisation should better be interpreted
with a caution. The Turkish experience, in this context, is more unique as corporate
sector liability dollarisation has been the highest in spite of substantial shifts in exchange
rate regimes over the period. This may indeed show also the importance of strong
macroeconomic policy stance and price stability for an endogenous dedollarisation
process (Galindo and Leiderman, 2005) along with precautionary/regulatory measures to
limit vulnerabilities caused by dollarisation.
17

.

15
See Kamil (2004) for a detailed information on the IADB database. The December (2003) issue of the
Emerging Markets Review is entirely devoted to studies using the IADB database (see, Galindo et al., 2003
for a review). The database covering around 2000 firms for 10 countries is one of the most comprehensive
source for firm level liability dollarisation capital structure. It is worth noting that the CBRT Company
Sector database covers a wide range of information for around 8000 firms annually after 1990.
16
Note that, as Kamil (2004) warns that the number of firms for Uruguay is very small (less than 30 for
most of the years) and thus may not be clearly representative. The small sample size problem is the case
also for Venezuela and Costa Rica.
17
See Kesriyeli et al. (2005) for the causes and balance sheet consequences of the liability dollarisation of
non-financial sectors in Turkey. The results by Kesriyeli et al. (2005) suggest that both sector-specific and
macroeconomic condition variables are significant in explaining the corporate sector liability dollarisation.
Firms are found to match only partially the currency composition of their debt with their income streams
making them potentially vulnerable to negative balance sheet affects of real exchange rate depreciation
shocks.
24

II.5. Corporate Sector Vulnerabilities and the Post-Crisis Adjustment
The analysis so for suggests that the non-financial firms in Turkey has been heavily
exposed almost all of the basic balance sheet risks. The corporate sector appears to be
excessively leveraged with relatively lower asset tangibility creating also a credit risk for
the lenders (basically banks). The figures for the maturity and currency composition of
the corporate sector debt show that firms rely heavily on foreign currency denominated
and short-term debt instruments in Turkey. Such a liability composition makes firms
vulnerable to both exchange rate and interest rate shocks through currency and maturity
mismatches. Interest rate increases may potentially lead to a rollover risk and a decline in
the net worth of the firms with higher short-term debt magnifying the conventional
interest rate channel as postulated by the financial accelerator mechanism (Bernanke,
Gertler and Gilcrist, 1999). Real exchange rate depreciations, whilst can potentially make
exporting firms more competitive, can negatively affect the balance sheets of firms that
do not produce tradable goods and thus can not hedge against exchange rate risks
(Aghion, Bacchetta and Banerjee, 2001). The results by Kesriyeli et al. (2005), indeed,
suggest that firms in Turkey match only partially the currency composition of their debt
with their income streams making them vulnerable to negative balance sheet affects of
Figure 12. Liability Dollarization in Latin America and
Turkey (%)
0
10
20
30
40
50
60
70
80
90
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
Turkey Argentina Bolivia Brazil
Chile Colombia Costa Rica Mexico
Peru Uruguay Venezuela
25
real exchange rate depreciation shocks. The high level of leverage ratios (or, the
relatively higher debt-equity ratios reflecting a capital structure mismatch) in Turkey
during the period can be expected to be an amplifying factor for the financial
vulnerability arising from currency and maturity mismatches.
Consistent with the recent balance sheet approach to financial crisis, the fragilities
of the corporate sector are found to be crucially important in triggering and determining
the output cost of the 2001 crisis in Turkey (Roubini and Setser, 2004; Koar and
zmen, 2005; and Keller and Lane, 2005). In this context we consider the developments
in the basic balance sheet items that potentially reflects fragility before, during and after
the 2001 financial crisis. The diamond-shaped chart plotted in Figure 13 summarises the
LRs, interest coverage risk ratio (ICRR = 100/ICR), liability dollarisation (FXD/TD) and
short-term debt (Short Term Debt/Total Debt) ratios (all are in percentage form) as
indicators for manufacturing firms fragility during 1996-2004.
18


Figure 13. Corporate Sector Vulnerabilities
0
100
Interest Coverage Risk
Liability Dollarisation
Leverage
Short Term Debt
1996-2000 2001 2002-2004




18
The ratios for the other non-financial firms showed a simmilar pattern with those for the manufacturing
industry firms, therefore not plotted to save the space.
26
The interest coverage risk increased sharply with the financial crisis of 2001 owing
both to the high leverage ratios and the jump in the interest rates. The risk decreased
sharply after 2001 even much below to the 1996-2000 average with the decline in
nominal and real interest rates. The leverage ratios improved only slightly during 2002-
2004 compared to the earlier periods. The major improvement in the LRs was observed
in 2004 as discussed earlier in the context of Figure 2. Liability dollarisation and short-
term debt maturity, however, tend to be persistent even after the 2001 crisis making firms
still extremely exposed to currency and interest rate risks. The monetary policy
credibility with price stability and better macroeconomic stance after the 2001 crisis has
been influential in reducing the LRs and ICRR and thus the interest rate and debt rollover
risks considerably. Currency composition of debt and its maturity which are potentially
sourced from the inertia of liability dollarisation and the relatively lower level of
financial intermediation, respectively, however still remain as important sources of risk
for the non-financial firms in Turkey.

III.1. Domestic Public Debt Finance and Financial Crowding-Out
A well-developed government securities market is often considered as helpful for
the development of a corporate bond market as it can provide the necessary market
infrastructure and investor base along with a reliable benchmark yield curve (IMF,
2005a,b). Government bonds and securities, especially those with short maturities,
provide liquidity services and can be used as financial collateral. As Woodford (1990)
and Holmstorm and Tirole (1998) convincingly show government debt as net wealth may
thus crowd-in private investment by relaxing liquidity constraint in non-Ricardian
economies with imperfect financial intermediation. However, these beneficial effects of
the government debt finance instruments may not be invariant to financial depth and to
the debt level (thus the sustainability) and the through which public sector borrows. High
levels of government borrowing from domestic markets may drain limited sources for
investment. The impact of such a financial crowding out might be more severe for bank
dependent firms when the public heavily borrows from the commercial banks.
19
In an

19
Financial deepening and banking system development can potentially limit the crowding out affect
(Caballero and Krishnamurthy, 2004). However, as Kuttner and Lown (1999) shows, bank holdings of
27
environment where financial markets are relatively shallow, over-barrowing by the public
sector may easily crowd out private sector activity as substantially large shares of public
debt in the domestic financial system may reduce the overall liquidity as a result of
increasing the country risk premium and thus reducing capital inflows (Caballero and
Krishnamurthy, 2004). In such a case, fiscal contractions can be expansionary as they
alleviate the credit constraint of firms.
Considering the overall benefits and costs of the government domestic debt finance,
it may be argued that there is an endogenous threshold beyond which an increase of it
may cause financial crowding out. The threshold is expected to increase with financial
deepening, the development of a corporate sector bond market, better governance and
sound macroeconomic policy stance. Consistent with the recent debt intolerance
arguments (Reinhart, Rogoff, and Savastano, 2003), the threshold is expected to be lower
in developing countries. Beyond the threshold level of the debt finance, expansionary
fiscal policies can indeed be contractionary as it can lead to financial crowding out and
even to a lower financial depth
20
.
The low levels of financial intermediation and the banking system preference to
finance corporate sector are already discussed as amongst the major causes of the firms
financial constraints in Turkey. Bank-based financial system in Turkey that is not deep
compared to those of the countries with similar development levels (Figure 6 in Section
II.3), appears to prefer to finance public deficits with its rather limited sources. Relatively
low share of bank loans in liabilities of corporate sector may be attributed to financial
crowding out of government debt finance via the commercial banking system in Turkey.
In Turkey, high and persistent fiscal deficits have often been interpreted as one of
the major sources of macroeconomic instability and chronic inflation sustained for more
than three decades. Before the financial crisis of 2001, the share of the public debt in
GDP was not very high with total debt (external and domestic) reaching just around 50%

public debt tend to displace lending to the non-bank private sector even in a country like the US with well
developed financial markets.
20
This can be interpreted as being perfectly consistent with the expansionary fiscal contractions
arguments in the literature. See Giavazzi et al., 2000 for a survey and Favero and Giavazzi, 2004 and
zatay, 2005 for the Brazilian and Turkish evidence, respectively.
28
in 2000
21
. However, the basic problem has been its mode of financing. After the financial
liberalization programme of 1980, financing domestic debt via deposit money banks
(DMBs) has become the major mode, especially after the mid-1980s. During the period,
DMBs absorbed around 80-90% of the Treasury cash security issuances in the primary
market. Given the relatively lower level of financial depth, the growth of the public debt
and the consequent heavier reliance on domestic debt finance through the banks after the
late 1990s have yielded high real interest rates and extremely short debt maturities, thus
has led to an interest payments explosion
22
.
Figure 14 plots the shares of government securities (PUB/TA) and credits to private
sector (PRIV/TA) in total assets (TA) of deposit money banks (DMBs) in Turkey during
1990-2004. The figure plots also the ratios of the DMBs government securities
(PUB/GDP), private sector credits (PRIV/GDP) and total assets (TA/GDP) to GDP
during the period. Note that, the substantial rise in the government securities share in
2001 is due to the fiscal cost of the financial crisis arising from government contingent
liabilities and bank bail outs.


21
With the 2001 financial crisis, however, public debt increased sharply due to the public sector contingent
liabilities and the realisation of the fragilities of the banking system under government bailout guarantees as
prospective public sector deficits (Koar and zmen, 2006). The values for the total public debt stock
(domestic debt) as a percent of GDP were 102 (69) in 2001, 90 (55) in 2002, 80 (55) in 2003 and 75 (53) in
2004.
22
Interest payments on domestic debt (as a percent of GDP) increased from 2.4 in 1990 and to 10.6 in 1998
and to 15 in 2000. After the crises it reached 23 per cent of GDP by the end of 2001 and gradually
declining thereafter with around 19 % in 2002, 16 % in 2003 and 13 % in 2004. See zatay (1997) and
zmen and Koar (1998) for the earlier studies on the Turkish public sector deficits finance before the
crisis. zatay and Sak (2003) and Koar and zmen (2006) are among the studies discussing the role of
the mode of the domestic debt finance in triggerring the financial crisis of 2001.
29
Figure 14. Government Securities and Private Sector Credits
Portfolios of DMBs (%)
0
10
20
30
40
50
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
0
20
40
60
80
100

T
A
/
G
D
P
PUB/GDP PRIV/GDP PUB/TA
PRIV//TA TA/GDP

Source: CBRT
By definition, over-borrowing of the public sector is expected to financially crowd
out private sector credit if it is not accommodated with a corresponding growth of
banking system assets. From the figure, it may be inferred that, the government
borrowing from the commercial banking system helped indeed financial deepening until
the mid 1990s. This may be plausible given the fact that the government borrowing from
DMBs was relatively moderate and the monetary policy stance was largely
accommodative with no effective control on the growth of the monetary aggregates
during the period. This is indeed consistent with the view that public debt finance
provides liquidity and thus can crowd in private sector investment by relaxing the
liquidity constraint under market imperfections (Woodford, 1990 and Holmstorm and
Tirole, 1998). However, the threshold level of domestic debt finance beyond which it
leads to crowding out appears to be somewhat reached after the mid 1990s. The heavier
reliance on the debt finance through the banks with the growth of public debt has begun
to severely crowd out private credits by the late 1990s. After 1999, the banking system
assets as a percent of GDP tend to remain stagnant (even decline after the crisis) despite
substantial increase in government borrowing.
23
The Turkish commercial banking system

23
This is consistent with the view that, continuously large public sector borrowing from the domestic
banking sector can have substantial adverse implications for financial development (Hauner, 2006, p. 3).
30
appears to allocate around 40% of its assets to public debt finance during the 2000s.
24
Despite the severe fiscal contractions and a credible monetary policy stance with
substantially decreased inflation rates after the 2001 crisis, the fiscal dominance on the
commercial banking system tends to be persistent. This is a crucially important policy
issue for sustainable growth and financial stability since the domestic debt finance
through the DMBs not only creates a direct fragility linkage between the balance sheets
of the public and banking sectors but also leads non-financial corporate sector to be
severely bank credit constrained (and thus financially constrained in the absence of
corporate sector bond markets) due to the financial crowding out.

III.2. Financial Asset Holdings of Non-Financial Firms under Macroeconomic
Uncertainty
Non-financial firms hold financial assets including government securities to hedge
themselves against liquidity and interest rate risks and maturity mismatches.
25
In the
conventional Modigliani and Miller (1958) world with perfect complete markets firms
can raise funds instantaneously to finance their profitable projects and thus they may
have no uncertainty induced precautionary demand for liquidity. However, as Holmstorm
and Tirole (2000) argue, the presence of credit rationing, informational constraints and
moral hazard may increase a demand for liquidity. The most liquid financial asset in the
firms portfolio is their cash balances the demand for which can be explained by the

The decline in the total bank assets to GDP ratio during the 2000s supports also the argument that rising
share of public debt to private assets reduces financial depth (Caballero and Krishnamurthy, 2004, p. 1)
24
This is, indeed, an extremely high ratio when compared internationally. The Turkish experience appears
to be an outlier according to the figures reported in Sy (2005), Hauner (2006) and Hanson (2003). Sy
(2005, Table 1) reports the 2003 values of the shares of government securities in total banking system
assets for a number of developing and industrial countries. The mean share for the 24 country sample is
13% and the Turkish banking system has the highest government securities ratio (37%) representing an
outlier (along with India and Indonesia with the shares 32% and 31%, respectively). In the same vein,
Hauner (2006, Table A1) reports the 2001-2003 average shares of the banks credits to public sector in
their total credits for a sample of 75 developing countries. Accordingly, the credits the public sector
constitute around 65% of total bank credits in Turkey representing one of the highest when compared
internationally. According to Hanson (2003) Turkey appears to have the highest ratio of banking system net
government credit to deposits amongst 25 developing countries with the largest banking systems.
25

Non-financial corporate sector firms hold substantial amounts of liquid financial assets in many countries
(Dittmar et al., 2003 and IMF, 2006). According to IMF (2006), the recent acceleration of the liquid
financial asset holdings of the firms in the G-/ countries is one of the striking changes in the global
financial landscape. According to IMF (2006, p.135), since the early 2000s, companies in many industrial
countries have moved from their traditional position of borrowing funds to finance their capital
expenditures to running financial surpluses that they are now lending to other sectors of the economy.
31
conventional transactions (Tobin, 1958) and precautionary (Miller and Orr, 1966)
motives. Firms cash balances, in this context, reduce transactions costs and provide a
buffer to absorb adverse shocks (Keynes, 1936). Not only cash balances but also cash-
like liquid assets including interest bearing bank deposits and short-term securities can
provide a financial buffer to absorb unexpected changes in transactions and investment
opportunities. Higher uncertainty sourced by macroeconomic instability may increase the
demand for financial assets instead of investing these resources into long-term investment
projects. In this context, the firms demand for financial assets in general may be
considered as due to their transactions/precautionary-cum-speculative motives.
26

Figures 15 and 16 plot the shares of government securities and total financial assets
(bank deposits, government securities and repurchase agreements) in current assets
during 1996-2004, respectively.
27
The shares of government securities (GS) and financial
assets (FA) tend to increase with firm size in manufacturing industry. The firms holding
of government securities tend to decline gradually during the period until the 2001
financial crisis. After the crisis, the share of government securities in current assets
appears to be remained stable at relatively lower levels. The firms holdings of financial
assets, on the other hand, tend to be stable during the whole period. Especially large sized
manufacturing firms invest more heavily in financial assets.


26
There is now a growing body of theoretical and empirical literature attempting to explain non-financial firms
demand for financial assets and liquidity. The recent contributions include Opler et al. (1999), Holmstorm and Tirole
(2000), Dittmar et al. (2003), Almeida et al. (2004), zkan and zkan (2004), IMF (2006) and Baum et al. (2006).
Kaplan et al. (2006) provide a recent survey of the literature and an empirical application to the Turkish case.
27
The share of current assets in total assets typically constitutes around a half of the total assets with exhibiting
negligible variation across firm groups and time periods. Therefore, dividing the numbers in Figures 15 and 16 by
two gives a broad measure proximating the shares in terms of total assets.
32
Figure 15. Government Securities/Current Assets, %.
0
2
4
6
8
10
1996 1997 1998 1999 2000 2001 2002 2003 2004
All NF Firms M_Small M_Medium
M_Large Manuf.


The FA holding of the manufacturing firms in Turkey is not low when compared
internationally.
28
However, the firms holdings of cash (typically less than 0.5% of their
current assets) have been minimal during the period. Under chronic high inflation rates,
economic agents, including non-financial firms, can be expected to minimise their cash
(and non-interest bearing demand deposits) holdings.
29
A sustained severe inflationary
process in a country may not only preclude domestic fiat money demanded as a store of
value, but may also reduce its role as a medium of exchange with the availability of
alternative liquid financial assets which can be used as an inflation hedge whilst
providing liquidity to a certain extent. This may plausibly explain the minimal holdings

28
For example, Baum et al. (2006) report that US and Germany corporations hold around 10% and 6% of
their total assets in liquid financial assets (cash and marketable securities), respectively. Dittmar et al. (2003)
consider a cross-section of firms from 45 countries and find that the median ratio of liquid financial assets to
net assets (total assets minus cash and marketable securities) is 6.6 %. The median liquid financial asset ratio
reported by Dittmar et al. (2003) for some selected countries are as follows: 3.1% (Chile), 6.4% (US), 7.3%
(Brasil), 7.4% (Germany), 8.1% (UK), 11.1% (France), 13.4% (Turkey), 15.5% (Japan), and 20.9% ( Israel).
In the same vein, Himmelberg et al. (2003) consider a cross-section of firms from 27 European countries and
find that the mean (median) ratio of liquid financial assets to net assets is 18.0 % (6.4%). The data sets by
Dittmar et al. (2003) and Himmelberg et al. (2003) both show that the liquid financial asset ratio varies
widely across (and within) countries. This suggest that there may be no optimal liquidity ratio for
nonfinancial firms invariant to industry/firm specific characteristics and the prevailing policy stance in the
country. A companion paper, Kaplan, zmen and Yaln (2006), empirically investigate the causes and
consequences of the financial asset holdings of the non-financial firms in Turkey.
29
There is no data for the firms holding of demand deposits with banks. However, it may be plausably
expected that the share of domestic currency denominated demand deposits is minimal as for the cash
holdings under the severe inflationary period.
33
of cash by the firms. However, the fact that the FA holdings of the Turkish firms is
roughly comparable with those for the countries enjoying much lower inflation and
stronger macroeconomic policy stance needs a further explanation.
Figure 16. Financial Assets/Current Assets, %.
0
5
10
15
20
1996 1997 1998 1999 2000 2001 2002 2003 2004
All NF Firms M_Small M_Medium
M_Large Manuf.

The liability structures of the firms, including maturity and currency composition of
their debt, are among the important determinants of their demand for liquid assets. A
short-term debt dominated structure forces the firm to be more liquid whilst long-term
debt allows it to be more flexible against liquidity shocks. A firm whose liabilities is
made up of mainly foreign currency denominated short-term debt and whose earnings are
mainly in domestic currency tends to liquidate FX assets against a currency risk. As
already discussed in the earlier sections, the bulk of the corporate sector firms debt in
Turkey have been short term and denominated in foreign currency that make firms
vulnerable against shocks. The opportunity costs of remaining liquid by holding TL cash
and TL denominated demand deposits have been very high in the face of the sustained
high inflation rates until very recently. Under these conditions, firms may need to hold
alternative liquid financial assets to avoid underlying risks.
Given the fact that the maturity of financial contracts, including government
securities and banking system time deposits, has been extremely short in Turkey (Koar
and zmen, 2006), non-financial firms tend to hold interest bearing FA also for
satisfying their liquidity needs. The Turkish banking system is heavily dollarised with FX
34
denominated deposits constituting around a half of the total deposits during the period
(Ylmaz, 2005 and Aknc, Barlas-zer and Usta, 2006). Consequently, non-financial
firms in Turkey have been able to hedge themselves against currency risk to a certain
extent and to remain relatively liquid by holding FX deposits with the banking system.
The firms demand for liquid assets depends also on the degree of financial
constraints that they face up with (Opler et al. 1999 and zkan and zkan, 2004).
Financially more constrained firms are expected to hold more liquid assets for
precautionary reasons as it may be more difficult for them to borrow when needed. The
degree of a financial constraint may decrease with firm size as large firms can have
access to capital markets and face a lower degree of asymmetric information problems
(Myers and Majluf, 1984). Consequently, we may expect that small firms tend to hold
more liquid assets as a financial buffer to hedge against negative shocks (the
precautionary-cum-transactions motive). However, the Turkish evidence suggests that the
FA holdings tend to increase with firm size in manufacturing industry which is not
consistent with these arguments (Figure 16). This contrasting evidence may, however, be
justified if FA are viewed not only as financial buffers against liquidity shocks but also as
a portfolio choice substituting fixed investments.
Non-financial firms face a choice between allocating their resources into fixed and
financial investments (Vickers, 1987 and Holmstorm and Tirole, 2000). By providing the
necessary liquidity services due to the transactions-cum-precautionary motive, liquid
financial assets can be complementary to fixed investments. In this context, the holding
of liquid financial assets including government securities can crowd in fixed investments
(Woodford, 1990 and Holmstorm and Tirole, 1998). However, under macroeconomic
instability and thus high uncertainty, non-financial firms may prefer to defer fixed
investments and hold financial assets also for their speculative motive. In such a case,
financial assets and fixed investments may become substitutes leading the former to
crowd out the latter.
30,31


30
Such a financial crowding-out behaviour under uncertainty is neatly propsed by Vickers (1987): Money
may be held when the uncertainties surrounding economic prospects make it desirable to defer the
commitment of resources to real investment and the pursuit of real economic activities. To the extent that this
is so, available real resources will not be utilized as fully as would othervise be possible (p. 11). In the same
vein, Ersel and Sak (1997) propose the notion of uncertainty induced liquidity preference to explain
35
The fact that large manufacturing firms hold substantially more (about the twice)
financial assets than the small and medium sized firms in Turkey (Figure 16) is consistent
with the view that firms may hold financial assets not only for their liquidity services but
also for high return from financial assets under macroeconomic uncertainty. This
speculative motive appears to be the case especially for large sized manufacturing firms.
Small and medium sized manufacturing firms, which are more financially constrained,
tend to be relatively less flexible for holding financial assets for interest income apart
from their liquidity.

Figure 17. Interest Income/Operating Profits, %.
0
10
20
30
40
50
60
1996 1997 1998 1999 2000 2001 2002 2003 2004
-10
0
10
20
30
40
R
e
a
l

I
n
t
e
r
e
s
t

R
a
t
e
s
All NF Firms M_Small M_Medium
M_Large Manuf. R_Int




corporate sector holding of liquid financial assets including government securities and FX assets as a
financial buffer under conditions of enhanced uncertainty. Accordingly, the distribution of the working
capital between production related assets and financial assets depends upon perceived risks over the
production cycle of the corporation (p.4). The financial crowding out, according to Ersel and Sak (1997), is
temporary as firms transfer the accumulated financial assets to finance real investments to the next production
cycle. The empirical results by Ersel and Sak (1997) support the uncertainty induced liquidity preference
hypothesis for the Turkish data and suggest that non-financial firms holding of government securities not only
cushioned the impact of the 1994 crisis but also allowed them to have a faster post-crisis recovery.
31
The argument about the crowding out affect of the financial assets holdings of non-financial firms may
also be relevant for the recent US experience. According to IMF (2006, p. 136), the recent acceleration of the
corporate sector holdings of financial assets has offset one-half of the increase in government and household net
borrowing, thereby helping to mitigate the impact on the external deficit.
36
Figure 17 plots the ratio of interest income to operating profits of the non-financial
firms during the period along with ex ante real interest rates (R_Int) on government
securities.
32
Not surprisingly, the interest income ratio for all firm groups tends to follow
a similar path with the real interest rates. Consistent with their financial assets holding
behaviour as depicted by Figure 16, the interest income ratio appears to increase with
firm size in manufacturing industry. During most of the period, their heavier investment
on financial assets yielded large sized manufacturing firms to obtain substantial interest
income reaching about 45% and 35% of their operating profits with the jump of the real
interest rates in 1999 and 2001, respectively. The interest income ratios for all the firm
groups tend to decline gradually after the 2001 financial crisis. This may plausibly
interpreted as resulting from the decline in macroeconomic uncertainty and thus in the
real interest rates with enhanced monetary policy credibility and better macroeconomic
stance after the 2001 financial crisis.
Figure 18 shows the ratios of interest income and net profits to net sales of the non-
financial firms. From Figure 18a, it may be strikingly inferred that there was a structural
change in the manufacturing industry income and profit generating activity behaviour
after the 2001 financial crisis. Before the 2001 crisis, interest income appears to be the
major source of manufacturing firms profits. Prior to the crisis, the profit ratios of
manufacturing firms had been gradually decreasing whilst their interest income had been
on an increasing trend. After 2001, compared to the pre-crisis period, we observe an
opposite path for the evolutions of interest income and profit ratios with a gradual decline
in interest income while an increase in profit ratios. This observation is consistent with
the view that income from real investments rather than that from financial asset holdings
has become the major source of revenue for manufacturing firms as macroeconomic
stability came into the picture during the post crisis period.
The profit and interest income ratios for the large sized manufacturing firms (Figure
18b) follow a similar path with those of the manufacturing industry in general suggesting
that the latter is indeed dominated by the former. The substantial amount of interest

32
It may be preferable to consider alternative measures such as net profits or profits before taxes. However,
for some observations these alternative measures have negative values precluding a meaningful interpretation
of the ratios based on them.
37
income obtained by large sized firms during the financial crisis of 2001 appears to be
effective in muting the unfavourable output impacts of the crisis for these firms. Based on
the corporations quoted at the Istanbul Stock Exchange, Ersel and Sak (1997) find that
financial asset holdings of firms cushioned the impact of the 1994 crisis and enabled
them to grow in 1995. Financial assets can be interpreted as to have a similar buffering
role especially for large sized manufacturing firms also during the 2001 crisis. The
financial asset holdings of small, and to a certain extend medium, sized manufacturing
firms have been relatively modest (Figure 16) leading to their profitability basically
determined by macroeconomic and firm/sector specific conditions. Small and medium
sized manufacturing firms with rather limited financial flexibility can be expected to hold
liquid financial assets mainly for transactions and precautionary purposes. Large sized
firms, on the other hand, are more flexible in allocating their assets into financial and real
investments. This flexibility allowed large firms to hold also speculative motive led
financial assets potentially crowding out their real investments especially before the 2001
crisis.
38
Figure 18. Interest Income and Net Profits (% of Net Sales)
Figure 18a. Manufacturing Industry
-1
0
1
2
3
4
5
6
1996 1997 1998 1999 2000 2001 2002 2003 2004
Int_Sales Prof_Sales

Figure 18b. Large Firms
0
1
2
3
4
5
6
7
1996 1997 1998 1999 2000 2001 2002 2003 2004

Figure 18c. Medium Firms
-6
-4
-2
0
2
4
6
1996 1997 1998 1999 2000 2001 2002 2003 2004


Figure 18d. Small Firms
-8
-6
-4
-2
0
2
4
1997 1998 1999 2000 2001 2002 2003 2004


39
IV. Concluding Notes
The balance sheets of firms contain valuable information that enables us to assess
potential vulnerabilities in the corporate sector and to understand the transmission
mechanism through which macro policies including monetary policy transmit into
financial and real activity of firms. One may expect that non-financial firms tend to adopt
financial strategies through which they could survive in an environment of long lived
macroeconomic instability, less developed financial market and institutions. In such an
environment, non-financial firms in Turkey have been heavily exposed to balance sheet
risks. Therefore, firms tended to avoid investing into capacity creation activity instead
they preferred to invest heavily in financial assets over 1990s. Eventually, potential
output of Turkey declined considerably in Turkey. We pay attention to some stylized
facts that reflect the basic structure of the balance sheets of firms in this study.
Turkish firms can be considered as highly indebted on average even though they do
not get finance directly from market by issuing bonds. In other words, contrary to
financially developed countries, trade credits and credits from subsidiaries or parents
constitute a very large portion of corporate sectors external finance. The share of bank
loans in total liabilities in Turkey is almost the same that of other emerging markets that
have a substantial amount of bond finance. This may imply that potential bank dependent
firms which are generally made of small sized companies, are less likely to have access to
bank finance in Turkey, that is, they are more financially constrained consistent with low
degree of financial deepening in Turkey. In fact, the share of bank finance for small firms
is lower than that of large and medium sized firms and the share of small firms declines
further when economy contracts. In general, small firms rely heavily on trade credits
especially during the recessions while large firms tend to use more bank loans and
internal funds. Trade credits are expected to mute the fluctuations of small firms activity
especially in bad times. Extensive use of trade credits by small firms allows large firms as
intermediary institutions. In fact, heavily investing in liquid assets, relatively easy access
to bank finance and operating with high profit margins in relatively less competitive
market conditions allow large firms to provide finance to their sub-contractors or small
firms in trade activity in form of trade credit.
40
We observe a marked shift in the finance structure of corporate sector in Turkey
after the crisis. All firms, predominantly large ones, tend to use more internal funds in
financing their real activity after the 2001 crisis. Similarly, in the same period, especially
small and medium sized firms invest more in tangible assets whose share in total assets in
corporate sector on average is very low compared to that of other emerging and
developed markets. The increase in the share of tangible assets of small and medium
sized firms reaches that of large firms that has been historically higher in 2004.
The maturity structure across firms classified in terms of size has stayed stable over
time. The maturity of debt is increasing with firm size as expected. The firms with high
tangible ratios have high share of long term debt. In fact, we observe a gradual decline in
the share of short term debt for small and medium sized firms as the shares of tangible
assets increase in their balance sheets. Debt maturity of corporate sector in Turkey
appears to be very short compared to those of emerging economies. This is believed to be
stemmed mainly from macroeconomic instability reflected in form of high inflation and
therefore Turkish corporate sector is more vulnerable to external shocks. Therefore,
Turkish corporate sector tend to stay more liquid compared to countries with longer debt
maturity even Turkey experiences high inflation during the period. One may easily notice
the risk of default with this debt structure when look at interest coverage ratios that
realized below 100 percent for small and medium sized firms in the 2001 crises. The
postions of all firm groups have improved substantially after the crisis while the
improvement is more considerable as firms get large. However, we observe a
deteriorating trend in the interest coverage ratio for small firms in 2004.
The figures for the maturity and currency composition of the corporate sector debt
show that firms rely heavily on foreign currency and short-term debt instruments in
Turkey. Such a liability composition makes firms vulnerable to both exchange rate and
interest rate shocks through currency and maturity mismatches. Interest rate increases can
lead to a rollover risk and a decline in the net worth of the firms with higher short term
debt magnifying the conventional interest rate channel as postulated by the financial
accelerator mechanism. Real exchange rate depreciation, whilst it may potentially
improve exporting firms more competitive, can negatively affect balance sheets of non-
exporting firms.
41
Although in case of the lack of private bond market, government securities may be
helpful in functioning of financial market by providing liquidity, the high level of
government borrowing from domestic market tend to drain funds and thus crowd out
private investment in Turkey. Bank dependent firms are expected to be severely affected
from this structure that has been the main source of macroeconomic instability. However,
the declining trend of budget deficits as a result of tight fiscal policy and the
improvement in the macroeconomic conditions including substantial achievements
towards price stability may be expected to contribute to the corporate sector in achieving
a stronger balance sheet structure. This may, in turn, be expected to be enhancing
macroeconomic stability and sustainable growth.


42
REFERENCES
Aknc, ., Barlas-zer, Y. and B. Usta (2005) Dolarizasyon Endeksleri: Trkiyedeki
Dolarizasyon Srecine likin Gstergeler, TCMB Aratrma ve Para Politikas Genel
Mdrl alma Teblii No. 05/07.
Aghion, P., Bacchetta, P., and A. Banerjee (2001) Currency Crises and Monetary Policy in an
Economy with Credit Constraints, European Economic Review, Elsevier, Vol. 45(7), 1121-
1150.
Aghion, P., Bacchetta, P., and A. Banerjee (2004) A Corporate Balance Sheet Approach to
Currency Crises, Journal of Economic Theory, Elsevier, Vol. 119(1), 6-30.
Allen, M., Rosenberg, C., Keller, C., Setser, B. and N. Roubini (2002) A Balance Sheet Approach
to Financial Crisis, IMF Working Paper 02/210.
Allen, F., Chui, M.K.F and A. Maddaloni (2004) Financial Systems in Europe, the USA, and
Asia, Oxford Review of Economic Policy, 20(4), 490-508.
Almeida, H., Campello, M. and M.S. Weisbach (2002) Corporate Demand for Liquidity, NBER
Working Paper No. 9253.
Baum, C.F., aglayan, M., zkan, N. and O. Talavera (2006) The Impact of Macroeconomic
Uncertainty on Non-Financial Firms Demand for Liquidity, Review of Financial Economics
(in press).
Beck, T., Demirg-Kunt, A. and R. Levine (2000) A New Database on Financial Development
and Structure, World Bank Economic Review, September, 597-605.
Beck, T., Demirguc-Kunt, A., Laeven, L. and V. Maksimovic (2005) The Determinants of
Financing Obstacles, Forthcoming, Journal of International Money and Finance.
Beck, T., Demirguc-Kunt, A., Laeven, L. and V. Maksimovic (2005a) Financial and Legal
Constraints to Growth: Does Firm Size Matter? The Journal of Finance, 60 (1), 137-177.
Bernanke, B, M. Gertler and S. Gilchrist (1999), The financial accelerator in a quantitative
business cycle framework, in: Taylor, J B and M. Woodford (editors), Handbook of
Macroeconomics, Amsterdam: Elsevier Science.
Bleakley, H. and K. Cowan (2004) Maturity Mismatch and Financial Crises: Evidence from
Emerging Market Corporations, Working Paper No. 200416 (San Diego: University of
California).
Booth, L., Aivazian, V., Demirguc-Kunt, A. and V. Maksimovic (2001) Capital Structure in
Developing Countries, Journal of Finance, 56, 87 130.
Burkart, M. and T. Ellingsen (2004) In-Kind Finance: A Theory of Trade Credit, American
Economic Review, 94, 569-590.
Byrne, J.P. and E.P. Davis (2004) A Comparison of Balance Sheet Structures in Major EU
Countries, National Institute Economic Review, 180, 83-95.
Caballero, R.J., and A. Krishnamurthy (2004) Fiscal Policy and Financial Depth, NBER Working
Paper No. 10532.
Calvo, G.A., Izquierdo, A. and L.F. Meja (2004) On the Empirics of Sudden Stops: The
Relevance of Balance Sheet Effects, NBER Working Paper No. 10520.
Cspedes, L.F., Chang, R. and A. Velasco (2004) Balance Sheets and Exchange Rate Policy, The
American Economic Review, 94(4), 1183-93.
Claessens, S., Djankov, S. and T. Nenova (2001) Corporate Risk around the World, in Glick, R.,
Moreno, R. and M. Spiegel, eds., Financial Crises in Emerging Markets, Cambridge, Mass:
Cambridge University Press.
Cowan, K., Hansen, E. and L.O. Herrera (2005) Currency Mismatches, Balance Sheet Effects and
Hedging in Chilean Non-Financial Corporations, Inter-American Development Bank,
Working Paper No. 521.
Davis, E.P. and M.R. Stone (2004) Corporate Financial Structure and Financial Stability, Journal
of Financial Stability, 1, 6591.
43
Demirguc-Kunt, A. and V. Maksimovic (2002) Firms as Financial Intermediaries: Evidence from
Trade Credit Data, Working Paper, World Bank and the University of Maryland.
Dittmar, A., Mahrt-smith, J. and H. Servaes (2003) International Corporate Governance and
Corporate Cash Holdings, Journal of Financial and Quantitative Analysis, 38(1), 111133.
Eichengreen, B., Hausmann, R., and U. Panizza (2003) Currency Mismatches, Debt Intolerance
and Original Sin: Why They Are Not the Same, Why It Matters, NBER Working Paper No.
10036.
Ersel, H. and G. Sak (1997) Corporate Sector Behavior Under Uncertainty: The Case of Turkey
in the 1990s, Economic Research Forum 4th Annual Conference on Regional Trade, Finance
and Labour Markets in Transition, Conference Proceedings, pp. 249-254, September, Beirut,
Lebanon.
Fan, J.P., Titman, S. and G.J. Twite (2004) An International Comparison of Capital Structure and
Debt Maturity Choices, Center for Institutions and Governance Working Paper No. 2005-E-
3-CIG, Available at: http://www.baf.cuhk.edu.hk/research/cig/.
Favero, C. and F. Giavazzi (2004) Inflation Targeting and Debt: Lessons from Brazil, NBER
Working Paper No. 10390.
Fisman, R. and I. Love (2003) Trade Credit, Financial Intermediary Development and Industry
Growth, Journal of Finance, 58, 353374.
Frank, M. and V. Maksimovic (1998) Trade Credit, Collateral, and Adverse Selection, University
of Maryland, Working Paper.
Galindo, A.J., Panizza, U. and F. Chiantarelli (2003) Debt Composition and Balance Sheet
Effects of Currency Depreciations: A Summary of the Micro-Evidence, Emerging Markets
Review, 4(4), 33039.
Galindo, A.J. and L. Leiderman (2005) Living with Dollarization and the Route to
Dedollarization, Inter-American Development Bank Working Paper, May 2005, Available at:
http://www.iadb.org/res.
Gertler, M. and S. Gilchrist (1994) Monetary Policy, Business Cycles and the Behaviour of Small
Manufacturing Firms, The Quarterly Journal of Economics 109(2), 309-340.
Gertler, M., Gilchrist S. and F. M. Natalicci (2003) External Constraints on Monetary Policy and
the Financial Accelerator, Available at: http://www.nyu.edu/econ/user/gertlerm/paper_1.pdf,
October 2003.
Giavazzi, F., Pagano, M. and T. Japelli (2000) Searching for Non-linear Effects of Fiscal Policy:
Evidence from Industrial and Developing Countries, European Economic Review, 44, 1259-
1289.
Glen, J. and A. Singh (2003) Capital Structure, Rates of Return and Financing Corporate Growth:
Comparing Developed and Emerging Markets, in The Future of Domestic Capital Markets,
R. E. Litan, M. Pomerleano and V. Sundararajan, (eds.), Brookings Press, Washington
DC.2003, pp. 373-416.
Glen, J. (2004) Debt and Firm Vulnerability, in M. Pomerleano and W. Shaw, eds., Corporate
Restructuring: International Best Practices, Washington: World Bank.
Hanson, J.A. (2003) Banking in Developing Countries in the 1990s, World Bank Policy Research
Working Paper No. 3168.
Hauner, D. (2006) Fiscal Policy and Financial Development, IMF Working Paper No. 06/26
(Washington: International Monetary Fund).
Himmelberg, C.P., Love, I. and V. Sarria-Allende (2003) Cash Holding at the Firm Level: Can
Transaction Costs Explain it All?, Working Paper.
Holmstrom, B. and J. Tirole (1998) Private and Public Supply of Liquidity, The Journal of
Political Economy, 106(1), 1-40.
Holmstrom, B. and J. Tirole (2000) Liquidity and Risk Management, Journal of Money, Credit,
and Banking, 32, 295-319.
44
IMF (2005a) Global Financial Stability Report: Market Developments and Issues, April, Ch. 4,
Washington DC: International Monetary Fund.
IMF (2005b) Global Financial Stability Report: Market Developments and Issues, September, Ch.
4, Washington DC: International Monetary Fund.
IMF (2006) World Economic Outlook, Globalization and Inflation, April, Ch. 4, Washington DC:
International Monetary Fund.
Kamil, H. (2004) A New Database on Currency Composition and Maturity Structure of Firms
Balance Sheets in Latin America: 19902002, (unpublished; Washington: Inter-American
Development Bank). Available at: www.iadb.org/res/files/databases/CowanKamil.pdf.
Kaplan, C., zmen, E. and C. Yaln (2006) The Determinants and Implications of Financial
Asset Holdings of Non-Financial Firms in Turkey: An Empirical Investigation, Mimeo, The
Central Bank of the Republic of Turkey.
Keller, C. and C. Lane (2005) Balance Sheet Developments Since the Crisis, Chapter IV in
Turkey at the Crossroads: From Crisis Resolution to EU Accession, IMF Occasional Paper
No. 242.
Kesriyeli, M., zmen, E. and S. Yiit (2005) Corporate Sector Debt Composition and Exchange
Rate Balance Sheet Effect in Turkey, Central Bank of Turkey, Research and Monetary Policy
Department, Working Paper No. 05/16.
Keynes, J.M. (1936) The General Theory of Employment, Interest and Money, London: Harcourt
Brace.
Koar, .. and E. zmen (2006) Sectoral Balance Sheet Fragilities and the Turkish Financial
Crisis of 2001, Paper Presented at the Annual Meeting of the ASSA-Middle East Economic
Association, Boston, MA., January, 6-8, 2006.
Krugman, P. (1999) Balance Sheets, Financial Crises, and the Transfer Problem, Available at:
http://web.mit.edu/krugman/www/whatsnew.html.
Kuttner, K. and C.S. Lown (1999) Government Debt, the Composition of Bank Portfolios, and
the Transmission of Monetary Policy, in K.A. Chrystal (ed.), Government Debt Structure and
Monetary Conditions, London: Bank of England, pp. 165-189.
Miller, M.H. and D. Orr (1966) Model of the Demand for Money by Firms, Quarterly Journal of
Economics, 80, 413-435.
Modigliani, F. and M.H.Miller (1958) The Cost of Capital, Corporation Finance and the Theory
of Investment, American Economic Review 48(3), 261297.
Myers, S. and N. Majluf (1984) Corporate Financing and Investment Decisions when Firms Have
Information that Investors Do Not Have, Journal of Financial Economics 13, 187-221.
Nivorozhkin, E. (2005) Financing choices of firms in EU accession countries, Emerging Markets
Review, 6, 138169.
Obstfeld, M. (2004) Globalization, macroeconomic performance, and the exchange rates of
emerging economies, Bank of Japan Monetary and Economic Studies, 22(S-1), 2955.
Opler, T., Pinkowitz, L., Stulz, R. and R. Williamson (1999) The determinants and Implications
of Cash Holdings, Journal of Financial Economics, 52, 346.
zatay, F. (1997) Sustainability of Fiscal Deficits, Monetary Policy and Inflation Stabilization:
The Case of Turkey, Journal of Policy Modeling, 19(6), 661-81.
zatay, F. (2005) Monetary Policy Challenges for Turkey in European Union Accession Process,
Central Bank of Turkey, Research and Monetary Policy Department Working Paper No.
05/12.
zatay, F. and G. Sak (2003) Banking Sector Fragility and Turkeys 2000-01 Financial Crisis, in
S.M. Collins and D. Rodrik, Brookings Trade Forum.
zkan, A. and N. zkan (2004) Corporate cash holdings: An empirical investigation of UK
companies , Journal of Banking and Finance, 28(9), 2103-2134.
zmen E. and .. Koar (1998) Sustainability of Budget Deficits in Turkey with a Structural
Shift, METU in Studies in Development, 25(1), 107-128.
45
zmen, E. and D. Arnsoy (2005) The Original Sin and the Blessing Trinity: An Investigation,
Journal of Policy Modeling, 27, 599-609.
Rajan, R. and L. Zingales (1995) What do we know about capital structure? Some evidence from
international data, Journal of Finance, 505, 14211460.
Ratha, D., Mohapatra, S. and P. Suttle (2003) Corporate Financial Structures and Performance in
Developing Countries, Ch. 4 in World Bank Global Development Finance.
Reinhart, C.M., Rogoff, K.S. and M. Savastano (2003) Debt Intolerance, Brookings Papers on
Economic Activity, Spring, 1, 1-74.
Reinhart, C.M.and K.S. Rogoff (2004) The Modern History of Exchange Rate Arrangements: A
Reinterpretation, Quarterly Journal of Economics, Vol. CXIX No. 1, 1-48.
Roubini, N. and B. Setser (2004) Bailouts or Bail-ins? Responding to Financial Crises in
Emerging Economies, Institute for International Economics, Washington DC.
Singh, A., Belaisch, A., Collyns, C., De Masi, P., Krieger, R., Meredith, G. and R. Rennhack
(2005) Stabilization and Reform in Latin America: A Macroeconomic Perspective on the
Experience Since the Early 1990s, IMF Occasional Paper No. 238.
Sy A. (2005) Managing the Interest Rate Risk of Indian Banks Government Securities
Holdings, International Monetary Fund, Working Paper No: 05/78.
Tobin, J. (1958) Liquidity Preference as Behavior Towards Risk, Review of Economic Studies,
25, 65-86.
Trkan, E. (2004) Trk Ekonomisinde Makro Kredi Kanal: lek ve Kalite Asndan Bir
Deerlendirme, Mimeo, Central Bank of the Republic of Turkey.
Vickers, D. (1987) Money Capital in the Theory of the Firm: A Preliminary Analysis,
Cambridge: Cambridge University Press.
Woodford, M. (1990) Public Debt as Private Liquidity, American Economic Review, Papers and
Proceedings, 80, 382-388.
Yaln, C., ulha, O.Y. and P.. zl (2005) Mali Yap ve Mali Derinliin Ekonomik
Bymedeki Rol: irketler Dzeyinde Analiz, TUSIAD-Ko niversitesi Ekonomik
Aratrma Forumu Tartma Teblileri Yayn No. EAF.2005-06.001.
Ylmaz, G. (2005) Financial Dollarization, (De)Dollarization and the Turkish Experience,
Turkish Economic Association Discussion Paper 2005/5.