## 1. Does Kazakhstan Have Dutch Disease?

## Source details

**Canonical URL:** [1. Does Kazakhstan Have Dutch Disease?](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13291.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13291.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13291.pdf.json)

---

### Introduction and objectives
- Kazakhstan is among the world’s top 20 oil producers with estimated reserves of 40 billion barrels, and about 2 percent share in global oil production.
- Kashagan oil field was discovered in 2000 and is the largest new field found in the world during the past 30 years.
- Chapter objective: develop a structural model calibrated to Kazakhstan to analyze macroeconomic implications of alternative public investment scaling-up scenarios, allowing for:
  - positive effects of public investment on growth; and
  - public investment inefficiencies and absorptive capacity constraints.
- High-level simulation insight:
  - Ambitious public investment can boost growth but also risk adverse macroeconomic outcomes, notably loss of competitiveness in the nonoil tradable sector (Dutch disease), and depleting accumulated buffers or accumulating costly external debt.

### Fiscal management of oil revenues and the National Fund (NFRK)
- Historical fiscal behavior:
  - During 2003–08 much of the windfall was saved or used to repay public debt; government spending remained relatively constant as a share of GDP.
  - During the 2008–09 crisis a stimulus estimated at 7.5 percent of GDP and about USD10 billion out of accumulated oil savings were used to fund the stimulus.
  - After the crisis the stimulus was unwound; the nonoil deficit has been declining but remains above the estimated sustainable level of 6 percent of GDP.
- NFRK specifics:
  - Established in 2000 as an off-budget fund; assets managed abroad by the National Bank of Kazakhstan on behalf of the government.
  - Direct taxes from the oil sector (corporate income tax, excess profit tax, royalties and share under production-sharing agreements) accrue directly to the fund and account for about 90 percent of all taxes from the oil sector; the remaining 10 percent accrue to the central or local budgets.
  - The oil fund balance reached 29 percent of GDP as of end 2012.
- Legislative anchors governing oil revenues:
  - Annual spending out of NFRK is fixed in nominal terms (at USD8 billion) with limited flexibility for countercyclicality.
  - Minimum balance of NFRK currently set at 20% of the current year GDP; expected to be raised to 30% during upcoming revisions.
  - Target that by 2020 the deficit of the unconsolidated budget net of the transfer from NFRK should not exceed 3% of GDP (roughly equivalent to 6% of GDP on consolidated basis).
  - Legislation includes a provision that interest payments on government debt should not exceed interest income earned by NFRK.

### Broader public sector, quasifiscal operations, and transparency
- Scope and systemic elements:
  - The public sector extends beyond the government budget; the quasifiscal sector—including Samruk Kazyna (SK), the systemic state investment holding company—is systemic.
  - SK’s assets are about the size of half of GDP, the bulk of which are in the natural resources sector.
  - The quasifiscal sector played a key role in the stimulus during the crisis and participates actively in long-term development programs.
- Recent reforms to improve transparency:
  - Legislative amendment requiring SK companies’ costs incurred due to participation in government’s non-commercial programs be covered from the government budget.
  - Consolidation of main development institutions into a new extra-budgetary entity (Baiterek) to separate commercial and non-commercial roles and potentially increase transparency.
  - Authorities intend to introduce legislation to keep the stock of the broader public sector debt under control, e.g., by setting ceilings for total government and quasifiscal debt.

### Development and diversification strategy: scale and risks
- Strategic ambitions:
  - Authorities’ long-term strategy (Strategy 2050) aims for accelerated diversification and to be among the world’s 30 most developed economies by 2050.
- Industrialization Map:
  - Includes 779 projects to be implemented during the course of several years, with about 20% of the projects currently being implemented.
  - Total cost KT11 trillion, which is equal to nearly a third of GDP in 2013.
- Financing modalities across the broader public sector include direct budget support, loans or loan guarantees, subsidized interest rates, equity financing, local content requirements, and public-private partnerships.
- Authorities project that all development and diversification programs combined are expected to ensure 6.8 percent growth rate in the medium term.
- Concerns and risks:
  - State-led, top-down industrialization poses risks given international experiences of failures associated with such policies.
  - Policy focus on state-led industrialization could increase the role of the state in the economy, rather than boosting private-sector-led growth.
  - Lack of clarity on scope, timeframe, costs, and expected benefits of programs makes comprehensive assessment difficult.

### Evidence and mechanisms of Dutch disease (Box 1)
- Observed empirical patterns:
  - A real effective exchange rate appreciation trend is discernible starting 2003; timing coincides with the start of the oil boom.
  - Prices of nontradables (proxied by services in the CPI basket, excluding administratively controlled utility prices) have grown consistently faster than prices of tradables (proxied by non-food goods).
  - Tradables have been losing share in both GDP and employment; investments and FDI have been highly concentrated in oil and gas and related transport sectors.
  - Rapid real wage growth (controlling for productivity) has been higher economy-wide and in all main sectors except for construction over the last 12 years.
  - Unit labor costs have stayed broadly stable; manufacturing prices grew at an annual average rate of 17 percent correlated with oil prices.
- Two primary Dutch disease channels:
  - Factor movement effect: resource sector attracts labor and capital from other sectors → “direct de-industrialization” and higher prices for nontradables → real exchange rate appreciation.
  - Spending effect: resource revenue spending increases aggregate demand → higher nontradable prices and resource reallocation away from tradables → “indirect de-industrialization.”
- Interpretation and ruling out alternatives:
  - Evidence points to the spending effect as the dominant channel.
  - Balassa-Samuelson effect is not supported by a clear positive productivity differential between tradables and nontradables in Kazakhstan.
- Policy implication:
  - Higher public spending to finance development and diversification could exacerbate Dutch disease, damaging the nonoil tradable sector and undermining diversification objectives.

### Model-based framework for analyzing management of oil wealth
- Model type and main features:
  - Dynamic stochastic general equilibrium for an oil-producing small open economy; combines elements of Buffie et al (2012), Berg et al. (2013), and Melina et al. (2013); abstracts from long run trend growth and the nominal side of the economy.
- Households:
  - Two types: optimizing households (access to capital and financial markets) and rule of thumb households (liquidity constrained, consume all disposable income each period).
  - Optimizing households can acquire domestic government bonds and international bonds with portfolio adjustment costs.
- Production sectors:
  - Three sectors: oil production, nontraded goods, non-oil traded goods.
  - Oil production exogenous AR process; international oil prices taken as given and follow an AR process.
  - Nontraded and non-oil traded sectors use Cobb-Douglas production functions with labor, private capital, and public capital. Nonoil traded sector TFP subject to learning-by-doing externalities.
  - Parameters ρ and related terms control the degree of Dutch disease via feedback from traded sector production to productivity.
- Public capital and investment:
  - Public capital enters production functions and provides a positive externality that can crowd in private investment.
  - Public investment subject to absorptive capacity constraints and investment inefficiency: effective investment = φ(g^G_t) · g^G_t where 0 < φ_t ≤ 1; efficiency falls from φ to φ̄ when expenditure exceeds a threshold.
  - Law of motion of public capital: k^G_t = (1-δ_t) k^G_{t-1} + ī^G_t, with depreciation rising when effective investment fails to maintain existing capital.
- Fiscal block:
  - Government expenditure financed by taxes on oil and non-oil sectors, interest income from oil savings, and domestic and foreign borrowing.
  - Borrowing subject to a risk premium depending on deviation of external public debt/GDP from steady state.
  - Budget surplus (excess oil revenues) saved in oil fund; deficits financed by withdrawals unless oil fund falls below a pre-specified level. When oil fund lower bound binds, government may borrow externally, adjust taxes, or cut expenditures.

### Calibration and scenarios for public investment scaling-up
- Calibration notes:
  - Public investment efficiency φ steady-state value: 0.6; when absorptive capacity constrained φ falls to 0.4 for additional investments above the initial level.
  - Output elasticity with respect to public capital, α_G, set to 0.15.
  - Oil fund floor set to 30% of GDP.
  - Simulations report deviations relative to a no scaling-up path; assume no shocks to oil production.
- Public investment objectives and paths:
  - Government calibration goal: achieve sustainable long term growth rate of 7 percent (policy objective in government documents).
  - Scaling-up scenarios calibrated to achieve specific nonoil growth objectives.
- Oil price scenarios:
  - Baseline: WEO projections until 2018 and minor fluctuations afterwards.
  - Adverse: large negative shock with price falling to $51.7. a barrel in 2016 and recovering to $80 in 2019.
- Budget treatment note:
  - All additional public investment assumed channeled through the budget; comparison to authorities’ fiscal plans would require incorporating broader public sector spending.

### Simulation results — prudent versus aggressive scaling-up
- Prudent path (target: average of 0.3% above trend nonoil growth)
  - Public investment needs to increase to about 6.4% of GDP.
  - Capital expenditures historically around 5% of GDP; under the prudent path capital expenditures increase by 1.4% of GDP (assuming all increase is in the budget).
  - Oil fund floor binds for the first few years; government borrows externally to finance investment.
  - External debt increases by only 5% of GDP over the current comfortable level in the adverse oil price scenario.
  - Oil fund trajectory by 2030:
    - Baseline oil price scenario: above 70% of GDP.
    - Adverse oil price scenario: about 50% of GDP.
  - Difference between interest income earned by oil fund and interest paid on public foreign debt remains positive throughout simulation.
  - Public capital increases by almost 20 percent by 2030 compared to no scaling-up (slightly less in adverse scenario).
  - Investment efficiency: does not decline much; absorptive capacity constraints bind only slightly.
  - Private consumption:
    - Baseline: grows steadily as external debt remains sustainable and taxes stay low.
    - Adverse: falls somewhat initially to make room for higher public investment, reflecting higher taxes.
  - Private investment: increases in the nonoil economy due to positive externality from productive public capital.
  - Sectoral effects and Dutch disease:
    - Real exchange rate appreciates by about 4 percent initially and then gradually returns to steady state; appreciation smaller in adverse scenario.
    - Traded output falls initially below trend for the first ~10 years in the baseline oil price scenario.
    - Private investment in traded sector: lower than in no scaling-up case; nontraded sector investment higher.

- Aggressive path (target: average of 0.6% above trend nonoil growth)
  - Public investment needs to increase to about 12% of GDP.
  - Oil fund floor binds over the entire simulation period in both scenarios; government resorts to large-scale external borrowing.
  - To keep external debt sustainable the government needs to increase taxes (labor tax and consumption tax).
  - In the adverse oil price scenario:
    - External debt increases by considerably more.
    - Labor income and consumption tax rates have to almost double in addition to higher taxes on the use of public capital.
    - Interest payments on foreign debt net of oil fund interest earnings reach about 1 percent of GDP.
  - Public capital increases by nearly 60 percent above no scaling-up case by 2030, despite investment efficiency falling from 0.6 to 0.4.
  - Costs and trade-offs:
    - Private consumption falls substantially and for a long period, especially in the adverse scenario.
    - Growing external debt, exchange rate appreciation, and higher taxes negatively affect private nonoil sector.
    - Real exchange rate appreciation: about 6 percent in the adverse oil price scenario and about 8 percent in the baseline scenario.
    - Private investment in the nonoil tradable sector is considerably lower compared to the prudent path due to loss of competitiveness.
    - Higher borrowing costs for the private sector as public external debt increases risk premia.

### Key quantitative outcomes and indicators (as reported)
- Public investment required:
  - Prudent path: about 6.4% of GDP.
  - Aggressive path: about 12% of GDP.
- Public capital change by 2030:
  - Prudent: almost 20% above no scaling-up.
  - Aggressive: nearly 60% above no scaling-up.
- Investment efficiency:
  - Steady-state φ = 0.6; falls to φ = 0.4 for additional investments above initial level under absorptive constraints.
- Oil fund floor: 30% of GDP.
- Oil price in adverse scenario: $51.7. a barrel in 2016; recover to $80 in 2019.
- External debt increase under prudent path (adverse oil price scenario): 5% of GDP over current comfortable level.
- Interest payments net of oil fund earnings under aggressive path (adverse scenario): about 1 percent of GDP.
- Real exchange rate appreciation:
  - Prudent: about 4 percent initially.
  - Aggressive: about 6 percent (adverse) and about 8 percent (baseline).
- Nonoil growth targets:
  - Prudent: average 0.3% above trend (adverse scenario growth increases by an average of 0.26% above trend).
  - Aggressive: average 0.6% above trend.
- Historical capital expenditures baseline: around 5% of GDP; prudent path implies increase by 1.4% of GDP if all additional investment is in budget.

### Conclusion — policy implications
- Trade-offs summarized:
  - Ambitious (aggressive) scaling up can generate higher nonoil growth but at high funding costs: substantially lower private consumption and household welfare, larger external borrowing, higher taxes, and stronger exchange rate appreciation that harms tradables.
  - Prudent scaling up can boost nonoil growth with much smaller costs: manageable external borrowing, limited declines in investment efficiency, positive spillovers to private investment, and smaller adverse effects on traded sector competitiveness.
- Policy guidance implicit in model results:
  - Favor calibrated/prudent increases in public investment that respect absorptive capacity and maintain fiscal and external sustainability.
  - Monitor investment efficiency, absorptive capacity thresholds, and the oil fund floor to avoid prolonged binding of fund constraint and excessive external borrowing.
  - Recognize distributional and sectoral effects: public capital can crowd in nontraded sector activity but risks Dutch Disease effects on tradables via real exchange rate appreciation.

### Comprehensive management of oil wealth and fiscal metrics
- Key findings:
  - Adopt a comprehensive approach to management of oil wealth.
  - Given the large role of the quasifiscal sector in Kazakhstan, it is particularly important to look beyond traditional metrics of fiscal accounts.
  - Quasifiscal operations and contingent liabilities to the state should be appropriately taken into account.
  - To guard against Dutch Disease it is important to be mindful of the impact of policy decisions on the private sector.

### Development and diversification strategy — practical implications
- Key recommendations:
  - Careful design of Kazakhstan’s development and diversification strategy is important.
  - Productive investment into high quality public service delivery would ensure that Kazakhstan’s oil wealth benefits all.
  - Top-down industrialization policies are less likely to provide broad based benefits and carry the risk of further increasing the role of the state in the economy.
  - Good public financial management is key, including careful cost-benefit analysis of various programs and evaluation of outcomes.

### Appendix 1. Kazakhstan: Calibration of Key Parameters
- Selected calibrated parameters and values:
  - α୒, α୘: labor income share in nontraded and traded sector — 0.5 — assumption
  - αୋ: output elasticity with respect to public capital — 0.15 — falls into the literature range of 0.05 – 0.2
  - ߜீ: steady-state annual depreciation rate of public capital — 0.07 — Berg et al. (2013)
  - ߩ௓், ߩ௒்: learning-by-doing parameter — 0.1 — Berg et al. (2013), mild externality
  - ߳: steady-state efficiency of public investment — 0.6 — between developed and developing countries, 0.3 – 1
  - ߳̅: lower efficiency when absorptive capacity is constrained — 0.4 — assumption, only for additional investment

*Prepared by Gohar Minasyan (MCD) and Susan Yang (RES); July 18, 2013.*

### 1. Does Kazakhstan Have Dutch Disease? __________________________________________________ 7

### 1. Does Kazakhstan Have Dutch Disease?

### Introduction and objectives
- Kazakhstan is among the world’s top 20 oil producers with estimated reserves of 40 billion barrels, and about 2 percent share in global oil production.
- Kashagan oil field was discovered in 2000 and is the largest new field found in the world during the past 30 years.
- The chapter develops a structural model calibrated to Kazakhstan to analyze macroeconomic implications of alternative public investment scaling-up scenarios, allowing for:
  - positive effects of public investment on growth; and
  - public investment inefficiencies and absorptive capacity constraints.
- Simulation results indicate ambitious public investment can boost growth but also risk:
  - adverse macroeconomic outcomes, notably loss of competitiveness in the nonoil tradable sector (Dutch disease); and
  - depleting accumulated buffers or accumulating costly external debt.

### Fiscal management of oil revenues and the National Fund (NFRK)
- During 2003–08 much of the windfall was saved or used to repay public debt; government spending remained relatively constant as a share of GDP (Figure 1).
- During the 2008–09 crisis a stimulus estimated at 7.5 percent of GDP and about USD10 billion out of accumulated oil savings were used to fund the stimulus.
- After the crisis the stimulus was unwound; the nonoil deficit has been declining but remains above the estimated sustainable level of 6 percent of GDP (Figure 2).
- NFRK specifics:
  - Established in 2000 as an off-budget fund; assets managed abroad by the National Bank of Kazakhstan on behalf of the government.
  - Direct taxes from the oil sector (corporate income tax, excess profit tax, royalties and share under production-sharing agreements) accrue directly to the fund and account for about 90 percent of all taxes from the oil sector; the remaining 10 percent accrue to the central or local budgets.
  - The oil fund balance reached 29 percent of GDP as of end 2012.
- Legislative anchors governing oil revenues:
  - Annual spending out of NFRK is fixed in nominal terms (at USD8 billion) with limited flexibility for countercyclicality.
  - Minimum balance of NFRK currently set at 20% of the current year GDP; expected to be raised to 30% during upcoming revisions.
  - Target that by 2020 the deficit of the unconsolidated budget net of the transfer from NFRK should not exceed 3% of GDP (roughly equivalent to 6% of GDP on consolidated basis).
  - Legislation includes a provision that interest payments on government debt should not exceed interest income earned by NFRK.

### Broader public sector, quasifiscal operations, and transparency
- The public sector extends beyond the government budget; the quasifiscal sector—including Samruk Kazyna (SK), the systemic state investment holding company—is systemic.
  - SK’s assets are about the size of half of GDP, the bulk of which are in the natural resources sector.
  - The quasifiscal sector played a key role in the stimulus during the crisis and participates actively in long-term development programs.
- Recent reforms to improve transparency:
  - Legislative amendment requiring SK companies’ costs incurred due to participation in government’s non-commercial programs be covered from the government budget.
  - Consolidation of main development institutions into a new extra-budgetary entity (Baiterek) to separate commercial and non-commercial roles and potentially increase transparency.
  - Authorities intend to introduce legislation to keep the stock of the broader public sector debt under control, e.g., by setting ceilings for total government and quasifiscal debt.

### Development and diversification strategy: scale and risks
- Authorities’ long-term strategy (Strategy 2050) aims for accelerated diversification and to be among the world’s 30 most developed economies by 2050.
- The Industrialization Map of Kazakhstan:
  - Includes 779 projects to be implemented during the course of several years, with about 20% of the projects currently being implemented.
  - Total cost KT11 trillion, which is equal to nearly a third of GDP in 2013.
- Financing modalities across the broader public sector include direct budget support, loans or loan guarantees, subsidized interest rates, equity financing, local content requirements, and public-private partnerships.
- Authorities project that all development and diversification programs combined are expected to ensure 6.8 percent growth rate in the medium term.
- Concerns:
  - State-led, top-down industrialization poses risks given international experiences of failures associated with such policies.
  - Policy focus on state-led industrialization could increase the role of the state in the economy, rather than boosting private-sector-led growth.
  - Lack of clarity on scope, timeframe, costs, and expected benefits of programs makes comprehensive assessment difficult.

### Evidence and mechanisms of Dutch disease (Box 1)
- Observed patterns:
  - A real effective exchange rate appreciation trend is discernible starting 2003 (aside from spikes related to the Russian crisis in 1998 and the 2008–2009 crisis); the timing coincides with the start of the oil boom.
  - Prices of nontradables (proxied by services in the CPI basket, excluding administratively controlled utility prices) have grown consistently faster than prices of tradables (proxied by non-food goods).
  - Tradables have been losing share in both GDP and employment; investments and FDI have been highly concentrated in oil and gas and related transport sectors.
  - Rapid real wage growth (controlling for productivity) has been higher economy-wide and in all main sectors except for construction over the last 12 years.
  - Unit labor costs have stayed broadly stable; in manufacturing favorable price dynamics (manufacturing prices grew at an annual average rate of 17 percent correlated with oil prices) have offset productivity/wage dynamics.
- Two primary Dutch disease channels described:
  - Factor movement effect: resource sector attracts labor and capital from other sectors, leading to “direct de-industrialization” and higher prices for nontradables → real exchange rate appreciation.
  - Spending effect: resource revenue spending increases aggregate demand, raising prices of nontradables and attracting resources away from tradables → “indirect de-industrialization.”
- Ruling out alternative explanations:
  - The observed price dynamics and sectoral employment/investment patterns point to the spending effect rather than the factor movement effect as the dominant channel.
  - The Balassa-Samuelson effect is not supported by a clear positive productivity differential between tradables and nontradables in Kazakhstan.
- Policy implication from evidence:
  - Higher public spending to finance development and diversification could exacerbate Dutch disease, damaging the nonoil tradable sector and undermining diversification objectives.

*Prepared by Gohar Minasyan (MCD) and Susan Yang (RES); July 18, 2013.*

### Box 1. Continued. Does Kazakhstan Have Dutch Disease?

### Box 1. Continued. Does Kazakhstan Have Dutch Disease?

### C. A model-based framework for analyzing management of oil wealth
- Model type: dynamic stochastic general equilibrium for an oil-producing small open economy; combines elements of Buffie et al (2012), Berg et al. (2013), and Melina et al. (2013). Main difference from Melina et al. (2013): abstracts from long run trend growth and the nominal side of the economy.
- Households:
  - Two types: optimizing households (access to capital and financial markets) and rule of thumb households (liquidity constrained, consume all disposable income each period).
  - Optimizing households can acquire domestic government bonds and international bonds with portfolio adjustment costs.
  - Private sector pays a constant premium over the government interest rate on external debt.
- Production sectors:
  - Three sectors: oil production, nontraded goods, non-oil traded goods.
  - Oil production is modeled as an exogenous autoregressive process with resource production shock; international oil prices taken as given and evolve as an autoregressive process with resource price shock.
  - Nontraded and non-oil traded sectors: Cobb-Douglas production functions using labor, private capital, and public capital. Nonoil traded sector TFP subject to learning-by-doing externalities depending on previous period’s traded output.
  - Parameters ρ and related terms control the degree of Dutch disease via feedback from traded sector production to productivity.
- Public capital and investment:
  - Public capital enters production functions and provides a positive externality that can crowd in private investment.
  - Public investment subject to absorptive capacity constraints and investment inefficiency: effective investment = φ(g^G_t) · g^G_t where 0 < φ_t ≤ 1; investment efficiency assumed to fall from φ to φ̄ when expenditure exceeds a threshold.
  - Law of motion of public capital includes time-varying depreciation: k^G_t = (1-δ_t) k^G_{t-1} + ī^G_t, with depreciation rising when effective investment fails to maintain existing capital.
- Fiscal block:
  - Government expenditure financed by taxes on oil and non-oil sectors, interest income from oil savings, and domestic and foreign borrowing.
  - Borrowing subject to a risk premium depending on deviation of external public debt/GDP from steady state.
  - Budget surplus (excess oil revenues) saved in oil fund; deficits are financed by withdrawals unless oil fund falls below a pre-specified level. When oil fund lower bound binds, government may borrow externally, adjust taxes, or cut expenditures.

### D. Analysis of alternative public investment scaling-up paths — calibration and scenarios
- Calibration notes:
  - Public investment efficiency φ steady-state value chosen as 0.6; when absorptive capacity constrained φ falls to 0.4 for additional investments above the initial level.
  - Output elasticity with respect to public capital, α_G, set to 0.15.
  - Oil fund floor set to 30% of GDP.
  - Simulations: deviations reported relative to a no scaling-up path; assume no shocks to oil production.
- Public investment objectives and paths:
  - Government goal used in calibration: achieve sustainable long term growth rate of 7 percent (noted as a policy objective in government documents).
  - Public investment scaling-up scenarios calibrated to achieve specific nonoil growth objectives.
- Oil price scenarios:
  - Baseline: WEO projections until 2018 and minor fluctuations afterwards.
  - Adverse: large negative shock with price falling to $51.7. a barrel in 2016 and recovering to $80 in 2019. (The magnitude of this negative shock is roughly the same as in 2009.)
- Notes on budget treatment:
  - All additional public investment assumed channeled through the budget; comparison to authorities’ fiscal plans would require incorporating broader public sector spending.

### Simulation results — prudent versus aggressive scaling-up
- Prudent path (target: average of 0.3% above trend nonoil growth)
  - Public investment needs to increase to about 6.4% of GDP.
  - Capital expenditures historically around 5% of GDP; under the prudent path capital expenditures increase by 1.4% of GDP (assuming all increase is in the budget).
  - Oil fund floor binds for the first few years; government borrows externally to finance investment.
  - External debt increases by only 5% of GDP over the current comfortable level in the adverse oil price scenario (and by less in the baseline).
  - Oil fund trajectory by 2030:
    - Baseline oil price scenario: above 70% of GDP.
    - Adverse oil price scenario: about 50% of GDP.
  - Difference between interest income earned by oil fund and interest paid on public foreign debt remains positive throughout simulation.
  - Public capital: increases by almost 20 percent by 2030 compared to no scaling-up (slightly less in adverse scenario).
  - Investment efficiency: does not decline much; absorptive capacity constraints bind only slightly.
  - Private consumption:
    - Baseline: grows steadily as external debt remains sustainable and taxes stay low.
    - Adverse: falls somewhat initially to make room for higher public investment, reflecting higher taxes.
  - Private investment: increases in the nonoil economy due to positive externality from productive public capital.
  - Sectoral effects and Dutch disease:
    - Real exchange rate appreciates by about 4 percent initially and then gradually returns to steady state; appreciation smaller in adverse scenario.
    - Traded output: falls initially below trend for the first ~10 years in the baseline oil price scenario.
    - Private investment in traded sector: lower than in no scaling-up case; nontraded sector investment higher.

- Aggressive path (target: average of 0.6% above trend nonoil growth)
  - Public investment needs to increase to about 12% of GDP.
  - Oil fund floor binds over the entire simulation period in both scenarios; government resorts to large-scale external borrowing.
  - To keep external debt sustainable the government needs to increase taxes (labor tax and consumption tax).
  - In the adverse oil price scenario:
    - External debt increases by considerably more.
    - Labor income and consumption tax rates have to almost double in addition to higher taxes on the use of public capital.
    - Interest payments on foreign debt net of oil fund interest earnings reach about 1 percent of GDP.
  - Public capital: increases by nearly 60 percent above no scaling-up case by 2030, despite investment efficiency falling from 0.6 to 0.4.
  - Costs and trade-offs:
    - Private consumption falls substantially and for a long period, especially in the adverse scenario.
    - Growing external debt, exchange rate appreciation, and higher taxes negatively affect private nonoil sector.
    - Real exchange rate appreciation: about 6 percent in the adverse oil price scenario and about 8 percent in the baseline scenario.
    - Private investment in the nonoil tradable sector is considerably lower compared to the prudent path due to loss of competitiveness.
    - Higher borrowing costs for the private sector as public external debt increases risk premia.

### Key quantitative outcomes and indicators (preserved as reported)
- Public investment required:
  - Prudent path: about 6.4% of GDP.
  - Aggressive path: about 12% of GDP.
- Public capital change by 2030:
  - Prudent: almost 20% above no scaling-up.
  - Aggressive: nearly 60% above no scaling-up.
- Investment efficiency:
  - Steady-state φ = 0.6; falls to φ = 0.4 for additional investments above initial level under absorptive constraints.
- Oil fund floor: 30% of GDP.
- Oil price in adverse scenario: $51.7. a barrel in 2016; recover to $80 in 2019.
- External debt increase under prudent path (adverse oil price scenario): 5% of GDP over current comfortable level.
- Interest payments net of oil fund earnings under aggressive path (adverse scenario): about 1 percent of GDP.
- Real exchange rate appreciation (prudent): about 4 percent initially.
- Real exchange rate appreciation (aggressive): about 6 percent (adverse) and about 8 percent (baseline).
- Nonoil growth targets:
  - Prudent: average 0.3% above trend (adverse scenario growth increases by an average of 0.26% above trend).
  - Aggressive: average 0.6% above trend.
- Historical capital expenditures baseline: around 5% of GDP; prudent path implies increase by 1.4% of GDP if all additional investment is in budget.

### E. Conclusion — policy implications
- Trade-offs:
  - Ambitious (aggressive) scaling up of public investment can generate higher nonoil growth but at high funding costs: substantially lower private consumption and household welfare, larger external borrowing, higher taxes, and stronger exchange rate appreciation that harms tradables.
  - Prudent scaling up can boost nonoil growth with much smaller costs: manageable external borrowing, limited declines in investment efficiency, positive spillovers to private investment, and smaller adverse effects on traded sector competitiveness.
- Policy guidance implicit in model results:
  - Favor calibrated/prudent increases in public investment that respect absorptive capacity and maintain fiscal and external sustainability.
  - Monitor investment efficiency, absorptive capacity thresholds, and the oil fund floor to avoid prolonged binding of fund constraint and excessive external borrowing.
  - Recognize distributional and sectoral effects: public capital can crowd in nontraded sector activity but risks Dutch Disease effects on tradables via real exchange rate appreciation.

*Source: Box 1. Continued. Does Kazakhstan Have Dutch Disease? — IMF staff estimates and Kazakhstani Authorities (from the provided PDF content).*

### 34.      The analysis in this chapter stresses the benefits of adopting a comprehensive

### 34. The analysis in this chapter stresses the benefits of adopting a comprehensive approach to management of oil wealth

### Comprehensive management of oil wealth and fiscal metrics
- Key findings:
  - Adopt a comprehensive approach to management of oil wealth.
  - Given the large role of the quasifiscal sector in Kazakhstan, it is particularly important to look beyond traditional metrics of fiscal accounts.
  - Quasifiscal operations and contingent liabilities to the state should be appropriately taken into account.
  - To guard against Dutch Disease it is important to be mindful of the impact of policy decisions on the private sector.

### Development and diversification strategy
- Key findings and policy implications:
  - Careful design of Kazakhstan’s development and diversification strategy is important.
  - Productive investment into high quality public service delivery would ensure that Kazakhstan’s oil wealth benefits all.
  - Top-down industrialization policies are less likely to provide broad based benefits and carry the risk of further increasing the role of the state in the economy.
  - Good public financial management is key, including:
    - Careful cost-benefit analysis of various programs.
    - Evaluation of outcomes.
  - These elements are emphasized as central for prudent and effective management of oil wealth, consistent with IMF’s recent work on resource rich countries.

### Appendix 1. Kazakhstan: Calibration of Key Parameters
- Parameters, values, and notes:
  - α୒, α୘: labor income share in nontraded and traded sector — 0.5 — assumption
  - αୋ: output elasticity with respect to public capital — 0.15 — falls into the literature range of 0.05 – 0.2
  - ߜீ: steady-state annual depreciation rate of public capital — 0.07 — Berg et al. (2013)
  - ߩ௓், ߩ௒்: learning-by-doing parameter — 0.1 — Berg et al. (2013), mild externality
  - ߳: steady-state efficiency of public investment — 0.6 — between developed and developing countries, 0.3 – 1
  - ߳̅: lower efficiency when absorptive capacity is constrained — 0.4 — assumption, only for additional investment

### References (selected items listed in source)
- Arestoff, F. Hurlin, C., 2006. Estimates of Government Net Capital Stocks for 26 Developing Countries, 1970-2012. World Bank Policy Research Working Paper.
- Berg, A., Portillo, R. Yang, S.-C. S., Zanna, L.-F., 2013. Public Investment in Resource Abundant Developing Countries. IMF Economic Review 61 (1), 92-129.
- Buffie, E.F., Berg, A. , Patillio, C., Portillo, R., Zanna, L.-F., 2012. Public Investment, Growth and Debt Sustainability: Putting Together the Pieces. IMF Working Paper WP/12/177, International Monetary Fund.
- Other references in the source include analyses on Dutch Disease, Kazakhstan macroeconomic studies, and IMF country reports and working papers.

*REPUBLIC OF KAZAKHSTAN  INTERNATIONAL MONETARY FUND*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13291.pdf_