Karthik has provoked me sufficiently with his comment on my last post on NRHM and reference to this older post by Jeff Hammer and Jishnu Das. And the article is a teachable exhibit.
First, let's get some facts out of the way. The Das-Hammer observations are till the 2008 period and this is for the 2005-06 to 2015-16 period. The IMR outcomes do not seem to have tapered.
Be that as it may, even if this were not true, I still have several problems with the blogpost. I think it takes theory and the quest for evidence to some extremes. Just a few that pop out egregiously are as follows
1. C-sections get done in the bigger institutions, community health centres (CHCs) and above, rarely in Primary Health Centres (PHCs). While they are undoubtedly a big concern in general, they cannot be the reason for any relative stagnation in the decline in mortality rates due to deliveries in PHCs.
2. I agree with the logic of the marginal cost/returns and that is a part of the larger state capacity issue itself - state is trying to address a vastly increased scope of its activities with pretty much the same set of institutional, human, and financial resources.
But I am not sure whether the marginal cost curve has reached the diminishing returns part in case of PHCs and institutional deliveries. My guess would have to be that the average monthly institutional deliveries in PHCs would be 15-20 (and very few would have, say, more than 75). In any case, what is the authors' basis for making their own assumption about the marginal cost argument? It appears to be only a hypothesis? It would have been useful to engage with this hypothesis if it was backed by more evidence of the like I just mentioned.
3. Surely the standards on promotion of institutional delivery and drugs validation process cannot be equated! Do governments need to wait for evidence before promoting institutional deliveries? Or should governments wait to fix state capacity before they promote institutional deliveries? Or should governments stand aside and ask local communities to iterate on institutional deliveries?
4. Again the author's argument about respecting the decisions of the poor is as much as their thought exercise as the practitioner's explanation of their theory of change. Does it mean that we now question the narrative about people not delivering in hospitals because of ignorance, customs, social norms, religious beliefs, and so on? Do we need evidence on that? In remote areas, and these are very large swathes, transportation is a big constraint on institutional delivery. Do we need evidence to show that this is indeed a constraint?
In all such examples, there are no one or two "starting points" to explain trends and behaviours. There are many, and it varies from region to region. And trying to figure that out on a context basis is obviously ideal, but of no practical relevance in our Bayesian world.
5. Some other arguments are plain baffling (italics mine),
Government encouragement of institutional deliveries is based on the idea that poor people choose to deliver at home either out of ignorance or an inability to make the right decisions or due to cultural norms and the exercise of (male) power. But an alternate starting point is that people were not using institutions to begin with precisely because quality was low, and that increasing quality would also bring more people in. In fact, this is the most obvious explanation for the correlation between increasing institutional deliveries and lower child mortality. To base policy on the belief that we can make better decisions over the lives of those who are about to be born than their parents is a stand that minimally requires the onus of proof to be on those who claim such knowledge.
How can we say with any degree of certainty that which of the two, supply or demand sides, are the bigger constraint on institutional delivery? Further, the answer could vary widely across regions, rural-urban, and so on. In fact, there is most certainly no one constraining factor across the general population. Frankly, not only do we not have answers to many of these issues, nor are we likely to ever have satisfactory enough answers to them.
Right now, we know that given the context in these developing countries, it is a good thing to encourage institutional deliveries. But we need to be simultaneously cognisant of the systemic weaknesses and try to address them. The NRHM has certain institutional features which tried to make some (obviously limited) attempts at this. As a counterfactual, Sarva Siksha Abhiyan allocates nothing to the school to improve learning outcomes (apart from a meagre annual grant to the teacher to make teaching materials).
Look, I hold no brief for India's health care system, which I will emphatically say is broken and needs fixing big time. But try coming up with a politically acceptable and administratively feasible action agenda (a program to address maternal and child health issues, leave aside a policy for healthcare), conditional on the world as it exists, for scaled up implementation across India!
Let us dispense criticism with some realism. I can understand some academics critiquing other academics for the latter's unquestioning embrace of some models like the NRHM or Chiranjeevi voucher scheme. In fact, this blog has consistently questioned the obviously inflated claims of many government programs. But I cannot understand academic critiques of governments for focusing on institutional deliveries.
If critiques are to be constructive, it would have to focus on more practical issues. Conditional on the focus on institutional deliveries, what additional resources are required to support the PHCs? What is a reasonable number of deliveries an institution can perform and is there some mechanism to support institutions where they exceed this number?
May be they are not amenable to academic research publications. Instead, if we disown any priors and open up all boundaries, we are left with idle critiques of no relevance to practitioners. Of course, it is fair to say why should academics be held to that test. Well, that is a different matter and for a latter post.