This episode is the ninth in a miniseries of weekly short episodes featuring young scholars entering the academic job market who discuss their latest research. In this episode, Shruti talks with Kim Fe Cramer about her job market paper, “Bank Presence and Health.” They discuss how access to banking and credit benefits health, why those benefits happen, whether healthcare should be subsidized and much more. Cramer is a Ph.D. candidate in finance and economics at Columbia Business School. She researches how the financial sector affects development outcomes and has conducted experiments in countries including India, Kenya and Ecuador.
SHRUTI RAJAGOPALAN: Welcome to Ideas of India, a podcast where we examine academic ideas that can propel India forward. My name is Shruti Rajagopalan, and this is the 2021 job market series where I speak with young scholars entering the academic job market about their latest research on India. I spoke with Kim Fe Cramer, a Ph.D. candidate in finance and economics at Columbia Business School. She has a B.Sc. in general management from European Business School.
We talked about her job market paper titled “Bank Presence and Health,” where we explored various mechanisms through which bank presence and access to bank credit, for both businesses and households, impact health outcomes in India.
For a full transcript of this conversation, including helpful links of all the references mentioned, click the link in the show notes or visit mercatus.org/podcasts.
Hi, Kim. Thank you so much for coming on the show. It’s a pleasure to have you here.
KIM FE CRAMER: Hi, Shruti. Thank you very much for having me. I’m honored to tell you a little bit about my research today.
Does the Presence of Banks Affect Health?
RAJAGOPALAN: Your research sounds very exciting because it talks about two things that we normally don’t put together in the same breath. You are working on financial sector penetration. In your paper, you’re looking at this 2005 RBI [Reserve Bank of India] policy in India, which is trying to incentivize banks to open branches in traditionally underbanked districts.
What you show is quite interesting. It’s that financial sector penetration, or rather, increase in access to banking and credit, actually improves health outcomes in these districts which were previously underbanked, and now they’ve improved in financial sector penetration. Can you tell us a little bit more about this paper and the mechanism at play?
CRAMER: Imagine that you live in a village or a small city in a developing country. There’s no bank in your area. That means there is little credit access for businesses, and you yourself only have a low-income job. Now somebody in your family falls ill with fever. You’re concerned because you don’t have enough money in your pockets from your job to pay for a visit to the doctor.
Additionally, you cannot rely on your savings account, an emergency bank loan or a health insurance payout since these services are simply not available in your area. You also know that even if you could afford a visit to the doctor, healthcare supply is often scarce and of low quality. Ultimately, your family remains in poor health.
Now imagine that a bank enters your area. First, the bank provides credit to the local businesses, which flourish, and you can benefit from higher income. Additionally, you can go to the bank branch and ask for traditional products such as a savings account and a bank loan. Very importantly, in the context of many developing countries, banks also directly provide health insurance products at the local branch.
This is the case because banks act as intermediaries between health insurance providers—big companies in faraway cities—and the local households, to decrease, for example, distance between the both or information asymmetries between the both. Either due to higher income or due to increased financial access, you can now, potentially, increase your healthcare demand.
Finally, banks potentially provide credit access for healthcare providers, which allows them to improve supply of healthcare services. All these potential mechanisms could mean that your family recovers quickly and you remain in good health. This small narrative provides strong motivations that banks can play a vital role in improving health.
Finding a Causal Connection
CRAMER: Despite these strong motivations, we actually have no causal empirical evidence on the relationship between bank presence and health. As you mentioned, what I’m going to show in my job market paper—I’m going to use a nationwide natural experiment that allows me to identify the causal effect of bank presence on health.
I’m using a policy of the Reserve Bank of India, the RBI, that was introduced in 2005 and is still in place until today. This policy basically has the objective to increase bank presence in underserved locations. In order to achieve this objective, the RBI relies on its authority to issue licenses for new bank branches in India.
Now, the policy states that banks can increase their chance to obtain a license for a favorite location, for example, Mumbai, if they strengthen their presence in so-called underbanked districts. An underbanked district is what allows for causal identification in my paper.
An underbanked district is defined as a district that has a very high population but a low presence of banks—that has a population-over-bank-branch ratio that is larger than the national average. At the same time, districts that have a ratio of population over bank branch that is lower than the national average are called banked districts. We have underbanked districts just above the national average and banked districts just below the national average.
Actually, the design of the policy allows me to apply a regression discontinuity analysis in which I compare districts that are just underbanked, where banks are incentivized to open extra branches, to districts that are banked. In these districts, I observe health outcomes a couple of years after the policy. I use multiple household-level surveys 6 years after the policy was introduced and 10 years after the policy was introduced.
First, I show that these underbanked districts indeed benefit from increased bank presence. These banks really set up more branches in these underbanked districts. Then, as a next step, I show that household health improves. I find that households are a third less likely to receive an illness such as fever, diarrhea or cough a couple of years after the policy was introduced. Their health status significantly improves. It also improves along other dimensions such as increased vaccination rates or decreased risks associated with pregnancies.
How Are Banks Improving Health?
RAJAGOPALAN: I want to understand that there are many channels through which financial sector penetration improves health outcomes, but can you maybe give us an insight on which are the dominant factors or which are the more prominent or salient channels? What I mean by that is, is the bulk of the work in your story being done by the economic growth that is made possible through financial sector penetration?
People get richer, so presumably they have access to clean drinking water and slightly better resources at the household level. So they get sick less, and the children get sick less, and you see all these improved health outcomes. That’s one possibility.
A second possibility is now, because households themselves have greater access to banking and therefore savings accounts, they’re able to smooth out their consumption. They can actually borrow and rely on credit for any health emergencies that they might have or to make sure that they have health consumption exactly when they need it as opposed to putting it off, which also improves health outcomes.
Is it because there is greater access to credit on the supply side, which is directly for the hospitals in a particular given area? Or is it through greater access on the supply side because overall economic growth is increased, and therefore the demand for hospitals is naturally increased as, overall, the economic growth increases? Or is it the fifth mechanism, which is, with increased access to credit and economic growth, the dual impact of that is going to lead more households to buy health insurance?
You can imagine all these different channels being the bridge between the financial sector access and penetration and the health outcomes. Which of it is it? If it’s all of the above, what’s the story there? And what’s the contribution of each of these mechanisms?
CRAMER: Thank you very much for your question. Indeed, you’re right. There are many potential mechanisms that could drive this relationship between bank presence and health. Maybe the first one we are thinking of is that banks stimulate business activity and households become a little bit richer. Now they can invest in health. I indeed find some evidence that households become richer in my data. That’s one potential mechanism.
On the other hand, we have studies that provide large cash transfers to households, very large cash transfers. These are actually studies in Kenya. They show, even if you give these huge cash transfers to households, there are no improvements in health in the short or the long term.
Gaining Access to Financial Products
CRAMER: These other studies suggest that simply providing an increase in income, an increase in cash available, is potentially not sufficient to improve health. Based on that, I’m turning toward other potential mechanisms that could be at play. The second one that you were mentioning was that households gain access to financial products: savings accounts, credit products, and health insurance.
Let’s first talk about savings accounts. I do find that households are significantly more likely to own savings accounts after the policy, and this, as you said, could potentially help them to smooth their consumption. However, there are other studies, big RCTs—randomized control trials—that show that simply only providing savings accounts are not likely to drive major welfare impacts. Yes, we do see an increase in savings accounts, but I don’t think this is the whole story here.
Credit—I actually don’t even see that households take up significantly more credit. It’s often that some households take it up, but on average, not a lot of households seem to take it up.
What’s really, really interesting here is the financial instrument of health insurance. Health insurance in this context is often directly provided by banks. That’s something that might seem a little bit surprising to many of us. If you’re maybe in the context of the U.S., you would think that banks traditionally provide savings accounts or credit products, but health insurance is mostly provided by insurance companies.
In this other context, why do banks provide health insurance? To understand this term, let me tell a little story. Imagine you’re a health insurance provider, and you’re in a relatively large city in India. Now you want to gain access to potential customers, but you have two problems: They are far away and you know very little about them.
What are you going to do? You’re going to partner with banks who are already present in these very far distant locations, and these banks help you to (a) gain access to these potential customers, and (b) also collect some information about these potential customers. For example, the banks have already verified the identity of these potential customers.
That’s the channel why we see that banks actually directly provide access to health insurance to households. I think that’s a very interesting story, and that’s something that we haven’t seen much in the literature about yet—that banks actually directly provide health insurance. That could play a big role in improving the health outcomes that we observe.
Then there is a third story, which I think is also very interesting, and it hasn’t been investigated much in the literature yet. Banks provide credit to healthcare providers. Again, maybe if you’re in the context of the U.S., you think of a healthcare provider of a hospital as these big institutions with many, many employees. But actually, in the reality on the ground, these are often relatively small establishments.
What we understand as a hospital—on average, these establishments only have six employees. They’re relatively small. There’s maybe one or two doctors, and there’s limited equipment, for example, to diagnose or to treat.
These kinds of establishments—they don’t have unlimited access to resources. Imagine, for example, that a healthcare provider wants to make an important investment. They want to, maybe, buy some equipment that would help them diagnose. Now, this equipment is relatively expensive. Where banks come in—they provide credit to these healthcare providers, which now allow these healthcare providers to expand and improve their supply of services, which ultimately benefits households.
Normative Policy Implications
RAJAGOPALAN: Here, it’s really interesting because you’re right. India is unique in that the health insurance coverage is also being provided through the banks, which normally only do financial intermediation.
Now, what are some of the normative implications of this research? Is it that developing countries like India need to focus on greater health insurance coverage? With or without the banking sector, the government should announce big programs, or it should subsidize these big programs because the heavy lifting in the story is being done by access to health insurance—which is also, of course, increasing the supply of healthcare. More people have health insurance. More small hospitals and clinics and dispensaries start opening up.
Or is it that there’s a very robust demand for health insurance? It is just a question of access and identification, and you need something like a bank, which is the financial intermediary, to make that happen. You don’t necessarily need large-scale government programs. What’s your hunch? Which is the policy direction that developing countries should take, given your findings?
CRAMER: I think that the area of health insurance is extremely interesting and very promising in this context, and that we’re going to see a lot more research on this in the future. Actually, I have another project in Kenya on health insurance. What I show in the context of my paper is that with bank presence, with increased access to health insurance, health improves for households. But there happens a second thing. At the same time that health insurance increases, there’s an increase to the credit access to healthcare supply.
It’s basically going both through the demand—like bank presence allows to increase demand—and increase supply at the same time. This could be a potentially crucial part of the puzzle—that we need to increase both. Again, when we see, for example, only demand increasing because we just give cash to people, we don’t see big improvements in health. But in the context of this study, when we increased both demand and supply, there seems to be a positive and strong and robust effect on health.
Now, your question was, if we, say, expand bigger government programs and regular health insurance programs, how do they benefit households and potentially have the ability to also similarly improve health outcomes? That’s a question I think we need more research on, but it might be the case that simply providing access to health insurance alone is not sufficient because the supply side cannot react.
For example, imagine there’s an increase in demand in health insurance now, a gradual increase in healthcare demand. So the healthcare provider on the ground says, “Oh, more and more people want to come in. They can now afford my services. They want better diagnostics. At the same time, I cannot really provide these services to them because I do not have access to invest in proper equipment.” If the supply remains really low, a gradual increase in demand might not be the thing that we need to really improve health. Maybe it needs to happen on both dimensions.
Kenya and Subsidizing Health Insurance
RAJAGOPALAN: Absolutely. What are some of your findings in Kenya? I’m just curious because, as an interesting comparison to the work that you’ve done in India, can you tell us a little bit about what’s happening in Kenya?
CRAMER: In Kenya, we’re using randomized control trials, so RCTs, to investigate the question whether health insurance should be subsidized. In my job market paper, I show that with increased access to banks, health insurance take-up increases. This basically removes this first barrier of, let’s say, distance or access to health insurance.
But there are many other barriers, and one a surprise. We see that many households say that they simply cannot afford the health insurance. It’s too expensive. One question that arises is whether governments or other players should subsidize health insurance such that households can afford them. This question depends on two considerations.
First, it depends on how much do households benefit from health insurance. We’re going to do an impact evaluation. One important outcome, of course, is going to be health. The second consideration is, what are the cost implications of subsidies for health insurance providers? Because we do not only care about the customers; it also needs to be a sustainable system for the healthcare providers. In particular, if you subsidize health insurance, that may mean that other types of potential customers select into the health insurance, or that they adjust their behavior based on the price that they paid.
We partnered with an innovative health insurance company in Kenya to both see what are the benefits of health insurance and what are the potential cost implications of subsidizing health insurance for health insurance providers.
RAJAGOPALAN: Is it simply a question of demand, which is, of course, completely backed up by price and the subsidy that you offer? Or is it also a supply-side story? I think that’s a very interesting area of research. In India, particularly where private sector provides a bulk of healthcare infrastructure that people rely on, most of the hospital services are actually provided by small, private-sector healthcare clinics and dispensaries. This is especially true in terms of employment. Most of the doctors and nurses work in private-sector hospitals in India. I think that is the part of the story which is critical going forward, as you show.
CRAMER: I think you’re right, the supply side is very interesting, and that’s something I want to focus on in the future as well. Potentially, we can really identify the causal effect of increased access to credit to healthcare providers on health status of households. That’s something we almost have no research about.
Interestingly, policymakers seem to be aware of that connection between banks and healthcare providers. In light of the recent COVID-19 crisis, the Reserve Bank of India issued a new policy that basically is saying that banks should provide credit to healthcare providers. There’s basically increased liquidity for healthcare providers up to $7 billion. It’s a very interesting area for both academics and policymakers.
RAJAGOPALAN: Yes. I don’t know how you would think about doing this, but the COVID wave in India hit different districts at different points in time. You see this massive increase in demand for healthcare services, right? The healthcare supply is not elastic. It doesn’t adjust quite as quickly as demand in the middle of a pandemic.
There, I think it would be quite interesting to see what bank presence has done in terms of an increase in demand for these services. I think that could be an interesting area to look at, specific to the pandemic, because it’s one of those rare moments when you have this surging demand without having an accompanying surge in supply because those processes are just slower.
CRAMER: Absolutely, yes.
RAJAGOPALAN: Speaking about the pandemic, I know you live in New York City and you’re at Columbia. I used to live in that neighborhood, by the way, on 117th Street on the other side of Morningside Park. What have you been up to during the pandemic?
CRAMER: Actually, I moved back to Germany in the pandemic. [laughs] I had the pleasure to spend a lot of time with my family, and actually, I spent a lot of time with my dog.
RAJAGOPALAN: Oh, that’s wonderful.
RAJAGOPALAN: What kind of dog do you have?
CRAMER: It’s a Labrador retriever.
CRAMER: Oh yes? Very cute. [laughs]
RAJAGOPALAN: I have two of them.
CRAMER: Mine’s a little bit orange. Yes, we were taking very long walks, and it was really a pleasure to just spend more time with my family and my dog. He loved the attention.
RAJAGOPALAN: One last question before I let you go: What have you been binge-watching?
CRAMER: What have I been binge-watching? Okay, I was actually binge-watching the show called “Money Heist.”
CRAMER: Yes. It’s about a robbery of a central bank, I think in Spain. That’s very fun.
RAJAGOPALAN: Oh, that’s fantastic. Where are you in Germany right now? Just out of curiosity.
CRAMER: I was located mostly in Berlin, but in the suburban area, so it is very rural. We have some wild pigs, a lot of forest.
RAJAGOPALAN: That sounds great. I spent a semester in Hamburg when I was an Erasmus student. I have spent a lot of time in Germany, and I have lots of friends in the Hamburg area, and I keep going back.
Thank you so much, Kim. It was a pleasure to have you on the show. Thank you for making time to do this, and I look forward to reading more of your research.
CRAMER: Thank you very much, Shruti.