LIFE presents plenty of financial challenges and opportunities for rich and poor alike: getting a job, marrying, setting up and furnishing a home, and educating children. These are human goals that are blind to levels of wealth. Each of us likes to feel that we have the means to pursue dreams and to grasp opportunities when they arise. And as the petty pace of life creeps along, we all worry about how to prepare for old age.
Richer families take advantage of loans, insurance, and savings plans to produce the right sums at the right time: a mortgage, a car loan, an education plan, a pension. Financial planners advise the well-off to hold part of their savings in reserve, ensuring that there are funds available for other opportunities—perhaps buying a property to rent out or a share in a business. A separate pot of long-term savings should be accumulated over time, to be prudently drawn down in retirement.
This is a world away from the financial lives of the poor households that we came to know. Yet the previous chapters show that, even for the poorest among them, life is more than just scraping by, day by day, and fending off emergencies. Chapter 2 showed that none of the diary households live hand to mouth, not even those living on one dollar a day and less. Putting together large sums for big events, though, is at least as big a challenge as managing the day-to-day basics. Yet many of the financial diaries households managed to do both.
The previous chapter painted a discouraging picture. The households are rarely able to build up enough savings or arrange for enough insurance to recover quickly from major crises. Insured or not, households in all three countries coped with emergencies by exhausting meager savings, seeking debt quickly, selling precious assets, and leaning heavily on neighbors and relatives—often with damaging long-term costs. Putting together large sums would thus seem beyond the reach of most poor households.
But this turns out to be too pessimistic a view. Many diary households did create usefully large sums during the year, and used them to buy household goods like pots and pans, or assets like bicycles or fans, or to seize new business opportunities or buy land and buildings. Many households needed large sums for social occasions. In Bangladesh and India, even the poor host elaborate weddings, and in India, a quarter of our diary households had to manage a wedding in their own household during the research year.
Each household acquired the necessary sums in its own way, but all of them assembled the funds piecemeal. Nearly always households drew on their entire portfolios, simultaneously running down savings and assets while running up debt and seeking donations from friends and family. Equating acquisition of a lump sum only with saving misses key elements of the households’ strategies and possibilities.
Standard economic surveys that are built to capture “balance sheets” of assets and liabilities at a fixed point in time do not tell us much about how lump sums are put together. They tend to understate households’ ability to generate large sums because many of those sums are created quickly and are not held for long periods.
The same would be true of our own data if we looked only at the balance sheets. In the diary households, year-end balances of financial assets and liabilities are indeed small relative to incomes. Of the 42 households in Bangladesh, for example, only one—a relatively wealthy landowning farmer—held monetary savings worth more than the average annual household income for the sample as a whole. Unlike households in rich communities, the diary households tend not to hold large long-term debt such as a home mortgage. Nor are they building up formal retirement funds (especially outside of South Africa). Nor do young households deposit into long-term education plans to see their children through college.1 In South Africa, far more often than in India or Bangladesh, we did find households that were able to build up long-term financial assets, using, for example, a retirement plan (provident fund) with their employer. For the sample as a whole though, they were exceptional.
But when we turn away from the balance sheets and look at the flows, the diaries show that many of the households in all three countries nevertheless created and spent large lumps during the course of the year. Seldom did we observe households converting these sums into a longer-term financial asset: they were built to be spent.
In this chapter we show the strategies used by households to accumulate these sums. In some ways, the strategies work, but they reveal a striking inability to accumulate over the long term. Without long-term accumulation, households have a hard time building toward bigger goals like better schooling for their children, the chance to migrate in search of better jobs, or securing a stable retirement.
Chapter 6 gives reason to believe that major improvements are possible. There, we describe the remarkable success that Grameen Bank has had in helping its customers build up savings over spans of five and 10 years through innovative “pension” products—which are in practice used for many purposes other than retirement. The evidence here suggests that Grameen’s successes can be spread more broadly: tiny incomes need not inevitably condemn poor households to trivial savings balances and low-value short-term debt. The instruments households already use, and the way they use them, point to the potential the poor have to save and borrow bigger sums over longer durations.
The financial capacity of the poor is constrained not just by low incomes but also by the characteristics of the instruments available to them today. New financial services may not be able to address low incomes, but they can do a lot by ensuring access to financial tools that provide the right doses of discipline, security, flexibility, and incentives. In this, the age-old strategies employed by the diary households anticipate solutions to the blend of economic, psychological, and social constraints that are explored in the new field of behavioral economics.2
It’s surprising that there is room in the household budgets of those living on small incomes to set aside substantial amounts to save and repay loans. It’s hard to imagine that households can maintain the discipline needed to save regularly and to ensure that loans get repaid on time.
Nomsa’s story illustrates the mechanics of saving by the very poor. She is a 77-year-old living with her four grandchildren in the rural village of Lugangeni, South Africa. The two youngest grandchildren, aged 7 and 14, whose mother died of AIDS, arrived just before the research year started. Before they came, Nomsa might have been considered reasonably well off, but now the five of them struggle on her government old-age grant of $115 a month. She has repeatedly asked social workers for a foster care grant that would more than double her income, but despite being eligible, she has been turned away. She supplements her income by selling vegetables from her garden, but she often has to take loans to make ends meet. All the same, she manages to keep up with monthly payments of $40 into her informal savings clubs (which we discuss extensively in a later section). Table 4.1 shows what her budget looked like each month.
Nomsa may seem extraordinary, saving a third of her monthly budget, but her savings patterns are not much different from most of her neighbors. Nomsa has a bank account into which she receives her monthly government grant, but she withdraws the entire amount every month. Likewise, she has a place in the house where she keeps spare cash, but this rarely amounts to much by the end of the month. Like her neighbors, Nomsa was able to save so much of her monthly income thanks to her two informal savings clubs.
Table 4.1 Nomsa’s Typical Monthly Budget
We found that, in all three countries, all families, even the poorest, attempted to accumulate lump sums of money over time through building up savings and paying off loans. Take a household like Sultan and Kanon’s. This Bangladeshi couple rented a yard where they sorted and sold waste scavenged in their Dhaka slum, but Sultan was in his fifties and ailing, and the income he raised was rarely more than $1.50 a day. Just before the research year their 15-year-old daughter Sweetie had found a job in a garments factory, at $28 a month plus occasional overtime, much of which she saved for her wedding while contributing her bit to the housekeeping: she married and left home just before the end of the year. Kanon was a client of a microcredit provider, and before the year had taken a loan of $110 that they used for a string of needs: drugs for Sultan’s health problems, repaying old loans from neighbors, consumption, and paying overdue rent on the waste-sorting yard. In addition, Kanon’s older daughter, already married and away from home, gave her microcredit loan to Sultan and Kanon to help fund Sweetie’s marriage. Sultan and Kanon gritted their teeth and kept up with the weekly loan payments on both these loans: $3.76 a week, week in, week out. On top of that, they saved another 75 cents each week with the microcredit NGOs. So, for months on end, they managed to squeeze $4.51 out of a weekly income of $20 or less, to repay their loans and save at the microcredit meetings.
Or take Sita, whom we met in chapter 2, a widow from the India rural site with low and very uneven income as a farm laborer. Sita lives with the eldest and youngest of her three sons: Udal, whose new bride came to join him in the research year, and Lalla. The whole household is illiterate but adult and able-bodied, earning the bulk of income through forms of wage labor—on local farms, on construction sites locally and in the regional capital of Allahabad. Lalla was contracted to a local grain trader to work for 43 cents per day (just under half the local market rate) in order to pay off a loan of $64 taken to pay for Udal’s wedding. Sita has title to 3.5 acres of land but only one acre is fertile, the rest rocky and unirrigated, and the fertile section was mortgaged two years earlier to raise money for bail for Udal, who had been charged with a robbery in the village, leaving the family shackled by court fees. The farm income, at $10 for the rice paddy season, was less than half of what Sita had expected. All this added up to an annual income of $353, averaging just under $30 per month.
Despite such low and uneven income, the household managed to put money aside from daily needs to go toward longer-term costs and debts. Over the year, they saved and repaid about $63, a little under a fifth of their annual income. Most of this was saved up from wages and kept in the home to repay a private loan secured by a land mortgage; the balance was in the form of deductions from Lalla’s wages, which went toward repaying his debt to his employer grain-trader.
It seems, then, that at both ends of the spectrum of households in our sample—from the responsibility-burdened Nomsa, to the fit but precarious Sita, to the frail and elderly Sultan—there was room in the budget to set aside money on a regular basis. Other research suggests that this may be true for poor households worldwide. In a 2007 paper, MIT economists Abhijit Banerjee and Esther Duflo report that surveys from around the globe show that the poor do not spend every available cent on food, leaving room in the budget for financial transactions that lead to larger expenses.3 The financial diaries data show that most South African households spent no more than 75 percent of income on goods and services: the balance went toward financial intermediation such as insurance, savings, or debt servicing. The next, crucial step is to find ways to protect the money that has been set aside and to transform it into usefully large sums.
Our focus in the chapter is on how Nomsa, Sita, Sultan, and their neighbors built usefully large lump sums of money. Their strategy, as we have seen, was to patch together resources from multiple points. Sometimes, however, a single instrument contributed a substantial sum, and much is revealed by looking at how that was done. We defined a “larger” sum as any sum formed completely in a single instrument, rather than patched together through different instruments, and equal to or exceeding one month’s household income. In Bangladesh and India, this benchmark was averaged for the country sample and set at about $50. In South Africa, where we had more precise income figures, it was set at each household’s own average monthly income.4 Altogether, our households between them acquired and spent 298 such sums during the research year. The total value of these sums, $80,857, is broken down by country in table 4.2, which also shows the average values of the sums for each country.
In India and Bangladesh the typical household extracted usefully large sums from financial tools with an average value of around three months’ income. In all three countries, when we compare wealthier households to poorer ones in the same neighborhood, we find that the poor households are able to draw, relative to their incomes, lump sums that are a larger proportion of income than those of a richer neighbor. Joseph, who lives in the shack areas of urban Langa, outside Cape Town, South Africa, has a stop order on his bank account that transfers a set amount of money to a savings account every month. He managed to save $630 this way over the course of one year, about one and a half times his monthly income. His neighbor, Nobunto, who earns half as much, saved $407, about two and a half times her monthly income, through a savings club. Many poor households, then, do manage to create substantial sums in their financial instruments.
Table 4.2 Lump Sums from a Single Instrument Spent in the Research Year, by Country
Table 4.3 Types of Instruments Used to Form Lump Sums
Table 4.3 puts another lens on the accumulation of large sums. It shows the type of financial tool—saving, borrowing, or insurance—used to create the 298 lump sums on which we’re focusing. A clear difference between South Asia and South Africa emerges. The relatively low shares for saving in Bangladesh and India show how hard it is to save up more than a month’s income there—in both countries it is far more common to borrow other people’s savings than to build your own. In contrast, most of the large sums in South Africa were raised by building up savings, partly in bank accounts but mostly in savings clubs. Insurance features significantly only in South Africa, which we explored at some length in the previous chapter.
Table 4.4 Primary Use of 298 Large Sums
Table 4.4 outlines three very general categories of the uses of these lump sums. “Emergencies” include all sudden-onset occurrences that threatened life, health, or property. Under “life cycle” uses we include household consumption, as well as expenditure on births, marriages, and deaths. “Opportunity” is the broadest class, broken down further in table 4.5 and discussed below. Of course, many lump sums were put to more than one use. In those cases, we allocated the lump sum to the use class for which the majority of the sum was used. We discuss these three categories in turn below.
Life Cycle Uses. For economists, the simplest theory of why households borrow and save hinges on life-cycle motivations. It asserts that households aim to match income and expenditure patterns over the long swings of a lifetime, from early on as a young worker, to the years of building a family, and, eventually, to retiring. The theory posits that households borrow when young, before income is sufficient to meet major needs like buying a house. As soon as practical, saving starts with an eye to retirement. And those savings are then gradually drawn down once retirement begins.
But the life-cycle theory captures only a part of what we see here. Its application is limited since even elderly people in our sample work late in life, like 77-year-old Nomsa, who was selling vegetables to care for her grandchildren after her daughter died of AIDS. Yet the life-cycle motive is hardly absent, though it can be hard to see at first.
If we turn from the 298 sums that were spent during the year to the subset of larger financial assets still sitting on the year-end balance sheets of our portfolios, we find that little of this money is in instruments designed to safeguard old age. In South Africa, only 15 percent of the diary households would have enough savings and assets to finance more than five years of retirement.5 Those wealthier households that do have retirement savings tend to do so through employer-provided instruments, such as retirement or pension funds, but poor households tend to hold very little long-term financial wealth. Making provision for old age just isn’t done directly by means of financial tools in our diary households. Still, many conversations with diary households suggest that the desire for security in old age is often behind their financial transactions.
Khadeja from Dhaka, whom we found borrowing to buy gold in chapter 1, saw a gold necklace as a valuable hedge against future uncertainty: the very real possibility, in the environment she lived in, that she could be widowed prematurely, or perhaps deserted or divorced. She used a financial tool—a microcredit loan—to buy gold. Her case is a good example of how poor households may use the short-term financial tools that are available to them to create stores of wealth as substitutes for the long-term financial tools, like pension plans, that aren’t available.
Vishaka, a diarist in a Delhi slum, would have seen eye to eye with Khadeja. Unlike Khadeja, who used microcredit, Vishaka used a savings club as her short-term saving instrument. When she received her payout from the club, her husband, Om Pal Singh, suggested that they store the money nearby with a moneyguard, Vishaka’s mother. Om Pal Singh pointed out that, with their expenses increasing—they had four children by then—they might need to draw on the money at short notice. This was exactly what Vishaka feared could happen, undermining her longer-term savings goals. Instead, she deposited the money with a goldsmith, who would keep the money at further reach. When she’s saved a bit more, she’ll purchase some gold—her savings for the future.
A very common use for lump sums is buying land, which we’ve classed as an opportunity (see table 4.5 for a breakdown of opportunity uses). The precise motive for buying land can be influenced by cultural norms: one of the better-off households living in the slums of Langa, Cape Town, in South Africa was clearly thinking about their last days when they told us that they were investing all of their savings in the home they were building back in the Eastern Cape. This was not because they planned to move there soon, but because, if you have a family of your own, “you can’t be buried from your parents’ home—they must take the coffin from your own home.” But in all three countries, land is seen with an eye toward future security. In Bangladesh and India urban households often sent funds, built through loans or through savings, back to the home village for investment in land or buildings.
Weddings, Funerals, and Broader Life-Cycle Uses. While the basic life-cycle theory of saving focuses on retirement, a broader model would account for major life events along the way. In Bangladesh and India weddings were by far the most common expensive “life-cycle” event. In South Africa, as we saw in the previous chapter, most households did hold specialist tools designed to produce large sums for the most common costly life-cycle expenditure there—funerals. Other important South African life-cycle events, such as the paying of lobola (bride price) or initiation school for boys, were also key financial events for diaries households. As Abhijit Banerjee and Esther Duflo find across a range of countries,6 expenses on religious and social events account for a large share of the expenses of poor households, and often require expenditures in large sums.
In India a quarter of our diary households had to manage a wedding in their own household during the research year, and 45 percent contributed to a wedding outside the home. Not surprisingly, then, of the 42 larger sums that were used for life-cycle events in India (table 4.4), almost all were spent on weddings. But they were rarely enough to cover the whole cost. For those rural Indian households where a child married during the research year, the wedding was an astonishing 56 percent of the total spending for the year. More commonly, then, these larger sums were a major element in the patchwork of resources that had to be assembled from a mix of interest-free loans, gifts, savings, and various forms of credit.
The enormity of weddings in the financial lives of rural Indians bears some explanation. The dowry amassed and the lavishness of the ceremony itself not only honor a daughter leaving home and support her well-being in her husband’s family, they are also strategies of aspiration whereby the social-economic standing of her whole family can be improved through a “good marriage.”
The situation is similar in Bangladesh. Ataur, the head of one of our rural diary households there, married out a daughter and a son during the research year. For the daughter’s wedding he sold assets (a cow for $50, a goat for $10, and some bamboo for $40), saved furiously at home (peaking at $240 just before the marriage), borrowed sparingly on the local market ($10, repaid along with another $10 interest, and then $40, repaid after two months with another $14 interest), and used a $200 loan taken earlier from a microcredit lender. For the son’s marriage, a few months later, they were on the receiving end of dowry payments—$100 in cash and $13 worth of jewelry—that more than offset their share of the marriage expenses.7
Emergencies. Life-cycle events provide a motive for saving, one that Princeton economist Angus Deaton terms “low frequency” saving because the events are generally predictable well in advance and savings strategies can be put in place without requiring substantial revision.8 “High frequency” saving, in contrast, refers to the sort of everyday consumption smoothing and cash-flow management that we described in chapter 2. Expensive emergencies are another matter and require larger sums. In the diary households, however, emergencies account for a rather small share of the 298 large sums we identified, as table 4.4 shows.
This low share was not for any shortage of crises, as we’ve shown in chapter 3. Rather, it is because households were unable to respond to emergencies with tailor-made large sums, there being no systematic insurance tools to allow them to do so. Instead, emergencies were met with a mosaic of smaller loans and savings combined with asset sales. In Bangladesh, microcredit loans were seldom available for emergencies, because they were disbursed on an annual cycle, with prepayments (which would lead to the early release of a fresh loan) not allowed. If microcredit loans contributed to emergencies, they did so indirectly: a microcredit client might be able to secure a private loan, for example, by assuring the lender that she was due to get a microcredit loan within the next few months and thus would be positioned to repay the moneylender. Likewise, many savings clubs in all three countries pay out only at pre-specified times, leaving a saver with a sudden emergency unable to access funds.
Opportunities. It turns out that most of the larger sums were spent to seize opportunities of various sorts. Table 4.5 shows that investments in land and buildings were major uses everywhere, though investments in land tended to be perceived differently in diary households in South Africa than in those in India and Bangladesh. In South Africa, the rural land populated by low-income households is rarely seen as a financial investment, but rather one that is made in order to fulfill cultural obligations and desires. Even in urban townships in South Africa, the secondary market is only just beginning to function and provide owners with the opportunity to buy and sell their properties, holding out slim hope for a return on investment. But in both India and Bangladesh, land is a fast-appreciating store of wealth, and the motive behind investing in land in these two countries is most certainly economic. In India, banks were becoming part of the land investment process. Of those sums used for property acquisition in India, about a quarter were formed in banks, through products that included savings as well as farm loans.
For other subcategories there was considerable variation between the three countries, especially between South Africa on the one hand, and South Asia on the other. Inputs into small business inventory in trading enterprises were the most common use in the investment category in both Bangladesh and India, but didn’t feature at all in South Africa. This does not mean that there were few small businesses among our South African sample, but business income was a much smaller share of household income than in the South Asian samples. As a result, in South Africa the lump sums used to finance working capital did not reach the value of the average monthly income that we used as a benchmark.
Table 4.5 Primary Use of 194 Large Sums Used for Opportunities
In Bangladesh, the microcredit providers, whose self-declared job it was to provide business capital to poor households, contributed to these numbers, but were responsible for only a minority share of them: about three times as many sums went into businesses from informal private sources than from microlenders. In India, where microlenders were thin on the ground in the diary areas, an even larger proportion of lump sums in the opportunities category were used for business. The majority (58 percent) of these were formed in the informal sector, but a substantial proportion were formed in the formal sector and just a handful through microcredit loans. Nearly all “formal” lump sums used for business came from bank or credit cooperative loans to farmers, another demonstration of the banks’ commendable outreach to (larger) farmers, and their poor outreach to other occupation groups.
In our sample, South Asians appear more likely than South Africans to use the sums they formed as a basis for lending to others. Where lump sums can be raised cheaply relative to other means, it makes sense to arbitrage: we have recounted stories where microcredit borrowers quickly lend their capital to others who not only repay and service the loan but pay additional interest, perhaps in the form of contributing the savings deposits into the microfinance account, as Hanif, Mumtaz’s boarder, did in chapter 2.
We found several sums being used to actually pay down other debt. In Bangladesh, microfinance loans are cheaper than private moneylender loans and are often used to pay off the latter. In India, several lump sums were borrowed by rural respondents to repay valuable lenders such as wholesalers and banks with strict deadlines.
When we began to look at the relative shares of saving and borrowing in the strategies used by the diary households to build larger sums, we were struck by an unexpected observation. Saving and borrowing turn out, in practice, to be surprisingly similar. Both involve steady, incremental pay-ins—saving week after week in small amounts, say, or paying back loans week after week in small amounts. The same is true of the way that common informal devices are designed, like the local savings and credit clubs described below—and, indeed, it is similar to the way that many richer people pay for insurance or contribute to pensions. It is one of the features that microcredit pioneers adapted to form new financial innovations.
We pay attention to this and other special features of the devices and strategies used by the diary households. We focus first on the borrowing side (the “accelerators”) and then turn to the various savings devices (the “accumulators”). In both their borrowing and their lending, households have discovered ways to deal with the economic, psychological, and social forces that make the job of amassing large sums of money so difficult.
We started this chapter with the idea that large sums are formed by patching together resources—putting luck, skill, and assets together to amass a needed lump. Loans are part of that process everywhere.
Loans are “accelerators” in the lump-sum-building process in more than one way. Obviously, they give households access to cash immediately rather than after a slow process of saving. But loans often have features that speed up the process even further. Price is one of them. In a slum in Vijayawada, a town in southern India, Seema negotiated a loan of $20 from a moneylender, at 15 percent a month, just after leaving a meeting of her local savings cooperative where she had $55 in a liquid savings account. This struck us as an expensive, perhaps even an irrational choice. But asked why she had done it, Seema said, “Because at this interest rate I know I’ll pay back the loan money very quickly. If I withdrew my savings it would take me a long time to rebuild the balance.”
This logic is used even by wealthier respondents. Delhi-based Satish’s $1,232 of cash assets at year end were the third highest in the whole Indian sample, after those of two wealthy farmers. And yet he loved to borrow, and to borrow above all on interest. He ended the year with $575 worth of debts, over half of it interest-bearing. His explanation was that the pressure of interest charges encouraged him to repay quicker, which he liked. Seema and Satish have their equivalents in wealthy countries: the pattern of borrowing at high cost even when adequate savings are in the bank is a regularity noted by economists working with data from low-income credit card holders in the United States.9
Seema and Satish used the pressure of price to make sure they put money aside. Khadeja, who took a loan at 36 percent a year and spent much of it on gold jewelry that she saw as a vital store of value for her future, used the pressure that the weekly discipline of her microcredit provider exerted on her. Like Seema, Khadeja saw the truth of an odd-sounding paradox: if you’re poor, borrowing can be the quickest way to save. Khadeja knew that without some external force to help her, her chances of saving enough money to buy the gold necklace were small. So when a microfinance NGO offered her the chance to turn a year’s worth of small weekly payments into a usefully large sum, she took it.
We watched as another Bangladeshi discovered the same thing. Surjo, an educated but poor young man, headed a large household that included his widowed mother and several siblings. He tried his hand at various ways of assembling money for the many needs of their growing household. When we met him, he told us that he was determined to save, and that he had opened a bank account to do so. That month he duly deposited $10 out of his $55 monthly factory wage into the account. Next month was the Eid festival, so he excused himself from depositing. The next month he made another excuse. We never saw him deposit again.
But his mother joined a microfinance group, and discovered a much more reliable way to get Surjo to save. During the year, she took a loan of $180 that she used to lease land in their home village, which they then sharecropped out to secure a supply of rice. Surjo realized he would be better off repaying this loan than pursuing his failed attempt to save in the bank. We watched as the family regularly made the weekly repayment of just under $4 right through the year.10
Table 4.6 Where Large Sums Were Formed
If Khadeja and Surjo had savings tools that were safe, and as disciplined as the microfinance loans, they would have been better off, since they would not have had to pay interest. But given the options available to them, their “borrowing to save” strategy makes sense.
We noted in table 4.3 that 83 percent of the large sums that the Bangladeshi diary households took and spent in the research year, and 73 percent of those spent by the Indian households, were formed by borrowing rather than saving. And though they also saved in informal clubs and other devices, for many of our South Asian diarists, borrowing proved the most manageable way to turn their capacity to set money aside into useful sums.
One reason for the preponderance of borrowing in Bangladesh is that, spurred by Grameen Bank’s success, the country has many microfinance providers, and, as we have seen, many of our diary households held memberships in these organizations. Table 4.6 reveals the differences between the three countries in the roles played by the formal, semiformal, and informal sectors in the creation of usefully large sums.
South Africa’s formal sector reached many of the diary households (Nomsa had her own bank account), but it has a poorly developed “semiformal” (or microfinance) sector, so 58 percent of sums were formed in the informal sector—mostly in the various clubs that we referred to earlier in this chapter. Bangladesh’s rich set of microfinance institutions reached a majority of our diary households there, with the result that the semiformal sector was involved in the formation of more than a third of large sums—mostly through loans rather than through savings accounts. India has the highest proportion of informal sector formation of sums: so far, it has fewer semiformal providers than Bangladesh, and its banks and insurance companies were less successful than South Africa’s in reaching diary households.
Thus far in this chapter, we’ve referred several times to households using informal savings clubs, a very common accumulator throughout the developing world. We saw that Nomsa, who featured at the start of this chapter, saved a large portion of her monthly income with her savings clubs, and she was typical of the South African households as a whole. Although she used a bank account to receive her government old-age grant, the financial instruments responsible for the bulk of her savings were her savings clubs: community-based organizations that are tried and tested methods for helping poor people squeeze savings out of their budget month after month. They are informal in the sense of not being legally incorporated and not relying on legal contracts. Instead they build on the trust and mutual obligations that bind neighbors together.
At the time we knew her, Nomsa was in two different sorts of clubs. The simpler was a saving-up club. It consisted of a group of women from the neighborhood who each deposited about $9 a month. The secretary of the club kept the money in her house. At the end of the year, clubs like these pay out the accumulated amount, splitting it among the members. Nomsa expected to receive $99 (11 months at $9) from this savings club in December.
Nomsa’s membership in the club poses a puzzle. After all, she has an account at the bank in her own name, and is used to transacting there. Why would Nomsa not bank this money for herself, avoiding the bother of the club (she has to attend its meetings) and its undoubted risks (what if the money is stolen from the secretary’s house?)? Many South African diary households belonged to clubs of this sort, and their most common answer to this question was that club membership was the surest way to discipline themselves to save for a particular event. “You feel compelled to contribute your payment. If you don’t do that, [it] is like you are letting your friends down. So it is better because you make your payment no matter what.”
Savings clubs, then, do the job that automated payments into savings accounts, or “stop orders” do for earners in rich economies: they shift money into a “hands off” account, acting as a guard against the temptation to spend spare money in trivial ways. In this, they play important psychological and social roles, building on commonsense notions that have only been recently recognized by behavioral economists.11 The basic idea is that many people, both rich and poor, are often caught in a bind. They feel the need to put aside resources for the future, but they are also impatient to spend today (often with good reason, if, say, health and nutrition needs loom). If impatience outruns concern for the future, little will be saved for later needs.
Given this bind, devices that permit households to commit to save steadily in a prescribed pattern (or to commit to pay down debt quickly) can make the households better off. The devices keep impatience in check and help households bring their two competing desires into balance. In essence, they allow users to exercise their self-control at a critical early moment—in the act of entering into a months-long arrangement—rather than having to battle competing desires (consume now? save for later?) several times each day or whenever important purchases are contemplated. A study in the Philippines, for example, shows that bank customers save much more when offered a type of savings account that allows them to commit to making regular deposits at fixed intervals over a given period.12 As noted above, richer households have many devices that do these jobs—like automatic salary deductions into retirement accounts. Poorer households usually have to rely on informal arrangements of their own making.13
It is not always simple impatience for consumption that poses a dilemma for poor households, but the uncertainty about the weight of an immediate need compared to that of a more distant one. And it is not always one’s own desires and needs that create the conflict. It may be one’s spouse who has a different idea of how to allocate funds. Or relatives may unexpectedly arrive with requests for assistance. The requests can accumulate and stretch the limits of generosity. Devices like Nomsa’s saving-up club and the other forms of informal savings clubs discussed below provide a common way to set and respect fair and reasonable bounds.14
FIGURE 4.1. Cash-flow schematic for Nomsa’s saving-up club. US$ converted from South African rand at $ = 6.5 rand, market rate.
Saving-Up Clubs. Figure 4.1 illustrates the cash flows for the first of Nomsa’s clubs—the saving-up club. It reveals the common feature of all such devices: they mobilize small, steady flows—$9 a month for Nomsa—and transform them into one large sum. The relatively small size of the monthly inputs allows them to be made without too much difficulty, but they are large enough to accumulate to a meaningful size over time. The simple idea of the steady schedule underlies these savings devices. Usually, in a saving-up club, the fund builds up in the bank or in a member’s home and isn’t touched until an agreed term has finished—and very often the term is set to end just before a major expensive festival such as Christmas, Eid, or Dewali.
The “slow and steady” schedule is similar to microfinance loan repayment schedules that so helped Khadeja and Surjo. This is the sense in which saving and borrowing often share a very similar process: small sums are steadily set aside in return for a single large sum received at the appointed date. From this vantage, a key difference between saving and borrowing is when the large sum is received: at the very start with a loan, or at the very end through saving.
RoSCAs. Nomsa’s second savings club was a “RoSCA,” or rotating savings and credit association. In a RoSCA the members save the same amount as each other every period—a month, say—and the total amount saved each period is given in whole to one of the members. This continues until everyone has received the “prize,” at which point the RoSCA comes naturally to a close—though of course its members may choose to start another cycle immediately or at any later time. One of the beauties of the RoSCA is that it requires neither storage of group-held funds (there are none) nor complicated bookkeeping (all that is required is a list of who has received the prize and who remains in line).
Nomsa’s RoSCA had just three members, close friends, and that made things especially simple. Each of them put in $31 each month, taking turns to come away from the meeting with $93 once every three months. The first time she was paid out, Nomsa used some of the money to repair her rondavel (a traditional round building with a grass roof), bought a pot, and paid off one of her debts. The second time she was paid out, she used the money to make further repairs to the rondavel, pay someone with a tractor to till the soil of her garden, and, again, to pay off one of her short-term debts.
RoSCAs are flexible enough to accommodate almost any number of members, any interval between payments, and any value of pay-in. They can also change all of their terms from cycle to cycle.15 On the other hand, they impose strong discipline through their structured regularity.
FIGURE 4.2. Cash-flow schematic for Nomsa’s RoSCA. US$ converted from South African rand at $ = 6.5 rand, market rate.
ASCAs. A third form of savings club popular among our South African households is the ASCA, or accumulating savings and credit association. Unlike the simpler saving-up club, both RoSCAs and ASCAs make use of the saved money while it is being accumulated, rather than just storing it. In an ASCA, which is a step closer to a credit union or credit cooperative than a RoSCA, members save regularly, but they do not “zero out” the fund each meeting by giving it in whole to one member in the way that a RoSCA does. Instead, the ASCA lends the fund in part to individual members (and in some cases to nonmembers), in varying amounts, charging interest on the loans and agreeing to a repayment schedule with the borrower. It may also accumulate any unlent part of the fund, storing it with the group’s treasurer or in the bank.
Nomsa didn’t belong to an ASCA, but Sylvia, another of our South African diarists, did. It had 33 members who each paid in $30 per month. As the pooled fund accumulated, members were obliged by the club’s rules to take part of the fund and lend it to nonmembers during the month. Sylvia usually took quite large amounts of money from the ASCA to lend to her neighbors. From July to November alone, she lent to a total of sixteen people an average of $60 each. They were charged 30 percent per month, the rate stated in the club’s rules. The interest earned on these loans was paid into the club, where it further increased the fund’s size. At the end of an agreed period the club closed, and savings and profits were distributed back to the members in proportion to their savings and lending record.
ASCAs like Sylvia’s obviously do more than help members save. They are designed to help members profit from their savings (which we discuss further in chapter 5), a feature that can make them unstable, as we shall see in a later section. Nevertheless, they take their place alongside saving-up clubs and RoSCAs as popular accumulator devices used by our South African households to overcome the difficulties of saving. In all, 67 percent of South African diarists belonged to at least one saving-up club, RoSCA, or ASCA.
Mutation, Adaptation, and Evolution of Informal Clubs. We have used South African examples to illustrate savings clubs because, as table 4.3 showed, saving was the preferred way of forming “usefully large sums” there, whereas in South Asia borrowing was more common. Nevertheless, India and Bangladesh have rich traditions of savings clubs of their own. Saving-up clubs, RoSCAs, and ASCAs are all found in the South Asian diaries. As in South Africa, they take many forms, since one of the benefits of informal clubs is that the members’ requirements can quickly shape the instrument’s structure. As needed, the members can simply change the rules by consensus, without the burden of consulting board members or applying to regulators as formalized institutions must do. The result is the evolution of an almost endless number of different formats of savings clubs, as each tries to get ever closer to a perfect match between the lump-sum needs and the cash flows of its members.
Some arrangements are not so much clubs as informally recognized reciprocal bonds. In the previous chapter we saw them at work in South Africa in one variant of the burial society, where no money changes hands until a funeral takes place, when all the households that form the social network of the bereaved contribute to the costs. Such ties are part of a household’s “credit rating,” just as in wealthy environments credit card accounts are maintained, though perhaps seldom used, in order to maintain options. These obligations are not “drawn down” continuously, but maintained in good standing against the time they will be needed. In this way, they function as risk-sharing devices rather than simply ways to save or borrow. In India, we see the same tradition used to finance weddings—the most expensive festival in the South Asian setting. One of our Indian diarists, Rajesh, told us that he used to give substantial gifts, totaling some $385, to family members to finance weddings when he was running a successful carpet-knotting unit. He then fell on hard times and earned most of his money through off-farm wage labor in the local market town. Just before our research, he arranged the marriage of his own daughter and financed it largely by getting his past gifts reciprocated, even though some of them were given years ago.
In South Asia, as in Asia generally, RoSCAs are common. There are many variants, and they can be distinguished by the method used for determining the order in which members take their “prize.” A few do it by consensus, agreeing the order before the cycle begins, a system that works well when the members intend to repeat their RoSCA cycle after cycle, so that after a few cycles it hardly matters what your “order number” is—you just get a prize at a regular fixed interval. Nomsa’s three-member RoSCA was of this sort. Nasir and his brothers, tailors for leather export factories in Delhi, joined a consensus RoSCA at the instigation of a colleague who became the manager. There were ten members contributing $21 each month for ten months. All came from the same district of Bihar (six of them from Nasir’s own village, including his two brothers, his first cousin, and his wife’s brother) and worked in the same company.
Because of close kin relations and high trust between the brothers and the RoSCA manager (who is also a source of interest-free loans), the RoSCA rules were flexible: fines went uncollected and the brothers regularly paid up for each other as well as for other members. When Nasir and one of his brothers lost their jobs following a protest about wage rates, their eldest brother took responsibility for all three payments to the RoSCA ($64 each month) in return for the others covering his living costs. The RoSCA survived because it had a core of close kin who helped each other out. The three brothers were saving toward the same common goals as a joint family, and the RoSCA, like some of South Africa’s savings clubs, brought existing reciprocal bonds into a formal structure.
Such RoSCAs can vary the value of the payments, the number of members, and the frequency of meetings, to arrive at a balance between the timing of contributions and prize-taking that suits particular needs. In the slums of Nairobi, Kenya,16 for example, Mary’s RoSCA helped her reconcile her tiny business—buying a basketful of vegetables daily from the market and selling them to her close neighbors—with the demands of being a single mother. She was in a seven-day daily-payment RoSCA, which delivered her a prize each week the value of which matched the value of her vegetable stock-in-trade. Whenever Mary needed to pay for some unexpected event—her son had fallen from a tree and needed a visit to the doctor on the day we interviewed her—she had to take cash from her tiny business capital, but she found that if she was faithful to her RoSCA savings, which she was, then within a week it replenished her capital in full. For this reason she valued the savings club highly, and returned to it after an unsuccessful try at being a borrower from a well-intentioned microfinance institution that had suddenly arrived in her slum. The problem was that the microfinance loan had a one-year term, which didn’t match Mary’s cash-flow needs.
In the Philippines, Taiwan, Pakistan, and Egypt, to mention just four countries where we have observed it in action, there is a RoSCA tradition that has found another way of matching needs with cash flows. There, we find RoSCAs that are one-cycle affairs, but each starts with the lump sum need of a particular individual, who then devises a schedule to suit the remaining members. A schoolteacher in rural Philippines who wants to buy a suite of furniture for a new home, for example, will “call” the start of a RoSCA of, say $100 (the amount she needs for her furniture). Her colleagues join in as an act of solidarity, but only if she works out a schedule that suits them—say $20 a month over the next five months, so that the whole thing is over before the expensive Christmas season starts.
Yet another form of RoSCA uses a lottery, so that the prizewinner isn’t known until the hour of the meeting, when a name is drawn from a hatful of eligible names (that is, all the members who haven’t yet received their prize) and a big smile breaks out on the lucky member’s face. This is simple to administer and it helps to make the order in which the prize is given seem fair. The “lottery RoSCA” may be the most common form of RoSCA in South Asia. In the Indian diary research we found prizewinners of lottery RoSCAs who, sometimes for a price, passed the prize on to other members who needed it more at the time, or who even on-lent it to outsiders, rendering the device more responsive to individual cash flows. Members with larger needs may also be allowed to hold more than one share, or “name” in the RoSCA.
A more refined solution, used by two of our Indian diarists, was another type of RoSCA, the “auction” RoSCA, where those members still eligible for a prize bid for it, with the prize going to whoever puts in the biggest bid. The bid money is then distributed among the members equally, so that those willing to refrain from bidding until the later rounds, when bids are smaller (since there are few bidders left), get a bigger than average prize for a smaller than average bid, as well as enjoying “interest” income from their share of the distributed bid money paid by others. Auction RoSCAs, therefore, cleverly attract savers (who bid late and are well rewarded for it) and borrowers (who bid early and pay heavily for it), and the current price of money is determined at each auction, driven by demand for it at that moment among that group of people. All this sophisticated matching of savers with borrowers, and all the associated accounting, is done without conscious analysis and with no need for pencil and paper. Moreover, the money flows directly from savers to borrowers without any down time or middlemen to cream off their percentage.
In this way, auction RoSCAs could be considered the world’s most efficient intermediation system. Perhaps not surprisingly, in India auction RoSCAs have developed into a licensed financial industry, known as “chit funds,” with tens of thousands of licensed chit fund managers running RoSCAs on behalf of their members, in return for a fee.
Indian RoSCAs, as everywhere, show their vitality through intriguing local variations. For example, we heard of auction RoSCAs in Delhi that draw lots when there are no bidders, but deduct a fixed sum from the prize that is then distributed to members. This creates a disincentive for those who want the money today but might otherwise be inclined to simply sit and wait, hoping no one else will ask for it. Some Indian auction RoSCAs reward the manager (who may be the originator of the RoSCA, as in the Filipino example described above) by allowing him or her to take the whole prize (nothing deducted in the bidding) on the first round, so that bidding actually begins only in the second round.
We found that people we met through the diaries took their RoSCAs and other similar clubs seriously. They were important to them on two fronts simultaneously: the social and the financial. Kenneth, one of our South African diarists, was a well-respected 81-year-old man from the urban area of Langa. He had both a job and a pension from a previous job, and enjoyed an income of $320 a month. Kenneth was one of only two respondents in the study who held unit trusts (mutual funds): he had $2,900 invested in a well-known income fund. But he prized his informal stokvel, a local word for various kinds of saving club, even more highly. He had been in the stokvel for many years, and rather than miss a payment into it, he would borrow—as we saw him do once during the year. His stokvel worked on a rotating basis, though not in the exact same way as a RoSCA. At each meeting, one member received a payment from all the other members. But the amount was not fixed: it depended instead on how much the recipient gave each donor in previous rounds. The rule was to give back a little more than you received. So if he gave $325 to a member when it was his turn, he would expect to receive, say, $355 in his own turn. This type of stokvel tends to invite more middle-income and wealthy people in the neighborhood and can generate very high payments: in the latest we saw the receiving member get $14,900 in total!
Kenneth’s stokvel is not a simple RoSCA or ASCA as we have defined them. It is in fact a very sophisticated way of bringing reciprocal one-on-one lending and borrowing into a structured context to strengthen it. Kenneth has a series of “contracts” with each club member individually—with differing amounts for different people. But this set of reciprocal bilateral deals is played out in public using the machinery of savings clubs—formal set meetings at regular intervals. As a result, the peer pressure, and the trust built from reiterated promises kept, are harnessed to discipline and strengthen the one-to-one deals. Although we did not find this mechanism in Bangladesh and India, we have found it on the other side of the world from South Africa—in the mountain villages in the northern Filipino island of Luzon. There, the ubbu-tungngul functions just like Kenneth’s stokvel.17 Indeed, the Filipino villagers report that the discipline produced by this device is so strong that ubbu-tungnguls can last from generation to generation, with children inheriting membership from their parents. A money-management device strengthens social ties, and the social ties in turn strengthen the money management—a symbiotic relationship that is one of the strongest virtues of informal finance.
RoSCAs and their like blur the distinction between saving and borrowing. In a RoSCA, members are transformed, one by one, from net savers into net borrowers. This happens because the basic mechanism is the intermediation of a series of small pay-ins into a single larger payout, and this mechanism is true for both savers and borrowers.
The stories we have told in this chapter are evidence of poor households’ determination to save or borrow their way to usefully large sums, and of the widespread distribution of powerful and sometimes elegant informal mechanisms to help them do so. We have also seen the emergence in South Asia, and above all in Bangladesh, of new semiformal providers (i.e., microlenders like Grameen Bank, BRAC, and ASA) with equally powerful tools based on loans repaid in small regular installments.
But these are imperfect instruments, and in this section we will review their most common shortcomings: low reliability, inflexible schedules, and terms that can be too short.
Savings clubs, as powerful a savings device as they can be, are not always reliable. They may be unreliable in small ways—a member may not make the expected contribution at the exact time that you need the payout. In Nomsa’s saving-up club, for example, not everyone, despite sincerely held intentions and loudly voiced avowals, paid on time every month (including Nomsa), throwing the timing and the amount of the payout in doubt.
Or such clubs may be unreliable in more devastating ways, as Sylvia discovered. Sylvia, as described earlier in this chapter, was in an ASCA whose members lent out a good part of the fund to nonmembers, at a high rate of interest. Unfortunately, Sylvia did not earn as much as she expected from the payout of this ASCA. First, when some of her borrowers failed to repay, she had to do so from her own pocket, seriously eroding her profits. Second, just before the payout, the treasurer of the ASCA was robbed and killed on her way back from the bank. As it happened, she was only carrying part of the ASCA members’ money. Sylvia received $246 from the member who was holding the other half of the money, but she had expected to receive twice as much.18
It is not just in South Africa that savings clubs can fail. Nearly half the incidences where Delhi households reported being cheated out of money involved ASCAs. In the Bangladesh sample, Surjo, the youngster from Dhaka, tried hard to stop his sister from joining an ASCA at her factory precisely because he himself had just been part of a 10-man club—a RoSCA—that had collapsed when several members failed to pay in. He lost about $14. As it happened, his sister’s ASCA, which was run by workers who shared the same floor of a factory and were paid similar amounts, worked well. Surjo told us that through the two experiences he had “learned a lot. . . . Now I know what kind of people you should let into your club, and how to run it.” But he didn’t hurry to join another one.19
The dearth of the “right” kind of people to join a RoSCA was a key issue for Delhi respondents. Nasir, as we saw earlier in the chapter, enjoyed well-run RoSCAs, but two of his neighbors said they didn’t have sufficiently trusting relations with anyone in their neighborhood, or even in Delhi, to depend on them to pay their dues. A respondent from another slum said he’d been trying to join a RoSCA for some time and couldn’t find one that would have him as a member. Finally, he met a manager of a RoSCA, who told him he could join only if he agreed to take the prize last. Two of his neighbors were excellent RoSCA members, but they had to travel all the way across Delhi to the meetings each month. Neither felt they would find anything suitable closer to home.
Two other RoSCAs used by our Delhi diarists were put under strain, if not broken, by the diary members’ own failure to pay their dues. The first instance was that of Sultan, a small businessman, who received his reduced prize after struggling to pay for several months. After deciding that he couldn’t continue, he arranged to be replaced by an old client of his who still owed him money. The client was to contribute the balance owed to the RoSCA each month, in lieu of paying Sultan. This move was smartly arranged, but risky if the client hadn’t fulfilled his obligations. Another case involved Mohammed Laiq, who failed to pay his two final RoSCA installments. A year later he had still not cleared these obligations. Shortly before we completed our research, he announced he’d found another RoSCA that would have him in spite of his poor track record. Among the members, he explained, were friends who had been required to vouch for him and guarantee his full participation.
On the whole, though with some disappointing exceptions, the Bangladesh microcredit loans discussed earlier in the chapter worked reliably. As we noted in an earlier chapter, users greatly appreciated their “contractuality”—the fact that workers came to the meetings on time, gave loans on the date and in the amount they promised, and didn’t take bribes. But they did suffer from a second general problem that our diary households encountered: schedules that fit poorly with cash flows.
In part, this is a problem that savings and repayment regimes everywhere battle—the tension between flexibility and discipline. At one level this is a mental battle waged inside the head of the user: we all know we should save regularly, but we also know how difficult it is to carry out our good intentions. We seek external help—automatic payments, accounts with penalties for early withdrawal or missed payments—or we devise mental tricks, keeping the rent money in a special place (the teapot that belonged to grandma) and erecting taboos against dipping into it. These “mental accounts” have been the subject of much recent enquiry.20
But at another level this is a practical matter. In Bangladesh, to keep things simple, the microcredit lenders offered only one loan term—a year—and only one repayment schedule—equal invariable weekly installments. Such a tight schedule is wonderful for discipline—but quite tough on borrowers with very small and very variable cash flows. So in Bangladesh, we found that the very poorest have been either unable to join microcredit schemes, or, having joined, soon leave after failing to complete a repayment on time. These “very poorest” are typically landless farm laborers, who have between-harvest “down” months when very little income comes in. They can pay each week most months, but not in every month of the year. Several of our poorest rural diary households had quit microcredit schemes after such an experience, a few were experiencing them during the research year, and others were reluctant to take a loan for fear of failing.
Sita, the Indian diarist we met earlier in this chapter, had a disappointing time as a microcredit client. She had taken her first microfinance loan the year before we met her, after saving for a few months. The loan of $43, to be repaid over a year, was invested in a grocery store on the advice of the loan officer. Within five months, the store had gone bad and she sold off the stock, purchasing a cow with the $22 saved from it. She continued to repay the loan from her wages (faltering briefly when her daughter-in-law fell ill), and when the local microfinance operation closed toward the end of her loan term, the last two installments were cleared using her compulsory savings. The microfinance institution left the village because there was inadequate demand for its loans. Although, unlike many, she repaid the loan fully, Sita is now convinced she has no use for such loans.21
Sita was unusual in having some bank savings, originating from a three-year-old government handout of $426 that she was given to construct a new house. Of this sum she had spent about $170 on building materials, but most of it was used on the marriage of her eldest son, with a small proportion ($45) put aside in a fixed deposit savings account at a bank (due to mature at $53 after five years). But toward the end of our research she was confronted with two emergencies: the funeral of her daughter-in-law, who had died at her parents’ home; and the deteriorating health of her oldest son, who needed treatment for tuberculosis. Unable to raise enough from neighbors, Sita’s response was to go to the bank to ask to release her fixed deposit six months before its due date. But the branch manager refused. Instead, she used $43 of savings she had been collecting at home for the purpose of releasing her only fertile land from mortgage. Hopes of using the mortgaged land for the next farming season were dashed. So her fixed deposit at the bank remained intact but at considerable cost. It is because of circumstances like this that we saw large lump sums from single instruments so rarely being used for emergencies.
During the year Sita proved that she was able to save and repay loans. She borrowed and repaid three times from neighbors and relatives. She saved continuously in her home, and her youngest son Lalla consistently serviced a debt to his employer through deductions in his wages. And yet because of the mismatch between the products she used and her needs, her saving efforts were in constant jeopardy.
Saving cash with the formal sector (banks and the Post Office), as Sita had done, is far more common among poorer people in India and South Africa than in Bangladesh. India’s contractual and fixed deposit schemes (maturing after five or 10 years) tend to have shorter terms than the 15-year LIC endowments described in the previous chapter but longer terms than RoSCAs (where terms are rarely more than two years). Eleven of our Indian households, or 23 percent, held fixed deposits like Sita, or contractual savings, and only two of these households were from the wealthiest of our three ranked groups. These kinds of long-term saving products (and the savings capacity they mobilize) could provide the foundations for creating secured loans—which use the savings deposits as collateral—with a potential for greater flexibility than the typical nonsecured loans offered by microfinance institutions.
A third limitation of the tools that our diary households most successfully used to form usefully large sums was that the terms are too short, so that savings plans or loans that require multiple years to fund—such as home mortgages or pension plans—cannot be achieved. In the informal sector, where, as we saw, most lump sums were formed, there are natural limits to term length.
For very good reasons, most informal saving devices are time-bound, and a general rule is that the shorter their term the better their chances of working well. Saving-up clubs targeted at particular dates, like a major festival, last for a year or less. RoSCAs are by nature time-bound: their life equals the number of members multiplied by the interval between meetings, and most of those that work well are over and done with in a year, and often less, though of course they may chose to start another cycle. In Bangladesh, Surjo’s sister’s successful RoSCA lasted seven months—14 members times the 15-day interval between payday meetings. In South Africa, Nomsa’s lasted three months, and, in Kenya, Mary’s just seven days. Most well-performing ASCAs also last a year or two at most.22
A short life-span provides a regular test of the health of the device. When the scheme ends and the savings and profits are returned to the members, they either get paid in full or they don’t. The payout acts as an “action audit.” If all is well, members can start over. If not, they can walk away, as Surjo did from his group, having learned not to be in a club with those members ever again. Clubs that continue for long periods are subject to many risks: members may move away, quarrel, or their circumstances may change so that they can no longer participate. Cash left with treasurers can be embezzled. As cash builds up, members, or worse outsiders, can be tempted to try to capture it. Since these are private clubs, not protected by law, recourse when trouble breaks out is hard to find. Better to cash-up and walk away or start over. For all these reasons, it is much easier for Filipina schoolteachers to get together for a few months to fund housefuls of furniture for each other than to get together for many years to fund their pensions.
Much the same is true—and more obviously—of borrowed sums. Informal lenders, whether lending socially or for profit, limit their loans to sums that they can be reasonably sure to recover within a predictable time span during which they expect to be able to keep tabs on the borrower. The microcredit lenders in Bangladesh worked to create a model in which loans were to be invested in small businesses, with loan values calibrated to the capacity of the business to repay it from business surpluses within a short term so that new capital—of a greater value—could be pumped in via a new loan after about a year. But even if this model hadn’t dictated their short loan terms, it is doubtful that they could have risked lending, long-term and uncollateralized, to poor households lacking secure legal identities.23 It is much more reasonable to expect them to explore how to develop long-term savings plans for their clients—an idea we return to in chapter 6.
Like richer households, poorer households need to finance the big things in life. For this, they need big chunks of money. Putting together large sums is, not surprisingly, far harder for the poor. How do they do it?
The first answer is that they do so piecemeal. Large sums are cobbled together from smaller ones: loans are taken, gifts received, savings depleted. Financial tools capable of producing really big sums—simply and in a single place—are rarely there.
But this isn’t a pessimistic story. The second answer is that households use financial instruments to trap and hold the small amounts they can squeeze out of a monthly budget. The poor households whose lives we followed did have room in their budgets to set aside funds for saving or repaying loans, and most used that capacity during the research year. Although balance sheets don’t show many large-scale items, our households did form several usefully large sums each year—sums that were multiples of an average month’s income.
The instruments that helped them leverage their capacity to save into these larger sums were of two kinds. There were the “accumulators” that allowed them to save regularly at fast rates, and the “accelerators” that encouraged them to pay down large loans quickly. The accumulators were mostly, though not exclusively, in the informal sector, and consisted of several kinds of savings clubs. The accelerators were found in the informal, semiformal (microfinance), and to a lesser extent, the formal sectors.
The underlying mechanism was the same in the two kinds of instrument. Both help poor households maximize their budgeting capacity by exchanging usefully large sums for a series of small regular payments. In this way, the act of saving and borrowing often looks quite similar in practice (except, of course, borrowers get hold of their sum sooner). In both cases, the sums can be used for any purpose. Microloans, for example, are by no means always used for—nor repaid from—microenterprise profits.
The accumulators and accelerators are often only part of a process. By patiently using accumulator or accelerator devices, poor households can sometimes put together funds that they then transform once more, buying value-preserving assets like precious metals and real estate that can provide security, not least in old age. In this way, poor households can use the short-term instruments at hand to substitute for the longer-term instruments they lack. However, the shift from a short-term instrument into an asset brings the lump sum briefly back into the hands of the household, putting it at risk of being whisked into providing for another, more urgent, need rather than saved for the longer term. We more often observed funds being accumulated and used within the short term than saved beyond the study year.
Existing financial devices, then, have many positive features. But this doesn’t mean that the poor should be left to make do with these instruments only. Accumulators and accelerators are not always available or reliable. They are not always able to offer schedules that match household cash flows or to be available for sudden emergencies. And their terms are often too short, hindering long-term accumulation.
By building on the established financial habits of poor households, providers interested in serving the poor can begin devising instruments that offer improved, longer-term versions of accumulating and accelerating devices. Chapters 6 and 7 describe ways that this creation of new instruments is already happening.
