Poor Economics

by Abhijit Banerjee

  But her financial portfolio did not end with the six ROSCAs. She had taken a loan from one of her ROSCA savings pools in early May 2009 (a little over two months before we met her) to buy maize worth 6,000 KES ($105 USD PPP). She was also a member of the village savings bank, where she had a savings account, though it was currently almost empty. She had used that money to buy shares in the village bank worth 12,000 KES ($210 USD PPP). Along with some shares she already had (each share entitled the borrower to borrow up to 4 KES from the village bank), this allowed her to borrow 70,000 KES ($1,222 USD PPP) and build herself a house. She also had little stashes of money hidden in various parts of her house to deal with small emergencies such as health needs, although as she pointed out, sometimes the health money was used for feeding visitors. Finally, she was owed money by a variety of people, including 1,200 KES by her clients and 4,000 KES by a former member of her joint liability group in the village savings bank. He had defaulted on the loan when he still owed the bank 60,000 KES ($1,050 USD PPP), obliging the group members to cover for him, and he was only now slowly paying them back.

  As a market vendor married to a farmer, Jennifer Auma probably lived on much less than $2 a day. Yet she had an array of finely tuned financial instruments. We see this kind of financial ingenuity time and time again.

  Yet all the ingenuity the poor employ to save may simply be a symptom of the fact that they don’t have access to the more conventional and simpler alternatives. Banks don’t like managing small accounts, largely because of the administrative costs of running them. Deposit-taking institutions are heavily regulated, for good reason—the government is worried about fly-by-night operators running away with people’s savings—but this means that managing each account requires bank employees to fill out some amount of paperwork, which can quickly become too burdensome, relative to any money that the bank can hope to make from these tiny accounts. Jennifer Auma explained to us that her savings account at the village savings bank was not a good place to save small amounts, because the withdrawal fees were too high. The fees were 30 KES for withdrawals less than 500 KES, 50 KES for withdrawals between 500 KES and 1,000 KES, and 100 KES for a larger withdrawal. As a result of such administrative fees, most of the poor may not want a bank account even when they are entitled to one.

  The fact that the poor have to substitute for lack of access to proper bank accounts by adopting complicated and costly alternative strategies to save might also mean that they save less than they would if they had a bank account. To find out whether this was the case, Pascaline Dupas and Jonathan Robinson paid the opening fees for a savings account at a local village bank, on behalf of a random sample of small business owners (bicycle taxi drivers, market vendors, carpenters, and the like) in Bumala. The bank had an office in the main marketplace where all these people operated their businesses. The accounts didn’t pay any interest. Instead, they charged a fee for each withdrawal.3

  Few men ended up using the accounts that were offered to them, but about two-thirds of the women deposited money at least once. And these women saved more than comparable women who were not offered an account, invested more in their businesses, and were less likely to draw on their working capital when ill. After six months, they were able to purchase on average 10 percent more food for themselves and their family, day in and day out.

  Although the poor do find sophisticated ways to put some money aside, these results show that they would be better off if it were much cheaper to start a bank account. As it is, each account in Kenya costs 450 KES to open, and on average about 5,000 KES got deposited in any account that was used at least once. This means that if Dupas and Robinson had not paid the fee for them, these poor clients would have had to pay a “tax” of nearly 10 percent for the privilege of having an account, not counting the withdrawal fees. To this, we have to add the cost for the poor of going to the bank, usually in a town center, far from where they live. The cost to the bank of managing small amounts of savings has to go down a lot before savings accounts for the poor can be economically viable.

  The “self-help groups” popular in India and elsewhere represent one way to reduce costs, leveraging the idea that if members pool their savings and coordinate their withdrawals and deposits, the total amount in the account will be larger, and the bank will be happy to take it. Technology can also play a role. In Kenya, M-PESA allows users to deposit money into an account linked to their cell phones and then use the cell phone to send money to other people’s accounts and to make payments. Someone like Jennifer Auma, for example, could deposit cash at one of the many local grocery shops that happens to be an M-PESA correspondent. This would credit her M-PESA account. She could then send a text message to her cousin in Lamu, who would be able to present the text message to his local correspondent to get his money. Once he gets the cash, the money would be deducted from her M-PESA account. Once M-PESA is linked to banks, people will be able to wire money in and out of their savings accounts using a local M-PESA correspondent, without having to trek all the way to the bank.

  Of course, no technology would remove the need for regulation of bank accounts. A part of the problem, however, comes from the fact that under the current regulations only highly paid bank employees are generally allowed to handle depositors’ money. This is probably unnecessary. Instead, the bank could use a local shopkeeper to take deposits. As long as the local shopkeeper issues the depositor a receipt for the money that the bank is legally obligated to honor, the depositor is protected. Then it is the bank’s problem to make sure that the shopkeeper doesn’t run away with the saver’s money. If the bank is willing to take that risk—and many banks would be happy to—then why should the regulator care? This realization has been percolating through the system in recent years, and a number of countries have passed new laws permitting this kind of deposit taking (in India, for example, this is called the Banking Correspondent Act). This might eventually revolutionize the whole business of savings.

  There is currently an important international effort, led in particular by the Bill & Melinda Gates Foundation, to increase access to savings accounts for the poor. Microsaving is poised to become the next microfinance revolution. But is the lack of access to formal saving accounts the only issue? Should we concentrate exclusively on making it easy and safe to save? Dupas and Robinson’s results suggest that this is not the whole story. First, there was the disturbing fact that most men did not use their (free) accounts. Many women did not use them either, or used them very little. Forty percent of women did not make a single deposit, and less than half made more than one; many who had started to use the account stopped after a while. In Busia, Kenya, in another study,4 only 25 percent of the couples who were offered up to three accounts for free (one for each member of the couple and a joint account) deposited any money in any of the accounts. This went up to only 31 percent among those who also received a free ATM card to make withdrawals and deposits easier and cheaper. Savings accounts clearly help some people. However, their absence is not the only thing that stops the poor from saving.

  We have already seen, in the previous chapter, another example of people who had lucrative opportunities to save but did not use them: the fruit vendors from Chennai, who borrowed about 1,000 rupees ($45.75 USD PPP) each morning at the rate of 4.69 percent per day. Suppose that the vendors decided to drink two fewer cups of tea for three days. This would save them 5 rupees a day, which could be used to cut down on the amount they would have to borrow. After the first day with less tea, they would have to borrow 5 rupees less. This means that at the end of the second day, they would have to repay 5.23 rupees less (the 5 rupees they did not borrow, plus 23 paisas in interest), which, when added to the 5 rupees they saved that second day by again drinking less tea, would allow them to borrow 10.23 rupees less. By the same logic, by the fourth day, they would have 15.71 rupees that they could use for buying fruit instead of borrowing. Now, say they go back to drinking their two cups more tea but continue to plough the 15.71 rupe
es they had saved from three days of not drinking so much tea back into the business (that is, borrowing that much less). That accumulated amount continues to grow (just as the 10 rupees had turned into 10.71 after two days) and after exactly ninety days, they would be completely debt-free. They would save 40 rupees a day, which is the equivalent of about half a day’s wages. All just for the price of six cups of tea!

  The point is that these vendors are sitting under what appears to be as close to a money tree as we are likely to find anywhere. Why don’t they shake it a bit more? How can we square this with the sophisticated financial planning that we encountered with Jennifer Auma?

  THE PSYCHOLOGY OF SAVINGS

  Understanding the way people think about the future can help resolve these apparent contradictions. Andrei Shleifer, probably the best exponent of the theory that many people sometimes do silly things (he coined, or at least popularized, the term “noise traders” to characterize the behavior of naïve stock traders who are ruthlessly exploited by sophisticated traders), who had just returned from Kenya, shared with us something that he had noticed there: a huge difference between the farms run by a group of nuns, which were lush and vibrant, and those run by their neighbors, which were much less impressive. The nuns were using fertilizer and hybrid seeds. Why, he asked us, were the farmers not able to do what the nuns were doing? Could it be a sign that they were much more impatient (the nuns’ profession presumably inclines them to patience because the rewards are mainly in the afterlife)?

  He had hit on something that had long been a puzzle for us. In surveys conducted over several years, Michael Kremer, Jonathan Robinson, and Esther found that only about 40 percent of the farmers in the Busia region in western Kenya (not far from Sauri, the village where Jeffrey Sachs and Angelina Jolie met Kennedy, the young farmer who had not been using fertilizer before the project gave it to him) had ever used fertilizer, and just 25 percent used fertilizer in any given year.5 Conservative estimates, based on offering a random group of farmers free fertilizer to use on a small part of their fields and then comparing the harvest on that plot to that on a similar plot of land belonging to the same farmer, suggest that the average annual return to using fertilizer exceeds 70 percent: For $1 paid in fertilizer, the average farmer would get $1.70 worth of extra maize. Not quite the returns the fruit vendors could make, but seemingly well worth the effort of saving a little. Why were they not doing it more? It may be that farmers do not know exactly how to use fertilizer. Or they could underestimate the returns. But if that were true, then at least the farmers who got the offer of free fertilizer (and a demonstration of how best to use it) and earned the high returns should be hugely enthusiastic about using it in subsequent seasons. In fact, Esther, Kremer, and Robinson found that the farmers who were given free fertilizer one season were 10 percentage points more likely on average to use fertilizer in the very next season after the study, but that still meant that a majority had gone back to not using fertilizer. It was not that they were unimpressed with what they saw: The vast majority claimed to be convinced and initially said they would surely use fertilizer.

  When we asked some farmers why they did not end up using fertilizer, most replied that they did not have enough money on hand to buy fertilizer when it was time to plant and use it. What is surprising is that fertilizer can be purchased (and used) in small quantities, so this is an investment opportunity that seems easily accessible to farmers with even a small amount of savings. It suggests that the issue, once again, is that farmers find it difficult to hold on to even very small sums of money for the period from harvest to planting. As Michael and Anna Modimba, a couple who farm maize near Budalengi in western Kenya, explained, saving is hard. On their farm, they had used fertilizer in the last growing season, but not the one before, because they had had no money left to buy it then. Saving at home is difficult, they explained, because there is always something that comes up that requires money (someone is sick, someone needs clothes, a guest has to be fed), and it is hard to say no.

  Another farmer we met the same day, Wycliffe Otieno, had found a way to solve this problem. He always made the decision about whether or not to buy fertilizer just after the harvest. If the harvest was sufficient to pay for school fees and provide food for the family, he immediately sold the rest of his crop and used the money to purchase hybrid seeds, and if he had any leftover money, fertilizer. He stored the seeds and the fertilizer until the next planting season. He explained to us that he always bought the fertilizer in advance, because, like the Modimbas, he knew that money kept in the house would not be saved: When there is money in the house, things always happen, he said, and the money disappears.

  We asked him what he did when he had already purchased fertilizer (but not yet used it) and someone got sick. Wasn’t he tempted to resell it at a loss? His answer was that he never found the need to resell the fertilizer. Instead, he tended to reevaluate the true urgency of any need when there was no money lying around. And if something really needed to be paid for, he would kill a chicken or work a bit harder as a bicycle taxi driver (a job he did on the side when he was not too busy with farming). Although they had never purchased fertilizer in advance, the Modimbas had the same view. If a problem came up but they had no money (say, because they had purchased fertilizer), they would figure something out—perhaps borrow from friends or, as they put it, “suspend the issue”; but they would not resell the fertilizer. It was their opinion that it would be a good thing for them to be forced to find an alternative solution, instead of using the cash at home.

  So to help people like the Modimbas, Esther, Kremer, and Robinson designed the Savings and Fertilizer Initiative (SAFI) program. Right after the harvest—when farmers have money in hand—they are given the opportunity to purchase a voucher entitling them to fertilizer at sowing time.6 ICS Africa, an NGO working in the area, implemented the program. Fertilizer was sold at market price, but an ICS field officer visited the farmers at home to sell the vouchers, and the fertilizer was delivered to their homes when they wanted it. The program increased the fraction of farmers who used fertilizer by at least 50 percent. To put this in perspective, the effect of this program was greater than the effect of a 50 percent reduction in the price of fertilizer. Just as Michael and Anna Modimba and Wycliffe Otieno had predicted, as long as it was brought to their door at the right time, farmers were very happy to buy fertilizer.

  But that didn’t explain why the farmers did not buy the fertilizer in advance on their own. A huge majority of the farmers who bought the vouchers went for immediate delivery, then stored the fertilizer and used it later on. In other words, just as Wycliffe Otieno had told us, once they had fertilizer, they didn’t resell it. But if they really want fertilizer, why don’t they go ahead and buy it themselves? We asked the Modimbas. Their answer was that the fertilizer shops did not always have fertilizer available immediately after harvest—they only got it later, just before planting. As Michael Modimba said: “When we have money, they don’t have fertilizer. When they have fertilizer, we don’t have money.” For Wycliffe Otieno, this was not such a problem: Because his job as a bicycle taxi driver brought him into town all the time, he was able to regularly check whether fertilizer had come in, and then buy it from whatever shop happened to have it. But for farmers like the Modimbas, who lived about an hour’s walk from the market town and had few reasons to go there, checking the stores was more difficult. It was this small inconvenience of keeping an eye out for fertilizer delivery (asking a friend to check, calling the store) that was holding back their savings and productivity. All our intervention really did was to remove this minor bottleneck.

  Savings and Self-Control

  The experience of the Indian fruit sellers and the Kenyan farmers suggests that a lot of people fail to save even when they have access to good saving opportunities. This suggests that barriers to savings are not all externally imposed. Part of the problem comes from human psychology. Most of us have some memory of trying to explain to
an irate parent that we were just sitting next to the cookie jar and the cookies somehow vanished. We knew eating the cookies would mean trouble, but the temptation was too strong.

  As we discussed in Chapter 3 on preventive health, the human brain processes the present and the future very differently. In essence, we seem to have a vision of how we should act in the future that is often inconsistent with the way we act today and will act in the future. One form that this “time inconsistency” takes is to spend now, at the same time as we plan to save in the future. In other words, we hope that our “tomorrow’s self” will be more patient than “today’s self” is prepared to be.