The Kenya Financial Diaries: 4 Flaws in the Social Network

The Kenya Financial Diaries was led by researchers at Digital Divide Data and Bankable Frontier Associates and funded by Financial Sector Deepening (FSD) of Kenya and the Bill & Melinda Gates Foundation via the GAFIS project. One of the largest studies of its kind, the project tracked the detailed cash flows of 300 low-income families for a full year, shedding new light on how the poor manage money in the context of a vibrant, dynamic financial service market. This is the first post in a two-part series. Read the second.

Since October Monicah had been growing increasingly sick. The 36-year-old mother of four had been staying home with her children after the recent birth of her son. Now, she was getting tired and losing weight. She couldn’t keep down food. In and out of the hospital, she waited for a diagnosis that only came in late November: a throat tumor. It would cost KSh 23,000 (about $260), to remove. Isaac and Monicah had already used up their savings; Isaac, a fisherman, even sold two rooms they were renting out to pay for the admission fee and x-rays that had gotten them this far. Where would they get this money? Monicah was discharged from the hospital and told to come back when she could pay for the surgery. She went to rest at her mother’s home. On Boxing Day, she died.

Immediately, there was an outpouring of gifts and contributions from friends and relatives, within a single day surpassing what was needed to pay for the surgery that might have saved Monicah’s life. For Isaac, his social network worked, but too late to save his wife. Isaac reflected later, “Wakati ukienda ni kama moshi haurudi tena,” or “Time is like smoke. Once it is gone, it does not come back.”

Observing the cash flows and life events of 300 low-income Kenyan families over the course of one year, the Kenya Financial Diaries project has provided us a new window to better understand the financial behaviors and constraints faced by the poor. In many instances, we have been astounded by how powerful informal risk sharing can be as a coping mechanism. But, in too many cases, like Isaac and Monicah’s, we have seen the social network fall short. And in those areas of weakness, we find opportunities for the financial sector to offer new kinds of solutions.

A barber cuts hair A barber cuts hair

Photo Credit: FSD Kenya

Four Flaws

1. Networks work, but too late. As Isaac and Monicah show, the network can work, but often not in time for urgent needs. Why the delay? There could be information asymmetries that make givers less certain about the urgency of a need. Givers may be unsure of whether their gift will be used for the intended purpose or whether it will be effective in saving a person’s life. A death on the other hand is clear and verifiable. The immediacy of need is readily apparent.

2. Networks are exclusive. Not everyone has access to a social network that can support them through a time of struggle. Out of the 300 households in our study, 11% received no contributions from their social network over the entire year. For 30% of households, these contributions were relatively small, amounting to less than 5% of total income during the study. And, men tend to benefit from these kinds of contributions much less than women. Fewer than half of men in our study received any contributions from their social networks, and when they did, it was in much smaller values than women. Whether that stems from men’s pride or networks’ bias is unclear, but it does mean that more men are on their own when faced with difficult circumstances.

3. Relying on networks merely shifts burdens. We found some evidence that social networks are redistributive rather than simply reciprocal. Net givers tend to be better off than net receivers. But, half of the net givers in our sample still fell below the $2/day poverty line. When they give in a crisis, it can wipe out wealth that might have otherwise helped their own households invest.

4. Networks are limited in size and can be inadequate. The networks of the poor are finite in size and often full of other poor people with limited giving capacity. Sometimes the size of a need far outstrips the capacity of the network to meet it. Four-year-old respondent Robert, for example, needs heart surgery that doctors told his family would cost KSh 500,000 (about $5,880). His family called on all of their friends and relatives for help and managed to raise just KSh 15,000 (about $176), which was only enough to have his condition evaluated. The cost of the surgery represents more than five years of earnings for Robert’s entire household – more than they could dream of raising among their family and friends.

Reinforcing Networks

Recognizing these limitations, we see opportunities for financial service providers to reinforce social networks, by helping them work better and providing complementary options that can help alleviate some of the strain that networks are often forced to bear on their own.

1. Reconsider the “market.” Given that slightly better off relatives and friends are the ones who often bear the cost of risk events for their relatively worse off kin, it may be worth considering marketing financial services—particularly those that help manage risk—at the social network givers rather than only those most likely to experience the risk directly.

2. Help iron out information asymmetries. By producing official “bills” that can be shareable with the social network, providers might help those in need communicate to their networks the urgency and scale of different kinds of needs and even enable direct payments to providers. This might generate faster and bigger reactions, particularly in health crises that require urgent action.

3. Continue providing options for the excluded. While all households need some complementary financial tools for coping with risk, those who cannot rely on their social networks for support are particularly vulnerable and might be a particularly ripe market segment for savings, loan, and insurance products that make them better able to manage risk in their lives. In our sample, it appeared that urban households and men were particularly likely to fall into this segment.

4. Larger pools of risk sharing for the biggest risks. There are some needs for which the social networks of poor people may never cover, like Robert’s heart surgery. For these kinds of very large, but lower frequency needs, there is a particularly strong need to pool risk at a higher level.

It is not easy to get any of that right, but a more grounded understanding of the realities of poor families at least can help us move in the right direction.

Join Julie Zollman, a Senior Associate at Bankable Frontier Associates and the Principal Investigator of the Kenya Financial Diaries, for a presentation and discussion about the Kenya Financial Diaries at CGAP headquarters on Thursday, September 9. More information is available at


28 August 2014 Submitted by Nambuwani Wasike (not verified)

This is an interesting piece, considering the role of social networks in Kenya's social fabric. Right here, lies a case for investigating Network-2-Person (N2P) transactions in group initiatives in Kenya!

29 August 2014 Submitted by Dr S Santhanam ... (not verified)

A new look at the micro level fund management of low income group families. But a social network more often would not be the solution provider. It would be better if they are connected to an MFI and more preferable if they are clients of an MFI or made mandatory to become a client of an MFI. An experiment on similar lines (with suitable guidelines and funding by agencies like Bill Gates Foundation) with an MFI as the project implementer who may be allowed to decide whether a loan or a grant is needed to select families for various financial needs over a year would help to understand the inside financial story of such vulnerable groups. Of course, the identified sample families would need the hand holding in maintaining diaries and recording both financial and non-financial events on a day to day basis. If needed, I can design a proposal for the researchers.

Add new comment