Journal Issue
Share
Article

Republic of Kazakhstan: Staff Report for the 2015 Article IV Consultation

Author(s):
International Monetary Fund. Middle East and Central Asia Dept.
Published Date:
September 2015
Share
  • ShareShare
Show Summary Details

Context

1. Like other countries in the region, Kazakhstan was hit in 2014 by large external shocks that are expected to be mostly long lasting. Against the backdrop of these shocks, economic growth has decelerated sharply, financial conditions have tightened, and external imbalances are emerging. While both oil prices and the Russian ruble are off their respective lows reached at end-2014, improvements in sentiment are tempered by the realization that low oil prices and uncertainties surrounding the Russian economy are likely to persist.

2. The authorities’ large stimulus package has helped blunt the initial impact of the shocks, but recovery will likely be gradual. The stimulus package (“Nulrly Zhulu”), which went into effect in 2014 and which involves public investment programs supported by multilateral development banks (MDBs), is partially offsetting the impact of the shocks. The stimulus is enhanced by a structural reform program aimed at improving the business climate and public sector efficiency. However, the extent of the shocks, the appreciation of the tenge in real effective terms, and a weak lending environment could reduce the prospects for quick economic recovery. At the same time, Kazakhstan’s upcoming accession to the World Trade Organization (WTO) could provide a boost to investor confidence.

3. President Nazarbayev won early elections overwhelmingly, and has suggested devolving power to the parliament and the regions. The elections took place on April 26, 2015 (previously scheduled in 2016). The President won reelection with 98 percent of the vote, and has signaled his interest in strengthening the role of parliament in decision making, including speeding up the process of transferring new powers to parliament and, through it, to the prime minister. The “100 steps” reform program announced by the President shortly after his reelection includes a strong emphasis on decentralization and strengthening the rule of law.

Recent Developments

4. Economic growth has decelerated significantly, but remains positive, as a result of the external shocks. GDP growth slowed to an annualized 2 percent during the first quarter of 2015, down from around 4 percent in 2014 and 6 percent in 2013. In addition to weaker external demand, slower growth was driven by the impact of lower income and profitability (resulting from lower oil prices) and confidence effects (reflecting regional developments) on private consumption and domestic investment. Private consumption expanded by 0.5 percent in 2014 (vs. 11.8 percent in 2013), while investment contracted by 3.4 percent (vs. 9 percent growth in 2013). The stimulus package has helped absorb some of the impact on consumption and investment. In the face of slowing demand and limited exchange rate movements, headline inflation fell sharply, from 7.4 percent year-on-year at end-2014 to 3.9 percent at end-June, 2015—well below the authorities’ objective range of 6-8 percent. The recorded unemployment rate remained at 5 percent in 2014.

5. The fiscal stimulus aimed at supporting growth has increased financing needs. The overall fiscal surplus fell from 5 percent of GDP in 2013 to 1.7 percent of GDP in 2014. Among other factors, the decline reflects a 1.8 percent of GDP increase in expenditures and net lending (partly due to the stimulus) and a 0.5 percent of GDP fall in oil revenues. The non-oil fiscal deficit increased from 6.8 percent of GDP in 2013 to 9.6 percent of GDP in 2014, while the stock of government debt increased from 12.9 percent of GDP to 14.5 percent of GDP over the same period. On July 14, 2015, with aim of financing a growing fiscal deficit and in response to reduced spreads in recent months, the authorities issued $4 billion in 10-year and 30-year global bonds. The issuance was well received and oversubscribed, with spreads of 285 bps and 335 bps, respectively, over U.S. treasuries.

6. Monetary conditions have been tight for most of the past year, which have reduced pressures on the exchange rate recently, but, together with lower economic activity, slowed lending sharply. During the second half of 2014, due to rapid depreciation of the ruble, heightened speculation of tenge devaluation, and increased dollarization of bank deposits, domestic money market interest rates rose significantly. Since early 2015, with the use of overnight currency swap and repo instruments, the National Bank (NBK) has managed to stabilize money market rates below 10 percent and dampen pressure on the tenge. As a result, and following the successful global bond issuance, the NBK widened its exchange rate band from 170-188 to 170-198 tenge/dollar, on July 15, 2015. Meanwhile, while Kazakhstan’s sovereign bond spreads have declined somewhat, domestic lending rates remain elevated and overall private sector credit has grinded to a halt, contracting by 3.5 percent year-on-year as of May 2015 (vs. annualized growth of 6 percent in 2014 and 13 percent in 2013). Credit conditions are expected to remain anemic in the short term (Box 1 and Annex I).

7. The external position has weakened largely due to the fall in oil prices. Kazakhstan’s real effective exchange rate has appreciated over the past year, mainly reflecting the depreciation of the ruble and sharp appreciation of the U.S. dollar, against which the tenge is managed. The current account balance turned negative in the second half of 2014, although there has been some improvement in the first quarter of 2015, in part reflecting the recent appreciation in the ruble and the rebound in oil prices. Based on a range of approaches used by staff, the tenge could be considered modestly overvalued by 4-14 percent (Annex II).1 Based on the IMF’s Assessing Reserve Adequacy (ARA) methodology, NBK’s gross reserves (which exclude the national oil fund) remain below the suggested adequacy range of 100-150 percent of the ARA composite metric. At the same time, the oil fund, with an accumulated value of around $75 billion (35 percent of GDP), provides a significant buffer for the economy as a whole.

8. Administrative and prudential measures have succeeded in lowering NPLs significantly, but vulnerabilities in some segments of the financial sector have increased (Box 2 and Annex III). Recent preliminary estimate indicate that NPLs have fallen from 34 percent of total loans a year ago to 10 percent in July 2015 (preliminary data) following the revocation of BTA’s banking license—after the bank absorbed all the bad assets of the merged KKB-BTA entity—and the removal of tax, accounting, and other legal obstacles to write-offs and transfers to special purpose vehicles (SPVs). NPL write-offs lowered the system-wide capital to risk-weighted assets (CAR) marginally, from 18 percent at end-2013 to 17 percent at end-2014, still well above the regulatory minimum of 12 percent. Nevertheless, some vulnerabilities have increased. In particular, deposit dollarization and declining lending activity are impacting banks’ profitability and increasing currency and maturity mismatches. Banks are particularly vulnerable to indirect foreign currency risk, due to the prevalence of lending to unhedged borrowers. Moreover, despite limits on credit concentration, lending mostly takes place in a few non-oil sectors that are currently under stress.

9. Progress has been made in implementing FSAP recommendations (Annex IV). The NBK has started to undertake bottom-up stress tests for banks. The authorities have also revised legislation to exclude the requirement for depositor and creditor approval in P&A and bridge bank resolution. They are initiating discussions with bank supervisors in other jurisdictions to strengthen cross-border supervision. The authorities have also made legislative amendments to implement recommendations in the insurance sector and on pensions, and drafted a law to address regulatory shortcomings in the securities market. Progress, however, has been limited in adopting risk-based assessment tools and extending supervision beyond compliance with prudential norms.

10. The authorities have embarked on an ambitious structural reform program, bolstered by extensive MDB engagement. Public investment under the Nurly Zhulu program aims to address infrastructure bottlenecks and generate jobs. Moreover, following the recent Presidential election, the authorities have launched a wide ranging reform program (“100 steps”) aimed at improving the efficiency of public administration and the rule of law, diversifying the economy, and addressing governance challenges. MDBs’ increased engagement strengthens the structural reform component of the authorities’ crisis response, with projects targeting the labor market, civil service and public expenditure reforms, business climate improvements, and support to small- and medium-sized enterprises (SMEs). In June, 2015, Kazakhstan completed negotiations to become a formal member of the WTO.

Outlook and Risks

11. The growth outlook has weakened, with predominantly downside external risks (Annex V). Staff projects real GDP growth to decelerate to 2 percent this year despite the fiscal stimulus. Weaker demand from Russia and China, lower oil prices, confidence effects, and continuing delays in the Kashagan oil field are the main factors behind the projected slowdown. Next year, growth is projected to pick up to 3.25 percent driven by a gradual recovery in oil prices and output, and improved external demand. Still, the medium-term growth outlook is less favorable than projected last year, in light of lower projected medium-term oil prices and continued slow growth in Russia. Staff now estimates potential growth at around 4.75 percent (vs. previous estimates of 5.5-6 percent). Against this backdrop, staff has also cut its headline inflation forecasts to 5.2 percent this year and 6 percent over the medium term (from 6 percent and 7 percent, respectively). The current account is expected to gradually improve in the coming years, due to a projected increase in oil exports once Kashagan comes on line, but remain in small deficit over the medium term. Downside risks to the growth outlook are mainly external and related to oil prices and regional uncertainty, although the recent improvements in the outlook for oil prices and the Russian economy lessen the tail risks. The authorities’ real GDP growth projection for 2015 is 1.5 percent. For the medium term, the authorities broadly shared staff’s outlook, but viewed the risks as more balanced, including because of the recent stability in oil prices and the Russian ruble.

12. Under an adverse scenario, with further decline in oil prices, anemic growth and larger imbalances could persist over the medium term. In a downside scenario, where oil prices are assumed to decline by 25 percent (permanently)2 relative to staff’s baseline projections, growth rates are expected to remain below potential in the medium term. While the deviation from baseline growth is not large, the adverse cumulative impact on the current account, FX reserves, fiscal balance, and public debt is more severe.

Staff’s baseline vs. alternative scenarios
201620172020
Growth
Baseline3.24.85.0
Downside2.74.34.6
Severe2.03.74.0
Current Account (percent of GDP)
Baseline−2.9−1.9−0.8
Downside−4.8−3.8−2.5
Severe−7.6−6.4−4.9
Fiscal Balance (percent of GDP)
Baseline−1.6−1.0−1.0
Downside−3.1−2.4−2.2
Severe−5.1−4.4−4.1
Gross Debt (percent of GDP)
Baseline20.522.930.7
Downside23.827.437.9
Severe28.533.848.9
FX Reserves ($ billion)
Baseline28.928.928.9
Downside26.625.423.8
Severe24.723.321.6
Oil Prices ($ per barrel)
Baseline65.769.774.0
Downside48.651.556.5
Severe26.729.634.6

Box 1.Credit Intermediation and Economic Growth in Kazakhstan

Credit market characteristics

Despite occasional bursts of growth, the credit market remains shallow. Financial intermediation has been falling since the 2007/2008 crisis. At 34 percent, Kazakhstan has a low credit to GDP ratio compared to peers.

Financial disintermediation?

The credit market is narrow with key sectors relying on other sources of financing. Energy and extractives rely primarily on foreign financing. Construction and real estate relied on wholesale foreign funding until it stopped in 2007. Consumer oriented borrowing (households, trade, services) dominates with over 70 percent of outstanding loans. Most businesses are cut-off from bank credit altogether. According to the latest World Bank survey less than 20 percent of firms have a loan or a line of credit.

Loans by sector January 1, 2014

(Source NBK, Staff estimates)

Loans by sector January 1, 2015

(Source NBK, Staff estimates)

Given its limited depth and scope, the domestic credit market has had a modest impact on economic activity. Analysis of quarterly credit data since 2004 reveals some correlation between credit growth and non-oil economic activity and no clear causality. The correlation is stronger in the run up to 2007/2008 crisis as the real estate bubble inflated. Given the limited scope of the credit market the main causality chain may run from oil prices through the economy to credit growth. (Annex I).

Recent developments

Credit growth slowed significantly over 2014 and so far in 2015. Overall bank credit to the private sector in May 2015 was lower by 3.5 percent in nominal terms relative to a year ago with a shift from foreign to national currency loans and from corporate to individual borrowers.

The slowdown in credit is exacerbated by high interest rates, tight domestic liquidity, dollarization, and, to a lesser extent, prudential policies on unsecured consumer lending. Weighted average interest rates on tenge loans to corporations jumped from 10 percent at the end of 2013 to 18 percent during 2015Q1. Broad money (M2) shrank by 8 percent in 2014 and a further 2 percent in 2015 mainly due to dollarization of deposits. Dollarization is increasing currency risks (Box 3), further deterring lending. Tight domestic liquidity, high interest rates and dollarization are byproducts of a tightly managed exchange rate in the context of high devaluation expectations. Prudential measures enacted in 2013 to limit the growth of unsecured consumer credit also contributed to the slowdown.

Debt is concentrated in construction and real estate

Probability, conditional on financial variables

The NPL overhang at 24 percent continues to restrict the banks’ ability to extend credit. Trade and services account for the highest share of NPLs (25 percent). The share of NPLs to total credit outstanding remains highest in construction, at almost 40 percent. NPL’s tie up client collateral and bank buffers which dampens credit activity.

The few businesses which have access to domestic credit are highly leveraged. Construction and real estate, which are still working though the fallout from the 2007/2008 crisis, have debt to equity ratios three times the average (chart). Real estate prices have remained stagnant.

Short-term prospects and policies

Staff expects credit to contract by 5.5 percent in nominal terms in 2015 but grow moderately, by 3.7 percent, in 2016. The forecast is broadly consistent with economic developments, taking into consideration the limiting factors described above. The IMF Macro-financial Forecast Consistency Template (Annex I) indicates that projected credit growth is broadly consistent with growth in other relevant variables.

Credit activity is likely to remain concentrated in trade and consumer lending. The prospect of a slowdown in real income growth could threaten these sectors and by extension private consumption.

Despite the weak link between credit and growth, anemic credit could trigger second round effects on economic activity and asset prices, further weakening banks’ balance sheets. Credit is essential for private sector led growth and diversification. The government’s program aimed at extending tenge 500 billion of subsidized loans to SME’s of which Tenge 250 billion have been disbursed so far seeks to mitigate the impact of tight credit on the economy. MDBs, including the ADB, are carrying out similar programs. It is too early to measure the impact of these programs. To reverse the fall in credit and strengthen the link between credit and growth it is necessary to address the key obstacle to growth: tight liquidity, high levels of NPL and lack of access to finance.

Box 2.Lower Oil Prices and Financial Sector Vulnerabilities

As an oil exporting economy, lower oil prices have a negative impact on Kazakhstan’s financial sector and on economic activity more generally. This box describes the solvency stress test that was performed on Kazakhstan’s banks to assess the impact on bank solvency of two shocks: lower oil prices and exchange rate depreciation. Stress tests on bank balance sheets as of end 2014 (latest available) reveal that the banking system is in a weak position to withstand shocks, particularly severe shocks. In the baseline, before any shocks are applied, three banks have capital adequacy ratios (CAR) below the regulatory minimum of 12 percent. The downside scenario would have the same 3 banks below the regulatory minimum. The severe scenario, however, would bring 17 banks (89 percent of banking system assets) below the regulatory minimum, and recapitalization needs would amount to 3 percent of GDP. The results are subject to the usual uncertainties associated with this type of tests.

The assumptions underlying the stress test are:

  • In the baseline scenario, estimates of oil prices reflect the WEO baseline, and real GDP growth and the exchange rate are based on staff’s macroeconomic projections. The projections assume further weakening in oil prices from the current $62.8pb to $58.9pb. Real GDP growth moderates from 4.3 percent in 2014, to 2 percent in 2015 and rises to 4.4 percent by 2019. The exchange rate depreciates from 179.2 (tenge/dollar) in 2014 to 185.8 in 2015 and to 188 by 2019 (i.e., within the current tenge/dollar exchange rate band).

  • In the downside scenario, a 0.5 historical standard deviation on the oil price is assumed. Under this scenario, oil prices fall to $43.2pb in 2015 and $48.6bp in 2016. Real GDP grows by 1.2 percent in 2015 and by 2.7 percent in 2016. The exchange rate is subjected to a 1.5 standard deviation shock, which implies a depreciation to 198 in 2015 and to 200.5 in 2016. The larger standard deviation shock on the exchange rate compared to oil prices mimics the large currency swings during periods of stress.

  • In the severe scenario, a 1.2 standard deviation from the baseline is assumed. Oil prices fall to $21.3pb in 2015 and $26.7pb in 2017. Real GDP grows by 0.1 percent in 2015 and by 2 percent in 2016. The exchange rate is subjected to a 3.5 standard deviation shock, which puts it at 15.5 percent depreciation from the current level in 2015. It depreciates to 215 in 2015 and to 217 in 2016.

The historical relationship between macroeconomic variables and NPLs is analyzed using a vector autoregression (VAR) model. The variables included in the VAR model are: the ratio of NPLs to total loans (NPL), the percent change in the oil price, Kazakhstan’s real GDP growth, partner countries’ real GDP growth, and the tenge/dollar exchange rate. Unit root and lag length tests were performed and adjustments made accordingly. The model was estimated on quarterly data from 2000 to 2014. The relationship shows that NPLs respond to shocks in macroeconomic variables. NPLs are most sensitive to changes in domestic activity, followed by the exchange rate and oil prices A percentage point increase in real GDP growth results in a 0.318 percentage point decline in NPLs; a percentage point increase in the oil price results in a 0.006 percentage point decline in NPLs; a percentage point increase (depreciation) in the exchange rate results in a 0.047 percentage point increase in NPLs. Single equation estimation shows similar but slightly larger coefficients on the exchange rate and oil prices.

The tests simulated balance sheets and profit/loss accounts for the individual banks given the NPL forecasts upon applying the effects of the shocks. In the simulation, liabilities are held constant. Given that historical balance sheet data are not available to staff, interest margins on current loans and liabilities, as well as net non-interest income are based on 2014 data. Given that estimates of collateral are unreliable, new NPLs are assumed to be provisioned for at 100 percent. This further dents profits. When net income is positive, the test assumes that 80 percent of profits are retained and the rest is paid out as dividends. A bank accumulates capital buffers when its capital ratio declined in the previous year. Capital covers the loss when net income is negative.

The results reveal that a number of banks are unable to withstand higher NPLs due the shocks described above. No adjustments were made on capital to take into account of the phase-out of capital instruments ineligible under Basel III as was done in the FSAP. Even without this adjustment, banks are still highly susceptible, especially in the face of severe shocks. The results are presented in the table below. The downside scenario brings system CAR down to 16 percent while the severe scenario brings it down to 6 percent, well below the regulatory minimum. Banks have low levels of profitability, thus, a low buffer against shocks. An exchange rate only shock of a 15.5 percent depreciation from the current level results in 10 banks being below regulatory CAR, while a 25 percent depreciation results in 17 banks below regulatory CAR. A reverse stress test reveals that an increase in NPLs by 16 percent brings the system wide CAR to the regulatory minimum of 12 percent (assuming a 70 percent level of provisioning). At this level, 6 banks would be below the regulatory minimum. NPLs decline over the projection period reflecting improvements in macroeconomic variable over time.

Stress Test Assumptions and Results
HistoricalBaselineDownsideSevere
2014201520162017201520162017201520162017
Assumptions
Domestic real GDP growth4.32.03.24.81.22.74.30.12.03.7
Oil price (in U.S. dollars)96.258.964.267.143.248.651.521.326.729.6
percent change over latest price (62.8)−31.212.45.9−66.025.110.8
Exchange rate179.2185.8188.0188.0198.3200.5200.5214.9217.1217.1
percent change over latest rate (186.1)6.61.10.015.51.00.0
Impact
Nonperforming loans (percent of total loans)32.213.05.81.521.711.53.537.624.69.9
Capital adequacy ratio17.417.024.523.116.124.830.46.015.526.0
CAR < 8 percent Number of banks11112111221
8 percent < CAR > 12 percent Number of banks2200100520
CAR < 12 percent Number of banks35353737353737213437
Recapitalization to achieve 12 percent CAR
Million Tenge97,914297,288510,326138,759295,659501,0761,233,496442,955497,869
Percent of 2014 GDP0.20.81.30.40.71.33.11.11.3

Policy Discussions

13. Policy discussions centered on the near-term policy response needed to revive the economy and the progress in strengthening policy frameworks. Regarding the short run, staff’s assessment is that (i) the fiscal stimulus is justified on countercyclical grounds, provided credible medium-term consolidation measures are identified; (ii) the tight monetary policy stance, which helped reduce pressure on the exchange rate, may need to be loosened in view of rapidly falling inflationary pressures; (iii) further steps are needed to lower NPLs and bolster financial sector resilience; and (iv) more rapid implementation of structural reforms is needed. As for the longer run, staff highlighted the imperative of further bolstering the macroeconomic policy architecture to help the economy absorb current and future shocks and to ensure durable growth and shared prosperity. The authorities agreed with the staff on the proposed short-term policy response and broadly shared the view on the need to enhance macroeconomic policy frameworks.

A. Fiscal Policy

14. The fiscal stimulus is tailored to support growth. The 3-5 year Economic Support Package introduced last year includes $12 billion (5.8 percent of GDP) spending from the National Fund of the Republic of Kazakhstan (NFRK) and $7 billion (2.9 percent of GDP) in loans contracted from Multinational Development Banks (MDBs) (see table). The package is aimed at modernizing critical infrastructure, promoting SME lending via quasi-fiscal development institutions, and recapitalizing the PLF. While the large stimulus come at the expense of higher current account deficit and lower reserves in the oil fund, staff supported the package, given its’ growth-enhancing, countercyclical, structural, and frontloaded nature. It also urged the authorities to ensure transparency and high quality spending and welcomed the involvement of MDBs in monitoring and selecting the investment projects.

15. The authorities agreed with staff that medium-term fiscal consolidation should be an essential component for the success of the fiscal stimulus. Given the size of the stimulus, a credible medium-term plan, especially on the revenue side, is needed to ensure fiscal sustainability. Assuming unchanged policies, total revenues are expected to decline from 23.8 percent of GDP in 2014 to 19.7 percent of GDP in 2020. The non-oil fiscal deficit is expected to rise sharply this year, to 11.4 percent of GDP (from 9.6 percent of GDP in 2014), and to decline only gradually to 8.7 percent of GDP in 2020, or around 3 percentage points of GDP higher than staff’s estimated sustainable level of 5.5 percent of GDP. To close the projected gap, the authorities have suggested improvements in tax administration and unspecified large cuts in expenditures (equivalent to 5 percentage points of GDP over three years) that are not based on concrete measures. Staff suggested that to maintain a reasonable and credible path for expenditures, and in view of relatively low tax revenue as a share of GDP, the authorities should aim at raising the revenue base through implementing measures such as (i) strengthening the enforcement of tax collection; (ii) reducing tax exemptions; including in the Special Economic Zones; and (iii) making income tax rates more progressive.

Stimulus Packages 2014–2019(In percentage of GDP)
201420152016201720182019
Financed from NFRK1.62.90.90.40.00.0
Financed from MDB0.00.00.90.90.80.4
Total1.62.91.91.30.80.4

16. To support fiscal transparency and ensure medium-term sustainability, staff emphasized the urgent need to further strengthen fiscal accounting and the fiscal policy framework. While important steps have been taken in improving fiscal coverage, more work is needed to bolster the quality of fiscal accounting (Annex VI). Concerns remain regarding the treatment of the oil fund transactions, some revenue and expenditure items, and the definition of the fiscal anchor. Staff underscored the need for bringing fiscal accounts into compliance with the GFS 2001 manual, particularly in the context of the fiscal consolidation strategy. The authorities agreed to consider staff’s recommendations and if necessary request TA in this area.

17. Public debt is expected to rise in the medium term, with increased international borrowing, but overall levels would remain relatively low. To help finance the fiscal deficit, the government has requested a $2 billion Development Policy Loan from the World Bank and $1 billion Countercyclical Support Facility from the Asian Development Bank. In addition to borrowing from MDBs and domestic debt issuance, the authorities noted their plans to tap international capital markets again this year when external market conditions allowed. The first such issuance took place on July 14. The global issuance also aims at helping build a yield curve in the absence of a well-functioning domestic debt market. According to staff’s calculations and assuming unchanged policies, the general government’s external debt is expected to rise from 3.8 percent of GDP at end-2014 to 6.4 percent of GDP by 2020. At the same time, total government debt is projected to increase from 14.5 percent in 2014 to 30.7 percent in 2020 (Annex VII).

B. Monetary and Exchange Rate Policy

18. Tight monetary policy has helped lower inflation, but the authorities concurred with staff that weak output and credit growth require caution about allowing market interest rates to remain too high for long periods. In view of rapidly declining inflationary pressures, subdued output growth, and weak lending conditions, the authorities are carefully evaluating the tradeoffs between the need for a tightening bias to contain the pressure on the currency and the goal of reviving lending and economic activity and stabilizing inflation around its objective range. Staff also urged efforts to avoiding periodic heightened volatility in money market interest rates.

19. The authorities have taken important steps to overhaul the monetary policy framework in support of inflation targeting. These steps include strengthening communication, preparing the grounds for the imminent introduction of new policy interest rate instruments, and enhancing governance. In particular, the recently published monetary policy guidelines for 2015 and 2020 stress the planned introduction of a new policy interest rate, supported by open market operations, during the early phase of the transition to inflation targeting, which the authorities see adopting within 3-5 years. The NBK has also recognized the need to strengthen its governance structure, including by setting up a monetary policy committee (MPC) and achieving greater central bank independence in policy decision making. Moreover, the NBK plans to establish a money market committee responsible for the implementation of monetary policy, including the daily assessment of banking sector liquidity conditions and interbank market developments. Staff noted that modernizing the monetary policy framework will also help the recently announced plans to develop Astana as an international financial center.

20. Staff has stressed the need to gradually introduce greater exchange rate flexibility in tandem with improving monetary policy instruments, while avoiding a step-devaluation. Staff emphasized that greater exchange rate flexibility would (i) support a more independent interest rate policy; and (ii) help reduce imbalances. Staff supported the NBK’s opposition to another outright step-devaluation, given that previous episodes resulted in increased dollarization, re-fixing the exchange rate, and further expectation of devaluation. A large devaluation would also put immediate stress on private-sector balance sheets of unhedged FX borrowers (Box 3). In contrast, a gradual adjustment would give banks more time to address balance-sheet vulnerabilities, while correcting the exchange rate misalignment over time. In view of this, staff considered that greater flexibility could initially be introduced by gradually widening the exchange rate band, with the aim of removing it altogether when conditions allow.

21. The authorities agreed that greater flexibility would support the reform efforts underway and alleviate imbalances. In light of this, and given reduced pressures on the tenge in recent months, on July 15 the NBK widened its exchange rate band from 170-188 to 170-198 tenge/dollar. To ensure stability, the authorities also agreed to communicate their plans consistently and credibly and continue to strengthen monetary policy instruments.

C. Financial Policies

22. Staff and the authorities broadly concurred on the need to bolster financial sector resilience. Staff noted the progress made in halting the surge in uncollateralized consumer loans seen in recent years. However, three banks have capital below the regulatory minimum and banks’ balance sheets appear to be highly sensitive to shocks. Supervision should be strengthened and corrective actions be taken for the undercapitalized banks to bring their capital above the regulatory minimum. Furthermore, staff urged the authorities to shore up system strength by developing plans to bring capital up to shock resilient levels and improve governance and risk management to help reduce risks. There was agreement that exchange rate risk remains a concern for the banking system, especially given corporate sector exposure. Further, credit concentration remains high. Staff recommended introducing higher risk weights or exposure caps on corporate lending; reducing (or prohibiting) FX lending to unhedged borrowers; and tightening net open position (NOP) limits to mitigate exchange rate risk (Box 3 and Annex VIII). The authorities are considering raising risk weights on loans denominated in foreign currency to help reduce banks’ exposure to indirect foreign currency risk. However, they consider the current limit on the NOP to be sufficient in minimizing on-balance sheet currency mismatches and do not see the need to limit or prohibit lending to unhedged borrowers as this, in their view, will curtail lending. Staff shared the concern about the need to avoid undermining an already weak lending environment but stressed the consequences of excessive lending to unhedged borrowers. Staff also urged the authorities to enhance the quality of data on corporate and household balance sheets to allow for a more informed assessment of balance-sheet vulnerabilities. The authorities agreed to further assess vulnerabilities and argued that they are monitoring bank soundness.

23. There was agreement that further reduction of NPLs, especially of the newly merged KKB-BTA entity, is a necessary condition to improving banking sector soundness. Staff welcomed the efforts underway to reduce the level of system NPLs, including the efforts to achieve the 10 percent ceiling by end-2015. But it expressed concern that NPL provisioning may be overstated because it relies on underlying loan collateral that is overstated—mainly real estate that has not yet recovered from the 2009 bubble bust—and is not regularly revalued. Staff also welcomed the refocus of NPL resolution at the merged KKB-BTA entity, which accounts for 70 percent of the universe of NPLs and about 25 percent of banking system assets. In this regard, staff advised that the authorities develop a broad strategy to address the factors that resulted in weaknesses in these banks, including conducting an asset quality review (AQR) of the merged bank. The strategy should address the restructuring of the bank to ensure long-term viability and how any capital needs identified by the AQR will be addressed. Regulatory and supervisory standards should also be improved, and consideration should be given to sequencing of reforms with the AQR, and the specific standards against which the banks will be assessed.

24. Staff welcomed progress made in implementing FSAP recommendations. It urged for further progress in adopting risk-based assessment tools and supervision, which remains oriented toward compliance. The NBK prefers to maintain the current supervisory approach for the time being. The authorities plan to continue making progress toward finalizing legislative amendments in insurance, pensions, and the securities market.

Box 3.Macroprudential Policies in Kazakhstan

Reducing exchange rate risk: Exchange rate risk remains one of the largest risks for the banking sector. FX risks are especially elevated in the corporate sector—the largest borrowers from the banking system—where leverage and the net open FX position to equity are very high. This sector has also experienced a significant slowdown in activity and is associated with already high NPLs to the banking system. To help limit risks emanating from high exposure to the corporate sector, the NBK should consider introducing risk weights or exposure caps for corporate exposures. This will help the banking system withstand corporate credit losses by building capital buffers (see IMF Staff Guidance on Macroprudential Policy). Risk weights can help build buffers by affecting the supply of credit indirectly while FX exposure caps curtail the supply of credit directly. Reducing FX lending to borrowers whose earnings are not in FX should also be considered. This can be done through higher risk weights on such exposures. Tighter net open position limits are recommended to help reduce exchange rate risk.

Reducing credit risk: Even with concentration limits, credit concentration remains a significant risk to financial stability. Though concentrated large exposures have fallen from levels during the global financial crisis, at 208 percent of capital, they remain large. In 2014, the construction sector received the largest share of new corporate sector credit (24 percent) while it also had the largest share of new NPLs (41 percent, rising from 15 percent the previous year). The NBK should restrict bank exposures to single counterparties or groups of connected counterparties.

Proper implementation and enforcement of macroprudential tools is critical. Inconsistent implementation of macroprudential measures, as has been in the past, could add to regulatory uncertainty and result in ineffective incentive structures and a loss in confidence in the domestic financial system.

D. Structural Reforms

25. The authorities recognized that the recent shocks to the economy reinforce the urgency of implementing structural reforms, to ensure durable long-term growth and shared prosperity. Diversification away from the extractive industry and the reduction of the state footprint in the economy are necessary to sustain improvements in public welfare and to help create jobs for the growing population. Addressing infrastructure bottlenecks, strengthening human capital, building institutions, bolstering the rule of law, enhancing financial intermediation, and improving business climate have been identified by both the authorities and development partners as key areas where further improvements are essential to achieve broad and inclusive growth. The authorities’ approach of working with MDB’s to implement public spending and structural reforms should ensure greater efficiency and focus on priority areas. Staff also noted that additional efforts were needed to reverse the slowdown in key social indicators and protect the most vulnerable from the impact of the economic slowdown. Despite the authorities’ declared commitment to ring fence social spending, budget cuts aimed at creating space for capital spending are adversely affecting key social programs (Annex IX).

26. The public investment and reform program offers an opportunity to balance the short-term need for countercyclical state intervention and the long term goal of private sector-led growth. Staff stressed that public investment should focus on areas beyond the reach of the private sector without expanding the state’s footprint in the economy. Moreover, staff noted that reforms aimed at improving the business climate should include easing liquidity constraints and improving access to finance. The authorities’ focus on large scale infrastructure projects and the SME subsidized loan program helps in this regard.

Staff Appraisal

27. Amid slower economic growth, tighter financial conditions, and emerging imbalances, policies should balance ensuring sustainability while alleviating the impact of shocks in the near term. In recent years the authorities have successfully harnessed oil resources to bolster economic growth and build buffers. However, economic growth has now decelerated largely as a result of likely long-lasting external shocks and is expected to remain subdued this year and next year. Moreover, the outlook for growth is subject to predominantly downside external risks. In view of weaker growth, the large and likely long-lasting nature of the shocks, and the accumulated buffers, the policy response in the short term should be geared toward supporting the economic recovery. Over the medium term, there is a need to further strengthen macroeconomic policy frameworks, to bolster resilience to shocks and promote durable growth.

28. To ensure fiscal sustainability, the stimulus must be accompanied by credible medium-term fiscal consolidation and more transparent fiscal policy framework. The stimulus is justified on countercyclical grounds and is appropriately frontloaded and tailored to support growth. However, to ensure a sustainable path for the non-oil deficit, the stimulus must be accompanied by credible medium-term consolidation measures, especially on the revenue side. In particular, there is scope to strengthen the enforcement of tax collection, reduce tax exemptions, including in the Special Economic Zones, and make income tax rates more progressive. Moreover, enhancing the fiscal policy framework is critical to ensuring transparency and medium-term sustainability. Key priorities include expanding the budget coverage to all fiscal activity, in line with GFSM 2001, and integrating fiscal policy into a broader macroeconomic policy framework.

29. Greater exchange rate flexibility in tandem with the introduction of new monetary policy instruments is needed to enhance the policy architecture and address imbalances. The authorities have taken confidence-building measures to overhaul the monetary policy framework in support of their medium-term goal of adopting inflation targeting, and have widened the exchange rate band. However, with the aim of more effectively managing liquidity and signaling the stance of policy, the authorities should speed up the planned introduction of a new policy interest rate, supported by open market operations. Moreover, strengthening the policy architecture requires further exchange rate flexibility, which will support a more independent interest rate policy and help reduce imbalances. To avoid undermining financial stability, and anchor expectations about policy intentions and operations, the authorities should communicate their plans openly and consistently.

30. In view of rising vulnerabilities, further actions are needed to bolster financial sector resilience. Efforts underway to reduce the level of NPLs are paying off. However, more needs to be done to bring down the level of NPLs to sustainable levels, while ensuring that the achievement of the 10 percent prudential ceiling by end-2015 does not compromise proper loan classification and provisioning. In this regard, while the recent plans to refocus the NPL resolution framework at the merged KKB-BTA entity is appropriate, weaknesses in the bank should be addressed and an asset quality review be undertaken within a broad strategy to ensure long-term viability. Strengthening financial sector resilience also requires introducing higher risk weights or exposure caps on corporate lending, limiting FX lending to unhedged borrowers, and tightening net open position limits to mitigate credit risk. Further steps in implementing the FSAP recommendations include adopting risk-based assessment tools and supervision and finalizing legislative amendments in insurance, pensions, and the securities market.

31. The structural reform agenda is appropriately ambitious, but effective implementation is essential to achieving sustainable and inclusive growth. Priority areas include strengthening human capital, building institutions, bolstering the rule of law, enhancing financial intermediation, and improving the business climate. Close collaboration with MDBs in these areas should facilitate greater efficiency in the procurement and implementation process. Moreover, diversification away from the oil sector and the reduction of the state footprint in the economy are necessary conditions to ensuring successful implementation of the broader private-sector-led growth strategy.

32. It is proposed that the next Article IV consultation take place on the standard 12-month cycle.

Figure 1.Kazakhstan: Economic Developments

Sources: Kazakhstani authorities and IMF staff estimates.

Figure 2.Kazakhstan: Banking Sector Developments1/

Sources: Kazakhstani authorities, GFSR, and IMF staff estimates.

1/Authorities did not provide systemwide CAR data during BTA’s second debt restructuring, hence the gap in the CAR graph.

2/Accounting for exchange rate valuation effects.

Figure 3.Kazakhstan: Key Financial Soundness Indicators, Cross-Country Comparison

Sources: IMF FSI database and World Economic Outlook.

Figure 4.Kazakhstan: Capital Markets and Expected Default

Source: Bloomberg.

Figure 5.Kazakhstan: Monetary and External Sector Developments

Sources: Kazakhstani authorities, Bloomberg, and IMF staff estimates.

Figure 6.Kazakhstan: Fiscal Sector Developments and Outlook

Sources: Kazakhstani authorities and IMF staff estimates.

Figure 7.Kazakhstan: Business Environment and Governance Indicators

Table 1.Kazakhstan: Selected Economic Indicators, 2012–20
201220132014201520162017201820192020
Projections
(Annual percent change, unless otherwise indicated)
National accounts and prices
Real GDP 1/5.06.04.32.03.24.84.64.45.0
Real oil−2.23.2−1.3−0.40.46.44.63.35.5
Real non-oil8.07.06.32.84.14.34.64.74.8
Real consumption9.99.81.50.57.15.35.24.44.7
Real investment10.85.41.78.7−1.52.02.84.82.3
Real exports3.4−0.91.82.32.35.94.74.17.3
Real imports19.45.0−6.75.85.94.44.34.54.5
Output gap (in percent of potental GDP)0.11.00.7−1.6−2.6−1.9−1.2−0.60.7
Crude oil and gas condensate production (million tons)798281818186909398
Consumer price index (p.a.)5.15.86.75.25.55.45.76.06.0
Core consumer price index (p.a.)5.64.36.74.25.05.45.75.75.7
GDP deflator4.89.77.4−2.67.06.06.16.05.5
Exchange rate (tenge per U.S. dollar; eop)1.52.218.73.10.00.00.00.00.0
(In percent of GDP, unless otherwise indicated)
General government fiscal accounts
Revenues and grants26.925.323.819.820.820.820.520.019.7
Of which: Oil revenues13.411.811.38.18.88.78.48.07.6
Expenditures and net lending22.420.322.123.122.321.921.321.120.7
Overall fiscal balance4.55.01.7−3.2−1.6−1.1−0.8−1.0−1.0
Financing−4.1−3.31.93.21.61.10.81.01.0
Domestic financing2.72.11.52.62.53.43.94.65.0
Foreign financing0.20.21.60.91.71.11.00.60.2
NFRK−7.0−5.5−1.3−0.3−2.6−3.4−4.1−4.1−4.2
Gross public debt (percent of GDP)12.412.914.518.120.522.925.528.230.7
Non-oil fiscal balance (percent of GDP)−8.9−6.8−9.6−11.4−10.4−9.8−9.3−9.0−8.7
Non-oil fiscal balance (percent of non-oil GDP)−13.0−9.4−13.4−13.8−12.6−11.9−11.3−10.8−10.3
(Annual percent change, eop, unless otherwise indicated)
Monetary accounts
Reserve money1.9−2.220.86.97.58.38.38.38.3
Broad money7.910.22.414.510.411.111.111.111.1
Broad money velocity (annual average)2.93.03.32.92.92.92.92.92.9
Credit to the private sector 2/11.612.75.7−5.53.76.011.010.610.8
Credit to the private sector (percent of GDP) 2/41.540.237.937.737.537.537.537.537.5
NBK refinance rate (eop; percent)5.55.55.5
(In billions of U.S. dollars, unless otherwise indicated)
External accounts
Current account balance (percent of GDP)0.50.42.1−3.3−2.9−1.9−1.2−1.1−0.8
Exports of goods and services91.890.785.460.064.670.676.180.186.4
Oil and gas condensate56.457.253.632.735.839.743.144.948.1
Imports of goods and services61.563.056.252.255.758.661.564.567.5
Foreign direct investments (net, percent of GDP)−5.8−3.4−2.7−2.5−2.3−2.1−2.0−1.8−1.7
NBK gross reserves (eop) 3/28.324.728.928.928.928.928.928.928.9
In months of next year’s imports of goods and services5.45.36.76.25.95.65.45.45.1
NFRK assets (eop)57.970.873.674.280.288.9100.6113.6128.2
Total external debt 4/136.9149.9157.1168.9181.3193.1203.7214.3224.2
In percent of GDP67.364.771.279.878.675.371.668.164.3
Excluding intracompany debt (percent of GDP)34.132.835.341.342.341.840.639.337.7
Memorandum items:
Nominal GDP (in billions of tenge)30,34735,27539,53039,28543,36548,19453,50259,15665,522
Nominal GDP (in billions of U.S. dollars)203.5231.9220.6211.7230.7256.3284.6314.7348.5
Saving-Investment balance (percent of GDP)0.50.42.1−4.5−3.4−2.1−1.5−1.3−0.7
Crude oil, gas cnds, production1.651.701.681.681.681.791.871.942.04
(millions of barrels/day) 5/
Oil price (in U.S. dollars per barrel)105.0104.196.258.964.267.169.971.071.5
Sources: Kazakhstani authorities and Fund staff estimates and projections.

The base year for real GDP calculations has been changed from 1994 in previous Fund documents to 2007.

Private sector includes nonbank financial institutions, public and private nonfinancial institutions, nonprofit institutions, and households.

Does not include NFRK.

Gross debt, including arrears and other short-term debt.

Based on a conversion factor of 7.6 barrels of oil per ton.

Sources: Kazakhstani authorities and Fund staff estimates and projections.

The base year for real GDP calculations has been changed from 1994 in previous Fund documents to 2007.

Private sector includes nonbank financial institutions, public and private nonfinancial institutions, nonprofit institutions, and households.

Does not include NFRK.

Gross debt, including arrears and other short-term debt.

Based on a conversion factor of 7.6 barrels of oil per ton.

Table 2.Kazakhstan: Balance of Payments, 2012–20(In billions of U.S. dollars, unless otherwise indicated)
201220132014201520162017201820192020
Projections
Current account1.10.94.6−7.0−6.8−4.8−3.5−3.3−2.8
Trade balance38.134.835.614.015.518.921.622.926.2
Exports (f.o.b.)86.985.679.154.658.964.569.573.178.8
Oil and gas condensate56.457.253.632.735.839.743.144.948.1
Non-oil exports30.528.325.421.923.124.826.428.230.6
Imports (f.o.b.)48.850.843.440.643.345.647.850.252.5
Oil and gas condensate2.32.80.21.31.51.51.61.71.8
Non-oil, gas imports46.548.043.239.241.844.046.248.550.8
Services and income balance−36.0−32.2−29.3−19.4−20.5−21.8−23.0−23.9−26.4
Services, net−7.9−7.1−6.4−6.1−6.6−6.8−7.1−7.3−7.4
Income, net−28.1−25.1−22.9−13.2−13.9−14.9−16.0−16.6−19.0
Of which: Income to direct investors24.822.519.910.811.512.413.414.016.4
Current transfers−1.0−1.6−1.7−1.6−1.8−1.9−2.1−2.3−2.6
Capital and financial account4.3−0.3−7.3−7.0−6.8−4.8−3.5−3.3−2.8
Foreign direct investment−11.9−7.9−5.9−5.4−5.3−5.4−5.6−5.7−5.9
Portfolio investment, net17.56.10.8−0.40.30.60.91.01.1
Of which: National Fund14.77.96.5−2.72.65.17.68.59.5
Other investment−1.31.5−2.2−1.3−1.70.01.21.42.0
Errors and omissions−1.1−3.6−8.10.00.00.00.00.00.0
Overall balance4.32.4−3.90.00.00.00.00.00.0
Financing: Net international reserves of NBK−4.3−2.43.90.00.00.00.00.00.0
Memorandum items: 1/
(In percent of GDP)
Current account0.50.42.1−3.3−2.9−1.9−1.2−1.1−0.8
Exports of goods42.736.935.825.825.525.124.423.222.6
Oil exports27.724.724.315.415.515.515.114.313.8
Non-oil exports15.012.211.510.410.09.79.38.98.8
Imports of goods24.021.919.719.218.817.816.815.915.1
(Annual growth rate, in percent)
Exports2.0−1.5−7.6−31.07.89.57.85.27.8
Non-oil exports1.6−7.0−10.3−13.85.37.46.66.68.8
Volume on non-oil exports14.8−4.0−3.02.26.16.06.16.24.3
Average price of non-oil exports−11.5−3.2−7.6−15.6−0.81.30.50.40.4
Imports20.94.1−14.5−6.66.85.25.04.94.7
Oil and gas imports−9.725.7−93.4606.512.03.14.73.35.6
Non-oil imports22.93.1−9.8−9.26.65.25.04.94.7
Volume on non-oil imports32.85.4−6.13.04.34.34.54.54.5
Average price of non-oil imports−7.7−2.2−4.1−9.30.20.50.30.30.2
Exports of oil and gas condensate (in MT)68.170.768.267.968.272.475.477.482.4
NBK gross international reserves (in billions of U.S. dollars)28.324.728.928.928.928.928.928.928.9
In months of next year’s imports of g&n.f.s. 2/5.45.36.76.25.95.65.45.45.1
National Fund (including interest), e.o.p.57.970.873.674.280.288.9100.6113.6128.2
External debt in percent of GDP67.364.771.279.878.675.371.668.164.3
Public external debt (percent GDP)2.72.43.74.75.15.35.35.25.1
Private external debt (percent GDP)64.662.267.575.173.570.066.362.859.2
Excluding intra-company loans (percent GDP)34.132.835.341.342.341.840.639.337.7
World oil price (U.S. dollars per barrel)105.0104.196.258.964.267.169.971.071.5
Sources: Kazakhstani authorities and Fund staff estimates and projections.

Estimates and projections are based on GDP at market exchange rates.

The number reflects months of same year’s imports of g&n.f.s. in 2019.

Sources: Kazakhstani authorities and Fund staff estimates and projections.

Estimates and projections are based on GDP at market exchange rates.

The number reflects months of same year’s imports of g&n.f.s. in 2019.

Table 3.Kazakhstan: Financial Soundness Indicators of the Banking Sector, 2008–141/
2008200920102011201220132014
Capital adequacy(In billions of tenge)
Regulatory capital1948−91418211956207223462629
Of which: Tier 11525−105414191492156116912063
Common shares and retained earnings974−150636212348237528472969
Preferred shares5487248250213212214
Hybrid instruments103674443383946
Of which: Tier 2495312451515510661641
Of which: Tier 31122122
Risk Weighted assets13106113271031211331115341269815226
Of which: Credit risk125288824943910470109051201914397
Of which: Market risk2432028296282235296318
Of which: Operational risk336475578579394384512
(In percent)
Regulatory capital as percent of risk-weighted assets *15−81817181817
Regulatory Tier I capital to risk-weighted assets *12−91413141314
Capital as percent of assets *12−81110141313
Asset quality
Nonperforming loans (NPL) as percent of gross loans *5222431303124
Provisions as percent of NPL69756370778380
NPL net of provisions as percent of tier I capital *10−485665524132
Large exposures as percent of tier I capital *166−279216252264256243
(10-largest credit to net credits) 25-largest credit to total loans27313436353235
(In billions of tenge)
Written off loans**43821621532353082,230
Earnings and profitability(In percent)
Gross profits as percent of average assets (ROAA) *0−24120222
Gross profits as percent of average equity capital (ROAE) *1−1198843−3131313
Net interest margin (net interest income as percent of interest bearing assets) *6434455
Gross income as percent of average assets741651388
Net interest income as percent of gross income771051671276057
Non-interest income as percent of gross income23−58429734043
Trading net income as a percent of gross income *3−430−2−342
Non-interest expenses as percent of gross income *35652168683839
Personnel expenses as percent of non-interest expenses *32342631123737
Liquidity
Liquid assets as percent of total assets *14192121181718
Liquid assets as percent of short-term liabilities *49536458474853
Foreign currency loans as percent of total loans *52575044383837
Foreign currency liabilities as percent of total liabilities *60544440353851
Deposits as percent of assets (excluding interbank deposits)39535761616462
Of which: Household deposis13171922252624
Of which: Corporate deposits26353839373838
Of which: Interbank deposits3221122
Of which: Other191611111
Other wholesale/market-based funding as percent of assets26382927222021
Loans as percent of deposits *198156131132135134124
Sensitivity to market risk
Off-balance sheet operations as percent of assets82796376767486
Gross asset position in derivatives as a percentage of tier I capital *12−832568
Gross liability position in derivatives as a percentage of tier I capital *9−754567
Net (long) open position in foreign exchange as a percentage of tier I capital *6162−3−1−5−3−2
Source: NBK.

Data include BTA unless otherwise indicated.

Core and encouraged set of indicators. Indicates available aggregate data for all banks.

Excluding BTA and credit cards.

Source: NBK.

Data include BTA unless otherwise indicated.

Core and encouraged set of indicators. Indicates available aggregate data for all banks.

Excluding BTA and credit cards.

Table 4.Kazakhstan: Monetary Accounts, 2010–14
20102011201220132014
(End of period, in billions of tenge)
Monetary Survey
Net Foreign Assets4,4825,2615,7886,4267,408
Net Domestic Assets4,1254,9645,2585,9226,578
Domestic Credit10,03711,59313,08114,78915,578
Net claims on central government158302482594579
Net claims on other government78775
Credit to the private sector 1/9,87211,28312,59214,18814,993
Claims on households2,2432,4833,0263,8074,201
Other items, net−5,912−6,629−7,823−8,868−9,000
Broad money8,4839,75210,52211,59811,882
Currency in circulation1,1481,3661,5281,5121,398
Total deposits7,3348,3878,99410,08610,484
Nonliquid liabilities116148169208203
Statistical discrepancy1234736071,1631,098
Accounts of NBK
Net foreign assets 2/4,0544,2694,1713,7425,199
Net international reserves 2/4,0874,2694,1823,7245,100
Net domestic assets 2/−1,320−949−572−140−525
Net domestic credit−527−79457671871
Net claims on central government−313−382−660−733−1,081
Net claims on other government00000
Net claims on the private sector 1/132147171179292
Net claims on banks−418−64403553701
Other items, net−793−870−1,029−810−1,396
Reserve money2,5722,8372,8902,8263,414
Currency in circulation1,3061,5481,7371,7631,382
Liabilities to banks7227287248441,500
Required reserves2926316658041,399
Other liabilities429975939101
Demand deposits544561429219531
Other deposits00000
Other liquid liabilities3891022829
Deposit money banks
Net foreign assets4289921,6172,6832,209
Net domestic assets7,4007,8618,4799,01711,278
Domestic credit11,68912,83814,04615,76917,416
Net claims on central government399463600656701
Net claims on other government78775
Credit to the private sector 1/9,74011,13612,42014,00914,758
Net claims on other financial corporations346316377560514
Banks’ reserves1,5421,2311,0181,0971,951
Other items, net−4,289−4,977−5,567−6,752−6,138
Banks’ liabilities7,8288,85410,09611,70013,487
Demand deposits2,0682,7622,6002,6362,971
Other deposits4,6845,0555,8647,2038,163
Other liabilities1,0761,0371,6321,8612,353
Memorandum items:
Reserve money (percent change, y-o-y)5.010.31.9−2.220.8
Broad money (percent change, y-o-y)13.315.07.910.22.4
Credit to private sector (percent change, y-o-y)5.614.311.612.75.7
Velocity of broad money2.62.82.93.03.3
Money multiplier3.33.43.64.13.5
Foreign currency deposits (in percent of total deposits) 3/3733303758
Foreign currency loans (in percent of total loans) 3/3732282827
Sources: Kazakhstani authorities and Fund staff estimates.

Private sector includes nonbank financial institutions, public and private nonfinancial institutons, non-profit institutions, and households.

Does not include oil fund resources.

Commercial banks only.

Sources: Kazakhstani authorities and Fund staff estimates.

Private sector includes nonbank financial institutions, public and private nonfinancial institutons, non-profit institutions, and households.

Does not include oil fund resources.

Commercial banks only.

Table 5.Kazakhstan: General Government Fiscal Operations, 2012–20
201220132014201520162017201820192020
Projections
(In billions of tenge)
Total revenue and grants8,1708,9119,4207,7979,00910,01610,96611,85112,910
Tax revenue7,5078,1768,5836,9178,0929,0239,86810,62111,531
Oil4,0764,1504,4863,2013,8134,1884,5134,7115,001
Non-oil3,4314,0254,0973,7154,2804,8355,3555,9096,530
Income from capital transactions393851515151515151
Nontax revenue6246987868298659421,0461,1791,328
Grants000000000
Total expenditure and net lending6,8107,1518,7339,0569,68610,53111,41312,46113,588
Total expenditure6,6767,0678,3678,6639,53310,42111,29012,32513,437
Current expenditure5,2915,6686,7016,2246,9897,8098,6599,64410,676
Capital expenditure1,3851,3991,6662,4392,5452,6122,6312,6812,761
Net lending 1/13483366393153111123136151
Overall balance1,3601,761687−1,259−677−515−447−610−678
Non-oil balance−2,716−2,390−3,799−4,460−4,490−4,704−4,960−5,322−5,679
Statistical discrepancy1275951,427000000
Financing−1,234−1,1667401,259677515447610678
Domestic financing, net8177185631,0131,0501,6092,0752,6883,254
Foreign financing, net6854651345724536536348160
Privatization receipts142021212121212121
NFRK 2/−2,133−1,957−496−120−1,118−1,651−2,185−2,447−2,757
(In percent of GDP)
Total revenue and grants26.925.323.819.820.820.820.520.019.7
Tax revenue24.723.221.717.618.718.718.418.017.6
Oil13.411.811.38.18.88.78.48.07.6
Non-oil11.311.410.49.59.910.010.010.010.0
Income from capital transactions0.10.10.10.10.10.10.10.10.1
Nontax revenue2.12.02.02.12.02.02.02.02.0
Grants0.00.00.00.00.00.00.00.00.0
Total expenditure and net lending22.420.322.123.122.321.921.321.120.7
Total expenditure22.020.021.222.122.021.621.120.820.5
Current expenditure17.416.117.015.816.116.216.216.316.3
Capital expenditure4.64.04.26.25.95.44.94.54.2
Net lending 1/0.40.20.91.00.40.20.20.20.2
Overall balance4.55.01.7−3.2−1.6−1.1−0.8−1.0−1.0
Non-oil balance−8.9−6.8−9.6−11.4−10.4−9.8−9.3−9.0−8.7
Financing−4.1−3.31.93.21.61.10.81.01.0
Domestic financing, net2.72.01.42.62.43.33.94.55.0
Foreign financing, net0.20.21.60.91.71.11.00.60.2
Privatization receipts0.00.10.10.10.00.00.00.00.0
NFRK 2/−7.0−5.5−1.3−0.3−2.6−3.4−4.1−4.1−4.2
Memorandum items:
Non-oil balance (in percent of non-oil GDP)−13.0−9.4−13.4−13.8−12.6−11.9−11.3−10.8−10.3
Non-oil revenues (in percent of non-oil GDP)19.618.717.414.314.614.814.714.514.4
Gross public debt (percent of GDP) 3/12.412.914.518.120.522.925.528.230.7
NFRK assets (in billions U.S. dollars) 1/57.970.873.674.280.288.9100.6113.6128.2
Nominal GDP (in billions KZT)30,34735,27539,53039,28543,36548,19453,50259,15665,522
Sources: Kazakhstani authorities and Fund staff estimates and projections.

Includes staff’s estimates of the announced lending package from the NFKR for the period 2014-16.

National Fund of the Republic of Kazakhstan. (-) is accumulator in the Fund.

General government, including republican and local budgets.

Sources: Kazakhstani authorities and Fund staff estimates and projections.

Includes staff’s estimates of the announced lending package from the NFKR for the period 2014-16.

National Fund of the Republic of Kazakhstan. (-) is accumulator in the Fund.

General government, including republican and local budgets.

Annex I. Credit Intermediation and Economic Growth in Kazakhstan1

1. Examination of quarterly credit and GDP data for the past ten years shows co-movement of credit and non-oil GDP growth. Within that dynamic, nominal credit growth outstripped nonoil GDP growth up to the global financial crisis and lagged behind ever since. As a result, financial deepening turned into disintermediation from 2007.

Credit and Growth

2. However, a causal link from credit to non-oil GDP is difficult to discern. Unrestricted VAR of log credit and log non-oil GDP (quarterly data) with a 4 period lag reveals no causality in either direction. The addition of key exogenous, variables such as the oil price and a dummy for the financial crisis does not improve the outcome. The results are consistent with available information. The credit boom in the run up to the financial crisis was driven by wholesale funds from abroad feeding a real estate bubble. Dynamics in the credit market since 2008 have been dominated by the legacy of the crisis especially the NPL overhang.

3. Credit is found to have a marginal impact on non-oil GDP growth. To examine the relationship between credit and growth we fitted a regression with the following specification:

LNGDP_XO=C+αLCREIDT_N + βNGDPXO(-4)+γDUM+u

Where LNGDP_XO and LNGDP_XO(-4) are the logs of nominal non-oil GDP for t and t-4. LCREDIT_N is the log of nominal credit outstanding and DUM is a dummy for the 2008 financial crisis. The output below shows that a 1 percent change in nominal credit affects nominal non-oil GDP on average by 0.12 percent.

Dependent Variable: LNGDPXO
Method: Least Squares
Date: 06/29/15 Time: 11:15
Sample (adjusted): 2004Q1 2014Q4
Included observations: 44 after adjustments
VariableCoefficientStd. Errort-StatisticProb.
C0.1714710.5976230.2869220.7757
LCREDIT_N0.1239900.0526642.3543460.0236
LNGDPXO(-4)0.8730920.04473819.515810.0000
DUM−0.0740180.061429−1.2049330.2353
R-squared0.983637Mean dependent var14.87075
Adjusted R-squared0.982410S.D. dependent var0.632069
S.E. of regression0.083829Akaike info criterion−2.033565
Sum squared resid0.281093Schwarz critericn−1.871366
Log likelihood48.73843Hannan-Quinn criter.−1.973414
F-statistic801.5339Durbin-Watson stat1.553109
Prob(F-statistic)0.000000

4. The weak link between credit and growth can be explained by the shallowness and compartmentalization of the credit sector. The chart below depicts the various segments of the credit sector and the dynamics affecting them.

Credit Is Compartmentalized Into Distinct Markets with Different Sources of Funding
Sectors/Categories of BorrowersSources of FundsCurrent Dynamics
Energy and extractives, subsidiaries of foreign companies and state owned/quasi-sovereign entitiesWholesale foreign borrowing including syndicated loans and bonds. FDI in the form of intracompany debt. Limited domestic borrowing, mainly short termSlowdown due to fall in oil and other commodity prices, lower demand from China and to some extent public spending cuts.
Construction and real estate, mainly private sector companiesWholesale funding from abroad up to 2007, limited domestic borrowing sinceYet to recover from the 2007/2008 crash caused by the sudden stop of external financing. Depressed real estate values, NPL overhang and lack of long term domestic financing continue to weigh on the sector
Manufacturing, agricultures, trade, and services, mainly private sector SME’sPrimarily, own resources, some domestic borrowing, including subsidized state and MDB supported loansPotential for growth hampered by high borrowing costs and shortage of long term domestic financing. Most businesses have no access to finance
Private consumptionDomestic uncollateralized borrowingRobust demand but risk from slowdown in real income growth
Annex II. Assessment of Kazakhstan’s External Position1

Kazakhstan’s external position has weakened over the past year. This assessment is based on (i) growing estimated exchange rate misalignment; (ii) widening current account deficit; and (iii) worsening reserves-adequacy position.

1. Staff assesses the real exchange rate as modestly overvalued. The real effective exchange rate has appreciated over the past year, mainly reflecting the depreciation of the ruble and sharp appreciation of the U.S. dollar, against which the tenge is tightly managed. Based on staff’s estimates and projections, the misalignment is judged to be in the range of 4-14 percent depending on what approach is used. The estimates are naturally subject to methodological uncertainties and data limitations. Based on the Fund’s External Balance Assessment (EBA-lite) methodology, the tenge’s real effective exchange rate is viewed as overvalued by around 14 percent (Table 1). The EBA-lite analysis reflects values for the cyclically adjusted fiscal balance, consistent with medium-term sustainability, and other policy values (e.g., private credit/GDP and the capital control index) that are consistent with established international benchmarks. Based on CGER calculations, the tenge is assessed as overvalued in the range of 4-14 percent (Table 2). The authorities, in contrast, do not view the exchange rate as significantly overvalued, in light of last year’s devaluation and the recent rebound in oil prices and Russian ruble appreciation. Moreover, they point out that, given the high volatility of the ruble and the predominantly speculative nature of local money markets, any estimated misalignment should be considered with a great deal of caution. The authorities expect the misalignment to disappear by end-2015.

Table 1:EBA-lite Assessment
Current ValueDesirable Value
Current Account (percent of GDP) 1/−2.50.1
Cyclically adjusted fiscal balance (percent of GDP)−2.93.6
Change in reserves (percent of GDP)0.00.0
Private credit (percent of GDP)33.360.0
Capital control index0.80.5
Real Exchange Rate Gap14.2

For the current account, Current Value reflects the last four quarters (2014Q3-2015Q2p), while the Desirable Value reflects staff’s estimated current account norm.

For the current account, Current Value reflects the last four quarters (2014Q3-2015Q2p), while the Desirable Value reflects staff’s estimated current account norm.

Table 2.CGER Assessment 1/
Underlying CA (Percent of GDP)CA Norm (Percent of GDP)Estimated over(+)/under(-) valuation
Macroeconomic Balance−0.90.15.7
Equilibrium Real Exchange Raten.a.n.a.4.1
External Sustainability 2/−0.91.613.6
Sources: Kazahkstani authorities and IMF staff estimates and projections.

Based on coefficients estimates for net oil exporters from “Are Middle Eastern Current Account Imbalances Excessive” WP/11/195.

NFRK assets included. The CA norm is the adjusted current account balance needed to stabilize the average 2018-20 NFA position.

Sources: Kazahkstani authorities and IMF staff estimates and projections.

Based on coefficients estimates for net oil exporters from “Are Middle Eastern Current Account Imbalances Excessive” WP/11/195.

NFRK assets included. The CA norm is the adjusted current account balance needed to stabilize the average 2018-20 NFA position.

Real Effective Exchange Rate (REER)

(December 2000=100)

Source: National Bank of Kazakhstan; last data point: May 2015.

2. The current account position is expected to worsen in the near term. The current account deficit during the first quarter of 2015 was 0.5 percent of GDP—smaller than previously expected. Still, given the external backdrop (lower oil prices, slowing external demand, strong tenge), staff projects a current account deficit of around 3½ percent of GDP for the full year and 3 percent of GDP next year. Moreover, following sustained current account surpluses in earlier years, the current account position is only expected to gradually improve and remain in small deficit over the medium term. Similarly, Kazakhstan’s international investment position is expected to decline over the coming years.

Kazakhstan: Current Account Deficit

(in percent of GDP)

Sources: NBK and IMF staff estimates.

Kazakhstan: International Investment Position (IIP) and Capital Flows

(in USD billion)

Sources: NBK, and IMF staff estimates.

1/Capital flows include net foreign direct investment, net portfolio investment, and other net investment.

3. NBK’s reserves have been under pressure, while the oil fund remains large. Based on the IMF ARA methodology,1 central bank gross reserves (i.e., excluding the national oil fund) remain below the suggested adequacy range of 100-150 percent, and are also below those in emerging markets and CCA peers (Panel Figure). Moreover, while headline gross reserves figures have been stable in recent months (around $29 billion), reserves include increased FX swaps and banks’ correspondent accounts with the NBK (the latter reflects the sharp increase in the dollarization of bank deposits). At the same time, the oil fund, with an accumulated value of around $75 billion (35 percent of GDP), provides a significant buffer for the economy as a whole.

Reserves as Percent of ARA Metric: Kazakhstan

Reserves/ARA Metric

(Minimum adequacy threshold: 100%)
Annex III. Kazakhstan: Stress Testing the Banking System1

As an oil exporting economy, lower oil prices have a negative impact on Kazakhstan’s financial sector and on economic activity more generally. This annex describes the solvency stress test that was performed on Kazakhstan banks to assess the impact on bank solvency of two shocks: lower oil prices and exchange rate depreciation. Real GDP growth responds to oil prices given elasticity from staff’s macroeconomic framework. Stress tests on bank balance sheets as of end 2014 (latest available) reveal that the banking system is in a weak position to withstand shocks. In the baseline, before any shocks are applied, three banks have capital adequacy ratios (CAR) below the regulatory minimum of 12 percent. The downside scenario has 3 banks below the regulatory minimum. In the severe scenario, 17 banks (89 percent of banking system assets) do not meet regulatory minimum standards and recapitalization needs amount to 3 percent of GDP. The 2014 FSAP also found the banking system unable to withstand shocks (tail risks) based on stress tests on balance sheets as of end 2013. However, system resilience to shocks has decreased further with weaker of bank balance sheets in an environment of low domestic activity and profitability, and the legacy of high nonperforming loans (NPLs) that continue to hamper credit.

A. Insights from the 2014 FSAP Stress Tests

1. The tests showed that banks are in a weak position to withstand shocks:

  • The initial baseline was adjusted to take account of the 2014 exchange rate devaluation and the phase-out of capital instruments ineligible under Basel III. This adjustment resulted in 7 banks (accounting for nearly half of the system’s assets) failing to meet regulatory minimum with a capital shortfall equivalent to 2½ percent of GDP.

  • In the Stress scenarios, two additional banks failed to meet capital requirements, bringing the total to 9 banks accounting for slightly more than half of the banking system’s assets. Under the more severe scenario, the capital deficiency reaches 5 percent of GDP. An oil price shock with prices at $40 per barrel (pb), assessed in combination with the risk of lower-than-anticipated growth potential in emerging markets—especially in Russia and China resulted in a decline in real GDP of 9 percentage points over two years, corresponding to a one-standard deviation move in two-year cumulative growth, or the highest decline registered since the break-up of the Soviet Union. Nine banks failed to meet the regulatory minima.

  • The stress tests revealed the main risks to be foreign exchange (FX) and concentration risks. The exchange rate was found to be a key driver of indirect credit risk for both the corporate and the household/SME sectors, and oil prices are also important determinants of corporate credit quality. Direct FX risk, however, appeared low due to the cap on open positions. Concentration risk, measured by the impact of a default of the largest 3 exposures of each bank, was found to be high as capital ratios fell by around 7 percentage points from their baseline level. 11 banks (about a third of the banking system’s total assets) would breach current regulatory minimum requirements if their 3 largest exposures defaulted. Sovereign and other market risks were found to be low.

  • Despite weaknesses in some banks, Kazakh banks on aggregate were found sufficiently liquid.

B. The Outcome in 2014

2. The FSAP stress test’s oil price assumption was partially realized in 2014 but other assumptions did not quite materialize. As a result only three, rather than seven, banks breached the regulatory minimum capital:

  • While oil prices declined and reached below $40pb briefly in the second half of 2014, the average oil price for the year was high at $96.2pb, compared with the FSAP assumption of $40pb for the year. This was still one of the highest oil price levels of the past 20 years. Thus, the overall impact on macroeconomic variables and NPLs was moderate compared to that assessed by the stress tests. At end-2014, 3 banks (accounting for 18 percent of total banking system assets) were below the capital regulatory minimum of 12 percent. An average annual price below $40 is likely, given historical trends—oil prices were below $40pb on 10 of the last 20 years, and even below $20pb in 1995 and 1998.

  • The oil price shock during the FSAP was assessed in combination with weaker economic activity to a recession and a large depreciation of the exchange rate. In actuality, economic activity was stronger and exchange rate depreciation less. Further, the oil price decline took place in the second half of 2014 and the impact feeding through other macroeconomic variables had not materialized fully as of end-2014.

Petroleum Oil Spot Price, 1995-2014

(U.S. Dollars)

3. Banks balance sheets, however, still weakened in 2014. Capital adequacy declined as profitability weakened but also given that banks began writing off NPLs that were not well provisioned for. Bank profitability declined and banks’ level of provisioning for NPLs also dropped (from 83 to 80 percent of NPLs). Provisioning estimates include collateral. In Kazakhstan, collateral is mostly real estate—a sector that experienced a bubble burst in 2009 and is yet to recover. Banks do no regularly revalue collateral and estimates are still largely based on high prices. This results in an overestimation of provisioning and, thus, the strenght of bank balance sheets. As a consequence of the adverse developments over the last year, the impact of the FSAP stress scenario would now be even higher than in 2014.

4. The distribution of CARs among banks is quite wide. In the baseline—before shocks are applied—three banks are below the 12 percent CAR minimum while several banks are close to it. There are also a number of banks with CAR far above the minimum. In addition to being close to the minimum CAR, a number of banks also have low levels of profitability. Banks would normally use profits as the first line of defense before dipping into capital. When profits are low, shocks have a larger impact on capital. The environment of a large level of non-performing loans in the banking system (23 percent in May, down from 34 percent a year ago) and weak economic activity (economic growth at 2 percent in the first quarter of 2015, down from 4 percent in 2014 and 6 percent in 2013) has reduced bank profitability. NPLs declined further in July to 13 percent with the revocation of BTA bank’s banking license. One bank has a negative return on assets (ROA)—a measure how well bank management is employing the bank’s assets to make a profit. About 19 banks are below the 1.5 percent level of ROA that banks generally strive to achieve.

Banking System Capital and Assets

C. The Updated Solvency Stress Test

5. The solvency stress test assesses the impact of lower oil price, an exchange rate depreciation, and weak economic activity on bank solvency. The assessment of solvency risk is top-down (as was the case in the FSAP) given that bottom-up bank level stress tests are not available. The baseline scenario is based on the latest World Economic Outlook assumptions and projections as well as desk estimations and projections of macroeconomic variables (real GDP and exchange rate for Kazakhstan). The historical relationship between macroeconomic variables and NPLs is analyzed using a vector autoregression (VAR) model. Single equation estimation is also undertaken to assess robustness. Three scenarios are considered: baseline, downside, and severe; over 2015 to 19 and they are presented below:

  • Baseline scenario: In the baseline scenario, estimates of oil are the WEO baseline and real GDP growth and the exchange rate are based on staff’s macroeconomic projections. The projections assume further weakening in oil prices from the current $62.8pb to $58.9pb. Real GDP growth moderates from 4.3 percent in 2014, to 2 percent in 2015 and rises to 4.4 percent by 2019. The exchange rate depreciates from 179.2.4 in 2014 to 185.8 in 2015 and to 188 by 2019.

  • Downside Scenario: In the downside scenario, a 0.5 historical standard deviation on the oil price is assumed in. Oil prices fall to $43.2pb in 2015 and $48.6bp in 2016. Real GDP grows by 1.2 percent in 2015 and improves to 2.7 percent in 2016. The exchange rate is subjected to a 1.5 standard deviation shock, which puts it at 6.6 percent depreciation from the current level. It depreciates to 198 in 2015 and to 200.5 in 2016. The larger standard deviation shock on the exchange rate compared to oil prices mimics the large currency swings during periods of stress.

  • Severe Scenario: In the severe scenario, a 1.2 standard deviation from the baseline is assumed. Oil prices fall to $21.3pb in 2015 and $26.7pb in 2017 and go back to the baseline in 2017. Real GDP grows by 0.1 percent in 2015 and rises to 2 percent in 2016. The exchange rate is subjected to a 3.5 standard deviation shock, which puts it at 15.5 percent depreciation from the current level in 2015. It depreciates to 215 in 2015 and worsening to 217 in 2016.

Kazakhstan: Exchange Rate1/

1/ an increase is a depreciation/or devaluation

6. The exchange rate shocks assessed in the scenarios above are mild given past experiences with the exchange rate in Kazakhstan. The exchange rate tends to exhibit low levels of volatility with large exchange rate adjustments. The exchange rate devalued by over 25 percent in 2009 and by 20 percent in 2014. There were larger exchange rate adjustments in earlier periods including over 74 percent in 1999. At the same time, the authorities have taken measures to improve banking system sensitivity to exchange rate shocks including by introducing NOP limits and banning foreign wholesale borrowing. The latter contributed significantly in the crisis of 2008/09. The size of fx loans to total loans has declined over the years.

The Macroeconomic Model

7. A VAR model is used to assess the response of NPLs to macroeconomic variables, which is used in turn to assess the impact of changes in NPLs (in response to macroeconomic variables) on bank solvency. The variables included in the model are: the ratio of NPLs to total loans (NPL), the percent change in the oil price (OIL), Kazakhstan real GDP growth (KAZG), partner countries real GDP growth (PARTGDP), and the exchange rate (EXCHR). The variables are tested for the presence of a unit root and those with a unit root are first differenced before being included in the model. The model is estimated on quarterly data over 2000 to 2014. The lag length used is 3, selected to minimize the Akaike information criterion, the sequential modified LR test statistic, and the final prediction error.

Kazakhstan: Nonperforming Loans and Exchange Rate

8. Upon eyeballing the data, there is a clear pattern of NPLs responding to exchange rate depreciation. Whenever there was a large depreciation of the exchange rate, NPLs increased. This happened clearly in 2009 and following the devaluation in early 2014.

9. NPLs are highly sensitive to changes in oil prices and a drop in output albeit with a delayed impact. An increase in oil prices results in a decline in NPLs in 4-5 quarters. Furthermore, the domestic real GDP impact on NPLs is persistent as the maximum impact happens about 11 to 12 quarters after the initial shock (Table 1). Economic activity in partner countries appears to account for a significant share of movements in NPLs, followed by domestic economic activity and the exchange rate. Translated into coefficients (elasticity), a percentage point increase in domestic real GDP growth results in a 0.318 percentage point decline in NPLs; a percentage point increase in the oil price results in a 0.006 percentage point decline in NPLs; a percentage point increase (depreciation) in the exchange rate results in a 0.047 percentage point increase in NPLs. Oil prices have a much larger effect on the exchange rate and on real GDP growth compared to the impact directly on NPLs, as further revealed through variance decompositions, also presented below. The results also show high dependence of the domestic economy on activity in partner economies.

Table 1.Impulse Response Functions for NPLs

10. Similar results to the VAR appear in single equation estimation. The results show a higher impact of domestic economic activity on NPLs compared to the other macroeconomic variables.

Single Equation Estimation of Determinants of Bank NPLs in Kazakhstan
VariableCoefficient
NPL (L1)0.9216***
Domestic GDP growth0.0235
Domestic GDP growth (L1)−0.3117***
EXCHR0.0308
Exchange rate (L1)−0.0361
Exchange rate (L2)0.1708*
Exchange rate (L3)−0.0668**
Oil price change0.0164*
Oil price change (LI)−0.0275**
Constant3.5811**
R-squared0.9742
Adjusted R-squared0.9692
F-statistic197.05
Prob(F-statistic)0.0000
Prob(Wald F-statistic)0.0000

Note: *** p < 0.01, ** p < 0.05, * p < 0.1

Note: *** p < 0.01, ** p < 0.05, * p < 0.1

11. The tests simulate balance sheets and profit/loss accounts for the individual banks given the NPL forecasts upon applying the effects of the shocks. In the simulation, liabilities are held constant. Given that historical balance sheet data is not available to staff, interest margins on current loans and liabilities, as well as net non-interest income, are given each banks performance in 2014. Given that estimates of collateral are unreliable, new NPLs are assumed to be provisioned at 100 percent. This further dents profits. When net income is positive, the test assumes that 80 percent of profits are retained and the rest is paid out as dividends. A bank accumulates capital buffers when its capital ratio declined in the previous year. Capital covers the loss when net income is negative. NPLs are deducted from total loans and interest income is earned on standard loans that exclude NPLs.

Table 2.Variance Decompositions

The Solvency Stress Test Results

12. The results reveal that a number of banks are unable to withstand higher NPLs due the shocks described above. It should be noted that the shocks are applied directly on bank balance sheets. No adjustments have been made on capital to take into account the phase-out of capital instruments ineligible under Basel III as was done in the FSAP neither to correct for provisioning. Even without this adjustment and with very mild shocks, banks are still highly susceptible to shocks. In the downside scenario, one bank goes further below the regulatory minimum CAR. System CAR goes down to 16 percent (from 17.3 percent in the baseline) while the severe scenario brings it down to 6 percent, below the regulatory minimum. In the severe scenario, as much at 17 banks have CAR below the regulatory minimum. Recapitalization needs amount up to 3 percent of GDP. NPLs decline over the projection period reflecting their response to assumptions on macroeconomic variables, which improve over the horizon, and together with favorable assumption on reinvested earnings, capital improves. NPLs by sector reveal that the construction and the trade sectors account for a larger share of NPLs in the system while agriculture as well as trade and communication sectors account for very small shares.

Stress Test Assumptions and Results
HistoricalBaselineDownsideSevere
2014201520162017201520162017201520162017
Assumptions
Domestic real GDP growth4.32.03.24.81.22.74.30.12.03.7
Oil price (in U.S. dollars)96.258.964.267.143.248.651.521.326.729.6
percent change over latest price (62.8)−31.212.45.9−66.025.110.8
Exchange rate179.2185.8188.0188.0198.3200.5200.5214.9217.1217.1
percent change over latest rate (186.1)6.61.10.015.51.00.0
Impact
Nonperforming loans (percent of total loans)32.213.05.81.521.711.53.537.624.69.9
Capital adequacy ratio17.417.024.523.116.124.830.46.015.526.0
CAR < 8 percent Number of banks11112111221
8 percent < CAR < 12 percent Number of banks2200100520
CAR > 12 percent Number of banks35353737353737213437
Recapitalization to achieve 12 percent CAR
Million Tenge97,914297,288510,326138,759295,659501,0761,233,496442,955497,869
Percent of 2014 GDP0.20.81.30.40.71.33.11.11.3

13. Several other simulations were done to assess the robustness of the results—a finding that the banking sector is highly susceptible to shocks. The first assessed only the impact of the exchange rate. The relationship between NPLs and the exchange rate was determined using a single estimation equation. The response of NPLs to the exchange rate is slightly high than in a multiple variable estimation given that the other variables are not present to account for all the movements in NPLs. The results are shown below and reveal that even with a very mild exchange rate shock (15 percent depreciation from the current level of the exchange rate), 10 banks have CAR below the regulatory minimum. This is a mild shock considering that much larger exchange rate adjustments that occurred in the past, including 20 percent in 2014 and a 25 percent one in 2009. A much larger exchange rate adjustment of 70 percent occurred in 1999. A 25 percent exchange rate depreciation results in 17 banks going below minimum CAR and recapitalization needs are estimated at 3.2 percent of GDP. Bank’s net open foreign exchange positions (NOP) are small; however, there is extensive lending to unhedged borrowers (borrowers with income in tenge). Data on bank by bank NOP was not made available to the team. The relationship between NPLs and the exchange rate likely reflects also the indirect impact that the exchange rate has on NPLs.

Exchange Rate Shocks
15.5 percent exchange rate shock
Stress Test Assumptions and Results
HistoricalBaselineExchange rate shock
2014201520162017201520162017
Assumptions
Exchange rate179.2185.8188.0188.0214.9217.1217.1
percent change over latest rate (186.1)15.51.00.0
Impact
Nonperforming loans (percent of total loans)32.217.619.119.829.430.731.2
Capital adequacy ratio17.417.115.914.612.013.515.1
CAR < 8 percent Number of banks1111332
8 percent < CAR < 12 percent Number of banks2222754
CAR > 12 percent Number of banks35353535283032
Recapitalization to achieve 12 percent CAR
Million Tenge96,058315,613535,990473,106585,577704,104
Percent of 2014 GDP0.20.81.41.21.51.8
25 percent exchange rate shock
Stress Test Assumptions and Results
HistoricalBaselineExchange rate shock
2014201520162017201520162017
Assumptions
Exchange rate179.2185.8188.0188.0232.6234.8234.8
percent change over latest rate (186.1)25.00.90.0
Impact
Nonperforming loans (percent of total loans)32.217.619.119.837.939.039.3
Capital adequacy ratio17.417.115.914.65.86.88.1
CAR < 8 percent Number of banks11111299
8 percent < CAR < 12 percent Number of banks2222573
CAR > 12 percent Number of banks35353535212226
Recapitalization to achieve 12 percent CAR
Million Tenge96,058315,613535,9901,267,2791,186,4751,221,174
Percent of 2014 GDP0.20.81.43.23.03.1

14. A reverse stress test also reveals banking system vulnerability. The test reveals that an increase in NPLs by 16 percent brings the system wide CAR to the regulatory minimum of 12 percent (assuming a 70 percent level of provisioning). At this level, 6 banks would be below the regulatory minimum.

Annex IV. Kazakhstan: Implementation of the 2014 FSAP Key Recommendations1
Recommendations and Authority Responsible for ImplementationTime1Status of Implementation
Financial Stability
Closely monitor quality of foreign currency-denominated loans. [NBK]ILimited progress. The NBK reports that it monitors the quality of foreign currency-denominated loans. However, lending in foreign currency to clients with earnings in tenge is still widespread.
Closely monitor banks’ concentrated large exposures. [NBK]IIn progress. The NBK regularly collects information on large exposures to monitor conformity to the prudential limits on concentrated large exposures.
Banking Oversight
Rebalance the emphasis of supervision towards a more risk-based approach [NBK]MTLimited progress. Bank supervision remains largely compliance based. They still rely on the CAMELOS system (as per the guidance number 600 of December 31, 2014 and number 29 from February 26, 2014). However, the NBK reports that it has developed draft Guidelines to off-site supervision of the banking sector that include the principles of risk-based supervision. They have been receiving technical assistance from an international donor to strengthen supervision with the aim of eventually conducting inspections with a risk-based approach.
Support the supervisor’s capacity to challenge banks’ decisions on provisioning [NBK]IIn the process of being implemented. The authorities have been working to formulate a technique of determining collateral value and provisioning for second-tier banks and organizations engaged in certain types of banking operations.
Intensify the supervision of the cross-border operations of Kazakh banks and signing MoU [NBK]NTIn progress. In 2014 and 2015 the NBK took part in drafting Memorandums of Understanding with the following foreign financial regulators: the Reserve Bank of India, the Central Bank of the Islamic Republic of Iran, the Central Bank of the Republic of San Marino, the Central Bank of the Russian Federation, the Polish Financial Supervision Commission, and the Central Bank of Bahrain. In 2015, the NBK plans to hold a joint meeting with representatives of the board of the Central Bank of Russia to discuss consolidated (cross-border) oversight of the banking subsidiaries of Kazakhstan banks, banks participating in banking conglomerates located in the Russian Federation, and the banking subsidiaries of Russian banks in the Republic of Kazakhstan. The NBK plans to conduct a banking college with the regulator board of the Russian Federation.
Analyze regularly indirect credit risk and market risk, including foreign exchange rate risk [NBK]IIn progress. As a part of its stress-testing, the NBK analyzes indirect credit risk and market risk. This, however, has not yet been taken up in the supervision of banks.
Monitor the impact of the adopted macroprudential policies and conduct assessments of effectiveness [NBK]MTBeing implemented. The NBK has been assessing the impact of macroprudential policies that were adopted in early 2014. Overheating in uncollateralized consumer lending declined following the introduction of macroprudential policies. The growth of unsecured consumer lending was 7.9 percent at end 2014, well below the 30 percent limit.
Financial Safety Net, Resolution of NPLs, and Systemic Liquidity Management
Revise purchase and assumption and bridge bank resolution options to exclude a requirement for depositor and creditor approval [NBK]MTImplemented. The NBK reports that a new law was adopted on April 27, 2015 (“On amendments and additions to certain legislative acts of the Republic of Kazakhstan on insurance and Islamic finance”), which includes removal of the requirement for depositor and creditor approval.
Limit emergency liquidity assistance to institutions that are solvent and financially capable of paying a penalty rate of interest [NBK]NTNot Implemented
Develop a procedure for documenting financial stability analysis in cases of provision of state support to the financial sector [NBK]NTRecommendation is being implemented during the development of the financial stability framework
Implement a multi-track approach for resolving the overhang of NPLs [NBK]NTPartially implemented. The authorities have removed legal, tax, and accounting obstacles to NPL write-offs and transfers to special purpose vehicles. However, the authorities are pursuing a different strategy that excludes the PLF in NPL resolution. They plan place funds that were previously intended to capitalize the PLF as a deposit in KKB for liquidity support.
Reduce procedural costs of enforcement obstacles arising from non-registered or junior pledge holders in foreclosures [Government/Legislator].NTLegal amendments have been made to help implement this recommendation. The authorities report that they have made amendments in the law on “On amendments and additions to certain legislative acts of the Republic of Kazakhstan on insurance and Islamic finance,” adopted on April 27, 2015.
Revise the insolvency law to strengthen protection for legal rights of secured creditors by giving them a higher priority in creditors’ ranking [Government/Legislator].NTIn the process of being implemented. The NBK has drafted the amendments to the Law on bankruptcy, the Tax Code, and the Banking Law in order to eliminate the legal barriers identified by the FSAP. Some amendments are also included in the draft law on “On amendments and additions to certain legislative acts of the Republic of Kazakhstan on non-performing loans and assets of banks, provision of financial services, and the activities of financial institutions and the National Bank of the Republic of Kazakhstan.” The amendments have been submitted to the government.
Incentivize out-of-court restructuring by providing tax incentives at creditor’s and debtor’s level for debt write-offs, partial forgiveness, bad debt and collateral sales [Government/Legislator].NTPartially implemented. Amendments were adopted in the Law “On amendments and additions to certain legislative acts of the Republic of Kazakhstan on rehabilitation and bankruptcy, taxation” dated as of 7 March 2014. The NBK has also drafted the amendments to the Law on bankruptcy, the Tax Code and the Banking Law in order to eliminate the legal barriers identified by the FSAP; and the draft Law “On amendments and additions to certain legislative acts of the Republic of Kazakhstan on non-performing loans and assets of banks, provision of financial services, and the activities of financial institutions and the National Bank of the Republic of Kazakhstan.” The amendments have been submitted to the government. The amendments, however, do not address partial forgiveness, bad debt, and collateral sales.
Operationalize the PLF by providing for its broad mandate in NPLs resolution, including bundling of NPLs and adequate financial and staffing resources [NBK/Government/Legislator].NTNot implemented.
Facilitate NPLs transfers into SPVs including by revising bank secrecy rules and property rights registration of the debt assignments and collateral transfers [Government/Legislator].NTImplemented. Amendments were taken in the Law of the Republic of Kazakhstan of April 27, 2015 “On amendments and additions to certain legislative acts of the Republic of Kazakhstan on insurance and Islamic finance.”
Public Pensions, Insurance, and Securities Market Oversight
Include in UAPF’s Charter a clear mandate for UAPF to maximize the retirement income for its members [Government/Legislator].IImplemented. The NBK reports that the Pension Assets Management Council approved a Pension Asset Management Concept that outlines that in the long term the investment return on pension assets of UAPF should be at the level or higher than the inflation rate.
Adjust the mandatory worker’s compensation to avoid collapse of the insurance sector [Government/ Parliament].IImplemented. The NBK reports that legal amendments (in the law “On amendments and additions to some legislative acts on insurance and Islamic finance” and in the law “On compulsory insurance of employees against accidents in the performance of his employment duties.”) have been made to improve the system of compulsory insurance of an employee against accidents at the execution of employment duties. They also resolve a number of issues related to the implementation of this type of compulsory insurance.
Reduce uncertainty in continuity of market rules; confirm legal fact and time of settlement finality [Government, NBK].IIn progress. The NBK has drafted the Law “On payments and payment systems.” The draft was included in Government Legislative Drafting Plan for year 2015 and will be sent to Parliament in December 2015. The Draft Law is currently under consideration by relevant government organizations and market participants.
Source: Kazakhstan Financial System Stability Assessment, 2014, IMF.

“I-Immediate” is within one year; “NT-near-term” is 1-3 years; “MT-medium-term” is 3-5 years.

Source: Kazakhstan Financial System Stability Assessment, 2014, IMF.

“I-Immediate” is within one year; “NT-near-term” is 1-3 years; “MT-medium-term” is 3-5 years.

Annex V. Kazakhstan: Risk Assessment Matrix1
Nature/Source of Main ThreatsLikelihoodExpected Impact on the Economy if Risk Is Realized
Persistently low oil pricesMediumGrowth would further decelerate over the medium term due to reduced oil revenues, lower investments in the oil sector, and lower investor confidence in the Kazakh economy in general. An expansionary fiscal policy response may be adequate to tackle a temporary oil-price shock to address the growth slowdown; but in case of a permanent shock fiscal policy adjustment over the medium term is critical to maintain sustainability.
Sharp growth slowdown in China in 2015-16LowKazakhstan would be adversely affected primarily through lower FDI from China and declining exports to China, both of which have risen in absolute and relative terms in recent years. Similar to the above, if the sharp slowdown in China is seen as temporary, the expansionary fiscal policy response may be appropriate, subject to measures taken to ensure medium-term sustainability
Russia-Ukraine conflict re-intensifiesMediumFurther intensification of the conflict may lead to further sanctions against Russia. As a result, Kazakhstan would be negatively affected through lower export demand and investment flows from Russia. Moreover, political uncertainly may increase, putting further downward pressure on the tenge. Similar to the above, if the conflict intensification is seen as temporary, the expansionary fiscal policy response may be appropriate, subject to measures taken to ensure medium-term sustainability.
Tighter or more volatile global financial conditions:

Sharp asset price adjustment and decompression of credit spreads
HighImpact expected to be limited given the small scale external borrowing by Kazakhstan’s government and corporates. That said, the yield spreads on existing Kazakhstan’s sovereign and quasi-sovereign bonds are expected to rise in line with other emerging market yields. Consideration should be given to speeding up the enhancements of the monetary policy framework and operations, including allowing greater exchange rate flexibility to absorb the shocks. Coupled with enhanced communication strategy, these will help boost confidence in central-bank policy making.
Persistent dollar strengthHighThe severity of the impact of persistent dollar strength depends on the degree to which the tenge continues to be tightly managed against the dollar. Under the prevailing tight exchange rate arrangement, the tenge could continue to come under pressure, especially if dollar strength is combined with sustained lower growth and oil prices. Consideration should be given to speeding up the enhancements of the monetary policy framework and operations, including allowing greater exchange rate flexibility to absorb the shocks. Coupled with enhanced communication strategy, these will help boost confidence in central-bank policy making.
Annex VI. Comparison in the Authorities and Staff’s Fiscal Accounting Frameworks1

1. The authorities and staff use different methodologies to track fiscal accounts. In the authorities’ approach: (i) most national oil fund accounts, including most oil export revenues are kept off-budget; (ii) oil export revenues are only recognized when funds are transferred from the national fund to the budget; (iii) tax on oil exports is treated as non-oil revenues; (iv) capital gains on national oil fund assets, including the effects of devaluation are treated as revenues; (v) direct lending (on-lending) from the national oil fund to quasi-sovereign entities is not recognized as expenditure; and (vi) only a fraction of quasi-sovereign entity’s foreign loans, which is directly guaranteed by the government, are recognized as contingent liabilities.

2. Staff, on the other hand, follows the Government Finance Statistics Manual (GFSM) 2001 framework with revenues and expenditures consolidated into single general government budget. In particular: (i) oil revenues, including oil export duty, are recorded on a cash basis; (ii) interest income of the national oil fund is recognized as revenues but not capital gains or valuation differences; (iii) lending to SOEs is treated as expenditure; (iv) net transfers from the national oil fund are recorded below the line; and (v) debts of SOEs, where there is expectation that the state may become liable for repayment in case of default, are treated as contingent liabilities.

3. The different methodologies result in diverging estimates and forecasts. This is illustrated below for the 2014 fiscal figures and the projected medium-term path for the non-oil fiscal deficits. In particular, the authorities’ approach underestimates oil revenues, and consequently the non-oil deficit, by half relative to GFSM.

Non-oil fiscal deficits as share of GDP
201520162017201820192020
IMF−11.4−10.4−9.8−9.3−9.0−8.7
Ministry of Economy−9.7−7.6−6.7−5.3−4.3−4.1
Source: Kazakhstani authorities and IMF staff estimates
Source: Kazakhstani authorities and IMF staff estimates
2014 Fiscal Figures(In billions of tenge)
AuthoritiesIMF staff
On-budget
Total revenues7,3727,330
Oil revenuesTransfers from the oil fund1,955Oil revenues, of which:4,220
Proceeds from the sale of oil and3,503
Oil export duty717
Non oil revenuesof which:5,4173,110
Oil export duty and717
Interest income and capital gains of oil fund including unrealized devaluation effect.1,862Only interest income of the oil fund is included in revenues
Expenditures8,460Expenditures, of which:8,935
on-lending from national oil fund resources.475
Overall balance(1,088)(1,605)
Non-oil balance(3,043)(5,825)
Below the lineNet change in oil fund assets3,402
Off-budget
Oil fund income, of which:5,366
Proceeds from oil sale of oil3,503
Profits including1,862
Other1.4
Oil fund expenditure, of which:1,964
Transfers to national budget1,955
Other9
Change3,402
Annex VII. Kazakhstan: Debt Sustainability Assessment

Kazakhstan Public Sector Debt Sustainability Analysis (DSA)—Baseline Scenario

(In percent of GDP unless otherwise indicated)

Source: IMF staff.

1/ Public sector is defined as general government

2/ Based on available data.

3/ EMBIG (bp).

4/ Defined as interest payments divided by debt stock (excluding guarantees) at the end of previous year.

5/ Derived as [r -π(1 + g) - g + ae(1 + r)]/(1+g+π+gπ) times previous period debt ratio, with r = effective nominal interest rate; π = growth rate of GDP deflator; g = real GDP growth rate; a = share of foreign-currency denominated debt; and e = nominal exchange rate depreciation (measured by increase in local currency value of U.S. dollar).

6/ The real interest rate contribution is derived from the numerator in footnote 5 as r - n (1+g) and the real growth contribution as -g.

7/ The exchange rate contribution is derived from the numerator in footnote 5 as ae(l + r).

8/ Includes asset changes in the oil fund.

9/ Assumes that key variables (real GDP growth, real interest rate, and other identified debt-creating flows) remain at the level of the last projection year.

Kazakhstan Public DSA—Composition of Public Debt and Alternative Scenarios

Source: IMF staff.

Kazakhstan: External Debt Sustainability Framework, 2010–2020(In percent of GDP, unless otherwise indicated)
ActualProjections
20102011201220132014201520162017201820192020Debt-stabilizing non-interest current account 6/
1Baseline: External debt79.966.667.364.771.279.878.274.570.566.862.9-6.8
2Change in external debt−18.0−13.20.6−2.66.58.6−1.6−3.6−4.0−3.7−3.90.0
3Identified external debt-creating flows (4+8+9)−25.1−27.0−11.2−12.1−1.4−0.7−1.6−3.6−3.8−3.5−3.90.0
4Current account deficit, excluding interest payments−3.0−7.3−2.6−2.0−3.91.91.60.70.10.0−0.26.8
5Deficit in balance of goods and services−14.4−20.3−14.8−11.9−13.2−3.7−3.9−4.7−5.1−5.0−5.4
6Exports44.247.645.139.138.728.428.027.626.725.524.8
7Imports29.927.330.227.225.524.624.122.921.620.519.4
8Net non-debt creating capital inflows (negative)−2.6−4.6−5.6−3.4−2.6−2.5−2.3−2.1−1.9−1.8−1.6−1.6
hideNet foreign direct investment, equity2.54.65.83.42.72.52.32.12.01.81.7
hideNet portfolio investment, equity0.10.0−0.20.0−0.1−0.1−0.1−0.10.00.00.0
9Automatic debt dynamics 1/−19.6−15.1−3.0−6.65.1−0.1−1.0−2.2−2.0−1.7−2.0−5.2
10Contribution from nominal interest rate2.11.92.11.61.81.41.31.21.11.01.01.0
11Contribution from real GDP growth−5.6−4.7−3.1−3.5−2.9−1.5−2.3−3.4−3.1−2.8−3.0−2.8
12Contribution from price and exchange rate changes 2/−16.1−12.3−2.0−4.76.2−3.3
13Residual, incl. change in gross foreign assets (2-3) 3/7.113.811.89.58.09.30.0−0.1−0.2−0.2−0.10.0
External debt-to-exports ratio (in percent)180.5140.0149.2165.3183.9281.3279.1270.4263.9262.5253.8
Gross external financing need (in billions of US dollars) 4/19.213.421.524.421.835.439.039.139.841.442.6
in percent of GDP13.07.110.610.59.910-Year10-Year16.716.915.314.013.112.2
Scenario with key variables at their historical averages 5/79.865.252.540.930.521.0-8.9
Key Macroeconomic Assumptions Underlying BaselineHistorical AverageStandard DeviationFor debt stabilization
Real GDP growth (in percent)7.37.55.06.04.36.43.02.03.24.84.64.45.05.0
GDP deflator in US dollars (change in percent)19.718.23.17.5−8.811.614.3−5.95.66.06.16.05.55.5
Nominal external interest rate (in percent)2.73.03.32.72.62.80.61.81.81.71.71.61.71.7
Growth of exports (US dollar terms, in percent)36.436.72.5−1.1−5.915.627.6−29.77.69.37.75.37.9
Growth of imports (US dollar terms, in percent)13.516.019.92.4−10.810.618.4−7.26.75.25.04.94.7
Current account balance, excluding interest payments3.07.32.62.03.91.73.9−1.9−1.6−0.7−0.10.00.2
Net non-debt creating capital inflows2.64.65.63.42.66.13.22.52.32.11.91.81.6

Derived as [r - g - r(1+g) + ea(1+r)]/(1+g+r+gr) times previous period debt stock, with r = nominal effective interest rate on external debt; r = change in domestic GDP deflator in US dollar terms, g = real GDP growth rate, e = nominal appreciation (increase in dollar value of domestic currency), and a = share of domestic-currency denominated debt in total external debt.

The contribution from price and exchange rate changes is defined as [-r(1+g) + ea(1+r)]/(1+g+r+gr) times previous period debt stock. r increases with an appreciating domestic currency (e = 0) and rising inflation (based on GDP deflator).

For projection, line includes the impact of price and exchange rate changes.

Defined as current account deficit, plus amortization on medium- and long-term debt, plus short-term debt at end of previous period.

The key variables include real GDP growth; nominal interest rate; dollar deflator growth; and both non-interest current account and non-debt inflows in percent of GDP.

Long-run, constant balance that stabilizes the debt ratio assuming that key variables (real GDP growth, nominal interest rate, dollar deflator growth, and non-debt inflows in percent of GDP) remain at their levels of the last projection year.

Derived as [r - g - r(1+g) + ea(1+r)]/(1+g+r+gr) times previous period debt stock, with r = nominal effective interest rate on external debt; r = change in domestic GDP deflator in US dollar terms, g = real GDP growth rate, e = nominal appreciation (increase in dollar value of domestic currency), and a = share of domestic-currency denominated debt in total external debt.

The contribution from price and exchange rate changes is defined as [-r(1+g) + ea(1+r)]/(1+g+r+gr) times previous period debt stock. r increases with an appreciating domestic currency (e = 0) and rising inflation (based on GDP deflator).

For projection, line includes the impact of price and exchange rate changes.

Defined as current account deficit, plus amortization on medium- and long-term debt, plus short-term debt at end of previous period.

The key variables include real GDP growth; nominal interest rate; dollar deflator growth; and both non-interest current account and non-debt inflows in percent of GDP.

Long-run, constant balance that stabilizes the debt ratio assuming that key variables (real GDP growth, nominal interest rate, dollar deflator growth, and non-debt inflows in percent of GDP) remain at their levels of the last projection year.

Kazakhstan: External Debt Sustainability; Bound Tests1/2/

Sources: International Monetary Fund, Country desk data, and staff estimates.

1/ Shaded areas represent actual data. Individual shocks are permanent one-half standard deviation shocks. Figures in the boxes represent average projections for the respective variables in the baseline and scenario being presented. Ten-year historical average for the variable is also shown.

2/ For historical scenarios, the historical averages are calculated over the ten-year period, and the information is used to project debt dynamics five years ahead.

3/ Permanent 1/4 standard deviation shocks applied to real interest rate, growth rate, and current account balance.

4/ One-time real depreciation of 30 percent occurs in 2010.

Annex VIII. Macroprudential Policies in Kazakhstan1

Background

1. The NBK has introduced several prudential and macroprudential tools, mostly following the 2009 financial crisis. The tools aim to address some of the risks already facing the banking sector and some tools will be implemented over the coming years. The existing macroprudential and prudential tools are presented below:

To reduce exchange rate risk and dollarization:

  • Differentiated reserve requirement ratios (RR): Minimum RRs on the share of bank liabilities associated with cash in vault and correspondent accounts with the NBK were introduced in May 2006. There were amendments in November 2012 to differentiate minimum RRs between domestic and foreign liabilities and further in 2015 to refine the structure. The 2012 amendments aimed at limiting the inflow of short-term speculative capital; diversifying the portfolio of banks; developing the market of long-term bank instruments; and reducing pressure on the foreign exchange market. The RR ratios are currently (i) 2 percent on tenge short-term domestic liabilities; (ii) 4 percent on tenge short-term foreign liabilities; (iii) 0 percent on tenge long-term domestic liabilities; (iv) 2 percent on tenge long-term foreign liabilities; (v) 2 percent on FX short-term domestic liabilities; (vi) 6 percent on FX short-term foreign liabilities; (vii) 0 percent on FX long-term domestic liabilities; and (viii) 2 percent on FX long-term foreign liabilities.

  • Limits on the net open FX positions (NOFP) or currency mismatches: Limits on the net open FX position were introduced in 2002, as follows: (i) 12.5 percent for currencies of countries with the S&P rating of at least “A” (or equivalent); (ii) 12.5 percent for euro; (iii) 5 percent for currencies of countries with the S&P rating below “A” (or equivalent); and (iv) a 25 percent net foreign currency position limit. There is also a 30 percent limit on open FX off-balance sheet positions, effective from July 2014. The measure is aimed at preventing speculative pressure in the spot and FX swap market and to control foreign exchange risk off-balance sheet on a net basis.

To reduce credit risk:

  • Measures in non-collateralized consumer lending: Effective from April 1, 2014, the NBK introduced: (i) a 50 percent cap on debt-to-income ratio; (ii) an increase in the retail lending risk weight for capital adequacy to 100 percent (from 75 percent); and (iii) a 30 percent limit for growth in consumer loans for each bank. The measures were aimed at minimizing risks associated with rapid growth in consumer lending.

To contain concentration risk:

  • Credit concentration limits: The NBK has the following limits on exposure to a single borrower: i) 10 percent of a bank’s capital for related counterparties except for subsidiaries acquiring bank’s doubtful and loss claims in order to manage them. The aggregate amount of risk for borrowers related to bank insiders should not exceed an amount of bank’s capital; ii) 25 percent of a bank’s capital for other counterparties (10 percent for unsecured lending, unsecured contingent liabilities related to borrower or to third parties). Banks are also required to submit on a monthly basis reports to the NBK on maximum exposure to a single counterparty and group of connected counterparties.

To reduce liquidity risk:

  • Liquidity Requirements: Introduced in 2005, amended in 2009 and in 2012. Maximum value of liabilities of non-residents to equity (authorities’ k9 ratio) was relaxed from 2 to 3, starting in July 2009. The current liquidity ratio (average monthly highly liquid assets of a bank to average monthly demand liabilities; k4) was introduced to differentiate between tenge and foreign currency liquidity, with the regulatory value set at 0.3. In 2012, liquidity ratios were tightened, from July, 2012—at least 0.08, and from July, 2013—at least 0.09, adding contingent liabilities to the calculation of capital adequacy ratio (k1-1).

To strengthen solvency:

  • Dynamic Provisioning (DP): DP was introduced in January 2013 to smooth provisioning costs over the business cycle by gradually building a countercyclical loan-loss reserve in good times and then using it to cover losses as they arise in bad times. The DP has been suspended since January 2014, while retaining its volume in the structure of capital as accumulated at end-2013.

  • General Countercyclical Capital Buffer (CCB): The CCB is planned to be introduced from 2016.

  • Capital Surcharges on SIFIs: Capital surcharges are planned to be introduce from 2016.

Assessment

2. The assessment of effectiveness of the macroprudential tools needs to be put in the historical context. Prior to the 2009 financial crisis there was significant foreign wholesale borrowing by banks (at 44 percent of GDP between 2005 and 2007). This borrowing helped finance a boom in credit to the non-tradable sector (mainly construction). The tightening in external financial conditions forced aggressive deleveraging; resulting in a bust in asset prices (particularly real estate). The 20 percent devaluation in 2009 added to FX pressures, spreading to unhedged FX borrowers. Since then, measures have been taken to significantly limit foreign wholesale borrowing and unhedged FX borrowing has also declined. Banks had large net open FX positions at the time of the crisis and they have declined significantly as supervision began to enforce the existing NOFP limits. However, the NOP has widened again in 2014—though not to the extent of 2009—and liabilities this time are mostly funded from deposits rather than foreign wholesale borrowing.

3. Recommendations to improve the macroprudential toolkit are aimed at addressing the major risks on financial stability. These are mainly on reducing exchange rate and credit risks. The planned implementation of the CCB and the phasing in of Basel III capital requirements should help enhance resilience of the banking sector.

Exchange rate risk:

  • The risk: Exchange rate risk, however, remains the largest risk for the banking sector. Fears of devaluation are still looming and are contributing to a rise in FX liabilities relative to FX loans as the net open FX position of banks is negative. Essentially, liabilities upon which interest must be paid exceed interest earning assets. FX risks are especially elevated in the corporate sector—the largest borrowers from the banking system—where leverage and the net open FX position to equity are extremely high.

  • The recommendation: To help limit risks emanating from high exposure to the corporate sector, the NBK should consider introducing risk weights or exposure caps for corporate exposures. This will help the banking system withstand corporate credit losses by building capital buffers (see IMF Staff Guidance on Macroprudential Policy). Risk weights can help build buffers by affecting the supply of credit indirectly while FX exposure caps curtail the supply of credit directly. Reducing FX lending to borrowers whose earnings are not in FX should also be considered. This can be done through higher risk weights on such exposures. Tighter net open position limits are also recommended to contain exchange rate risk.

Credit risk:

  • The risk: Even though there are concentration limits, credit concentration remains a risk to financial stability. Though concentrated large exposures have fallen from levels during the global financial crisis, at 208 percent of capital, they remain large. The FSAP found that if three largest bank exposures were to default, about 11 banks would breach the regulatory minimum capital requirement. In 2014, the construction sector received the largest share of new corporate sector credit (24 percent) while it also had the largest share of new NPLs (41 percent, rising from 15 percent the previous year). On the other form of credit risk, macroprudential measures introduced by the NBK to contain consumer credit were effective in reducing such lending.

  • The recommendation: The NBK should ensure compliance with exposure limits to restrict bank exposures to single counterparties or groups of connected counterparties.

4. Proper implementation and enforcement of macroprudential tools is critical. The FSAP warned that inconsistent implementation of macroprudential measures, as has been in the past (including forbearance for some banks resulting in ineffective incentive structures), could add to regulatory uncertainty and result in ineffective incentive structures and a loss in confidence in the domestic financial system.

Annex IX. Kazakhstan: Inclusive Growth1

Kazakhstan’s economic growth has been broadly inclusive, but further efforts are needed to reduce growing income gaps. Efforts to reduce income inequality by promoting faster employment growth have been relatively weak, in part because the enclave nature of extractive industries leaves wide welfare gaps among regions and along the urban-rural divide. The twin priorities of economic diversification and private sector development require addressing key structural weaknesses, namely human capital and institutionsareas where Kazakhstan lags behind emerging market peers. Staff carried out an analysis on inclusive growth, focusing on income inequality and employment, under the 2014 Article IV consultations.2 Important challenges highlighted by the analysis are resource dependence, public sector dominance, and governance.

Income Inequality

1. Based on available data, poverty has declined considerably over the past decade, from 47 percent in 2001 to 3 percent in 2013, with urban and rural poverty falling, respectively, from 36 percent to 1.3 percent and from 59 percent to 5 percent. Urban-rural disparity, however, remains relatively high: the rural poverty gap—defined as rural poverty rate minus the national poverty rate—is wider than in some poorer neighboring countries. Regional disparities are also high with poverty ranging from 1.7 percent in Astana to over 10 percent in south Kazakhstan. High poverty rates are observed in both non-oil and oil-rich regions, reflecting the enclave nature of the oil industry, which is capital intensive and thus does not create significant economic spillover effects.

Employment

2. Thanks to robust economic growth, the unemployment rate in Kazakhstan has declined rapidly, from over 10 percent in the early 2000s to around 5 percent today. High female labor force participation (67 percent) is another positive feature of the labor market in Kazakhstan. Yet, given the high level of self-employment, which reflects in many ways informal part-time employment, the low recorded level of unemployment needs to be treated with caution. Moreover, job creation has not been commensurate with economic growth (especially since the global financial crisis), particularly in the manufacturing sector where it is anemic, despite the accelerated industrialization program. Employment in the agricultural sector has continuously fallen, because of the unproductive farm structure, which reflects limited access to commercial credit and lack of long-term investment. The low degree of job creation also reflects a capital intensive economy dominated by the oil and public sectors. In 2012, in part as an attempt to address some of these gaps, the authorities introduced new labor market policies such as (i) supporting job search assistance and intermediation services; (ii) development of entrepreneurship in rural areas through micro-credits; (iii) free training and retraining, including support for relocation; and (iv) provision of employment through infrastructure and communal service programs.

Business climate and governance

3. Kazakhstan scores relatively low on several business environment and governance indicators. For example, while it has seen some improvement in the World Bank’s Doing Business indicators, the diagnostic of firm-level business constraints reveals important weaknesses in labor skills, taxation, and corruption. Governance issues are mostly rooted in weak transparency prevalent in off-budget operations, including through large quasi-fiscal institutions. The large role of the state in the economy—for example, the size of the state investment holding company Samruk Kazyna (SK) has grown over time, reaching roughly 50 percent of GDP (in assets)—represents an additional barrier to enhanced private-sector led business environment. The government is launching a number of programs in cooperation with MDB to strengthen governance indicators, including through greater transparency of quasi fiscal institutions.

Fiscal policy

4. The scale of government spending on critical social sectors (including health, education and social protection) has been broadly adequate in Kazakhstan compared to countries with similar income per capita. In addition, the government has attempted to protect critical spending (as a share of GDP) during times of fiscal adjustment. However, the 2015 budget cuts have affected some important social programs including the reduction of the budget for Health and Social Services ministry by 15 percent. Quality of social spending, as measured by outcomes, on the other hand, needs improvement. The World Bank has identified some areas of concern, especially related to education. In particular, Kazakhstan’s students’ performance on PISA (Program for international Student Assessment) has been below their peers’ in comparator countries. Finally, there are significant disparities in educational enrollment across residence and income status within the country. Staff will deepen the analysis of these issues in the context of its future diagnostic of inclusive growth in Kazakhstan.

These estimates, derived from the External Balance Assessment (EBA) methodology and CGER calculations, are broadly in line with market indicators, e.g., non-deliverable forward (NDF) rates suggest (as of end-June) depreciation expectations of around 10-15 percent over the next 6-12 months.

This is equivalent to a permanent oil-price shock applied in 2015 as a 0.5 percent standard deviation from historical oil prices.

Prepared by Yahia Said.

Prepared by Matteo Ghilardi and Yahia Said.

The Assessing Reserve Adequacy (ARA) Metric captures the adequacy of central bank reserves in terms of GDP and balance of payments developments, including imports, portfolio and other investment liabilities, and short-term external debt (at remaining maturity). Since Kazakhstan falls into the group of resource dependent countries with oil exports exceeding 50 percent of GDP, this suggests the need for higher buffers to reflect the added volatility in oil prices. Recalculating the ARA metric with oil prices one standard-deviation lower than currently projected results in a wider shortfall from the adequacy threshold.

Prepared by Nombulelo Duma.

Prepared by Nombulelo Duma.

The Risk Assessment Matrix (RAM) shows events that could materially alter the baseline path (the scenario most likely to materialize in the view of IMF staff). The relative likelihood of risks listed is the staff’s subjective assessment of the risks surrounding the baseline (“low” is meant to indicate a probability below 10 percent, “medium” a probability between 10 and 30 percent, and “high” a probability of 30 percent or more). The RAM reflects staff views on the source of risks and overall level of concern as of the time of discussions with the authorities. Non-mutually exclusive risks may interact and materialize jointly.

Prepared by Matteo Ghilardi and Yahia Said.

Prepared by Nombulelo Duma.

Prepared by Yahia Said.

See IMF Country Report No. 14/243 Republic of Kazakhstan—Selected Issues “Assessment of Inclusive Growth.”

Other Resources Citing This Publication