The webinar Social expenditure, legislative frameworks and findings of impact evaluations of social protection in South Asia was the third event of the Social Protection in South Asia series. It took place on 23 July 2020 and was jointly organised by the IPC-IG and UNICEF. The moderator was Usha Mishra (UNICEF) who was joined by presenters Luca Lazzarini, Carolina Bloch, and Yannick Markhof (all IPC-IG).

You can see the full presentation here and watch the recording here.

Under discussion were the main findings and takeaways of three recent IPC-IG papers in collaboration with the UNICEF Regional Office for South Asia (ROSA). Specifically, presenters explored some of the following questions: What are the main features of existing legal and regulatory frameworks that lead to promoting children’s rights through social protection in South Asia? Do the legal frameworks of the region’s non-contributory social protection programmes comply with a human rights-based approach and global UNICEF social protection policy? What is the current level of public social expenditure in South Asian countries and how does the region compare globally? And finally: What evidence is there for the impact of social protection programmes in South Asia on economic growth and improvements in human development indicators?

The first presentation, by Luca Lazzarini, started with an analysis of social protection legislation from international, national and regional perspectives. He provided an assessment of the compliance of programmes’ regulatory frameworks against the criteria of the human rights-based approach to social protection.

The international outlook

The main finding is that every country in the region has ratified the Convention on the Rights of the Child and is therefore bound to implement it. At the national level, the provisions relate to the human rights to social security and to an adequate standard of living. Both belong to economic, social and cultural rights: as such, they will be achieved gradually and are subject to the availability of resources. Yet, States are required to ensure, with immediate effect, certain minimum levels of human rights—known as the ‘international minimum threshold’.

The regional perspective

The presentation then shifted to the regional perspective: South Asia is the only global region which currently lacks a regional human rights monitoring system. Therefore, there are neither regional mechanisms for determining State accountability on alleged human rights violations, nor engagement in the monitoring and promotion of human rights. Furthermore, Luca spoke about strengthening the regional cooperation in social development within the context of the South Asian Association for Regional Cooperation (SAARC).

The national level

From a constitutional perspective, it was noted that the constitutions of every country—except for Afghanistan—contain provisions related to social protection. Even so, Afghanistan has a rule dedicated to the assistance of children and other vulnerable groups. Legal frameworks for social protection are lacking in Bhutan, Bangladesh, India and Pakistan—however, a social protection act is currently being discussed in Bangladesh. Afghanistan, Maldives, Nepal and Sri Lanka have all enacted social protection legislation within the past six years. This trend can be interpreted as an important step towards building social protection systems anchored in a rights-based approach.

Regarding child-focused legislation, countries tend to regulate single matters through individual pieces of legislation. Over time, this has resulted in a significant amount of legislation, which has sometimes resulted in conflicting provisions. Afghanistan, Maldives and Nepal have enacted child-focused legislation, which explicitly includes children’s right to social protection and further provisions to fulfil this right. Moreover, the regulations are particularly commendable for their effort in systematising national legislation establishing children’s rights. All laws were enacted in between 2018 and 2019.

The programme level

Of a total of 51 programmes mapped, 31 are not embedded in regulatory frameworks. The remaining 20 programmes that do have a regulatory framework present significant differences in terms of compliance with the human rights-based approach to social protection, as depicted in the figure below.

The most important takeaway is that accessible complaint and appeal mechanisms are significantly lacking in the legislation analysed. It is worth noting that the ability to file a complaint (and eventually to appeal) is the most basic form of guaranteeing accountability and transparency, and represents a crucial step in the shift towards a human rights-based approach to social protection, where the institutions—as duty bearers—are held accountable.

Government intent vs. action

The second presenter of the webinar, Carolina Bloch, discussed the spending on social protection in South Asia. As poverty and social exclusion have garnered growing attention in South Asia since the early 2000s, governments in the region have strengthened their commitments to ensure equal access to basic health care, education and an adequate income. However, the path to materialise government promises into well-being for the population has been unsteady, as poverty and inequality are still high in most countries in the region and structural factors have hindered the provision of social services.

As illustrated in the following figure, she highlighted that on average South Asian governments spend less than other regions on health, education and social assistance as a share of gross domestic product (0.95, 3.4 and 0.9 per cent of GDP on average, respectively), but there is a great heterogeneity within the region, both in terms of total spending and of government priorities.

Health, education and social assistance spending by country, as a percentage of GDP.

Health spending in South Asia

Government provision of health services is crucial to reduce the burden of excessive health costs on households and avoid catastrophic spending or inadequate treatment, which are a reality for a large share of South Asia’s population. While the government is responsible for over 70 per cent of health care financing in Bhutan and Maldives, and almost 50 per cent in Sri Lanka, other countries have huge shares of out-of-pocket spending.

The regional comparison shows that countries with similar levels of spending can have wildly different outcomes. Carolina highlighted the case of Sri Lanka, which achieved outstanding performance with a lower level of spending than most of its neighbours. The country concentrates the largest share of its health expenditure on services mostly used by the low-income population and out-of-pocket expenditures are incurred mainly by richer households. However, structural factors should also be considered: in Maldives, for instance, the maintenance costs of health services are very high as the country is composed of hundreds of islands. 

Education spending in South Asia

Overall education seems to be a government priority in the region. Most countries are below, but not far from UNESCO’s Framework for Action proposal, which is to allocate between 4 per cent and 6 per cent of GDP and from 15 per cent to 20 per cent of total government spending to education. According to UNESCO’s estimates, there are 11 million out-of-school children of primary age in South Asia, and 19 million of lower secondary age. However, countries such as Sri Lanka, Maldives and Nepal are close to achieving universal basic education. Again, similar proportion of public spending on education can lead to completely different patterns in education indicators. Afghanistan and Maldives, for instance, spend about the same share of GDP (4.1 per cent), but have completely different outcomes and challenges. Possible explanations for this heterogeneity include each country’s income level, institutions, demography, demand-side factors such as participation rates and geographical remoteness, which should be key inputs to determining their current and future financial needs.

Social assistance expenditures in South Asia

While many South Asian countries have strengthened commitments to improve social protection systems, a large share of the population remains excluded. ASPIRE estimates that the average spending on social assistance in South Asia is at 0.9 per cent of GDP, which places the region at the bottom of the global distribution. India, Nepal and Maldives are the only countries to exceed 1 per cent of GDP, though not by much. India, which has the highest spending—at 1.5% of GDP—directs the largest share of these expenditures to food and in-kind transfers. Nepal allocates more funds than any other country in the region to unconditional cash transfers, and Maldives spends considerably more on social pensions. On the other side of the distribution, Bangladesh, Sri Lanka, Pakistan and Bhutan spend less than 1 per cent of GDP (there were no data available for Afghanistan).

Fiscal space for social spending

Despite structural challenges, South Asia is the fastest-growing region in the world, and even countries with a tight budget have the potential to increase, or at least improve investment in the social sector. Most countries underexploit tax collection as a financing mechanism. Increasing direct taxation, particularly individual income taxes, would create more room for progressive taxation, which would be praiseworthy in all countries, but especially in Maldives and Sri Lanka, where inequality is highest. Bangladesh could also be in a particularly good position to implement tax reforms, since its tax revenues remain low despite strong economic growth over the last years.

Assessing inefficiencies and re-programming public expenditures is an option that all countries could consider. This means consistently evaluating expenditures and redirecting resources from high-cost/low-impact spending towards more progressive social policies, but also improving institutional capacity to implement development budgets, including dealing with budget under-execution, implementation delays and heterogeneity of governance between government units, which are problems commonly identified in the region.

A rigorous evidence basis

In the last presentation of the webinar, Yannick Markhof finally turned to the evidence linking social protection programmes in South Asia with economic and human development outcomes. The IPC-IG study he presented conducted a literature review of quantitative impact evaluations and built a sample comprising 37 impact evaluations of 16 social assistance programmes in six South Asian countries.

Equipped with this selection of studies, the authors aimed to compile a rigorous evidence basis for general and programme-specific takeaways. To this end, they proceeded in two steps.

First, they classified the effects found by each study into desirable, insignificant, and undesirable impacts. In doing so, they also discern between impacts applicable to the general population of programme beneficiaries and those accruing only to a subgroup.

Second, since different impact evaluations study similar concepts, such as poverty or women’s empowerment, through slightly different proxies, the authors considered those indicators assessing a similar concept together. Through this aggregation step, the study ended up summarising the findings of each study and for each programme along 31 broad indicators of impact.

Putting social assistance programmes in South Asia to the test

By considering the findings from many different studies, we can get a broad picture of the performance of different social assistance programmes in South Asia. In the webinar, Yannick focused on only a small selection of the most pertinent takeaways for indicators belonging to the categories of poverty and finances, labour market, education, sexual, maternal and child health, and gender.

Poverty and Labour Markets

Concerning poverty impacts, Yannick pointed out the positive examples of BISP in Pakistan, which would lead to lasting, desirable impacts on consumption, assets, and poverty, and the EGPP in Bangladesh. Here, evidence supports the notion that the EGPP provides a steady income stream across the year and substitutes for loan-financed consumption spending during seasonal unemployment.

Earlier, Yannick had noted that the world’s largest public works programme—India’s MGNREGA, led to significant increases in income. This would be especially the case for women, reflecting their otherwise limited opportunities on the labour market. Turning to labour market impacts, he then pointed out that this effect seems driven precisely by higher labour force participation and desirable general equilibrium impacts on women’s wages. This would be especially interesting as not just beneficiaries would receive higher wages, but the general gender pay gap would decrease in districts where the programme is in place. Furthermore, he highlighted that the FSSSP, a secondary school scholarship for girls in Bangladesh, BISP and MGNREGA all incentivise desirable changes in the sector of occupation. Finally, no programme in the sample adversely affected child labour.

Education, sexual, maternal and child health, and gender

Regarding education, Yannick emphasised the importance of a comprehensive assessment of effects capturing the extensive and intensive margins, as well as performance. The conditional cash transfer component of BISP, for example, increased enrolment but left attendance unaffected, while the FSSSP improved educational attainment and secondary school completion rates.

A number of programmes assessed aimed to promote safe delivery practices for pregnant women. While most were able to fulfil this mandate, it was worth highlighting that supply-side considerations, such as hospital service quality, and programme design—such as trust in benefit pay-out—mattered for outcomes. Another interesting observation was that building human capital seemed to influence women’s reproductive decisions by delaying marriage and reducing fertility rates.

Lastly, Yannick pointed to the gender category as an encouraging example. Here, he emphasised the importance of a gender-sensitive programme design to elicit desirable impacts on women’s social status and political participation, but also leading to changes in women’s actual behaviour and rights. Perhaps unsurprisingly, important means of empowerment here seemed to be the opportunity to work and education.

Evaluating social assistance programmes in South Asia: the takeaways

Yannick concluded his presentation by focusing on six succinct takeaways for policymakers and researchers. He also emphasised that a many more programme-specific recommendations can be found in the study itself, which is due for publication later this summer.

The presentations were followed by a fruitful Q&A session, which can be accessed here.



(forthcoming). “Social spending in South Asia: an overview of government expenditure on health, education and social assistance”. IPC-IG Research Report. Brasília: International Policy Centre for Inclusive Growth. 

(forthcoming). “Social protection legislative frameworks in South Asia from a child-rights perspective”. IPC-IG Research Report. Brasília: International Policy Centre for Inclusive Growth.

(forthcoming). “Evidence linking social protection programmes in South Asia with child poverty, economic growth, and improvements in human development.” IPC Research Report. Brasilia: International Policy Centre for Inclusive Growth.


This was the third webinar of the Series ''Social Protection in South Asia – the landscape before COVID-19, a snapshot into responses to the crisis and the paths ahead'', jointly organised by the IPC-IG, UNICEF Regional Office for South Asia (ROSA), and UNICEF Country Offices.

Social Protection Programmes: 
  • Social assistance
Social Protection Building Blocks: 
  • Policy
    • Laws and Policies
    • Expenditure and financing
    • Monitoring and evaluation systems
Cross-Cutting Areas: 
  • Education
  • Gender
  • Health
    • Child health
    • Maternal health
  • Human rights
  • Labour market / employment
  • Poverty reduction
  • Afghanistan
  • Bangladesh
  • Bhutan
  • India
  • Maldives
  • Nepal
  • Pakistan
  • Sri Lanka
  • South Asia
The views presented here are the author's and not's