Republic of Azerbaijan: 2021 Article IV Consultation-Press Release; and Staff Report
Author:
International Monetary Fund. Middle East and Central Asia Dept.
Search for other papers by International Monetary Fund. Middle East and Central Asia Dept. in
Current site
Google Scholar
PubMed
Close

1. A robust and swift policy response to the pandemic set the floor on the economic contraction and ignited a recovery. Containment, health, and economic relief measures helped slow infections and protect the vulnerable. Defeating the pandemic and firming up the recovery— including through a broadening of the vaccine rollout—is the immediate priority. With oil reserves likely to be exhausted in the next few decades, multifarious reforms are needed to strengthen long-term fiscal sustainability and spur private sector-led, diversified and inclusive growth. Tackling these long-term challenges requires integrated policy frameworks and transparency, to support decision-making, foster credibility, and attract foreign investment.

Abstract

1. A robust and swift policy response to the pandemic set the floor on the economic contraction and ignited a recovery. Containment, health, and economic relief measures helped slow infections and protect the vulnerable. Defeating the pandemic and firming up the recovery— including through a broadening of the vaccine rollout—is the immediate priority. With oil reserves likely to be exhausted in the next few decades, multifarious reforms are needed to strengthen long-term fiscal sustainability and spur private sector-led, diversified and inclusive growth. Tackling these long-term challenges requires integrated policy frameworks and transparency, to support decision-making, foster credibility, and attract foreign investment.

Context and Recent Developments

1. A robust and swift policy response to the pandemic set the floor on the economic contraction and ignited a recovery. Containment, health, and economic relief measures helped slow infections and protect the vulnerable. Defeating the pandemic and firming up the recovery— including through a broadening of the vaccine rollout—is the immediate priority. With oil reserves likely to be exhausted in the next few decades, multifarious reforms are needed to strengthen long-term fiscal sustainability and spur private sector-led, diversified and inclusive growth. Tackling these long-term challenges requires integrated policy frameworks and transparency, to support decision-making, foster credibility, and attract foreign investment.

2. The confluence of the pandemic and oil shock pushed the economy into a recession. Real GDP declined by 4.3 percent in 2020, driven by oil GDP, which fell by 7.0 percent, in part owing to the OPEC+ commitments (Figure 2).1 Nonoil GDP dropped by 2.6 percent in 2020, with contact-intensive and transportation sectors (accounting for nearly half of the nonoil economy) affected the most.2 Alongside strict containment measures aimed to slow the spread of infections, economic relief measures helped protect jobs and incomes (Annex I). Contraction in private credit was mild, owing to borrower support and interest rate cuts. Amid the recession, inflation remained below 3 percent, and the real effective exchange rate appreciated (by 5.5 percent).

Figure 1.
Figure 1.

Azerbaijan: Economic Impact of COVID-19

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: COVID-19 Government Response Tracker, John Hopkins University, National authorities, Our World in Data, WEO, and IMF staff calculations.
Figure 2.
Figure 2.

Azerbaijan: Real Sector Indicators

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, Haver, and IMF staff calculations.
uA001fig01

Azerbaijan: Fiscal and Monetary Measures, 2020

(In percent)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: IMF Covid Policy Tracker, National authorities and IMF staff calculations.

3. The authorities suspended the fiscal rule and rolled out a sizeable relief package (4.8 percent of GDP), to address public health needs and support individuals and firms affected by the pandemic (Annex I). The overall deficit turned from a surplus of 8.2 percent of GDP in 2019 to a deficit of 6.5 percent in 2020, owing to the decline in oil revenues and pandemic-related spending increases, amplified by the GDP decline (Table 5). Nonoil primary balance (NOPB) widened from -27 percent in 2019 to -30 percent of nonoil GDP in 2020. The fiscal package was financed by drawing down State Oil Fund of Azerbaijan (SOFAZ) assets and reallocating spending. Government debt remained below 35 percent of GDP, as of January 2021 (Annex II).3

Table 1.

Azerbaijan: Selected Economic and Financial Indicators, 2018–26

article image
Sources: National authorities, and IMF staff estimates and projections.

Includes the production and processing of oil and gas.

Consolidates State Budget, State Oil Fund of Azerbaijan (SOFAZ), Nakhchevan Autonomous Region (NAK) and State Social Protection Fund.

Includes a transfer of CBA profits (AzN 650 million) in 2019.

Includes the impact of an extraordinary SOFAZ transfer ($1.4 bn to the CBA in 2017) and expenditures for the NPL program in 2019 (AzN 650 mil).

Starting in 2021, includes guarantees issued to Aqracredit for its acquisition of distressed assets from the IBA.

Table 2.

Azerbaijan: Balance of Payments, 2018–26

(In millions of U.S. dollars, unless otherwise specified)

article image
Sources: National authorities, and IMF staff estimates and projections.
Table 3.

Azerbaijan: Balance of Payments, 2018–26

(In percent of GDP, unless otherwise specified)

article image
Sources: National authorities, and IMF staff estimates and projections.
Table 4.

Azerbaijan: Statement of Consolidated Government Operations, 2018–26

(In millions of manat)

article image
Sources: National authorities, and IMF staff estimates and projections.

Defined as non-oil revenue minus total expenditure (excl. interest payments) and statistical discrepancies.

Starting in 2021, includes guarantees issued to Aqracredit for its acquisition of distressed assets from the IBA.

Tax revenue excluding AIOC and SOCAR profit tax, and social contributions.

Table 5.

Azerbaijan: Statement of Consolidated Government Operations, 2018–26

(In percent of non-oil GDP)

article image
Sources: National authorities, and IMF staff estimates and projections.

Defined as non-oil revenue minus total expenditure (excl. interest payments) and statistical discrepancies.

Starting in 2021, includes guarantees issued to Aqracredit for its acquisition of distressed assets from the IBA.

Tax revenue excluding AIOC and SOCAR profit tax, and social contributions.

4. A sharp reversal of the current account surplus in 2020 underscored high vulnerability to oil shocks. Oil export revenues declined by 40 percent, and import compression was smaller, given the undiversified structure of the economy (Figure 3, Tables 23). The 9.1 percent of GDP surplus of 2019 turned into a 0.5 percent deficit. FDI, mainly to the oil sector, contracted towards the end of the year. Central Bank of Azerbaijan’s (CBA’s) international reserves increased by 2.3 percent, to $6.4 billion, and valuation gains partially offset the drawdown of SOFAZ assets to finance the budget deficit.4 Together with SOFAZ’s assets of $43.6 billion, external buffers reached 117 percent of GDP at end-2020. Nonetheless, the external position deteriorated in 2020 and is deemed by staff to be substantially weaker than the level consistent with medium-term fundamentals and desirable policies (Annex III).

Figure 3.
Figure 3.

Azerbaijan: External Sector

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: WITS, Haver, and IMF staff calculations.

5. Monetary policy was eased, in the context of the de facto peg. With oil prices plummeting in March-April 2020, pressures on the manat intensified, and SOFAZ expanded foreign exchange (FX) sales (Figure 5). The subsequent rebound in oil prices allowed the authorities to gradually reduce the policy rate, starting in June, by 125 basis points to 6.25 percent by end-2020. In tandem, the CBA tightened reporting requirements for currency conversion in June 2020.5 Although easing from the peaks of March–April 2020, financial conditions remain tighter than before the pandemic, mainly owing to manat real appreciation and sticky real lending rates.

Figure 4.
Figure 4.

Azerbaijan: Fiscal Sector

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, Haver, IMF World Economic Outlook, IMF Public Investment Management Assessment, and IMF
Figure 5.
Figure 5.

Azerbaijan: Monetary and Exchange Rate Developments

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, Haver, and IMF staff calculations.
uA001fig02

Azerbaijan: Financial Condition Index

(Standard Deviations)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Source: IMF staff calculations.Note: The financial condition index is constructed based on a list of financial variables in Azerbaijan (i.e., real short-term interest rates, sovereign spreads, manat lending rate term spreads, FX lending rate term spreads, VIX, change in NEER, and oil price) using principal component analysis.

6. Notwithstanding recent improvements in line with available data, vulnerabilities in the banking sector persist. By the start of the pandemic, bank buffers had recovered from the lows of the 2015–16 crisis. Moreover, in April 2020, CBA approved a package of measures aimed at supporting the banking sector and closed four medium-sized banks with long-standing solvency issues. As a result, the sectoral capital and liquidity ratios edged up in 2020, although improvements have not been uniform. Bank profitability worsened, likely owing to weak credit demand and loan impairment charges (Figure 6). Although FX exposures have declined since 2015–16, several large banks continue to run negative net open FX positions (NOFP). Deposit dollarization continued to decline in 2020 from the highs of the previous crisis but remains high, close to 50 percent, and concentrated in a few banks.

Figure 6.
Figure 6.

Azerbaijan: Banking Sector

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, Haver, and IMF staff calculations.1/ According to the 2015 FSAP report, for statistical purposes only overdue payments are classified as non-performing loans (NPLs). The reported time series of NPLs does not fully reflect actual NPLs.

Outlook and Risks

7. Economic growth is expected to recover in 2021. Staff projects a 2.3 percent increase in real GDP this year, driven by the nonoil sector. The economy is reopening following the start of vaccination in mid-January (with almost 10 percent of population fully vaccinated as of early June). Fiscal policy remains supportive. Increased gas production, owing to the opening of the Southern Gas Corridor (SGC) to Europe, is projected to offset the decline in oil output.

uA001fig03

Azerbaijan: Azeri-Chirag-Gunashli Oil and Gas Production

(In millions of barrels per day)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: British Petroleum and IMF staff calculations.

8. Medium-term prospects remain subdued. Weak nonoil productivity growth compounds the impact of slowing capital accumulation. Scarring from the pandemic including a prolonged decline in demand in contact-intensive sectors such as tourism and transportation, increased unemployment, disruptions in learning, long-term health effects of COVID, and slowdown in investment weighs on the outlook (Table 1). Hydrocarbon output is projected to remain stable as higher gas production compensates for the declining productivity of the aging oil fields.

uA001fig04

Azerbaijan: Potential Nonoil Output Growth

(In percent, HP filtered)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, and IMF staff calculations.
uA001fig05

Azerbaijan: Contribution to Nonoil Real GDP Growth

(Percent, production function asumes 70 percent capital share)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

9. Risks are broadly balanced (Annex IV). The emergence of more dangerous virus strains, and problems in the global rollout of vaccines could delay the global reduction in infections and the recovery, while putting pressure on oil demand and prices. Domestically, the third wave of infections in the spring of 2021 and challenges in vaccination could slow the re-opening of the economy, while faster-than-expected structural decline in productivity could weaken growth. Besides oil price volatility, higher-than-expected Karabakh reconstruction needs and weaknesses in SOE and bank balance sheets present fiscal risks. Policy fragmentation and weaknesses in policy frameworks, if allowed to persist, are another source of downside risks. On the upside, a faster global recovery and higher oil prices could boost growth and improve fiscal and external balances. Over the medium term, a significant boost in structural reform efforts could improve economic prospects.

10. The authorities broadly shared staff’s assessment of risks but were more optimistic about economic growth prospects. While recognizing uncertainty about the future course of the pandemic, the authorities noted that the vaccination is already showing positive results and expressed hope that it would lead to a reopening of the economy and pick-up in activity. The authorities expected that reforms to be designed and initiated in line with the recently launched National Priorities for Socio-Economic Development over the next decade would improve SOE’s performance and boost productivity in the nonoil sector in the coming years.

Policy Discussions

Staff recommended: (i) continuing to support the economy in the near term and phase out the exceptional support gradually, to avoid jeopardizing the recovery; (ii) recalibrating fiscal plans and anchors to ensure long-term sustainability; (iii) aligning the de jure and de facto exchange rate regimes; (iv) fostering sustainable, inclusive, diversified and private sector-led growth; and (v) improving resilience and management of fiscal and financial risks, including contingency planning, to evaluate available policy space.

A. Fiscal Policy

11. The authorities and staff agreed that an accommodative fiscal stance is appropriate for 2021. The 2021 budget implies a significant widening of the deficit, based on a conservative oil price assumption of $40 per barrel. Staff projects a smaller deficit, reflecting a higher assumed oil price. Still, this implies a moderate additional stimulus, with NOPB deteriorating by 1.8 percentage point (pp) of nonoil GDP. This stance is reasonable, given the continued need to fight COVID-19 and support the recovery as well as some fiscal space available in the near term. Support should remain in place until the recovery is firmly underway and after that it should be phased out gradually, to avoid cliff effects. Health spending needs to be sufficient to ensure a successful vaccine rollout and defeat the pandemic.

12. All recognized that fiscal policy should remain flexible. The authorities and staff agreed that if downside risks were to materialize, automatic stabilizers should be allowed to operate and additional, targeted on-budget fiscal support should be provided, as necessary, within the available fiscal space (Annex IV). Staff recommended saving the windfall gains from the oil price exceeding the 2021 budget assumptions and reallocating instead spending to cover additional needs, to limit a further deterioration in the long-term fiscal position and buffers. However, the authorities signaled their intention to revise spending up in the near term to cover additional reconstruction and vaccination needs, as well as subsidies to agriculture in response to the increase in diesel tariffs in early 2021.

13. Staff emphasized that weakening long-term fiscal sustainability requires formulating credible reform plans. Deficits are expected to persist (Tables 45), as oil prices are projected to remain below fiscal breakeven levels and the reconstruction needs of the Karabakh region – not yet costed but potentially high – will be largely met from public sources (the 2021 budget allocated 2.5 percent of GDP to reconstruction). Moreover, oil revenues as a share of oil GDP have been historically well above nonoil revenues as a share of nonoil GDP, and under the current tax structure, the projected decline in the share of oil GDP in total GDP will also lead to a decline in total revenue as a share of GDP (Table 5). Thus, declining oil revenues would require increasing effective taxation of the nonoil sector in the coming years. High oil dependence (oil revenues account for half of total government revenues) and sizeable contingent liabilities underscore the need for adequate precautionary buffers. A large gap of 21 pp of nonoil GDP from the long-term Permanent Income Hypothesis (PIH)-consistent NOPB benchmark of 9 percent of nonoil GDP implies that, on current policies, the fiscal position is not sustainable in the long run.

uA001fig06

Azerbaijan: Fiscal Dynamics

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, and IMF staff calculations.1/ The gap between actual nonoil primary balance and non-oil primary balance recommend by the Permanent Income Hypothesis. The higher the value the greater the effort needed to restore intergenerational equity.

14. There was consensus that significant, multi-year fiscal adjustment would be needed to improve intergenerational equity and reduce external misalignment. The authorities considered that such consolidation would also support the de facto currency peg. Staff projected in the baseline that consolidated government’s net financial assets (NFA) would decline to 7.1 percent of GDP by 2030 and the PIH gap would reach 14 pp of non-oil GDP in 2030. The baseline assumed, in line with the authorities’ previously announced plans, that NOPB would improve by 1 pp of nonoil GDP per year.

15. The authorities and staff agreed that, given the continued weakening of the fiscal position, a sustained consolidation would be needed after the recovery from the pandemic takes hold. Based on domestic and international experiences,6 staff recommended a steady NOPB improvement of 1.5 pp of nonoil GDP per year on average during 2022–30. Such an adjustment would strike an appropriate balance among improving long-term fiscal sustainability, addressing development needs, and mitigating the adverse impact on the economy.7 The authorities generally agreed. They explained that, although their preliminary plans assumed medium-term consolidation (especially in 2022 as the pandemic spending was expected to unwind), they were considering the possibility of spending a (still to be identified) portion of windfall gains on reconstruction. Staff emphasized the importance of developing credible medium-term fiscal plans, including identifying consolidation measures, and evaluating their impact on the economy.

uA001fig07

Azerbaijan: Financial Assets

(In percent of GDP)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities and IMF staff calculations and estimates. Notes: Baseline scenario assumes the nonoil primary balance increases on average by about 1 pp per year during 2022–30. Adjustment scenario assumes the nonoil primary balance increases on average by about 1.5 pp per year during 2022–30.

Azerbaijan: Fiscal Scenarios, 2021–26

(In percent of nonoil GDP, unless stated otherwise)

article image
Sources: National authorities and IMF staff calculations and estimates.

Starting in 2021, includes guarantees issued to Aqracredit for its acquisition of distressed assets from the IBA.

16. The composition of fiscal consolidation should be designed with a view to alleviating its likely adverse impact on economic growth while mobilizing nonoil revenues. In line with the authorities’ policy plans, staff’s baseline assumed that consolidation would be achieved mainly through current expenditure restraint and rationalization, resulting in slower growth of spending on wage bill, goods and services, and transfers, while pandemic-and reconstruction-related spending declines gradually. In the adjustment scenario, staff recommended the following fiscal reforms to achieve larger consolidation in a growth-friendly manner:

  • Broad-based improvements in spending efficiency (especially for infrastructure where high quality was achieved at a relatively high cost in the past) through expenditure reviews, performance budgeting and enhancements in the public investment management framework and governance. Improvements in investment efficiency would also improve productivity.

  • A gradual reduction of the wage bill as a share of GDP (following the 40 percent rise in public sector wages in 2019, the public wage bill increased by 3 pp of GDP in the last two years to 9 percent of GDP, the second highest in the region). Anchoring wage increases on labor productivity 2 is essential to prevent the loss of competitiveness. Public employment is among the highest in EMDCs and some gradual downsizing (for example, through attrition and the reduction of “ghost” employees8) could be another source of savings.

  • Phasing out generalized subsidies (which account for 15 percent of GDP, mostly owing to below-cost recovery domestic prices for energy products) would reduce the need for fiscal support to the SOEs,9 improve fairness, and help “green” the economy by containing energy consumption and green-house-gas (GHGs) emissions. In January 2021, the authorities already raised prices for lower-octane gasoline (by 11 percent), diesel (33 percent), and water and sewage fees (100 percent) and deregulated prices of some oil products. The timeline and scope of future subsidy reforms would need to be identified, in the context of broader SOE reforms and with adequate social protection to the vulnerable groups of population (see below).

  • Revenue mobilization measures and tax reforms, focusing on personal income taxes (no new exemptions, progressive taxation), modernizing property taxes (market value assessments), and reducing tax expenditures from extensive exemptions (appending tax expenditure estimates to the budget) would also support fiscal consolidation (Annex V). While pointing to room for revenue mobilization using the latter two measures over the medium term, staff recognized that any personal income tax reforms would need to be delayed until after the seven-year sunset clauses of the 2019 reforms expire.10

  • The above reforms should create space for ensuring appropriate social spending (on health, education, and support of the vulnerable).11 This is critical for promoting inclusive growth and raising productivity, against the backdrop of the pandemic-induced increases in unemployment (Figure 2), poverty and scarring. Another benefit is facilitating subsidy and SOE reforms, and transformations caused by diversification, digitalization and climate change.12 Staff recommended reform priorities centered on improving the adequacy of social assistance in the poorest quintiles by increasing the size of benefits paid to the poor (Figure 7, Table 8). Extending social insurance programs to include self-employed and informal workers is also important, as the COVID-19 experience demonstrated. Efforts to improve the efficiency and effectiveness of social spending need to continue, through better targeting and monitoring. Recent improvements in the e-government systems (including e-transfers of cash to the unemployed during the pandemic and the “ASAN” one-stop shop for public e-services) and the centralization of social contributions for benefit programs are welcome in this regard.

Figure 7.
Figure 7.

Azerbaijan: Social Spending

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: ASPIRE, National authorities, IMF (2019), IMF EAT, UNWHO, World Development Indicators, World Economic Outlook, and IMF staff calculations.1/ Education spending (from the WB and UNESCO) includes current and capital spending as well as transfers. Health spending (from the WHO) includes internal transfers and grants, transfers, subsidies to voluntary health insurance beneficiaries, NPISH or enterprise financing schemes as well as compulsory prepayment and social health insurance contributions, only on current expenditures. For more details on the definition of social spending, see the IMF’s 2019 Strategy on Social Spending, available at www.imf.org. Social Protection and Security data is sourced from the April 2021 World Economic Outlook. Data for EM and CCA are the latest pre-pandemic data available (mostly 2018).2/ The CCA average excludes Azerbaijan.3/ Data for Azerbaijan refer to the 2021 approved budget.
Table 6.

Azerbaijan: Summary Accounts of the Central Bank, 2018–26

(In millions of manat, unless otherwise specified)

article image
Sources: National authorities, and IMF staff estimates and projections.

Includes CBA holdings of Aqrarcredit’s bonds as a part of the SPV, and IBA deposits.

Table 7.

Azerbaijan: Monetary Survey, 2018–26

(In millions of manat, unless otherwise specified)

article image
Sources: National authorities, and IMF staff estimates and projections.
Table 8.

Azerbaijan: Inclusive Growth Indicators

article image
Sources: IMF World Economic Outlook, World Bank, World Economic Forum, International Labour Organization, Transparency International, UNDP, Oxfam International.

Indicators use official sources and surveys to summarize perceptions of the quality of governance and business environments.

uA001fig08

Azerbaijan: Quality of Infrastructure and Capital Expenditure

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: World Economic Forum Global Competitiveness Report 2019, World Economic Outlook, and IMF staff calculations.
uA001fig09

Azerbaijan: Wage Bill and Public Employment

(In percent)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, IMF Expenditures Assessment Tool, and IMF staff calculations.Note: Comparison of public wage bill faces challenges due to differences in definitions, coverage, and data availability across countries.
uA001fig10

Azerbaijan: Post-tax Subsidies, 2017 1/

(In percent of GDP)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: IMF FAD Energy Subsidies, and IMF staff calculations.1/ Post-tax energy consumer subsidies exist if the consumer prices are below the supply costs plus the efficient levels of taxation.

17. The authorities expected consolidation to be largely expenditure-driven and concurred it should be growth-friendly. They have not yet identified specific policy measures to support their consolidation plans but indicated their intention to focus on improving spending efficiency while enhancing cost recovery of SOEs and slowing public wage bill growth. While broadly agreeing with staff’s recommendations concerning revenue mobilization, the authorities noted that they were proceeding with estimation of tax expenditures and planned to review specific tax reform options proposed by the IMF’s technical experts. The authorities did not see social spending reforms as an immediate priority, pointing out that social spending per capita was higher in Azerbaijan than in most regional peers while poverty levels were lower than in global peers.13 Moreover, social spending increased significantly in recent years on the heels of increases in wages and other payments (introduced in part to offset the impact of past devaluations). Like staff, the authorities considered it important that any future wage increases are aligned with productivity growth.

18. A well-designed fiscal anchor would enhance policy credibility and long-term fiscal sustainability while reducing procyclicality. Following the temporary suspension of the fiscal rule during the pandemic, the authorities have been working to address the design issues and plan to reintroduce the rule in 2022, while updating the medium-term expenditure framework (MTEF) and Debt Management Strategy.14 They emphasized their intention to limit future general government debt issuance and keep the debt ratio (excluding Aqrakredit guarantees) below 20 percent of GDP. As the authorities are considering new parameters for their fiscal rule, staff recommended setting the floor for the average annual NOPB improvement at 1.5 pp in 2022–26 and the general government debt target at 40 percent of GDP (including Agrarkredit guarantees), with the ceiling at 45 percent, to provide buffers for adverse shocks and contingent liability risks. Staff did not recommend setting an operational target for the stabilization portion of SOFAZ assets, although estimating the value (of SOFAZ assets)-at-risk would be useful. To complement the fiscal rule, consideration could be given over the medium term to developing an asset-liability management framework, which would help evaluate risk-return trade-offs of public financial assets and liabilities in an integrated manner. The authorities and staff agreed that deeper public financial management reforms, including of the MTEF, monitoring, and reporting of fiscal performance, are important to support the implementation and credibility of the rule. Further building of the Ministry of Finance’s macro-fiscal capacity would ensure quality forecasts and timely data and support reporting and compliance with the rule.

19. Better risk management, transparency and reporting need to accompany fiscal reforms. Some large SOEs have been operating at a loss, owing to operational inefficiencies, outdated equipment, and low regulated prices. Improving SOE profitability would reduce the drain on public resources and increase room for productive spending. This requires reforms to enhance SOE operational efficiency as well as financial control and reporting, alongside gradually eliminating generalized subsidies. Better monitoring and management of risks from SOE guaranteed debt is also crucial.15,16 In this context, the authorities emphasized progress in recent SOE reforms, including the establishment of the Azerbaijan Investment Holding, and performance indicators and corporate governance guidelines for SOEs. Welcoming these important reforms, staff recommended also putting in place a comprehensive framework for fiscal risk management,17 accompanied by increased transparency and timely publication of data and statements on fiscal risks and SOE performance. In addition, continued efforts to ensure transparency of government spending, especially for COVID-19 and reconstruction, are important, including through publication of the ex post budget execution reports and audits (as outlined in the Law on Budget Systems of Republic of Azerbaijan). SOEs reforms and increased transparency of public spending will also help lessen vulnerabilities to corruption.

Azerbaijan: Financial Indicators for Utility and Railway SOEs, Latest Available

article image
Source: Staff estimates based on financial statements of the companies. Notes: For Azerenerji and Azersu the figures are for 2019, while for Azerbaijan Railways the figures are for 2017 (the latest year for which information is publicly available).

B. Monetary and Exchange Rate (ER) Policy

20. Although monetary policy is subordinated to the de facto peg, staff saw room for further easing. Despite recent interest rate cuts, financial conditions were tighter than prior to the pandemic, real interest rates for manat loans remained in double digits, and the spread between CBA and Fed rates was sizeable. Low financial integration limited short-term capital flows. Inflation was projected to increase in 2021 to close to the mid-point of CBA’s band, mainly owing to one-off increases in administered prices and higher global food prices; core inflation remains below 3 percent on a cumulative y/y basis in 2021. With the output gap still negative, slow money growth, and nominal effective ER appreciating, this increase was expected to be short-lived. Inflation expectations were well anchored, and risks to inflation were balanced. Against this backdrop, consideration could be given to further easing, in staff’s view, to support the nascent recovery, especially if downside risks were to materialize. While recognizing that room for easing might exist to support the recovery, CBA considered the current monetary stance appropriate, since its mandate focused on price stability and the outlook for inflation was uncertain. CBA intended to continue to reassess the outlook regularly by reviewing monetary and nonmonetary factors in a forward-looking manner.

21. Looking ahead, staff advocated a gradual move to a more flexible ER regime, to facilitate adjustment to oil price shocks and foster market development. The de facto peg has been effective in keeping inflation under control in Azerbaijan’s undiversified economic setting where the ER is the main channel for the transmission of inflationary shocks. However, the peg limits CBA’s ability to provide monetary support to the economy in times of stress and puts a higher burden on fiscal policy. Maintaining competitiveness is challenging under the peg since Azerbaijan’s main trading partners have more flexible regimes. Greater ER flexibility would support diversification and encourage the development of financial markets, and better management of currency risks. It is preferable to introduce increased ER flexibility from the position of strength, while external buffers are projected to remain sufficient in the coming years. Although currency mismatches in the banking and corporate sectors have declined since the 2015–16 crisis and appear manageable, continued efforts to de-dollarize the economy are still needed, including by improving policy credibility and possibly widening the differential in reserve requirements for FX and manat deposits. In this context, staff underscored risks to policy credibility stemming from the difference in Azerbaijan’s de jure and de facto ER regimes and from framing policy communication in inflation-targeting (IT) terms, urging the authorities to address these inconsistencies.

22. While reaffirming IT as their long-term goal, the authorities believed the de facto peg to be optimal for the foreseeable future. They believed that the manat was at a fair value but recognized that their empirical analysis was backward-looking and country-specific, in contrast to staff’s forward-looking, PIH-based, and multilaterally consistent EBA-Lite framework (Annex III). In any event, CBA saw fiscal and structural policies as the main policy levers for addressing any potential competitiveness issues. Regarding the difference between the de facto and de jure ER regimes, CBA did not see the need for an official change in the ER regime on the grounds that market participants and the public understood well CBA’s policy of keeping the ER pegged to the dollar. Staff advised periodically reviewing the appropriateness of the peg, to assess the evolving balance between its costs and benefits, and developing contingency plans, in the event the country is hit by a large and prolonged adverse shock, for example, to oil prices. All agreed that ongoing efforts to improve policy coordination (particularly, the recent establishment of an Economic Council), CBA liquidity forecasting and research capacity, and communication18 as well as initiatives to strengthen CBA independence in monetary policy making could help CBA enhance its role and lay the ground for an eventual transition to IT. Alongside, stepped up efforts to improve the functioning of interbank, money and FX markets, and monetary transmission are needed to support financial market development and transition to the new economic growth model.

C. Financial Sector Policy

23. The authorities and staff agreed that prudential policies need to pivot carefully from incentivizing banks to use their buffers to rebuilding these buffers and reducing problem assets over time. While reversing the exceptional prudential easing prematurely may adversely affect bank soundness indicators, there are also important risks to delaying the reversal of such forbearance. In particular, this could mask the growing vulnerabilities in the banking sector, ultimately undermining credit quality and market discipline, putting future stability at risk. Given the still outstanding restructured loans portfolio in the banking sector, staff recommended that CBA uses the available breathing space to ascertain implications of the support measures and the likely impact of their reversal on the recovery and bank balance sheets strength. Such evidence-based analysis, in tandem with data reporting transparency and close communication with financial market participants, would ensure a more targeted and time-bound plan to phase out financial sector support measures.

24. In this context, the authorities indicated their intention to start with measures that are less binding for the banking sector and phase out other measures as the recovery solidifies. Staff welcomed the recent replacement of a blanket deposit insurance, in place since the previous banking crisis, with a limited guarantee program, and a list of tightening measures that have been implemented or on the schedule. Rolling back exemptions on borrowers’ credit histories soon would help mitigate the adverse impact on lending standards. The risk weight on mortgage loans, reduced to 50 percent last year, should be complemented by stronger prudential measures (i.e., LTV requirement in line with Basel III standards) to prevent deterioration in the quality of the mortgage portfolio.

25. Staff saw strengthening financial oversight as an important complement to medium-term reforms aimed to improve access to finance (Annex VI). The latter is essential for transition to the new growth model, as financial development and access to finance have been declining since the 2015–16 crisis and lag peers. However, past credit expansions were driven largely by weaker banks, which may cause bank distress and output loss. Stronger supervision and regulation are needed to prevent the repeat, including strengthening financial oversight of both lenders’ risk-taking activities and borrowers’ credit risk.

26. Staff’s and the authorities’ stress-testing exercise confirmed that systemic banks are resilient against credit risks yet vulnerable to direct and indirect FX risks. The authorities agreed that some vulnerabilities in the banking sector exist: the restructured loan portfolio quality was deteriorating, with some systemic and mid-sized banks remaining exposed to the FX risks mostly due to their outstanding net open FX positions, and bank profitability being still subdued as a consequence of weak credit demand and increased loan impairment charges. However, the authorities reckoned that these vulnerabilities were unlikely to pose a serious threat to banking stability. The IMF staff expressed its readiness to continue supporting CBA in their important efforts to enhance risk-based supervision and stress-testing.

uA001fig11

Azerbaijan: Capital-to-Asset Ratio and Loan Growth

(Average, In Percent)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities and IMF staff calculations.Note: Banks with fast (slow) loan growth are the five banks with the highest (lowest) annual changes in gross loans.
uA001fig12

Azerbaijan: NPL Ratios for Business Loans in Systemic Banks

(In Percent)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, and IMF staff calculations.Note: The chart presents estimates of NPL ratios for business loans in four systemic banks. The baseline stress scenario uses staff’s baseline macroeconomic forecast. The adverse scenario assumes that the nonoil real GDP growth rate is 2 percentage points below the baseline forecast during 2021–23.

27. Staff also emphasized that greater CBA independence and transparency are also essential for its ability to fulfill its mandate. Amendments to the CBA Law are needed to increase its independence not only in monetary policymaking but also financial supervision. Staff welcomed the authorities’ plans to publish the Financial Stability Report this year, including financial soundness indicators and stress tests results. Staff urged CBA to encourage banks to share more details on the ultimate beneficiary ownership and enhance fit and proper tests for those who own and control banks, to better assess shareholders’ financial strength and sources of funding. Staff also recommended that the authorities continue making progress in enhancing the effectiveness of the AML/CFT framework, including in preparation for the upcoming MONEYVAL assessment and to mitigate potential pressure to correspondent banking relationships. Preventive measures are important in this context, including in relation to domestic politically exposed persons and risk-based supervision for banks, as well as bolstering entity transparency through the swift adoption and operationalization of the national registry for beneficial ownership, and enhancing prosecution, conviction, and confiscation of ill-gotten proceeds.

D. Structural Policy

28. Deep-rooted impediments hinder private sector growth and job creation. They include weak governance and institutions, limited SME access to credit; inefficient labor, product, and service markets; and sizeable skill gaps. The combined result is high informality, limited nonoil FDI, weak entrepreneurship and SME development, and slow productivity growth. Scarring from the pandemic (including increased insolvency, unemployment, learning losses, etc.) needs to be addressed. Improving inclusiveness requires expanding economic opportunities for women and the young. Continued digitalization of the economy, building on broad access to internet and expanded e-government services, is important for raising productivity growth. Policies also need to address climate change, increase green investment, and improve energy efficiency.

uA001fig13

Azerbaijan: Individuals using the Internet

(In percent of population)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: The Global Carbon Project (2020), The World Bank World Development Indicators, and IMF staff calculations.
uA001fig14

Azerbaijan: CO2 Emissions

(In Millions of Tons)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

29. Designing and implementing comprehensive reforms in line with the recently released national priorities would go a long way towards promoting sustainable, inclusive, and green growth over the longer term. Among other things, the recently created Azerbaijan Holding Company is expected to play a key role in improving corporate governance and efficiency of SOEs. Applying the Anti-Monopoly law would lower entry barriers and improve market efficiency. Identifying ways to improve competition in the financial sector, where net interest margins appear high by comparison to peers, could help improve access to finance. The WTO accession would expose domestic producers to international competition, encouraging improvements in productivity and diversification. Developing the green growth agenda, as envisaged under the National Priorities, is an important task, alongside reforming generalized energy subsidies. Applying the OECD’s Investment Guidelines could help attract FDI and know-how, while minimizing tax expenditures. Following up on the most macro-critical aspects of the governance recommendations outlined the 2019 Article IV report remains critical and would augment the impact of other structural reform efforts.

30. The authorities concurred with staff on the need to accelerate structural reforms with a view to raising nonoil growth and fostering economic diversification. The authorities noted that the recently developed broad-ranging reform agenda, embedded in their ten-year development plan, focuses on improving corporate governance of SOEs, and later their productivity and efficiency, through the creation of Azerbaijan Investment Holding. They saw recent deregulation of prices for utilities and oil products as a step towards greater market price determination and reduction of government subsidies. At the same time, the authorities believed that tax incentives would boost private sector investment in the nonoil sector, employment, and export diversification.

Staff Appraisal

31. Azerbaijan faced an unprecedented challenge in 2020, but timely and appropriate policy response mitigated the impact of the pandemic on the economy. A significant relief package helped cushion the impact of the pandemic and the decline in oil prices. With vaccination progressing and lockdown measures gradually eased, the economy has started to recover. The near-term challenge is to properly calibrate the phasing out of the COVID-19-related stimulus, balancing the need to avoid endangering the recovery while preparing the ground for addressing structural challenges. In the long term, Azerbaijan’s running out of oil in the coming decades poses the fundamental challenge as the economy needs to wean off its dependence on oil revenues.

32. While fiscal policy has been appropriately eased during the pandemic, a growth-friendly fiscal consolidation needs to start in 2022. The accommodative fiscal stance in 2021 is appropriate and, in the case of adverse shocks, automatic stabilizers should be allowed to operate and additional, targeted on-budget fiscal support should be provided, as necessary, within the available fiscal space. Windfall revenue from higher-than-budgeted oil prices should be saved. Looking ahead, continued expansionary stance would not be compatible with preserving the de facto peg and fiscal sustainability and would also make it more difficult to attain intergenerational equity in the face of a projected exhaustion of oil resources. Thus, consolidation should start in 2022 and be sustained over the long term, to gradually achieve a significant reduction in the nonoil primary deficit. While the authorities see a key role for expenditure consolidation, measures to improve nonoil revenues will eventually have to contribute to adjustment. Consolidation needs to be accompanied by measures to protect the vulnerable and support education and health, with a view to enhancing inclusiveness and productivity. Stepped structural and financial sector reforms are also critical, to mitigate the adverse impact of consolidation on economic growth.

33. Planned fiscal adjustment should be supported by the revised fiscal rule. In response to the outbreak of the pandemic, the authorities have temporarily suspended the fiscal rule. They intend to revise and re-launch it in 2022, using the nonoil primary balance and debt-to-GDP targets to anchor fiscal policies. Public financial management reforms, including of the medium-term fiscal framework, fiscal risk management and transparency, are needed to support the implementation of the rule and its credibility. Establishing a solid track record of the rule implementation is particularly important given its suspension in 2020 during the pandemic, a year after its introduction.

34. Further monetary easing could be considered to support the nascent recovery, especially if downside risks were to materialize. Although CBA has reduced its policy rates since the outbreak of the pandemic, financial conditions remain tighter than when the pandemic started. Despite the de facto peg, low financial integration and a large spread over the Fed rates provide room for further rate cuts. In addition, following a recent temporary uptick, inflation is projected to remain close to the mid-point of the CBA target range.

35. A gradual move to a more flexible ER regime should be considered in the medium term. The de facto peg has been effective in keeping inflation under control, but as Azerbaijan’s main trading partners have more flexible regimes, maintaining competitiveness is challenging under the peg. A more flexible exchange rate would facilitate adjustment to oil price shocks, reduce adjustment burden on fiscal policy, and support diversification of the economy and market development. The current difference between the de jure flexible exchange rate regime and the de facto pegged one and the framing of policy communication in inflation-targeting terms poses risks to policy credibility and needs to be addressed promptly. More broadly, the external position is deemed to be substantially weaker than the level consistent with medium-term fundamentals and desirable policies (Annex III).

36. Targeted and time-bound plan to phase out exceptional prudential easing introduced in response to the pandemic is needed. The authorities need to balance the risk to bank soundness arising from prematurely phasing out the forbearance measures and the risks from delaying the reversal to credit quality and market discipline. Staff welcomes the recent replacement of a blanket deposit insurance, in place since the previous banking crisis, with a limited guarantee program, a list of regulatory tightening measures that have been already implemented or on the schedule, and the efforts in improving financial stability analysis capacity. In the medium term, strengthening financial oversight of both lenders’ risk-taking activities and borrowers’ credit risk is an important complement that is needed to ensure financial stability and improve access to finance. The authorities should also continue making progress in enhancing the effectiveness of the AML/CFT framework.

37. Impediments to private sector growth and job creation need to be tackled to support private sector development and diversification of the economy. Weak institutions, limited SME access to credit; inefficient labor, product, and service markets; and sizeable skill gaps result in high informality, discourage nonoil FDI and slow productivity growth. Addressing these weaknesses, including by strengthening the governance and anti-corruption frameworks, attending the scarring impact from the pandemic, and pursuing the green agenda envisaged under the Azerbaijan 2030 Priorities, are needed to achieve inclusive and green growth. The recently released national priorities set ambitious goals for social and economic development and would need to be supported by well-designed and consistently implemented policy reforms.

38. It is recommended that the next Article IV consultation takes place on the standard 12-month cycle.

Annex I. Response to the Pandemic

1. The 2020 recession was driven by the oil sector, which contracted by 6.4 percent owing to OPEC+ cuts on top of a gradual structural decline in oil production. The impact of the pandemic on the nonoil output was smaller (3 percent decline), broadly in line with that in global and regional peers. The pandemic mostly affected contact-intensive sectors, especially hospitality, which contracted by 59 percent but accounts for a small share of the economy. Construction and trade activity also declined (by 8.8 percent and 1.3 percent respectively). Given high labor intensity of contact-intensive sectors, unemployment rose from 4.8 percent in 2019 to 7.2 percent in 2020, and so did poverty. The large informal sector was also severely affected. However, manufacturing, transportation, and agriculture proved to be resilient, expanding by 10.4 percent, 4.6 percent, and 2 percent, respectively.

2. A sizeable fiscal relief package, amounting to 4.8 percent of the 2020 GDP, cushioned the impact of the pandemic. Measures were targeted to the affected sectors, firms, and individuals. Main instruments included loan guarantees, subsistence cash transfers to the unemployed and temporary public jobs, salary payments and support to microentrepreneurs, and support to passenger transportation (mostly the national airline). Spending and tax relief measures terminated at the end of 2020. The package was financed mainly through the budget, by reallocating spending and increasing transfers from SOFAZ. Some additional financing came from the Solidarity Fund, which included private donations.

uA001fig15

Azerbaijan: Fiscal Support in Response to COVID-19

(In percent of 2020 GDP, cumulative since March 2020)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: IMF COVID-19 Tracker and IMF staff calculations.Notes: Includes announced discretionary fiscal packages including both above and below the lines support.

3. The coverage of the package, at one fifth of the population, was broadly adequate. This is reflected, among other things, in its substantial uptake by end-2020. (The duration of some measures (especially social support) was extended when the pandemic turned out to be more prolonged than originally expected.)

COVID-19 Relief Measures

(In percent of GDP)

article image
Sources: National authorities and IMF staff calculations.

Mainly the use of the reserves of the State Social Protection Fund to sustain social payments (pensions and benefits) to the population in the face of reduced insurance contributions.

4. Cash transfer programs helped people to cope with the crisis. The largest cash transfer program to address the pandemic consisted of lump-sum subsistence payments to unemployed workers, including from the informal sector, who registered with the State Employment Service. The distribution of support benefitted from the recently introduced digital social systems, which were expanded during the pandemic. Although the amount of the cash transfer per individual appeared reasonable by historical standards, the principal cash assistance to the unemployed replaced only 27 percent of average monthly income, which is somewhat low by international comparison. Additional social assistance for the most vulnerable persons during the pandemic was meagre (less than 0.1 percent of GDP). Even prior to the pandemic, Azerbaijan’s social assistance expenditure and its coverage of the poorest quintile were well below the average for EM economies. Unemployment insurance remains a small program with tight eligibility criteria.

5. Financial sector policies also helped cope with the crisis. The impact on the financial sector was mitigated by a gradual reduction of the CBA policy rate, easing requirements on bank capital buffers and loan loss provisions, loan forbearance on interest payment and credit histories, and a temporary suspension of on-site inspections. Borrower support programs accounted for a significant portion of loan portfolios in the affected sectors. The scope for using liquidity support was limited by the de facto peg.

uA001fig16

Azerbaijan: Financial Measures

(Proportion of countries implementing measure)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: IMF Covid Policy Tracker, National authorities and IMF staff calculations.

Annex II. Debt Sustainability Analysis (DSA)

Figure A.II.1.
Figure A.II.1.

Azerbaijan: Public Sector Debt Sustainability Analysis (DSA)

(In percent of GDP unless otherwise indicated)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Source: IMF staff projection.1/ Public sector is defined as general government.2/ Based on available data.3/ EMBIG.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 = 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 – π (1+g) and the real growth contribution as -g.7/ The exchange rate contribution is derived from the numerator in footnote 5 as ae(1+r).8/ Includes asset changes and interest revenues (if any). For projections, includes exchange rate changes during the projection period. In 2021, reflects inclusion to debt of Aqracredit guarantee, and drawdown of SOFAZ assets in 2022–26.9/ 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.
Figure A.II.2.
Figure A.II.2.

Azerbaijan: Public DSA – Composition of Public Debt and Alternative Scenarios

(In percent of GDP unless otherwise indicated)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Source: IMF staff projection.1/ Contingent liability shock refers to standard financial sector liability shock with one-off increase in non-interesat expenditures equal to 10 percent of banking sector assets; SOE/Financial sector shock combines financial sector and SOE contingent liability shocks, with non-interest expenditures increased by 20 percent of SOE guarantees excluding Aqracredit guarantees.

Annex III. External Sector Assessment

Staff assesses the external position to be substantially weaker than the level consistent with medium-term fundamentals and desirable policies. Amid the pandemic, a decline in oil export revenues caused a sharp deterioration in the external position in 2020. While overall external buffers remain high, the current account (CA) balance is significantly weaker than the level needed to ensure sufficient savings for future generations.

1. A large CA surplus turned into a small deficit in 2020. Following three years of surpluses, the CA reversed from 9.1 percent of GDP in 2019 to -0.5 percent of GDP in 2020 owing to a sharp decline in global oil prices and oil production cuts under the OPEC+ agreements in the midst of the pandemic. A recovery in oil prices in 2017–19, coupled with a reduction in oil-related construction and business services (as many projects were completed), led to a rebound of the CA to a surplus averaging 8.7 percent of GDP in the 2017–19 period. The nonoil CA deficit remains stubbornly high, at 12.5 percent of GDP, driven by large nonoil imports. Nonoil exports are low at 4 percent of GDP in 2019 and 2020. The CA is expected to swing back to a surplus of 2.4 percent of GDP in 2021 and then to gradually decline in tandem with oil export volumes.

uA001fig18

Azerbaijan: Real Effective Exchange Rates and Contributions to Current Account

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, IMF World Economic Outlook, and IMF staff calculations.

2. External reserve coverage remains more than adequate. Despite the CA deterioration, CBA’s gross international reserves increased moderately, reaching $6.4 billion (15 percent of GDP, 4.5 months of next year’s imports of goods) by end-2020. CBA’s reserves are complemented by the liquid part of foreign assets held by the Oil Fund (SOFAZ). This liquid part, mostly fixed income instruments, equities and gold, represents about 75 percent of the $43.5 billion in assets (102 percent of GDP) at end-2020. Together, CBA and SOFAZ liquid assets are equivalent to 25 months of next year’s imports. Azerbaijan scores much better than most EMs on all criteria of the IMF’s Assessing Reserve Adequacy (ARA) metric, which combines information on exports, broad money, short-term debt and other investment liabilities (yellow diamonds lower left chart). This is the case even after adjusting the ARA metric for the country’s heavy reliance on oil exports (green diamond in same chart). Reserve coverage scores are projected to remain strong into 2026, given expected CA surpluses, rising reserves, and limited increases in external debt (red diamonds).

3. The external position has substantially weakened in 2020 compared to the level implied by medium-term fundamentals and desirable policy settings.

  • The EBA-Lite commodity approach1 is preferable for assessing the CA position in Azerbaijan because it is based on the permanent income hypothesis (PIH) and reflects the fundamental goal of achieving intergenerational equity in this country with nonrenewable resources. Assuming the annuity is constant in real terms, the CA norm is estimated at 18.5 percent of GDP in 2020, implying a negative CA gap of 16.1 percent of GDP.2 The gap is expected to remain negative over the medium term at 11.2 percent of GDP and will require sustained fiscal consolidation to narrow it further over time.

  • The current account model, which estimates an equilibrium level of the CA consistent with the gamut of structural and policy factors in the IMF’s multilaterally consistent External Balance Assessment-Lite framework, also suggests that the CA was weaker than fundamentals in 2020. The CA norm is estimated at 5.7 percent of GDP, implying a negative gap of 1.8 percent of GDP, after adjusting for the temporary impact of COVID-19. Policy changes that would reduce the CA gap over the medium term include: (i) deeper and sustained fiscal consolidation than assumed in the baseline; (ii) financial deepening, for bank credit to the private sector to rise closer to the EM average; (iii) an increase in public health expenditures from current 2 percent of GDP by 0.5 percent of GDP in the medium term, closer to the EM average while improving spending efficiency.

uA001fig19

Azerbaijan: Forecasted CA and CA Norm under Bems-Carvalho (Oil Resource) Adjustment

(In percent of GDP)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Source: IMF staff calculations.

4. The manat continued to appreciate in 2020. It has been de facto pegged to the dollar at a rate of 1.7 since April 2017. The real effective exchange rate (REER) was 12 percent stronger than 2017. The recent appreciation has been driven mostly by the U.S. dollar nominal appreciation vis-à-vis trading partners’ exchange rates (particularly Russia and Turkey). The current account and REER models suggest that the exchange rate is significantly overvalued, with the REER gap in a range of 6.0 to 6.8 percent, consistent with the negative CA gap over the medium term. The permanent income model could indicate a much larger REER gap, in line with the larger CA gap implied by a constant real annuity.

5. Overall, staff assesses the external position to be substantially weaker than the level consistent with medium-term fundamentals and desirable policies. This assessment is based on the results above, giving the relevance of the external sustainability approach based on the PIH. Specifically, staff assesses the CA gap to be -11.2 percent of GDP, based on the constant real annuity estimated by the EBA-Lite commodity approach, and the medium-term REER gap to be at least 6.0 to 6.8 percent, also considering the current account and REER models.

6. Azerbaijan has been a net supplier of capital. Capital and financial accounts are generally open but domestic financial markets are underdeveloped compared to those in regional and global peers. Capital inflows are dominated by FDI, mostly to the oil sector, and portfolio inflows, while outflows include trade credits, bank outflows, and outward FDI, mainly by the national oil company. Risks of increased bank outflows are mitigated through administrative and reporting requirements that were introduced in response to the 2015–16 crisis and largely remain in effect.3

Figure A.III.1.
Figure A.III.1.

Azerbaijan: Exchange Rates and Reserve Adequacy

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, Haver, IMF FAD Reserve Adequacy Tool, and IMF staff calculations.Note: The data for EM refer to a sample of 47 emerging markets.

Annex IV. Risk Assessment Matrix

article image
article image
article image
article image

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 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 between 30 and 50 percent). 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.

Annex V. Priorities for Tax Reforms

As Azerbaijan strives to reduce the dependence of its economy on oil production and informal activity, a greater tax effort will be required to sustain its public finances. The groundwork for formulating and administrating future tax reforms must be laid as soon as possible.

1. With oil production on a declining trend, mobilization of nonoil tax revenue is critical for strengthening long-term fiscal sustainability. Nonoil tax revenue as a share of nonoil GDP has not increased to compensate for the gradually declining importance of oil revenues. To maintain consolidated revenues at 30 percent of GDP beyond 2026, nonoil tax effort will need to increase by 1.3 percent of nonoil GDP in 2027, and gradually rise by 10.8 percent of nonoil GDP in 2040. Clearly, as Azerbaijan transitions to a more diversified economy, additional revenues will need to be mobilized from the nonoil sector.

uA001fig20

Azerbaijan: Oil Revenues and Non-Oil Tax Revenues

(In percent of GDP, unless stated otherwise)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities and IMF staff calculations.

2. Compared to its peers, Azerbaijan has higher rates for VAT, comparable for CIT, and lower for PIT. The VAT rate is 18 percent with registration obligatory for taxpayers with annual turnovers exceeding AZN 200,000 ($120,000). The CIT rate is 20 percent, while the PIT structure depends on the sector of employment. The marginal PIT rates are 14 percent on monthly taxable incomes up to AZN 2,500 and 25 percent on the portion of income exceeding AZN 2,500, but the top rate applies only to the public and petroleum sectors. A key tax reform that took effect on January 1, 2019 exempted from PIT for a period of seven years employees working in the nonoil & gas and non-state sectors and earning up to AZN 8,000 per month (equivalent to $56,000 annual income). The tax rate applicable on the portion of income above AZN 8,000 is 14 percent. The top PIT rate for the non-public and nonoil sectors (14 percent) is lower than the average of the group of comparator countries and the CIT rate is slightly below the average rate of the comparators.1 Azerbaijan also derives less revenues from property taxes.

3. Azerbaijan has recently adopted several tax policy and administration measures, with the goal of reducing the sizeable informal economy and expanding the tax base. The administrative measures included expanding electronic information exchange with government agencies for tax control purposes; offering «Cash-back» incentives for consumers to promote “whitening” of turnovers in retail trade and service, while tightening financial sanctions for illegal employment and undeclared income. Recent tax policy measures included the PIT tax holiday in the nonoil, non-public sector (to encourage formal employment contracts), tighter eligibility restrictions for the simplified tax regime (to move microenterprises from the simplified regime to profit tax), reducing the turnover tax rate of the simplified tax regime (to induce the self-employed to register their activities), cutting the income tax rate for micro-businesses (to promote growth), and expanding the use of tax incentives to attract investment and create positive agglomeration effects.

uA001fig21

Azerbaijan: General Government Revenues, 2018

(In percent of GDP)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Notes: EMEU= Emerging Market Europe; LMI= Lower Middle Income countries; UMI= Upper Middle Income countries.
uA001fig22

Azerbaijan: Growth of Total Tax Revenues and Nominal GDP

(In percent)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, and IMF staff calculations.

4. Tax revenue as a share of GDP is lower than in most peers. In 2019, tax revenues were 14.6 percent of GDP, an improvement of 1.5 pp over 2018, stemming mainly from increased VAT collection. VAT C-efficiency in Azerbaijan (0.52) in 2018 was comparable to the average for lower-middle income countries (0.50), while below the average for upper-middle income countries (0.62). PIT revenues have been on a downward trend, falling by 0.7 pp of GDP during 2016–19.

5. The buoyancy of non-oil tax revenues is weak.2 In the few years preceding the global financial crisis, total tax revenue growth exceeded GDP growth, while the sharp GDP contraction in 2009 brought an even steeper fall in revenues, suggesting that overall tax buoyancy has been larger than one during this period, while the pattern is less clear in the subsequent years. A simple linear regression indicates that over the period 2000–19, on average a 1 pp change in GDP led to a 1.1 pp change in total tax revenues in the same year. The buoyancy stems entirely from oil sector CIT, underscoring revenue risks from the long-term slowdown in oil GDP growth. VAT buoyancy equals 1, while PIT exhibits weak buoyancy (with a coefficient of 0.85) and non-oil CIT shows buoyancy of 0.94 with respect to non-oil GDP. These findings further underscore the dependence of tax revenues on the oil sector and the weak response of other tax instruments to GDP growth.

6. Tax reforms will be needed to reorient revenue mobilization to nonoil sector development. Several directions are possible over the longer term, but the groundwork for these reforms, including strengthening the capacity of the tax administration, must be laid sooner.

  • i. PIT is the primary tax instrument for achieving fairness in taxation across income groups. The progressivity of PIT helps to offset the regressive aspect of the VAT, whereby the non-rich bear a disproportionate amount of tax compared to their income levels, either directly in tax payments or indirectly by a pass-through of the tax to consumers. Hence, PIT is both an important source of revenue mobilization and a key instrument for reducing inequality. Following the scheduled end of the PIT holiday for non-public, nonoil wage earners in 2026, a progressive PIT system based on end-of-year declaration of all sources of income should be put into place, with taxes withheld credited against the end-of-year tax liability. A PIT schedule with a small exemption level and at least one additional (non-zero) marginal tax rate, with the top rate designed to increase the tax paid by the richest individuals in the country, would help reduce the regressivity of taxation and increase revenues.

  • ii. Property tax is levied on the net book value of the fixed assets of legal entities, while for physical persons the tax is based on an amount per square meter of the area of the residential building, which depends on its location. A gradual replacement of the existing property tax with the one based on the market values of residential and nonresidential properties would improve revenue buoyancy and fairness. A modern residential property tax is less distortive than income tax and proxies wealth tax on owner-occupied dwellings.

  • iii. Tax exemptions are numerous3 and may cause unnecessary revenue leakages via abusive transfer pricing practices and disruptions to the VAT chain, or because domestic investment and employment relocate to the tax-exempt areas without generating net increases in amounts. Moreover, studies show that infrastructure and the ease of doing business are more important than tax holidays for attracting FDI. The nature of exemptions from profit tax is that the size of the benefit is not directly related to the size of the investment and some of these investments would have occurred even without the tax break. The cost of the tax exemptions, in terms of forgone revenues, should be calculated and assessed in relation to their measured benefits. A recommended practice is to append an annual tax expenditure report to the budget. Reducing tax exemptions would broaden the tax base and likely improve revenue performance.

Annex VI. Financial Development, Inclusion, and Stability

Finance is an engine of growth and needs to play a key part in Azerbaijan’s transition the new growth model. Yet if credit expansion is supply-driven, it raises risks to financial stability. Strengthening prudential oversight, stress-testing and transparency can help mitigate these risks.

1. Azerbaijan lags its peers in financial development and inclusion. Accessibility and depth are low, financial markets are underdeveloped, and weaknesses in bank balance sheets persist. High collateral requirements and low use of bank credit for business finance are among key obstacles.

uA001fig23

Azerbaijan:Financial Development Index

(AZE, CCA and EMDE Averages, 1980–2018)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: World Bank Enterprise Survey, IMF staff calculations.
uA001fig24

Azerbaijan: Firm Credit Access Measures

(In percent)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: World Bank Enterprise Survey, IMF staff calculations.

2. Past credit expansions were driven by vulnerable banks and accompanied by a weakening of lending standards. When credit booms are propelled by banks’ increased lending capacity, weaker banks tend to take advantage of less binding constraints and expand risky lending aggressively. Without strong prudential oversight, rapid credit expansions tend to be accompanied by excessive risk-taking, exacerbating bank distress in an economic downturn.1 An analysis using the future loan loss provision ratio to gauge current lending standards2 shows that the 2013–15 credit boom in Azerbaijan was driven by banks with low capital ratios and weak lending standards.3

uA001fig25

Azerbaijan: Selected Financial Indicators

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Sources: National authorities, and IMF staff calculations.Note: In panel 1, the RCOs are normalized by standard deviations in 2012–20. In panel 2, the estimates are α2 in Yi,t+h = α0 + α1Xi,t + α2Xi,t x S_Boomt + T’Ctrlsi,t + γi + γt + εi,t where Yi,t+h is h-year ahead credit growth or LLP ratio, S_Boomt is a time dummy for supply-driven credit booms, Xi,t is bank capital ratio. Dashed lines or dots indicate 90 percent confidence intervals.

3. Busts in supply-driven credit booms amplify economic downturns while demand-driven credit booms have lasting economic benefits. In a vector autocorrelation,4 increased credit supply boosts GDP in the first 4 months but the impact turns negative after 9 months, lasting for 2.5 years. An increase in credit demand, however, is followed by a 2.5-year boost to GDP.

4. To ensure sustainable financial deepening where increases in credit supply and demand go hand in hand, reforms need to include measures to strengthen bank supervision and regulation, promote SME financial literacy and improve financial sector transparency.

uA001fig26

Azerbaijan: Impulse Response of Real GDP Growth After Credit Shock

(Annualized additional monthly growth in percentage points)

Citation: IMF Staff Country Reports 2021, 278; 10.5089/9781616359591.002.A001

Source: IMF staff calculations.
1

Oil GDP includes natural gas production which is becoming an important part of the sector.

2

In Q1 2021, real GDP was flat compared to Q1 2020, with nonoil GDP increasing by about 2 percent. Inflation picked up to 4.2 percent in May, driven by increases in global food prices.

3

This estimate includes the guarantee granted to Agrarkredit, a government-controlled credit agency, on the bond it issued to acquire bad assets from the IBA during the 2015–16 crisis. The government is highly likely to service the bond because Agrarkredit may not be able to recover the bad assets in full. The outstanding amount was AZN 9.48 billion (13.1 percent of the 2020 GDP), as of January 1, 2021. The debt service cost is limited, given 30-year maturity, 5-year grace period, and interest rate of 0.15 percent. Other guarantees account for additional 22 percent of GDP.

4

SOFAZ’s transfer of its 2020 asset valuation and profit gains to the CBA offset the $1.8 billion decline in the CBA reserves reported in the balance of payments.

5

This measure is being assessed by staff using the Institutional View on Liberalization and Management of Capital Flows, as it might constitute a capital flow management measure.

6

Azerbaijan consolidated by 1.6 pp of GDP per year on average during 2016–19. In three-to-four year-long adjustment episodes, half of countries reduced their primary balance by 5 percent of GDP, and a quarter by 7 percent of GDP. Escolano, J. and others, 2018, How Much Is A Lot? The Maximum Size of Fiscal Adjustments, Journal of Applied Economics, 21(1), 137–59.

7

However, even under this adjustment scenario, NFA decline to 18 percent of GDP by 2030. The debt ratio in the authorities’ scenario is lower than that in the staff’s scenario because of different debt coverage (see Footnote 3), lower deficits, higher nominal GDP, and larger drawdown of SOFAZ assets.

8

OECD, 2016, Fourth Round of Monitoring of the Istanbul Anti-Corruption Action Plan.

9

The government has been supporting SOEs through transfers and capital injections, “bridge financing” for debt service, and budget co-financing of investment projects. In 2020, such on-budget support amounted to over 2 percent of GDP. The authorities expect it to fall to 0.5 percent of GDP in 2021.

10

The PIT tax holiday was legislated to take effect on January 1, 2019 and expire at end-2025, as part of the government’s strategy to encourage informal operators to register formal labor contracts. The authorities attributed an increase in registered labor contracts in 2019–20 to the announcement of the holiday.

11

The IMF’s 2019 Strategy on Social Spending and Annex V of the 2019 Article IV Consultation Report, available at www.imf.org.

12

By reducing the need for precautionary saving, higher social spending would also help close the external gaps (Annex III).

13

See Figure 7 for cross-country comparisons of social spending in per capita terms, as a share of GDP and government spending, and the social protection coverage of the vulnerable groups.

14

The authorities also intend to strengthen their macroeconomic forecasting and policy analysis systems to ensure consistency of fiscal policies, monetary and structural policies and increase policy formulation transparency.

15

Government-guaranteed SOE debt stood at 35 percent of the 2020 GDP in January 2021, including Agrarkredit debt to the CBA, i.e., local currency debt related to the past banking crisis (13.1 percent of GDP); and foreign currency debt related to the Southern Gas Corridor commercial enterprise (9 percent of GDP), SOEs in energy, oil, airline, railroad and other sectors (5 percent of GDP), and the unclassified (1 percent of GDP). The government has been budgeting annually for payments on guaranteed debt and has set up an off-budget contingency fund to make payments on guaranteed debt while providing a line of credit to SOEs facing payment difficulties.

16

Another source of risk stems from bank weaknesses, as evidenced by the IBA restructuring, the recent closure of four banks and the need to provide support to the Azerbaijan Deposit Insurance Fund.

17

Specific measures include establishing an ownership monitoring framework based on the OECD’s Guidelines on Corporate Governance of SOEs, conducting an SOE sectorization into public corporations and general government units according to GFSM 2014, and creating a risk management unit in the Ministry of Finance.

18

In recent years, CBA started producing press releases after each monetary policy decision, quarterly monetary policy reviews, and annual monetary policy guidelines. It also publishes information on FX auctions.

1

Bems, R., and I. de Carvalho Filho, 2009, “Exchange Rate Assessments: Methodologies for Oil-Exporting Countries,” IMF Working Paper 09/281.

2

The estimated norm is sensitive to the choice of model parameters, such as the GDP growth rate, interest rate, and population growth rate, as well as the underlying oil prices.

3

The administrative and reporting requirement measures are being assessed by staff using the Institutional View on Liberalization and Management of Capital Flows, as they might constitute capital flow management measures.

1

The comparator countries are the oil producers Colombia, Kazakhstan, Mexico, Turkmenistan, and Uzbekistan.

2

A buoyancy rate of 1 would imply that an extra percent of GDP would increase tax revenue by 1 percent, leaving the tax-to-GDP ratio unchanged. A tax buoyancy exceeding 1, however, would increase tax revenue by more than GDP and lead to improvements in the fiscal balance, as well as providing automatic stabilization against income shocks.

3

Exemptions from profit tax include a 50 percent exemption for legal entities holding an investment promotion certificate and a ten-year exemption for entities resident in industrial and technology parks. VAT exemptions have been applied for all types of goods imported by the residents of industrial parks for the purposes of their activities for the period of five years until 1 May 2021. The first Free Economic Zone (FEZ) became operational in May 2020, to promote the emerging trade and logistics hub in the Alat settlement on the Caspian Sea coast. Goods and services imported into the FEZ are exempted from duties and VAT, and legal entities operating in the zone and their employees, as well as Free Zone residents, are exempt from all taxes related to their activities. The profits of SME clusters are exempted from tax for 7 years and 75 percent of the profit of micro entrepreneurs are exempt from profit tax and property tax. Startups are also exempt from income and profit tax for three years, while agricultural producers are exempted from profit tax, property tax, and VAT for 10 years.

1

Jiménez, G. and others, 2014. “Hazardous Times for Monetary Policy: What Do Twenty-Three Million Bank Loans Say About the Effects of Monetary Policy on Credit Risk-Taking?” Econometrica. 82(2), 463–505. Igan, D., and M. Pinheiro. 2011. “Credit Growth and Bank Soundness: Fast and Furious?” IMF paper. Igan, D., and N. Tamirisa. 2008. “Are Weak Banks Leading Credit Booms? Evidence from Emerging Europe.” Comparative Economic Studies. 50(4), 599–619.

2

Fahlenbrach, R., and others, 2017. “Why Does Fast Loan Growth Predict Poor Performance for Banks?” The Review of Financial Studies. 31(3), 1014–63.

3
The riskiness of credit originations is measured as:
RCOtX=ΣiHighGrowthXi,tNt,HighGrowthΣiLowGrowthXi,tNt,LowGrowth
where High Growth refers to banks with credit increases in the top quartile, Low Growth in the bottom quartile, Xi,t is bank riskiness (the capital ratio or z-score), and (Nt,HighGrowth, Nt,Low Growth) are the numbers of banks in each group. Lower Xi,t means greater risk, and negative ROCtX indicates credit expansion driven by weaker banks. Greenwood, R., and S. Hanson. 2013. “Issuer Quality and Corporate Bond Returns.” The Review of Financial Studies. 26(6), 1483–1525.
4

Uhlig, H., 2005. “What are the effects of monetary policy on output? Results from an agnostic identification procedure.” Journal of Monetary Economics. 52, 381–419.

  • Collapse
  • Expand
Republic of Azerbaijan: 2021 Article IV Consultation-Press Release; and Staff Report
Author:
International Monetary Fund. Middle East and Central Asia Dept.
  • View in gallery
    Figure 1.

    Azerbaijan: Economic Impact of COVID-19

  • View in gallery
    Figure 2.

    Azerbaijan: Real Sector Indicators

  • View in gallery

    Azerbaijan: Fiscal and Monetary Measures, 2020

    (In percent)

  • View in gallery
    Figure 3.

    Azerbaijan: External Sector

  • View in gallery
    Figure 4.

    Azerbaijan: Fiscal Sector

  • View in gallery
    Figure 5.

    Azerbaijan: Monetary and Exchange Rate Developments

  • View in gallery

    Azerbaijan: Financial Condition Index

    (Standard Deviations)

  • View in gallery
    Figure 6.

    Azerbaijan: Banking Sector

  • View in gallery

    Azerbaijan: Azeri-Chirag-Gunashli Oil and Gas Production

    (In millions of barrels per day)

  • View in gallery

    Azerbaijan: Potential Nonoil Output Growth

    (In percent, HP filtered)

  • View in gallery

    Azerbaijan: Contribution to Nonoil Real GDP Growth

    (Percent, production function asumes 70 percent capital share)

  • View in gallery

    Azerbaijan: Fiscal Dynamics

  • View in gallery

    Azerbaijan: Financial Assets

    (In percent of GDP)

  • View in gallery
    Figure 7.

    Azerbaijan: Social Spending

  • View in gallery

    Azerbaijan: Quality of Infrastructure and Capital Expenditure

  • View in gallery

    Azerbaijan: Wage Bill and Public Employment

    (In percent)

  • View in gallery

    Azerbaijan: Post-tax Subsidies, 2017 1/

    (In percent of GDP)

  • View in gallery

    Azerbaijan: Capital-to-Asset Ratio and Loan Growth

    (Average, In Percent)

  • View in gallery

    Azerbaijan: NPL Ratios for Business Loans in Systemic Banks

    (In Percent)

  • View in gallery

    Azerbaijan: Individuals using the Internet

    (In percent of population)

  • View in gallery

    Azerbaijan: CO2 Emissions

    (In Millions of Tons)

  • View in gallery

    Azerbaijan: Fiscal Support in Response to COVID-19

    (In percent of 2020 GDP, cumulative since March 2020)

  • View in gallery

    Azerbaijan: Financial Measures

    (Proportion of countries implementing measure)

  • View in gallery
    Figure A.II.1.

    Azerbaijan: Public Sector Debt Sustainability Analysis (DSA)

    (In percent of GDP unless otherwise indicated)

  • View in gallery
    Figure A.II.2.

    Azerbaijan: Public DSA – Composition of Public Debt and Alternative Scenarios

    (In percent of GDP unless otherwise indicated)

  • View in gallery

    Azerbaijan: Real Effective Exchange Rates and Contributions to Current Account

  • View in gallery

    Azerbaijan: Forecasted CA and CA Norm under Bems-Carvalho (Oil Resource) Adjustment

    (In percent of GDP)

  • View in gallery
    Figure A.III.1.

    Azerbaijan: Exchange Rates and Reserve Adequacy

  • View in gallery

    Azerbaijan: Oil Revenues and Non-Oil Tax Revenues

    (In percent of GDP, unless stated otherwise)

  • View in gallery

    Azerbaijan: General Government Revenues, 2018

    (In percent of GDP)

  • View in gallery

    Azerbaijan: Growth of Total Tax Revenues and Nominal GDP

    (In percent)

  • View in gallery

    Azerbaijan:Financial Development Index

    (AZE, CCA and EMDE Averages, 1980–2018)

  • View in gallery

    Azerbaijan: Firm Credit Access Measures

    (In percent)

  • View in gallery

    Azerbaijan: Selected Financial Indicators

  • View in gallery

    Azerbaijan: Impulse Response of Real GDP Growth After Credit Shock

    (Annualized additional monthly growth in percentage points)