## cr1765

## Source details

**Canonical URL:** [cr1765](https://www.imf.org/-/media/files/publications/cr/2017/cr1765.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2017/cr1765.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2017/cr1765.pdf.json)

---

### Gender Inequality Index and GDP per Capita — Overview and key findings
- Morocco ranked 117th out of 155 countries on the United Nations gender inequality index; comparator MENAP countries ranked: Tunisia 48, Algeria 85, Jordan 102.
- Countries with higher gender inequality are, at the global level, poorer and grow more slowly.
- Closing overall gender gaps would help Morocco close its GDP per capita gap with benchmark countries by up to 1 percentage point.
- Simulations show that gradually closing gender gaps in the labor force participation rate could lead to significant income gains over the long term.
- Policy recommendations highlighted: investing in secondary education for women and in infrastructure, and reforming gender-discriminatory tax policies and laws.

### Gender gaps in Morocco — education, literacy, and labor
- Education enrollment:
  - Female-to-male enrollment at the primary level rose from around 70 percent in the mid-1990s to 95 percent today.
  - Gender gap for secondary and tertiary levels narrowed significantly, "up to 85 percent and 90 percent respectively."
  - Improvements have been driven by urban areas; rural areas lag behind.
- Adult literacy and rural disparities:
  - In rural areas, 60 percent of women are illiterate compared to 35 percent for men.
- Female labor force participation:
  - The rate of women’s participation in the labor force is 25 percent (noted as just at the average for the MENAP oil importers’ region).
  - Female labor force participation has been declining over the past decade, mainly driven by falling participation for women over 25 years of age.
  - Rural women participate in the labor market in greater numbers when the economy performs well but are the first to be excluded during downturns.
  - There are disparities between rural and urban areas, with the gender gap being wider in urban areas.
- Employment and unemployment:
  - The unemployment rate is only slightly higher for women than for men and similar for youth, but educated women have a much higher unemployment rate than educated men.
  - Labor market mismatches and regulations (e.g., high minimum wages and payroll taxes) are associated with higher unemployment rates and lower formality rates, especially among youth and women.
  - Low levels of education limit women’s opportunities and force many into low-quality jobs.

### Quantifying the impact of gender gaps on growth — empirical approaches and results
- Empirical approaches used:
  - Decomposition of differences in average real GDP per capita growth rates between Morocco and benchmark groups, attributing portions of the shortfall to gender gaps and other factors.
  - A general equilibrium occupational choice model (Cuberes, Newiak and Teigner (2016) framework) to quantify GDP losses due to misallocation of women in the labor force.
- Growth regression:
  - Sample: 103 countries from regions including MENA, LAC, SSA, and Asia, plus selected advanced economies for 1990–2014.
  - Method: robust two-step GMM to control for endogeneity.
  - Finding: gender inequality has a negative impact on growth for all countries; when including gender and income inequalities, the negative impact is pronounced for lower income countries only, including Morocco.
- Growth differential decomposition:
  - Reducing gender inequality and improving education to the levels of the Asian, Emerging Europe and Latin American benchmark countries could boost Morocco’s real GDP per capita growth rates relative to these countries by:
    - 1 percent,
    - 1.5 percent,
    - 0.75 percent.
  - Educational gaps and gender inequality are significant contributors to Morocco’s real GDP per capita shortfall versus benchmark groups.

### Occupational choice model — female labor force participation and frictions
- Model structure and calibrated frictions:
  - Agents receive a random entrepreneurship skill and choose to be employers, self-employed (own-account), or employees.
  - Female labor market frictions parameterized by μ, μ0 and λ:
    - Only a fraction μ of women can choose their occupation freely; a share 1-μ cannot become an employer.
    - Out of women excluded from being employers, only a fraction μ0 can choose to be self-employed.
    - Only a fraction λ can join the labor market in general; a share 1-λ is excluded from all occupations.
  - Parameters are calibrated to match observed ratios: female-to-male employers, female-to-male own-account workers, and women-to-men in the labor force.
- Purpose: frictions capture discrimination, differences in choices, and demand/supply constraints that lead to misallocation of women across occupations and consequent income losses.

### Drivers of female labor force participation (region and Morocco-specific)
- Regionally specific factors: history, religion, culture, and social norms affect low female labor force participation.
- Morocco-specific drivers:
  - Slow pace of structural transformation limiting creation of manufacturing jobs for women with secondary education.
  - Marriage and household composition influence participation: educated women are more likely to marry educated men who may support the household, lowering women’s participation incentives.
  - The probability of participation decreases with the number of children below six in urban areas.

### Policy implications emphasized
- Invest in secondary education for women.
- Invest in infrastructure that supports female labor force participation.
- Reform gender-discriminatory tax policies and laws to reduce barriers to women’s labor market participation and entrepreneurship.
- Address labor market regulations and mismatches (for example, minimum wages and payroll taxes) that disproportionately affect youth and women.

### Morocco’s current income losses and simulations (section 11)
- Current costs:
  - Costs associated with gender gaps in labor force participation and entrepreneurship are currently as high as 46 percent of income per capita compared to a situation where women have the same level of labor force and entrepreneurship participation as men.
- Simulation outcomes for eliminating gender gaps:
  - Overall income gains of about 27 percent in 2040 if gender gaps are closed in 50 years.
  - GDP gains decrease the longer it takes to eliminate gender gaps:
    - 13.6 percent if gender gaps were eliminated in 100 years,
    - 9.2 percent if eliminated in 150 years.
- Demographic context:
  - Morocco is experiencing a demographic transition with an increase in the working-age population, but population growth is slowing and the United Nations population division projects a rise in the dependency ratio by 2040.
- Parental leave provisions:
  - Maternity leave: 14 weeks at 100 percent of a woman’s wages, payable from a national social security fund, meeting ILO standards on duration of maternity leave.
  - Paternity leave: 3 days.

### Policy considerations and recommendations (from section 11)
- Legal framework:
  - Existing laws promoting gender equality include the (revised) Labor Code (2004), the 2011 Constitution, and a revised family code (2004).
  - Remaining legal inequalities: married women are not permitted to be the head of the household; inequality between sons and daughters and between female and male surviving spouses in inheritance rights.
  - Recommendation: provide for equality in inheritance rights to create opportunities for women to own housing or land and reduce gender gaps in labor force participation.
- Paternity leave:
  - Recommendation: increase paternity leave (currently 3 days) to contribute to gender equality at work and intra-household equality.
- Infrastructure:
  - Recommendation: invest in safe public transportation, improved road accessibility, and public childcare facilities to reduce women’s travel time and free women’s time to attend school and join the labor market.
- Gender budgeting:
  - Morocco introduced gender budgeting beginning in 2002; annual Gender Report covers more than 30 departments and ministries and is included as part of each year’s Finance Bill following an organic finance law approved in 2014.
  - Recommendation: enhance oversight, audit, and monitoring since there is currently no comprehensive system in place for monitoring or evaluation; increase the role of parliament or civil society organizations in gender budgeting work.
- Tax policy:
  - Remove gender-discriminatory tax practices; example: male taxpayers can claim a dependent deduction for both spouse and children, whereas female taxpayers may not claim the same deduction unless they prove legal guardianship.
- Education and employment:
  - Follow national employment strategy recommendations, including:
    - Use conditional transfers for education (Tayssir) to promote better access to secondary education for girls.
    - Implement literacy programs for women in rural areas and vocational training programs for all women.
    - Create more local jobs in activities that require more women.
    - Support female entrepreneurship.

### Empirical methodology (Annex I) — estimation details
- Growth equation controls: log of initial GDP per capita, fixed capital formation (% of GDP), total average years of schooling, an inflation dummy for average inflation of 15 percent and above, an institutional quality index, and an infrastructure index (first principal component of mobile phones and internet per 100 people, access to electricity and water, total air transportation of passengers per year).
- Inequality measures included for robustness: share of income held by top 20 percent relative to bottom 40 percent, net Gini coefficient, and the UN’s Gender Inequality Index (GII).
- Estimation technique: robust two step system GMM.
- Sample: 103 countries from MENA, LAC, SSA, Asia, and selected advanced economies over 1990 to 2014. Variables averaged over five-year periods except net Gini where the first value of the five years is used (or previous year if missing).
- Key empirical interpretations:
  - Inequalities have a negative impact on growth, particularly for countries in early stages of development.
  - Income inequality significantly and robustly impedes growth for low-income countries.
  - Gender inequality is negatively related to growth in the whole sample and significantly negatively impacts growth for low-income countries alone in some specifications.

### Financial development and credit — key statistics and trends
- Financial development:
  - Morocco’s Financial Development index is at the emerging market average.
  - Private sector credit to GDP: 63 percent.
  - Lower middle-income peers average private sector credit to GDP: 45 percent.
  - Egypt: 26.5 percent of GDP.
  - Philippines: 42 percent of GDP.
  - Upper middle-income peers average: 113.5 percent of GDP.
  - Turkey and Panama: 80 percent of GDP.
  - Malaysia: 125 percent of GDP.
- Historical sectoral credit shifts (2003–2008):
  - Real estate sector share of total private sector credit increased from 11 percent to 24 percent.
  - Credit to the financial sector increased from 13 percent to 20 percent.
- Current borrower composition (outstanding credit):
  - Private enterprises: 55 percent.
  - Households: 38 percent.
  - Public sector: 7 percent.
- Bank credit to the nonfinancial public sector: 5 percent of total credit (or 4 percent of GDP).
- Total financing to public enterprises: 23 percent of GDP (mostly foreign financing and some capital market access for a few large public enterprises).
- Sectoral distribution (mid-2016):
  - Households (individual loans and mortgages): 33 percent.
  - Productive sectors (particularly manufacturing): 35 percent.
  - Financial sector: 15 percent.
- Sectoral concentration index: 0.13 (BAM 2015).

### Credit concentration, SMEs, and household balance sheets
- Large exposures to single parties/groups of counterparties = 302 percent of the banking system’s Tier 1 capital in June 2016 (against 376 percent in 2009).
- Total credit to the top three largest corporates declined to 6.2 percent of total banking credit from 8.6 percent a year ago.
- World Bank Enterprise Survey (2013):
  - 52 percent of Moroccan firms had a line of credit (against 33 percent in 2007).
  - Large enterprises access: 72 percent.
  - SMEs access: 50 percent.
  - World average: 35 percent.
  - Middle East average: 26 percent.
- Share of credit to SMEs declined from 36 percent (three years prior to June 2016) to 33 percent of total non-financial sector loans in June 2016.
- Household credit and debt:
  - Household debt: 31 percent of GDP.
  - Mortgage loans increased from 11 percent of GDP (2006) to 18 percent of GDP (2016).
  - Consumer loans increased from 2 percent of GDP (2006) to 5 percent of GDP (2016).
  - Household debt is estimated at about 50 percent of household income.
  - By 2015, 25 percent of a sample of households had a debt to income ratio higher than 40 percent (BAM 2015).
  - Household NPLs averaged around 6.3 percent in mid-2016 and have been decreasing since 2014.
  - Household debt is denominated in local currency; house prices have been slightly decreasing since 2013.

### Bank profitability, NPLs, and liquidity
- Bank funding composition: 72 percent customer deposits, 57 percent of deposits are non-remunerated demand deposits.
- Since 2013, deposits increased and liquidity conditions improved while credit growth was low.
- Banks increased purchases of government bonds: 9.5 percent of their assets today (against 6.5 percent in 2012).
- NPLs increased to 7.9 percent in October 2016 from 4.8 percent in 2011.
- NPLs from the nonfinancial corporate sector around 11 percent and are the main drivers of the increase.
- Sectors with highest NPL levels: tourism, manufacturing, and trade.
- Payment delays and arrears:
  - SMEs face largest payment delays (almost 7 months).
  - Real estate development: about 18 months delay.
  - Public works and industries: more than three months delay.

### Determinants of cost of credit and access
- Lending rates in Morocco are lower than in comparator and emerging market countries and have been declining in recent years.
- Percentage of loans requiring collateral declined from 90 to 84 percent between 2007 and 2013 but remains higher than global and MENA averages.
- Average value of collateral required: 166 percent.
- For small enterprises, average collateral requirement: 224 percent.
- Enterprises consider access to finance one of the top obstacles for doing business in Morocco.

### Structural model of credit supply and demand — estimation results
- Sample: 2009M7–2016M7; Included observations: 85; Total system observations: 170.
- Estimation method: Generalized Method of Moments (GMM).
- Supply equation (CREDIT = C(1) + C(2)*SPREAD + C(3)*NPL + C(4)*BSLP):
  - Bank structural liquidity position coefficient: 0.0001405; Std. Error: 5.27E-05; t-Statistic: 2.664587; Prob.: 0.0085 (statistically significant).
  - Lending-deposit spread coefficient: 4.404970.
  - Non-performing loan growth coefficient: -0.046775 (not significant).
  - Supply equation R-squared: 0.761978; Adjusted R-squared: 0.753162.
- Demand equation (CREDIT = C(6) + C(7)*SPREAD + C(8)*GDPNA + C(9)*REI):
  - Lending-deposit spread coefficient: -20.146981; Std. Error: 10.20299; t-Statistic: -1.974616; Prob.: 0.0500 (borderline significant).
  - Real estate index coefficient: 0.000887; Std. Error: 0.0002164; t-Statistic: 4.097845; Prob.: 0.0001 (statistically significant).
  - Non-agricultural GDP growth coefficient: -0.288099 (not significant).
  - Demand equation R-squared: 0.565283; Adjusted R-squared: 0.549182.
- Interpretation:
  - Supply curve is shifted by banks’ liquidity position.
  - Demand curve is shifted by real estate prices.
  - Negative relationship between demand for credit and interest rates; positive relationship (not significant) with supply.
  - Over the long term, GDP and non-performing loans have a significant relationship with credit.

### Demand and supply curve dynamics, projections, and macro linkages
- Past dynamics:
  - Predicted demand and supply curves follow observed credit growth closely over the past eight years.
  - Both credit demand and supply diminished rapidly after the 2008 financial crisis.
  - Drivers of the decline:
    - Demand side: lower real estate prices in the post-financial crisis period.
    - Supply side: lower liquidity for banks, lower interest rates, and the increase in NPLs.
  - Since 2014: the supply curve has been relatively stable while predicted demand has surpassed observed credit growth.
- Projecting a consistent level of credit growth:
  - Medium-term expected average: 4.5–5 percent.
  - Low-end estimate: 2.5 percent.
  - High-end estimate: 7 percent.
  - Assumption: the share of credit in GDP remains constant.
- Impact of credit on GDP (VAR results):
  - A one percent increase in credit corresponds to a 0.2 percent increase in GDP levels in later periods.
  - A VAR including credit, inflation, import, exports, CBSI stock index, cement sales, tax revenue, oil prices, GDP, employment, unemployment rate, real estate prices, and industrial production shows inflation, the CBSI stock index, employment, and real estate price levels move upward in response to higher credit growth.

### Policy recommendations on credit, banking sector, and SME access
- Monetary policy:
  - Policy action by BAM in the form of a further reduction in the policy rate in 2016 has contributed to supporting a nascent recovery.
- Credit supply support:
  - Public guarantees for SME credit have helped boost credit supply to SMEs.
- Structural and regulatory actions:
  - Reducing payment delays is key to improving the financial situation of SMEs and their access to credit.
  - Reducing collateral constraints, particularly for SMEs, including by approving the draft law on collateral, would help improve credit access to SMEs.
  - Containing the increase in NPLs is essential; corporate debt restructuring, in particular, may help support credit allocation to sectors that have been lagging since the mid-2000s credit boom.
  - Increasing household indebtedness should continue to be monitored closely, particularly regarding consumption credit (for which NPLs are higher).

### External competitiveness and labor cost policy considerations (section 16)
- Productivity and wages:
  - Productivity growth should match or exceed nominal wage growth to maintain low relative ULCs.
  - Non-wage labor compensation, such as social security contributions, can contribute to high and sticky nominal wages.
- Exports and global value chains:
  - Efforts to integrate exports into global value chains should continue; degree of integration affects elasticity of exports to the REER.
- Structural reforms:
  - Improving education and training, or labor market reforms to facilitate reallocation of productive resources toward more productive, more competitive sectors.
- Exchange rate flexibility:
  - Greater exchange rate flexibility may help preserve competitiveness and support emerging manufacturing sectors, particularly where productivity gains have been slow.

*Source: IMF staff estimates and analysis as presented in the cited chapter.*

### 1. Gender Inequality Index and GDP per Capita  ________________________________________  3

### 1. Gender Inequality Index and GDP per Capita

### Overview and key findings
- Morocco ranked 117th out of 155 countries on the United Nations gender inequality index; comparator MENAP countries ranked: Tunisia 48, Algeria 85, Jordan 102.
- Countries with higher gender inequality are, at the global level, poorer and grow more slowly.
- Closing overall gender gaps would help Morocco close its GDP per capita gap with benchmark countries by up to 1 percentage point.
- Simulations show that gradually closing gender gaps in the labor force participation rate could lead to significant income gains over the long term.
- Policy recommendations highlighted: investing in secondary education for women and in infrastructure, and reforming gender-discriminatory tax policies and laws.

### A. Gender gaps in Morocco — education, literacy, and labor
- Education enrollment:
  - Female-to-male enrollment at the primary level rose from around 70 percent in the mid-1990s to 95 percent today.
  - Gender gap for secondary and tertiary levels narrowed significantly, "up to 85 percent and 90 percent respectively."
  - Improvements have been driven by urban areas; rural areas lag behind.
- Adult literacy and rural disparities:
  - In rural areas, 60 percent of women are illiterate compared to 35 percent for men.
- Female labor force participation:
  - The rate of women’s participation in the labor force is 25 percent (noted as just at the average for the MENAP oil importers’ region).
  - Female labor force participation has been declining over the past decade, mainly driven by falling participation for women over 25 years of age.
  - Rural women participate in the labor market in greater numbers when the economy performs well but are the first to be excluded during downturns.
  - There are disparities between rural and urban areas, with the gender gap being wider in urban areas.
- Employment and unemployment:
  - The unemployment rate is only slightly higher for women than for men and similar for youth, but educated women have a much higher unemployment rate than educated men.
  - Labor market mismatches and regulations (e.g., high minimum wages and payroll taxes) are associated with higher unemployment rates and lower formality rates, especially among youth and women.
  - Low levels of education limit women’s opportunities and force many into low-quality jobs.

### B. Quantifying the impact of gender gaps on growth
- Two empirical approaches used:
  - Decomposition of differences in average real GDP per capita growth rates between Morocco and benchmark groups, attributing portions of the shortfall to gender gaps and other factors.
  - A general equilibrium occupational choice model (Cuberes, Newiak and Teigner (2016) framework) to quantify GDP losses due to misallocation of women in the labor force.
- Growth regression:
  - Sample: 103 countries from regions including MENA, LAC, SSA, and Asia, plus selected advanced economies for 1990–2014.
  - Method: robust two-step GMM to control for endogeneity.
  - Finding: gender inequality has a negative impact on growth for all countries; when including gender and income inequalities, the negative impact is pronounced for lower income countries only, including Morocco.
- Growth differential decomposition:
  - Reducing gender inequality and improving education to the levels of the Asian, Emerging Europe and Latin American benchmark countries could boost Morocco’s real GDP per capita growth rates relative to these countries by:
    - 1 percent (relative to one benchmark group),
    - 1.5 percent (relative to another benchmark group),
    - 0.75 percent (relative to the third benchmark group).
  - Educational gaps and gender inequality are significant contributors to Morocco’s real GDP per capita shortfall versus benchmark groups.

### C. Female labor force participation and growth — occupational choice model
- Model setup and frictions:
  - Uses a general equilibrium occupational choice model in which agents receive a random entrepreneurship skill and choose to be employers, self-employed (own-account), or employees.
  - Female labor market frictions are modeled as follows:
    - Only a fraction μ of women can choose their occupation freely; a share 1-μ cannot become an employer.
    - Out of women excluded from being employers, only a fraction μ0 can choose to be self-employed.
    - Only a fraction λ can join the labor market in general; a share 1-λ is excluded from all occupations.
  - Parameters μ, μ0 and λ are calibrated to match observed ratios: female-to-male employers, female-to-male own-account workers, and women-to-men in the labor force.
- Purpose and interpretation:
  - These frictions capture barriers such as discrimination, differences in choices, and demand/supply constraints that lead to misallocation of women across occupations and consequent income losses.

### D. Drivers of female labor force participation (Box 1)
- Regionally specific factors affecting low female labor force participation include history, religion, culture, and social norms.
- Morocco-specific drivers:
  - Slow pace of structural transformation limiting creation of manufacturing jobs for women with secondary education.
  - Marriage and household composition influence participation: educated women are more likely to marry educated men who may support the household, lowering women’s participation incentives.
  - The probability of participation decreases with the number of children below six in urban areas.

### Policy implications emphasized in the chapter
- Invest in secondary education for women.
- Invest in infrastructure that supports female labor force participation.
- Reform gender-discriminatory tax policies and laws to reduce barriers to women’s labor market participation and entrepreneurship.
- Address labor market regulations and mismatches (for example, minimum wages and payroll taxes) that disproportionately affect youth and women.

*Prepared by Vincent Dadam, Lisa Kolovich, and Anta Ndoye; chapter excerpt from “Morocco: Implications of Gender Inequality for Growth.”*

### 11.      The results show that Morocco is currently losing out on a significant share of income

### 11.      The results show that Morocco is currently losing out on a significant share of income

### Key quantitative findings
- Costs associated with gender gaps in labor force participation and entrepreneurship are currently as high as 46 percent of income per capita compared to a situation where women have the same level of labor force and entrepreneurship participation as men.
- Simulations indicate policies to eliminate gender gaps could lead to overall income gains of about 27 percent in 2040 if gender gaps are closed in 50 years.
  - GDP gains decrease the longer it takes to eliminate gender gaps: 13.6 percent if gender gaps were eliminated in 100 years, and 9.2 percent if eliminated in 150 years.
- Demographic context:
  - Morocco is experiencing a demographic transition with an increase in the working-age population, but population growth is slowing and the United Nations population division projects a rise in the dependency ratio by 2040.
- Maternity leave:
  - Morocco offers 14 weeks of maternity leave, at 100 percent of a woman’s wages, payable from a national social security fund, meeting ILO standards on duration of maternity leave.
- Paternity leave:
  - Paternity leave in Morocco is currently 3 days.

### Policy considerations and recommendations
- Continue and expand existing measures that reduce gender inequality, and consider complementary policies:
  - Legal framework
    - Morocco has laws promoting gender equality, including the (revised) Labor Code (2004), the 2011 Constitution, and a revised family code (2004) expanding rights in areas such as guardianship, marriage, child custody, and access to divorce.
    - Remaining legal inequalities include: married women are not permitted to be the head of the household; inequality between sons and daughters and between female and male surviving spouses in inheritance rights.
    - Recommendation: provide for equality in inheritance rights to create opportunities for women to own housing or land and reduce gender gaps in labor force participation.
  - Paternity leave
    - Recommendation: increase paternity leave (currently 3 days) to contribute to gender equality at work and intra-household equality.
  - Infrastructure
    - Recommendation: invest in safe public transportation, improved road accessibility, and public childcare facilities to reduce women’s travel time and free women’s time to attend school and join the labor market.
  - Gender budgeting
    - Morocco is notable for an advanced gender budgeting initiative; recommendation: enhance oversight, audit, and monitoring since there is currently no comprehensive system in place for monitoring or evaluation.
  - Tax policy
    - Remove gender-discriminatory tax practices; example: male taxpayers can claim a dependent deduction for both spouse and children, whereas female taxpayers may not claim the same deduction unless they prove legal guardianship.
  - Education and employment
    - Follow national employment strategy recommendations, including:
      - Use conditional transfers for education (Tayssir) to promote better access to secondary education for girls.
      - Implement literacy programs for women in rural areas and vocational training programs for all women.
      - Create more local jobs in activities that require more women.
      - Support female entrepreneurship.

### Morocco’s gender budgeting efforts (Box summary)
- Morocco introduced gender budgeting beginning in 2002, focusing on Millennium Development Goals, increasing women’s public employment, and collecting gender-disaggregated data.
- Annual Gender Report published by the Ministry of Economy and Finance covers more than 30 departments and ministries and is included as part of each year’s Finance Bill following an organic finance law approved in 2014.
- The organic finance law requires gender equality be considered when defining performance objectives, results, and indicators in all line budgets, and mandates inclusion of the Gender Report in the Finance Bill.
- The Government Plan for Equality calls for strengthening the law by increasing transparency, improving fiscal performance, and generalizing evaluation, audit, and accountability procedures.
- Recommendation: increase the role of parliament or civil society organizations in Morocco’s gender budgeting work.

### Empirical methodology (Annex I)
- Growth equation estimated with GDP per capita growth as dependent variable and controls including: log of initial GDP per capita, fixed capital formation (% of GDP), total average years of schooling, an inflation dummy for average inflation of 15 percent and above, an institutional quality index, and an infrastructure index (first principal component of mobile phones and internet per 100 people, access to electricity and water, total air transportation of passengers per year).
- Inequality measures included for robustness: share of income held by top 20 percent relative to bottom 40 percent, net Gini coefficient, and the UN’s Gender Inequality Index (GII).
- Estimation technique: robust two step system GMM method to address endogeneity concerns.
- Sample: 103 countries from MENA, LAC, SSA, Asia, and selected advanced economies over the period 1990 to 2014. Variables averaged over five-year periods except net Gini where the first value of the five years is used (or previous year if missing).

### Empirical results and interpretation
- Main empirical conclusion: inequalities have a negative impact on growth, particularly important for countries in early stages of development.
- Income inequality:
  - Income inequality significantly and robustly impedes growth for low-income countries for both the Gini coefficient and the top 20/bottom 40 income share ratio.
  - In specifications accounting for both gender and income inequality, the negative impact of income inequality remains.
- Gender inequality:
  - Gender inequality is negatively related to growth in the whole sample (model specification (3)).
  - In model specification (4), gender inequality significantly and negatively impacts growth for low-income countries alone.
  - Interpretation: gender inequality tends to be higher in early development stages and decreases as the economy grows.
- Selected estimation details preserved from tables:
  - Sample period: 1990 to 2014.
  - Number of countries in the main sample: 103.
  - Example reported statistics in Table 1 (for illustrative coefficients and diagnostics):
    - Initial income per capita (log) coefficients: -1.595***, -1.606***, -1.801*** in specifications (1)–(3).
    - Infrastructure index coefficients: 0.555**, 0.435**, 0.103***.
    - Institutional quality coefficients: 0.129**, 0.137***, 0.279*.
    - Diagnostics: Arellano-Bond AR(1) and AR(2) p values reported for various specifications (e.g., AR(1) p value 0.006, AR(2) p value 0.027 in one specification).
    - Observations and country counts reported (e.g., Observations 380, 450, 369; Number of countries 97 in some specifications).
  - Table 2 reports further specifications including interactions with LICs, and control variable coefficients such as fixed capital investment (% GDP) and schooling (years), with statistical significance markers (***) p>0.001, (**) p>0.05, (*) p>0.1.

*Source: IMF staff estimates and analysis as presented in the cited chapter.*

### 3.      Morocco’s level of financial development is close to the emerging market average. In

### 3.      Morocco’s level of financial development is close to the emerging market average. In

### Financial development and credit levels
- Morocco’s Financial Development index is at the emerging market average but there is room for increasing access and developing both institutions and markets (Sahay and others, 2015).
- Private sector credit to GDP: 63 percent.
- Lower middle-income peers average private sector credit to GDP: 45 percent.
- Egypt: 26.5 percent of GDP.
- Philippines: 42 percent of GDP.
- Upper middle-income peers average: 113.5 percent of GDP.
- Turkey and Panama: 80 percent of GDP.
- Malaysia: 125 percent of GDP.

### Historical credit developments and sectoral allocation
- Real estate sector share of total private sector credit increased from 11 percent to 24 percent between 2003 and 2008.
- Credit to the financial sector increased from 13 percent to 20 percent over the same period.
- By borrower type (current outstanding credit):
  - Private enterprises: 55 percent.
  - Households: 38 percent.
  - Public sector: 7 percent.
- Bank credit to the nonfinancial public sector: 5 percent of total credit (or 4 percent of GDP).
- Total financing to public enterprises: 23 percent of GDP (mostly foreign financing and some capital market access for a few large public enterprises).
- Sectoral distribution (mid-2016):
  - Households (individual loans and mortgages): 33 percent.
  - Productive sectors (particularly manufacturing): 35 percent.
  - Financial sector: 15 percent.
- Sectoral concentration index: 0.13 (BAM 2015).

### Credit concentration and firm-level access
- Large exposures to single parties/groups of counterparties = 302 percent of the banking system’s Tier 1 capital in June 2016 (against 376 percent in 2009).
- Total credit to the top three largest corporates declined to 6.2 percent of total banking credit from 8.6 percent a year ago.
- World Bank Enterprise Survey (2013):
  - 52 percent of Moroccan firms had a line of credit (against 33 percent in 2007).
  - Large enterprises access: 72 percent.
  - SMEs access: 50 percent.
  - World average: 35 percent.
  - Middle East average: 26 percent.

### Trends in credit to SMEs
- Share of credit to SMEs declined from 36 percent (three years prior to June 2016) to 33 percent of total non-financial sector loans in June 2016.

### Credit and business cycles
- Credit fluctuations have been much larger than GDP fluctuations.
- During the 2007 peak, growth of credit to GDP surpassed the 10 percent threshold identified in Dell’Ariccia and others (2012).
- Since 2012, actual credit-to-GDP growth has been below the more recent trend (a negative credit gap).

### Corporate deleveraging and sector dynamics
- Private corporate liabilities to domestic banks rose from 25 percent of GDP to 51 percent of GDP between 2004 and 2011; corporate bank credit declined thereafter and is currently at 42 percent of GDP absorbing 44 percent of total private credit.
- Credit to the financial sector stable at about 10 percent of GDP.
- Between 2012 and September 2016:
  - Mortgage lending increased by 16 percent in nominal terms.
  - Lending to developers decreased by 5 percent.
- BAM survey (2010–2014): debt ratio of large enterprises increased from 38 to 41 percent; SMEs decreased from 20 to 13 percent.
- Listed firms sample (many from real estate): indicators point to higher indebtedness and lower profitability.

### Household credit and financial stress
- Household debt: 31 percent of GDP.
- Mortgage loans increased from 11 percent of GDP (2006) to 18 percent of GDP (2016).
- Consumer loans increased from 2 percent of GDP (2006) to 5 percent of GDP (2016).
- Household debt is estimated at about 50 percent of household income.
- By 2015, 25 percent of a sample of households had a debt to income ratio higher than 40 percent (BAM 2015).
- Household NPLs averaged around 6.3 percent in mid-2016 and have been decreasing since 2014.
- Household debt is denominated in local currency; house prices have been slightly decreasing since 2013.

### Bank profitability, risk, and liquidity
- Bank profitability declined up until 2015 due to increasing NPLs, higher provisions, increasing cost of risk, subdued credit growth, and declining net interest margins; a recovery in credit growth and profitability observed in 2016.
- Moroccan banks’ funding composition: 72 percent customer deposits, 57 percent of deposits are non-remunerated demand deposits.
- Since 2013, deposits increased and liquidity conditions improved while credit growth was low.
- Banks increased purchases of government bonds: 9.5 percent of their assets today (against 6.5 percent in 2012).

### Non-performing loans and sectoral impacts
- NPLs increased to 7.9 percent in October 2016 from 4.8 percent in 2011.
- NPLs from the nonfinancial corporate sector around 11 percent and are the main drivers of the increase.
- Sectors with highest NPL levels: tourism, manufacturing, and trade.
- Payment delays and arrears:
  - SMEs face largest payment delays (almost 7 months).
  - Real estate development: about 18 months delay.
  - Public works and industries: more than three months delay.
- Distribution of credit by sector (mid-2016) NPL shares (selected): real estate 15.7 percent, industry 6.4 percent, finance 15.5 percent, individual loans 20.9 percent (sectoral shares and NPL shares shown in Figure 5).

### Determinants of cost of credit and access
- Lending rates in Morocco are lower than in comparator and emerging market countries and have been declining in recent years.
- Percentage of loans requiring collateral declined from 90 to 84 percent between 2007 and 2013 but remains higher than global and MENA averages.
- Average value of collateral required: 166 percent.
- For small enterprises, average collateral requirement: 224 percent (higher than global and MENA averages).
- Enterprises consider access to finance one of the top obstacles for doing business in Morocco.

### Structural model of credit supply and demand (estimation results)
- Sample: 2009M7–2016M7; Included observations: 85; Total system observations: 170.
- Estimation method: Generalized Method of Moments (GMM).
- Supply equation (CREDIT = C(1) + C(2)*SPREAD + C(3)*NPL + C(4)*BSLP)
  - Bank structural liquidity position coefficient: 0.0001405; Std. Error: 5.27E-05; t-Statistic: 2.664587; Prob.: 0.0085 (statistically significant).
  - Other supply variables: Lending-deposit spread coefficient: 4.404970; Non-performing loan growth coefficient: -0.046775 (not significant).
  - Instruments: Interbank rate, policy rate, central bank interventions, treasury note issuances (volume).
  - Supply equation R-squared: 0.761978; Adjusted R-squared: 0.753162.
- Demand equation (CREDIT = C(6) + C(7)*SPREAD + C(8)*GDPNA + C(9)*REI)
  - Lending-deposit spread coefficient: -20.146981; Std. Error: 10.20299; t-Statistic: -1.974616; Prob.: 0.0500 (borderline significant).
  - Real estate index coefficient: 0.000887; Std. Error: 0.0002164; t-Statistic: 4.097845; Prob.: 0.0001 (statistically significant).
  - Non-agricultural GDP growth coefficient: -0.288099 (not significant).
  - Instruments: Exports, oil price, tourist arrivals, bond yields.
  - Demand equation R-squared: 0.565283; Adjusted R-squared: 0.549182.
- Interpretation:
  - Supply curve is shifted by banks’ liquidity position.
  - Demand curve is shifted by real estate prices.
  - Negative relationship between demand for credit and interest rates; positive relationship (not significant) with supply.
  - Low significance for some variables attributed to the short observation period and correlation among explanatory variables.
  - Over the long term, GDP and non-performing loans have a significant relationship with credit.

*Source: IMF staff estimates.*

### 19.      The plots of the estimated demand and supply curves as a function of the lending rate

### 19.      The plots of the estimated demand and supply curves as a function of the lending rate

### Demand and supply curve dynamics (summary of Figure 7 and accompanying text)
- The predicted curves follow observed credit growth closely over the past eight years.
- Both credit demand and supply diminished rapidly after the 2008 financial crisis, coinciding with credit growth entering its current trend.
- Drivers of the decline:
  - Demand side: lower real estate prices in the post-financial crisis period.
  - Supply side: lower liquidity for banks, lower interest rates, and the increase in NPLs.
- Since 2014:
  - The supply curve has been relatively stable.
  - Predicted demand has surpassed observed credit growth.

### D. Projecting a Consistent Level of Credit Growth
- Medium-term credit growth projections based on macroeconomic fundamentals:
  - Expected average: 4.5–5 percent over the medium term.
  - Low-end estimate: 2.5 percent over the medium term.
  - High-end estimate: 7 percent over the medium term.
  - Assumption: the share of credit in GDP remains constant.
  - Rationale: consistent with expected GDP and inflation growth, envisaged investment and GDP growth paths, pre-credit boom trends (prior to 2006), and cross-country financial deepening patterns.

### E. Impact of credit on GDP (Figure 8 and VAR results)
- A vector autoregression including GDP, credit, and other high-frequency indicators shows:
  - Inflation, the CBSI stock index, employment, and real estate price levels move upward in response to higher credit growth.
- Quantitative effect:
  - A one percent increase in credit corresponds to a 0.2 percent increase in GDP levels in later periods.
- Note on the VAR system:
  - Full system includes credit, inflation, import, exports, CBSI stock index, cement sales, tax revenue, oil prices, GDP, employment levels, unemployment rate, real estate prices, and industrial production.
  - Y-axis in the impulse response functions shows the percent change in levels; x-axis shows lags.
  - Data are quarterly and monthly in frequency.

### E. Concluding Remarks and Policy Recommendations (summarized)
Findings on functionality and cyclical drivers:
- No major issues in the functioning of credit markets in Morocco.
- Credit growth aligns with economic cycles; larger-than-GDP fluctuations in credit reflect demand and supply factors related to increasing risks.
- Specific contributors since 2013:
  - Low credit growth largely resulted from the combination of the business cycle (international financial crisis and related macroeconomic uncertainties) and the credit cycle (post-credit boom in the real estate sector).
  - Demand-side deleveraging by corporations, particularly in the real estate sector and in SMEs most affected by payment delays.
  - Supply-side effects from higher NPLs and related provisioning requirements impacting bank profitability and credit allocation to affected sectors.

Policy recommendations and supportive actions:
- Monetary policy:
  - Policy action by BAM in the form of a further reduction in the policy rate in 2016 has contributed to supporting a nascent recovery, indicating effective monetary transmission mechanisms in Morocco.
- Credit supply support:
  - Public guarantees for SME credit have helped boost credit supply to SMEs.
- Structural and regulatory actions to improve credit access and financial health:
  - Reducing payment delays is key to improving the financial situation of SMEs and their access to credit.
  - Reducing collateral constraints, particularly for SMEs, including by approving the draft law on collateral, would help improve credit access to SMEs.
  - Containing the increase in NPLs is essential; corporate debt restructuring, in particular, may help support credit allocation to sectors that have been lagging since the mid-2000s credit boom.
  - Increasing household indebtedness should continue to be monitored closely, particularly regarding consumption credit (for which NPLs are higher).

*Italic: IMF staff analysis as presented in the source document.*

### 16.      This analysis holds several policy considerations for preserving Morocco's external

### 16.      This analysis holds several policy considerations for preserving Morocco's external

### Key policy considerations
- Productivity growth should match or exceed nominal wage growth to maintain low relative ULCs.
  - Low ULCs have thus far provided Morocco an advantage in external competitiveness relative to its peers, but this gap is narrowing.
  - Slow productivity growth would pose a constraint as nominal wages rise (for example, for textiles).
  - To maintain relative external competitiveness, productivity growth should match or exceed growth in nominal wages.
  - Non-wage labor compensation, an important component of the latter, such as social security contributions, can contribute to high and sticky nominal wages.

- Efforts to integrate exports into global value chains should continue.
  - The empirical analysis of export performance suggests that manufacturing industries remain sensitive to the REER despite recent shifts up global value chains.
  - It is the degree of integration within these chains, rather than the additional value add per se, which would work to weaken the elasticity of exports to the real exchange rate, a shift which is likely to take place in the future as these sectors develop further.

- Supportive structural reforms will be important.
  - Where external competitiveness allows returns to accrue to sectors with low relative unit labor costs, structural reforms can facilitate reallocation of productive resources.
  - Examples include improving education and training, or labor market reforms.
  - Such reallocation would move resources towards more productive, more competitive sectors, increasing welfare gains for the economy.

- Greater exchange rate flexibility may also contribute.
  - The findings highlight potential benefits for Morocco from greater exchange rate flexibility, including to preserve competitiveness and support emerging manufacturing sectors.
  - This is particularly relevant where productivity gains have been slow (such as for textiles), and in the case of nominal shocks.

### Supporting references (selected from the content)
- Ahmed, S, M. Appendino, and M. Ruta, 2015, “Global Value Chains and the Exchange Rate Elasticity of Exports,” IMF Working Paper 15/252.
- Amiti, M., O. Itskhoki, and J. Konings, 2014, “Why Hasn’t the Yen Depreciation Spurred Japanese Exports?” Liberty Street Economics.
- Ministry of Economy and Finance, DEPF, 2016, Décomposition de la compétitivité structurelle du Maroc: Marges intensives et extensives de nos exportations (Morocco).
- European Central Bank, 2012, “Revisiting the Effective Exchange Rates of the Euro,” Occasional Paper Series No. 134.
- International Monetary Fund, 1997, IMF Working Paper WP/97/71; 2012, IMF Working Paper WP12/107.
- Veeramani, C., 2008, “Impact of Exchange Rate Appreciation on India’s Exports,” Economic and Political Weekly, Vol. 43, No. 22 (May 31–June 6, 2008), pp. 10–14.

*cr1765 - 16.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2017/cr1765.pdf_