By Nicholas Mathers
The Government of Nepal’s FY2016/17 budget, announced on 28th May, includes a commitment to enhance and expand the Child Grant nationally to all children in poor households.
The proposed reforms represent an exceptional step forward in social policy for a country that is recovering from a devastating earthquake and is still striving to emerge from Least Developed Country (LDC) status. But does it make sense to provide the Child Grant only to poor households?
First, a little background. Since 2010, the Child Grant has provided Rs. 200 ($2) per month to families with under-fives - barely enough to make difference - and has been limited in coverage to the remote and highly food insecure region of Karnali and to Dalit households elsewhere in the country. While coverage averages 75-80 per cent of eligible children, implementation weaknesses mean that payments can be irregular - and sometimes missed altogether - and often the youngest children miss out due to the lengthy registration process.
There is reason for optimism, however. Although the transfer amount is small, recipients already use the money in the best ways that they can for the well-being of their children. There is evidence that the Child Grant leads to small but statistically significant increases in dietary diversity, expenditure on medicines, and access to credit which helps with consumption smoothing. Given these are important underlying factors of good nutrition, an increased benefit amount and improvements in delivery systems and process will likely lead to measurable changes in nutritional status.
Good news then, as the reforms will mean a doubling of the benefit value to Rs. 400 ($4) per month as well as initiation of the expansion beyond Dalit households. Government and development partners are also working on strengthening implementation systems so that delivery is more effective, efficient and accountable.
The issue of targeting at children in poor households, however, remains problematic.
The rationale for poverty targeting is that it is more efficient, allowing sometimes scarce resources to be concentrated on those who need it most. The efficiency argument can be appealing, but there are potentially significant costs associated with targeting - (mis)identification of the poor, complex and costly implementation, negative behavioural incentives, and social and political tensions between those who qualify and those who do not. Universal approaches, on the other hand, are by definition more inclusive and are less demanding to implement - an important consideration in countries - like Nepal - with widespread poverty and limited administrative capacity.
There are several good reasons, therefore, why poverty targeting the Child Grant doesn’t make sense.
• First, poverty rates, especially among children, are high. While the national poverty rate is around 25 per cent, 36 per cent of children under-five live under the poverty line. If you consider those who are also ‘near-poor’ - close to the poverty line and at risk of becoming poor - that leaves just 29 per cent who can be considered reasonably income secure. In this context, one of the problems that targeting aims to address – ‘leakage’ of benefits to the wealthy – may not be of such concern.
• Second, current targeting approaches to the Child Grant lead to the exclusion of a majority of poor under-fives by virtue of their very design. Maintaining the Dalit-only approach, or extending the use of the simple proxy means test (PMT) that is currently applied, at least in policy, to the Dalit Child Grant, will result in exclusion of more than 70 per cent of income poor under-fives (see table). A universal approach would logically result in no exclusion error (assuming perfect implementation) and, as indicated above, include just 29 per cent who can be considered truly better off.
Exclusion and inclusion errors for under-fives under different targeting designs
Data source: Nepal Living Standards Survey (NLSS) 2010
• Third, actual experience in Nepal suggests implementing poverty targeting is practically, politically and financially challenging. The simple PMT mentioned above is rarely applied in practice due to policy distortion, weak administrative capacity, and limited social acceptance - ‘we are all poor here’ is a common refrain. A more complex PMT trialled in two districts was estimated to cost 22 per cent of the programme budget for the targeting alone – considerably more than the average of 4 per cent estimated for programmes in middle income countries.
International evidence supports the idea that poverty targeting often brings little additional value, especially in low income contexts. World Bank analysis shows that child grants have been shown to perform as well as means tested and PMT targeted programmes in transferring resources to poor households, despite not being specifically designed to do so.
Poverty-targeting of social transfer programmes can be beneficial in certain circumstances. In the case of Nepal’s Child Grant, however, trying to reach the poor would lead to largely undesirable outcomes. And a universal Child Grant for under-fives is not pie-in-the-sky thinking from a financial perspective, costing an estimated 0.67 per cent of GDP at the new benefit levels.
As the Government follows through on its commitment to initiate an expanded and enhanced Child Grant it would do well to consider:
• Taking a universal approach such that all households with children under 5 years old are eligible for the programme.
• Recognising the need to incrementally expand the Child Grant over time, prioritise certain districts based on human development index (HDI), poverty rates or exposure to natural disasters or other risks.
• Alongside expansion, continue to invest in strengthening delivery systems such that unintentional exclusion and inclusion errors are minimised and the full potential of the Child Grant can be realised.
This blog is a summary of a UNICEF Nepal Policy Brief of the same title. Nicholas Mathers is Cash Transfers Specialist with the Social Policy and Economic Analysis Section, UNICEF Nepal.
 VARG & UNICEF (2014) A cross-sectional survey on reduction of child malnutrition through social protection in the Karnali zone of Nepal. UNICEF Nepal, Kathmandu.
 ‘Universal’ is sometimes used to refer to blanket coverage of the entire population. Here, as commonly understood, it means available to all within the defined social category regardless of income status.
 Based on a higher consumption line that is 1.6 times the national poverty line, similar to the difference between the international extreme and moderate poverty lines.
 The simple proxy means test (PMT) currently applied (in policy) to the Dalit CPG requires that households: do not own their own house; have less than a certain area of farmable land; or can only produce adequate food for less than 3 months of the year.
 Acharya A. et al (2010) Options for the Social Security Pilot Project. UNCDF, Nepal.