Selection of beneficiaries of the country's social protection schemes may be improved if a quality national household database is used. Currently, the beneficiary selection for social safety net programmes is debatable. A number of studies conducted at both local and international levels show that a significant number of deserving people have no access to the social security programmes.
It is learnt from government officials and experts familiar with the process that the selection (or targeting) is not accurate due to nepotism, corruption and political factors. Officials involved in the selection process rely on lists sent by local administrations and public representatives. Some selection criteria are based on survey results and according to the criteria defined by the ministry or division. Few other selection processes require medical certificates.
But the procedure is faulty. The government recently wanted to provide TK 2,500 each following a list of 3.7 million households sent from upazila levels. This was a part of the government-announced financial assistance for vulnerable people who lost their incomes due to covid-19. However, because of data mismatch, the finance division could disburse money to only 1.4 million people. There were discrepancies between NID numbers and corresponding mobile numbers. As a result, the list of remaining 2.3 million people was sent back for verification. The most outrageous part of the list was that it included names of several thousand people who were government employees. A significant number of them had investments in national saving certificates, while a there were others who received pension as retired public servants.
EXCLUSION ERROR: According to 2016 Household Income Expenditure Survey (HIES), there are more than 70 per cent "exclusion errors" or under-coverage when it comes to identifying poor persons. Under coverage or 'exclusion error' denotes the sum of actual poor wrongly classified as non-poor as a proportion of the total poor. On the other hand, leakage which is also known as 'inclusion error' is the sum of actual non-poor incorrectly classified as poor as a proportion of the total poor (Johannsen, 2006).
An analysis by Policy Research Institute of Bangladesh (PRI) used HIES data (BBS, 2017) to estimate the extent of exclusion and inclusion errors. Estimated exclusion errors and inclusion errors in 2016 were 71.0 per cent and 46.5 per cent respectively. Exclusion errors were lower in the rural locations (67.0 per cent) compared to the urban locations (84.0 per cent) due to the lower coverage of social protection system in urban locations. Inclusion errors were slightly higher in rural locations compared to the urban locations.
PRI analysis also showed that poor targeting has turned out to be costly in Bangladesh's pursuit of reducing poverty. Under the current social protection system, additional poverty impacts of 2.2 per cent Gross Domestic Product (GDP) spending on social protection programmes is only 0.42 per cent (i.e. the difference between poverty rate of 24.3 per cent with social protection and 24.81 per cent without social protection). However, improved (or perfect) targeting of deserving beneficiaries may have higher impact on poverty reduction.
To address the situation, the government has undertaken a project to develop a comprehensive database of all households (known as the National Household Database - NHD) and a selection method based on 'Proxy Means Test (PMT)' model to improve the selection of the poor and thereby reduce the high level of exclusion and inclusion errors. The prime objectives of the PMT model as stated in the NHD project (2019) are to: (i) maximise coverage of the poor and vulnerable population given limited budget; (ii) make the system fully consistent with the goal of universal coverage of the poor; and (iii) address the key challenge to build a cost effective system to identify the poor.
However, the outcomes of Bangladesh PMT model are not encouraging, according to PRI analysis. Bangladesh PMT model used HIES 2016 data. The 2016 PMT model is an updated version of the 2010 PMT model. From ten different PMT models, the best model has been selected. It is selected on the basis of key indicators such as under-coverage, leakages and targeting efficiency when a 20 per cent cut-off point is considered. According to the best fit Bangladesh PMT model (2016), under coverage is 60 per cent. Therefore, 40.0 per cent of the poor are left out of the social protection net. This is a theoretical error. If 5.0 per cent more is added to implementation error - the level of exclusion error is close to the current exclusion error of 70.0 per cent. The inclusion error is also not small at 34.0 per cent. Again, if a 5.0 per cent implementation error is added, the inclusion error (i.e. 39.0 per cent) is close to current exclusion error of 47.0 per cent. However, these results are not surprising. It is well known and accepted that at a lower level of coverage (e.g. 20.0 per cent), the exclusion and inclusion errors will be high. The level of errors reduces with the expanded coverage. It is clear that universal coverage of the poor will not be ensured using the PMT model.
OPINIONS OF POLICYMAKERS & EXPERTS: Bangladesh Bureau of Statistics (BBS) has conducted a census to produce a national household database and it may release the outcome by next August. It is expected that it will significantly reduce the exclusion error. The rate of exclusion error was much higher in the previously conducted HIESs -HIES, 2010 and HIES, 2005.
In this connection, Prof Dr Shamsul Alam, a member and senior secretary at the General Economics Division of the Planning Commission, said the Financial Express (FE): "We will be able to identify truly deserving people when we will start using a national household database." He also added that the database will be an effective instrument for the selection process as the census was made with a well-defined questionnaire along with other features that will help screen out eligible poor beneficiaries for the government benefits. He further added that the data will help use strategies, like the grievance mechanism, if the deserving beneficiaries are not given the benefit.
Currently, there are 140 programmes under some 22 ministries/divisions. To make an inclusive social protection programme, the National Social Security Strategy (NSSS) was approved by the cabinet sometime in 2015. The GED was entrusted to prepare the document. The core of the strategy is a lifecycle approach implying that it will cover people's needs throughout their lives, beginning from birth. But the inclusive coverage of deserving people has remained a far cry.
Admitting the deficiencies in the selection process, an official at the Ministry of Finance (MoF) told the FE that the selection will be much accurate when we start to use the national household database. "There are some programmes where proper targeting is possible like allowance for elderly people as the age is visible by checking NID," said Md Azizul Alam, additional secretary at the MoF. Mr Alam said the government wants to switch over to cash transfer programmes instead of in-kind transfer programmes like food-assisted ones as the former ensures accountability and transparency. He also said most of the leakages and corruption happen with in-kind transfers.
Mohammad Ismail, additional secretary at the Ministry of Social Welfare, said they will now make the disbursement among over 8.0 million deserving people across the country. He said his ministry has been running as many as 12 programmes. The large programmes under the ministry of social welfare are: old-age allowance, widows/deserted and destitute women and handicapped people. "We select handicapped people through a survey conducted in 2014 and this is an authentic list," Mr Ismail told the FE.
According to the survey, there are over 1.4 million of such people in the country. He said there are over 4.4 million elderly people and 1.7 million widow or divorcees who get the cash benefits. "We are going to sign an agreement with four mobile financial services, including bKash, for expanding cash transfer," Mr Ismail mentioned.
But the cash support is comparatively low ranging between Tk 500 and Tk 750 a month for the programmes the ministry is implementing. He said the number of beneficiaries under his ministry will expand to nearly 10 million in the fiscal year 2020-21.
Economists familiar with the country's social security programmes said out of 140 programmes, there are a number of schemes designed for the non-poor. They said pension meant for government retired employees, freedom fighters and some stipend programmes are of such examples.
Despite improper selection, the official statistics show that Bangladesh has ensured social protection for 56.54 million people, equivalent to 34 per cent of total population, in fiscal year 2019-20. The government eyes to expand it by more than 9.0 million people in the current fiscal year 2020-21 (FY21).
The beneficiary figures, however, include all types of government supports meant for social safety net programmes. They are cash transfer, food security and employment, stipend, credit supports, development activities and others.
Dr SM Zulfiqar Ali, a senior research fellow at the Bangladesh Institute of Development Studies (BIDS) told the FE that a significant number of poor are not getting the benefit due to the weakness of the selection process. He said on the other hand a significant number of the non-poor are getting the benefit as there are some programmes targeted them. "There are political affiliations, nepotism and corruption while selecting people for government supports that help raise exclusion and inclusion errors." He said many less-deserving people are now getting the benefit cashing in on an inefficient selection process.
He, however, said if there was an authentic national database, the selection process might improve. Dr Ali was critical of the slow progress of the NSSS. "I don't know why the cabinet-approved document is progressing so poorly" Dr Ali commented. The BIDS fellow said as per the document, the number of programmes will need to come down significantly. "All programmes, according to the NSSS, should come under one umbrella," he noted.
Dr Dipankar Roy, project director at the HIES of the BBS, told the FE that the problem of exclusion errors has been improving day by day."Although this is still high, the trend is that it is improving," he said.
CONCERN FROM DONORS: People at the donor agencies in Dhaka told the FE that reforms take time for implementation. "We cannot implement a reform overnight," said an official working at an international funding agency. He said the scope of the national social security strategy is large consisting of 140 programmes for 22 ministries/divisions. He, however, opined that the government may choose a small portion of functions, instead of all functions at a time to ensure effective implementation of the NSSS.
In the context of improving beneficiary selection for the social protection system, it may be important to note that several countries (including many developing countries) have opted for universal coverage of citizens since the trade-off between the investment requirements for universal coverage and cost of implementing targeted programme is rather small. Hence, adopting universal coverage at least for few programmes (such as child grant, disability grant and social pension) may be seriously considered in Bangladesh.
BUDGET: The government usually allocates resources for the social security programmes in its annual budget each year, which is on an average 2.5 per cent of the GDP. But it has proposed over 3.0 per cent of the GDP for the FY21 keeping in mind the adverse impacts of the Covid-19 on the poor people. It is important to improve the selection process and reduce leakages of the SP system in Bangladesh along with increasing allocations. However, FY21 budget did not provide any strategic direction in this regard.
Jasim Uddin Haroon is a special correspondent of FE.
This piece is produced under Bangladesh Policy Advocacy Initiative (PAI), conducted by Policy Research Institute of Bangladesh (PRI). To prepare the article, the writer receives valuable advises from Dr Bazlul Haque Khondker, Director, PRI.