## 1.     Interest Spread Decomposition, by Different Bank Groups

## Source details

**Canonical URL:** [1.     Interest Spread Decomposition, by Different Bank Groups](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2004/_cr04336.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2004/_cr04336.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2004/_cr04336.pdf.json)

---

### Background and recent developments
- Banking sector growth since 2000 driven by strong economic growth; number of licensed banks fell from 71 in 1998 to 36 in 2003.
- Bank branches: 458 in 1998; 367 in January 2003; 359 in January 2004.
- Deposit insurance scheme introduced in 1999 (compulsory in 2004).
- Deposits constitute almost 70 percent of banks’ total liabilities.
- Currency composition: around 55½ percent of credit and 47 percent of deposits are denominated in foreign currency (mainly U.S. dollars).
- Broad money compared to non-oil GDP in 2003: 29½ percent.
- Mortgage credit increased four times in 2003; share of lending to households is small but rising.
- Loan allocation since 2000: almost 60 percent to industry and trade; agriculture share between 10 to 12 percent.
- Financial intermediation: Kazakhstan moved from below-average in 2001 to slightly above average in 2002 among countries with PPP per capita GNI between US$3,000 and US$10,000.

### Interest spread and drivers
- Nominal interest spreads: declined by around 3½ percentage points since 2000, rose slightly in 2002, and decreased again in 2003.
- Deposit rates: subject to a ceiling for banks participating in the deposit insurance scheme; deposit rate has not varied substantially across years.
- Administrative measures affecting spreads: ceiling on deposit rates; limits on tax deductibility of interest paid on deposits above thresholds (led some banks to charge higher lending rates).
- Decomposition methodology: ex post spread = interest earnings (percent of loans) − interest expenses (percent of deposits); follows Randall (1998).
- Non-quantifiable improvements reducing spreads: better creditor information (Credit Register regulation adopted 1998; reporting thresholds: tenge 3 million for legal entities, tenge 1 million for individuals), automated subsystem under development; law on credit bureaus submitted for approval.
- Institutional environment: weaker property rights enforcement and judicial effectiveness associated with higher margins in cross-country evidence; Kazakhstan is less economically “free” than EU-accession countries and has the highest spread among its neighboring country group.

### Heterogeneity across bank groups (aggregate banking system; three largest banks; Halyk Savings Bank)
- Categories examined: (a) aggregate banking system (group of 34 banks, excluding state-owned banks); (b) three largest banks (by asset size); (c) Halyk Savings Bank (analyzed separately).
- Spread dynamics:
  - Halyk Savings Bank had the largest decline in spread from 1999 to 2002, ending 2002 below the average banking system; its spread increased in 2003 due to an increase in its lending rate and a decrease in its deposit rate.
  - Three largest banks tended to increase interest margins to improve profitability, particularly when provisioning rose.
- Profitability (ROA) pattern:
  - In 2002 and 2003, ROA (profits in percent of assets) was a little higher for the aggregate banking system than for the three largest banks.
  - Halyk Savings Bank: ROA especially low in 2002, improved in 2003 as privatization benefits began to materialize.
- Cost structure and provisioning:
  - Personnel expenses/deposits have fallen for all categories since 1999; the decline is more drastic for Halyk Savings Bank.
  - Since 2000 the three largest banks had lower personnel expenses relative to deposits than the aggregate banking system.
  - Loan-loss provisions relative to deposits/assets remain high compared to peer countries; adoption of new provisioning rules and SME lending increased provisioning for larger banks.
- Non-interest income:
  - Larger banks expanding into insurance, securities, and real estate; fees and commissions are a non-interest income source.
  - For the aggregate banking system in 2002: fees and commissions received were almost 12 percent of operating income; fees and commission expense were around 4 percent of total operating expenses.
  - Except for Halyk Savings Bank, fees and commissions have been declining relative to deposits.

### Quantitative indicators from Table 1 (selected series, exact values preserved)
- Interest Rate on Loans = Interest Received/Loans (percent):
  - Aggregate: 1999 19.18; 2000 15.91; 2001 12.63; 2002 14.93; 2003 12.47
  - Three largest: 1999 19.94; 2000 17.06; 2001 14.98; 2002 14.55; 2003 13.05
  - Halyk Savings: 1999 23.10; 2000 14.81; 2001 12.87; 2002 12.55; 2003 13.03
- Interest Paid on Deposits = Interest Paid/Deposits (percent):
  - Aggregate: 1999 8.36; 2000 6.79; 2001 6.26; 2002 6.27; 2003 5.61
  - Three largest: 1999 6.16; 2000 5.93; 2001 5.69; 2002 6.17; 2003 5.73
  - Halyk Savings: 1999 4.07; 2000 4.38; 2001 4.69; 2002 5.60; 2003 5.32
- Spread (percentage points):
  - Aggregate: 1999 10.82; 2000 9.12; 2001 6.36; 2002 8.65; 2003 6.86
  - Three largest: 1999 13.77; 2000 11.13; 2001 9.29; 2002 8.38; 2003 7.32
  - Halyk Savings: 1999 19.03; 2000 10.42; 2001 8.17; 2002 6.95; 2003 7.71
- Reserve Requirements (RR):
  - Required reserves changed from 0.08 (8 percent) in 1999–2001 to 0.06 (6 percent) during 2002–2003. (Required reserves are unremunerated.)
- RR * (Interest Rate on loans) (percent):
  - Aggregate: 1999 1.53; 2000 1.27; 2001 1.01; 2002 0.90; 2003 0.75
  - Three largest: 1999 1.59; 2000 1.36; 2001 1.20; 2002 0.87; 2003 0.78
  - Halyk Savings: 1999 1.85; 2000 1.18; 2001 1.03; 2002 0.75; 2003 0.78
- ROA * (Assets/Deposits) (percent):
  - Aggregate: 1999 5.31; 2000 2.42; 2001 2.31; 2002 2.60; 2003 2.85
  - Three largest: 1999 6.43; 2000 2.39; 2001 1.53; 2002 3.20; 2003 3.42
  - Halyk Savings: 1999 1.06; 2000 -0.24; 2001 0.07; 2002 1.90; 2003 3.79
- Personnel Expenses/Deposits (percent):
  - Aggregate: 1999 7.77; 2000 3.46; 2001 2.36; 2002 1.80; 2003 1.49
  - Three largest: 1999 6.97; 2000 4.90; 2001 3.20; 2002 2.42; 2003 2.00
  - Halyk Savings: 1999 9.23; 2000 2.98; 2001 2.36; 2002 1.93; 2003 1.97
- Loan-loss Provisions/Deposits (percent):
  - Aggregate: 1999 7.91; 2000 3.04; 2001 4.87; 2002 6.70; 2003 5.37
  - Three largest: 1999 9.31; 2000 4.56; 2001 5.39; 2002 5.72; 2003 4.88
  - Halyk Savings: 1999 3.51; 2000 1.72; 2001 4.33; 2002 5.23; 2003 6.46
- Fees and Commissions Received/Deposits (percent):
  - Aggregate: 1999 5.43; 2000 4.39; 2001 3.71; 2002 3.13; 2003 3.20
  - Three largest: 1999 4.87; 2000 4.60; 2001 3.48; 2002 3.43; 2003 3.35
  - Halyk Savings: 1999 5.87; 2000 4.21; 2001 4.27; 2002 3.79; 2003 4.19
- Net Foreign Exchange Revaluation Gains/Deposits (percent):
  - Aggregate: 1999 --; 2000 --; 2001 --; 2002 1.27; 2003 0.19
  - Three largest: 1999 6.07; 2000 -0.10; 2001 0.28; 2002 0.91; 2003 0.15
  - Halyk Savings: 1999 --; 2000 --; 2001 --; 2002 0.36; 2003 0.31
- Memorandum ROA (Gross before-tax profits percent of end-of-period Assets):
  - Aggregate: 1999 2.95; 2000 1.50; 2001 1.51; 2002 1.71; 2003 1.84
  - Three largest: 1999 3.73; 2000 1.45; 2001 1.00; 2002 1.95; 2003 1.98
  - Halyk Savings: 1999 0.91; 2000 -0.20; 2001 0.06; 2002 1.30; 2003 2.35

### Analysis and interpretation
- Decline in spreads has been uneven across bank categories; deposit-rate ceilings help explain low variation in deposit rates.
- Lower operating expenses in 2002 did not uniformly reduce spreads because higher provisioning and SME lending raised credit costs, prompting some banks to raise lending rates.
- Foreign exchange revaluation gains were an important buffer in 1999 (e.g., 3.1 percent of total assets and 6.1 percent of deposits in 1999 for the banking system) but fell by 2003 with tenge appreciation (aggregate banking system FX revaluation gains fell to 0.15 percent of deposits in 2003 from 0.91 percent in 2002).
- Cross-country evidence indicates a (weak) positive relationship between net interest margins and CPI inflation; higher and more variable inflation is associated with higher spreads.
- Institutional and informational improvements (creditor information dissemination, creditor rights enforcement) have contributed to lowering perceived credit risk and supporting financial intermediation.

### Policy implications and recommendations
- A highly concentrated banking system could signal lack of competition but must be assessed in conjunction with contestability (openness to entry). Lack of license applications since 2001 raises questions about attractiveness of entry.
- Competitive and contestability conditions will be further enhanced by:
  - continuation of strong macroeconomic and financial policies;
  - improvements in financial sector regulation and enforcement.
- Administrative measures to control interest rates (for instance, the ceiling on deposit rates introduced by the Deposit Insurance Fund and tax incentives) distort financial intermediation and are likely to be counterproductive.

---

### 9.      Profitability of all three categories of banks has been increasing since 2001

### Profitability trends and spreads
- Profitability of all three categories of banks in Kazakhstan has been increasing since 2001 despite the decline in spreads.
- Kazakhstan’s experience contrasts with other countries that have generally had stable spreads over the period.
- During 1999–2002, the average spread and profitability for other oil-producing and EU-accession countries did not vary significantly, but both spread and profitability were lower than those for the three largest banks in Kazakhstan.
- Neighboring countries, with lower spreads, varied in profitability over the years and were harder hit by the Russian crisis in 1998.
- Foreign exchange revaluation gains helped cushion profitability but are uncertain and volatile; they were especially high for the three largest banks in 2002 and declined with the appreciation of the tenge against the US dollar in 2003.

### Efficiency of banking operations
- Banks in Kazakhstan are small by international standards; maintaining large networks tends to increase overhead expenses and reduce profitability.
- Various efficiency indicators analyze the cost of lending operations and labor productivity, treating banks as producing loans using deposits as main funding and labor as an input.
- Loans per employee increased for all categories and is highest for the three largest banks; these banks have introduced labor-saving products (ATMs, credit and debit cards).
- Non-interest expense as a share of deposits (indicator of branch maintenance costs) has come down, but remains relatively high for Halyk Savings Bank.

### Operating costs, income composition, and productivity (Table 2 highlights)
- Pre-tax Return on Assets (ROA):
  - Aggregate (2002): 1.95; (2003): 1.98
  - 3-largest (2002): 1.71; (2003): 1.84
  - 10-largest (2002): 1.52; (2003): 1.46
  - Halyk (2002): 1.30; (2003): 2.35
- Net interest Income/Assets:
  - Aggregate (2002): 5.27; (2003): 4.43
  - 3-largest (2002): 5.47; (2003): 4.43
  - 10-largest (2002): 5.38; (2003): 4.36
  - Halyk (2002): 4.35; (2003): 5.02
- Operating Income/Assets:
  - Aggregate (2002): 7.83; (2003): 9.02
  - 3-largest (2002): 15.76; (2003): 8.98
  - 10-largest (2002): 14.99; (2003): 8.80
  - Halyk (2002): 12.81; (2003): 10.22
- Non-Interest Expenses/Assets:
  - Aggregate (2002): 15.99; (2003): 7.26
  - 3-largest (2002): 17.83; (2003): 7.17
  - 10-largest (2002): 13.37; (2003): 7.12
  - Halyk (2002): 10.73; (2003): 8.22
- Net Interest Income/Operating Income:
  - Aggregate (2002): 29.57; (2003): 49.05
  - 3-largest (2002): 34.71; (2003): 49.33
  - 10-largest (2002): 35.91; (2003): 49.54
  - Halyk (2002): 33.93; (2003): 49.08
- Extraordinary Income/Operating Income:
  - Aggregate (2002): 0.87; (2003): 2.32
  - 3-largest (2002): 0.30; (2003): 0.01
  - 10-largest (2002): 0.79; (2003): 1.96
  - Halyk (2002): 0.01; (2003): 0.01
- Non-interest Expenses/Operating Income:
  - Aggregate (2002): 89.69; (2003): 80.51
  - 3-largest (2002): 89.41; (2003): 79.88
  - 10-largest (2002): 88.72; (2003): 80.90
  - Halyk (2002): 88.83; (2003): 80.43
- Non-interest Expenses/Deposits:
  - Aggregate (2002): 26.22; (2003): 11.93
  - 3-largest (2002): 22.16; (2003): 11.33
  - 10-largest (2002): 20.44; (2003): 11.15
  - Halyk (2002): 16.61; (2003): 13.41
- Loans/Employee (thous. KZT):
  - Aggregate (2002): 28.44; (2003): 40.43
  - 3-largest (2002): 1,152.32; (2003): 3,434.51
  - 10-largest (2002): 5.32; (2003): 6.2
- Net Profits/Employee (thous. KZT):
  - Aggregate (2002): 70.08; (2003): 84.98
  - 3-largest (2002): 2,213.93; (2003): 71.29
  - 10-largest (2002): 9.23; (2003): 0.57
  - Halyk (2002): 74.1; (2003): (value not legible in table)
- Observations:
  - Net interest income as percent of operating income rose from around 30 percent to 50 percent between 2002 and 2003, with the increase higher at the aggregate level—implying faster growth of intermediation for smaller banks.
  - Non-interest expense (including loan-loss provisions) as a share of assets declined substantially in 2003, but non-interest expenses as percent of operating income remained high—more than 80 percent for all categories in 2003 (down from around 88 percent in 2002).
  - Extraordinary income disproportionately benefited medium-sized banks relative to the largest banks.

### Cost structure and international comparison
- Cost of operations in Kazakhstan’s banking sector remained high, though progress was made during 2002–03 in lowering them.
- The three largest banks are the most labor-productive and more profitable than medium-sized banks.
- Cross-country comparisons (2002) show Kazakhstan had relatively higher cost of operations than many comparable countries, which could explain a relatively higher spread in Kazakhstan.
- Specific cross-country indicators shown for 2002 include Net Fees and Commissions as percentage of Assets, Personnel Expenses as percentage of Assets, and Loan-loss Provisions as percentage of Assets (Kazakhstan compared with Kyrgyz Republic, Russia, Uzbekistan, and a range of oil-producing and EU-accession countries).

### Competition, contestability, and market segmentation
- Banks appear fairly competitive within their own group, but not necessarily between groups; three largest banks compete with each other but may not view some medium-sized and smaller banks as competitors.
- Competition is stronger in and around big cities; smaller towns are served by fewer banks.
- Large Kazakhstani business groups’ access to international markets with lower borrowing rates has led larger banks to focus more on SMEs and retail business.
- Market is segmented geographically and by clientele, enabling product differentiation and specialization (e.g., mortgage market: medium-sized banks using Kazakhstan Mortgage Company (KMC) securitization vs. largest banks’ own mortgage programs vs. Zhilstroi Bank catering to low-income groups).
- Proposed capital account liberalization and increasing access to international capital markets are expected to promote competition.

### Concentration metrics and implications
- More than 80 percent of assets, loans, and deposits were held by the top 10 banks in 2003.
- In 2003, the three largest banks held 61 percent of assets and 66 percent of loans.
- Personal deposits: more than 82 percent held by the 10 largest banks and 69 percent by the 3 largest banks.
- Law on Competition: a dominant position is defined if aggregate share of up to three market participants constitutes 70 percent or more.
- Herfindahl-Hirschman indices for 34 banks (excluding the three state-owned banks) in 2003:
  - Assets: 1,402
  - Loans: 1,509
  - Deposits: 1,426
  - Personal deposits: 1,645
- For 34 banks, an even distribution would yield an index of 294; deviations (H*N/100) from an even distribution are 350, 383, 356, and 459 percentage points for assets, loans, deposits, and personal deposits respectively.
- These indices represent an improvement since 1998.
- A relatively high H index could help explain high interest spreads, but H might not fully reflect competition; empirical work suggests competitiveness is positively related to greater foreign entry and fewer entry and activity restrictions and is not necessarily negatively related to concentration.

### Entry and activity restrictions (selected indicators)
- Fraction of entry denied (averages):
  - ECA: 8.19
  - MENA: 19.05
  - EU: 3.67
  - Kazakhstan: -- (no new applications since 2001)
- Restrictions on foreign bank entry:
  - ECA: 0.06
  - MENA: 0.10
  - EU: 0.00
  - Kazakhstan: 1.00
- New entry requirements (number of legal submissions required, range 0 to 8):
  - ECA: 7.39
  - MENA: 7.36
  - EU: 7.07
  - Kazakhstan: 8.00
- Activity restriction indices (range 1 to 4 for Securities, Insurance, Real estate):
  - Securities: ECA 1.51; MENA 1.45; EU 1.13; Kazakhstan 1.00
  - Insurance: ECA 2.27; MENA 3.45; EU 2.20; Kazakhstan 1.00
  - Real estate: ECA 2.27; MENA 3.73; EU 1.87; Kazakhstan 1.00
- Other characteristics:
  - 5-bank concentration of deposits:
    - ECA: 65.42; MENA: 72.04; EU: 59.19; Kazakhstan: 64.00
  - Initial capital stringency (index where lower values indicate higher stringency):
    - ECA: 1.86; MENA: 1.45; EU: 1.93; Kazakhstan: 2.00
- Observations:
  - Kazakhstan required all eight submissions for entry-into-banking requirements (highest among compared regions).
  - Kazakhstan shows no restrictions for securities, insurance, and real estate activities per the indices.
  - There were no new applications for entry into banking business in Kazakhstan since 2001 (with the exception noted elsewhere of state-owned Zhilstroi Bank licensed in 2003).

### Contestability and barriers to entry
- Contestability—openness of the market to new entrants—is important for competitiveness; lack of contestability can raise spreads if regulatory restrictions reduce entry threat.
- Lack of license applications since 2001 raises questions; reports mention cumbersome bureaucracy, incomplete implementation of laws, and questionable settlement of investment disputes as factors discouraging new banking entrants.
- Collaboration exists between the Agency on Regulation and Supervision of Financial Markets and Financial Institutions (FSA) and the Anti-Monopoly Authority to meet consumer protection mandates.

---

### 20.      Rapid progress is being made in developing an efficient banking sector, and

### Summary and key observations
- Rapid progress is being made in developing an efficient banking sector; further improvements will depend on good supervisory and regulatory enforcement and improving competitive conditions.
- The level of financial intermediation, although high among neighboring countries and growing, is still low in broader comparisons.
- The interest rate spread is higher than for other countries in the peer group.
- Interest margins have declined across the three different categories of banks, and profitability has been increasing since 2001.
- Banks are moving into businesses involving fee-based incomes, but net interest income has been growing as a proportion of operating income.
- Operational costs remain high compared to other countries, but are likely to improve both within and between different groups of banks, and in non-urban regions.
- Certain administrative measures like imposing a ceiling on deposit rates and limits on tax deductibility of interest paid by banks over a certain threshold are likely to be counterproductive in reducing spreads.
- Although the banking market is fairly concentrated, the lack of license applications, in spite of profitable banks, raises the question of desirability of doing banking business in Kazakhstan; this could indicate deeper problems involving institutional setups that were beyond the scope of this analysis.
- Capital account liberalization will lead to improvements in contestability conditions as may Kazakhstan’s eventual accession to the WTO.

### Interest spread decomposition and accounting framework
- Consolidated income identity (gross profits P and components):
  - P = Interest Income (II) + Fees and Commissions received (FC) + Net Foreign Exchange Revaluation Gains (FX) − Interest Paid (IP) − Personnel Expenses (PE) − Loan-loss Provisions (PROV) + Other factors (U)
- Expressed per deposits and loans, and using required reserve ratio RR, the interest rate spread identity is presented in the source as:
  - LDL APEPROVFCFX ii RRi ROAe DD    D    D D −≡   ++ +    − − +
  (Definitions preserved from source)
- Definitions given:
  - ROA = Return on Assets
  - L i = average interest charged on loans
  - D i = average interest charged on deposits
  - e = residual items not taken into account (reflecting flow/stock combination errors and simplifying assumptions)

### Concentration analysis and appendix notes
- Bank-level data (1998–2003) and cumulative shares of Deposits, Assets, Loans, Personal Deposits are provided in the source (appendix table).
- As of November 2003, Statistical Bulletin, National Bank of Kazakhstan.
- Explanation of indices used for entry and activity restrictions (appendix III):
  - Fraction of entry denied — percentage of total entry applications.
  - Restrictions on foreign bank entry — index 1 for any restrictions, 0 for no restrictions.
  - New entry requirements — up to eight documentary submissions; index ranges 0 to 8.
  - Restrictions on entry into other activities — coded 1 to 4; index ranges 1 to 4.
  - Initial capital stringency — index ranges 0 to 3; higher value signifies less stringency.

*Source: IMF staff report sections titled "1.     Interest Spread Decomposition, by Different Bank Groups", "9.      Profitability of all three categories of banks has been increasing since 2001", and "20.      Rapid progress is being made in developing an efficient banking sector, and" (PDF: _cr04336). *

### 1.     Interest Spread Decomposition, by Different Bank Groups ............................................12

### 1.     Interest Spread Decomposition, by Different Bank Groups

### Background and recent developments
- Banking sector growth since 2000 driven by strong economic growth; number of licensed banks fell from 71 in 1998 to 36 in 2003.
- Bank branches: 458 in 1998; 367 in January 2003; 359 in January 2004.
- Deposit insurance scheme introduced in 1999 (compulsory in 2004).
- Deposits constitute almost 70 percent of banks’ total liabilities.
- Currency composition: around 55½ percent of credit and 47 percent of deposits are denominated in foreign currency (mainly U.S. dollars).
- Broad money compared to non-oil GDP in 2003: 29½ percent.
- Mortgage credit increased four times in 2003; share of lending to households is small but rising.
- Loan allocation since 2000: almost 60 percent to industry and trade; agriculture share between 10 to 12 percent.
- Financial intermediation: Kazakhstan moved from below-average in 2001 to slightly above average in 2002 among countries with PPP per capita GNI between US$3,000 and US$10,000.

### Interest spread and drivers
- Nominal interest spreads: declined by around 3½ percentage points since 2000, rose slightly in 2002, and decreased again in 2003.
- Deposit rates: subject to a ceiling for banks participating in the deposit insurance scheme; deposit rate has not varied substantially across years.
- Administrative measures affecting spreads: ceiling on deposit rates; limits on tax deductibility of interest paid on deposits above thresholds (led some banks to charge higher lending rates).
- Decomposition methodology: ex post spread = interest earnings (percent of loans) − interest expenses (percent of deposits); follows Randall (1998).
- Non-quantifiable improvements reducing spreads: better creditor information (Credit Register regulation adopted 1998; reporting thresholds: tenge 3 million for legal entities, tenge 1 million for individuals), automated subsystem under development; law on credit bureaus submitted for approval.
- Institutional environment: weaker property rights enforcement and judicial effectiveness associated with higher margins in cross-country evidence; Kazakhstan is less economically “free” than EU-accession countries and has the highest spread among its neighboring country group.

### Heterogeneity across bank groups (aggregate banking system; three largest banks; Halyk Savings Bank)
- Categories examined: (a) aggregate banking system (group of 34 banks, excluding state-owned banks); (b) three largest banks (by asset size); (c) Halyk Savings Bank (analyzed separately).
- Spread dynamics:
  - Halyk Savings Bank had the largest decline in spread from 1999 to 2002, ending 2002 below the average banking system; its spread increased in 2003 due to an increase in its lending rate and a decrease in its deposit rate.
  - Three largest banks tended to increase interest margins to improve profitability, particularly when provisioning rose.
- Profitability (ROA) pattern:
  - In 2002 and 2003, ROA (profits in percent of assets) was a little higher for the aggregate banking system than for the three largest banks.
  - Halyk Savings Bank: ROA especially low in 2002, improved in 2003 as privatization benefits began to materialize.
- Cost structure and provisioning:
  - Personnel expenses/deposits have fallen for all categories since 1999; the decline is more drastic for Halyk Savings Bank.
  - Since 2000 the three largest banks had lower personnel expenses relative to deposits than the aggregate banking system.
  - Loan-loss provisions relative to deposits/assets remain high compared to peer countries; adoption of new provisioning rules and SME lending increased provisioning for larger banks.
- Non-interest income:
  - Larger banks expanding into insurance, securities, and real estate; fees and commissions are a non-interest income source.
  - For the aggregate banking system in 2002: fees and commissions received were almost 12 percent of operating income; fees and commission expense were around 4 percent of total operating expenses.
  - Except for Halyk Savings Bank, fees and commissions have been declining relative to deposits.

### Quantitative indicators from Table 1 (selected series, exact values preserved)
- Interest Rate on Loans = Interest Received/Loans (percent):
  - Aggregate: 1999 19.18; 2000 15.91; 2001 12.63; 2002 14.93; 2003 12.47
  - Three largest: 1999 19.94; 2000 17.06; 2001 14.98; 2002 14.55; 2003 13.05
  - Halyk Savings: 1999 23.10; 2000 14.81; 2001 12.87; 2002 12.55; 2003 13.03
- Interest Paid on Deposits = Interest Paid/Deposits (percent):
  - Aggregate: 1999 8.36; 2000 6.79; 2001 6.26; 2002 6.27; 2003 5.61
  - Three largest: 1999 6.16; 2000 5.93; 2001 5.69; 2002 6.17; 2003 5.73
  - Halyk Savings: 1999 4.07; 2000 4.38; 2001 4.69; 2002 5.60; 2003 5.32
- Spread (percentage points):
  - Aggregate: 1999 10.82; 2000 9.12; 2001 6.36; 2002 8.65; 2003 6.86
  - Three largest: 1999 13.77; 2000 11.13; 2001 9.29; 2002 8.38; 2003 7.32
  - Halyk Savings: 1999 19.03; 2000 10.42; 2001 8.17; 2002 6.95; 2003 7.71
- Reserve Requirements (RR):
  - Required reserves changed from 0.08 (8 percent) in 1999–2001 to 0.06 (6 percent) during 2002–2003. (Required reserves are unremunerated.)
- RR * (Interest Rate on loans) (percent):
  - Aggregate: 1999 1.53; 2000 1.27; 2001 1.01; 2002 0.90; 2003 0.75
  - Three largest: 1999 1.59; 2000 1.36; 2001 1.20; 2002 0.87; 2003 0.78
  - Halyk Savings: 1999 1.85; 2000 1.18; 2001 1.03; 2002 0.75; 2003 0.78
- ROA * (Assets/Deposits) (percent):
  - Aggregate: 1999 5.31; 2000 2.42; 2001 2.31; 2002 2.60; 2003 2.85
  - Three largest: 1999 6.43; 2000 2.39; 2001 1.53; 2002 3.20; 2003 3.42
  - Halyk Savings: 1999 1.06; 2000 -0.24; 2001 0.07; 2002 1.90; 2003 3.79
- Personnel Expenses/Deposits (percent):
  - Aggregate: 1999 7.77; 2000 3.46; 2001 2.36; 2002 1.80; 2003 1.49
  - Three largest: 1999 6.97; 2000 4.90; 2001 3.20; 2002 2.42; 2003 2.00
  - Halyk Savings: 1999 9.23; 2000 2.98; 2001 2.36; 2002 1.93; 2003 1.97
- Loan-loss Provisions/Deposits (percent):
  - Aggregate: 1999 7.91; 2000 3.04; 2001 4.87; 2002 6.70; 2003 5.37
  - Three largest: 1999 9.31; 2000 4.56; 2001 5.39; 2002 5.72; 2003 4.88
  - Halyk Savings: 1999 3.51; 2000 1.72; 2001 4.33; 2002 5.23; 2003 6.46
- Fees and Commissions Received/Deposits (percent):
  - Aggregate: 1999 5.43; 2000 4.39; 2001 3.71; 2002 3.13; 2003 3.20
  - Three largest: 1999 4.87; 2000 4.60; 2001 3.48; 2002 3.43; 2003 3.35
  - Halyk Savings: 1999 5.87; 2000 4.21; 2001 4.27; 2002 3.79; 2003 4.19
- Net Foreign Exchange Revaluation Gains/Deposits (percent):
  - Aggregate: 1999 --; 2000 --; 2001 --; 2002 1.27; 2003 0.19
  - Three largest: 1999 6.07; 2000 -0.10; 2001 0.28; 2002 0.91; 2003 0.15
  - Halyk Savings: 1999 --; 2000 --; 2001 --; 2002 0.36; 2003 0.31
- Memorandum ROA (Gross before-tax profits percent of end-of-period Assets):
  - Aggregate: 1999 2.95; 2000 1.50; 2001 1.51; 2002 1.71; 2003 1.84
  - Three largest: 1999 3.73; 2000 1.45; 2001 1.00; 2002 1.95; 2003 1.98
  - Halyk Savings: 1999 0.91; 2000 -0.20; 2001 0.06; 2002 1.30; 2003 2.35

### Analysis and interpretation
- Decline in spreads has been uneven across bank categories; deposit-rate ceilings help explain low variation in deposit rates.
- Lower operating expenses in 2002 did not uniformly reduce spreads because higher provisioning and SME lending raised credit costs, prompting some banks to raise lending rates.
- Foreign exchange revaluation gains were an important buffer in 1999 (e.g., 3.1 percent of total assets and 6.1 percent of deposits in 1999 for the banking system) but fell by 2003 with tenge appreciation (aggregate banking system FX revaluation gains fell to 0.15 percent of deposits in 2003 from 0.91 percent in 2002).
- Cross-country evidence indicates a (weak) positive relationship between net interest margins and CPI inflation; higher and more variable inflation is associated with higher spreads.
- Institutional and informational improvements (creditor information dissemination, creditor rights enforcement) have contributed to lowering perceived credit risk and supporting financial intermediation.

### Policy implications and recommendations (as presented)
- A highly concentrated banking system could signal lack of competition but must be assessed in conjunction with contestability (openness to entry). Lack of license applications since 2001 raises questions about attractiveness of entry.
- Competitive and contestability conditions will be further enhanced by:
  - continuation of strong macroeconomic and financial policies;
  - improvements in financial sector regulation and enforcement.
- Administrative measures to control interest rates (for instance, the ceiling on deposit rates introduced by the Deposit Insurance Fund and tax incentives) distort financial intermediation and are likely to be counterproductive.

* _cr04336 - 1.     Interest Spread Decomposition, by Different Bank Groups_

### 9.      Profitability of all three categories of banks has been increasing since 2001

### 9.      Profitability of all three categories of banks has been increasing since 2001

### Profitability trends and spreads
- Profitability of all three categories of banks in Kazakhstan has been increasing since 2001 despite the decline in spreads.
- Kazakhstan’s experience contrasts with other countries that have generally had stable spreads over the period.
- During 1999–2002, the average spread and profitability for other oil-producing and EU-accession countries did not vary significantly, but both spread and profitability were lower than those for the three largest banks in Kazakhstan.
- Neighboring countries, with lower spreads, varied in profitability over the years and were harder hit by the Russian crisis in 1998.
- Foreign exchange revaluation gains helped cushion profitability but are uncertain and volatile; they were especially high for the three largest banks in 2002 and declined with the appreciation of the tenge against the US dollar in 2003.

### Efficiency of banking operations
- Banks in Kazakhstan are small by international standards; maintaining large networks tends to increase overhead expenses and reduce profitability.
- Various efficiency indicators analyze the cost of lending operations and labor productivity, treating banks as producing loans using deposits as main funding and labor as an input.
- Loans per employee increased for all categories and is highest for the three largest banks; these banks have introduced labor-saving products (ATMs, credit and debit cards).
- Non-interest expense as a share of deposits (indicator of branch maintenance costs) has come down, but remains relatively high for Halyk Savings Bank.

### Operating costs, income composition, and productivity (Table 2 highlights)
- Pre-tax Return on Assets (ROA):
  - Aggregate (2002): 1.95; (2003): 1.98
  - 3-largest (2002): 1.71; (2003): 1.84
  - 10-largest (2002): 1.52; (2003): 1.46
  - Halyk (2002): 1.30; (2003): 2.35
- Net interest Income/Assets:
  - Aggregate (2002): 5.27; (2003): 4.43
  - 3-largest (2002): 5.47; (2003): 4.43
  - 10-largest (2002): 5.38; (2003): 4.36
  - Halyk (2002): 4.35; (2003): 5.02
- Operating Income/Assets:
  - Aggregate (2002): 7.83; (2003): 9.02
  - 3-largest (2002): 15.76; (2003): 8.98
  - 10-largest (2002): 14.99; (2003): 8.80
  - Halyk (2002): 12.81; (2003): 10.22
- Non-Interest Expenses/Assets:
  - Aggregate (2002): 15.99; (2003): 7.26
  - 3-largest (2002): 17.83; (2003): 7.17
  - 10-largest (2002): 13.37; (2003): 7.12
  - Halyk (2002): 10.73; (2003): 8.22
- Net Interest Income/Operating Income:
  - Aggregate (2002): 29.57; (2003): 49.05
  - 3-largest (2002): 34.71; (2003): 49.33
  - 10-largest (2002): 35.91; (2003): 49.54
  - Halyk (2002): 33.93; (2003): 49.08
- Extraordinary Income/Operating Income:
  - Aggregate (2002): 0.87; (2003): 2.32
  - 3-largest (2002): 0.30; (2003): 0.01
  - 10-largest (2002): 0.79; (2003): 1.96
  - Halyk (2002): 0.01; (2003): 0.01
- Non-interest Expenses/Operating Income:
  - Aggregate (2002): 89.69; (2003): 80.51
  - 3-largest (2002): 89.41; (2003): 79.88
  - 10-largest (2002): 88.72; (2003): 80.90
  - Halyk (2002): 88.83; (2003): 80.43
- Non-interest Expenses/Deposits:
  - Aggregate (2002): 26.22; (2003): 11.93
  - 3-largest (2002): 22.16; (2003): 11.33
  - 10-largest (2002): 20.44; (2003): 11.15
  - Halyk (2002): 16.61; (2003): 13.41
- Loans/Employee (thous. KZT):
  - Aggregate (2002): 28.44; (2003): 40.43
  - 3-largest (2002): 1,152.32; (2003): 3,434.51
  - 10-largest (2002): 5.32; (2003): 6.2
- Net Profits/Employee (thous. KZT):
  - Aggregate (2002): 70.08; (2003): 84.98
  - 3-largest (2002): 2,213.93; (2003): 71.29
  - 10-largest (2002): 9.23; (2003): 0.57
  - Halyk (2002): 74.1; (2003): (value not legible in table)
- Observations:
  - Net interest income as percent of operating income rose from around 30 percent to 50 percent between 2002 and 2003, with the increase higher at the aggregate level—implying faster growth of intermediation for smaller banks.
  - Non-interest expense (including loan-loss provisions) as a share of assets declined substantially in 2003, but non-interest expenses as percent of operating income remained high—more than 80 percent for all categories in 2003 (down from around 88 percent in 2002).
  - Extraordinary income disproportionately benefited medium-sized banks relative to the largest banks.

### Cost structure and international comparison
- Cost of operations in Kazakhstan’s banking sector remained high, though progress was made during 2002–03 in lowering them.
- The three largest banks are the most labor-productive and more profitable than medium-sized banks.
- Figure 6 (cross-country comparisons, 2002) shows Kazakhstan had relatively higher cost of operations than many comparable countries, which could explain a relatively higher spread in Kazakhstan.
- Specific cross-country indicators shown for 2002 include Net Fees and Commissions as percentage of Assets, Personnel Expenses as percentage of Assets, and Loan-loss Provisions as percentage of Assets (Kazakhstan compared with Kyrgyz Republic, Russia, Uzbekistan, and a range of oil-producing and EU-accession countries).

### Competition, contestability, and market segmentation
- Banks appear fairly competitive within their own group, but not necessarily between groups; three largest banks compete with each other but may not view some medium-sized and smaller banks as competitors.
- Competition is stronger in and around big cities; smaller towns are served by fewer banks.
- Large Kazakhstani business groups’ access to international markets with lower borrowing rates has led larger banks to focus more on SMEs and retail business.
- Market is segmented geographically and by clientele, enabling product differentiation and specialization (e.g., mortgage market: medium-sized banks using Kazakhstan Mortgage Company (KMC) securitization vs. largest banks’ own mortgage programs vs. Zhilstroi Bank catering to low-income groups).
- Proposed capital account liberalization and increasing access to international capital markets are expected to promote competition.

### Concentration metrics and implications
- More than 80 percent of assets, loans, and deposits were held by the top 10 banks in 2003.
- In 2003, the three largest banks held 61 percent of assets and 66 percent of loans.
- Personal deposits: more than 82 percent held by the 10 largest banks and 69 percent by the 3 largest banks.
- Law on Competition: a dominant position is defined if aggregate share of up to three market participants constitutes 70 percent or more.
- Herfindahl-Hirschman indices for 34 banks (excluding the three state-owned banks) in 2003:
  - Assets: 1,402
  - Loans: 1,509
  - Deposits: 1,426
  - Personal deposits: 1,645
- For 34 banks, an even distribution would yield an index of 294; deviations (H*N/100) from an even distribution are 350, 383, 356, and 459 percentage points for assets, loans, deposits, and personal deposits respectively.
- These indices represent an improvement since 1998.
- A relatively high H index could help explain high interest spreads, but H might not fully reflect competition; empirical work suggests competitiveness is positively related to greater foreign entry and fewer entry and activity restrictions and is not necessarily negatively related to concentration.

### Entry and activity restrictions (Table 4 highlights)
- Indices compare Europe & Central Asia (ECA), Middle East and North Africa (MENA), European Union (EU), and Kazakhstan (latest available data, as of end-2003).
- Fraction of entry applications denied and restrictions on foreign bank entry (index range 0 to 1):
  - Fraction of entry denied (averages):
    - ECA: 8.19
    - MENA: 19.05
    - EU: 3.67
    - Kazakhstan: -- (no new applications since 2001)
  - Restrictions on foreign bank entry:
    - ECA: 0.06
    - MENA: 0.10
    - EU: 0.00
    - Kazakhstan: 1.00
- New entry requirements (number of legal submissions required, range 0 to 8):
  - ECA: 7.39
  - MENA: 7.36
  - EU: 7.07
  - Kazakhstan: 8.00
- Activity restriction indices (range 1 to 4 for Securities, Insurance, Real estate):
  - Securities:
    - ECA: 1.51
    - MENA: 1.45
    - EU: 1.13
    - Kazakhstan: 1.00
  - Insurance:
    - ECA: 2.27
    - MENA: 3.45
    - EU: 2.20
    - Kazakhstan: 1.00
  - Real estate:
    - ECA: 2.27
    - MENA: 3.73
    - EU: 1.87
    - Kazakhstan: 1.00
- Other characteristics:
  - 5-bank concentration of deposits:
    - ECA: 65.42
    - MENA: 72.04
    - EU: 59.19
    - Kazakhstan: 64.00
  - Initial capital stringency (index where lower values indicate higher stringency):
    - ECA: 1.86
    - MENA: 1.45
    - EU: 1.93
    - Kazakhstan: 2.00
- Observations:
  - Kazakhstan required all eight submissions for entry-into-banking requirements (highest among compared regions).
  - Kazakhstan shows no restrictions for securities, insurance, and real estate activities per the indices, consistent with high net fees and commissions to total assets.
  - There were no new applications for entry into banking business in Kazakhstan since 2001 (with the exception noted elsewhere of state-owned Zhilstroi Bank licensed in 2003).

### Contestability and barriers to entry
- Contestability—openness of the market to new entrants—is important for competitiveness; lack of contestability can raise spreads if regulatory restrictions reduce entry threat.
- Lack of license applications since 2001 raises questions; reports mention cumbersome bureaucracy, incomplete implementation of laws, and questionable settlement of investment disputes as factors discouraging new banking entrants.
- Collaboration exists between the Agency on Regulation and Supervision of Financial Markets and Financial Institutions (FSA) and the Anti-Monopoly Authority to meet consumer protection mandates.

*Source: IMF staff report section titled "9.      Profitability of all three categories of banks has been increasing since 2001" (PDF: _cr04336 - 9.      Profitability of all three categories of banks has been increasing since 2001).*

### 20.      Rapid progress is being made in developing an efficient banking sector, and

### 20.      Rapid progress is being made in developing an efficient banking sector, and

### Summary and key observations
- Rapid progress is being made in developing an efficient banking sector; further improvements will depend on good supervisory and regulatory enforcement and improving competitive conditions.
- The level of financial intermediation, although high among neighboring countries and growing, is still low in broader comparisons.
- The interest rate spread is higher than for other countries in the peer group.
- Interest margins have declined across the three different categories of banks, and profitability has been increasing since 2001.
- Banks are moving into businesses involving fee-based incomes, but net interest income has been growing as a proportion of operating income.
- Operational costs remain high compared to other countries, but are likely to improve both within and between different groups of banks, and in non-urban regions.
- Certain administrative measures like imposing a ceiling on deposit rates and limits on tax deductibility of interest paid by banks over a certain threshold are likely to be counterproductive in reducing spreads.
- Although the banking market is fairly concentrated, the lack of license applications, in spite of profitable banks, raises the question of desirability of doing banking business in Kazakhstan; this could indicate deeper problems involving institutional setups that were beyond the scope of this analysis.
- Capital account liberalization will lead to improvements in contestability conditions as may Kazakhstan’s eventual accession to the WTO.

### Interest spread decomposition and accounting framework
- The consolidated income statement defines gross profits (profit before taxes), P, as Interest Income (II) plus non-interest income that includes fees and commissions received (FC) and Net Foreign Exchange Revaluation Gains (FX) minus interest expense (IP), operating cost that can be broken down into Personnel expenses (PE) and overheads (not shown here), and Loan-loss Provisions (PROV). Other (U) factors like other non-interest income and expenses, not taken into account, could also affect profitability.
- The identity expressed as the interest margin is shown in the source as:
  - IIIPPE    PROVP    FC    FX    U−≡ +   +− − +
- Dividing by Deposits (D) and using loans (L) and assets (A) yields:
  - ** IILIPPEPROVPAFCFXU LDD  DD    AD D  D D −≡ +   +  − − + **
- Using the fact that 1() L required reserves ratio  RR D ≡−, and that interest incomes and expenses are average lending rate ( L i ) and deposit rate ( D i ) times volumes, the resulting expression for the interest rate spread is:
  - ** LDL APEPROVFCFX ii RRi ROAe DD    D    D D −≡   ++ +    − − + **
- Definitions given:
  - ROA = Return on Assets
  - L i = average interest charged on loans
  - D i = average interest charged on deposits
  - e = residual items not taken into account (not shown in the table), also reflects errors that might result from combining flow data and stock data, as well as the simplifying assumption that loanable funds consist of deposits net of required reserves.

### Concentration analysis (appendix highlights)
- A table of bank-level data (1998–2003) and cumulative shares of Deposits, Assets, Loans, Personal Deposits is provided in the source (appendix table).
- As of November 2003, Statistical Bulletin, National Bank of Kazakhstan.
- The difference between individual bank data for 22 banks and aggregate data distributed evenly between the rest of the banks is noted.

### Entry and activity restrictions: explanation of indices (appendix III)
- Fraction of entry denied — measured as percentage of total entry applications.
- Restrictions on foreign bank entry — index is 1 for any restrictions, and 0 for no restrictions.
- New entry requirements — eight possible documentary submissions; each assigned 1 if required or 0 if not. Documents: (1) Draft By-laws, (2) Intended Organizational Chart, (3) First 3-year financial projections, (4) Financial information on main potential shareholders, (5) Background/experience of future directors, (6) Background/experience of future managers, (7) Sources of funds to be used to capitalize the new bank, and (8) Intended differentiation of new bank from other banks. Index ranges from 0 to 8.
- Restrictions on entry into other activities — degree to which banks are allowed to engage in fee-based activities. Responses coded 1 for ‘unrestricted,’ 2 for ‘permitted,’ 3 for ‘restricted,’ and 4 for ‘prohibited.’ Index ranges from 1 to 4.
- Initial capital stringency — three questions with answers assigned 1 for ‘yes’ and 0 for ‘no’: (1) Can initial and subsequent infusions of regulatory capital include assets other than cash or government securities; (2) Can the initial infusion of capital be based on borrowed funds; (3) Are the sources of funds that count as regulatory capital verified by the regulatory or supervisory authorities. The index ranges from 0 to 3. Higher value signifies less stringency.

*Source: _cr04336 - 20.      Rapid progress is being made in developing an efficient banking sector, and*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2004/_cr04336.pdf_