Trade Finance and Trade Collapse during the Global Financial Crisis:
Evidence from the Republic of Korea*
Changyong Rhee†
E. Young Song‡
February, 2013
Abstract
This study examines the role of trade finance in the trade collapse of 2008-09 from the perspective of the
Korean economy. We use two approaches. Firstly, we investigate aggregate data on trade finance, on which
Korea has relatively rich data. We find that foreign trade loans and cross-border accounts payable were stable or
even increased during the crisis. In contrast, documentary bills purchased, domestic import usances and crossborder accounts receivable all dropped fast during the crisis. However, documentary bills purchased and crossborder accounts receivable fell at a slower pace than exports. Overall, aggregate data do not seem to support the
view that trade finance played an active role in causing the trade collapse.
Secondly, using quarterly data on listed firms in Korea, we conduct panel estimations to test whether some
financial characteristics of firms intensified export contraction during the crisis. We find that firms who used
more external finance suffered more contraction in the exports/sales ratio during the crisis. In contrast, firms
who relied more on cross-border accounts payable experienced less fall in the exports/sales ratio. We also find
that contrary to the popular belief, small exporters did not suffer more during the crisis. However, small
exporters who also carried large stocks of inventory are found to have suffered more.
When we restrict our attention to export changes in the intensive margin, where we have more reliable data,
we find stronger influence of financial factors. In this case, small exporters indeed were more vulnerable to the
crisis. Trade receivables, trade payables, cross-border trade receivables and cross-border trade payables all
contributed in intensifying export contraction through their interaction with the smallness of firms. It seems that
the inter-firm trade credit market tightened more for small exporters during the crisis.
JEL classification: F10, G01
Keywords: Credit constraints, Firms' heterogeneity, Trade credit
*
This study was commissioned by Asian Development Bank. Song wishes to thank Sogang University for the
partial financial support under the Research Grant of 201310029.01.
†
Chief Economist, Asian Development Bank, Mandaluyong City, 1550 Metro Manila, Philippines.
Tel: +63-2-632-4900/6620, Fax: +63-2-636-2357, e-mail: [email protected]
‡
Professor, Department of Economics, Sogang University, C. P.O. Box 1142, Korea.
Tel: +82-2-705-8696, Fax: +82-2-705-8180, e-mail: [email protected]
I. Introduction
As soon as the Global Financial Crisis of 2008-09 broke out, economists, national policy
makers and the leaders of the world economy expressed concerns about the adverse effects of
financial contraction on world trade, and called for coordinated efforts to prevent a drying-up
of trade finance. The G-20 quickly convened and agreed on the need to expand trade finance
liquidity, and national governments and international agencies set up programs to supply
additional liquidity to finance trade transactions.
During 2009, world trade volume contracted 10.5 percent while world GDP contracted 0.6
percent. The disproportionately large contraction of trade during the Global Financial Crisis
is named the Great Trade Collapse and many studies offer explanations. Now there seems to
have developed a consensus that most of the fall in trade can be explained by demand
contraction caused by the Great Recession. The recession hit hard the demand for consumer
durables, manufactured intermediate goods and capital goods, and world trade is concentrated
on these kinds of goods. However, many observers find this observation inadequate for
explaining trade collapse far larger than GDP contraction. The fall in trade is much bigger
than most econometric and CGE models predict. There must be supply-side causes, such as a
shortage of trade finance or a breakdown of supply chains.
Now we have a fairly large body of literature that studies the role of trade finance during
the trade collapse of 2008-09, a part of which will be summarized in the next section. Studies
published in academic journals tend to present evidence that trade finance played a
significant role in causing trade collapse. In contrast, many studies using survey results and
aggregate or firm-based data on trade finance report that trade finance was a factor, but it
played a rather limited role. We need more country-based studies to make an objective and
comprehensive assessment on the role of trade finance in the trade collapse.
2
This study examines the role of trade finance in the trade collapse of 2008-09 from the
perspective of the Korean economy. We use two research strategies. Firstly, we look at
aggregate data to find a clue on the link between trade finance and trade volume. Korea has
relatively rich data on trade finance and trade credit. The Bank of Korea publishes aggregate
data on “foreign trade loans”, which are loans extended by commercial banks to exporters for
the purpose of providing working capital. The Bank also publishes aggregate data on crossborder accounts receivable and payable of private firms, allowing us to track the volume of
inter-firm trade credit. The Financial Supervisory Service of Korea reports two sets of data on
trade finance: “documentary bills purchased” by commercial banks and “domestic import
usances”, which are banker’s usances extended to domestic importers. We find that foreign
trade loans and cross-border accounts payable were stable or even increased during the crisis
while trade volume was shrinking fast. In contrast, bills purchased, import usances and crossborder accounts receivable all dropped quickly during the crisis. However, except for import
usances, which dropped at a faster rate than imports, bills purchased and accounts receivable
fell at a slower pace than exports. Aggregate data do not seem to support the view that trade
finance or credit played an active role in causing the trade collapse.
The second approach of this paper is an econometric evaluation using firm-level data.
Using quarterly data on listed firms in Korea, we conduct panel estimations to test whether
exporters with more dependence on external finance, or with more reliance on inter-firm
trade credit or with weaker financial ratios were more vulnerable to the crisis. We adopt firmlevel estimations for two reasons. A major problem in estimating the effect of trade finance is
to discern whether the fall in trade was due to inadequate trade finance or lower demand. In
panel estimation using firm-level data, we can easily control for industry-specific demand
shocks by industry-time dummies. The second reason for using firm-level data is related to
the structure of the Korean economy. Most exports from Korea are made by top 30 giant
3
exporters, almost every one of which belongs to a large business group called Chaebol. It is
unlikely that these firms were seriously credit-constrained during the crisis. If some firms
suffered from a shortage of trade finance, they would be small or medium-sized firms. The
effects through SMEs would not show up at industry-level exports where giant firms
dominate.
There are a growing number of studies that use firm-level data to test the influence of
finance on exports using the methodology pioneered by Rajan and Zingales (1998). This
paper is one of them, but it can be differentiated from other studies in three aspects. Firstly,
our data set allows us to check whether some financial characteristics of firms intensified the
fall in the ratio of exports to domestic sales, not the fall in the absolute amount of exports,
during the crisis period. This approach is more reasonable in that the challenge here is to find
out whether trade finance can explain the fall in trade larger than the fall in output. Secondly,
we focus on the effect of finance on small exporters, especially on the interaction between the
smallness of firms and their financial characteristics. The reason for this has been explained
above. Lastly, we utilize firm-level data on cross-border accounts receivable and payable.
Past studies had to rely on data on total trade credit, domestic plus foreign, even though data
on cross-border credit are required, because data on cross-border trade credit were not
available.
We find somewhat mixed results. The two financial characteristics of firms are found to
have robust effects on exports: dependence on external finance and cross-border accounts
payable. Firms who used more external finance in investment suffered more contraction in
the exports/sales ratio during the crisis. In contrast, firms who relied more on cross-border
accounts payable experienced less contraction, suggesting that inter-firm trade credit can be a
substitute for trade finance. We also find that contrary to the popular belief, small exporters
4
did not suffer more during the crisis. However, small exporters who also carried large stocks
of inventory suffered more, maybe because they needed more working capital for exports.
When we restrict our attention to export changes in the intensive margin, where we seem to
have more reliable data, we find stronger influence of financial factors. Now we find that
small exporters were more vulnerable to the crisis. In addition, we find that trade receivables,
trade payables, cross-border trade receivables and cross-border trade payables all
significantly intensified export contraction when they are interacted with the smallness of
firms. It seems that the inter-firm trade credit market tightened more for small exporters
during the crisis.
The paper is organized as follows. Section II summarizes related literature. Section III
presents a snapshot of the Korean experience during the crisis based on aggregate data.
Section IV reports estimation results for the firm-level analysis. Section V concludes.
II. Related Literature
The literature on the effects of financial crises on exports has developed quickly following
empirical studies on the link between finance and growth. Particularly influential is the study
by Rajan and Zingales (1998). They present evidence that sectors that depend more on
external finance grow relatively faster in countries with more developed financial systems.
This finding suggests that financial development reduces credit constraints on firms arising
from financial market imperfections and thereby fosters real growth. Using a similar
framework, Braun (2003) finds that sectors with more tangible assets perform relatively
better in terms of growth and contribution to GDP in countries with poorly developed
financial institutions. This evidence suggests that tangible assets are collateralizable and thus
can solve contractibility problems in external finance, which would be more severe in
financially underdeveloped countries. Fisman and Love (2003) also use an analogous
5
approach, but they emphasize inter-firm credit as an alternative to bank finance. They find
that in countries with the low quality of financial intermediation, sectors that depend more on
inter-firm trade credit tend to grow relatively faster.
Parallel studies can be pursued in search for the link between finance and comparative
advantage. A theoretical lead is provided by Kletzer and Bardhan (1987) who had shown that
differences in the cost of capital can arise due to credit market imperfections between
countries with identical technology and factor endowments. Beck (2003), Hur, Raj and
Riyanto (2006) and Manova (2008a) back up the Kletzer-Bardhan hypothesis by presenting
evidence that sectoral differences in external finance dependence or asset tangibility can play
a role in shaping comparative advantages among countries with differing levels of financial
development.
More relevant to the present paper are Kroszner, Laeven, and Klingebiel (2007) and
Dell'Ariccia, Detragiache and Rajan (2008). These studies focus on the impact of financial
crises on industrial output. Here the key idea is that the period of a banking crisis can be
regarded as the period of sudden deterioration in the quality of financial intermediation, and
thus its effect on sectoral growth would be similar to that of financial underdevelopment.
They find that sectors that are more dependent on external finance experience greater
contraction during banking crises. The study that has a direct bearing on the debate on the
role of trade finance in the Great Trade Collapse of 2008-2009 has been initiated by Iacovone
and Zavacka (2009). Instead of output, they investigate the effect on exports, but the
methodology used is essentially identical to those used by aforementioned papers. They find
that sectors with higher reliance on external finance in the sense of Rajan and Zingales (1998),
or sectors with less tangible assets in the sense of Braun (2003), or sectors with less
dependence on trade credit in the sense of Fisman and Love (2003) suffered more in terms of
export growth during the past banking crises.
6
The paper by Chor and Manova (2012) is among the first that applied the Rajan-Zingales
framework to data on the trade collapse of 2008-09. Using US imports data, they find that
exports from sectors with more dependence on external finance or those with less tangible
assets or less reliance on inter-firm trade credit were more vulnerable to interest rate hikes
during the global financial crisis. In contrast, Levchenko, Lewis and Tesar (2010) find no
evidence that US imports or exports shrank more or less in sectors with more reliance on
inter-firm trade credit during the crisis. However, they do not control for setoral differences
in demand shifts during the crisis, making it difficult to compare their result with Chor and
Manova (2012). In parallel with the growing popularity of heterogeneous firm models of
trade, now more and more studies on trade finance use firm-level data. Among the most
innovative is the study by Amiti and Weinstein (2011). Using Japanese firm-level data during
the Japanese financial crisis of 1990s, they find that the exports of firms whose main banks
were in trouble suffered more drops in exports, suggesting a direct link between trade finance
and exports. The study by Bricongne et al. (2012) is the one that is closely followed by the
current paper. In their study, a history of credit defaults plays the role of an unhealthy main
bank in the study of Amiti and Weinstein (2011). The logic is that firms that had defaulted on
credit in the past would be more likely to be credit-constrained during the crisis and thus their
exports would be more adversely affected. Using French firm-level data, they discover this
tendency among firms in sectors with high reliance on external finance, but they also find that
its quantitative effect is quite limited. Minetti and Zhu (2011) also conduct a firm-level study.
They identify credit-constrained firms by Italian survey data, and find that these firms tend to
export less than non-constrained firms. Behrens, Corcos and Mion (2010) also find in Belgian
firm-level data that some variables measuring the financial vulnerability of firms significantly
affect exports. However, they find no evidence that these variables affect trade more than
domestic sales, suggesting no special role of trade finance in causing trade to drop more than
7
output during the global financial crisis. While studies published in academic journals tend to
find evidence that trade finance or credit played an important role in causing trade collapse,
the general picture seems to be different. A lot of studies using survey results, aggregate data
on trade finance or firm-based data report the limited role of trade finance in the Great Trade
Collapse. Chapters in the World Bank volume edited by Chauffour and Malouche (2011)
summarize these studies and echo this kind of skepticism.
III. A Snapshot of aggregate data
<Figure 1>
In this section, we draw a snapshot of Korean exports and financial markets during the
Global Financial Crisis using some aggregate data. Figures 1 and 2 trace the value, price and
quantity of Korean exports. According to Figure 1, the dollar value of exports that had been
increasing over 20% per year started to decline sharply around July, 2008. The decline started
with a plunge in the dollar prices of exports. The dollar price index of exports declined fast
until March 2009 and then slowly recovered. The quantity index of exports also dropped
sharply until January, 2009, but it recovered quickly. By the middle of 2009, it had almost
recovered the level that had been reached before the crisis. Because of the falls in the price
and the quantity of exports, the dollar value of exports dropped 50% from its pre-crisis level
when it hit the bottom in January, 2009.
<Figure 2>
8
The picture becomes substantially different if we use the local currency won as the
measure of value. The won/dollar exchange rate has been rising slowly before the Lehman
shock in September, 2008. The exchange rate then shot up with the Lehman crisis, and by
March, 2009, at the peak of financial turmoil, the won depreciated more than 40% from the
pre-crisis level. Because of the sharp depreciation, the won prices of Korean exports did not
fall despite the sharp fall in dollar prices, and the won price index actually remained above
the pre-crisis level at the heights of the financial crisis. Because of stable export prices, the
won value of exports moved in tandem with the quantity of exports during the crisis. When it
hit the bottom in January, 2009, the won value of exports fell 34 % from the pre-crisis level.
Then the value recovered quickly as the quantity did.
<Figure 3>
Figure 3 summarizes the credit market situation using data on three kinds of interest rates:
bank lending rate for large companies, bank lending rate for small and medium companies,
and discount rate on commercial paper. Various interest rates had been increasing slowly as
the subprime mortgage crisis was developing in the US and had a brief surge in January, 2008
when the turmoil shook up the international short-term financial markets. After a brief relax,
interest rates mounted again and shot up with the Lehman Crisis in October, 2008. However,
they declined fast as the Bank of Korea conducted a series of big policy-rate cuts after
October, 2008. As the year 2009 began, the financial squeeze was almost over in terms of
credit prices. Other interest rates showed similar patterns during the crisis.
<Figure 4>
9
An important question to ask is whether the Korean economy experienced a great trade
collapse or a disproportionally large contraction of trade during the global financial crisis.
Figure 4 shows that exports indeed dropped proportionally more than GDP during the crisis.
The exports/GDP ratio dropped 14 percentage points from top to bottom. However,
exports/industrial output moved little and dropped only 2 percentage points from top to
bottom. Exports contracted in tandem with manufacturing output and the exports/GDP ratio
fell as manufacturing output fell proportionally more than services. In this sense, Korea did
not experience a great trade collapse during the global financial crisis.
Korea has relatively rich sets of data on trade finance and trade credit. In the following, we
will use trade finance and trade credit in different senses. Trade finance will be used to denote
credit created through financial intermediaries to support international trade. In contrast, trade
credit will be used to denote credit created through inter-firm international transactions of
goods and services. We first look three measures of trade finance.
Figure 5 traces the behavior of foreign trade loans around the crisis period. Foreign trade
loans denote credits extended by commercial banks to domestic exporters and export-related
firms for the purpose of providing them with working capital necessary for export-related
business. The Bank of Korea collects and publishes monthly data on foreign trade loans.
These loans are treated as domestic local currency loans of commercial banks because they
are denominated in Korean won. We can see from the figure that except for seasonal drops in
December, the amount of foreign trade loans was stable during the crisis period. With stable
trade loans and rapidly shrinking exports, the ratio of trade loans to exports peaked in January
2009, in the middle of the crisis. The ratio remained high during 2009 and began to fall as the
value of exports began to recover from early 2010. It is hard to believe that that foreign trade
loans played an active role in the export collapse of late 2008 and early 2009.
10
<Figure 5>
Figure 6 presents the behavior of documentary bills purchased by commercial banks. They
are mostly bills of exchange purchased by commercial banks from exporters. They are called
documentary bills because exporters should back up these bills with shipping and collection
documents. Domestic exporters are direct recipients of these credits, but because foreign
importers or their banks are liable for them, they are treated as external foreign currency
assets of banks. Quarterly data are collected by the Financial Supervisory Service of Korea.1
According to Figure 6, documentary bills purchased by banks dropped sharply during Q4
2008 and Q1 2009. However, the value of export dropped at the same time at a faster rate,
increasing the ratio of bills purchased to exports in Q1 2009. As exports recovered and bills
purchased remained stagnant, this ratio fell during Q2 2009 and afterwards. It would be hard
to argue that the contraction of trade finance in the form of bills purchased initiated the export
collapse in Q4 2008 and Q1 2009. However, its recovery was sluggish even as the value of
exports quickly recovered from early 2009. Trade finance in the form of bills purchased
might have played a role in dampening the speed of export recovery.
<Figure 6>
Figure 7 deals with our last measure of trade finance. Domestic import usances are banker’
usances extended by commercial banks to domestic importers. These are treated as domestic
foreign currency assets of banks because domestic importers are liable for them in foreign
1
The data do not include credits supplied by the local branches of foreign banks. They might have played an
important role.
11
currency. Quarterly data are collected by the Financial Supervisory Service of Korea. 2
Domestic import usances dropped sharply during Q4 2008 and Q1 2009. The value of
imports also dropped sharply, but in this case, import usances declined faster, resulting in a
slight fall of usances/imports ratio in Q4 2008 and in Q1 2009. Afterwards, both usances and
imports recovered steadily and the usances/imports ratio remained stable. It is possible that
the initial drop in domestic import usances was a factor in causing the sharp drop in imports
immediately after the Lehman shock.
<Figure 7>
Next we investigate the behavior of inter-firm trade credit collected from the Korean
International Investment Position. The Bank of Korea records on the external asset side of the
International Investment Position cross-border trade credit extended by domestic private
firms. We look at only short-term credits. Figure 8 shows that external trade credits extended
by domestic private firms fell from Q3 2008 through Q1 2009. However, the value of exports
dropped more sharply, and the ratio of trade receivables to exports increased from Q4 2008
through Q3 2009.
<Figure 8>
Figure 9 investigates the short-term external trade-credit liability of private firms. There
are credits extended to domestic firms by foreign firms in the form of trade payables. It is
quite interesting to note that external trade payables of domestic firms increased despite the
2
Again, the data do not include credits supplied by the local branches of foreign banks.
12
financial stress and the sharp fall of imports during Q4 2008 and Q1 2009.
Because of this
mysterious development, the ratio of trade payables to imports peaked up in Q1 2009, and
was kept at a high level during the crisis period. It could be the case that domestic importers
who were declined banker’s usances shifted to inter-firm trade credit to finance imports. It is
difficult to argue that a freeze in trade credit markets acted as a major impediment to imports
for Korean firms.
<Figure 9>
The data sets above allow us to track the Korean economy both on the path of trade finance
and inter-firm trade credit using official statistics rather than survey results. It is repeatedly
observed that a measure of trade finance or trade credit dropped sharply during the crisis, but
the value of trade declined at a faster rate such that the ratio of credit to trade value increased
during the trade collapse. Except in the case of import usances, it seems more likely that trade
collapse caused credit contraction, not the other way round. However, our data can be made
consistent with the hypothesis that trade finance caused trade collapse if one could argue that
a 1 percent drop in trade finance caused a more than 1 percent drop in trade through a
multiplier process.
Some observers argue that the failure to find in aggregate data convincing evidence for
trade finance as a significant cause of trade collapse may be explained by the possibility that
firms who were declined inter-firm credit during the crisis had to switch to banks to find an
alternative source of financing. The increased demand for bank finance and the resulting
increase in the supply of trade finance may explain why we do not observe sharp declines in
the official statistics of trade finance. This line of argument does not seem convincing in the
13
case of Korea. As we have seen in figures 8 and 9, inter-firm trade credits did not fall more
sharply than measures of trade finance.
<Figure 10>
Another way of checking the possibility of substitution between trade finance and trade
credit is to examine exports classified by payment types. The Korea Customs Service
classifies exports according to methods of payment. The largest category is remittance, which
includes cash-in-advance, COD (cash on delivery) and CAD (cash against documents), and
open-account transactions. L/C transactions are those in which payments are made through
letters of credit issued by banks. Collection refers to transactions in which payments are made
using documentary collection by banks, but that do not involve trade finance. The three types
of exports do not sum to 100 percent as there are other kinds of transactions, of which the
biggest category is processing trade. Figure 10 shows variations in the composition of exports
by methods of payment. It seems that the percentage of exports by letter of credit declined
faster in early 2009 despite it was on a long-term downward trend, while the percentage of
exports by direct remittance rose faster on its upward trend during early 2009. It is not clear
whether this rise in the percentage of remittance was caused by an increase in cash-inadvance transactions or open-account transactions. However, the data suggest that a squeeze
in trade finance pressured firms to shift to cash payments or inter-firm trade credit, not the
other way round.
14
IV. Firm-level Analysis
In this section, we test using firm-level data whether finance played an independent role in
the contraction of Korean exports during the Global Financial Crisis. The econometric
method that we will use is a difference-in-differences analysis at a firm level similar to the
studies by Behrens et al. (2010) and Bricongne et al. (2012).
<Table 1>
We adopt a firm-level approach for two reasons. Firstly, during the Global Financial Crisis
of 2008-2009, the predominant concern of the Korean government and economists was its
effects on small and medium enterprises. Korean exports are dictated by a small number of
giant firms, almost all of which belong to family-controlled business groups called Chaebols.
Table 1 shows that in 2007, top 10 firms exported 60 percent, and top 20 firms exported 74
percent of the total merchandise exports by listed firms in Korea3. Almost all these firms
belong to Chaebols specially monitored by the Korea Fair Trade Commission, have large
cash reserves, and enjoy privileged access to commercial banks and bond markets. If
financial factors had any independent supply-side effect on the exports of Korean firms, it
would show up thorough small and medium exporters. However, the effect on SMEs would
be too small to be detectible by industry-level data on exports.4 Indeed, a major response to
the financial crisis of the Korean government and public financial institutions was to expand
public funds for small and medium firms.
3
During the sample period, listed firms contributed on average 80 percent of total Korean exports
4
These giant exporters depend on small and medium-sized domestic firms for the supply of parts and
components, and a financial crisis can affect giant exporters through supply chains. This is a possibility although
we do not believe that it is very likely given the tight control of Chaebols over their supply chains.
15
The second reason that we use firm-level data is that we can control for demand shifts and
other time-varying industry-specific effects through industry-time dummies. As we saw in the
previous section, most of the contraction in trade finance or trade credit during the crisis is
likely to be the one induced by the contraction in demand for Korean exports. A crucial step
in detecting the independent role of finance on exports is to purge the influence of these
demand shifts from export data. For example, in some industry-level studies, researchers find
that export collapse was relatively greater in industries that are more vulnerable to financial
shocks, but these industries could be exactly those where external demand dropped more.
Carefully-constructed controls for industry-specific demand shocks can avoid this problem,
but they often are difficult to construct. A firm-level analysis allows us to use industry-time
dummies to control for demand shifts and other time-varying industry-specific variables.
1. Data
All firm-level data that we will use come from the WISEfn database, which compiles the
quarterly financial statements of listed firms in Korea published by the Korean Listed
Company Association (KLCA). The database contains total exports and domestic sales of
1,798 listed firms, along with various income statement and balance sheet variables. For
years from 2005 through 2009, on average, 62 percent of listed firms reported positive
exports. It is unfortunate that the database does not break total exports by destination and
product code.
A potential problem is dealing with zeroes in exports. If we ignore them, all variations of
exports at the extensive margin will be eliminated from data and bias can be introduced in
estimation. To avoid this problem, we will use mid-point growth rates of exports as in
Bricongne et al. (2012).
16
=
t.
(
)
(
))
(
× 100.
(1)
is the year-on-year mid-point growth rate of exports by i th firm in industry k at quarter
is the won value of exports. The only difference from the standard growth rate is that
we put in the denominator the average of the past value and the current value of x instead of
the past value alone. Thus the growth rate is well-defined unless both the past value and the
current value of x equal zero. The mid-point growth rate moves closely with (log
log
))
(
× 100 when both
and
(
)
−
are positive. It takes the minimum value of
−200 when the value of exports drops from a positive number to zero and the maximum value
of 200 when it increases from zero to a positive number. In this way, the growth rate can
capture variations at the extensive margin. However, assigning the absolute value of 200 to
whatever change at the extensive margin can be considered as arbitrary and it may result in
the situation where estimation results depend excessively on tiny changes in exports of small
firms around zero.
We can classify
when
into four categories. Growth at the positive extensive margin occurs
= 200, growth at the negative extensive margin when
positive intensive margin when
margin when
=∑
= −200, growth at the
is in (0, 200), and growth at the negative intensive
is in (−200, 0). The aggregate growth at each margin at time t can be
calculated by summing up
margins.
over (i, k) for which
belongs to each of the four
is the weight for firm i and calculated as follows.
(
(
) )
(
(
(2)
))
17
<Table 2>
The period that we will examine is 20 quarters from Q1 2005 through Q4 2009.5 We will
restrict our analysis to firms in manufacturing industries. Table 2 reports aggregate growth
decomposed into the four margins. We can see that variations at the extensive margin are very
small compared to those at the intensive margin. This pattern is repeatedly observed in other
studies, but it is partly due to the limitation of our data. Some of growth that was classified
into the intensive margins would shift to the extensive margins if we used exports data
divided by product and destination.6 We can also note that the export growth rate of listed
firms in manufacturing moves closely with the growth rate of goods exports in Korean won
as reported by the Bank of Korea (their correlation equals 0.86). However, there are some
notable discrepancies between the two series around the Great Financial Crisis period. The
total mid-point export growth of listed firms never fell below zero during the crisis period
and it recovered quickly from Q4, 2009. In contrast, negative export growth rates were
maintained from Q2 2009 through Q4 2009 in the BOK data. An interesting feature of Table
2 is the extent to which the value of Korean exports is subject to exchange rate shocks. Goods
exports in Korean won had been growing at almost 40% just before the Lehman shock as the
won/dollar exchange rate was increasing fast. The won value of exports still was growing in
Q1 2009 in the middle of the crisis as the exchange rate was depreciating at the rate of 48
5
The quarterly reports of financial statements of listed firms became mandatory from 2004. At the time of
writing, there were too many missing data in the financial statements for years 2010 and 2011.
6
Another issue is how to treat entries into and exits from stock exchanges by firms. Because the exports of a
firm before an entry or after an exit are unobservable to us, we simply treat them as missing. We do not treat
them as zeroes because most exporters were exporting before it was listed and kept on exporting after it was
delisted. Thus growth at the extensive margin in our data set is mostly due to changes in the exports of small
listed firms from and to zero.
18
percent. In terms of the won value of exports, the export collapse during the financial crisis of
2008-2009 was not really more severe than the pace of export contraction in 2005 when the
Korean won was fast appreciating.
<Table 3>
In the analysis below, we will define the crisis period as the two quarters from Q1 2009
through Q2 2009. During this period, the manufactured exports of listed firms grew at nearzero rates, while in Q4, 2008 and in Q3, 2009, the growth rate was above or near 10 percent.
The KLCA classifies manufacturing firms as belonging to one of 80 industries based on the
major products of firms. The industry classification is identical to the Korean Standard
Industrial Classification at 3-digit level. We regroup them into 18 bigger industries as shown
on the first column of Table 3. The second column shows how the exports from each industry
were affected during the crisis period of Q1 2009 and Q2 2009 compared to the non-crisis
period. The numbers were obtained by estimating the industry fixed effects of the crisis from
a regression of the mid-point export growth rates of firms on industry dummies and industry
dummies interacted with the crisis dummy that takes the value of 1 during the crisis period.
The three industries that are most adversely affected by the crisis are motor vehicles and parts,
basic metal products, and other machinery. The three that are most favorably affected are
pharmaceuticals and medical chemicals, domestic appliances, and non-metallic mineral
products.
<Table 4>
19
Table 4 shows by industry the summary statistics of key financial variables that we will
employ in regression. The 9 financial variables are defined as follows.
!"#$%&'()$*+,- .*
EXTFIN ≡ 1 − 0 × 100,
'"%&')#"$*"*&'%$'(&/-"$**"%*
TANG ≡ %$'(&/-"$**"%*
TPAY ≡ )
TREC ≡ INV ≡ % %$-$**"%*
%#$1"!$2$/-"*
*% ,(
× 100,
1** -1
× 100,
%#$1"#")"&3$/-"*
*$-"*
&'3"'% #&"*
*$-"*
INTCOST ≡ STBFOR ≡ × 100,
× 100,
'"%&'%"#"*%"4!"'*"*
56789:
× 100,
*+ #% %"#;/ ## .&'(*&', #"&(')<##"')&"*
TPAYFOR ≡ *$-"*
%#$1"!$2$/-"*&', #"&(')<##"')&"*
TRECFOR ≡ "4! #%*
× 100,
%#$1"#")"&3$/-"*&', #"&(')<##"')&"*
"4! #%*
× 100,
× 100.
EXTFIN is a variable that measures the degree to which a firm depends on external finance
for investment. This measure of financial vulnerability is the most popular in the literature on
finance and growth after Rajan and Zingales (1998). We expect that during a credit crunch, a
firm whose has a higher value of EXTFIN will suffer more in export growth because of its
higher financial vulnerability. Braun (2003) proposed TANG as a measure of secure access to
external finance. We expect that a firm who has a higher value of TANG will suffer less in a
tight credit situation because collateralized borrowing will be easier. TPAY measures the
dependence of a firm on inter-firm credit offered by its suppliers. Fisman and Love (2003)
argue that a firm whose has a higher value of TPAY would suffer less in a tight credit market
20
because it depends less on commercial banks for working capital finance. However, during
the Global Financial Crisis, inter-firm credit could have contracted more than commercial
banking, and the opposite situation might have been the case. We tried alternative ways of
calculating TPAY such as the ratio of trade payables to the stock of debt or its flow version.
However, they did not work better. In a similar concept, TREC can be used to measure the
dependence of a firm’s sales on inter-firm credit, now the firm acting as a credit supplier. One
can argue that a seller who has a higher value of TREC can maintain more stable flows of
sales during a crisis because its customers have less need to borrow from banks. However,
with the increased vulnerability of trade partners during a crisis, sellers can cut credit to their
customers more sharply than bankers, and the opposite situation can develop. As in the case
of TPAY, we tried alternative ways of calculating TREC, but their performances were not any
better. INV is added to check whether a firm who maintains a higher stock of inventory
relative to sales suffered more during the crisis. This tendency might develop simply because
firms with higher levels of inventory become more vulnerable to contraction in short-term
finance as they need more working capital during production process. However, there can be
more complicated reasons. Alessandria et al. (2010) and Zavacka (2012) argue that firms
with large inventories can be made more vulnerable to recessions through inventory
dynamics.
In addition to these variables, we test the influence of variables that measure more directly
the financial vulnerability of firms. One may expect that firms who had borrowed more
would suffer more during a crisis because they are more likely to be credit-rationed. There
are many measures of debt reliance such as debt-to-equity ratio or interest-expense to sales
ratio. Among these financial ratios, we report only results using INTCOST, which is the ratio
of net interest expenses to EBITDA. In most of regressions, we find that this variable has a
more significant effect on export contraction than other debt reliance measures. An attraction
21
of our dataset is that it reports short-term borrowings and trade credits denominated in foreign
currencies. Unfortunately, to accommodate them in our analysis, we have to significantly
reduce the sample size because many firms did not report these items. STBFOR measures the
ratio of short-term borrowings in foreign currencies to total sales. We may expect that a firm
that maintains a higher value of this ratio would be hurt more by hikes in interest rates or
exchange rates during a crisis. The data on trade credits denominated in foreign currencies
are quite valuable as data on cross-border trade credit at firm-level are rare. TPAYFOR is the
ratio of trade credit payable in foreign currencies to exports. Using imports rather than
exports in the denominator will be logical, but we had to use exports because imports data
were not available. TRECFOR is the ratio of trade credit receivable in foreign currencies to
exports.
Table 4 reports for each of 18 industries the medians and the standard deviations of the 9
financial ratios defined above. We calculated nine-year (2001-2009) averages of these
variables for each firm, and then obtained the median and the standard deviation of each
variable across firms in each industry. For each industry, the upper row shows the medians,
and the lower row the standard deviations. On the second column, we can see that metal
product industries, electronic and electrical components industries, and other machinery
industries have relatively high medians of EXTIN. This result can be expected, but it may be a
surprise to find that industries like textile, apparel and footwear or furniture and other
manufacturing also have very high median values, and the motor vehicle industry has a
negative median. The standard deviations of EXTFIN are extremely large in most industries
because operating cash flows and fixed investments vary enormously across firms. This
phenomenon does not go away when we work with a finer industry classification with 80
industries, and the variable is quite volatile across time too. We doubt the wisdom of treating
this variable as a measure that captures an exogenous characteristics of an industry. We find
22
from Table 4 that INTCOST, TPAYFOR and TRECFOR have high standard deviations relative
to medians, but to a far less degree than EXTFIN. In contrast, TANG, TPAY, TREC and
STBFOR have low standard deviations relative to medians. These variables also have
medians that are stable across industries as can be found on the last row, which reports the
standard deviations of medians across industries.
2. Estimation
For our benchmark regressions, we will use the following specification.
= => + =@ ABC
+ =D ABC
∗ FCGHGH + =I ln KBLMCNCMO
+ = ln PQNHRBCP
+S
+T
.
(3)
VAR is one of the nine financial ratios defined above. We add another variable ASSET, a
dummy that takes the value of 1 when a firm is in the bottom quintile of its industry in terms
of asset size. The small-firm dummy is added to test whether small firms were more
adversely affected during the financial crisis. CRISIS is a dummy that equals 1 if t is in the
crisis period. ln KBLMCNCMO is the natural log of labor productivity calculated as the ratio
of value-added to the number of employees. ln PQNHRBCP is the natural log of a firm’s
export share in its industry.
Note that (3) is written in the form of a difference-in-differences model. What we aim to
test is whether a firm who had a higher value of VAR experienced a lager drop in the export
growth rate during the crisis period. VAR is included in the regression equation because firms
with difference values of VAR were growing at different rates during the non-crisis period,
and what we want to compare across firms is the difference between the growth rate during
the non-crisis period and that during the crisis period. What we are most interested in is the
value of =D . A negative (positive) =D implies that firms with higher values of VAR
23
experienced lager (smaller) drops in the export growth rate during the crisis. In this way, we
may infer that the financial factors captured by VAR exerted a negative (positive) effect on
exports during the crisis.
A potential problem in applying the difference-in-differences method to a firm-level
analysis is that financial factors represented by VAR might be correlated with the error term T.
One may argue that the problem might be more serious in a firm-level study than in an
industry-level study because the financial ratios frequently used in industry-level studies can
be regarded as capturing exogenous and technical properties of industries.7 For example, a
finding that firms who were hit harder by the crisis also had higher values of EXTFIN may
imply that these firms were losing competitiveness and profits even before the crisis and thus
had to rely more on debt finance for their investment. To mitigate the endogeneity problem,
we employ two methods. Firstly, we use the past values of VARs in regressions. For example,
EXTFINikt for quarter t in year s is calculated by the three-year average of EXTFIN from years
(s−4) through (s−2).8 Other VARs are calculated in the same way. Secondly, we add on the
right-hand side ln KBLMCNCMO to control for the competitiveness of firms. We also include
ln PQNHRBCP. This variable is for capturing the convergence dynamics of firms: young and
small firms may grow faster than mature firms, or firms who experience unusual adverse
shocks may tend to recover and grow faster in future periods. Again, we use the past values
of these variables in regressions. These two variables could be interacted with the crisis
dummy, but we found that the interaction terms are not significant and do not alter our main
results. Thus we will focus on the simpler case.
7
However, one can cast serious doubts on the conventional wisdom. The financial characteristics of industries
are altered by shifts in comparative advantages and vary across countries and time.
8
Firms usually are required to submit the financial statements for the past three years when they apply for bank
loans. The value for year (s−1) was not used because
(
)
is used to calculate
year (s−1). Annual values were used to increase the availability of the data.
24
and (t −4) is a quarter in
S
are industry-quarter dummies capturing fixed effects that vary across industries and
quarters. They capture variations in price deflators and exchange rates that can change with (k,
t). However, this variable serves a more important purpose. S
are included to control for
demand shifts that vary across time and industries. A potential problem is that our industry
classification is not fine enough to adequately control for demand shocks faced by different
industries. To check the robustness of our results, we will later report results using a finer
industry classification. In our benchmark regressions, we will use robust standard errors
clustered by industries because the export shocks faced by firms in the same industry are
likely to be correlated. When we work with a finer industry classification with 80 industries,
standard errors will also be clustered accordingly.
<Table 5>
Table 5 reports the results of our benchmark regressions. The dependent variable is the
mid-point growth rates of firms. Across all regressions, we find that the effects of
ln KBLMCNCMO and ln PQNHRBCP on export growth are highly significant. A 1 percent
increase in labor productivity raises the export growth rate by about 10 percent. We also find
that there is a strong tendency for convergence. A 1 percent increase in the export share tends
to lower the export growth rate by about 2 percent. In the below, we will not further mention
about these two variables as the effects of these two variables are consistently strong across
all regressions.
Of the ten variables under investigation, only two variables have significant effects on
export contraction: EXTFIN and TPAYFOR. A higher dependence on external finance in
investment increased export contraction during the crisis and the effect is significant at 1
percent. This result is in line with studies that use data from other countries, but the estimated
25
coefficient of −0.002 is quite small. However, as we saw in Table 4, firms show extremely
large variation in this financial ratio. The coefficient implies that a firm whose EXTFIN was
higher by one standard deviation (6668.9) experienced an additional 13.3 percentage point
drop in the export growth rate during the crisis. In contrast, a higher dependence on trade
payables in external trade reduced export contraction during the crisis. The effect is
significant at 1 percent level, and the coefficient of 0.006 implies that an increase in
TPAYFOR by one standard deviation (583.4) results in a 3.5 percentage point lower drop in
the export growth rate during the crisis. This result suggests that trade credit extended by
foreign suppliers to Korean exporters was relatively stable during the crisis period.
All other variables are found insignificant. The availability of tangible assets for collateral
loans (TANG), total (domestic plus foreign) trade credit received and extended (TPAY and
TREC), inventories (INV), interest expenses (INTCOST), short-term borrowing in foreign
currencies (STBFOR), and trade receivables in foreign currencies (TRECFOR) all had
insignificant effects on export contraction. It is also interesting to note that smaller firms did
not seem to experience additional contraction during the crisis. We find the same result
whether we use different percentiles to define small firms or use the number of employees
instead of assets to define smallness. This finding runs counter to the expectation of many
observers who believe that small firms are discriminated in the Korean financial markets.
We investigate further into the experience of small firms during the crisis because this was
the central concern of policy makers, and promoting small and medium size firms became the
primary object of economic policy in Korea. For this purpose, we regress export growth rates
on the nine financial variables interacted with the asset dummy:
= U> + U@ ABC
+ UD ABC
+ UW ln KBLMCNCMO
∗ FCGHGH +UI ABC
+ UX ln PQNHRBCP
∗ BHHPV
+S
+T
26
.
+ U ABC
∗ BHHPV
∗ FCGHGH
(4)
<Table 6>
The coefficient U tests whether firms who were small and had the high value of VAR
experienced a larger export contraction during the crisis compared to firms who were not
small and had the same value of VAR. In other words, we are asking whether smallness
increased export contraction through its interaction with the financial factor represented by
VAR. In Table 6, we see that smallness significantly intensified export contraction through its
interactions with two financial variables. The estimated value of U
for INV is negative and
significant at 5 percent. The coefficient of −0.916 implies that a firm whose value of INV
equals 12.3, which is the average of the median values of INV in 18 industries, is expected to
experience an 11.3 percentage point larger contraction in export growth when it is small.
Because INV alone does not exert significant influence on export, this means than firms who
are small and carry large stocks of inventory are vulnerable to credit crunch. We can make a
similar observation for the variable of STBFOR. The estimated value of U equals −2.121
and is significant at 1 percent. The coefficient implies that a firm whose value of STBFOR
equals 4.9, the average of the industry medians, is expected to experience a 10.4 percentage
point larger contraction in export growth when it is small. Exporters who were small and
borrowed heavily in foreign currencies suffered more during the crisis.
3.
Robustness
In this subsection, we check for the robustness of our benchmark regressions. We try five
different specifications for each variable that we found significant. The first one is to cluster
27
standard errors by firms, instead of industries. The second one is to run within regressions to
see if the times series variation of VAR within a firm affects export growth. What we are
asking here is whether a firm would be hurt more by the crisis when its value of VAR during
the crisis is higher than its normal value. This regression had an advantage of avoiding biases
that arise from omitting time-invariant characteristics of firms from regressions, but we do
not think that this method asks a question that we want to answer. The third robustness check
is more important. We estimate the effects of variables on net growth, which is the mid-point
growth rate of exports minus the mid-point growth rate of domestic sales. The big question
that is being asked in the literature is whether trade finance or credit can explain the Great
Trade Collapse, world trade contraction much larger than world GDP contraction during the
Global Financial Crisis. If one of our financial variables contributes toward this question, it
should influence exports more than domestic sales. The fourth and fifth specification use a
finer industry classification with 80 industries. The industry classification coincides with the
Korean Standard Industrial Classification at 3-digit level, but in our sample, firms were
exporting in only 70 industries of them.
<Table 7>
Table 7 reports the robustness tests for the estimation results for equation (3), by which we
estimated the effect that a financial variable exerts by itself. We found that two variables
EXTIN and TPAYFOR have significant effects on export contraction. In Table 7, five columns
under each of these two variables report results on five robustness checks explained above.
Both variables are quite robust except in within regressions where the significance level
drops. Higher values of EXTIN (TPAYFOR) resulted in larger (lower) contractions of exports
28
than of domestic sales whether we use the industry classification with 18 industries or the
finer industry classification.
<Table 8>
Table 8 reports the results on robustness checks for equation (4), by which we estimated
the effect of smallness when it is interacted with financial variables. We found that INV and
STBFOR had significant effects on export growth when they were combined with small firms.
Table 8 shows that the interacted effects of INV survive in the alternative specifications
although in the within regression, the effect becomes less significant. In contrast, STBFOR
does not seem to have robust effects. Regressions (37) and (39) show that we have no ground
to believe that foreign borrowings reduced the export growth of small firms more than their
domestic sales growth during the crisis. In addition, the variable becomes much less
significant when we use the finer industry classification.
The final aspect of robustness that we consider is the way that we calculate the export
growth rates of firms. We have been using the mid-point growth rate to include in our
analysis the responses of firms in the extensive margin. When the export of a firm recovers
from zero to a positive number, its export growth is recorded as 200 percent, and when the
export falls from a positive number to zero, its export growth becomes −200 percent. The
number 200 that we assign to any change, however small, in the extensive margin can be
considered as arbitrarily large. Furthermore, these changes in the extensive margin almost
always occur on tiny firms whose international business sporadically turns on and off. They
might carry important information, but they can produce a lot of noise. We rerun all previous
regressions now using only data on changes in the intensive margin to see how the picture
changes. The Appendix reports these results.
29
In Table A1, we report the estimation results for the effects of financial variables when they
act by themselves. In the sample where we include only changes in the intensive margin,
three more variables acquire significance in addition to EXTIN and TPAYFOR. TANG and
ASSET now are significant at 5 percent level, and INTCOST becomes significant at 10
percent. Tables A2 and A3 check the robustness of these additional variables. To judge
robustness, we will focus on whether a variable remains significant in the regressions that use
net growth as the dependent variable, and in those using the finer industry classification.
Table A2 shows that TANG and INTCOST do not have robust effects on export contraction. In
contrast, ASSET survives all five tests. In the intensive margin, small firms suffered more
export contraction during the crisis, and the ratio of exports to domestic sales dropped more
for small firms than for big firms.
In Table A4, we run regressions estimating the effects of nine financial variables when they
are interacted with smallness. Compared to Table 6, where only INV and STBFOR are
significant, we have four additional variables that are significant: TPAY, TREC, TPAYFOR
and TRECFOR. Tables A5 through A7 check the robustness of these results using five
alternative specifications. All six variables seem to have robust effects on export contraction.
STBFOR is the weakest of them: the significance of STBFOR vanishes in the regression of
net growth using the finer industry classification as regression (A54) reports.
In summary, two variables seem to have robust effects on export contraction during the
Global Financial Crisis: EXTFIN and TPAYFOR. Firms who used more external finance in
investment suffered more contraction of export growth, and the dependence on external
finance caused more contraction of export growth than domestic sales growth. In contrast,
firms who relied more on cross-border trade payables experienced less contraction of export
growth during the crisis. This evidence might be interpreted as suggesting that the trade
finance of banks tightened more than inter-firm trade credit during the crisis. However, the
30
ground for this interpretation becomes rather weak if we also consider our findings that
variables directly measuring financial vulnerability, such as debt-equity ratio or the ratio of
interest expenses to sales or profits, did not have any systematic effect on export. For small
firms, we have to give a serious consideration to the effect of an additional variable: INV.
Small firms who carried large stocks of inventory suffered more contraction of export growth
during the crisis, absolutely and relatively to the contraction of domestic sales growth. Small
firms who need more working capital because of larger inventory stocks seem vulnerable to a
financial crisis.
If we restrict our attention to export changes in the intensive margin, where we seem to
have more reliable data, we can find more evidence for the influence of financial variables.
Now the small firm dummy acquires significance and its effect is quite robust. When nine
financial variables are interacted with the small-firm dummy, we find that six variables (INV,
STBFOR, TPAY, TREC, TPAYFOR and TRECFOR) can cause additional contraction in the
export growth of small firms. Of the six, the latter four are significant only for export growth
in the intensive margin, and their effects are quite robust in the intensive margin. It is
interesting to note that all four variables are related to the inter-firm credit market. They all
intensify export contraction during the crisis when they are interacted with smallness. It is
also interesting to note that TPAYFOR reduces the export contraction of non-small firms, but
it intensifies the export contraction of small firms. Our results suggest that during the crisis,
inter-firm trade credit, both domestic and cross-border, tightened more for small exporters.
V. Conclusion
Official data on trade finance are valuable because they are rare. Aggregate data on various
measures of trade finance are available in Korea and we investigated their behavior during
31
the financial crisis of 2008-09. However, these data do not seem to support the view that
trade finance played an active role in causing the observed trade collapse. An interesting
phenomenon that we found is that cross-border accounts payable actually increased during
the financial crisis. This suggests that firms moved to the inter-firm credit market to tap an
alternative source of finance when bank credit tightened. Data on exports by the methods of
payment and our firm-level analysis are consistent with this interpretation.
The results on our firm-level analysis are somewhat mixed. Results are sensitive to
whether we look at all margins or only intensive margins. Some of our interesting findings
are that small exporters who were carrying large stocks of inventories or relied a lot on interfirm trade credit seem to have suffered more during the crisis. To get a more reliable answer,
we need to investigate further on the experience of small firms using data that contain both
listed and unlisted firms because small unlisted firms are more likely to be credit-constrained
in a financial crisis.
References
Alessandria, G., Kaboski, J.P. and Midrigan, V. (2010), “The Great Trade Collapse of 2008–
09: An Inventory Adjustment?” IMF Economic Review 58: 254–294.
Amiti, M. and D. Weinstein (2011), “Export and Financial shocks,” Quarterly Journal of
Economics 126: 1841-1877.
Beck, T. 2003. “Financial Dependence and International Trade," Review of International
Economics, 11(2): 296-316.
Braun, M. 2003. “Financial Contractibility and Asset Hardness," mimeo, Harvard University.
Behrens, K., Corcos, G., Mion, G., (2010), “Trade Crisis? What Trade Crisis?” CEPR
Discussion Paper No. 7956.
32
Bricongne, J.C., Fontagné, L., Gaulier, G., Taglioni, D., Vicard, V., (2012), “Firms and the
Global Crisis: French Exports in the Turmoil.” Journal of International Economics 87:
134–146.
Chauffour, Jean-Pierre, and Mariem Malouche (2011), Trade Finance during the Great
Trade Collapse. Washington, DC: World Bank Group.
Chor, Davin, and Kalina Manova (2012), “Off the Cliff and Back? Credit Conditions and
International Trade during the Global Financial Crisis,” Journal of International
Economics 87: 117–133.
Dell'Ariccia, G., Detragiache E., and Rajan R. 2008. “The Real Effect of Banking Crises,"
Journal of Financial Intermediation 17(1): 89-112.
Eaton, J., Kortum, S., Neiman, B., Romalis, J., (2010), “Trade and the Global Recession,”
University of Chicago, Mimeo.
Fisman, R., Love, I., 2003. “Trade Credit, Financial Intermediary Development, and Industry
Growth.” Journal of Finance 58: 353–374.
Hur, J., Raj, M., Riyanto, Y., (2006), “The Impact of Financial Development and Asset
Tangibility on Export.” World Development 34: 1728–1741.
Iacovone, Leonardo, and Veronika Zavacka. 2009. “Banking Crises and Exports: Lessons
from the Past.” Policy Research Working Paper 5016, World Bank, Washington, DC.
ICC (International Chamber of Commerce) (2009), “Rethinking Trade Finance 2009: An ICC
Global Survey.” ICC, Paris.
——— (2010), “Rethinking Trade Finance 2010: An ICC Global Survey.” ICC, Paris.
——— (2011), “Rethinking Trade and Finance 2011: ICC Global Survey on Trade and
Finance.” ICC, Paris.
IMF-BAFT (International Monetary Fund-Bankers’ Association for Finance and Trade)
(2009), “Trade Finance Survey: A Survey among Banks Assessing the Current Trade
33
Finance Environment,” Report by FImetrix for IMF and BAFT, Washington, DC.
IMF and BAFT-IFSA (International Monetary Fund and Bankers’ Association for Finance
and Trade-International Financial Services Association) (2010), “Trade Finance Services:
Current Environment & Recommendations: Wave 3.” Report by FImetrix for IMF and
BAFT-IFSA, Washington, DC.
Kletzer, K., Bardhan, P., 1987. “Credit markets and patterns of international trade.” Journal
of Development Economics 27: 57–70.
Kroszner, R.S., Laeven, L., Klingebiel, D., (2007), “Banking Crises, Financial Dependence,
and Growth,” Journal of Financial Economics 84, 187–228.
Levchenko, Andrei A., Logan T. Lewis, and Linda L. Tesar, (2010), “The Collapse of
International Trade during the 2008–2009 Crisis: In Search of the Smoking Gun.” IMF
Economic Review 58 (2): 214–53.
Minetti, R., Zhu, S. (2011), “Credit Constraints and Firm Export: Microeconomic Evidence
from Italy,” Journal of International Economics 83: 109–125.
Rajan, Raghuram G., and Luigi Zingales (1998), “Financial Dependence and Growth.”
American Economic Review 88 (3): 559–86.
Hwang, Sangyeon and Hyejoon Im, (2012), “Financial Shocks and Trade Finance: Evidence
from Korea.” Mimeo, Yeungnam University.
Zavacka, Veronika (2012), “The Bullwhip effect and the Great Trade Collapse.” EBRD
Working Paper No. 148.
34
Figure 1: Export price, quantity and value (in dollar)
export price
export quantity
export value
Indexes, July 2008 = 100
140
120
100
80
60
40
20
Jan/07
Jan/08
Jan/09
Jan/10
Jan/11
Source: Economic Statistics System, Bank of Korea
All variables are indexes. value = price × quantity. The indexes in July 2008 are normalized
to 100.
Figure 2: Export price, quantity and value (in Korean won)
exchange rate
export price
export quantity
export value
Indexes, July 2008 = 100
160
140
120
100
80
60
40
Jan/07
Jan/08
Jan/09
Jan/10
Jan/11
Source Economic Statistics System, Bank of Korea
All variables are indexes. value = price × quantity. The indexes in July 2008 are normalized
to 100. Price and value indexes in KRW are obtained by multiplying dollar indexes by
exchange rates.
35
Figure 3: Interest rates
bank lending rate for large companies
bank lending rate for SMEs
discount rate on commercial paper
9
8
7
% 6
5
4
3
janv./07
janv./08
janv./09
janv./10
janv./11
Source Economic Statistics System, Bank of Korea
Figure 4: Trade crisis?
exports/GDP
exports/industrial output
120
110
100
%
90
80
70
60
Q1 2007
Q1 2008
Q1 2009
Source Economic Statistics System, Bank of Korea
36
Q1 2010
Q1 2011
Figure 5: Foreign trade loans and exports
Billion KRW
trade loans
exports
trade loans/exports
60 000
0,60
50 000
0,50
40 000
0,40
30 000
0,30
20 000
0,20
10 000
0,10
0
janv.-07
janv.-08
janv.-09
janv.-10
janv.-11
Source: Economic Statistics System, Bank of Korea
Trade loans and exports are shown on the left axis. The value of exports was converted into
KRW using average monthly exchange rates. The ratio of foreign trade loans to exports are
shown on the right axis.
Figure 6: Documentary bills purchased and exports
bills purchased
exports
bills/exports
0,30
150
Biilion USD
0,25
100
0,20
0,15
50
0,10
0,05
0
Q1 2007
0,00
Q1 2008
Q1 2009
Q1 2010
Q1 2011
Source: Financial Supervisory Service of Korea
Observations are quarterly. Bills purchased and exports are shown on the left axis. The ratio of
foreign trade loans to exports are shown on the right axis. The amount of bills purchased was
converted to dollar value using exchange rates at the end of periods.
37
Figure 7: Domestic import usances and imports
import usances
imports
usances/imports
160
0,30
140
0,25
Billion USD
120
0,20
100
0,15
80
60
0,10
40
0,05
20
0
Q1 2007
0,00
Q1 2008
Q1 2009
Q1 2010
Q1 2011
Source: Financial Supervisory Service of Korea
Observations are quarterly. Import usances and imports are shown on the left axis. The ratio
of foreign trade loans to exports are shown on the right axis. The amount of bills purchased
was converted to dollar value using exchange rates at the end of periods.
Figure 8: External trade credit (Accounts receivable)
trade credits
exports
trade credits/exports
160
0,25
140
0,20
120
Billion USD
100
0,15
80
0,10
60
40
0,05
20
0
Q1 2007
0,00
Q1 2008
Q1 2009
Q1 2010
Q1 2011
Source: Economic Statistics System, Bank of Korea
Observations are quarterly. External short-term trade credits of private firms are shown on
the left axis and the ratio of the credits to exports are shown on the right axis.
38
Figure 9: External trade credit (Accounts payable)
trade credits
imports
trade credits/imports
150
0,20
Billion USD
0,15
100
0,10
50
0,05
0
Q1 2007
0,00
Q1 2008
Q1 2009
Q1 2010
Q1 2011
Source: Economic Statistics System, Bank of Korea
Observations are quarterly. External short-term trade credits of private firms are shown on
the left axis and the ratio of the credits to imports are shown on the right axis.
Figure 10 : Exports by methods of payment
Remittance
%
L/C
Collection
70
60
50
40
30
20
10
0
janv.-05 janv.-06 janv.-07 janv.-08 janv.-09 janv.-10 janv.-11
Source: Economic Statistics System, Bank of Korea through Korea Customs Service.
Remittance includes cash-in-advance and open-account and transactions. L/C refers to
transactions in which payments are made using letters of credit, and collection refers to
transactions in which payments are made using documentary collection by banks. The three
categories do not sum to 100 % as there are unclassified transactions.
39
Table 1
Top 20 merchandise exporters in year 2007
Rank
Company
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
SAMSUNG ELECTRONICS CO,.LTD
HYUNDAI MOTOR CO
LG ELECTRONICS INC.
HYUNDAI HEAVY INDUSTRIES CO.,LTD
LG DISPLAY CO., LTD.
KIA MOTORS CORPORATION
S-OIL CORP
SK HYNIC INC.
SAMSUNG HEAVY INDUSTRIES CO.,LTD
DAEWOO SHIPBUILDING & MARINE ENGINEERING CO., LTD.
POSCO
LG CHEM LTD
HYUNDAI MOBIS CO.,LTD
HYOSUNG CORPORATION
SAMSUNG SDI CO.,LTD
HYUNDAI MIPO DOCKYARD CO.,LTD
DOOSAN INFRACORE CO., LTD.
SAMSUNG ELECTRO-MECHANICS CO.,LTD
SAMSUNG TECHWIN CO.,LTD
STX OFFSHORE & SHIPBUILDING CO.,LTD.
Percent of
total
19.7
6.8
6.5
5.3
5.1
4.3
3.5
3.2
3.0
2.7
2.6
2.5
1.9
1.5
1.3
1.1
0.9
0.9
0.8
0.8
Source: WISEfn
40
Cumulative
percent
19.7
26.4
33.0
38.3
43.3
47.6
51.1
54.3
57.3
60.0
62.6
65.1
67.0
68.5
69.9
71.0
71.9
72.8
73.6
74.4
Large enterprise
SAMSUNG
HYUNDAI MOTOR
LG
HYUNDAI HEAVY INDUSTRIES
LG
HYUNDAI MOTOR
SK
SAMSUNG
DAEWOO SHIPBUILDING & MARINE ENGINEERING CO., LTD.
POSCO
LG
HYUNDAI MOTOR
HYOSUNG
SAMSUNG
HYUNDAI HEAVY INDUSTRIES
DOOSAN
SAMSUNG
SAMSUNG
STX
Table 2
Decomposition of mid-point export growth rates (year on year, percent)
Quarter
Extensive
positive
Extensive
negative
Extensive
margin
Intensive
positive
Intensive
negative
Intensive
margin
Total
2005q1
2005q2
2005q3
2005q4
2006q1
2006q2
2006q3
2006q4
2007q1
2007q2
2007q3
2007q4
2008q1
2008q2
2008q3
2008q4
2009q1
2009q2
2009q3
2009q4
0.02
0.17
0.07
0.07
0.10
0.02
0.04
0.04
0.03
0.05
0.10
0.05
0.06
0.09
0.10
0.07
0.04
0.07
0.10
0.03
-0.02
-0.04
-0.11
-0.03
-0.06
-0.15
-0.03
-0.04
-0.10
-0.02
-0.04
-0.04
-0.04
-0.02
-0.04
-0.04
-0.03
-0.02
-0.02
-0.01
0.00
0.13
-0.04
0.04
0.05
-0.13
0.01
0.00
-0.08
0.03
0.06
0.01
0.02
0.07
0.06
0.03
0.01
0.05
0.08
0.02
10.57
7.12
7.12
12.12
9.90
10.18
12.41
9.96
11.05
13.86
14.11
16.89
21.53
25.63
25.90
19.87
12.70
11.77
18.34
19.48
-5.01
-9.36
-8.24
-6.06
-4.34
-3.49
-4.07
-5.61
-4.47
-3.14
-4.01
-3.86
-3.07
-1.31
-2.83
-5.85
-9.36
-11.11
-8.61
-8.71
5.56
-2.24
-1.12
6.06
5.56
6.69
8.34
4.35
6.58
10.72
10.10
13.03
18.46
24.31
23.07
14.02
3.34
0.65
9.73
10.77
5.56
-2.11
-1.17
6.10
5.61
6.56
8.35
4.35
6.50
10.75
10.15
13.03
18.48
24.38
23.13
14.05
3.35
0.71
9.82
10.79
Source: WISEfn, Bank of Korea and own calculations
41
export value in
KRW
(BOK, goods)
-0.98
-3.50
3.23
6.65
8.63
8.94
10.09
2.61
9.79
10.65
7.70
17.59
21.66
36.06
39.57
26.65
5.60
-2.87
-3.15
-1.32
KRW/USD
exchange rate
-12.75
-13.25
-10.91
-5.21
-4.40
-5.71
-7.21
-9.51
-3.95
-2.23
-2.81
-1.90
1.82
9.41
14.49
48.03
48.04
26.75
16.77
-14.25
Table 3
Effects of the crisis on growth rates by industry
Industry
fixed effects
food and beverages
textile, apparel and footwear
wood, paper and printing
chemicals, chemical products and oil refinery
pharmaceuticals & medical chemicals
rubber and plastic products
non-metallic mineral products
basic metal products
fabricated metal products
electronic components
computers and communication equipment
medical, precision and optical instruments
electrical equipment
domestic appliances
other machinery
motor vehicles and parts
ships and other transport equipment
furniture and other manufacturing
22.5
3.5
4.4
3.7
34.1
15.7
25.4
−25.9
−8.6
3.5
14.1
12.3
−8.2
33.7
−13.4
−38.1
9.6
3.7
Note: The fixed effects of the crisis are calculated from a regression of mid-point growth rates on industry dummies and industry dummies interacted with the crisis dummy that takes the
value of 1 in Q1 2009 and Q2 2009. They are normalized such that the weighted average equals zero.
42
Table 4
Means and standard deviations of financial variables by industry
Industry
food and beverages
textile, apparel and footwear
wood, paper and printing
chemicals, chemical products and oil refinery
pharmaceuticals & medical chemicals
rubber and plastic products
non-metallic mineral products
basic metal products
fabricated metal products
electronic components
computers and communication equipment
medical, precision and optical instruments
electrical equipment
domestic appliances
other machinery
EXTFIN
-59.7
(1803.8)
48.7
(6999.1)
-75.2
(6768.6)
-27.5
(1820.0)
-26.3
(2594.2)
-19.4
(5946.7)
-25.5
(3741.0)
14.1
(3528.1)
31.5
(4455.8)
20.2
(13287.8)
-8.7
(11995.6)
10.6
(1334.5)
15.5
(6921.3)
-30.8
(9994.1)
25.8
(35283.0)
TANG
36.7
(16.0)
28.5
(19.6)
49.6
(19.1)
39.4
(16.9)
29.0
(14.5)
34.5
(15.8)
34.9
(19.0)
39.0
(16.0)
33.6
(14.6)
32.3
(18.4)
18.0
(14.4)
23.0
(15.4)
28.1
(15.1)
29.6
(16.5)
26.6
(15.5)
TPAY
10.0
(7.6)
8.7
(13.4)
6.8
(29.7)
9.5
(27.6)
10.1
(9.6)
9.6
(17.7)
10.8
(8.6)
7.9
(7.8)
13.1
(17.4)
9.0
(10.6)
11.3
(14.8)
12.1
(10.7)
11.8
(8.8)
12.4
(11.6)
15.6
(12.3)
43
TREC
12.3
(9.1)
14.4
(12.4)
17.0
(8.3)
15.7
(36.0)
35.6
(19.8)
20.3
(14.5)
22.0
(13.3)
14.7
(8.3)
20.1
(11.1)
15.7
(11.4)
17.8
(29.0)
22.9
(17.3)
18.4
(14.2)
18.1
(10.4)
24.0
(21.5)
INV
9.5
(12.4)
20.6
(22.2)
10.7
(8.0)
11.3
(49.4)
14.6
(16.8)
8.7
(11.0)
13.4
(11.5)
14.3
(9.1)
13.1
(26.4)
10.8
(13.7)
12.3
(17.1)
14.0
(13.6)
11.4
(13.3)
9.6
(14.0)
14.4
(15.2)
INTCOST
5.4
(264.0)
4.2
(89.3)
10.7
(94.5)
4.4
(113.7)
2.4
(107.6)
0.7
(75.3)
7.1
(187.3)
10.0
(217.8)
8.3
(66.0)
1.8
(210.9)
-1.2
(200.9)
0.2
(44.5)
1.4
(185.5)
-2.2
(41.0)
1.3
(176.8)
STBFOR
4.7
(14.8)
8.2
(22.8)
7.6
(11.2)
4.0
(6.1)
3.2
(9.4)
4.5
(4.1)
3.3
(5.5)
6.9
(8.4)
5.4
(10.2)
4.7
(13.3)
4.4
(16.5)
7.8
(14.1)
4.7
(6.8)
2.2
(2.5)
5.0
(18.5)
TPAYFOR
6.6
(26.6)
4.1
(93.4)
6.3
(1968.4)
5.4
(932.2)
12.9
(1947.1)
5.7
(13.4)
12.8
(136.5)
3.9
(229.2)
3.0
(14.2)
3.4
(46.3)
3.9
(3732.3)
4.1
(340.5)
5.4
(401.3)
3.8
(5.3)
3.8
(460.9)
TRECFOR
6.6
(13.1)
8.4
(135.7)
6.5
(10.4)
8.0
(38.3)
12.8
(129.2)
10.1
(10.7)
13.1
(158.3)
6.5
(12.2)
13.4
(38.6)
12.4
(34.6)
14.6
(1097.4)
19.1
(55.9)
13.8
(62.5)
12.3
(12.4)
20.3
(38.2)
Table 4 (continued)
Medians and standard deviations of financial variables by industry
Industry
motor vehicles and parts
ships and other transport equipment
furniture and other manufacturing
Average median
EXTFIN
-14.7
(1046.5)
-27.6
(2109.0)
102.3
(410.4)
TANG
37.7
(15.1)
35.1
(16.0)
23.8
(12.0)
TPAY
15.4
(9.5)
11.4
(10.7)
12.8
(6.1)
TREC
17.7
(8.6)
19.9
(12.1)
29.3
(15.3)
INV
5.7
(7.5)
7.2
(8.8)
20.6
(15.6)
INTCOST
4.0
(115.1)
-1.0
(72.4)
10.4
(536.1)
STBFOR
2.1
(4.6)
3.4
(6.2)
6.8
(4.2)
TPAYFOR
1.4
(37.2)
1.0
(87.3)
9.3
(29.2)
TRECFOR
10.6
(74.2)
10.6
(14.1)
28.1
(18.9)
-2.6
32.2
11.0
19.8
12.3
3.8
4.9
5.4
12.6
(6668.9)
(16.1)
(13.0)
(15.2)
(15.9)
(155.5)
(9.9)
(583.4)
(108.6)
41.3
7.3
2.4
5.6
3.9
4.1
1.9
3.3
5.5
Average standard deviation
Standard deviation of medians across industries
Source: WISEfn and own calculations.
Note: For each industry, the numbers on the upper row are the medians of financial variables, and those on the lower row in the parentheses are the standard deviations. In each industry, the
median and the standard deviation for each variable are calculated from nine-year averages (2001-2009) of the variable for firms belonging to the industry. Average median is equal to the
average of medians across industries. Average standard deviation is the average of standard deviations across industries.
44
Table 5
Effects of financial variables on the export collapse
VAR
EXTFIN
TANG
TPAY
TREC
INV
INTCOST
STBFOR
TPAYFOR
TRECFOR
ASSET
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
0.000
0.181*
0.084
0.000
−0.009
0.009
−0.095
−0.000
−0.006***
−1.014
(1.99)
(0.89)
(0.01)
(−0.52)
(1.02)
(−0.71)
(−1.42)
(−3.94)
(−0.41)
−0.036
−5.358
(−0.55)
(−0.78)
(1.26)
VAR∗CRISIS
−0.002
***
(−2.94)
ln LABORPROD
9.703
***
(4.38)
ln EXPSHARE
−1.994
0.152
0.119
0.131
−0.380
−0.026
−0.036
0.006
(0.86)
(0.35)
(0.52)
(−1.22)
(−1.68)
(−0.06)
(3.68)
10.313
***
(4.34)
***
−2.217
9.699
***
(4.40)
***
−1.963
9.768
***
(4.43)
***
−1.974
9.541
***
(4.61)
***
−1.993
9.725
***
(4.33)
***
−1.990
15.435
***
(4.68)
***
−2.647
8.696
***
**
(2.81)
***
−1.850
9.467
***
(4.14)
**
−1.795
9.592***
(4.10)
***
−2.096***
(−2.94)
(−3.06)
(−3.00)
(−3.01)
(−2.96)
(−2.96)
(−2.92)
(−2.16)
(−2.97)
(−3.57)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,775
14,775
14,775
14,775
14,775
14,775
7,230
9,946
12,642
14,775
R2
0.050
0.051
0.049
0.049
0.049
0.050
0.087
0.071
0.056
0.049
Industry-Quarter
fixed effects
Note: The dependent variable is GROWTHMDP. Each column reports regression results when VAR is equal to the financial variable on the top row. CRISIS is a dummy variable that takes
the value of 1 in Q1 2009 and Q2 2009. ASSET is a dummy that takes the value of 1 when a firm belongs to the bottom 20 percent in asset size. The numbers in the parentheses are t-ratios
calculated using robust standard errors (clustering by industries). Constants are not reported.
45
Table 6
Effects of financial variables interacted with an asset-size dummy
VAR
EXTFIN
TANG
TPAY
TREC
INV
INTCOST
STBFOR
TPAYFOR
TRECFOR
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
0.001
0.186
0.051
−0.009
−0.012
0.010
−0.014
−0.000
−0.006***
(2.12)
(0.77)
(−0.27)
(−0.59)
(0.87)
(−0.10)
(−1.37)
(−3.85)
(1.30)
VAR∗CRISIS
VAR∗ASSET
VAR∗ASSET∗CRISIS
ln LABORPROD
ln EXPSHARE
−0.002
0.166
0.242
0.235
−0.218
−0.031
0.356
(−2.49)
(1.02)
(0.64)
(0.78)
(−0.72)
(−1.67)
(0.75)
0.001
−0.049
0.276
0.066
0.063
−0.004
−0.922
(1.74)
(−0.71)
(1.47)
(0.54)
(0.42)
(−0.25)
(−2.50)
(−3.25)
(−1.07)
−0.002
−0.124
−0.829
−0.425
−0.916
0.016
−2.121
0.027
−0.129
(−0.84)
(−0.54)
(−1.62)
(−1.44)
(−2.32)
(0.61)
(−4.37)
(0.11)
(−1.21)
9.703***
10.121***
9.877***
9.782***
9.520***
9.735***
14.919***
8.901***
9.332***
(4.43)
(4.12)
(4.29)
(4.20)
(4.43)
(4.37)
(4.48)
(2.83)
(4.08)
−1.994
**
***
−2.343
***
−1.797
**
−1.933
***
−2.025
**
***
−1.996
***
−3.068
0.006
***
(3.65)
**
***
***
−0.013
−2.002
0.005
(0.11)
***
**
−0.048
−1.928***
(−2.94)
(−3.39)
(−2.77)
(−3.37)
(−2.98)
(−3.01)
(−3.63)
(−2.34)
(−3.06)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,775
14,775
14,775
14,775
14,775
14,775
7,230
9,946
12,642
R2
0.051
0.051
0.050
0.049
0.050
0.050
0.090
0.072
0.057
Industry-Quarter
fixed effects
Note: The dependent variable is GROWTHMDP. Each column reports regression results when VAR is equal to the financial variable on the top row. CRISIS is a dummy variable that takes
the value of 1 in Q1 2009 and Q2 2009. ASSET is a dummy that takes the value of 1 when the firm belongs to the bottom 20 percent in asset size. The numbers in the parentheses are t-ratios
calculated using robust standard errors (clustering by industries). Constants are not reported.
46
Table 7
Effects of financial variables on the export collapse: robustness
EXTFIN
(20)
Clustering by
firms
VAR
−0.002
***
***
(4.67)
ln EXPSHARE
(23)
(24)
(25)
Within
Effects on
Finer
Effects on
Clustering by
−0.000
0.000
9.703
(22)
net growth
(−2.67)
ln LABORPROD
(21)
a
(1.33)
VAR∗CRISIS
TPAYFOR
−1.994
0.000
(−0.79)
(0.25)
−0.002
*
−0.002
(−1.85)
(−3.36)
8.670
**
9.791
(2.21)
***
−20.475
***
−2.453
(finer ind.) b
0.000
0.000
(0.07)
−0.002
**
(−2.42)
10.441
***
(3.53)
***
net growth
class b
(1.31)
***
(3.30)
***
industry
−2.260
−0.002
**
(−2.17)
10.491
***
(3.17)
***
−2.714
firms
(27)
(28)
(29)
Within
Effects on
Finer
Effects on
net growth
industry
a
class
b
net growth
(finer ind.)
−0.000
−0.002***
−0.000
−0.000
−0.000
(−1.26)
(−15.59)
(−1.20)
(−1.63)
(−1.39)
**
**
0.006**
0.006
***
(2.97)
8.696
***
(3.65)
***
(26)
−1.850
*
0.007
(2.00)
(2.35)
*
*
0.007
6.578
8.132
(1.84)
***
−20.308
(2.95)
10.087
(1.92)
***
−2.403
0.006
(2.12)
***
(2.96)
**
−2.248
9.521**
(2.16)
***
−2.732***
(−3.81)
(−5.19)
(−3.57)
(−3.69)
(−4.39)
(−2.80)
(−3.68)
(−2.51)
(−2.97)
(−3.46)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,775
14,775
14,396
14,775
14,396
9,946
9,946
9,656
9,946
9,656
R2
0.050
0.202
0.039
0.112
0.098
0.072
0.245
0.052
0.147
0.125
Industry-Quarter
fixed effects
b
Note: The dependent variable is GROWTHMDP except for (22), (24), (27) and (29) where the dependent variable is net growth defined as the mid-point growth rate of
exports minus the mid-point growth rate of domestic sales.. CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the parentheses
are t-ratios calculated using robust standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms.
b
t-ratios calculated using robust standard errors clustered by 70 industries
47
Table 8
Effects of financial variables interacted with an asset-size dummy in the intensive margin: robustness
INV
STBFOR
(30)
Clustering by
firms
(31)
(32)
(33)
(34)
(35)
Within
Effects on
Finer
Effects on
Clustering by
net growth
industry
net growth
a
class
VAR
VAR∗CRISIS
VAR∗ASSET
VAR∗ASSET∗CRISIS
(38)
(39)
Within
Effects on
Finer
Effects on
net growth
industry
a
b
class
b
net growth
(finer ind.)
−0.012
−0.064
−0.014
−0.070
−0.014
0.575
−0.171
0.005
−0.179
(−0.49)
(−1.10)
(−1.69)
(−0.45)
(−1.76)
(−0.10)
(1.42)
(−0.99)
(0.03)
(−1.06)
−0.218
−0.118
−0.118
−0.213
−0.143
0.356
0.114
0.383
0.584
0.395
(−0.69)
(−0.44)
(−0.36)
(−0.67)
(−0.48)
(0.89)
(1.34)
(0.93)
−0.073
−0.922
−0.959
−1.404**
0.625
0.256
0.021
0.009
(0.40)
(0.83)
(0.12)
(0.05)
−0.916
**
9.520
***
(4.58)
ln EXPSHARE
(finer ind.)
firms
(37)
−0.116
(−2.05)
ln LABORPROD
b
(36)
−2.025
*
−0.819
−1.120
(−1.83)
(−2.92)
8.995
**
9.389
(2.22)
***
−20.474
***
***
(3.12)
***
−2.558
−0.893
**
(−2.06)
10.187
***
(3.57)
***
−2.342
(0.36)
(0.79)
*
−0.149
−1.361
(−0.39)
(−1.93)
(−0.25)
(−2.47)
(−1.59)
(−2.04)
−0.961
−2.121
−2.078
−0.512
−1.990
*
−0.389
(−0.62)
(−1.96)
(−0.38)
**
(−2.06)
9.939
**
(3.02)
***
−2.906
***
(−2.69)
14.919
***
(4.82)
***
−3.068
***
(−3.30)
12.675
13.688
(1.81)
***
−18.976
**
***
(3.30)
***
−3.929
15.427
***
(4.16)
***
−3.161
13.586***
(3.05)
***
−3.825***
(−3.68)
(−5.30)
(−3.63)
(−3.69)
(−4.43)
(−3.94)
(−3.18)
(−3.55)
(−3.78)
(−3.54)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,775
14,775
14,396
14,775
14,396
7,230
7,230
7,046
7,230
7,046
R2
0.050
0.202
0.039
0.112
0.098
0.091
0.263
0.073
0.199
0.173
Industry-Quarter
fixed effects
b
Note: The dependent variable is GROWTHMDP except for (32), (34), (37) and (39) where the dependent variable is net growth defined as the mid-point growth rate of exports minus the
mid-point growth rate of domestic sales. CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the parentheses are t-ratios calculated using robust
standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms; b t-ratios calculated using robust standard errors clustered by 70 industries
48
Appendix
Table A1
Effects of financial variables in the intensive margin
VAR
EXTFIN
TANG
TPAY
TREC
INV
INTCOST
STBFOR
TPAYFOR
TRECFOR
ASSET
(A1)
(A2)
(A3)
(A4)
(A5)
(A6)
(A7)
(A8)
(A9)
(A10)
0.000
0.173**
−0.026
−0.051
0.001
0.012
0.026
0.000
0.002
0.498
(2.29)
(−0.44)
(−1.18)
(0.06)
(1.14)
(0.24)
(0.49)
(0.49)
(0.26)
0.019
0.031
−15.220**
(2.18)
VAR∗CRISIS
−0.002
***
(−2.94)
ln LABORPROD
8.037
***
(4.42)
ln EXPSHARE
−2.393
0.297
**
(2.13)
8.641
(0.53)
***
(4.47)
***
−2.620
0.184
8.023
(0.06)
***
(4.45)
***
−2.379
7.927
***
(4.38)
***
−2.442
−0.110
−0.027
*
−0.143
0.005
(−0.51)
(−1.86)
(−0.21)
(3.31)
7.983
***
(4.63)
***
−2.384
8.042
***
(4.38)
***
−2.394
12.567
***
(4.89)
***
−2.831
6.173
***
(0.72)
**
(2.34)
***
−2.044
8.012
(−2.66)
***
(4.07)
***
−2.179
7.890
(4.17)
***
−2.464***
(−5.77)
(−6.17)
(−5.77)
(−5.51)
(−5.82)
(−5.89)
(−3.70)
(−3.60)
(−4.75)
(−7.01)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,154
14,154
14,154
14,154
14,154
14,154
6,968
9,682
12,288
14,154
R2
0.068
0.069
0.066
0.067
0.067
0.067
0.099
0.086
0.066
0.067
Industry-Quarter
fixed effects
Note: The dependent variable is GROWTHMDP. Each column reports regression results when VAR is equal to the financial variable on the top row. CRISIS is a dummy variable that takes
the value of 1 in Q1 2009 and Q2 2009. ASSET is a dummy that takes the value of 1 when the firm belongs to the bottom 20 percent in asset size. The numbers in the parentheses are t-ratios
calculated using robust standard errors (clustering by industries). Constants are not reported.
49
Table A2
Effects of financial variables in the intensive margin: robustness
TANG
INTCOST
(A11)
(A12)
(A13)
(A14)
(A15)
(A16)
(A17)
(A18)
(A19)
(A20)
Clustering by
Within
Effects on
Finer
Effects on
Clustering by
Within
Effects on
Finer
Effects on
net growth
industry
net growth
net growth
industry
firms
a
class
VAR
VAR∗CRISIS
ln LABORPROD
0.173***
0.088
(3.30)
(0.50)
*
0.335
(1.78)
(2.61)
0.297
8.640
***
(5.02)
ln EXPSHARE
−2.620
5.611
**
−19.118
0.089
0.012
(1.37)
(2.74)
(1.67)
(1.23)
0.048
−0.027
7.894
0.219
(1.54)
***
(2.92)
***
class
0.165
(0.45)
(1.87)
***
(finer ind.)
firms
b
0.099
0.098
*
b
a
−3.018
9.367
(0.18)
***
(3.45)
***
−2.892
8.440
(−2.04)
**
(2.60)
***
−3.240
**
8.042
***
(4.65)
***
−2.394
net growth
(finer ind.)
0.009
0.016
0.025***
0.030***
(0.52)
(1.53)
(2.84)
(3.37)
−0.017
−0.030
−0.039
(−1.01)
(−1.58)
(−2.30)
5.327
*
7.591
(1.76)
***
b
−19.144
***
(2.93)
***
−2.908
9.016
**
***
(3.25)
***
−2.723
−0.039
(−1.48)
8.340**
(2.59)
***
−3.194***
(−6.09)
(−5.67)
(−7.75)
(−6.26)
(−5.95)
(−5.58)
(−5.96)
(−7.59)
(−5.91)
(−5.96)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,154
14,154
13,784
14,154
13,784
14,154
14,154
13,784
14,154
13,784
R2
0.069
0.206
0.046
0.132
0.105
0.067
0.206
0.047
0.132
0.107
Industry-Quarter
fixed effects
b
Note: The dependent variable is GROWTHMDP except for (A22), (A24), (A27) and (A29) where the dependent variable is net growth defined as the mid-point growth
rate of exports minus the mid-point growth rate of domestic sales.CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the
parentheses are t-ratios calculated using robust standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms.
b
t-ratios calculated using robust standard errors clustered by 70 industries
50
Table A3
Effects of financial variables in the intensive margin: robustness
ASSET
(A21)
(A22)
(A23)
(A24)
(A25)
Clustering by
Within
Effects on
Finer
Effects on
net growth
industry
firms
a
class
VAR
9.286*
0.498
(0.21)
VAR∗CRISIS
(1.85)
−15.220
**
(−2.39)
ln LABORPROD
7.890
***
(4.52)
ln EXPSHARE
−2.464
0.045
−12.092
(−2.30)
5.192
7.384
(1.74)
***
−18.699
−18.977
**
(2.84)
***
−3.034
(−0.40)
−16.715
**
(−2.60)
8.762
***
(3.15)
***
−2.807
−18.718**
(−2.09)
8.000**
(2.51)
***
−3.348***
(−5.28)
(−6.10)
(−8.37)
(−5.55)
(−5.76)
Yes
Yes
Yes
Yes
Yes
Obs.
14,154
14,154
13,784
14,154
13,784
R2
0.067
0.206
0.047
0.131
0.106
Industry-Quarter
fixed effects
b
−0.855
(0.04)
**
(−2.80)
*
(finer ind.)
0.084
(0.02)
**
net growth
b
Note: The dependent variable is GROWTHMDP except for (A32) and (A34) where the dependent variable is net growth defined as the mid-point growth rate of exports
minus the mid-point growth rate of domestic sales. CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the parentheses are tratios calculated using robust standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms.
b
t-ratios calculated using robust standard errors clustered by 70 industries
51
Table A4
Effects of financial variables interacted with an asset-size dummy in the intensive margin
VAR
VAR∗CRISIS
VAR∗ASSET
VAR∗ASSET∗CRISIS
ln LABORPROD
ln EXPSHARE
EXTFIN
TANG
TPAY
TREC
INV
INTCOST
STBFOR
TPAYFOR
TRECFOR
(A26)
(A27)
(A28)
(A29)
(A30)
(A31)
(A32)
(A33)
(A34)
0.000
0.173**
−0.039
−0.061
−0.004
0.010
0.047
0.001
0.002
(2.45)
(2.35)
(−0.65)
(−1.22)
(−0.22)
(0.91)
(0.44)
(0.49)
(0.55)
−0.002**
0.326**
0.336
0.200
0.070
−0.031*
0.359
0.005***
0.055
(−2.46)
(2.50)
(0.85)
(0.61)
(0.31)
(−1.99)
(0.79)
(3.22)
(1.83)
0.010
−0.309
−0.010
(0.51)
(−0.84)
(−3.92)
(−0.34)
0.003
−3.019
−0.449
−0.635**
0.010
0.158
(0.84)
(0.19)
(0.86)
−0.002
−0.316
−1.177
(−0.95)
(−1.76)
(−2.78)
(−3.07)
(−2.84)
(0.09)
(−4.32)
(−3.09)
(−2.79)
8.008***
8.565***
8.008***
7.836***
8.028***
8.020***
12.189***
6.413**
7.810***
(4.48)
(4.31)
(4.25)
(4.13)
(4.58)
(4.39)
(4.74)
(2.38)
(3.86)
−2.394
−2.669
***
−2.349
0.152
−0.020
0.000
***
0.068
***
(0.75)
**
***
−0.746
−2.446
(1.15)
***
***
−1.206
−2.363
**
***
−2.383
***
−3.028
***
***
−2.273
***
***
−2.368***
(−5.78)
(−7.23)
(−6.20)
(−6.35)
(−5.74)
(−5.86)
(−4.06)
(−4.09)
(−4.62)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,154
14,154
14,154
14,154
14,154
14,154
6,968
9,682
12,288
R2
0.068
0.069
0.067
0.068
0.068
0.067
0.103
0.088
0.068
Industry-Quarter
fixed effects
Note: The dependent variable is GROWTHMDP. Each column reports regression results when VAR is equal to the financial variable on the top row. CRISIS is a dummy variable that takes
the value of 1 in Q1 2009 and Q2 2009. ASSET is a dummy that takes the value of 1 when the firm belongs to the bottom 20 percent in asset size. The numbers in the parentheses are t-ratios
calculated using robust standard errors (clustering by industries). Constants are not reported.
52
Table A5
Effects of financial variables interacted with an asset-size dummy in the intensive margin: robustness
TPAY
TREC
(A35)
(A36)
(A37)
(A38)
(A39)
(A40)
(A41)
(A42)
(A43)
(A44)
Clustering by
Within
Effects on
Finer
Effects on
Clustering by
Within
Effects on
Finer
Effects on
net growth
industry
net growth
net growth
industry
firms
a
class
VAR
VAR∗CRISIS
VAR∗ASSET
net growth
(finer ind.)
0.019
−0.023
0.043
−0.061
−0.021
−0.074*
−0.055
−0.064
(−0.77)
(−1.11)
(0.29)
(−0.43)
(0.47)
(−1.43)
(−0.57)
(−2.00)
(−0.94)
(−1.21)
0.336
0.419
0.356
0.413
0.279
0.200
0.341
0.172
0.305
0.182
(1.08)
(1.03)
(0.81)
(0.88)
(0.56)
(0.86)
(0.88)
(0.49)
(0.85)
(0.42)
0.158
0.390
0.121
0.156
0.099
0.068
0.190
0.051
0.069
0.042
(0.81)
(0.87)
−0.731
−1.518
(−1.59)
(−2.62)
−1.177
**
8.008
***
(4.59)
ln EXPSHARE
class
b
−0.049
(−2.42)
ln LABORPROD
(finer ind.)
firms
b
−0.039***
(0.95)
VAR∗ASSET∗CRISIS
b
a
−2.350
5.193
*
7.486
(1.77)
***
−18.919
(0.71)
**
***
(2.89)
***
−2.913
−1.184
(0.72)
***
(−2.72)
8.908
***
(3.16)
***
−2.641
−1.484
(0.72)
**
(−2.24)
8.145
**
(2.54)
***
−3.178
−0.746
(−2.62)
7.836
***
(4.48)
***
−2.446
(0.51)
(0.84)
***
−0.514
−0.804
(−2.05)
(−2.49)
5.475
*
7.320
(1.78)
***
(0.59)
*
−18.850
**
**
(2.79)
***
−3.015
−0.766
(0.35)
**
(−2.40)
8.756
***
(3.11)
***
−2.728
−0.772*
(−1.95)
7.998**
(2.50)
***
−3.259***
(−5.19)
(−6.43)
(−7.78)
(−5.27)
(−5.70)
(−5.26)
(−6.17)
(−7.83)
(−4.82)
(−4.99)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,154
14,154
13,784
14,154
13,784
14,154
14,154
13,784
14,154
13,784
R2
0.067
0.206
0.047
0.131
0.106
0.068
0.206
0.047
0.132
0.106
Industry-Quarter
fixed effects
b
Note: The dependent variable is GROWTHMDP except for (A37), (A39), (A42) and (A44) where the dependent variable is net growth defined as the mid-point growth
rate of exports minus the mid-point growth rate of domestic sales. CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the
parentheses are t-ratios calculated using robust standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms; b t-ratios calculated using robust standard errors clustered by 70 industries
53
Table A6
Effects of financial variables interacted with an asset-size dummy in the intensive margin: robustness
INV
STBFOR
(A45)
(A46)
(A47)
(A48)
(A49)
(A50)
(A51)
(A52)
(A53)
(A54)
Clustering by
Within
Effects on
Finer
Effects on
Clustering by
Within
Effects on
Finer
Effects on
net growth
industry
net growth
net growth
industry
firms
a
class
VAR
VAR∗CRISIS
VAR∗ASSET
net growth
(finer ind.)
−0.058*
−0.007
−0.063**
0.047
0.346
−0.123
0.053
−0.140
(−0.21)
(−0.72)
(−1.78)
(−0.31)
(−2.08)
(0.40)
(0.90)
(−0.83)
(0.43)
(−1.06)
0.070
0.030
0.246
0.157
0.307
0.359
0.218
0.405
0.558
0.385
(0.26)
(0.17)
(1.16)
(0.61)
(1.52)
(0.95)
(0.63)
(0.94)
(1.39)
(0.97)
0.152
0.449
0.119
0.137
0.073
−0.309
0.526
−0.616
−0.319
−0.597
−1.206
(1.81)
***
8.028
***
(4.58)
ln EXPSHARE
class
b
−0.006
(−2.83)
ln LABORPROD
(finer ind.)
firms
b
−0.004
(1.20)
VAR∗ASSET∗CRISIS
b
a
−2.363
−0.958
(0.85)
**
(−2.23)
(−3.11)
5.530
7.306
(1.63)
***
−18.912
−1.500
(1.13)
***
***
(2.87)
***
−2.932
−1.303
(0.52)
***
(−2.68)
8.902
***
(3.24)
***
−2.674
−1.461
**
(−2.57)
7.894
**
(2.44)
***
−3.217
(−0.92)
(0.79)
−3.019
−2.595
***
(−3.98)
12.189
***
(4.41)
***
−3.028
***
(−2.93)
6.414
−19.354
(−0.64)
(−1.03)
−1.616
*
−2.471
−1.158
(−2.06)
(−2.36)
10.770
(1.04)
***
(−1.29)
***
(3.57)
***
−3.719
13.185
**
***
(4.03)
***
−3.099
(−1.11)
10.968***
(2.73)
***
−3.579***
(−5.22)
(−6.23)
(−7.65)
(−5.39)
(−5.61)
(−4.18)
(−5.29)
(−4.06)
(−4.61)
(−3.96)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
14,154
14,154
13,784
14,154
13,784
6,968
6,968
6,787
6,968
6,787
R2
0.068
0.206
0.047
0.132
0.106
0.103
0.264
0.076
0.220
0.180
Industry-Quarter
fixed effects
b
Note: The dependent variable is GROWTHMDP except for (A47), (A49), (A52) and (A54) where the dependent variable is net growth defined as the mid-point growth
rate of exports minus the mid-point growth rate of domestic sales. CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the
parentheses are t-ratios calculated using robust standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms; b t-ratios calculated using robust standard errors clustered by 70 industries
54
Table A7
Effects of financial variables interacted with an asset-size dummy in the intensive margin: robustness
TPAYFOR
TRECFOR
(A55)
(A56)
(A57)
(A58)
(A59)
(A60)
(A61)
(A62)
(A63)
(A64)
Clustering by
Within
Effects on
Finer
Effects on
Clustering by
Within
Effects on
Finer
Effects on
net growth
industry
net growth
net growth
industry
firms
a
class
VAR
0.001
(0.42)
VAR∗CRISIS
VAR∗ASSET
VAR∗ASSET∗CRISIS
(0.50)
0.000
(0.48)
(3.68)
(1.87)
(2.24)
(2.99)
−0.010
0.008
−0.001
−0.007
0.007
(−1.45)
(0.79)
−0.449
−0.525
***
6.413
***
(3.24)
ln EXPSHARE
(−7.16)
−2.273
**
(−2.69)
−18.900
0.005
**
class
b
0.002
−0.008***
0.002
0.002
(0.54)
(−3.47)
(0.49)
(0.57)
net growth
(finer ind.)
(0.66)
**
0.055
0.022
0.051
0.063
0.057
(2.20)
(1.54)
(0.63)
(1.46)
(2.22)
(1.53)
−0.002
−0.020
−0.042
−0.000
−0.023
−0.005
(−0.06)
(−0.37)
(−0.44)
(−0.00)
(−0.39)
(−0.10)
−0.441
−0.433
−0.404
−0.635
−0.622
−0.743
−0.613
−0.730***
−2.567
***
(−3.41)
6.647
**
(2.10)
***
−2.361
***
(−3.01)
6.317
−2.574
(−2.24)
7.810
(1.48)
***
**
***
(4.12)
***
−2.368
**
(−2.56)
(−3.41)
**
**
7.400
7.536
(2.48)
***
***
−22.549
(2.42)
***
−2.779
***
(−2.76)
8.735
***
(2.97)
***
−2.678
(−3.18)
8.097**
(2.37)
***
−3.058***
(−4.22)
(−3.41)
(−3.99)
(−4.13)
(−3.99)
(−4.73)
(−4.94)
(−5.26)
(−4.53)
(−4.83)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Obs.
9,682
9,682
9,396
9,682
9,396
12,288
12,288
11,947
12,288
11,947
R2
0.088
0.244
0.057
0.167
0.132
0.068
0.211
0.048
0.137
0.109
Industry-Quarter
fixed effects
b
0.002
(−1.02)
**
(1.61)
***
0.005
(0.59)
***
firms
b
(−0.21)
6.119
(0.55)
***
**
(−2.79)
1.986
(finer ind.)
0.001
0.005
(−3.14)
ln LABORPROD
0.000
*
0.005
***
−0.001
b
a
Note: The dependent variable is GROWTHMDP except for (A57), (A59), (A62) and (A64) where the dependent variable is net growth defined as the mid-point growth
rate of exports minus the mid-point growth rate of domestic sales. CRISIS is a dummy variable that takes the value of 1 in Q1 2009 and Q2 2009. The numbers in the
parentheses are t-ratios calculated using robust standard errors clustered by 18 industries unless noted otherwise. Constants are not reported.
a
t-ratios calculated using robust standard errors clustered by firms; b t-ratios calculated using robust standard errors clustered by 70 industries
55