More on why economists make bad plumbers

I've blogged earlier about why economists make bad plumbers. Another exhibit is this interview of Abhijit Banerjee and Esther Duflo. Banerjee says,

I think India is a good example of [a country] where they literally had not thought through their own plumbing. If you think of what happened to the urban migrants, India’s welfare system is actually completely designed on the assumption that people live in their stable families which live in one place for year after year. In your village, you’re entitled to apply for the public distribution, which is essentially nearly free food . . . and in rural areas there is the rural employment guarantee system. Both of those are designed for rural citizens who live in their own village. You’re not entitled to go to any village and say: ‘I want my employment guarantee.’ There might be as many as 50m of these low-income migrants who temporarily live in cities. They can’t connect to the welfare system. That’s why there were pictures in the first lockdown of people walking 1,000 kilometres . . . there was no way for them to survive. They just had to go home. That is pure plumbing failure.

This is pure rhetoric. It's the classic hatchet job - form your hypothesis (a system where migrant workers can access food and other welfare benefits), set up a straw man (the public distribution system, PDS, or any welfare benefit), demonstrate how the straw man fails the hypothesis test (the example of covid induced migration), and blame the system (the government "did not think through their own plumbing" on its programs). Before passing such sweeping judgement on something like the PDS or NREGS, it's useful to understand its original purpose and its trajectory of evolution. It's also classic hindsight-based judgement. 

The PDS, in existence since independence, has the objective of providing food security by distributing foodgrains to the poor living in villages and slums at subsidised rates. Despite its undoubted deficiencies, given its scale and the pervasive state capacity weakness, it's been a remarkable success and is arguably a practical model for food security in any developing country. As I've blogged earlier, along with NREGS, it's been one of the two policy instruments that have ensured India avoided starvation during the pandemic. It has undergone constant improvements over the years, with its effectiveness in terms of inclusion and exclusion errors and service levels being very good in some states. 

But the additional requirement of migrant coverage and describing its absence as a plumbing failure is unfair. Consider the critical requirements for a portable PDS - a portable identity (which allows credible identification of the beneficiary), a dynamic state-wise PDS allocation mechanism (which allows for shifting of PDS allocations across states based on migration trends), a dynamic supply chain (which allows PDS shops to adjust stocks for periodic changes in migration trends), and a country-wide real-time stock management system (that connects all PDS shops). I have overlooked several other factors. The same analysis could be done about employment guarantee schemes. 

Pause and reflect on the enormity of the challenge. Given the context of rural and urban India, a robust and dynamic logistics management of this nature would have been unthinkable till very recently. Now, this is real thinking through the plumbing. How many economists can claim this expertise?

No amount of thinking through the plumbing would have helped overcome these challenges. Each of the four requirements would have been daunting impossible. With advances in digital technologies and Aadhaar, it may now be possible to start pursuing this objective. Recognising this, and even before the migrant crisis, the Government of India had launched the One-Nation-One-Ration Card scheme to make ration cards portable. Portability was even a incentivised as a condition for state governments to access the additional borrowing permitted during covid.

In simple terms, castigating PDS with plumbing failure for non-portability and not covering urban migrants is like judging the batting strike rate of Sunil Gavaskar in comparison to that of the T20-era batsmen. The PDS was designed for a purpose and, on a purely relative terms and in the world of second-bests, it's done a reasonably good job over decades in ensuring food security and eliminating famines from India. Like anything, it has to evolve with time to accommodate the emerging challenge of migrant coverage and more. 

The issue of providing welfare to the 50 million or so migrants in cities with an acceptable enough level of leakages is perhaps one of the hardest targeting challenges in development. Migrants are a floating population, almost exclusively engaged in informal and outside the market activities, with little incentive and a lot of disincentive to being tracked. Therefore, despite the awareness of the problem, tracking them will remain a daunting challenge. In the circumstances, it's the perfect example of armchair analysis to nonchalantly call for transferring cash and other benefits to poor urban migrants. 

It's an altogether different matter that the government (and the vast majority of mainstream opinion makers and researchers) have seriously underestimated the sheer scale of internal migration. It was surely a governance failure that the system was unprepared to deal with the pandemic induced reverse migration. That's not a plumbing failure with PDS or with any specific prevailing government program. But it also needs to be acknowledged that even with the best technologies, targeting migrants with PDS and other welfare services in a reasonably efficient manner will remain elusive for some more time. 

As I have written earlier, instead armchair analysis of areas on which their understanding is limited and grandstanding to reinforce their pet ideas (demonstrating government plumbing failures in development), economists should focus on their research and let the evidence generated inform public debates. Plumbing should be left to plumbers.