Monday, October 24, 2016

Mobile Phones, Insurance and a Funeral: A Closer Look at South Africa’s Mobile Micro-Insurance Market

By IMTFI Fellow Christopher Paek 

About halfway through my fieldwork in Cape Town, South Africa, tragedy befell Goodwill Nxusani for the second time. He had been one of my key sources and interlocutors, connecting me to local residents of his township, Khayelitsha. Earlier that year, his grandmother had passed away and he was generous enough to invite me to her funeral. Just a few months later, he received word that his father-in-law, who lived in the Eastern Cape, had also passed. As the only income-earning household in the immediate family, Goodwill’s family was responsible to pay for the whole funeral.

A traditional Xhosa funeral in Khayelithsa, South Africa (Photo credit: Christopher Paek)    

Funerals are sacred among the Xhosa. Whether poor or rich, families do whatever they can to ensure that their beloved kin are sent off properly in death so that their souls can join with the ancestors. Goodwill’s father-in-law, the male head of household, was to be honored, as customs dictated, with a slaughtered cow. Since he died near Cape Town, transportation would also have to be arranged so that his body could be returned to the Eastern Cape, a common story for many Xhosa who had migrated to the Western Cape in search of work.

Between the transportation costs, the livestock, food, and the funeral ceremony itself, Goodwill faced a price tag of R42,040 ($3,123). If Goodwill had spent every rand he earned, which was R2,000 ($148) per month, it would still take him nearly 2 years to fully pay for the funeral. Fortunately, Goodwill was among the lucky few who had taken out a funeral insurance policy that covered R14,000 ($1,040) of the cost. Still, the death of his father-in-law posed a considerable financial burden on his family. As he broke the bad news, he informed me how he and his wife had gone three days without food in order to pay the first installment on the cow.

Economists and insurance professionals see Goodwill’s story, which is fairly common in communities across South Africa, as a story about financial risk. In their view, the financial toll imposed on a grieving family can be alleviated by finding ways to extend financial services into low-income spaces…no easy feat. Insurance, widely considered a grudge purchase, is a hard sell to even middle-upper class people. How do you convince the poor to spend what little they have on insurance?

South Africa is unique in this regard because demand for micro-insurance (insurance products designed for low-income clients) is high, driven by the cultural imperatives placed on funeral rituals. Of the nearly 62 million lives insured by micro-insurance on the African continent, South Africa alone accounts for more than half of these lives, making it one of the world’s largest micro-insurance markets.

While microfinance enthusiasts might see these numbers with unbridled optimism, there is an important caveat to consider. Micro-insurance sales in South Africa are almost exclusively driven by funeral insurance policies. Other products including life, health, and asset insurance have found no success in the low-income market. Many are hopeful that exposure to high-quality funeral insurance products can serve as a sort of Trojan horse into this market, but this is yet to be seen.

As might be expected, building profitable micro-insurance markets presents a number of challenges, especially the need to achieve scale, since the sustainability of insurance operations relies heavily upon building a sizable risk-pool. Fortunately, the advancement and proliferation of technology across the developing world, particularly mobile phones and its networks, have been a game-changer for many industries including micro-insurance. Since mobile penetration is deep in South Africa (mobile phone subscriptions per capita stand at 1.47, according to the World Bank), insurance companies have partnered with mobile network operators (MNOs) to tap into this expansive distribution network. Insurance products that are sold through and with mobile operators are commonly referred to as mobile insurance, or m-insurance for short.

By overlaying their operations upon a mobile infrastructure, insurance companies have been able to generate efficiency gains across the entire micro-insurance value chain from product design, marketing and sales all the way to enrollment and claims administration. From the MNO perspective, m-insurance is an appealing product insofar as it stimulates average revenue per user (ARPU) and reduces churn, i.e. increased loyalty/retention. And for the end-client, efficiency gains translate into affordable premium rates that compare favorably to traditional micro-insurance products or even their informal sources of insurance coverage. Sensing the market opportunity, insurance companies and MNOs launched several varieties of m-insurance products including (but not limited to):

1. Loyalty Based Models- Clients receives “free” coverage paid for by the MNO if the client behaves in an incentivized way (e.g. more airtime usage, data purchases, etc.)
2. Airtime Deduction Models- Clients can make their premium payments with their airtime balance.
3. “Dumb Pipe” Models- The mobile phone is used only for data capture, enrollment, and communications functions, but not for premium collection/payout.

A non-exhaustive typology of m-insurance products on the South African market    

It would seem, then, that South Africa, with its high demand for micro-insurance, a corporate commitment to m-insurance, and high levels of mobile penetration, would be fertile ground for the wide-scale uptake of mobile-based micro-insurance. But it came as a surprise to many in the industry when the anticipated m-insurance market failed to achieve scale. What happened? And what does this mean for other financial service providers who are looking to break into the low-income market through mobile channels?

The research I conducted in Khayelitsha, a large township outside Cape Town, indicated that a major reason why this market failed to materialize had to do with trust. Even longtime micro-insurance clients who were well familiar with how insurance worked, would not trust using their mobile phones to conduct financial transactions. What drove this mistrust?

To answer this question, it’s important to understand clients’ experience with m-insurance within a much wider context of mistrust in which they live and operate. For township residents, in particular, this environment is typically characterized by high crime rates, lack of formal legal recourse, a lack of consumer advocacy and education, countless experiences with money/phone scams, and high unemployment. Anthropologist Erik Bähre observed how, in the midst of such a volatile environment, township residents would seek out and form “islands of trust” where they felt safe enough to keep/grow their money (i.e. informal financial mutuals).

Filtered through this perspective, it’s useful to see m-insurance products as operating outside the boundaries of these islands of trust. M-insurance was instead interpreted through a lens developed and used over time to guard against fraud. For example, many respondents dismissed m-insurance because of their past experiences dealing with phone and money scams. When they come across so-called “free” insurance coverage (i.e. loyalty-based m-insurance), they are understandably skeptical.

What may have been the most unexpected finding was the extent to which even very poor clients were willing to pay a higher premium to deal with insurance sales staff face to face. When presented with an m-insurance product that had a stronger monetary value than traditional retail insurance, clients often expressed how important it was to them that their premium payments and claims were being administered in an office. An office is tangible, it can’t disappear in the night; it is, for lack of a better phrase, Bähre’s “island of trust.”

A funeral m-insurance product. 
A partnership between an insurance company, Hollard 
and a clothing retailer, Pep (Photo credit: Christopher Paek)    
Among m-insurance developers, there is an on-going debate as to the virtues and drawbacks between “high-touch” products, which incorporate sales agents into their models and “low-touch” products, which are typically passive models that eliminate sales agents in order to lower cost. Results from this project seem to suggest that at least initially, a more high-touch approach is required to first develop trust, especially in environments where the use of mobile phones to cross-sell financial products have become synonymous with fraudulent activities.

A related example may reinforce this point. When ATMs were first introduced into South African townships, initial reports suggested that there was widespread mistrust among residents. It took concerted time and effort—i.e. bank tellers would walk through each step with individual customers again and again—for clients to eventually trust ATMs enough to deposit their hard earned cash. Examples like this demonstrate that trust in m-insurance products can eventually be earned, but that an initial investment in time and financial resources may be required to do so.

As this research shows, efficiency, convenience, and price are necessary but not sufficient factors in building a successful m-insurance market. If the trust gap can be overcome, insurance companies may be in a good position to fully leverage the potential of mobile phones and networks to deliver financial services at a meaningful scale.

Read Christopher Paek's Final Report

Monday, October 17, 2016

Object-Centered Focus Group Discussions: Stimulating Conversations On Mobile Money Practices and Culture

By IMTFI Researchers Milcah Mulu-Mutuku (Egerton University, Kenya) and Castro Ngumbu Gichuki (Nanjing Agricultural University, China)

Conversations on personal financial practices are sensitive and many times difficult for researchers to actualize in the field. As many researchers have documented, personal financial practices are private. Getting people to talk about them often times requires great effort. Adding culture to these conversations creates a further challenge that seems almost insurmountable. Yet, even with all of these challenges, it is essential to understand how culture influences mobile money practices if appropriate policies and programs are to be put in place.

Participants during the Focus Group Discussion
Our IMTFI sponsored research involves determining the influence of mobile money on women micro-entrepreneurs’ control of productive resources. In many communities in the developing world, control of productive resources is a subject deeply entrenched in cultural practices rooted in money and land as the preserve of men. As we are realizing through our research, control also depends on the context of resource use. In this era of technological advancements, a slight change in external factors surrounding productive resources alters significantly the way control of these resources is experienced by those involved. For example, the use of mobile money significantly impacts the dynamics of resource control. However, this depends largely on the type of resource (e.g. money, land, information) under consideration and the importance attached to it by the people concerned. Money is probably the most highly rated productive resource in terms of its importance by our target group of women micro-entrepreneurs. How this resource is controlled and managed has a direct impact on their business performance.

Stimulating conversations on money, control and culture

To get women micro-entrepreneurs conversing on these matters, we designed two charts: a static one (Fig. 1) which we used to prompt participants to think through their financial practices (without intervention on our part) and an interactive one (Fig. 3) which required participants to perform some activities. Our approach was inspired by the IMTFI Fellows workshop held at the IMTFI Insight and Impact Conference on April 22, 2016 and by object-centered interview methodologies featured in IMTFI’s Consumer Finance Research Methods Toolkit (the toolkit is free and available for download here).

Fig 1: Mobile Money Services Chart
The Mobile Money Services Chart (Fig. 1) aimed to stimulate conversations on mobile money services utilization at personal, family and business levels. The participants were expected to view a representation of a mobile money service or product on the chart and then rate the usage of these services and products. We then used a flip chart to record the responses (Fig. 2). Using this chart, we explored the benefits and challenges of using these services as experienced by the participants. The main benefits of services frequently used included autonomy and time saving while major challenges to uptake and use were interference with family relations due to practices of secrecy and disclosure related to the personal identification number (PIN). 

Fig 2: "Services used and
not used"
The Productive Resources Control Chart (Fig. 3) aided us in discussing women micro-entrepreneurs’ control over productive resources and the influence of mobile money services on control and decision-making processes. Participants were asked to place color-coded sticky notes or plasticine (modeling clay) next to the person who makes decisions about the acquisition, management, and use of various productive resources. Choices included self, husband, brother/father, and friend(s).  

After the exercise, we counted the sticky notes or plasticine mounds to determine who had the highest control of a particular resource, which we then discussed with participants. We also discussed how mobile money has influenced this control and explored gender dynamics in table banking group activities in this era of mobile money use. Interestingly, fathers and brothers no longer interfere with women’s control of productive resources, whether single or married. In addition, mobile money services have provided women with a discreet method of controlling finances.

Fig 3: Productive Resources Control Chart

Our experience in using Object-Centered Focus Group Discussions

This being our first time to use such a technique in data collection, we did not know what to expect. Much as the idea of using objects in Focus Group Discussions excited us, and of course we were looking forward to it, we were alive to the possibility that our ability to communicate to the participants through these charts might not work as we hoped. However, our anxieties were immediately settled as soon as we began the first session and observed the respondents’ excitement. Discussions were lively, diverse, engaging and quite informative. This came as a surprise to us because we did not realize beforehand that visualizing mobile money services would trigger such diverse thoughts in the minds of the participants and therefore diverse conversations. One chairlady of a table banking group summed up her experience in the discussions, remarking, “na leo mmetuchangamsha kweli kweli! Tumejifunza mambo mengi leo!” (You have really excited us today! We have learnt many things today).

Mobile Money Services Chart in use
Much as we intended to collect data for our research, we also ended up raising awareness about available mobile money services. Participants discussed some mobile money services depicted in the first chart that they never thought were meant for them. One lady pointed at the Pesa Point ATM booth image and remarked, “hii huwa naiona town kila wakati ni kienda lakini sijawahijua inaweza kunisaidia”! (I always see this one every time I’m in town but I have never known it can assist me). Such remarks opened our eyes to the role of such data collection techniques in educating participants on matters otherwise taken for granted, and in creating awareness about services and products.

This experience was exciting and enriching for both researchers and respondents. Within the context of the activities, it was easy for respondents to talk about their personal financial practices and provide information, which under normal circumstances might be difficult to share. For instance, one participant shared with participants about her secret business money saving practices. In order to avoid interference and misuse of business money, she “hides” it in an M-Shwari Deposit Account, a paperless micro-savings product (offered by Safaricom in partnership with Commercial Bank of Africa and operated through M-Pesa). She reasoned that it is easy for her husband to track her savings on her M-Pesa Account but not so easy with her M-Shwari Deposit Account. This way, she is able to control her business finances without interference. However, she has never used the M-Shwari micro-credit product. She finances her business activities through her table banking group and through other micro-finance institutions such as Small and Micro-Enterprise Program (SMEP).

Productive Resources Control Chart in use
Almost all participants indicated they do not use the M-Shwari micro-credit product to finance business activities in spite of the favorable facility fee (compared with micro-finance institutions’ interest rates). Participants gave two major reasons for not utilizing this service. One, the amount of credit that can be accessed through M-Shwari depends on points earned through the usage of a M-Shwari Deposit Account, M-Pesa and other Safaricom services. These being the ‘base of the pyramid’ customers, earning points that would guarantee a sizeable amount of credit does not come easy. Much of the time, the amount of credit they can access through this service is too little for their business needs. Two, though the service providers have ensured that information on terms and conditions of use for the M-Shwari micro-credit product is in the public domain, participants reported lack of information about the product, making them apprehensive of the service. Many participants expressed a desire to understand how Safaricom calculates the points that determine the amount of credit they can access through the M-Shwari micro-credit product.

Concluding remarks

All in all, our experience in using Object-Centered Focus Group Discussions taught us that this is an excellent way of collecting data on sensitive topics. However, researchers must be alert and sensitive to remarks coming from the participants, which can be probed to yield more information and reveal unexpected details. In addition, depending on the topic of discussion, researchers need to be careful not to be sidetracked by the excitement of participants as they discuss their own subjects triggered by the images and writings on the charts.

Read Milcah Mulu-Mutuku and Castro Ngumbu Gichuki's final report here

If you would like to leave feedback about this blog post and the Consumer Finance Research Methods Toolkit, see here.

Erin B. Taylor and Gawain Lynch, with Ursula Dalinghaus (2016). Consumer Finance Research Methods Toolkit. IMTFI.