Selected Issues

Abstract

Selected Issues

Poverty, Inequality and Social Safety Nets in Pakistan1

This paper provides an overview of social safety nets (SSNs) in Pakistan and uses a frontier analysis approach to assesses their efficiency in reducing poverty and inequality. SSNs in Pakistan were significantly strengthened over time but remain small against regional and emerging markets’ averages. The analysis suggests that stepping up public expenditure in SSNs is needed to alleviate still high poverty and inequality. To this end, finalizing the update of BISP beneficiaries’ database, broadening its coverage, and stepping up educational transfers is key. In parallel, continuing the energy subsidies reform would create fiscal space to strengthen SSNs and priority social spending.

A. Poverty and Inequality in Pakistan

1. Pakistan has made significant improvements in reducing poverty over the last two decades. Over the period 2000–15, per-capita income nearly tripled, mostly reflecting sustained growth, and the poverty headcount more than halved from 64 percent in 2001 to 29.5 percent in 2013 based on the new poverty line (from 34 percent to about 9 percent based on the 2001 poverty line).2 Despite these improvements poverty incidence remains high,3 reaching about 36 percent of the population in rural areas (against 18 percent in urban areas).4 Poverty incidence also significantly varies across the different provinces, peaking at more than 50 percent in the Balochistan region.5 The multi-dimensional poverty index—measuring achievements in key dimensions of human development—points to a higher incidence of poverty at 38.8 percent of the population (55 percent multi-dimensional poverty incidence in rural areas and 70 percent in the Balochistan region).6

uA01fig01

Pakistan: Poverty has Declined but Remains High

(Poverty headcount in percent of total population)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: WDI (2016); Pakistan Authorities; IMF staff estimates.

2. Access to basic services has significantly improved above South-Asia average and further efforts are underway. The access of the population to basic services has significantly strengthened during 2000–15, and is now well above South Asia average. Access to electricity and improved water sources is close to universal while access to improved water sanitation facilities improved to about 60 percent of the population. Furthermore, local communities’ projects are being supported to improve access to basic services in rural areas, under the Sustainable Development Goals (SDGs) program launched in FY 2014/15. These projects aim to strengthen access to improved sanitation, water resources and electricity to achieve the SDGs.

uA01fig02

Access to Electricity is almost universal in Pakistan

(In percent of population)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: World Development Index (2016)
uA01fig03

Access to improved sanitation facilities in Pakistan is above regional average

(access to improved sanitation facilities; in percent of population)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: World Development Index (2016)

3. Children’s stunting and malnutrition remains a challenge. The stunting rate among children under five was at 45 percent in 2012, above South Asia average—and the wasting rate among children under-five was at 10½ percent.7 The incidence of stunting and malnutrition among children varies across different provinces, with the highest prevalence of stunting in the provinces of Balochistan and Sindh where it was above 50 percent (Figure 1).

Figure 1.
Figure 1.

Pakistan: Poverty is High and Children’s Malnutrition is Widespread

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

4. Despite improvements, education and health outcomes remain below the regional average, with public spending in these areas comparatively lower than in other emerging markets. Education and health outcomes have improved during 2000–15, with the primary gross enrollment ratio and youth literacy rate increasing to 94 percent and 73 percent, respectively, and infant mortality declining. Public spending in education and health increased to 2.5 and 0.9 percent of GDP, respectively, but remains well below emerging markets’ average, also reflecting capacity constraints at the provincial level in the implementation of development spending (Figure 2). However, education and health outcomes remain below the South-Asia average and Pakistan ranks low on the Human Development Index (147 out of 188 countries in 2015, unchanged from 2009).

Figure 2.
Figure 2.

Pakistan: Challenging Education and Health Outcomes amid Limited Public Spending

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Source: World Development Index (2016).

5. While income inequality is moderate, sizable gaps in education and health outcomes between the poorest and the richest in the population remain. The Gini coefficient slightly declined over the last two decades and stands at about 30 percent (2010), broadly in line with South Asia average. However, nonmonetary indicators highlight significant inequality along different dimensions. Health outcomes remain unequal among the population, with the rates of under-five mortality and children’s stunting in the poorest quintile almost threefold than in the richest one. Furthermore, the gap in education attainment is high, with average years of schooling among young adults (for 20–29 years old) in the richest quintile being more than double than in the poorest quintile and the gross enrollment ratio (for 13–14 years old) being almost four times higher in the richest quintile (Figure 3).8

Figure 3.
Figure 3.

Pakistan: High Inequality Across Many Dimensions

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

6. Despite some progress, gender inequality continues to be pervasive. Pakistan ranks very low in the Gender Gap Index (143 out of 144 countries in 2016), mostly reflecting poor economic participation and opportunities for women.9 Women’s labor force participation remains very low at 25 percent (against 83 percent for men) while the rate of women’s unemployment is almost twice as high as for men, at 9 percent (against 5 percent for men). About 35 percent of young women are illiterate (against 20 percent of young man) and girls are lagging behind boys in primary and secondary education enrollment, with the ratio differential in primary education at 15 percent. Several recent initiatives support women’s empowerment, including Gender Responsive Budgeting, a 10 percent women quota in public sector positions; and the Benazir Income Support Program provides cash transfers to women in poor households. Furthermore, women’s participation in federal and provincial assemblies increased to 17 percent of seats while in local bodies to one-third.

7. Furthermore, the large size of the informal economy constrains inclusiveness by limiting the creation of high quality and durable jobs and upward mobility (Box 1). The size of the informal economy is estimated to be large, in the range of 30–50 percent of the formal economy, above South Asia average. The presence of a large informal sector constrains economic growth and limits the creation of high quality and durable jobs, as workers employed in the informal sectors have no social protection coverage and limited career and upward mobility opportunities.10

uA01fig04

Pakistan: A large Informal Economy

(Informal economy; as a share of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Friedrich Schneider, “The shadow economy and work in the shadow: what do we (not) know?,” (2012).

B. Social Safety Nets, Targeting and Outcomes in Pakistan

8. Social safety nets have a critical role to play to support and protect the most vulnerable in Pakistan.11 Pakistan has made progress in strengthening social protection programs to reduce poverty. Within this context, several social safety nets programs are in place, both at the federal and provincial level, which differ in terms of size, efficiency and targeted population. At the federal level, the Benazir Income Support Program is the main social safety net, complemented by smaller programs as the Zakat program, the Bait-Ul-Mal program, non-contributory Social Security and Social Welfare programs, and the Workers Welfare Fund.12 Recently, the new Prime Minister’s Health Card Program has been launched to provide free health care services to the poor. In addition to these social safety net programs, electricity and food subsidies and subsidies to foster the development of the agricultural sector are in place. Overall, expenditures on social safety nets, excluding subsidies and provincial programs, represent 0.54 percent of GDP, well below South Asia average and emerging markets’ average.

uA01fig05

Pakistan: Social Safety Nets are Relatively Small

(Expenditures on social safety nets; in percent of GDP, latest available data)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan Authorities; IMF staff calculations; World Bank ASPIRE Database (2016)

A Large Informal Economy Constrains Inclusiveness in Pakistan

The size of the informal economy is estimated to be large, in the range of 30–50 percent of the formal economy. Schneider (2012), using a cross-countries MIMIC model, estimates the size of the informal economy at about 40 percent of the official GDP in Pakistan, above South Asia average, with about 60 percent of the employed population is the informal economy.1 Furthermore, the Heritage Foundation Index (2005) points to the informal economy to be large in Pakistan.2 A significant regulatory burden may contribute to driving firms in the informal economy. Despite having slightly improved, Pakistan’s ranking in the World Bank Doing Business remains low (144 out of 190 countries).

The very low affiliation to pensions schemes and high self-employment also point to the presence of a sizable informal economy. Labor market indicators of informality, as the lack of pension coverage and self-employment highlight the large size of the informal economy. In particular, affiliation rate to pension schemes is very low at 5 percent, well below comparator countries, and self-employment is high at above 60 percent.

uA01fig06

Self-employment is Significant in Pakistan

(In percent of total employment)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: WDI (2016).
uA01fig07

Affiliation to Pension Schemes is Very Low in Pakistan

(In a percent of labour force)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: ILOSTAT.

Large informality may undermine inclusiveness in Pakistan. The presence of a large informal sector may help reduce poverty but is likely to constrain economic growth owing to firms remaining small-scale and having low productivity, and limit job creation. Furthermore, workers employed in the informal sector have no social protection coverage and limited career and upward mobility opportunities. Strengthening the business climate could contribute to reduce informality and raise inclusiveness.

1Kemal and Qasim (2012) estimate the informal economy in Pakistan to be at about 90 percent of the formal economy using a consumption-based household survey and calculating the discrepancy to official GDP.2 The Heritage Foundation index is based on subjective perception of general compliance with the law, focusing on the role played by corruption.

9. The efficiency of social safety nets has been strengthened over time towards the implementation of best practices.13 The landscape of social safety nets programs has evolved through time and was significantly strengthened by the launch of Benazir Income Support Program in 2008, which provides targeted cash transfers to the poor. The efficiency of the Benazir Income Support Program was progressively strengthened over time towards best practices. Notably, the program relies on effective and transparent targeting based on proxy means testing. Furthermore, the coverage and size of transfers under the program were gradually strengthened, conditional cash transfers aiming at strengthening human capital accumulation were introduced, and program delivery is effected through modern mechanisms. Efforts to improve the targeting of smaller social safety nets programs, such as the Zakat program, are needed and consolidating them into the Benazir Income Support Program would improve the overall efficiency of social safety nets in Pakistan.

The Benazir Income Support Program

10. The Benazir Income Support Program (BISP) is the main social safety net program, providing targeted cash transfers to the poorest families. The program initially aimed to protect the poor from the negative effects of rising food and fuel prices through the provision of cash transfers, with the broader medium-term objective to provide a minimum income to the poorest and reduce poverty. The key element of BISP is the provision of unconditional cash transfers (UCTs) to poor families, which is paid on a quarterly basis directly to the woman head of the household in order to support women empowerment.14 Furthermore, education conditional cash transfer (CCTs) started to be rolled-out in 2012 to BISP families with a child (within 5–12 years) to support enrollment, attendance and completion of primary school.15 Disbursements under BISP have more than tripled since the launch of the program while, in parallel, untargeted energy subsidies were reduced.

uA01fig08

BISP is the Largest Social Safety Net in Pakistan

(In percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan authorities; and IMF staff calculations.

11. BISP’s targeting mechanism has been significantly strengthened over time and relies now on proxy means testing. At the onset of the program, the identification of BISP beneficiaries was entrusted to the members of the National Assembly since no reliable database of the poor was available.16 A new and more transparent targeting mechanism based on a proxy means test was introduced in 2009, with the support of the World Bank, providing an objective method of approximating households’ welfare and poverty status. A poverty scorecard was applied in a door-to-door country-wide survey covering 27 million households (almost the full population of Pakistan) and, based on a cut-off to the resulting poverty score, 7.7 million poor households (about 29 percent of overall surveyed households) were identified as eligible for BISP cash transfers.17 On this basis, the National Socio-Economic Registry, Pakistan’s most comprehensive and reliable database of the poor, was also established. Overall, BISP is perceived to be targeted relatively fairly and protect the poorest households.

12. The coverage and the size of cash transfers to the poor have increased over time. Since 2013, the scope and the coverage of BISP were scaled-up. The budgetary allocation for the program increased from PRs 50 billion in FY2013/14 to PRs 102 billion in in FY2016/17, with PRs 115 billion budgetary allocation for FY2017/18. During 2013–17, the size of UCTs was increased by more than 50 percent, with stipends increasing from PRs 3,000 to PRs 4,834 (about US$46) per beneficiary per quarter.18 In parallel, while still not reaching the full set of intended beneficiaries, the coverage of the program was significantly broadened to reach to about 5.4 million beneficiary households (17 percent of the national population) in FY 2016/17 (from 3.8 million households in FY 2012/13). Similarly, the coverage of CCTs (PRs 750 per child/per quarter) was gradually expanded from 5 to 32 districts to cover about 1.3 million children in FY 2016/17 (from 52,000 children in 2012).

uA01fig09

Disbursments under BISP have streghtened over time

(BISP disbursments; in percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan authorities; and IMF staff calculations.

13. BISP operations were also significantly strengthened thus improving service to beneficiaries. Despite some remaining delays, the frequency and predictability of payments to beneficiaries has improved substantially, both in terms of number of payments and amounts disbursed, strengthening confidence in the program. The modality of payments delivery was also significantly improved, from the initial money orders through the postal system to the introduction of a dedicated BISP debit card with access to any ATM in Pakistan. Most beneficiaries are now receiving their cash transfers through the BISP debit card while beneficiaries in remote communities with limited financial access still receive transfers via money orders. As a result, transparency improved and the proportion of beneficiaries reporting to have to pay a “fee” to an intermediary in order to collect their transfer has declined. Furthermore, a new biometric-based transaction system was rolled out in September 2016 aimed at reducing fraudulent behavior, and contracts with commercial banks are being strengthened.

14. BISP has contributed to lifting targeted households out of poverty, reducing the poverty gap, and improving welfare of beneficiary households. A recent impact evaluation analysis of BISP assessed outcomes for beneficiaries against key objectives, including poverty reduction, women’s empowerment and improved household and child nutrition (Oxford Policy Management, 2015).19 During 2011–14, BISP cash transfers are estimated to have contributed to reduce poverty, with the proportion of beneficiaries living below the poverty line declining by about 20 percentage points, relative to non-beneficiaries. Furthermore, BISP contributed to reducing the depth of poverty among beneficiaries, leading to a reduction in the poverty gap by 3 percentage points relative to non-beneficiaries. BISP has also strengthened women’s empowerment by increasing women’s access to cash and reducing dependence on their husbands’ support. BISP transfers also contributed to reduce malnutrition amongst girls, with rates of stunting falling by 4 percentage points, relative to non-beneficiaries. Despite increasing primary enrollment of BISP beneficiaries’’ children by 10 percent, the impact of BISP on children’s education remains limited, since the size of the educational cash transfer remains low compared to the cost of schooling, one of the main reasons cited by BISP beneficiaries for children not to attend school.20

15. BISP should be strengthened by updating the beneficiaries’ database and broadening coverage, and stepping up educational cash transfers. Challenges include an outdated beneficiaries’ database, potential lack of awareness, and beneficiaries not having identification cards which are required to receive the cash transfer. Broadening BISP coverage is important to strengthen the program’s impact and make a dent on poverty. Notably, reaching additional 1 million poor households could lift out of poverty about 1.5 million additional people and reduce the poverty rate by about 0.7 percent. However, further progress is constrained by several factors, including an outdated beneficiaries database, potential lack of awareness, and beneficiaries not having identification cards which are required to receive the cash transfers. Thus, ongoing efforts to further strengthen the program’s targeting and updating the beneficiaries database based on a planned new national survey are important and should be continued. To this end, the poverty scorecard methodology was updated and a pilot for the new survey was finalized in selected districts, with the full national survey expected to be completed by end-2017 and the new BISP beneficiaries’ database by mid-2018. To ensure high enrollment, efforts should focus on strengthening awareness among the poor, notably in rural areas, of their eligibility, enhancing capabilities to follow geographical movements of eligible beneficiaries, and helping beneficiaries to obtain identification cards. Finally, increasing the size of educational cash transfers provided under BISP is needed to cover costs of schooling and foster children’s school attendance and human capital accumulation.

Other Social Safety Nets in Pakistan

16. Several noncontributory social security and social welfare programs, managed by different Ministries, aim at meeting social welfare needs at the level of local communities. These programs include social pensions to the needy, extending social welfare programs in underdeveloped rural areas, and supporting non-profit organizations. Overall expenditures under the Social Security and Social Welfare programs reached about 0.1 percent of GDP in FY 2014/15, covering about 0.4 million beneficiaries.

17. The public Zakat program is a nation-wide social assistance program which supports poor and needy Muslims. The Zakat program, implemented by the Ministry of Religious Affairs, is based on the Islamic injunction of charity which mandates Muslims to give a percent of their wealth to the poor (zakat).21 Financed by voluntary private contributions, the Zakat program provides financial support to deserving and needy poor Muslims through monthly allowance to households (the Guzara allowance, the program’s main component), stipends to students, and health care coverage. Differently from BISP, the Zakat program lacks a formalized targeting mechanism, and the local zakat committees have substantial discretion on eligibility decisions. Disbursements under the public Zakat program have significantly declined over time to 0.02 percent of GDP in FY 2014/15. In parallel, the number of beneficiaries has also declined to 0.8 million from 2.5 million beneficiaries in the mid-2000s. However, zakat contributions outside the governmental program have continued supporting extensive private philanthropic initiatives so that the number of beneficiaries of the Zakat system outside the official program is much higher.

uA01fig10

Disbursements under the Zakat program have halved

(Disbursments under the Zakat program; in percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan authorities; and IMF staff calculations.

18. The Bait-Ul Mal program and the Workers Welfare Fund are the other main social safety nets programs. Launched in the early 90s, the Bait-Ul-Mal program is a budgetary-funded program which provides housing, financial support and education to orphans as well as financial assistance to widows, elderly and invalids. Disbursements under the Bait-Ul-Mal program have declined over the last decade, reaching 0.01 in percent of GDP in FY 2014/15, with the number of beneficiaries almost halving to 0.9 million. Furthermore, the Workers Welfare Fund, a federally-managed fund governed by a tripartite body—representing the government, employers and workers—and financed by industrial establishment, aims at improving industrial workers’ welfare through several initiatives, including the provision of low-cost housing, marriage and death grants, educational scholarships for workers’ children, construction of education and health facilities.22 Disbursements under the Workers Welfare Fund were at 0.01 percent of GDP in FY 2014/15, covering about 21,000 beneficiaries.

uA01fig11

Disbursments under the Bait-Ul-Mal have been declining

(Bait-UL-Mal disbursments; in percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan authorities; and IMF staff calculations.

19. In addition to public social safety nets, sizable private philanthropic initiatives aim at reducing poverty and fostering development. Private philanthropy in Pakistan is large, estimated at 1.1 percent of GDP, more than twice the size of the BISP program.23 In particular, individual philanthropic initiatives reach 0.8 percent of GDP while philanthropy by the diaspora abroad represent about 0.2 percent of GDP.

uA01fig12

Philanthropy is sizable in Pakistan

(disbursments under philantropic initiatives; percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan Centre for Philantrophy (2016).

The Role of Subsidies

20. Subsidies remain an important element through which social assistance is delivered in Pakistan. Overall subsidies at the federal level have significantly declined over the last years, from about 3 percent of GDP in FY 2011/12 to close to 0.8 percent in FY 2015/16. Electricity subsidies represent the bulk of subsidies, at about 0.6 percent of GDP in FY 2015/16. On a much smaller scale, a price subsidy scheme on fertilizers for agricultural producers to support the sector and subsidies on some food items, such as wheat and sugar, are in place.

uA01fig13

Subsidies are sizable in Pakistan

(Subsidies; in percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan authorities; and IMF staff calculations.

21. Electricity subsidies were significantly reduced owing to reforms efforts while, in parallel, targeted cash transfers to the poor were increased. In an environment of lower oil prices, electricity subsidies were reduced by 1½ percent of GDP during 2013–16, on the back of increases in electricity tariffs, the introduction of surcharges to better reflect costs, and the elimination of untargeted generalized subsidies for commercial and industrial consumers and for the highest-volume residential consumers.24 The targeting of electricity subsidies was strengthened, with subsidies maintained for selected consumer categories. Notably, a life-line tariff is charged to vulnerable consumers (electricity usage up to 50 kWh/per month) and concessional tariffs are charged for low to moderate consumption levels (usage up to 300 kWh/per month) and for agriculture tube wells to support the agricultural sector. In parallel to the subsidies reform, social safety nets were strengthened and cash transfers to the poor under BISP were increased by about 0.3 percent of GDP.

uA01fig14

Electricity subsidies were reduced and social safety nets strenghtened

(In percent of GDP)

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Pakistan authorities; and IMF staff calculations.

22. Continuing reform efforts to further reduce electricity subsidies are needed to achieve better targeting of the poor and free up resources for growth-supporting priority spending. The electricity consumption threshold for the concessional tariff remains high, with most households consuming less than 300 Kwh/per month and thus benefiting from subsidies, while subsidies for the life-line tariff for the smallest consumers (up to 50 Kwh/month) account for only about 8½ percent of the overall subsidies envelope. Lowering the consumption threshold for the concessional electricity tariff would contribute to making remaining electricity subsidies less regressive and reducing their level. Also, since many among the poor might not have electricity access, subsidies could be fully eliminated and targeted cash transfers to protect the poor could be stepped up.

C. Frontier Analysis: The Efficiency of Social Safety Nets and Social Programs in Pakistan

23. A frontier analysis approach provides useful insights on the efficiency of Pakistan’s public spending on social safety nets and social programs towards achieving their objectives. A cross-country comparison of public spending in social safety nets and in education and health against selected social indicators is used to assess how Pakistan’s spending on these programs measures up against progress in reducing poverty and inequality and improving education and health outcomes. However, these results have to be interpreted carefully since many factors contribute to the extent of poverty and inequality as well as education and health outcomes.

Efficiency of Spending in Social Safety Nets in Reducing Poverty and Inequality

24. Public spending on social safety nets in Pakistan remains relatively low amid still high poverty. While many other factors affect poverty outcomes, a cross-country comparison shows that spending on social safety nets, excluding subsidies and provincial outlays, in Pakistan remains relatively low and poverty has remained relatively high compared to other countries (Figure 4).25 First, most countries have a higher level of spending in social safety nets (in percent of GDP) than Pakistan which, in turn, tends to be associated with lower poverty headcounts and multi-dimensional poverty rates. Furthermore, most countries with a level of public spending on social safety nets close to or lower than Pakistan tend to have lower poverty and multi-dimensional poverty rates.

Figure 4.
Figure 4.

Pakistan: Lower Spending on Social Safety Nets and Higher Poverty and Inequality 1/

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: Human Development Reports (2016), ASPIRE Database (2016), WDI (206), and IMF Staff estimates.1/ The sample comprises middle income economies (2015 GNI per capital was between US$1,026 and

25. Similarly, the relatively low level of spending on social safety nets in Pakistan is associated with relatively higher inequality. Focusing on non-monetary indicators of inequality such as education and health gaps among the top and the bottom quintile, a cross-country comparison highlights that inequality is relatively high given the level of spending in social safety nets in Pakistan (Figure 4). Notably, countries with spending on social safety nets (in percent of GDP) close to or lower than Pakistan’s tend to have smaller gaps in years of schooling and stunting among children between the poorest and the richest quintiles.

Efficiency of social spending on education and health outcomes

26. Public spending on education remains relatively low amid weaker outcomes in Pakistan. A cross-country comparison highlights that Pakistan’s public spending on education (in percent of GDP) lags behind comparator countries, with most countries having higher public spending in education associated with better education outcomes (Figure 5).26 Furthermore, most countries with a level of public spending on education close to or lower than in Pakistan tend to have better education outcomes such as higher youth literacy rates and primary school enrollment ratios.

Figure 5.
Figure 5.

Pakistan: Lower Social Spending in Education and Health and Weaker Outcomes 1/

Citation: IMF Staff Country Reports 2017, 213; 10.5089/9781484309766.002.A001

Sources: WDI (2016).1/ The sample comprises middle income economies (2015 GNI per capital was between US$1,026 and

27. Similarly, public spending on health in Pakistan is very low compared to other countries and it is associated with weaker outcomes. Pakistan’s public spending on health (in percent of GDP) is among the lowest in the sample, with most countries having higher spending than Pakistan and stronger health outcomes (Figure 5). Furthermore, data suggests that countries with spending on health close to Pakistan’s tend to display better health outcomes as higher life expectancy and lower infant mortality rates.

D. Policy Recommendations and Conclusions

28. Significant progress has been made over the past decades to reduce poverty in Pakistan. However, about 30 percent of the population still lives below the poverty line, about 39 percent experiences multi-dimensional poverty, and children’s stunting and malnutrition remain high. Education and health outcomes are weaker than South-Asia average and social spending in these areas remains below emerging markets’ average, in part reflecting implementation and capacity constraints at the provincial level. While income inequality is relatively moderate, gaps in education and health outcomes between genders and the richest and the poorest quintiles are sizable. The size of the informal economy is estimated to be large, which is likely to contribute to these outcomes.

29. Generating higher and more inclusive growth is the top priority to make a dent poverty and inequality. Further efforts are needed to raise the living standards of the population, generating sustainable jobs and ensure shared equality. To this end, Pakistan will need to preserve macroeconomic stability, a necessary pre-condition to foster higher private investments and private-sector led growth. Furthermore, moving ahead with key growth-supporting structural reforms is crucial, in particular advancing the energy sector reform, restructuring and privatizing public sector enterprises, improving the business climate and strengthening governance. Dedicated efforts are also needed to strengthen gender equality, including through reforms to boost female labor force participation.

30. Stepping up expenditures on social safety nets is needed to support the most vulnerable. With high poverty and inequality, social safety nets have a critical role to play to support the poor and protect the most vulnerable. However, despite having strengthened over time, the size of social safety nets remains low against regional and emerging markets’ averages. Furthermore, efficiency frontier analysis suggests that low spending on social safety nets is associated with higher poverty and inequality. BISP, the main social safety net program, provides cash transfers to poor households using an effective targeting mechanism and it is underpinned by sound delivery infrastructure. However, it remains small and thus has a limited impact on poverty reduction.

31. Keeping the reform momentum is needed to further reduce electricity subsidies and free up public resources to strengthen social safety nets and growth-supporting priority spending. Subsidies remain sizable, with an overall envelope larger than the combined disbursements under social safety nets programs. Electricity subsidies, representing the bulk of subsidies, were substantially reduced in recent years but remain sizeable and poorly targeted. Further reducing electricity subsidies and improving their targeting are needed to free up public resources and increase targeted social safety nets and growth-supporting expenditures. To this end, reducing the consumption threshold to benefit from the concessional electricity tariff, or fully eliminating electricity subsidies while increasing targeted cash transfers to the poor should be explored.

32. Broadening BISP coverage, updating its beneficiaries’ database, and increasing educational transfers will improve the efficiency of social safety nets in reducing poverty and inequality. Swiftly extending BISP coverage to include all eligible poor households, while continuing to protect beneficiaries’ purchasing power, will be key to increase the program’s impact towards reducing poverty and inequality. In this vein, resources allocated to smaller and poorly targeted social assistance programs could be consolidated into BISP thus increasing overall efficiency. In addition, finalizing ongoing efforts towards strengthening BISP targeting and updating the BISP beneficiary database will allow to better reach the poorest. Finally, increasing educational conditional cash transfers under BISP would allow to better cover children’s schooling costs and raise the program’s impact on school enrollment, attendance and education outcomes.

33. Furthermore, stepping up public social spending is needed to improve education and health outcomes. Despite having slightly increased, Pakistan’s public spending in education and health remains very low compared to emerging markets’ average. Efficiency frontier analysis suggests that social spending in Pakistan is low compared to other countries, and is associated with weaker education and health outcomes. Thus, moving forward with reforms to mobilize higher fiscal revenues is needed to create additional fiscal space to increase growth-supporting social spending. Furthermore, strengthening the implementation capacity of the provinces is needed to ensure higher and more efficient social spending in education and health.

References

  • Alkire, S., Jindra, C., Robles, G., and Vaz, A., (2016), “Multidimensional Poverty Index 2016: Brief Methodological Note and Results,” OPHI Briefing 42, University of Oxford.

    • Search Google Scholar
    • Export Citation
  • Kemal A. and A. Qasim, (2014), “Precise Estimates of the Informal Economy,” Pakistan Institute for Development Economics.

  • Government of Pakistan, (2009–2015), “Annual PRSP Progress Reports

  • International Monetary Fund, 2014, “Towards New Horizons: Arab Economic Transformation Amid Political Transitions

  • International Monetary Fund, 2014, “Subsidies Reform in the Middle East and North Africa: Recent Progress and Challenges Ahead

  • Schneider F., 2012, “The Shadow Economy and Work in the Shadow: What Do We (Not) Know?,” Discussion Paper No. 6423, IZA.

  • World Bank, 2011, “Social Safety Nets,” An evaluation of World Bank Support, 2000–10.

  • World Bank, 2012, “Inclusion and Resilience: the Way forward for Social Safety Nets in the Middle East and North Africa

  • World Bank, 2015, “Addressing inequality in South Asia

  • World Bank, “The State of Social Safety Nets

  • World Bank, 2016, “Pakistan Development Update: Make Growth Matter

  • World Bank, 2016, “Residential Electricity Subsidies in Pakistan: Targeting, Welfare Impacts, and Options for Reform

1

This paper was prepared by Giorgia Albertin.

2

The government adopted in 2016 a new poverty line, with the World Bank support, which set a higher and more inclusive standard of well-being compared to the poverty line set in 2001. Based on the Cost of Basic Needs method, a PRs 3,032 per adult/per month poverty threshold was identified, leading to a poverty headcount of 29.5 percent of the population in 2013, with about 60 million people in Pakistan classified as poor. Under the former 2001 poverty line based on the Food Energy Intake method, the poverty threshold was at PRs 732.4 per month/per person leading to poverty headcount rate of 34.7 percent in 2001, declining to 9.3 percent in 2013. Back-casting the new 2016 poverty line, the poverty headcount was at 64 percent in 2001, with the same declining trend during 2001–13.

3

Pakistan has one of the lowest poverty incidence within the South-Asia region, based on the US$1.90 a day poverty line.

4

World Bank Development Indicators (2016).

5

The Balochistan region represents 6 percent of the overall population.

6

The Human Development Index (HDI) is a summary measure of average achievement in key dimensions of human development: a long and healthy life, being sufficiently educated, and have a decent standard of living. The health dimension is assessed by life expectancy at birth, the education dimension is measured by mean of years of schooling for adults aged 25 years and more and expected years of schooling for children of school entering age. The standard of living dimension is measured by gross national income per capita.

7

Stunting is defined as the percentage of children under five whose height-for-age ratio is two standard deviations or more below the World Health Organization (WHO) Child Growth Standards. Wasting is defined as the percentage of children under five whose weight-for-height ratio is two standard deviations or more below the WHO Child Growth Standards. World Bank Development Indicators used in the analysis are based on the Pakistan Demographic and Health Survey (PDHS) 2012–13. New data on stunting and malnutrition will be released in the PDHS 2017/18.

8

Data on gross enrollment ratio distribution per quintile is based on Pakistan Social and Living Standard Measurement Survey 2013–14.

9

The Gender Gap Index (World Economic Forum, 2015) measures the gap between men and women in four categories: economic participation and opportunity, educational attainment, health and survival, and political empowerment.

10

Few empirical studies have so far provided an empirical assessment of the impact of the informal economy. Elbadawy and Loayza (2008) show that an increase in informality leads to a decrease in economic growth thus hurting inclusiveness while having a positive impact reducing the incidence of poverty.

11

Social safety nets are noncontributory measures designed to provide regular and predictable support to poor and vulnerable people and are a component of larger social protection systems (World Bank, 2015). Social safety nets include noncontributory transfers (unconditional and conditional cash-transfers, food and other in-kind transfers, school feeding programs), noncontributory social pensions, measures to provide access to essential services as education, health and housing through fee-waivers, public works and social care services. Generalized subsidies are typically not considered as part of social safety nets due to their regressive nature.

12

This paper focuses on non-contributory social assistance programs at the federal level. Several provincial initiatives, including the Punjab Khidmat Card, the Insaf Card and low cost provision housing in Sindh also contribute to alleviate poverty. Furthermore, contributory social protection programs are also in place in Pakistan, such as the Employees Old Benefits Initiative and other social security programs.

13

Efficient social safety nets are characterized by broad coverage and appropriateness of benefits, poverty-based targeting using proxy means testing, consolidation of fragmented programs, unified registry of beneficiaries to be used by different programs, modern service delivery mechanisms, programs to strengthen human capital, strong governance and dissemination of information to the poor on available programs (IMF (2014), World Bank (2012)).

14

The BISP cash transfer is paid to the adult woman in the household, defined as every ever-married woman in the household with a valid Computerized National Identity Card.

15

In addition to UCTs and CCTs, other complementary programs are implemented under BISP as Waseela-e-Rozgar providing educational training to the youth, Waseela-e-Haq providing micro-loans to female beneficiaries and Waseela-e-Sehet providing health insurance to beneficiaries.

16

A cut-off income threshold for eligibility to BISP was established (PRs 6,000 monthly income per family), families could apply for income support, and each parliamentarian was assigned to review an equal number of applications, irrespective of political affiliation, and decided on the family’s eligibility.

17

The survey was conducted on the basis of a questionnaire relying on a wide range of questions, including composition and characteristics of household roster, age, education, employment and disability status of household members, nature of dwelling, moveable and fixed assets. The Poverty Scorecard generated a poverty score for each surveyed households (from 1–100) and household below the established cut-off score (of 16.17) were considered eligible to receive cash transfers through BISP.

18

BISP quarterly stipends were increased from PRs 3,000 at end-June 2013 to PRs 4,700 at end-June 2016. In particular, stipends were increased by 4.5 percent in FY 2015/16 in order to protect the beneficiaries’ purchasing power.

19

A second round impact evaluation analysis was conducted in 2014 to assess the impact of BISP on beneficiaries against key objectives of the program: poverty reduction, women’s empowerment, improved household and child nutrition, and increased asset retention. A quasi-experimental method was used, based on a comparison between a treatment group of beneficiaries of BISP against a control group of households’ non-beneficiaries of BISP but just above the BISP threshold scorecard (Oxford Policy Management, 2015). This followed the first round evaluation impact analysis conducted in 2013.

20

The cost of schooling includes costs of uniforms, books, supplies, transports and others (Oxford Policy Management, 2015).

21

A tax levy (2.5 percent) used to be imposed on financial assets of individuals, including bank deposits above a certain threshold, and collected into the Central Zakat Fund. Since 1999, the supreme court ruled for zakat to be voluntary.

22

The Workers Welfare Fund is financed by contributions from industrial firms with total income above PRs 5000,000, which have to provide to the Fund two percent of their profits.

23

This includes private, corporate and diaspora philanthropy (Pakistan Centre for Philanthropy).

24

The Tariff Differential Subsidy (TDS) is the main component of electricity subsidies, a transfer from the government to the power distribution companies (DISCOs) to compensate for the difference between tariff that would allow each DISCO to fully recover their costs and the Uniform Minimum Tariff notified at the national level. In addition, certain categories of consumers are protected by being charged a tariff which is below the determined minimum tariff.

25

To ensure cross-country comparison, poverty incidence data used in the frontier analysis are based on a US$3.1 per day poverty (World Bank Development Indicators). This leads for Pakistan to a poverty rate of 38 percent, higher than under the 2016 national poverty line (29.5 percent).

26

Private schooling is sizable in Pakistan, with enrollment representing more than 1/3 of enrollment in public schools.

Pakistan: Selected Issues
Author: International Monetary Fund. Middle East and Central Asia Dept.
  • View in gallery

    Pakistan: Poverty has Declined but Remains High

    (Poverty headcount in percent of total population)

  • View in gallery

    Access to Electricity is almost universal in Pakistan

    (In percent of population)

  • View in gallery

    Access to improved sanitation facilities in Pakistan is above regional average

    (access to improved sanitation facilities; in percent of population)

  • View in gallery

    Pakistan: Poverty is High and Children’s Malnutrition is Widespread

  • View in gallery

    Pakistan: Challenging Education and Health Outcomes amid Limited Public Spending

  • View in gallery

    Pakistan: High Inequality Across Many Dimensions

  • View in gallery

    Pakistan: A large Informal Economy

    (Informal economy; as a share of GDP)

  • View in gallery

    Pakistan: Social Safety Nets are Relatively Small

    (Expenditures on social safety nets; in percent of GDP, latest available data)

  • View in gallery

    Self-employment is Significant in Pakistan

    (In percent of total employment)

  • View in gallery

    Affiliation to Pension Schemes is Very Low in Pakistan

    (In a percent of labour force)

  • View in gallery

    BISP is the Largest Social Safety Net in Pakistan

    (In percent of GDP)

  • View in gallery

    Disbursments under BISP have streghtened over time

    (BISP disbursments; in percent of GDP)

  • View in gallery

    Disbursements under the Zakat program have halved

    (Disbursments under the Zakat program; in percent of GDP)

  • View in gallery

    Disbursments under the Bait-Ul-Mal have been declining

    (Bait-UL-Mal disbursments; in percent of GDP)

  • View in gallery

    Philanthropy is sizable in Pakistan

    (disbursments under philantropic initiatives; percent of GDP)

  • View in gallery

    Subsidies are sizable in Pakistan

    (Subsidies; in percent of GDP)

  • View in gallery

    Electricity subsidies were reduced and social safety nets strenghtened

    (In percent of GDP)

  • View in gallery

    Pakistan: Lower Spending on Social Safety Nets and Higher Poverty and Inequality 1/

  • View in gallery

    Pakistan: Lower Social Spending in Education and Health and Weaker Outcomes 1/