Interesting piece, and one I've been mulling over myself for some time now.
I think the article does two things well: correctly identifying this as a problem of incentives, and proposing constrained maximisation as the right framework response. I agree on both counts. Having spent many years working with both corporate India and donors/philanthropists, I'd add that a direct transplantation of EA concepts will not work here, and the CSR context makes this especially clear.
Most CSR funding decisions are driven by three factors: (i) alignment with national and government priorities; (ii) a license to operate locally, particularly for manufacturing companies; and (iii) the personal interests and convictions of founders and promoters. A significant portion of CSR budgets will always be allocated along these lines: that's simply the institutional reality. But given the rate at which CSR budgets are growing (broadly tracking India's nominal GDP growth, and expected to triple by 2035), even redirecting incremental spending toward more impactful charities could make a material difference. I would go deeper into how one can make impactful CSR genuinely aspirational, which means solving the incentives problem from the status angle, this seems like a promising avenue worth developing further.
I think your intermediary argument is overstated. There are 27,000+ companies contributing to India's CSR pool, and only a small fraction of them work with intermediaries at all. More importantly, CSR committees at the major spenders — Tata, Reliance, HDFC and their peers — are composed of C-suite executives and board directors. These individuals rarely interface with intermediaries directly; their decisions are shaped by the structural factors I outlined above, and only to a small degree by what Samhita or Sattva recommend. A steel company, for instance, will almost invariably direct a significant share of its CSR budget to communities near its plants, not because an intermediary guided them there, but because that's what social legitimacy requires. The same logic applies to mining. The gatekeeper framing, I think, applies to a narrower segment of the market than the article implies. Also, targeting the intermediaries adds one more layer of work: the intermediaries will still have to change behavior at companies.
On the GiveWell-for-India idea: I think this could be valuable, not just for CSR, but for Indian donors more broadly. A resource that maps cause areas and geographies by neglectedness and marginal impact potential would fill a real gap. The methodology isn't fully clear to me from this draft, but the directional idea is sound, and I'd be interested to see how you develop it.
Interesting piece, and one I've been mulling over myself for some time now.
I think the article does two things well: correctly identifying this as a problem of incentives, and proposing constrained maximisation as the right framework response. I agree on both counts. Having spent many years working with both corporate India and donors/philanthropists, I'd add that a direct transplantation of EA concepts will not work here, and the CSR context makes this especially clear.
Most CSR funding decisions are driven by three factors: (i) alignment with national and government priorities; (ii) a license to operate locally, particularly for manufacturing companies; and (iii) the personal interests and convictions of founders and promoters. A significant portion of CSR budgets will always be allocated along these lines: that's simply the institutional reality. But given the rate at which CSR budgets are growing (broadly tracking India's nominal GDP growth, and expected to triple by 2035), even redirecting incremental spending toward more impactful charities could make a material difference. I would go deeper into how one can make impactful CSR genuinely aspirational, which means solving the incentives problem from the status angle, this seems like a promising avenue worth developing further.
I think your intermediary argument is overstated. There are 27,000+ companies contributing to India's CSR pool, and only a small fraction of them work with intermediaries at all. More importantly, CSR committees at the major spenders — Tata, Reliance, HDFC and their peers — are composed of C-suite executives and board directors. These individuals rarely interface with intermediaries directly; their decisions are shaped by the structural factors I outlined above, and only to a small degree by what Samhita or Sattva recommend. A steel company, for instance, will almost invariably direct a significant share of its CSR budget to communities near its plants, not because an intermediary guided them there, but because that's what social legitimacy requires. The same logic applies to mining. The gatekeeper framing, I think, applies to a narrower segment of the market than the article implies. Also, targeting the intermediaries adds one more layer of work: the intermediaries will still have to change behavior at companies.
On the GiveWell-for-India idea: I think this could be valuable, not just for CSR, but for Indian donors more broadly. A resource that maps cause areas and geographies by neglectedness and marginal impact potential would fill a real gap. The methodology isn't fully clear to me from this draft, but the directional idea is sound, and I'd be interested to see how you develop it.