ABSTRACT
Exchange rate effects manufacturing sector through investment and balance sheet channels. The second channel seems to be dominant for Turkey where an increase in exchange rate increases the costs of firms and may diminish their profits. A possible common explanation for this phenomenon is the high degree of liability dollarization, supply side. That is apparent from the real sector balance sheet items especially when the capital inflows are reversed and domestic currency depreciates. On the other hand, it is not a homogenous effect for importing and exporting sectors. It depends on the production and financial dependency structure of each sector. Taking into account the importance of negativities in micro basis may convert to a macro scale negativity in the long-run. This study provides sectoral liability dollarization ratios during the years 2009-2012 for the Turkish economy. The ratios are calculated for total companies, total manufacturing and 22 sub-sectors in manufacturing and they are compared according to their total, short-run and long-run liability structures. The scales of the companies are taken into account and possible diversifications among the sectors in manufacturing are discussed. Possible effects of the domestic currency depreciation on the liability dollarization of the sectors are evaluated.
PDF Abstract XML References Citation
How to cite this article
DOI: 10.3923/ijaef.2014.62.81
URL: https://scialert.net/abstract/?doi=ijaef.2014.62.81
INTRODUCTION
Historically, the level of exchange rate, its volatility and direct or indirect effects on the main economic indicators are the subjects have been researched widely for the Turkish economy. Because, exchange rate has an essential role within overall prices, production and foreign trade dynamics. Government, banking sector and firms operating in the real sector generally and historically prefer to borrow credit from abroad. The domestic interest rates are higher than the foreign interest rates, the total amount of foreign currency reserves are restricted so by borrowing foreign currency, firms obtain a short-term easy wealth. Foreign trade and its finance are also essential in this foreign debt dynamics. Firms obtain credit to finance their import from foreign institutions. Therefore when the economy grows, current deficit also increases. Central Bank of the Republic of Turkey (CBRT) observes the foreign exchange market and seeks the Real Exchange Rate (REER) to be between 120 and 130. As seen in Fig. 1, when the Turkish lira depreciates, REER diminishes and CBRT keenly observes the level.
The firms in the real sector may use different types of foreign currency in their foreign trade transactions. Therefore the relationship between different currencies is also important. The Pearson correlations for the US dollar (USD) and Euro (EUR), USD and Euro-US Parity (EUR/USD) and USD and Real Exchange Rate (REER) are 0.84, -0.17 and -0.26, respectively and they are significant1. The scatterplots and the regression lines in the Appendix-1 confirm these measures. Between June, 2011 and September, 2013 USD/TL parity increased from 1.60-2.02. The EUR/TL increased from 2.30-2.69. The average monthly depreciation of Turkish lira between these dates are 0.92 and 1.81%, respectively. The level of REER was below 120 except for the months from January, 2013 to May, 20132. Therefore as of September, 2013 the level of 108 had been started to threaten the firms owning high foreign debt levels.
The high volatility of Turkish lira among the other currencies is one of the historical problems for the Turkish markets. During the years 1960-2012 the depreciation of Turkish lira is at one of the highest rates compared to the other OECD countries, following Chile (51%), Israel (31%) and Brazil (208%) that are Non-OECD economies (Table 1). During the post-2002 period, where Turkey adopted flexible exchange rate regime, Turkish lira fluctuated and depreciated slightly, however, the trend was upwards. The level of exchange rate continued to be one of the controversial issues (Fig. 1).
The price of Turkish lira is keenly observed by the Small and Medium Sized Entrepreneurships (SMEs) in Turkey. Because firms owning high open currency position in their balance sheets, had been affected negatively from the depreciation of Turkish lira in the past (Kucuksarac et al., 2012). During the post-2002 period, by the transition to the floating exchange rate regime, the volatility and level of foreign exchange rate has come to order. This high volatility and increase in exchange rate concerned the SMEs since currency depreciation directly and simultaneously increases production costs and alters export costs.
The credit usage of the SMEs is another common research interest for the Turkish economy. After changing the SME definition3 in 2012, the ratio banking sector credits to SMEs, increased according to Turkish Banking Regulation and Supervision Agency.
Table 1: | Nominal exchange rates, national units per USD, average percentage change (%) |
Brazil, China, India and South Africa are Non-OECD member economies. The countries are collected according to availability criteria. Source: OECD, Currency exchange rates, National units per USD, monthly average |
Fig. 1(a-d): | Exchange rate (January, 2003-September, 2013), (a) REER, (b) EUR, (c) EUR/USD and (d) USD. REER: Real exchange rate (CPI Based), EUR: Euro-Turkish lira effective exchange rate, sales. USD: US Dollars-Turkish lira effective exchange rate, sales effective sales, Source: CBRT |
Turkish Banking Sector extended 120 billion US dollars loans and 35 billion US dollars non-cash loans to SMEs as of August 2013. Nearly 64% of the FX loans (50 billion US dollars, 61 thousand) extended to medium enterprises. These ratios appear in Table 2. It is interesting to see that banks extend FX loans much more to medium enterprises than the micro and small enterprises. However, the number of micro enterprises borrowed TRY loans from banks is nearly 76% of total enterprises (744 thousand) and nearly fifteen times of the medium ones. The total loans extended to these micro and medium size enterprises are similar. Export FX loans are usually short-termed but the term structure of the FX loans for investment purposes is longer.
The motivation of firms for FX borrowings increased by diminishing interest rates. Table 3 shows a comparision of Turkey with the other countries. Up to year 2010, the low interest rate and volatile exchange rate policy had been adopted. But after October, 2010, when the new monetary policy tools such as reserve option mechanism and interest rate corridor have been started to be implemented, the interest rate has started to be more volatile as the high exchange rate volatility (Fig. 2).
During the last ten years, essential ammount of short-term capital had entered to Turkish economy. Since the foreign currency liquidity was high because of low interest rate atmosphere, SMEs easily borrowed money from foreign financial institutions.
Table 2: | SME loans ratios distributed according to enterprise size as of August, 2013 |
Source: Turkish Banking regulation and supervision agency, Turkish banking sector interactive monthly bulletin, million dollars |
Table 3: | Level of short-term interest rates, average percentage change (%) |
Countries are collected according to availability criteria. Source: OECD, immediate interest rates, call money, interbank rate, percent per annum |
When the SMEs benefited from the easy lending, their foreign currency liabilities started to increase. The balance sheets of the firms improved by the capital inflows and borrowing from abroad became easier. The economic growth had been affected positively from the high capital inflows. On the other hand, CBRT started to be concerned on the volatile capital inflows. Because according to their opinion, short-term easy money was also increasing the fragility and volatility. So, CBRT started to take macro prudential policy measures for a more stabilized exchange rate, growth rate and current account.
Fig. 2: | Current account and portfolio investment in Turkey, 1990-2012 (Million US Dollars), Source: CBRT |
The liability of the firms in terms of foreign currency increased fastly and monetary policy authorities had started to be concerned. At this point, as of July, 2013 CBRT tried to diminish the exchange rate volatility by selling foreign currency to the market. The CBRT decided to sell US dollars to the market to diminish the exchange rate which was above 2.15 as of the end of December, 2013. The role and function of Reserve Option Mechanism (ROM) in terms of automatic stabilizer function had started to be researched among economists.
During the post-2002, the effects of the credit expansion on the unemployment and the current account had been discussed among the economists. Credit expansion let the firms invest more and they hired more people. Since firms in Turkey are highly dependent to import for production, such as energy and interval goods, credit expansion could also increased the current account. When the investment increased economy grew, employment increased and unemployment decreased. An increase in investment also increases the current account through import channel. Therefore, a trade off between the current account and the employment had been shown itself.
FINANCIAL DOLLARIZATION LEVEL IN THE ECONOMY
Financial dollarization has two aspects: Asset and liability dollarization. These two have different motives. During the inflationary period, SMEs may choose to save their Turkish lira by increasing their foreign assets which is called as partial dollarization (Serdengecti, 2005). Bahmani-Oskooee and Domac (2002), also considered FX deposits as a percentage of total deposits as an indicator of dollarization. The ratio of foreign exchange deposit accounts (commercial and other institutions deposits, residents) to total deposits in US dollars was nearly stationary and from September, 2011 to August, 2013, it fluctuated around 11.72% but increased from 9.77-12.77% (Fig. 3). Same ratio was 15.56 for natural persons in average and was nearly constant. Ratio of official institutions was less than one. Overall average foreign exchange deposit accounts of residents (natural persons, commercial and other institutions, official institutions) were 27.97% and increased from 28-29% within this period.
Fig. 3: | TRY and FX deposits to total deposits ratio |
Table 4: | Foreign exchange assets and liabilities of non-financial companies (Million USD) |
Source: CBRT. 1As of the first months, 2Average of the years. Methodological explanation can be found in CBRT (2013) |
Asset dollarization (demand side) that is the ratio of FX-Denominated Deposits to Total Deposits increased from January, 1992 to December, 1996 and stayed nearly 70% up to 2003 (Kesriyeli et al., 2011). But started to decrease from January, 2003 to June, 2006 (Basci, 2011). However, there was a slight increase from June to September, 2006. As can be seen from the Fig. 3, the ratio of overall foreign currency deposits to total deposits diminished from 35-30% between January, 2006 and October, 2013 (Source: CBRT). It is also clear that when TRY deposits increase, FX deposits diminish simultaneously.
Table 4 provides a summary for the foreign exchange assets and liabilities of non-financial companies in million dollars. During the period 2002-2013, the liabilities of the non-financial companies increased more than the assets. Although, the external loans increased slightly, 6%, during the years 2011, 2012 and 2013, the domestic loans increased nearly 51% between these three years. It seems that the deposit, securities, export recievables and direct investment abroad of the firms should be increased4 to close the gap of increasing net foreign exchange position that is nearly 173 billion dollars as of December, 2013, which was 124 billion dollars as of 2011.
REAL SECTOR LIABILITY DOLLARIZATION (SUPPLY SIDE) ANALYSIS FOR THE SMES
The data over the years 2009-2012 for the Turkish companies is gathered from the Central Bank of the Republic of Turkey (CBRT), Company Accounts Database provided by Real Sector Data Division. The CBRT compiles data for the sectoral balance sheets. As of 2011, there are 8925 firms had been involved in the database. These firms are decomposed into small, medium and large scaled according to the numbers of their workers, higher than fifty and five hundred or not. There are 3870 firms and 78 thousand workers in the small companies. There are 4257 firms and 711 thousand workers in medium companies. There are 798 big companies and they have nearly 1.2 million workers. The sectoral balance sheet data for the year 2012 was available as of November, 2013. The number of firms involved in the database increased to 9468 in 2012. The number of small scaled companies increased to 4546 and the number of workers to 82050. There are 4107 medium scaled companies and they have 683551 workers. The number of large scaled companies also increased to 815 and their number of workers to 1202841 as of 2012. It is apparent that when the scales of the companies increase, their employment generation capacity also increases.
The firms in manufacturing sector are considered in the study. Companies in manufacturing sector constitute a significant portion of total companies. There are 1270 and 2027 middle scaled companies, 1535 and 1117 small scaled companies and 616 and 454 large scaled companies in manufacturing sector as of 2011 and 2012, respectively.
The liability dollarization (DB) for the real sector is calculated as a ratio of Foreign Currency Denominated Cash Credits to Total Cash Credits. As it can be seen from the Fig. 4 and 5, the liability dollarization of the firms diminished continuously from 1998-2012 and the diminish is more apparent for the post-2002 period. The short-term debt is nearly half of the 1998 level with a ratio of 41.16. Firms tend to borrow more in terms of long-term compared to short-term by the time being but long term debt diminished from 81.25-78.94 between 2011-2012. Total debt also diminished from 69.50-65.10 between 2011-2012. In 2012 the ratios were below the 1998-2012 averages of 56.84 (short), 86.61 (long) and 65.19 (total). This shows that the firms tend to borrow long-term debt more.
We initially provide the average liability dollarization between 2009-2011 and than separately for the year 2012 (Table 5-16).
Fig. 4: | Foreign exchange short-term position ratios (million dollars), Source: CBRT, the year 2013 is the average of first five months |
Fig. 5: | Liability dollarization, total firms and sectors, 1998-2012, Source: CBRT sectoral balance sheets database |
Table 5: | First five sectors in terms of DB, total, 2009-2011 averages |
Table 6: | Highest sectors in terms of DB, total, as of year 2012 |
Table 7: | Last five sectors in terms of DB, total, 2009-2011 averages |
When the average liability dollarization between 2009-2011 analyzed, for the total sectors, short-term and long-term weighted debts, we may gather interesting results. The highest sectors in terms of liability dollarization are observed for the large firms and the four of the five sectors owning the least liability debt belong to small companies for the three types of liability structure. The large scale companies have higher long-term liabilities compared to the medium and small companies.
Table 8: | Highest sectors in terms of DB, total, as of year 2012 |
Table 9: | First five sectors in terms of DB, short-term, 2009-2011 averages |
Table 10: | Highest sectors in terms of DB, short-term, year 2012 |
Table 11: | Last five sectors in terms of DB, short-term, 2009-2011 averages |
Table 12: | Lowest sectors in terms of DB, short-term, year 2012 |
Table 13: | First five sectors in terms of DB, Long-term, 2009-2011 averages |
Table 14: | Highests sectors in terms of DB, long-term, year 2012 |
Table 15: | Last five sectors in terms of DB, long-term, 2009-2011 averages |
Table 16: | Lowest sectors in terms of DB, long-term, year 2012 |
The histogram, PDF, CDF and QP plots at the Appendix 3 also confirms this fact and present the location and scale shifts in the distributions.
The total liability dollarization is highest for large companies in manufacturing of machinery and equipment but lowest for small companies in manufacture of food products. Short-term dollar liability is highest for large companies in manufacture of wearing apparel but lowest for small scaled manufacture of basic pharmaceutical products and pharmaceutical preparations. Concerning the long-term debt, large companies in manufacture of machinery and equipment is highest but lowest in manufacture of food products of small scaled companies. In general we may say that the large companies have higher liability dollarization than the small scaled firms.
Specifically, we analyzed the debt dollarization ratios for the year 2012 of among the sectors (Appendix 2). Concerning the total and long-term debt ratios, manufacture of machinery and equipment and manufacture of food products have the highest and lowest liability dollarization as of 2012. But for the short-term dollarization ratios, manufacture of main metals has the highest but manufacture of basic pharmaceutical products and pharmaceutical preparations have the lowest in terms of debt dollarization as of 2012.
POSSIBLE EFFECTS OF THE EXCHANGE RATE ON THE LIABILITY DOLLARIZATION
During the year 2009, USD/TL exchange rate increased by 20% and reached up to 1.56 TL. Within the year 2010, it diminished and became 1.51 TL. However, in 2011 it increased by 11% and became 1.80 TL. It continued to increase in 2012 and 2013 and reached nearly 1.95 TL as of July, 2013 which was historically a record. As of end of December, 2013 the markets observed USD/TL exchange rate nearly 2.17 TL. The year 2014 started by sharp exchange rate fluctuations and reached nearly 2.35 as of January, 28, 2014. The CBRT increased the short-term interest rates and policy rates to prevent the increase in exchange rate5. This high increases in the exchange rate had been observed similary on March, 2009 and December, 2011. There are two aspects of increase. One is related by the possible sources of this increase during the last four years. The other is the consequences of this increase for the real sector. The firms in the real sector have debt in the short and long-run (Appendix 2 and 3). That is a choice whether to borrow or not. That is another question and may be answered by using a more micro data. If the domestic currency depreciates, the debt of firms would increase. If a certain level is exceeded, firms start to make a pressure on government to take some measures. Even if the export sectors may benefit more than others, since they are dependent to import, they are affected negatively from the exchange rate volatility. The government may take some measures for minimizing the possible negative effects of the exchange rate volatility. Between 1996-2014, Turkey also took several legislative measures that may also effect the volatility of exchange rate. The changes in the banking law, changes towards sustaining the value of Turkish lira, foreign trade legislative arrangements, exchange rate regime shifts, restricting credit growth rates and changing credit-collateral rates and the news effects related to the board of monetary policy may also effected the volatility of exchange rate6. Qualitative Easing policies of the Fed in November, 2008, November, 2010 and September, 2012 also increased the volatility in exchange rate. Recently CBRT started using Reserve Option Mechanism (ROM) policy to diminish the credit boom which may also effect the exchange rate. It differentiated the Reserve Option Coefficient (ROC) for short and long term accounts so aimed banks to change their choices from short-term to long-term interest rates. Besides it increased the interest rate volatility to diminish the capital inflows (Sahin, 2013).
Figure 6 provides a graph for the monthly data January, 2005 to August 2013 concerning the USD/TL exchange rate and Fig. 7 provides the monthly volatility graph for more longer-term. It can be seen that the exchange rate increased during this period. According to Leiderman et al. (2006), during the high dollarization periods, the transivity of exchange rate to prices is also high.
Fig. 6: | USD/TL effective exchange rate and sales, January, 2005-August, 2013, Source: CBRT |
Fig. 7: | USD/TL real exchange rate volatility, sales, extreme values excluded, January, 1970-July, 2013, Source: CBRT |
The sectors such as food, textile, leather, wood, paper, chemical, pharmacy, rubber, basis metal, motor vehicle, furniture have been affected differently from the increase in exchange rate. The high transivity of exchange rate to prices may affect the cost of firms negatively and their profitability may decrease (Fig. 6 and 7).
Since high volatile capital inflows increase the fragility of the economy, a question arises whether the hot money is useful for the firms in terms of appreciation of TL or not. The CBRT intervenes to the FX market by arranging the quantity auctions. Or as in January, 2014 may increase the interest rates following the increase in exchange rate because of capital outflows, tapering of fed and political instability. So, CBRT does not want the value of TL to fluctuate so much. International organisations also started to be concerned by the level of exchange rate. The G-20 Central Bank Governors and Fiscal authorities stressed in September, 22, 2011 in USA that the high volatility and high irregularity in FX market is a threat for the economy. Because being able to export is the long-term target of most SMEs. SMEs may create high employment and has a high share among the total number of firms (Demirbas, 2009). Turkish firms export mainly to the EU countries. However, during the post economic crisis, SMEs also started to diversify its exports to other countries. For instance, Turkey exports mainly raw materials to China but imports technological products.
The findings in this study are consistent with Alp (2013) who claimed that the liability dollarization is the highest for small companies before 2001 and highest during the post-2001 for the large companies. She claimed by using panel regression that an increase in real exchange rate, appreciation of Turkish lira increases the liability dollarization by diminishing the cost of borrowing in terms of foreign currency. She also found that the liability dollarization is highest for Basis Metal, Textile and Transportation sectors but lowest for Leather, Electric and Optic sectors. Aklan and Nargelecekenler (2009) use dynamic panel data method and analyze of the effects of liability dollarization on balance sheet in manufacturing sector between 1998-2007. According to Bahmani-Oskooee and Domac (2002) inflation targeting regime can be applied by floating regime to prevent inflation volatility. According to them, an increase in the volatility of exchange rate may diminish the increase in dollarization rate. They consider dollarization as an increasing holding of foreign currency. According to Ozmen and Yalcin (2007), between 2001-2005 the ratio of real sector debts to total assets ratio diminished in Turkey. This diminish was because of the decrease in total assets, although the liabilities were increasing. According to Berkmen and Cavallo (2010), countries having high levels of liability dollarization try to stabilize the exchange rate. Besides, fear of floating is an argument for those countries benefiting from the fixed exchange rate regime.
Metin-Ozcan and Us (2009) explored that the asset dollarization diminished for the post-2002 period but liability dollarization increased because of the shift from demand side driven to supply side driven motivation for dollarization. They found a pairwise causality between asset dollarization and liability dollarization for the periods 1996-2001 and 1996-2006 by monthly data. However, for the years 2001-2006, they found that the asset dollarization does not cause liability dollarization. Kesriyeli et al. (2011) used the GMM method and estimated that the real appreciation of exchange rate diminish the cost of borrowing in terms of foreign currency and increase the liability dollarization of firms.
CONCLUSION
Possible suggestions can be extracted from the liability dollarization ratios. There are good and bad news. Good news are that Turkish economy had started to decrease liability dollarization. Besides, increasing exports and income in terms of FX may improve the asset side of the balance sheet. Bad news are about the effect of high exchange rate volatility. Possible persistent depreciation of Turkish lira may detorate pricing decisions of the firms. Since the small open economies are highly dependent to developed countries in terms of importing technology, exchange rate behavior also effects the competition level. Export oriented companies are highly keen to borrow money by foreign currency. The debt structure of firms and usage areas are essential for the job creation and aggregate growth figures. Possible suggestions for the firms are as follows:
• | Companies may benefit from the instruments such as hedging to avoid from the exchange rate risk |
• | The firms owning Turkish lira revenues but borrows money in terms of US dollars may be effected negatively more from exchange rate movements. Since financial cycles are longer than the business cycles as mentioned by Caruana and Cohen (2014), there may be a time mismatch for the firm when central bank intervenes to the exchange rate |
• | Countries such as Turkey, Brazil, Indonesia, Argentina and South Africa are highly dependent to exchange rate movements because of current account and finance problems. So, their economies are highly sensitive to the capital movements. Firms and policy makers should consider this during their decision making process |
• | The correlation between reel GDP in terms of US dollars and export in terms of US dollars is negative 0.24. As mentioned by Esteves and Rua (2013), firms may choose increasing exports when the domestic demand decreases. Therefore arranging the timely pass through between domestic and foreign demand in terms of asset side of real sector balance sheets is important |
• | Central Banks may increase the interest rates to appreciate the domestic currency but an increase in interest rates may also affect the sectors such as durable consumer goods negatively. An increase in inflation triggered by the depreciation of TL may diminish the investments of the firms through expectation channel. So central bank should adopt an optimal exchange rate and interest rate policy to support the economic stability |
• | Some sectors provide their interval goods used in production from abroad but some provide from the internal sources. So, there is a high heterogeneity among the sectors considering their sensitivity to the exchange rate movements. The more open to foreign trade sectors may borrow or lend more foreign currency compared to the low openness ratio sectors |
Appendix 1: | Scatter plots of parities with regression lines |
Appendix 2: | Detailed debt dollarization ratios |
Appendix 3: | Distribution analysis for the year 2012 data, (a) Histogram analysis, (b) Denstity (PDF), (c) Cumulative Distribution Function (CDF) and (d) Quantile plot inverse of CDF |
ACKNOWLEDGMENT
This study was presented in the 2nd International Conference on Critical Issues in Business and Economics, May, 2-3, 2014, Gumushane, Turkey orally with a title Exchange Rate and Small and Medium Entrepreneurships Liability Dollarization: An Evidence for Turkey (Without publishing the full text in the proceeding book of the conference).
1Spearman correlation coefficients give similar results
2Note that the REER diminishes when the USD/TL and EUR/TL increases
3We consider here the similar definition of EC (2012), which defines the SMEs as; micro, small and middle scale enterprneurships where less than 250 people work, yearly turnover is 50 million dolar and the yearly total balance sheet is not higher than 43 million euros. In Turkey KOSGEB also adopted the same definition and defined it as the companies employing less than 250 people, yearly net sale revenue or any item of its balance sheet is less than 40 million Turkish lira
4Mehmet Simsek, Minister of Turkish Finance told the importance of asset dollarization in one of his speeches within April, 2014. He stressed that firms owning high liability dollarization also have high asset dollarization because of high FX income. Detailed balance sheet information is necessary to evaluate the sources of FX income of firms such as from exports and domestic sales
5Similarly during the years July, 2006 and October, 2011 following the increase in exchange rate, CBRT also increased the interest rates and TL started appreciating
6During the second quarter of 2014 economic agents observed that following a sharp increase in short-term interest rates in Turkey by CBRT, consumer credit interest rates also increased and total amount of credits diminished. CBRT thought that increasing interest rates would diminish the exchange rate. However this high interest rate policy could detorate the growth figure through credit channel. Therefore government were making pressure on CBRT to decrease the interest rates. The recent two elections were also effective on these interest rate level debates
REFERENCES
- Aklan, N.A. and M. Nargelecekenler, 2009. Liability dollarization and exchange rate balance sheet effect in Turkey: A sectoral level anaylsis. Proceedings of the Anadolu International Conference in Economics: Developments in Economic Theory, Modelling and Policy, June 17-19, 2009, Eskisehir, Turkey.
- Berkmen, S.P. and E. Cavallo, 2010. Exchange rate policy and liability dollarization: What do the data reveal about causality? Rev. Int. Econ., 18: 781-795.
CrossRef - Demirbas, D., 2009. The political economy of government programs on export success: A logit model for a Turkish survey. J. East-West Bus., 15: 316-339.
CrossRef - Kesriyeli, M., E. Ozmen and S. Yigit, 2011. Corporate sector liability dollarization and exchange rate balance sheet effect in Turkey. Applied Econ., 43: 4741-4747.
CrossRef