Information about Middle East Oriente Medio
Journal Issue
Share
Article

Kyrgyz Republic: Selected Issues

Author(s):
International Monetary Fund. Middle East and Central Asia Dept.
Published Date:
February 2016
Share
  • ShareShare
Information about Middle East Oriente Medio
Show Summary Details

Financial Structure, Financial Inclusion, and Credit Trends in the Kyrgyz Republic1

While the Kyrgyz financial sector remains relatively small and underdeveloped, the recent credit boom and deepening may improve financial inclusion indicators. However, access to credit varies widely across regions. High and increasing dollarization in light of exchange rate volatility and concentration are key sources of vulnerabilities Prudent monetary and de-dollarization policies, with a set of macroprudential tools, could help mitigate these risks.

A. Financial structure and inclusion

1. The Kyrgyz financial sector remains relatively small and is dominated by banks. The overall level of credit to GDP is low relative to the Caucasus and Central Asia region and Central and Eastern Europe, particularly when compared with the lower middle income country (LMIC) average. The Kyrgyz financial system is a bank-based system, which accounts for 87 percent of financial system assets and provides about 80 percent of credit to the private sector. Nevertheless, Microfinancial Institutions (MFIs)2 play an important role (Table 1). Up until 2013, MFIs provided about one-quarter of the total credit to the private sector in the Kyrgyz Republic. However, since then MFIs have started to obtain banking licenses (Bai Tuschum in 2013, Finca in early 2015, and Kompanion underway), mostly to overcome hurdles in doing MFI business (including lending in foreign currency, access to the National Bank of the Kyrgyz Republic (NBKR) facilities, and wider range of sources of financing). By contrast, other financial institutions play a marginal role, and capital markets are very shallow, including government securities market.3

Table 1.Financial System Structure
20072008200920102011201220132014Jun-15
Number of institutions
Banks222122222223242423
Locally incorporated212021212122232322
Private domestic banks98910912111111
Foreign banks101010897998
State-owned banks222333222
Branches of Foreign banks111111111
Non-bank credit institutions and exchange offices9991089970906931810764660591
Credit-unions272248238217197183153135127
Micro-finance organizations233291359397454320277215176
FSCDCU111211111
Exchange Offices312353372290279306333309287
Pawn shops (business of providing securitized loans)181196
Other financial institutions
Securities companies1038185787753737373
Pension Funds112212111
Insurance companies161819191516171717
Assets in percent of GDP
Banks28.029.333.826.323.328.531.334.637.4
Locally incorporated27.628.833.325.822.928.030.634.037.0
Private banks5.75.66.47.75.88.625.428.030.8
State-owned banks2.62.92.77.06.36.35.16.16.2
Foreign banks19.620.824.611.510.813.115.317.519.3
Branches of Foreign banks0.40.50.50.50.40.50.70.60.4
Non-bank credit institutions and exchange offices
Credit-unions0.50.50.50.60.50.50.50.30.3
Micro-finance organizations2.95.15.46.66.66.26.16.94.3
FSCDCU0.30.20.20.20.20.10.10.10.1
Exchange Offices
Other financial institutions
Securities companies10.17.17.96.35.04.9nanana
Pension Funds0.010.010.010.010.010.010.010.010.01
Insurance companies0.30.30.40.40.40.50.50.60.6
Other
Insurance companies, gross premium in percent of GDP0.10.10.20.20.20.30.30.30.2
Market capitalization, percent of GDP29.923.725.126.321.421.018.821.020.8
Trading value at stock exchange, percent of GDP4.054.922.570.650.560.430.460.411.10
Turnover in payment system, times GDP2.12.32.42.02.02.22.43.6 na
Broad money, percent of GDP30.325.828.431.427.832.434.031.330.1
Currency in circulation, percent of broad money64.163.662.662.662.759.155.445.844.5
Credit to private sector, percent of GDP18.117.516.916.517.218.921.626.625.4
of which:
Banks14.713.612.512.511.713.816.120.721.8
Non-bank credit institutions3.43.94.445.55.15.55.93.6
Source: Kyrgyz authorities.
Source: Kyrgyz authorities.

Credit to Private Sector

Sources: IMF, International Financial Statistics, World Development

2. The supply of financial services is limited, while the economy remains heavily cash based. Banks offer a few financial instruments (loans/credit cards/ leasing), while supply of hedging instruments is very rare.4 Savings mobilization is still low with deposits accounting for less than 20 percent of GDP. The majority of deposits are demand deposits (about 55 percent) and only 12 percent of total deposits are with maturity of over a year. There are 253 depositors5 per 1000 adults in the Kyrgyz Republic, as opposed to averages of 115 and 364 for LICs and LMICs, respectively. The average weighted maturity of bank loans is 28.9 months (23.8 and 33.6 months for loans in national and foreign currency, respectively), while for MFIs it is 17 months, which does not provide sufficient support for investment. While the share of cash in the broad money has declined over the past two years, it still remains relatively high at about 40 percent. The five largest banks hold more than half of all deposits and loans, while the ten smallest banks hold about 10 and 12 percent of deposits and loans, respectively.

Banking and Monetary Indicators, 2013–15

Sources: NBKR data and IMF staff.

Concentration in the Banking Sector

(Share of total, June 2015)

Sources: The NBKR data and IMF staff calculations.

Bank Deposits and Credits by Regions

3. Financial inclusion indicators started to improve in 2012, but access to finance still varies across regions. Credit expansion over the past three years, together with efforts to pay pensions and wages via banking institutions, resulted in an increase in the number of accounts. This included both deposit and credit accounts at banks and MFIs. Similar to the regional income disparity, credit is mostly concentrated in the larger cities (Bishkek, Oss, and Jalal-Abad), while the vast majority of depositors—87 percent—are from Bishkek. In addition, according to the enterprise survey, only a small number of firms—30 percent—use credit lines, and mostly for working capital. Only 8 percent of firms use bank loans for investment.

Financial Inclusion Indicators
Key Indicators200920102011201220132014
Accounts
Account at a formal financial institution (% age 15+) /111.312.417.423.641.555
Credit
Loan from a financial institution in the past year (% age 15+) /215.7
Loan from a financial institution in the past year, female (% age 15+) 3/3.3
The number of women received loans at nonbank finance and credit
institutions (person)210,215198,273
Branch Penetration
Automated teller machines (ATMs) (per 1,000 sq km)2.43.14.14.9
Automated teller machines (ATMs) (per 100,000 adults)12.11114.417.3
Commercial bank branches (per 1,000 sq km)1.21.11.31.41.41.5*
Commercial bank branches (per 100,000 adults)6.55.76.87.37.47.5
Source: The NBKR data.

Number of adults with age above 15 with an account at a bank includes an amount of deposit accounts of individuals at a bank plus an amount of credits of individual at a bank plus an amount of credits of individual at a non-bank financial institutions

Number of individuals with age above 15 who received credit at a bank

Number of women with age above 15 with received credit at a bank

Source: The NBKR data.

Number of adults with age above 15 with an account at a bank includes an amount of deposit accounts of individuals at a bank plus an amount of credits of individual at a bank plus an amount of credits of individual at a non-bank financial institutions

Number of individuals with age above 15 who received credit at a bank

Number of women with age above 15 with received credit at a bank

4. MFIs provide credit to a wide range of borrowers across the country. MFIs provided credit to 402 out of 575,0006 registered borrowers at the credit bureau. Women are more likely to borrow from an MFI than from a bank. About ⅔ of all borrowers are between 30 and 60 years of age.

Borrowers by Type of Institution and Gender(Percent of total by institution/gender)
TypeFemaleMaleTotal
MFI825872
Bank184228
Source: Ishenim, credit bureauNote: registry includes data for 23 banks, and 204 MFI.
Source: Ishenim, credit bureauNote: registry includes data for 23 banks, and 204 MFI.
Borrowers by Age and Gender(Percent of total borrowers/ by gender)
AgeFemaleMaleTotal
<25587
25-3091411
31-40222624
41-50282426
51-60272024
60>979
total5842100
Source: Ishenim, credit bureauNote: registry includes data for 23 banks, and 204 MFI.
Source: Ishenim, credit bureauNote: registry includes data for 23 banks, and 204 MFI.

B. Credit and NPLs trends

5. Credit growth started to moderate in 2015, but it remains robust. Credit growth over the past five years was strong—increasing credit to GDP from 12.5 percent in 2010 to 20.8 percent in 2014. That was accompanied by strong deposit growth, which was financed mostly from internal sources and helped to keep leverage low. Credit growth as of end-September 2015 year-on-year was 27.0 percent. However, the major contributor to the credit growth in 2015 comes from the transition of an MFI to a bank (credit growth without this MFI is just 17.2 percent). Credit growth in real terms is 21 percent, and the constant exchange rate7 about 13.6 percent over the same period. Loans to agriculture, supported by a government subsidy scheme,8 along with loans to the trade sector, were the key drivers of this growth. The share of mortgages and construction loans in total loans declined relative to 2009 (Figure 1). The majority of loans to the trade and construction sectors, and for mortgages, are in foreign currency. Mortgages and loans to the trading and construction sectors account for 60 percent of all foreign exchange loans, while agriculture and trade account for 65 percent of all som loans.

Figure 1.Credit Trends and Structure by Sector

Credit Growth

(In percent)

Sources: Authorities data and IMF staff calculations.

Structure of Loans by Currency and Sector

6. Credit from MFIs started to shrink in the second quarter of 2015. In addition to one MFI becoming a bank, the main reason behind this was a reduction in the average size of MFI loans—US$100 on average9—due to slowdown in the Kyrgyz economy. NPLs are still low, as MFIs mostly lend (about 80 percent) to groups, thus ensuring multiple warrants for their loans. MFIs mostly lend to agriculture (40 percent), consumers (19 percent), and the trade sector (17 percent).

Micro-Finance Organizations Financial Indicators
2009201020112012201320142015 q1
MFO credit as a share of GDP4.45.05.55.05.45.83.5
MFO Non-performing loans to total loans3.94.63.44.32.83.4
Source: The NBRK data.
Source: The NBRK data.

7. NPLs have been gradually increasing. They remain low, but could be higher after allowing for restructuring and prolonging loans. With the recent exchange rate volatility and slowdown in the region, the share of restructured loans has increased to 5 percent.10 NPLs are somewhat higher in the trade, construction, and mortgage sectors. The highest incidence of loan restructuring is in the construction sector. NPLs are sensitive to exchange rate movements, reacting with a one-month lag.

Nonerforming loans and som-dollar exchange rate

Sources: Authorities data and IMF staff calculations.

Loans, NPLs, and Restructured Loans by Sector(Percent of total)
Dec-10Dec-11Dec-12Dec-13Dec-14Jan-15Feb-15Mar-15Apr-15May-15Jun-15Jul-15Aug-15
Share in Total Loans
Industry5.65.35.68.18.48.58.67.57.37.47.77.98.0
Agriculture12.813.115.417.618.818.518.119.721.021.521.121.120.5
Trade43.142.940.837.536.136.336.435.134.533.934.734.534.0
Construction7.85.65.55.86.26.46.36.15.15.15.15.25.2
Mortgage10.89.59.610.210.610.610.69.89.59.49.59.59.7
Consumer loans7.18.89.49.08.48.38.27.67.78.08.28.48.6
Other12.714.713.711.711.511.411.814.314.814.813.813.414.0
NPLs15.810.27.25.54.54.85.15.45.35.55.25.35.6
Industry12.812.17.710.05.76.15.66.25.85.75.36.96.9
Agriculture3.61.61.71.31.31.51.61.71.82.12.32.42.7
Trade15.38.46.35.24.85.25.76.46.46.86.76.56.9
Construction41.136.923.515.310.811.410.210.813.413.55.75.45.9
Mortgage13.39.26.44.63.64.04.14.55.15.35.56.26.6
Consumer loans7.22.93.42.42.42.72.83.13.13.23.43.13.1
Other22.817.112.28.26.76.57.76.86.06.16.86.76.7
NPLs in foreign currency11.67.85.34.03.43.63.94.14.04.13.83.74.0
NPLs in domestic currency4.262.401.871.551.051.151.281.291.371.471.481.601.59
Restructured Loans5.45.64.72.42.22.32.73.63.74.14.35.05.0
Prolonged Loans8.94.84.03.43.02.22.42.42.52.82.93.23.1
Source: Authorities data and IMF staff calculations.
Source: Authorities data and IMF staff calculations.

C. Dollarization trends

8. Dollarization has been declining since 2010, but the depreciation of the som over the past two years has started to reverse the trend. Dollarization of deposits fell below 50 percent in 2012, but increased sharply in 2014 and 2015. Even if dollarization is calculated at the constant exchange rate, deposit dollarization increased between 2014 and 2015, reflecting confidence effects of developments in Russia and Kazakhstan. The foreign currency loan-to-deposit ratio stabilized below 100 in June 2015, suggesting that dollar amounts of loans and deposits are similar. The limit on the net open foreign currency position is 15 percent of capital for a single currency, and 20 percent of capital in total. The foreign currency exposure for the banking system is negative as of end-July.

Dollarization Trends

D. Key vulnerabilities in the financial sector

9. Given its structure and recent trends, key vulnerabilities in the Kyrgyz financial system are emerging from:

  • Exchange rate related risks and dollarization. Even though foreign currency liabilities, upon which interest is paid, are similar to the foreign currency interest bearing assets, foreign liabilities may rise much faster than assets if the som depreciates due to developments in the region. In addition, a large depreciation may affect the ability of clients with unhedged banks to service loans, leading to a run on deposits.11 While unlikely to materialize, this would create a large currency mismatch for banks and diminish their capital. Stress tests conducted by the authorities at the end of August suggest that in the case of a 30 percent depreciation in the som—, the capital position of the banking sector would weaken.
  • Concentration of lending in key industries, namely trade and agriculture. Given that more than 50 percent of loans are concentrated in these industries and key bank clients may be connected, this may be a source of vulnerabilities for banks in the case of a further slowdown to the economy and exchange rate vulnerabilities. Large exposure are not yet an issue (the five largest exposures account for 56 percent of equity in the system and 80 percent for the five largest banks at end-July).
  • Credit risks mostly stem from currency-induced risks at the moment, but new ones may emerge. Loans, including consumer loans (except credit card), are collateralized and 90 percent of collateral is made up from real estate. The value of collateral varies across banks and types of projects, but usually exceeds value of loans as the haircut of value of collateral is applied. Some banks revise collateral with the fall of real estate prices, and occasionally ask borrowers to pledge additional collateral or repay the loans. However, the value of collateral may become an issue in the case of a real estate market crash12 and in the case of a severe slowdown in the economy, collateral may become illiquid, meaning a shortage of buyers for real-estate. Given small market and connected businesses, illiquidity or bankruptcy on one client can quickly transmit to the other clients.
  • A liquidity ratio of about 50 percent13 does not signal issues for concern, but maturity mismatches may give rise to new problems. Tight monetary policy to offset exchange rate and inflationary pressures may squeeze liquidity in banks too much. Liquidity stress test suggests that with an 18 percent deposit withdrawal, the liquidity ratio would fall by 2 percent for the largest banks.
  • While there is no evidence of weakening lending standards, the strong pace of credit growth suggests it is likely. Increased incidence of restructuring and prolonging (revolving credits) may mask the extent of the problems manifested in the balance sheet of borrowers.
  • Cross-border contagion is not a source of problem at the moment, due to limited integration in the international financial system. The role of Kazakh banks diminished following the 2009 crises, and these banks barely do any lending. However, following the Kyrgyz Republic’s accession to the Eurasian Economic Union (EEU), new banks could come to the Kyrgyz market and expand access to markets of other members of the EEU, which could in turn increase risks of cross-border contagion. Moreover, a number of banks have corresponding accounts in Russian and Kazakh banks, so if those banks encounter difficulties, Kyrgyz banks may face liquidity problems.

10. The pace of credit growth is not yet a concern. The BIS based credit-to-GDP gap suggests that credit is only 2 percent above the long-term trend and within normal range. The macro-financial forecast consistency check suggests that credit growth forecasts and economic dynamics are consistent. Despite being strong, the pace of credit growth may rather signal financial deepening than just a pre-crisis expansion.

Credit Gap

Sources: Authorities data and IMF credit-to-GDP

Distribution of Macro Outcomes, 2016

Sources: IMF macro-consistency check.

E. Going forward

11. Developing the financial sector will be critical for promoting growth, increasing financial inclusions, and providing a shock absorber. To address the above mentioned vulnerabilities, it is important to:

  • Maintain a stable macroeconomic environment and an appropriate policy mix that increases confidence in the banking system, including the NBRK, and lowers inflation. In this context, it is important to increase the traction of monetary policy and strengthen the interest rate channel. A symmetric and narrower corridor around the policy rate can help to increase its signaling power, while higher variety and volume of securities available for collateral can stimulate bank intermediation. In addition to measures aimed at strengthening the interbank market, it will be crucial to pass the draft banking code, which will help increase the autonomy of the NBKR, and consequently confidence in the banking system. Moreover, such measures would help attract additional bank deposits, and enable a healthy and balanced expansion of the sector. This would require the NBKR to balance the trade-off between exchange rate flexibility and the impact on the financial sector. In the event of exchange rates reaching the threshold for more serious problems in the banking sector, the NBKR should offer additional liquidity and SWAPs to banks. Moreover, the NBKR will need to balance the trade-off between the impact of tight liquidity on the banking sector, while curtailing the demand for foreign currency.
  • Implement market-driven and gradual de-dollarization policies aimed at building trust in the financial market. De-dollarization measures that have failed in the past in other countries include the forced conversion of foreign currency deposits and the suspension of access to foreign currency deposits. Therefore, such measures should be avoided. Similarly, forced conversion of loans is likely to create currency mismatches in banks, putting them in difficulty with servicing their loans, and distort market trends and interest rate structure, further impeding monetary policy transmission. Creating a market for local currency-denominated government securities—currently very shallow as parliament imposes a limit on these securities—would be an initial step. In addition, the securities market would help strengthen the transmission mechanisms and increase the effectiveness of monetary policy.
  • Develop financial hedging instruments. The NBRK should work with banks to develop hedging instruments, including SWAPs to enable banks’ clients to protect themselves against risks. Given the limited availability of SWAPs, their short maturity and high cost, an intervention by the regulator is needed. Thus, while protecting itself from foreign currency risk, the NBKR should begin to serve as a market maker and match banks with opposite SWAP demands. Ahead of any such measures, an intensive campaign aimed at educating the market about hedging instruments and forward exchange rates may be needed.
  • Strengthen the supervision of banks and regulations. In particular, rules on net open foreign exchange positions, classification of loans, provisioning, and restructuring should be carefully monitored and enforced. Moreover, continuing to strengthen banking sector supervision by developing a strategic plan to build supervisory capacity—and to provide for a robust supervisory program, including risk based supervision—will be crucial.
  • Design a complete macroprudential tool kit that enables the NBRK to react promptly when required. Such a kit should build on existing macroprudential measures introduced for different provisioning and reserve requirements for foreign currency assets and liabilities, and debt-to-income (DTI) ratios for consumer lending in foreign currency. The kit could for example include: (i) provisions for prohibiting lending in foreign currency to unhedged borrowers or for a particular sector; (ii) include higher provisioning for loans to the most vulnerable sectors (like mortgages, trade and construction); (iii) loan-to-deposit ratio for foreign currency; and (iv) DTI for mortgages and consumer lending in domestic currency etc. In designing such rules, the stability of the financial sector should take priority over pursuing government strategies to develop certain sectors/services (like provide affordable housing and affordable lending for productions).
  • Develop options for the NBRK to provide emergency loans in foreign currency to ensure that the DPA can pay out deposits on time and allow the government to become involved in times of systemic banking problems. A crises preparedness framework would ensure timely and coordinated action, as would a communication plan in the case of such problems, which would help to contain consumer panic.
1

Prepared by Dragana Ostojic.

2

Includes credit union, microcredit companies, microcredit agencies, microfinance companies, credit union, and special financial credit institutions.

3

The stock of government securities is 3 percent of GDP, of which banks hold 1.3 percent of GDP and institutional investors (social fund and deposit protection agency) hold 1.4 percent.

4

There are a few SWAP transactions, but with short maturity and high interest rates.

5

Currently, the Deposit Protections Agency insures deposits of individuals up to 100,000 som (1.4 percent of GDP per capita), covering 96.5 percent of individual account holders.

6

About 10 percent of population.

7

Som exchange rate from December 2013.

8

Affordable Loans for Farmers was introduced in 2013. The government provides subsidies for interest rates, so they are lower than market rates. The scheme is done via commercial banks, which can opt in or out.

9

Average size of loan reduced from US$700–800 to US$600–700 for the largest MFI (in a process of obtaining banking license. For the second largest, the average size of loan declined from US$550–600 to US$450–500.

10

According to regulation restructured loan should be classified as substandard until regular payments are received for 180 days. However, that is not the practice as at the moment.

11

While there is limited information on corporate and household balance sheets, anecdotal information suggests it takes a longer time to pay claims. The average time for accounts receivable has been extended from two weeks to two months.

12

Real estate market statistics are scarce. While some statistics suggest declines in real estate prices in Bishkek and Osh, it is difficult to assess the appropriate changes in prices.

13

Liquid assets over short-term liabilities.

Other Resources Citing This Publication