At a conference organized by Pakistan Poverty Alleviation Fund (PPAF) last week, Geof Woods, professor emeritus at University of Bath, made a comment that he is becoming “anti-targeting”, referring to the targeting of public services such as education, health etcetera, and to the targeted cash-transfer programmes in the absence of adequate public services by the state.
Geoff’s passing comment against “targeting” didn’t cause much debate or discussion at the PPAF moot. Perhaps if Marvi Memon from BISP was there, or other relevant federal/provincial government officials who were conspicuous by their absence, a discussion on targeting may have followed. Geoff’s anti-targeting ideas are probably influenced from his long experience in India where states have been realizing the futility of targeting as the panacea principle of social policy, and instead adopting universal education or healthcare programmes.
The idea of targeting, as Amatya Sen wrote in his book titled An Uncertain Glory, is “deceptively simple”. This idea revolves around focusing public resources on the poor, which “sounds like common sense way of ensuring that limited resources are well used from the point of view of poverty reduction. In practice, however, there are some serious problems with a targeting based system of social support.”
Some of those problems include the room for corruption, and inclusion/exclusion errors. However, the mechanism for targeted social safety nets has evolved considerably since 2013, when Sen published that book. Using bio-metrics, the mechanism in practice these days reduce the room for corruption. In the case of BISP here at home, the inclusion/exclusion errors have also been reduced by way becoming more methodologically robust. (See Brief Recording “BISP to transition towards dynamic registry and biometric payments: Marvi Memon, Chairperson BISP” Published Jan 18, 2017).
Be that as it may, the whole process of targeting is disjointed at the roots: the question of who is poor. Pakistan’s Planning Commission has been focusing on multi-dimensional poverty at the one end, but in so far as the money metric of poverty is concerned it uses Food-Energy-Intake method to compute headcount poverty and has been thinking to move to Cost of Basic Needs method. BISP on the other hand uses its own proxy means test. This inconsistency in number reeks of policy incoherence.
Second, pro-poor programmes and social safety nets abound in Pakistan. There is the BISP; there are microfinance institutions, the PPAF, the NRSP, Bait-ul-Mal, Zakat, Usher, EOBI and what not. Among all these BISP is the biggest. But there is no way to be sure how many of the beneficiaries are duplicated across these agencies/poverty programmes.
Third, a key factor ignored by targeting approach is the transitory nature of poverty. Recognizing this limitation, Dr Durre Nayab of PIDE, had advised BISP to take into account not just the poverty status of a household but also its dynamics vis-à-vis poverty since “a household above the poverty line could move below it and vice versa in the face of changing circumstances.”
“A recipient household could become ineligible due to poverty dynamics while an ineligible household could become eligible. Such changes need to be taken into account by the BISP design for the more rational and equitable distribution of cash assistance,” she co-wrote in her 2012 PIDE paper titled Effectiveness of Cash Transfer Programmes for Household Welfare in Pakistan: The Case of the Benazir Income Support Programme.
In cognizance of this BISP, is trying to make its National Socioeconomic Registry (NSER) more dynamic than static. However, keeping the costs in mind, the NSER is unlikely to be updated every year. Even as international best practices such datasets are updated within a 3-5 years period to ensure the validity of data, which implies even if Pakistan was being super efficient, it might be able to update its NSER every three years.
However, three years is a long time in poverty, as the poor and the non-poor who are bordering poverty are quite vulnerable to various types of shocks and hazards, be it natural/agricultural, economic/social, or health.
BR Research has been making inquires in research community to find out the intensity, frequency of shocks/hazards and the percentage of poor affected by it. It turns out the matter is little researched. A host of research by PIDE economists, such as G.M. Arif and Rashida Haq, suggests that adverse shock is rather pervasive. For example, one 2015 study by Haq observed that “one-third of rural households experience an adverse shock, be it natural/agricultural, economic, social or relating to health.”
It is difficult to get the complete profile of the intensity, frequency of shocks/hazards and the percentage of poor or non-poor-bordering-poverty affected by it. That’s because this kind of research requires a series of panel data survey; the more frequent that survey is conducted, the better idea we would have about the frequency of various adverse shocks/hazards that push the poor into an ultra-poor and the non-poor into a poor.
At present, such panel data surveys are conducted every five years, which only informs us about the frequency of shocks/hazards every five years. But that is a long time in poverty, and just because there is no data to support a higher frequency of shocks/hazards, it doesn’t necessary mean that these adverse shocks are not frequent.
Imagine a household, whose eldest child is going to grade-5 in a school. Imagine now that the bread earner of the family suffers from a health shock pushing the household in poverty and pushing the child out of school. In 3-5 years, some dataset figures point out that this household needs to be given a cash transfer or some kind magic bullet under education-targeting programme, the child’s life trajectory has already changed for the worse.
Targeting may well be serving its intended purpose, but the processes and mechanism behind targeting ensure that the purpose is served too little too late, at least in so far as households affected by shocks/hazards are concerned, which are quite in number. Targeting should not be made as the panacea principle of social policy, and must not come at the cost of universal public services by the state.