Selected Issue

Abstract

Selected Issue

Non-Financial Corporate Sector Debt in Turkey1

Non-financial corporate sector (NFC) debt2 has increased substantially in recent years, on the back of increased foreign currency (FX) leverage. This has translated into a large and widening net open FX position. Partially mitigating these developments, the maturity of FX debt has lengthened, thereby alleviating firms’ short term refinancing, and interest rate risks. Indeed, the corporate sector has been resilient through a period of marked exchange rate depreciation as well. However, a sizable portion of NFC FX debt is held by more vulnerable sectors with less favorable natural hedging and debt service capacity. These vulnerabilities have, on balance, become larger, and could become a source of brittleness when economic and financing conditions turn less supportive. Over the short term, concentration of NFC FX leverage in a few non-tradable industries may also pose risks for the domestic banking sector. The growing disconnection between higher NFC borrowing and private investment in recent years may also have implications for the country’s medium term growth potential.

A. Stylized Facts

1. NFC indebtedness rose significantly in most emerging economies (EMs) after the 2008 global crisis. Exceptionally accommodative global financing conditions since 2008, as well as the concomitant relaxation of domestic financial constraints in emerging market economies (EMs) have offered the NFC sector new venues to increase leverage both in FX and local currencies. As a result, NFC debt in EMs rose on average by 26 percentage points of GDP to US$18 trillion over 2004–14 (IMF GFSR 2015).

A03ufig01

Change in Corporate Debt: 2007-2014

(in percent of GDP)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

2. Turkey has seen a significant increase in NFC debt since 2008, among the highest in EMs. NFC debt in Turkey rose by 44 percentage points of GDP to about 69 percent of GDP over the last decade, well above the average change in EM NFC debt.

A03ufig02

NFC Debt

(Percent of G DP) 1/

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Sources: BRSA; and CBRT1/ Corporatebond issues are excluded.

3. The increase in NFC debt-to-GDP ratio was driven primarily by a rise in longer-term FX leverage. The NFC FX debt-to-GDP ratio rose from 14 percent in 2005 to 39 percent in 2015, accounting for 56 percent of the rise in total NFC debt. At the same time, the share of long term FX debt rose from 51 percent to 82 percent of total NFC FX debt, mitigating the short term rollover, exchange rate, and interest rate risks. Importantly, the CBRT data on NFC long term FX loans extended by domestic banks are presented on an original maturity basis, which may result in an overstatement of the long term FX debt and understatement of the short term FX debt compared to a presentation based on remaining maturity basis.

A03ufig03

Maturity Composition of NFC FX Loans 1/

(Share of Long term FX Loans; in percent of total NFC FX loans)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.Note:1/ Exclude import loans.
A03ufig04

NFC Net FX Position

(Billions of USD)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.

4. Higher FX leverage has translated into a large and widening net open FX position. As FX asset accumulation fell far short of FX leverage build-up, the net open FX position rose substantially from US$21.7 billion in 2005 to US$175.3 billion in September 2015, widening by 23.4 percentage points to 27.9 percent of GDP. More than three-fourths (18.3 percentage point of GDP) of the cumulative increase in net open FX position over 2005—September 2015 took place after the 2008 global crisis due to a rise in FX loans, mostly extended by domestic banks. This is, to some extent a reflection of the impact of regulatory changes in 2009 aimed at bolstering NFC FX lending by domestic banks through their on-shore branches rather than off-shore branches. However, it should also be noted that despite the large and widening net FX position, the NFC net short term FX position remains minor and stable.

A03ufig05

NFC Debt and FX Position

(Percent of GDP)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Sources: BRSA; and CBRT.

5. While a large portion of the NFC FX debt is held by big companies, a non-trivial portion of total FX debt is owed by SMEs that could face future constraints in accessing financing. SMEs accounted for 56 percent of total exports in 2014, while they owed 36 percent of total NFC loans, and 20 percent of NFC FX loans as of late 2015. While significant SME involvement in exports provides natural hedge buffers against FX indebtedness, a less favorable export outlook could mean lower natural hedge buffers, going forward. In addition, a possible increase in refinancing needs of big companies may further constrain SMEs access to bank lending.

A03ufig06

NFC Loans by Company Size

(Percent of respective total)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: BRSA.

6. Despite high FX indebtedness, the NFC sector has proved resilient thus far. While the NFC’s balance sheet has been stretched by a few currency depreciation episodes, no systemic company failures and credit defaults have been experienced so far. Nevertheless, the NFC’s aggregate balance sheet has responded to exchange rate shocks through a combination of stock and flow adjustments, including a deteriorated debt-to-equity mix, as well as lower profitability and a slowdown in fixed investment.

7. This resilience may be explained by multiple factors, some possibly more durable than others:

  • A supportive external financing environment may have helped NFCs to rollover due FX debt obligations smoothly.

  • Domestic banks have, in general, had high external debt rollover ratios, allowing for continued domestic financial intermediation of external bank funding into the NFC FX borrowing. In addition, the NFC’s own external debt rollover ratios have also remained high.

  • The lengthening of maturities of NFC FX loans by banks may have mitigated NFC’s short term re-financing risk and limited companies short-term cash outflows for financial expenses in the wake of currency weakness.

  • Resilient domestic growth may have helped company earnings, creating room for firms to pass on extra cost of currency weakness to their prices (in some sectors through FX-indexed pricing)

  • The relatively small short-term FX position meant that NFC had sufficient liquid FX assets to cover short-term FX debt obligations.

  • Banking sector forbearance may have allowed for restructuring and/or or reclassification of potentially problematic NFC FX loans.

  • Company owners’ unregistered personal financial holdings (especially in FX or gold, that are neither booked in the company accounts nor in the official NFC macro data) may have been used to finance NFC expenses during financial distress episodes.

NFC’s off-balance sheet financial hedges that are not recorded in the official data may have limited companies’ financial losses from currency weakness.

Selected NFC Financial Ratios

(Percent)

article image
Source: CBRT.

Total debt refers to total non-equity liabilities.

B. Data Analysis1

8. The key findings of the data analysis presented in the next section are summarized as follows:

  • The rise in NFC FX leverage has no statistically significant relationship with private investment growth.

  • Higher NFC FX leverage may put pressure on firm profitability which may in turn constrain firms’ tangible asset investments.

  • A large contribution to higher NFC FX indebtedness came from the non-manufacturing and lower-naturally-hedged sectors.

  • Sectors with weaker debt service capacity hold a sizable portion of total NFC FX debt.

  • Brittleness associated with NFC debt service capacity may increase with higher global risk aversion, slower domestic growth, and protracted exchange rate volatility.

9. Both domestic and external factors are associated with the rise in NFC FX leverage. A time series analysis is employed to explore the association between the Turkish NFC FX annual loan growth and the respective annual changes in the key domestic and global macroeconomic variables over the period of 2007:Q4–2015:Q3. On the external side, the analysis unsurprisingly suggests (see tables below) a statistically significant positive association between the annual changes in the NFC FX loans and global capital flows to non-bank entities in EMs. Accordingly, an increase of US$1 billion in annual global capital flows to EMs is found to be associated with an increase of US$85 million in annual in NFC FX loans in Turkey. On the domestic side, (i) the real GDP growth; (ii) the interest rate spread between commercial TL and US$ lending rates (adjusted for expected rate of change in the exchange rate); and (iii) the short-term trend of the real effective exchange rate, are all found to have a statistically significant positive association with the annual growth in NFC FX loans. The data suggest that a one percentage increase in annual real GDP growth rate is associated with an increase of US$1.71 billion in annual NFC FX loans while one percentage point increase in annual change in commercial lending rate spread is associated with an increase of US$197 million in annual NFC FX loans. On the other hand, there is no statistically significant relationship between the annual growth rates of private investment growth and NFC FX loans.

A03ufig07

Global Flows to EM Non-Banks and Turkish NFC FX Loans

(US$ billion)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Sources: BIS; and BIS.
A03ufig08

NFC FX Loans and Lending Rate Spread

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.
A03ufig09

NFC FX Loans and REER (PPI-based)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.

10. The corporate tax code has an inherent bias toward debt over equity finance. Firms’ interest expenses are deductible from the CIT base, while dividend payouts are not. In addition, firms are also liable for a withholding tax of 15 percent on their dividend payouts, implying “double-taxation” on shareholders’ equity. Recently, the tax code’s debt-bias has been somewhat mitigated after the government enacted (in July 2015) a new CIT allowance in the form of a new “notional interest deduction” for cash capital injections to non-financial firms (excluding SOEs).

11. Regulatory restrictions on domestic FX borrowing by firms with no FX income were eased in 2009. Before the 2009 regulatory amendments, only firms with FX income had been allowed to borrow FX-denominated loans from on-shore bank branches at maturities no longer than 18-months. These restrictions implicitly created a venue for firms with no FX income to borrow in FX-denominated terms through domestic banks’ off-shore branches that were free of regulatory restrictions on FX borrowing. In June 2009, authorities removed these restrictions, and instead allowed firms with no FX income to borrow FX-denominated loans from on-shore bank branches, provided that size and original maturity of FX-denominated loans shall be set at no shorter than US$5 million and 1 year, respectively. No such restrictions had applied for those firms with no FX income, should they have sought to borrow FX-indexed loans from on-shore branches Furthermore, the new regulation also offered firms with no FX income to be exempted from the new loan size and maturity limitations so long as they collateralized FX-denominated loans with FX deposits and securities parked at on-shore branches.

Model A: NFC FX Loans3

article image
t statistics in parentheses* p<0.1, ** p<0.05, *** p<0.001

12. These developments resulted in a substantial shift in residency composition of NFC FX assets and liabilities. In the aftermath of the regulatory amendment in June 2009, the composition of NFC FX loans and deposits predictably witnessed a structural shift in terms of residency constituency from domestic banks’ off-shore branches towards on-shore branches. The share of NFC FX loans extended by domestic banks on-shore branches rose from 33 percent of total NFC cash FX loans in 2009:Q2 to 68 percent in 2015:Q3, while the share of NFC FX deposits parked at the domestic banks on-shore branches rose from 44 percent of total NFC FX deposits to 75 percent over the same period. Considered together with the large and widening open NFC FX position, the direction and size of the post-2009 shifts in the residency composition of NFC FX deposits and loans suggest that NFC FX borrowing by firms with less robust FX income buffers may be sizable.

A03ufig10

Structural shift in composition of NFC Deposits

(percent of respective total)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.
A03ufig11

Structural shift in composition of NFC FX Loans

(percent of respective total)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.

13. Increased FX indebtedness appears to have been driven by non-manufacturing and lower naturally hedged sectors. In an analysis of NFC sectors by the asset-weighted historic averages over 2008–14 of (i) exporting capability (gauged by exports over total net sales) and (ii) FX leverage (gauged by FX loans over non-equity liabilities), individual sectors are distributed in the following four quadrants.12

A03ufig12

Sectoral Breakdown by Quadrants (2008)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.
A03ufig13

Sectoral Breakdown by Quadrants (2014)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

- Higher Risk Zone (Q1) which includes sectors with higher FX leverage and lower export ratios.

- Leverage Sensitive or Natural Hedge Zone (Q2) which includes sectors with higher export and FX leverage ratios.

- FX Sensitive Zone (Q3) which includes sectors with both lower export and FX leverage ratios.

- Lower Risk Zone (Q4) which includes sectors with higher export and lower FX leverage ratios.

  • Available corporate balance sheet data suggest that the overall FX Leverage Ratio (defined as FX Loans-to-Non-equity Liabilities) rose from 14.9 percent in 2008 to 26.2 percent in 2014, while the Export Ratio (defined as Exports-to-Net Sales) declined from 24.1 percent to 19.3 percent. On the other hand, the short term FX loans-to-exports ratio rose by 3.3 percentage points to 15.8 percent over the same period.

  • An analysis of sectoral sources of the changes in these ratios reveals that the sharp rise in the overall FX leverage ratio was driven significantly by those sectors with lower export ratios, hence weaker natural hedges (i.e., sectors located in Q1 and Q3).4 On the other hand, the overall export ratio was pulled down mainly by those sectors with relatively strong natural hedge buffers (i.e., those sectors located in Q2 and Q4). The rise in the short term FX loans-to-exports ratio was contributed by almost all sectors with the exception of those sectors located in the higher risk zone of Q1.

  • An alternative decomposition based on a classification of manufacturing and non-manufacturing sectors shows that the non-manufacturing sector accounted for a significant share of the deterioration in the FX Leverage and Export Ratios.

14. However, highly indebted sectors with weaker natural hedging capabilities have improved their maturity profiles. Based on the classification of sectors by quadrants along the lines described above, those sectors with lower export ratios (i.e., the ones in the quadrants Q1 and Q3, accounting for 60 percent of total assets), held around 55 percent of the total FX loans in 2013 and 2014. Whereas, their short term FX loans declined from almost 40 percent in 2013 to 30 percent in 2014 of the overall short term FX loans on the NFC balance sheet.

A03ufig14

Drivers of the Changes in Key Ratios

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.
A03ufig15

Distribution of Key NFC Aggregates by Quadrants

(Percent of the Respective Corporate Sector Aggregate)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.

15. Those sectors with weaker debt service capacity also hold a sizable portion of total FX debt. A similar analysis examined NFC FX debt at risk—i.e., the FX debt held by those sectors with a relatively weak debt service capacity (Interest Coverage Ratio (ICR)5 at or below 1.5). This analysis shows that the share of total FX loans held by sectors with weaker debt service capacity declined from some 29 percent of overall FX loans in 2013 to 20 percent in 2014. The recent improvement is encouraging, but the share is still relatively high.

16. A strong improvement can be observed in 2014 with regard to short term “FX debt at risk”. The portion of short term FX debt held by sectors with weaker debt service capacity declined from 35 percent of overall short term FX debt in 2013 to mere 4.4 percent in 2014 at the assumed ICR threshold of 1.5. If the ICR threshold for weak debt service capacity is set at 1.75, the short term FX debt at risk stands at a higher 21 percent of overall short term FX debt in 2014. This adds an element of caution, as there appear to be a significant number of firms on the cusp of debt servicing difficulties.

A03ufig16

Total FX Loans: Debt at Risk

(Percent of total FX loans)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

A03ufig17

Short Term FX Loans: Debt at Risk

(Percent of short term FX loans)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

17. Nevertheless, brittleness associated with debt service capacity remains a source of concern, if further shocks were to materialize. A simple empirical analysis was employed (over 1996–2014) to examine the impact of global risk appetite, real domestic GDP growth, and nominal exchange rate variation on the aggregate ICR. The findings suggest that ICR has a statistically significant positive association with domestic real GDP growth. It has negative associations with VIX and nominal exchange rate volatility, with the latter having a more pronounced impact on ICR than any other explanatory variable. A one percentage point increase in the coefficient of variation in exchange rate is associated with a decline of almost 11 percentage points in the ICR while one percentage point increase in real GDP growth is associated with a rise of almost 9 percentage point increase in the ICR. The findings suggest that in an unfavorable combination of shocks to these variables that may arise from heightened global risk aversion, debt service capacity may weaken notably.

Model B: Interest Coverage Ratio (ICR)6

article image
t statistics in parentheses* p<0.1, ** p<0.05, *** p<0.001

18. Higher NFC FX leverage may constrain firm-level investment by pressuring profitability. Based on sectoral balance sheet data, an empirical analysis was conducted over 1997–2014 to shed light on linkages between NFC profitability, tangible investments, and total bank borrowing.7 The analysis suggests a statistically significant negative association between net profit margin and borrowing. For instance, a one percentage point increase in the leverage ratio is associated with a 0.40 percentage point decline in the net profit margin. This is likely to be driven by a direct effect (higher financial expenses), as well as an indirect effect (resorting to more leverage as retained earnings fall). Empirical findings indeed also suggest a statistically negative association between leverage and tangible asset investment over the studied period, which speaks to the profitability-leverage linkages discussed above. Furthermore, in years of high leverage, real exchange rate appreciations appear to reduce the leverage ratio, possibly due to valuation effects as the real appreciation may lower local currency value of FX leverage and in turn the leverage ratio which is defined as total loans-to-total liabilities.

Model C: NFC Total Leverage8

article image
t statistics in parentheses* p<0.1, ** p<0.05, *** p<0.001

C. Implications

Short term implications

19. The banking system’s balance sheet exposure to NFC has risen substantially. The share of corporate debt (including TL and FX loans) extended by domestic banks’ on-shore branches (resident banks) rose from 59 percent in 2008 to 81 percent of total corporate debt in September 2015. This brings the outstanding corporate debt stock held by the resident banks to 56 percent of GDP, almost triple its 2008 share. NFC total and FX deposits accounted for 32 percent and 15 percent of the banking sector total non-financial deposits respectively, in September 2015.

20. NPL ratios for corporate loans have so far remained resilient. Despite slower growth and several episodes of currency depreciation since 2008, NPL ratios have remained low.9 At 0.1 percent of total FX loans, the NPL ratios for NFC FX loans remained very small compared to the NPL ratio for TL loans that hovered around 5 percent, which in part relates to banks sales of NPLs to the private asset management companies and regulatory forbearance arrangements, including reclassification of FX NPLs in TL terms. The existing banking regulations allow the non-performing FX loans to be re-classified as a non-performing TL loan at the discretion of borrower firms.

A03ufig18

NFC Financial Debt by Creditors

(Percent of GDP)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Sources: BRSA; and CBRT.
A03ufig19

NPL Ratios for NFC Loans 1/

(Percent of gross loans)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: BRSANote:1/ Shaded areas reprsents those periods of exchange rate depreciation of at least 20 percent from trough to peak.
A03ufig20

Sectoral Loans and NPLs - September 2015

(Percent of respective NFC totals)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: BRSA.

21. However, heterogeneity in sectoral NPL patterns is notable, with concentration and credit risks for some sectors. Construction and wholesale trade, which have higher than average NPL ratios also account for much larger shares within total sectoral loans and NPLs compared with other sectors. In terms of concentration of FX loans, energy, construction, and wholesale and retail trade sectors (which are not highly “tradable”) hold significantly more concentrated portion of FX loans, altogether accounting for one third of total NFC FX loans.

22. The large NFC open FX position creates indirect exchange rate risk for the banking sector. Recalling two central observations from the preceding section—that FX denominated debt appears to have increasingly permeated sectors that are less naturally hedged, and may also have relatively weaker debt service capacity—an unfavorable protracted exchange rate shock, particularly when combined with a less favorable interest rate environment, could trigger financial stress and deleveraging pressures for NFCs. This translates to both heightened credit risk, and also possible pressures on wholesale funding for the banking sector. Pressures from both channels may subsequently generate powerful adverse feedback loops in case of tightening of bank lending to the NFC.

A03ufig21

Sectoral FX Loans

(Percent of Total FX Loans, 2014)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT Sectoral Accounts.
A03ufig22

Sectoral Loans

(Percentoftotal assets)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.

23. Recent trends point to the possibility of increased risk aversion among banks and firms. Over 2015, banks began to tighten lending standards in line with heightened risk perceptions on industry or firm specific outlook. In addition to tighter credit standards on the supply side, firms’ demand for FX loans also started to decline, implying possible risk aversion on both the supply and demand side.10

A03ufig23

Banking Sector Risk Perceptions vs FX Loan Growth 1/

(Y-o-y, ER-adjusted)

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source:s BRSA, CBRT, and Bank Loans Tendency Survey.Note:1/ Banking sector risk perceptions are derived from the CBRT Bank Loans Tendency Survey. Positive values represent tighter bank lending standards.

Medium term implications

24. Despite rising NFC leverage, private investment growth has remained muted. Against the backdrop of subdued private investment since late 2011, NFC total loans continued to grow in real and exchange rate-adjusted terms by 11 percent yoy in 2015:Q3, suggesting an apparent disconnection between NFC loan growth and private investment, mirroring empirical findings presented in the preceding section. Continued NFC total loan growth despite the prolonged weakness in private investment demand may in part relate to firms’ lower operational profitability and increased working capital needs that may arise in a slower growth environment.

25. The same disconnection between higher leverage and subdued investment may also relate to refinancing needs linked to increased debt service expenses in the wake of prolonged exchange rate weakness. Indeed, the CBRT’s Bank Loans Tendency Survey suggests that the NFC sector’s financing need for fixed investments has declined, while firms have tended to borrow more for working capital and debt restructuring purposes, which may be a sign of increased NFC balance sheet strains.

A03ufig24

Corporate Credit and Private Fixed Investment Growth

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Sources: BRAS; CBRT; and TURKSTAT.

26. A persistent disconnection between NFC borrowing and private investment may have implications for medium term growth potential. Subdued private investment against the backdrop of increased NFC debt suggests that higher leverage may not have been adequately channeled into productive investment during a relatively prolonged period of favorable external financing conditions. Stretched balance sheets through the accumulation of leverage without a meaningful rise in productive capital accumulation may have a dampening effect on potential growth—a matter of significance for policy, given the high priority placed on convergence to high income country status.

A03ufig25

Bank Loans Tendency Survey Financing Needs of NFC Sector 1/

Citation: IMF Staff Country Reports 2016, 105; 10.5089/9781484338209.002.A003

Source: CBRT.Note: 1/Derived from the Bank Loans Tendency Survey. Positive values imply increase in demand and negative values imply decrease in demand.

D. Summary and Policy Recommendations

27. NFC debt in Turkey has increased substantially, with increasing risks specifically in certain sectors. The increase in NFC debt has been proportionally higher than in many other EMs, and driven primarily by the rise in NFC FX leverage. With NFC FX asset accumulation that fell far short of the rise in FX leverage, higher NFC FX leverage meant a large and widening net open FX position. FX debt may have become more concentrated in those sectors which are cyclically more sensitive, with lower export capabilities, and with weaker debt service capacity. Furthermore, potentially financially-constrained SMEs hold a notable portion of the NFC FX debt. The NFC sector has remained resilient so far, but exchange rate weakness, and higher borrowing costs—particularly if combined with slower growth—could unmask an underlying brittleness in the economy. Increased NFC indebtedness may also constrain corporate profitability and in turn constrain firms’ tangible asset investments. In the short term, higher NFC FX leverage and ensuing large open FX position poses an indirect exchange rate risk for the banking sector, via credit risk. In this sense, the resilience of the banking sector has mirrored that of the NFC sector. From a macroeconomic perspective, if NFC indebtedness fails to translate into productive private investment, medium term growth potential may also be affected.

28. The following policy options may be worth consideration:

  • Fill data Gaps. For a timely and accurate assessment of financial soundness of the NFC sector, authorities may consider identifying and bridging data gaps with more granular data particularly on NFC cross-border assets and exposures, on-balance sheet currency and maturity mismatches, off-balance sheet financial hedging activity, and corporate bond issuances.

  • Assess NFC financial soundness. The Authorities may consider conducting periodic financial soundness assessments on the NFC sector.

  • Re-assess regulatory arrangements. Regulatory arrangements that allow FX lending to firms with no FX income may be re-assessed. Bring the prudential regulation for FX-indexed loans in line with that for FX loans.

  • Strengthen macroprudential measures. Macroprudential measures that currently apply to commercial loans may be tightened. Further macroprudential measures may also be considered in the form of risk-weighted capital requirements, incremental provisioning requirements, and concentration limits with a view to contain NFC FX leverage and more efficient bank lending to financing of firms productive fixed investments.

  • Revisit the tax bias for debt financing. The Authorities may consider measures to further reduce tax code’s debt bias over equity finance by the NFC sector.

Appendix I. Econometric Specifications

Model A: NFC FX Loans

NFC FX Loanst = β0 + β1REERt + β2Real GDPt + β3 Lending Rate Spreadt–1 + ut

NFC FX Loanst = β0 + β1REERt + β2 Lending Rate Spreadt–1 + β3 Real Private Investmentt + β4 Capital Flowst + ut

article image
Data Sources: BIS, TURKSTAT, CBRT Sectoral Accounts Database, quarterly data over 2007:Q4 – 2015:Q3

Model B: Interest Coverage Ratio (ICR)

ICRt = β0 + β1 Real GDPt + β2 Exchange Rate Variationt + β3VIXt + β4REERt + ut

article image
Data Sources: TURKSTAT, CBRT Sectoral Accounts Database, annual data over 1996–2014.

Model C: NFC Total Leverage

ΔLeveraget = β0 + β1 ΔInvestmentt + β2 ΔExportt + β3 ΔReer x Leverage + β4 ΔGDP Growtht + μt

article image
Data Sources: TURKSTAT, CBRT Sectoral Accounts Database, annual data over 1997–2014.

References

  • Alp B., Yalcin C., 2015, “Liability Dollarization and Growth Performance of Non-Financial Firms in Turkey,” Working Papers 1501, Research and Monetary Policy Department, Central Bank of the Republic of Turkey.

    • Search Google Scholar
    • Export Citation
  • Aydin H., Kaplan C., Kesriyeli M., Ozmen E., Yalcin C., Yigit S., 2006, “Corporate Sector Financial Structure in Turkey: A Descriptive Analysis,” Working Papers 0607, Research and Monetary Policy Department, Central Bank of the Republic of Turkey.

    • Search Google Scholar
    • Export Citation
  • Hulagu T., Yalcin C., 2014, “Turkiye’de Firmalarin Yabanci Para Borclulugu ve Kur Riskine Iliskin Mikro Degerlendirmeler,” CBT Research Notes in Economics 1413, Research and Monetary Policy Department, Central Bank of the Republic of Turkey.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, Global Financial Stability Report, 2015, Chapter 3, Annex 3.1. Emerging Market Corporate Leverage: Data and Empirics, “Corporate Leverage in Emerging Markets—A Concern?,” International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Kesriyeli M., Ozmen E., Yiğit S., 2011, “Corporate Sector Liability Dollarization and Exchange Rate Balance Sheet Effect in Turkey,” Applied Economics, Taylor & Francis Journals, vol. 43(30), pages 47414747.

    • Search Google Scholar
    • Export Citation
  • Ozmen E., Sahinoz S., Yalcin C., 2012, “Profitability, Saving and Investment of Non-Financial Firms in Turkey,” Working Papers 1214, Research and Monetary Policy Department, Central Bank of the Republic of Turkey.

    • Search Google Scholar
    • Export Citation
  • Yildirim N., Tastan H., 2012, “Capital Flows and Economic Growth across Spectral requencies: Evidence from Turkey,” Working Papers 2009/2, Turkish Economic Association.

    • Search Google Scholar
    • Export Citation
  • World Bank Synthesis Report, 2011, Turkey Country Economic Memorandum, “Sustaining High Growth: The Role of Domestic Savings,” World Bank.

    • Search Google Scholar
    • Export Citation
1

The bubble sizes in the quadrant analysis represent respective asset size of each sector.

2

Sectoral analyses on NFC indebtedness are based largely on the CBT’s Sectoral Accounts Data on NFC which draw on unaudited and aggregated sectoral financial statements submitted to the CBRT by a sample firms. Sample varies on a 3-year rolling basis and does not necessarily represent the entire NFC sector. Therefore, interpretation of results calls for a degree of caution.

3

These regressions are illustrative because of limited number of observations and problems of endogeneity. Please see Annex for model specification and variable definitions with descriptive statistics.

4

Drivers of changes are calculated along the following formulation:

Z=XY, where Z is a ratio, while X and Y are levels. In terms of rate of growth:

(1+z)=(1+x)(1+y),wherex=ΔXX,y=ΔYYandz=ΔZz

In terms of percentage point change in Z:

ΔZ=Z[ΣΔXiXΣΔYiY(ΔZZΔYY)], where the third term in the bracket is labeled as residual in the charts.

5

ICR is defined as the ratio of Earnings Before Interest and Tax (EBIT) to Financial Expenses. Sectors with an ICR at or below 1.5 are assumed to have a weaker debt service capacity as their financial expenses account for at least two thirds of the their EBIT. On the other hand, FX debt at risk is defined as the share of FX debt held by those sectors whose ICR stands at or below 1.5.

6

Please see Annex for model specification and variable definitions with descriptive statistics.

7

FX loans comprise 62 percent of total NFC sample loans as of 2014.

8

Please see Annex for model specification and variable definitions with descriptive statistics.

9

NPL ratios exclude NPL sales to private asset management companies but include the effect from reclassification of NPLs in TL terms.

10

In loan data by company size, FX indexed loans are included in TL Loans.

Turkey: Selected Issue
Author: International Monetary Fund. European Dept.