## Estimating Fiscal Multiplier for Qatar (sipea2025016)

## Source details

**Canonical URL:** [Estimating Fiscal Multiplier for Qatar (sipea2025016)](https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025016.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025016.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025016.pdf.json)

---

### A. Introduction — role of public spending and policy direction
- State spending built nation’s LNG production/export capacity and broader infrastructure, including Lusail, Hamad International Airport, Hamad Port, Doha Metro, and other transport and social infrastructure.
- Public sector expenditure accelerated in the early 1990s to commercialize natural gas via (i) LNG and GTL for global export, (ii) pipeline gas for regional export, and (iii) pipeline gas for domestic petrochemicals, power generation plants, and industrial consumption.
- The investment surge ahead of the 2022 FIFA World Cup is estimated to have driven much of non-hydrocarbon sector growth in the past decade.
- The State has undertaken spending-efficiency reforms targeting subsidies, travel and office expenses, non-core services, and GRE activities; capital expenditure proposals are closely scrutinized.
- The State is shifting its role toward enabling private sector-led growth (Third National Development Strategy, NDS3, released January 2024), using public spending to support reforms and human capital development.

### B. GCC panel approach — estimation strategy and data
- Baseline model specification:
  - Follows Espinoza and Senhadji (2011) and Fouejieu et al. (2018).
  - Dependent variable: real growth rate of non-hydrocarbon output.
  - Independent variables: real growth rates of central government spending (total, current, capital) and controls for global output and oil prices; country fixed effects included.
- Extensions and robustness:
  - Regressions estimated with and without the UAE to account for potential GRE effects.
  - Include indicator of capital stock relative to GDP (from Penn World Table) to test whether multipliers decline as capital stock rises.
  - Introduce Qatar country dummy to obtain Qatar-specific coefficients while exploiting GCC panel information.
- Data:
  - Annual data span 1990–2023 for 6 GCC countries; some 2023 data are IMF WEO estimates.
  - Stock of capital from Penn World Table available up to 2019 (2020–23 extrapolated).
  - Correlation analysis: relatively strong correlation between real growth rates of non-hydrocarbon output and central government spending in the GCC; country-level correlations most significant for Qatar, followed by Oman and Saudi Arabia; UAE correlation is negative.

### C. GCC panel approach — estimated results and key statistics
- Key multiplier estimates (baseline regressions; Table 2):
  - Current spending implied fiscal multiplier: 0.2 (model 2).
  - Capital spending implied long-term fiscal multiplier: close to 0.9 (model 3).
  - When current and capital spending included together, capital spending multiplier falls to 0.7 (model 4); current spending loses significance.
- Other findings:
  - Oil prices significant when introduced without global output growth; global output growth significant in models 1–3, 5.
  - Excluding the UAE strengthens effects of current and capital spending (models 6–9).
- Impact of capital stock level on capital spending multiplier (Table 3):
  - Capital stock-to-GDP categorized into low, medium, high (top 1/3 labeled “high”); capital stock relative to GDP rose over time except for the UAE.
  - When capital stock/GDP is high, coefficients on lagged capital spending become statistically insignificant — growth impact of capital expenditure is smaller when stock of capital is relatively high.
  - Results robust to lowering the threshold for “high” (top 1/2) and to excluding the UAE.
- GCC and Qatar-specific extensions (Table 4; models 16–19):
  - For Qatar, estimated elasticity (without distinguishing capital stock level) implies long-run fiscal multiplier close to 1.5 (insignificant for GCC overall).
  - When capital stock is relatively high in Qatar, capital spending does not have significant growth effects.
  - Excluding the UAE increases general GCC spending impact but Qatar-specific elasticities remain broadly unchanged.
- Selected numeric model diagnostics and sample details:
  - Number of observations in baseline GCC regressions: N = 175 (some models N = 145 when UAE excluded).
  - Example R-squared values reported: 0.2830, 0.2310, 0.2680, 0.1960, 0.2910, 0.3630, 0.2850, 0.3060, 0.365.
  - Global output coefficients frequently significant at conventional levels (e.g., 0.0141***).

### D. Qatar single-country approach — estimation strategy and data
- Data span and regression framework:
  - Complementary single-country analysis for Qatar uses longer data span 1983–2023 (including estimates).
  - Regressions: non-hydrocarbon output (real growth) on central government spending (total, current, capital), global output, and oil prices (all in real growth rates).
- Methods:
  - Static regressions.
  - Dynamic regressions: vector auto-regression (VAR) and local projection.
  - Time-varying growth effects estimated via the same static model on a 12-year rolling basis.
- Findings:
  - Static and dynamic regressions confirm growth effects of lagged capital spending are important in Qatar.
  - Rolling regressions indicate that in the recent decade, growth effects of capital spending waned coinciding with maturing of the infrastructure investment cycle associated with the World Cup.

### E. Growth effects of World Cup-related and public capital spending
- Observations:
  - World Cup-related capital spending in the 2010s appears to have had “renewed” growth effects, particularly with greater persistence gauging from the second lag.
  - These investments may reflect a more comprehensive infrastructure investment strategy and/or higher spending efficiency.
  - The growth effects moderated as the infrastructure investment cycle started to mature.
- Caveats:
  - Interpretations are suggestive because the model does not control for other potential determinants and the rolling approach further reduces the sample size.
- Text Figure 4 (description):
  - Measure: 12-year rolling coefficient and 68 percent confidence interval.
  - Data and model: Non-hydrocarbon output regressed on current spending (L0 and L1), capital spending (L1 and L2), global output, and oil prices, all in real growth rates, on a 12-year rolling basis ending in the year indicated, using annual data for Qatar spanning 1983–2023. Coefficients on the first and second lag of capital spending are displayed.

### F. Conclusions from empirical analysis
- Main conclusion: Qatar’s strong capital expenditure multiplier became less impactful when the stock of capital reached a high level.
- Consistency with broader evidence:
  - Aligns with results for a panel of GCC countries and with Fouejieu et al. (2018) that fiscal consolidation in the GCC may be less costly as fiscal multipliers declined over time, to the extent that the stock of capital generally rose over time.
- Relevant literature highlights:
  - Reallocation to non-wage spending within current spending envelope, particularly to education spending that builds human capital, and higher capital spending relative to current spending tend to boost economic growth (Gupta et al., 2005; and Acosta-Ormaechea and Morozumi, 2013).
  - Seemingly productive expenditures, when used in excess, could become unproductive (Devarajan et al., 1996).
  - Investments in new areas to support a knowledge-based economy with higher value-added sectors and to enhance climate sustainability and promote “green” growth could have significantly larger multipliers than traditional infrastructure investment (IMF, 2023b).
  - Growth effects of public spending tend to be more sizable, long-lasting, and stable when institutional quality is higher (Avellan et al., 2020).

### G. Policy implications and recommendations for Qatar
- Strategic shift supported by evidence:
  - Empirical evidence supports Qatar’s strategy to shift the State’s role to an enabler of private sector-led growth.
  - Diminishing marginal impact of capital spending as public capital stock rises argues for reallocation toward policies that support private investment and efficient use of existing assets.
- Public spending priorities:
  - Expenditure to support human capital accumulation.
  - Implementation of broader reforms guided by the Third National Development Strategy (NDS3) to facilitate private sector-led growth.
  - Measures that enhance investment efficiency and project selection, given reduced multipliers when capital stock is high.
- Specific priorities and actions:
  - Reorient spending to support knowledge-based growth to boost growth effects of fiscal spending.
  - Improve human capital, both for nationals and expatriates.
  - Provide a more conducive environment for businesses.
  - Enhance climate sustainability and continue to adapt to the energy transition.
  - Further investment in climate adaptation to mitigate vulnerabilities to climate stressors.
  - More investment to facilitate decarbonization and promote renewables to help Qatar reach its emission reduction target and smooth the energy transition process.
  - Crowd in private sector investment with efficient public spending to further economic diversification and accelerate the transition to private sector-driven growth.
- Fiscal policy design considerations:
  - Use multiplier estimates to gauge the non-hydrocarbon growth efficiency of different spending components (current vs. capital).
  - Be cautious about large-scale capital projects when capital stock/GDP is already high, as marginal growth returns may be limited.

### H. Annex I — Additional results (selected statistics)
- Data and methods:
  - Sources: Haver, IMF WEO, and IMF staff calculations.
  - Note: OLS results. The dependent variable is real growth of non-hydrocarbon output. The independent variables are all in real growth rates.
- Selected table entries (Annex I. Table 1. Qatar: Determinants of Non-Hydrocarbon Output 1/):
  - N39
  - R^20.4930.3740.5930.638
  - Global output coefficients (various model specifications): 01.1981.4041.973**1.790**
  - Oil prices coefficients: 0-0.068-0.072-0.068-0.082*
  - Capital spending coefficients (lags and significance markers shown in table)
  - Current spending coefficients (lags and significance markers shown in table)
  - Significance legend: * p<.1; ** p<.05; *** p<.01
- Annex I. Figure 1 (description):
  - Title: Qatar: Impact of Fiscal Expenditure on Non-hydrocarbon Output 1/
  - Methods compared: Local projection and Vector autoregression.
  - Endogenous variables: current expenditure, capital expenditure, and non-hydrocarbon output (all in real growth rates) in this order.
  - Exogenous variables: global output and oil prices (both in real growth rates).
  - IRFs show the impact of the mentioned variables on non-hydrocarbon output.
  - Sources: Haver, IMF WEO, and IMF staff calculations.

*Source: IMF Selected Issues Paper SIP/2025/016, “Estimating Fiscal Multiplier for Qatar,” prepared by Ken Miyajima (completed January 8, 2025); sipea2025016 - Section 2 (IMF staff text and annex material).*

### Section 1

### Estimating Fiscal Multiplier for Qatar

### A. Introduction — role of public spending and policy direction
- State spending built nation’s LNG production/export capacity and broader infrastructure, including Lusail, Hamad International Airport, Hamad Port, Doha Metro, and other transport and social infrastructure.
- Public sector expenditure accelerated in the early 1990s to commercialize natural gas via (i) LNG and GTL for global export, (ii) pipeline gas for regional export, and (iii) pipeline gas for domestic petrochemicals, power generation plants, and industrial consumption.
- The investment surge ahead of the 2022 FIFA World Cup is estimated to have driven much of non-hydrocarbon sector growth in the past decade.
- The State has undertaken spending-efficiency reforms targeting subsidies, travel and office expenses, non-core services, and GRE activities; capital expenditure proposals are closely scrutinized.
- The State is shifting its role toward enabling private sector-led growth (Third National Development Strategy, NDS3, released January 2024), using public spending to support reforms and human capital development.

### B. GCC panel approach — estimation strategy and data
- Baseline linear model follows Espinoza and Senhadji (2011) and Fouejieu et al. (2018): dependent variable is real growth rate of non-hydrocarbon output; independent variables are real growth rates of central government spending (total, current, capital) and controls for global output and oil prices; country fixed effects included.
- Regressions estimated with and without the UAE to account for potential GRE effects.
- Extensions:
  - Include indicator of capital stock relative to GDP (from Penn World Table) to test whether multipliers decline as capital stock rises.
  - Introduce Qatar country dummy to obtain Qatar-specific coefficients while exploiting GCC panel information.
- Data:
  - Annual data span 1990–2023 for 6 GCC countries; some 2023 data are IMF WEO estimates.
  - Stock of capital from Penn World Table available up to 2019 (2020–23 extrapolated).
  - Correlation analysis shows relatively strong correlation between real growth rates of non-hydrocarbon output and central government spending in the GCC; country-level correlations are most significant for Qatar, followed by Oman and Saudi Arabia; UAE correlation is negative.

### B. GCC panel approach — estimated results and key statistics
- Baseline regression findings (Table 2):
  - Current spending implied fiscal multiplier: 0.2 (model 2).
  - Capital spending implied long-term fiscal multiplier: close to 0.9 (model 3).
  - When current and capital spending included together, capital spending multiplier falls to 0.7 (model 4); current spending loses significance.
  - Oil prices significant when introduced without global output growth; global output growth significant in models 1–3, 5.
  - Excluding the UAE strengthens effects of current and capital spending (models 6–9).
- Impact of capital stock level on capital spending multiplier (Table 3):
  - Capital stock-to-GDP categorized into low, medium, high (top 1/3 labeled “high”); capital stock relative to GDP rose over time except for the UAE.
  - When capital stock/GDP is high, coefficients on lagged capital spending become statistically insignificant — growth impact of capital expenditure is smaller when stock of capital is relatively high.
  - Results robust to lowering the threshold for “high” (top 1/2) and to excluding the UAE.
- GCC and Qatar-specific extensions (Table 4; models 16–19):
  - For Qatar, estimated elasticity (without distinguishing capital stock level) implies long-run fiscal multiplier close to 1.5 (insignificant for GCC overall).
  - When capital stock is relatively high in Qatar, capital spending does not have significant growth effects.
  - Excluding the UAE increases general GCC spending impact but Qatar-specific elasticities remain broadly unchanged.
- Selected numeric model diagnostics and sample details:
  - Number of observations in baseline GCC regressions: N = 175 (some models N = 145 when UAE excluded).
  - Example R-squared values reported: 0.2830, 0.2310, 0.2680, 0.1960, 0.2910, 0.3630, 0.2850, 0.3060, 0.365.
  - Global output coefficients frequently significant at conventional levels (e.g., 0.0141***).

### C. Qatar single-country approach — estimation strategy and data
- Complementary single-country analysis for Qatar uses longer data span 1983–2023 (including estimates).
- Regressions: non-hydrocarbon output (real growth) on central government spending (total, current, capital), global output, and oil prices (all in real growth rates).
- Methods:
  - Static regressions.
  - Dynamic regressions: vector auto-regression (VAR) and local projection.
  - Time-varying growth effects estimated via the same static model on a 12-year rolling basis.
- Data and inference:
  - Static and dynamic regressions confirm growth effects of lagged capital spending are important in Qatar.
  - Rolling regressions indicate that in the recent decade, growth effects of capital spending waned coinciding with maturing of the infrastructure investment cycle associated with the World Cup.

### D. Policy implications and recommendations (drawn from empirical findings)
- Empirical evidence supports Qatar’s strategy to shift the State’s role to an enabler of private sector-led growth:
  - Diminishing marginal impact of capital spending as public capital stock rises argues for reallocation toward policies that support private investment and efficient use of existing assets.
- Public spending should prioritize:
  - Expenditure to support human capital accumulation.
  - Implementation of broader reforms guided by the Third National Development Strategy (NDS3) to facilitate private sector-led growth.
  - Measures that enhance investment efficiency and project selection, given reduced multipliers when capital stock is high.
- Considerations for fiscal policy design:
  - Use multiplier estimates to gauge the non-hydrocarbon growth efficiency of different spending components (current vs. capital).
  - Be cautious about large-scale capital projects when capital stock/GDP is already high, as marginal growth returns may be limited.

*Source: IMF Selected Issues Paper SIP/2025/016, “Estimating Fiscal Multiplier for Qatar,” prepared by Ken Miyajima (completed January 8, 2025).*

### Section 2

### sipea2025016 - Section 2

### Growth effects of World Cup-related and public capital spending
- World Cup-related capital spending in the 2010s appears to have had “renewed” growth effects, particularly with greater persistence gauging from the second lag (right panel).
- These investments may reflect a more comprehensive infrastructure investment strategy and/or with high spending efficiency.
- The growth effects moderated as the infrastructure investment cycle started to mature.
- Interpretations are suggestive because:
  - The model does not control for other potential determinants.
  - The rolling approach further reduces the sample size.
- Text Figure 4 description:
  - Title: Qatar: Non-hydrocarbon Growth Impact of Public Capital Spending 1/
  - Measure: 12-year rolling coefficient and 68 percent confidence interval.
  - Data and model: Non-hydrocarbon output is regressed on current spending (L0 and L1), capital spending (L1 and L2), global output, and oil prices, all in real growth rates, on a 12-year rolling basis ending in the year indicated on the horizontal scale, using annual data for Qatar spanning 1983–2023. Coefficients on the first and second lag of capital spending are displayed.
  - Sources: Haver and IMF staff.

### Conclusions from empirical analysis
- Main conclusion: Qatar’s strong capital expenditure multiplier became less impactful when the stock of capital reached a high level.
- Consistency: This finding aligns with results for a panel of GCC countries and with Fouejieu et al. (2018) that fiscal consolidation in the GCC may be less costly as fiscal multipliers declined over time, to the extent that the stock of capital generally rose over time.
- Relevant literature highlights:
  - Reallocation to non-wage spending within current spending envelope, particularly to education spending that builds human capital, and higher capital spending relative to current spending tend to boost economic growth (Gupta et al., 2005; and Acosta-Ormaechea and Morozumi, 2013).
  - Seemingly productive expenditures, when used in excess, could become unproductive (Devarajan et al., 1996).
  - Investments in new areas to support a knowledge-based economy with higher value-added sectors and to enhance climate sustainability and promote “green” growth could have significantly larger multipliers than traditional infrastructure investment (IMF, 2023b).
  - Growth effects of public spending tend to be more sizable, long-lasting, and stable when institutional quality is higher (Avellan et al., 2020).

### Policy implications and recommendations for Qatar
- Reorient spending to support knowledge-based growth to boost growth effects of fiscal spending.
- Focus of public investment given existing infrastructure and excess supply in some areas:
  - Improving human capital, both for nationals and expatriates.
  - Providing a more conducive environment for businesses.
  - Enhancing climate sustainability.
  - Continuing to adapt to the energy transition.
- Specific priorities:
  - Investment in human capital (education and health) is a welcome key pillar of NDS3 and there is scope to improve spending efficiency in Qatar.
  - Further investment in climate adaptation to mitigate vulnerabilities to climate stressors.
  - More investment to facilitate decarbonization and promote renewables to help Qatar reach its emission reduction target and smooth the energy transition process.
  - Crowding in private sector investment with efficient public spending to further economic diversification and accelerate the transition to private sector-driven growth.

### Annex I — Additional results (selected statistics)
- Data and methods:
  - Sources: Haver, IMF WEO, and IMF staff calculations.
  - Note: OLS results. The dependent variable is real growth of non-hydrocarbon output. The independent variables are all in real growth rates.
- Selected table entries (Annex I. Table 1. Qatar: Determinants of Non-Hydrocarbon Output 1/):
  - N39
  - R^20.4930.3740.5930.638
  - Global output coefficients (various model specifications): 01.1981.4041.973**1.790**
  - Oil prices coefficients: 0-0.068-0.072-0.068-0.082*
  - Capital spending coefficients (lags and significance markers shown in table)
  - Current spending coefficients (lags and significance markers shown in table)
  - Significance legend: * p<.1; ** p<.05; *** p<.01
- Annex I. Figure 1 description:
  - Title: Qatar: Impact of Fiscal Expenditure on Non-hydrocarbon Output 1/
  - Methods compared: Local projection and Vector autoregression.
  - Endogenous variables: current expenditure, capital expenditure, and non-hydrocarbon output (all in real growth rates) in this order.
  - Exogenous variables: global output and oil prices (both in real growth rates).
  - IRFs show the impact of the mentioned variables on non-hydrocarbon output.
  - Sources: Haver, IMF WEO, and IMF staff calculations.

*Source: sipea2025016 - Section 2 (IMF staff text and annex material).*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025016.pdf_