## 2.   Online Annex. From Setbacks to Comebacks: Structural Reforms to Build Resilience and Prosperity

## Source details

**Canonical URL:** [2.   Online Annex. From Setbacks to Comebacks: Structural Reforms to Build Resilience and Prosperity](https://www.imf.org/-/media/files/publications/reo/mcd-cca/2023/october/english/ch2onlineannexes.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/mcd-cca/2023/october/english/ch2onlineannexes.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/mcd-cca/2023/october/english/ch2onlineannexes.pdf.json)

---

### Annex 2.1. Measuring Policy Space in the Region
- Methodology
  - Policy space index constructed by applying principal component analysis on normalized indicators listed in Annex Table 1.1.
  - Principal component analysis converts several variables into a single series—the principal component—while retaining as much information as possible from all variables included in the analysis.
  - Variable loadings, correlations, and the index itself are based on scores from principal component 1.
  - All indicators are annual and sourced from the World Economic Outlook database for the years between 2000 and 2022, as available for countries in the Middle East and Central Asia.
  - Growth and inflation enter the index as deviations from long-term averages, which are calculated using 10-year rolling averages so that for each year, the value of the long-term average is based on the preceding 10 years.
  - Public debt is assessed relative to the IMF–World Bank Debt Sustainability Analysis benchmarks for emerging market economies and low-income countries, respectively.
  - Underlying indicators are defined such that higher values mean more limited policy space; therefore, the index can be interpreted as measuring policy constraints.
- Underlying variables (Variable Description — WEO Series Code)
  - Public debt deviation: Public debt ratio deviation from Debt Sustainability Analysis benchmarks (70 percent for emerging markets and 55 percent for low-income countries) in percent of GDP — GGD_GDP
  - Public external debt to total debt ratio: Public external debt to total public debt ratio; deviation from Debt Sustainability Analysis benchmark for the ratio — DG_GDP, GGD_GDP
  - Exchange market pressure: Sum of the exchange rate deprecation (in percent) and reserve outflows (percent of base money) — EDNA, FAFAR, FALM
  - GDP growth deviation: Growth deviation from long-term average (percent) multiplied by -1 — NGDP_R_PCH
  - Inflation deviation: Inflation deviation from long-term average — PCPI_PCH
- Summary statistics by year (Annex Table 1.2)
  - 2010: N 22; Mean -0.11; SD 0.79; Min -2.04; Max 1.20
  - 2011: N 23; Mean -0.01; SD 1.18; Min -3.06; Max 2.34
  - 2012: N 23; Mean -0.10; SD 1.20; Min -2.83; Max 3.90
  - 2013: N 23; Mean -0.18; SD 0.85; Min -2.08; Max 2.06
  - 2014: N 25; Mean 0.08; SD 0.80; Min -1.22; Max 1.45
  - 2015: N 25; Mean 0.56; SD 0.93; Min -1.15; Max 2.99
  - 2016: N 26; Mean 0.42; SD 0.71; Min -0.99; Max 1.80
  - 2017: N 26; Mean 0.45; SD 0.90; Min -0.95; Max 3.00
  - 2018: N 26; Mean 0.41; SD 1.10; Min -1.13; Max 4.60
  - 2019: N 26; Mean 0.52; SD 1.23; Min -0.82; Max 4.02
  - 2020: N 26; Mean 1.33; SD 1.84; Min -0.52; Max 7.73
  - 2021: N 24; Mean 0.39; SD 1.75; Min -1.13; Max 8.16
  - 2022: N 24; Mean 0.17; SD 0.94; Min -1.32; Max 3.19

### Annex 2.2. World Governance Indicators
- Data and coverage
  - World Governance Indicators (WGI) project reports aggregate and individual governance indicators for over 200 countries and territories over the period 1996–2021 for six dimensions of governance.
- Six WGI dimensions (as reported)
  - Voice and Accountability
  - Political Stability and Absence of Violence/Terrorism
  - Government Effectiveness
  - Regulatory Quality
  - Rule of Law
  - Control of Corruption
- Regional series presented (labels in figure)
  - ME&CA, MENAP, MENA, MENA ex GCC, GCC, CCA, Rest of the world (RHS scale)
- Figure note
  - Visual index series plotted for 1996–2021; numeric labels and axes preserved in source figure.

### Annex 2.3. Structural Reform Indicator
- Overview
  - Structural reform indicators draw on the Fraser Institute’s Economic Freedom Database and the World Governance Indicators database, with component-specific constructions described below.
- 2.3.1 Labor
  - Data source: Fraser Institute’s Economic Freedom Database.
  - Index composition: Simple average of two components—Hiring and Firing Regulations and Centralized Collective Bargaining.
- 2.3.2 External Sector
  - Data source: Fraser Institute’s Economic Freedom Database.
  - Index composition: Simple average of four sub-indicators:
    - Tariffs (tariff revenues, tariff rate and volatility of tariffs)
    - Non-tariff trade barriers
    - Black-market exchange rate (disparity between official and parallel exchange rates)
    - Control of the movement of capital and people (degree of financial openness, restrictions to visitors, capital controls)
- 2.3.3 Credit Market
  - Data source: Fraser Institute’s Economic Freedom Database.
  - Index composition: Three components:
    - Ownership of banks (extent to which bank deposits are held in privately owned financial institutions)
    - Private sector credit (extent of government borrowing relative to private-sector borrowing; higher score for more private sector borrowing)
    - Interest rate controls (market-determined interest rates, stable monetary policy, low real-deposit and lending-rate spreads receive higher ratings)
- 2.3.4 Regulatory Quality
  - Data source: World Governance Indicators database.
  - Use: Regulatory Quality used as a proxy for business regulations/regulatory quality data; measures government ability to formulate and implement policies that permit and promote private sector development.
- 2.3.5 Governance
  - Index composition: Simple average of the six WGI components:
    - Voice and accountability
    - Political stability and absence of violence/terrorism
    - Government effectiveness
    - Regulatory quality
    - Rule of law
    - Control of corruption
- 2.3.6 Gender
  - Index: Gender Legislation Index derived from Cardarelli, Martin, and Lall (2022) based on the World Bank’s Women, Business and the Law (WBL) database (35 variables).
  - Construction: Standardize each sub-indicator to vary between zero and one, take unweighted average, and scale result to 100; a higher value points to less favorable conditions for female labor force participation (LFP) and employment.
  - Coverage: Includes legislation on mobility, workplace and pay conditions, marriage, parental leave, entrepreneurship, access to financial services, restrictions on mobility and household legal position (domestic violence, divorce, right to remarry).

### Annex 2.4. Calculating Non-oil Labor Productivity
- Definition
  - Labor productivity defined as GDP in constant USD prices per worker.
- Data and adjustments
  - Oil and non-oil GDP in constant local prices are available in the World Economic Outlook database.
  - Figures adjusted using the implied PPP conversion rate (national currency per international dollar).
  - Non-oil employment proxied by the International Labour Organization modeled estimates of employment in mining and quarrying (ISIC Rev. 4 Code B), which includes extraction of coal and ores, petroleum, and natural gas.
- Figure (Annex Figure 4.1) notes
  - Series units: Thousands, constant PPP USD.
  - Time range plotted: 2010–2021.
  - Country/label codes shown on the figure: GCCBHRKWTQATARESAUOMN

### Annex 2.5. Additional Case Studies
- Jordan: Trade and Private Sector Led Reforms (2000–04)
  - Reforms and context
    - Post-1999 reform initiative focused on trade liberalization and privatization, addressing inefficiencies, corruption, and indebtedness in SOEs.
    - Free trade agreements and accession: U.S. FTA signed October 2000; Association Agreement with the European Union signed November 1997 (effective May 2002); WTO accession in April 2000.
    - Preferential market access schemes: Qualified Industrial Zones and Special Economic Zones improved export performance.
  - Social and Economic Transformation Program (2002)
    - Emphasized private capital mobilization, human capital development, enhancement of government services, and community development.
  - Vision 2025 and subsequent developments
    - Vision 2025 launched in 2015 with pillars on governance, business environment, trade competitiveness, and social outcomes.
    - Trade facilitation and labor market reforms included enhanced trade infrastructure, addressing non-tariff barriers, upskilling public sector workers, labor market information system, fostering female labor force participation, and vocational training coordination.
    - IMF programs: Jordan entered an IMF program in 2016–19 and another in 2020 to support balance of payments and reforms.
    - Financial inclusion and secured lending: Financial Inclusion Strategy and Secured Lending Law enacted during 2016–18.
  - Recent recovery and modernization
    - Post-COVID-19 recovery: real GDP growing at an average of 2.4 percent over 2021–22.
    - Jordan’s Economic Modernization Vision launched mid-2022 expected to spur growth closer to 3 percent by 2024.
    - Unemployment rate: estimated at 22.9 percent in the fourth quarter of 2022.
    - Reform priorities: public sector administrative reforms including digitalizing government services; strengthening governance and transparency via Integrity and Anti-Corruption Commission; labor market reforms to support women’s participation, amend labor law, improve labor market flexibility, incentives to hire youth, encourage transition to formality; improve business environment via competition regulatory and legal framework, reduce entry barriers, and improve trade infrastructure.
- Kazakhstan: Reforms to Stem Against High Volatility (2013–15)
  - Context and actions
    - Reforms implemented amid sanctions on Russia in 2014, lower oil prices, and weaker demand from China and the European Union.
    - Progress on anti-corruption strategy, monetary and exchange rate framework, and business environment between 2013 and 2015.
    - Launch of “100 steps” wide-ranging reform program following presidential election in early 2015 to improve public administration efficiency, diversify the economy, and address governance challenges.
    - June 2015: completed negotiations to become a formal member of the World Trade Organization.
  - Outcomes
    - Improved governance effectiveness and regulatory quality, moving Kazakhstan from the third to the fourth quintile on respective WGI score distribution.
    - Many reforms were incremental and incomplete; the state footprint remains high.
- Armenia: High-Level Anti-corruption Efforts to Facilitate Wide-Ranging Reforms (2017–20)
  - Context and actions
    - Parliamentary elections in December 2018 won on platform of tackling corruption and reforming the economy.
    - Extensive anti-corruption campaign; anti-corruption strategy for 2019–23 introduced reforms to enhance courts, prosecutor’s office, and investigative bodies.
    - November 2017: signed Comprehensive and Enhanced Partnership Agreement with the European Union to improve investment climate and business environment.
    - January 2020: made audit and publication of SOE financial reports mandatory.
  - Outcomes and remaining gaps
    - Improvements in voice and accountability and control of corruption: rising from below the median to the top 30 and 20 percent in the global sample, respectively.
    - Remaining unfinished reforms: SOE accountability, fiscal oversight, enforcement of corporate governance and financial accountability requirements.
- Uzbekistan: High-Impact Economic Reforms Followed by a Deep Structural Reform Agenda (2017–Present)
  - Initial reform package (late 2016–2019)
    - Economic modernization program launched by President Shavkat Mirziyoyev in late 2016 emphasizing FX liberalization, tax reform, and improved economic statistics.
    - FX liberalization: unified official and parallel exchange rates at a heavily depreciated rate and eliminated all FX restrictions subject to IMF jurisdiction.
    - Tax reform: aimed to foster job creation by reducing the tax burden on private firms and workers.
    - Statistical improvements: availability and quality of economic statistics improved substantially.
    - Outcomes: credit and investment surged; number of firms in the standard tax net increased five-fold through 2019.
    - Initial steps taken against corruption.
  - COVID-19 period and subsequent reforms
    - Pandemic slowed reform implementation but progress included liberalization of cotton and wheat prices; reduced corruption opportunities via improved online procurement, digitalization, and increased transparency; substantial expansion of social safety net; start of privatization efforts; reduction in preferential lending; opening more activities to private sector participation while curbing SOE privileges.
    - Incomes improved over the past five years but remain below those of other emerging economies.
    - Remaining challenges: large state footprint and large informal sector; need to further reduce the role of the state, open markets, enhance competition, and improve governance and rule of law.
    - Post-2021: Following re-election in 2021, a new five-year development strategy was adopted in 2022 to continue deeper economic and social reforms.
  - Data note
    - Data for Uzbekistan is not available for the structural reform indicators discussed in this chapter; econometric results presented elsewhere do not include the potential positive impact of structural reforms in Uzbekistan.

### Annex 2.6. Estimations of the Impact of Structural Reforms on Outcome
- Estimation methodology
  - Sample:
    - 27 countries in the ME&CA region.
    - Period: 2000 to 2021 at annual frequency.
    - Countries included (based on data availability): Algeria, Armenia, Azerbaijan, Bahrain, Djibouti, Egypt, Georgia, Islamic Republic of Iran, Iraq, Jordan, Kazakhstan, Kuwait, Kyrgyz Republic, Lebanon, Libya, Mauritania, Morocco, Oman, Pakistan, Qatar, Saudi Arabia, Somalia, Sudan, Tajikistan, Tunisia, United Arab Emirates, and Yemen.
  - Baseline empirical framework:
    - Local projection method (Jordà, 2005), aligned with the SPR-RES Staff Discussion Note on structural reforms and following IMF (2019).
    - Baseline specification (equation 1.1):
      - y_{i,t+k} − y_{i,t−1} = β_k R_{i,t} + θ_{i,t} + α_i + γ_t + ε_{i,t}
      - y_{i,t} is the log of output (or investment, employment, or labor productivity).
      - t and i are time and country indices.
      - k = 0,1,2,...5.
      - R_{i,t} denotes the structural reform variable, defined as the change in the structural indicator.
      - θ_{i,t} is a set of control variables, including lags of the dependent variable and past reforms.
      - α_i denotes country fixed effects.
      - γ_t denotes percentage change in oil prices.
      - Standard errors: robust standard errors clustered at the country-level.
  - Regime-dependent specification:
    - Modified specification (equation 1.2) to allow responses to vary with a conditioning variable z_{i,t} (e.g., growth or Policy Space index):
      - y_{i,t+k} − y_{i,t−1} = β^L_k F(z_{i,t}) R_{i,t} + β^H_k [1 − F(z_{i,t})] R_{i,t} + θ_{i,t} + α_i + γ_t + ε_{i,t}
      - F(z_{i,t}) = exp(−γ z_{i,t})/(1+exp(−γ z_{i,t})), with γ = 1.5.
      - z_{i,t} is normalized to have zero mean and unit variance.
      - F(z_{i,t}) and 1 − F(z_{i,t}) interpreted as the probability of being in the “low” regime (e.g., low growth or limited policy space) and the “high” regime, respectively.
      - β^L_k and β^H_k are the structural reform multipliers in low and high regimes.
    - Equations 1.1 and 1.2 are estimated for each k = 0,1,2,...5.
    - Impulse response functions computed using estimated β_k (or β^L_k and β^H_k); confidence bands derived from estimated standard errors.
  - Controls and conditioning:
    - Some conditioning variables are also included in θ_{i,t} as control variables (for example, lagged growth and structural developments are included as controls).
  - Purpose and advantages:
    - Permits direct testing of whether reform effects vary across business cycle phases (e.g., slowdowns defined as positive output growth below a threshold) and across different policy spaces.
    - Allows effects to change smoothly by considering regime probabilities rather than splitting the sample into discrete regimes, improving stability and precision relative to separate SVARs by cycle.
- Sequencing of reforms
  - Approach to sequencing:
    - Use first-generation reform index (an average of governance, external sector and regulatory quality reform indices) as the conditioning variable z_{i,t}.
    - Purpose: examine whether impulse responses of other reforms (e.g., credit market reforms) vary by the level of first-generation structural development.
  - Interpretation:
    - If impulse responses of growth from other reforms are larger in countries with higher first-generation structural development, this implies first-generation reform should be prioritized to lay a foundation and enhance gains from subsequent reforms.
- Presentation of results (figures and interpretation details)
  - Impulse-response presentation:
    - Responses shown for a “major historical reform” defined as two standard deviations of the annual change in the structural index.
    - Time scale: x-axis measured in years where t = 0 is the first year the reform is implemented; k ranges up to 5 years.
    - Shaded areas denote 90 percent confidence bands.
  - Heterogeneity exercises shown in figures:
    - Baseline average effects of reforms on Investment, Employment, and Labor Productivity for reform categories: Governance; External Sector; Credit Market; Regulatory Quality; Labor Market.
    - Effects under weaker governance versus baseline (Annex Figure 6.2):
      - Blue line: baseline (sample-average) response to a major historical reform; shaded 90 percent confidence bands.
      - Red line: response when governance is weaker; solid (dashed) segments denote statistically significant (insignificant) responses at the 90 percent level.
    - Effects under low growth versus baseline (Annex Figure 6.3):
      - Blue line: baseline response to a major historical reform; shaded 90 percent confidence bands.
      - Red line: response when growth is low; solid (dashed) segments denote statistically significant (insignificant) responses at the 90 percent level.
    - Effects under limited policy space versus baseline (Annex Figure 6.4):
      - Blue line: baseline response to a major historical reform; shaded 90 percent confidence bands.
      - Red line: response when policy space is limited; solid (dashed) segments denote statistically significant (insignificant) responses at the 90 percent level.
  - Visualization notes:
    - Figures plot percent changes on the vertical axis; horizontal axis in years with t = 0 as reform implementation year.
    - Responses computed for investment, employment, and labor productivity by reform type and by regime (weaker governance, low growth, limited policy space).

*Source: ch2onlineannexes - 2.   Online Annex. From Setbacks to Comebacks: Structural Reforms to Build Resilience and Prosperity (PDF).*

### 2.   Online Annex. From Setbacks to Comebacks:

### 2.   Online Annex. From Setbacks to Comebacks: Structural Reforms to Build Resilience and Prosperity

### Annex 2.1. Measuring Policy Space in the Region
- Methodology
  - The policy space index is constructed by applying principal component analysis on normalized versions of the indicators listed in Annex Table 1.1.
  - Principal component analysis converts several variables into a single series—the principal component—while retaining as much information as possible from all variables included in the analysis.
  - Variable loadings, correlations, and the index itself are based on scores from principal component 1.
  - All indicators are annual and sourced from the World Economic Outlook database for the years between 2000 and 2022, as available for countries in the Middle East and Central Asia.
  - Growth and inflation enter the index as deviations from long-term averages, which are calculated using 10-year rolling averages so that for each year, the value of the long-term average is based on the preceding 10 years.
  - Public debt is assessed relative to the IMF–World Bank Debt Sustainability Analysis benchmarks for emerging market economies and low-income countries, respectively.
  - The underlying indicators are defined such that higher values mean more limited policy space; therefore, the index can be interpreted as measuring policy constraints.
- Underlying variables (Variable Description — WEO Series Code)
  - Public debt deviation: Public debt ratio deviation from Debt Sustainability Analysis benchmarks (70 percent for emerging markets and 55 percent for low-income countries) in percent of GDP — GGD_GDP
  - Public external debt to total debt ratio: Public external debt to total public debt ratio; deviation from Debt Sustainability Analysis benchmark for the ratio — DG_GDP, GGD_GDP
  - Exchange market pressure: Sum of the exchange rate deprecation (in percent) and reserve outflows (percent of base money) — EDNA, FAFAR, FALM
  - GDP growth deviation: Growth deviation from long-term average (percent) multiplied by -1 — NGDP_R_PCH
  - Inflation deviation: Inflation deviation from long-term average — PCPI_PCH
- Summary statistics by year (Annex Table 1.2)
  - 2010: N 22; Mean -0.11; SD 0.79; Min -2.04; Max 1.20
  - 2011: N 23; Mean -0.01; SD 1.18; Min -3.06; Max 2.34
  - 2012: N 23; Mean -0.10; SD 1.20; Min -2.83; Max 3.90
  - 2013: N 23; Mean -0.18; SD 0.85; Min -2.08; Max 2.06
  - 2014: N 25; Mean 0.08; SD 0.80; Min -1.22; Max 1.45
  - 2015: N 25; Mean 0.56; SD 0.93; Min -1.15; Max 2.99
  - 2016: N 26; Mean 0.42; SD 0.71; Min -0.99; Max 1.80
  - 2017: N 26; Mean 0.45; SD 0.90; Min -0.95; Max 3.00
  - 2018: N 26; Mean 0.41; SD 1.10; Min -1.13; Max 4.60
  - 2019: N 26; Mean 0.52; SD 1.23; Min -0.82; Max 4.02
  - 2020: N 26; Mean 1.33; SD 1.84; Min -0.52; Max 7.73
  - 2021: N 24; Mean 0.39; SD 1.75; Min -1.13; Max 8.16
  - 2022: N 24; Mean 0.17; SD 0.94; Min -1.32; Max 3.19

### Annex 2.2. World Governance Indicators
- Data and coverage
  - The World Governance Indicators (WGI) project reports aggregate and individual governance indicators for over 200 countries and territories over the period 1996–2021 for six dimensions of governance.
- Six WGI dimensions (as reported)
  - Voice and Accountability
  - Political Stability and Absence of Violence/Terrorism
  - Government Effectiveness
  - Regulatory Quality
  - Rule of Law
  - Control of Corruption
- Regional series presented (labels in figure)
  - ME&CA, MENAP, MENA, MENA ex GCC, GCC, CCA, Rest of the world (RHS scale)
- (Figure note: visual index series plotted for 1996–2021; numeric labels and axes preserved in source figure.)

### Annex 2.3. Structural Reform Indicator
- Overview: Structural reform indicators draw on the Fraser Institute’s Economic Freedom Database and the World Governance Indicators database, with component-specific constructions described below.
- 2.3.1 Labor
  - Data source: Fraser Institute’s Economic Freedom Database.
  - Index composition: Simple average of two components—Hiring and Firing Regulations and Centralized Collective Bargaining.
- 2.3.2 External Sector
  - Data source: Fraser Institute’s Economic Freedom Database.
  - Index composition: Simple average of four sub-indicators:
    - Tariffs (tariff revenues, tariff rate and volatility of tariffs)
    - Non-tariff trade barriers
    - Black-market exchange rate (disparity between official and parallel exchange rates)
    - Control of the movement of capital and people (degree of financial openness, restrictions to visitors, capital controls)
- 2.3.3 Credit Market
  - Data source: Fraser Institute’s Economic Freedom Database.
  - Index composition: Three components:
    - Ownership of banks (extent to which bank deposits are held in privately owned financial institutions)
    - Private sector credit (extent of government borrowing relative to private-sector borrowing; higher score for more private sector borrowing)
    - Interest rate controls (market-determined interest rates, stable monetary policy, low real-deposit and lending-rate spreads receive higher ratings)
- 2.3.4 Regulatory Quality
  - Data source: World Governance Indicators database.
  - Use: Regulatory Quality used as a proxy for business regulations/regulatory quality data; measures government ability to formulate and implement policies that permit and promote private sector development.
- 2.3.5 Governance
  - Index composition: Simple average of the six WGI components:
    - Voice and accountability
    - Political stability and absence of violence/terrorism
    - Government effectiveness
    - Regulatory quality
    - Rule of law
    - Control of corruption
- 2.3.6 Gender
  - Index: Gender Legislation Index derived from Cardarelli, Martin, and Lall (2022) based on the World Bank’s Women, Business and the Law (WBL) database (35 variables).
  - Construction: Standardize each sub-indicator to vary between zero and one, take unweighted average, and scale result to 100; a higher value points to less favorable conditions for female labor force participation (LFP) and employment.
  - Coverage: Includes legislation on mobility, workplace and pay conditions, marriage, parental leave, entrepreneurship, access to financial services, restrictions on mobility and household legal position (domestic violence, divorce, right to remarry).

### Annex 2.4. Calculating Non-oil Labor Productivity
- Definition
  - Labor productivity is defined as GDP in constant USD prices per worker.
- Data and adjustments
  - Oil and non-oil GDP in constant local prices are available in the World Economic Outlook database.
  - Figures are adjusted using the implied PPP conversion rate (national currency per international dollar).
  - Non-oil employment is proxied by the International Labour Organization modeled estimates of employment in mining and quarrying (ISIC Rev. 4 Code B), which includes extraction of coal and ores, petroleum, and natural gas.
- Figure (Annex Figure 4.1) notes
  - Series units: Thousands, constant PPP USD.
  - Time range plotted: 2010–2021.
  - Country/label codes shown on the figure: GCCBHRKWTQATARESAUOMN

### Annex 2.5. Additional Case Studies
- Jordan: Trade and Private Sector Led Reforms (2000–04)
  - Reforms and context
    - Post-1999 reform initiative focused on trade liberalization and privatization, addressing inefficiencies, corruption, and indebtedness in SOEs.
    - Free trade agreements and accession: U.S. FTA signed October 2000; Association Agreement with the European Union signed November 1997 (effective May 2002); WTO accession in April 2000.
    - Preferential market access schemes: Qualified Industrial Zones and Special Economic Zones improved export performance.
  - Social and Economic Transformation Program (2002)
    - Emphasized private capital mobilization, human capital development, enhancement of government services, and community development.
  - Vision 2025 and subsequent developments
    - Vision 2025 launched in 2015 with pillars on governance, business environment, trade competitiveness, and social outcomes.
    - Trade facilitation and labor market reforms included enhanced trade infrastructure, addressing non-tariff barriers, upskilling public sector workers, labor market information system, fostering female labor force participation, and vocational training coordination.
    - IMF programs: Jordan entered an IMF program in 2016–19 and another in 2020 to support balance of payments and reforms.
    - Financial inclusion and secured lending: Financial Inclusion Strategy and Secured Lending Law enacted during 2016–18.
  - Recent recovery and modernization
    - Post-COVID-19 recovery: real GDP growing at an average of 2.4 percent over 2021–22.
    - Jordan’s Economic Modernization Vision launched mid-2022 expected to spur growth closer to 3 percent by 2024.
    - Unemployment rate: estimated at 22.9 percent in the fourth quarter of 2022.
    - Reform priorities: public sector administrative reforms including digitalizing government services; strengthening governance and transparency via Integrity and Anti-Corruption Commission; labor market reforms to support women’s participation, amend labor law, improve labor market flexibility, incentives to hire youth, encourage transition to formality; improve business environment via competition regulatory and legal framework, reduce entry barriers, and improve trade infrastructure.
- Kazakhstan: Reforms to Stem Against High Volatility (2013–15)
  - Context and actions
    - Reforms implemented amid sanctions on Russia in 2014, lower oil prices, and weaker demand from China and the European Union.
    - Progress on anti-corruption strategy, monetary and exchange rate framework, and business environment between 2013 and 2015.
    - Launch of “100 steps” wide-ranging reform program following presidential election in early 2015 to improve public administration efficiency, diversify the economy, and address governance challenges.
    - June 2015: completed negotiations to become a formal member of the World Trade Organization.
  - Outcomes
    - Improved governance effectiveness and regulatory quality, moving Kazakhstan from the third to the fourth quintile on respective WGI score distribution.
    - Many reforms were incremental and incomplete; the state footprint remains high.
- Armenia: High-Level Anti-corruption Efforts to Facilitate Wide-Ranging Reforms (2017–20)
  - Context and actions
    - Parliamentary elections in December 2018 won on platform of tackling corruption and reforming the economy.
    - Extensive anti-corruption campaign; anti-corruption strategy for 2019–23 introduced reforms to enhance courts, prosecutor’s office, and investigative bodies.
    - November 2017: signed Comprehensive and Enhanced Partnership Agreement with the European Union to improve investment climate and business environment.
    - January 2020: made audit and publication of SOE financial reports mandatory.
  - Outcomes and remaining gaps
    - Improvements in voice and accountability and control of corruption: rising from below the median to the top 30 and 20 percent in the global sample, respectively.
    - Remaining unfinished reforms: SOE accountability, fiscal oversight, enforcement of corporate governance and financial accountability requirements.
- Uzbekistan: High-Impact Economic Reforms Followed by a Deep Structural Reform Agenda (2017–Present)
  - Initial reform package (late 2016–2019)
    - Economic modernization program launched by President Shavkat Mirziyoyev in late 2016 emphasizing FX liberalization, tax reform, and improved economic statistics.
    - FX liberalization: unified official and parallel exchange rates at a heavily depreciated rate and eliminated all FX restrictions subject to IMF jurisdiction.
    - Tax reform: aimed to foster job creation by reducing the tax burden on private firms and workers.
    - Statistical improvements: availability and quality of economic statistics improved substantially.
    - Outcomes: credit and investment surged; number of firms in the standard tax net increased five-fold through 2019.
    - Initial steps taken against corruption.
  - COVID-19 period and subsequent reforms
    - Pandemic slowed reform implementation but progress included liberalization of cotton and wheat prices; reduced corruption opportunities via improved online procurement, digitalization, and increased transparency; substantial expansion of social safety net; start of privatization efforts; reduction in preferential lending; opening more activities to private sector participation while curbing SOE privileges.
    - Incomes improved over the past five years but remain below those of other emerging economies.
    - Remaining challenges: large state footprint and large informal sector; need to further reduce the role of the state, open markets, enhance competition, and improve governance and rule of law.
    - Post-2021: Following re-election in 2021, a new five-year development strategy was adopted in 2022 to continue deeper economic and social reforms.
  - Data note
    - Data for Uzbekistan is not available for the structural reform indicators discussed in this chapter; econometric results presented elsewhere do not include the potential positive impact of structural reforms in Uzbekistan.

*Source: ch2onlineannexes - 2.   Online Annex. From Setbacks to Comebacks: Structural Reforms to Build Resilience and Prosperity (PDF).*

### Annex 2.6. Estimations of the Impact of Structural Reforms on Outcome

### Annex 2.6. Estimations of the Impact of Structural Reforms on Outcome

### Estimation methodology
- Sample:
  - 27 countries in the ME&CA region.
  - Period: 2000 to 2021 at annual frequency.
  - Countries included (based on data availability): Algeria, Armenia, Azerbaijan, Bahrain, Djibouti, Egypt, Georgia, Islamic Republic of Iran, Iraq, Jordan, Kazakhstan, Kuwait, Kyrgyz Republic, Lebanon, Libya, Mauritania, Morocco, Oman, Pakistan, Qatar, Saudi Arabia, Somalia, Sudan, Tajikistan, Tunisia, United Arab Emirates, and Yemen.
- Baseline empirical framework:
  - Local projection method (Jordà, 2005), aligned with the SPR-RES Staff Discussion Note on structural reforms and following IMF (2019).
  - Baseline specification (equation 1.1):  
    y_{i,t+k} − y_{i,t−1} = β_k R_{i,t} + θ_{i,t} + α_i + γ_t + ε_{i,t}
    - y_{i,t} is the log of output (or investment, employment, or labor productivity).
    - t and i are time and country indices.
    - k = 0,1,2,...5.
    - R_{i,t} denotes the structural reform variable, defined as the change in the structural indicator.
    - θ_{i,t} is a set of control variables, including lags of the dependent variable and past reforms.
    - α_i denotes country fixed effects.
    - γ_t denotes percentage change in oil prices.
    - Standard errors: robust standard errors clustered at the country-level.
- Regime-dependent specification:
  - Modified specification (equation 1.2) to allow responses to vary with a conditioning variable z_{i,t} (e.g., growth or Policy Space index):
    y_{i,t+k} − y_{i,t−1} = β^L_k F(z_{i,t}) R_{i,t} + β^H_k [1 − F(z_{i,t})] R_{i,t} + θ_{i,t} + α_i + γ_t + ε_{i,t}
    - F(z_{i,t}) = exp(−γ z_{i,t})/(1+exp(−γ z_{i,t})), with γ = 1.5.
    - z_{i,t} is normalized to have zero mean and unit variance.
    - F(z_{i,t}) and 1 − F(z_{i,t}) interpreted as the probability of being in the “low” regime (e.g., low growth or limited policy space) and the “high” regime, respectively.
    - β^L_k and β^H_k are the structural reform multipliers in low and high regimes.
  - Equations 1.1 and 1.2 are estimated for each k = 0,1,2,...5.
  - Impulse response functions computed using estimated β_k (or β^L_k and β^H_k); confidence bands derived from estimated standard errors.
- Controls and conditioning:
  - Some conditioning variables are also included in θ_{i,t} as control variables (for example, lagged growth and structural developments are included as controls).
- Purpose and advantages:
  - Permits direct testing of whether reform effects vary across business cycle phases (e.g., slowdowns defined as positive output growth below a threshold) and across different policy spaces.
  - Allows effects to change smoothly by considering regime probabilities rather than splitting the sample into discrete regimes, improving stability and precision relative to separate SVARs by cycle.

### Sequencing of reforms
- Approach to sequencing:
  - Use first-generation reform index (an average of governance, external sector and regulatory quality reform indices) as the conditioning variable z_{i,t}.
  - Purpose: examine whether impulse responses of other reforms (e.g., credit market reforms) vary by the level of first-generation structural development.
- Interpretations:
  - If impulse responses of growth from other reforms are larger in countries with higher first-generation structural development, this implies first-generation reform should be prioritized to lay a foundation and enhance gains from subsequent reforms.

### Presentation of results (figures and interpretation details)
- Impulse-response presentation:
  - Responses shown for a “major historical reform” defined as two standard deviations of the annual change in the structural index.
  - Time scale: x-axis measured in years where t = 0 is the first year the reform is implemented; k ranges up to 5 years.
  - Shaded areas denote 90 percent confidence bands.
- Heterogeneity exercises shown in figures:
  - Baseline average effects of reforms on Investment, Employment, and Labor Productivity for reform categories: Governance; External Sector; Credit Market; Regulatory Quality; Labor Market.
  - Effects under weaker governance versus baseline (Annex Figure 6.2):
    - Blue line: baseline (sample-average) response to a major historical reform; shaded 90 percent confidence bands.
    - Red line: response when governance is weaker; solid (dashed) segments denote statistically significant (insignificant) responses at the 90 percent level.
  - Effects under low growth versus baseline (Annex Figure 6.3):
    - Blue line: baseline response to a major historical reform; shaded 90 percent confidence bands.
    - Red line: response when growth is low; solid (dashed) segments denote statistically significant (insignificant) responses at the 90 percent level.
  - Effects under limited policy space versus baseline (Annex Figure 6.4):
    - Blue line: baseline response to a major historical reform; shaded 90 percent confidence bands.
    - Red line: response when policy space is limited; solid (dashed) segments denote statistically significant (insignificant) responses at the 90 percent level.
- Visualization notes:
  - Figures plot percent changes on the vertical axis; horizontal axis in years with t = 0 as reform implementation year.
  - The responses are computed for investment, employment, and labor productivity by reform type and by regime (weaker governance, low growth, limited policy space).  

*Source: Annex 2.6, “Estimations of the Impact of Structural Reforms on Outcome.”*

---


_Source: https://www.imf.org/-/media/files/publications/reo/mcd-cca/2023/october/english/ch2onlineannexes.pdf_