Relevance: GS Paper 2: Poverty and Hunger; GS Paper 4: Civil Service Values
Anyone familiar with the Indian welfare state would attest that, aside from a few activist bureaucrats, most people see it as a corrupt cesspool forced on them by politicians motivated by electoral incentives. The consensus is that the welfare state should be “reformed” and restricted to “target beneficiaries”.
Officers schooled in the sciences, notably engineering, now dominate the IAS. The Trivedi Centre for Political Data’s data on the Indian bureaucracy confirms this. In 2020, 80% of new IAS recruits were STEM graduates, compared to 18% in 1980.
As a result, the socioeconomic foundation from which IAS candidates are recruited has become more democratic and representative of Indian society. Anirudh Krishna, a former IAS officer and public policy professor, found that 85% of the 2009 IAS hailed from small towns and villages. 24 percent attended public schools, while 23 percent attended rural schools.
Do bureaucrats view welfare in the same way? A closer look revealed an intriguing theory.
First, to paraphrase a former IAS officer, STEM training promotes a mental paradigm that understands the world through algorithms. It simplifies real-world problems into technical (typically technological) fixes.
Real-world poverty is shaped by several layers of distinct and often unpredictable interactions of policy with individuals and institutions, as well as the underlying power dynamics that regulate these interactions.
As a technical cure, welfare properly identifies all that is broken – corruption, ineffective and apathetic bureaucrats, and a greedy elite seeking welfare loot. But the fixes don’t solve the complexity.
Technological advances reduce welfare failures to technical faults, to be remedied with data platforms, GPS tracking, and command and control. The fact that this mental model fosters a worldview that sees the welfare state as corrupt and inefficient, locked in distortionary networks that feed political cycles, is a major issue.
This view legitimises a strong dissatisfaction with welfare possibilities. That’s why the bureaucracy is always anxious about “identifying” the impoverished. Thus, direct benefit transfers (DBTs) have a higher priority than investments in public goods and safety nets like food subsidies and the Mahatma Gandhi National Rural Employment Guarantee Scheme.
To work, DBTs must be built on data platforms that allow citizens to be “verified” and “authenticated” as “beneficiaries”. The state’s welfare function is thus reduced to a palliative given only to those who can be “confirmed.” This ideological paradigm caused the bureaucracy to aggressively obstruct cash transfers to migrant workers during the lockdown crisis. How can we offer money when workers can’t be identified? Creating a new database became inevitable.
The second theory concerns the IAS’s changing socioeconomic basis. Today’s IAS reflects India’s societal realities. But does overcoming adversity affect ideas of what it takes to “get out of poverty”?
This shapes welfare perceptions, too. After all, the IAS exam is meant to find the best. Does success reward a belief system that values human capacity and entrepreneurship over structural inequality, which underpins the welfare state’s rationale? Is this why “growth” and “welfare” are framed as trade-offs?
These ideas are not conclusions, but rather suggestions for understanding why the welfare state frequently fails. To paraphrase political scientist James Wilson, bureaucracy is not a black box. Rather, bureaucratic behaviour is moulded by individuals’ belief systems, attitudes, and professional conventions.
But we rarely discuss this in our debates on civil service and welfare. Incentives for performance and technical skill sets are frequently discussed in reform debates. But no amount of reform will work if those in charge of delivering it do not believe in it.
The Indian State desperately requires new rules, values, and collective purpose frameworks. To reform the welfare state, officials must first believe in it. Then it can find the suitable algorithm.