Why does the prepaid model work so well and what are the lessons for business model innovation?

Increasingly, employment is becoming ad hoc and flexible. The gig economy and the informal sector share a common characteristic of incomes which are irregular and unpredictable, unlike the timely wages characteristic of formal employment. Both budgeting and planning thus become a challenge when there’s no predictable paycheck to rely on. Expenses are managed against cash flows to minimize volatility, and payments with calender deadlines become a challenge in planning.

It is in this scenario that the prepaid or pay as you go model works so well for the customer, one of the reasons why its ubiquity across the developing world drives the growth of mobile phones. It puts control over timing and amount of money spent in the hands of the user, allowing them juggle voice and data purchases against available cash in hand.

Here are the lessons for business model innovation applicable for a plethora of products and services, drawn from our decade of research into the financial frameworks underlying the operating environment characterized by unpredictability and volatility, and the success of the prepaid model.

Flexibility

The prepaid model is flexible. There is no rigid requirement on the amount that can be spent, beyond the voucher values of each telcom operator, nor are there periodic calender based deadlines such as those in a monthly bill. In Nigeria, traders have been found to top up their phones multiple times a week or even the same day, yet purchasing the smallest denomination of vouchers. High frequency of small amounts is a purchasing pattern that resembles their own cash flow while trading in the informal market. They don’t want to tie up their liquidity in airtime in case cash on hand is required for business, yet their trade is clearly dependent on mobile communication hence the frequent recharges.

This flexibility built into the business model clearly puts control over timing and amounts spent in the hands of the end-user who must manage a volatile cash flow situation.

Seasonality

In addition to the daily or weekly fluctuations in cash flow experienced by gig economy workers or those active in the developing country informal sectors, there are larger variations in income level over the course of the natural year. Unlike the regularity of a monthly salary, irregular incomes rise during peak seasons, such as festivals and holidays, and plunge during low seasons. Developing country economies are more closely linked to the seasonality of agriculture, given the greater proportion of the population’s dependence on farming. Incomes can vary as much as 300% for instance, for tea farmers in western Kenya’s Kisii region. Climatic effects also have greater impact on cash flows, and the current drought in East Africa is expected to depress livestock prices in the coming half year. On the upside, seasonal peaks in consumer durable sales are predictable as the regional harvest timings are a known factor. North India’s post harvest season in late October/November kickstarts an orgy of consumer spending during the festivals and the weddings which take place during this period.

Business models designed to take expected seasonal changes into account can minimize the dropout rate of customers when their income changes.

Liquidity

One of the biggest challenges we have wrapping our heads around when considering more rural or cash intensive economies is that liquidity is not equivalent to wealth, or even purchasing power. While this factor can apply to anyone relying on multiple income streams from a variety of sources, I’ll use the example of a small farmer to explain its importance to the design of business models.

The homestead is managed like an investment portfolio, with different sources of income maturing over different durations of time over the course of the natural year. This is also why control over Timing – frequency, periodicity – of payments, such as possible in the prepaid model, is so critical for the success of payment plans. A smartphone might be purchased after the major harvest of the annual cash crop, but its the daily cash from the sale of milk that would be used for recharges (and other basic necessities). Similarly, a calf may be purchased to fatten against the following year’s school fees.

Negotiability

This leads directly to a factor more relevant to heavily informal economies where variance in systems and structures means transactions are more human centered, depending on face to face communication, trusted references, and mutual compacts rather than legal contracts to enforce agreements. Negotiability of your business model, and its close relation, reciprocity – “the give and take” – is an element missing from faceless institutions that seek to serve this demographic.

This is one reason many prefer to seek solutions outside of formal banking institutions, for example, as their opening hours might not suit the trader’s business hours. In Busia, Uganda, most women traders had established trusted relationships with a mobile money agent, many of whom would show up at the end of the work day to assist the trader in transferring the cash earning safely onto the digital wallet. And, unlike the bank, the telco’s prepaid model allows customers to “negotiate” when and how much they’ll pay within the constraints of far more flexible terms and conditions than most other models.

A farmer has “purchased” this solar panel after coming to an agreement with the shopkeeper. He will pay off the total, over time, as and when he has spare cash, and collect the panel when payment is complete. There is no interest charge. The shopkeeper has put the farmer’s name on the panel but will keep hold of the item.

The greater the span of control over timing and amounts, the greater the success of the payment plan

The prepaid model bridges the critical gap between the predictable formal structures of the large institution and the dynamic challenges of the informal. The bottomline is that the flexibility, negotiability, and reciprocity of the model are more important factors for its success than the conventional understanding of permitting micropayments in advance. Numerous consumer product marketers entering emerging markets experienced this challenge when their micropayment hire purchase models failed customers who might have to miss one or two week’s payments due to illness or other emergencies – their products were repossessed without any recourse to adjustment. Its the rigid calender schedule embedded in a payment plan that is often the barrier to a high ticket purchase than the actual price itself.

None of these factors are insurmountable with today’s technology, and the field for business model innovation for irregular income streams such as those in the gig economy or the informal sector is still wide open for disruption.

Financial Behaviour Patterns Observed Among Households in Rural Informal Economy in Asia

This is the original working paper of the research conducted on rural household financial management, in developing country conditions, pioneering the use of methods from human centered design for discovery, during Nov 2008 to March 2009, aka the Prepaid Economy Project. It was peer reviewed by Brett Hudson Matthews, and I have incorporated his comments into the PDF.

This research study was carried out with the aid of a grant from the iBoP Asia Project (http://www.ibop-asia.net), a partnership between the Ateneo School of Government and Canada’s International Development Research Centre (www.idrc.ca)

The abstract:

The challenge faced by Bottom of the Pyramid (BoP) ventures has been the lack of knowledge about their intended target audience from the point of view of business development whereas decades of consumer research and insights are available for conventional markets. What little is known about the BoP’s consumer behaviour, purchasing patterns and decision making tends to assume that there are no primary differences between mainstream consumers and the BoP except for the amount of their income – pegged most often between $2 to $5 a day.

In practice, the great majority at the BoP manage on incomes earned from a variety of sources rather than a predictable salary from a regular job and have little or no access to conventional financial tools such as credit cards, bank accounts, loans, mortgages. This is one of the biggest differentiators in the challenge of value creation faced by BoP ventures, particularly among rural populations (over 60% of the global BoP population lives in rural areas).

Exploratory research was conducted in the field among rural Indian and rural Filipino populations in order to understand how those on irregular incomes managed their household expenses. Empirical data collected by observations, interviews and extended immersion led us to identify patterns of behaviour among the rural BoP in their management of income and expenditure, ‘cash flow’ and ‘working capital’ and the significance of social capital and community networks as financial tools. Practices documented include ‘conversion to goods’, ‘stored wealth’, ‘cashless transactions’, and reliance on multiple sources of income that mature over different times.

This paper will share our observations from the field; identify some challenges these behaviours create for business and also explore some opportunities for value creation by seeking to articulate the elements that BoP ventures must address if they are to do business profitably with the rural ‘poor’ based on their own existing patterns of financial habits and norms.

The Conclusion:

In sum, it can be concluded that the challenges for value creation can be quite different for BoP ventures interested in addressing the rural markets. From the observations made in the field, we can highlight three key implications for business development. These are:

Seasonality – with the exception of the salaried, everyone else in the sample pool was able to identify times of abundance and scarcity over the course of natural year in their earnings. Identification of a particular region or market’s local pattern of seasonality would benefit the design of payment schedules, timing of entry or new product and service launch, for example.

Relative lack of liquidity – The majority of the rural households observed tended to ‘store wealth’ in the form of goods, livestock or natural resources, relying on a variety of cashless transactions within the community for a number of needs. Conventional business development strategies need to be reformulated to take this into account as these patterns of behaviour may reflect the household’s purchasing power or income level inaccurately.

Increasing the customer’s span of control over the timing, frequency and amount of cash required – Since the availability and amount of cash cannot be predicted on calendar time, this implication is best reflected by the success of the prepaid mobile phone subscriptions in these same markets. When some cash is available, it can be used to purchase airtime minutes for text or voice calls, when there is no money, the phone can still receive incoming calls. Models which impose an external schedule of periodicity, frequency and amount of cash required may not always be successful in matching the volatile cash flow particular to each household’s sources of income.

Permanently seasonal

Photo: Niti Bhan, March 2017, New Delhi

Now Dinesh, the chick (bamboo or rattan blinds and awnings) maker offers a seasonal service in the warmer months, yet his signage and product display seems to imply a permanent storefront.

More likely, it’s permanent during the spring and summer, when the chicks are replaced and installed to protect and cool against the heat of the sun – Delhi’s summer temperatures are known to cross 50 Celsius and tend to average in the mid to high 40s.

If memory serves, he returns each year, thus assuring himself of a permanent yet temporary workspace seasonally, and thus dependably and reliably.

A matter of timing: seasonal opportunities

Temporary stall for festive goods (Photo: Niti Bhan, March 2017)

These stalls full of water pistols and balloons sprouted overnight a couple of days before the spring festival of Holi (March 13th 2017) – these vendors are neither local nor regulars in the market complex. They’re here to offer seasonal products and might even have been invited by the local shopkeepers to provide attractive temporary displays not unlike festival shopping at the mall.

Seasonal opportunities for special offers and custom products are not to be missed chances for a boost in sales. India’s FMCG majors can’t afford to ignore the seasons that guide the cash flow for the majority in the informal and rural economies over the course of the natural year.

Poverty is Dynamic and Flexible, Just like the Informal Economy: Evidence from India

…the concept of poverty today is fundamentally different from that of poverty three decades ago, and that safety nets need to be tailored to meet the needs of a society in transition.~ The Hindu, 2 Aug 2016

When quantitative data provided by the India Human Development Survey (the first large panel survey in India) provokes the academics involved to question their fundamental assumptions and premise of what poverty is, and what it might mean, its a noteworthy moment.

The survey, conducted by the University of Maryland and the National Council of Applied Economic Research (NCAER) for the same households at two points in time, viz. 2004-05 and 2011-12. Their analysis has led them to say:

Once we recognise that poverty is dynamic in nature, and that as per our conventional definition of poverty, poor households may move out of poverty and the non-poor may become poor over a period of time, we are forced to question the veracity of our fundamental assumptions about poverty. Perhaps poverty occurs not simply due to the accident of birth or as defined in terms of where and in which family people are born, but also due to the accident of life caused by the occurrence of disease, disability and unemployment. Achieving this recognition entails a complete transformation in our mindset.

I will leave them to their explorations from the perspectives of their disciplines, and explore the broader implications of their findings.

A few years ago, as part of my discoveries from more qualitative user research in the field on the informal sector’s financial context and operating environment, I had had my insight on the dynamic nature of poverty as it was conventionally defined.

It was when attempting to clearly distinguish between patterns of cash flow in the formal vs the informal economy, using the concept of the degree of control granted to the end user over the variables of time (duration, frequency, periodicity) and money (amount, cash or kind), that it struck me what kind of difference does control over timing mean for money.

That is, there is a complex value processing underneath each of the decisions on allocating available cash money, particularly in rural areas where cashless transactions can tend to be more common.

When one can control the timing of one’s payments – such as the advance purchase of airtime minutes to use a mobile phone – one’s income could be called dynamic. Within any particular set of calender based time eg a week or a month or a quarter; a vast majority of the lower income bracket cannot predict their total cash income nor feel confident enough to claim it. It can be affected by seasonality prevalent in their region, or it can be purely random volatility, one’s workshop burns down in an accidental fire.

Static income is that which is stuck, such as a fixed salary paid every calender period, regular in frequency, amount and periodicity.

As cash flows tend to be volatile, fluctuating with seasonal influences, chance, and the vagaries of daily life, those whose incomes are not as predictable as a periodic paycheck, are more often than not unable to clearly state (or even know) their monthly or weekly income.

That is, even as data gurus in development banks seek to segment people into neatly defined ranges such as $2 to $4 a day or whatever, it is neither a given that people will remain within this range over the course of the natural year, nor can it be a reliable and consistent indicator of their income level – Below Poverty Line (BPL) is the concept used in The Hindu’s article above.

Therefore, if the survey studied households in an agricultural region during its fallow season the first time, and then went back to study the same households during the post harvest season the second time, that simple little factor of calender time alone can create a difference of as much as 100% to the incomes being claimed during that period. If the study does not follow up the income question to ask if there was seasonality in their cash flows over the course of the natural year and if this question was being asked during the high season or the low season.

When I did the original fieldwork for the prepaid economy project on an IDRC grant, looking at the rural household financial management behaviour in rural India, Philippines, and Malawi, I found that depending on the local region’s primary cash crop harvest patterns over the natural year (say monsoon to monsoon, or Christmas to Christmas) the entire local economy felt the impact of the difference in cash flowing through their ecosystem during the high and the low season. Or, the wet and the dry season.

It was not the naming of the seasons that is important. It is the ability of the people to forecast known fluctuations in their income streams based on patterns recognized from experience and local wisdom. Within the context of an environment of uncertainty and volatility, it offered them some anchors for planning and financial management.

Given that the vast majority of the poor in the developing world, like in India and across Africa, are dependent on irregular, often unpredictable cash flows from a variety of sources, in an environment of higher risk and uncertainty, their incomes can confidently assumed to be dynamic, rather than a static salary.

And the dynamic nature of the informal sector precludes conventional classifications and categorizations of poverty, especially by any stated amount of money mapped against a particular duration of calender time. Time and money are themselves the uncertain elements requiring flexibility built into the systems if they are to work properly in this operating environment.

Thus, I can confidently state that what the Indian data is finally providing the evidence for are the findings from my qualitative research among the same segments of the population, using design ethnography methods. That is, we now have the quantitative data to support the insights derived from the qualitative research.

Full Stop.

Seasonality as an element of contextual planning for emerging consumer markets

Growing up as a Hindu expat in multicultural ‘West Malaysia’ of the 1970s and 80s, it was a matter of course that every festival would be a big occasion. We had Christmas in December, and Chinese New Year soon after, to be followed by Hari Raya (Eid) and Deepawali – each of them deserving of TV specials and decorations on the streets.

Seasonality of cash flows and income streams in the informal and rural economy translated in the urban areas as festivals triggered a boom in consumer sales. India’s formal economy still keeps watch on the onset of the annual monsoons, as those rains will have documented impact on their 3rd quarter sales in the peak festival season of October and November, leading into the wedding season.

In Eastern Africa, this seasonality is seen, among other things, in the lives of pastoralists and livestock farmers. As Eid Al Adhar approaches in a few days, livestock sales for the annual sacrifice are reaching their peak. Trade in meat is one of the staple income sources in the arid lands and the Port of Mombasa is one of the keys to the distribution networks.

The livestock trade to the Middle East accounts for 60 percent of Somaliland’s gross domestic product and 70 percent of its jobs.

This, however, is changing, as the Port of Berbera will soon receive millions of dollars of investment in improved infrastructure. The element of seasonal cycles over the course of the natural year, however, will not change. And this is worth noting for those considering the emerging consumer markets in the developing world.

Beyond word of mouth, however, it is hard to get a proper idea about the economic impact of Ramadan. Perhaps because of sensitivities around dealing with a religious institution, international organisations such as the World Bank, International Monetary Fund and United Nations Development Programme have not conducted research on the precise economic impact of the custom.

FMCG majors already feeling the pinch of shrinking domestic markets are finally taking note of this entire opportunity space. In Indonesia, Unilever, Beiersdorf and L’Oreal are making halal face creams and shampoos to court Muslims as sales in Western markets taper off.

There are patterns of trade around major holidays in each region, be it Chinese New Year or Dussehra, and the informal sector prepares for, and relies upon, these expected bumper ‘harvests’ in their cash flow. It will be interesting to watch what happens in the context of the African consumer market as the Asian giants begin to eye it seriously as the last frontier for significant growth.

Part 2: The Observations made during original research on rural economic behaviour

narrow_prepaid
One can roughly consider the relative income (or wealth) across three regions where observations were conducted on a continuum where the Indian village was the ‘wealthiest’ while the Malawians were living closest to the edge. However, on synthesizing the combined data collected across geographies, patterns of financial behaviour emerged that showed similarities of intention and goals.

For example, non-perishable food grains such as wheat in India or rice in the Philippines were considered a form of wealth that could be stored, acting as savings or insurance. A portion of the harvest would be held back, to be either sold on demand for cash, over the course of the year or as a source of food. Wealth was also stored, as security, for the longer term, in the form of silver ornaments (in India) or as an investment, in the short term, as livestock – pigs, chickens or a milch cow.

Also, people rarely held on to money in the form of cash for any length of time, for the most part due to lack of access to banks and/or the high cost of maintaining an account proportionate to their incomes. Available cash was usually converted to “kind” – either goods or livestock- the choice of which reflected careful prioritization. These tangible purchases then acted as financial tools depending on their “convertibility”-

long-term security (silver);
planned savings (buying building materials on a piecemeal basis over time until a house could be built);
insurance or a “cushion” against shocks (a pig that could be sold to raise cash or eaten as food) and finally,
investments (milk bearing cow, young piglets to rear to maturity, culling high margin ‘fighting cocks’ from chicks).

Cashless transactions, thus, were frequently observed. These behaviours were most complex in India; where a sophisticated mechanism allowed a group of farmers to negotiate the annual retainer for the services of a carpenter in the form of a number of sacks of wheat to be paid during the harvest and the local shops would set a ‘currency’ conversion rate of a kilo of wheat to the rupee to be used for buying sundry provisions. The shop that insisted on cash only transactions priced its goods about 10% cheaper than the rest. Barter was far simpler in Malawi, where a mobile phone could fetch its equivalent price in goats.

Here, it must be noted that very often each household’s resources such as a store of fuel (cow dung in India; firewood elsewhere), chickens or a kitchen garden and assets like milch goats or cows, would be pointed out with pride. For their possession implied an independence from cash money – in almost every interview, people would emphasize how little they needed to purchase in the store or nearest town for their daily needs as they were self sufficient in these demonstrated requirements. Often it would be added that in a city, you had no choice but to purchase everything you needed.

Thus the use of purchased resources were optimized for maximum cost/benefit and their use extended as much as possible before replenishment. For example, if a household had access to cooking gas, they would still use firewood or charcoal for foods that took longer to cook while the more expensive fuel was used for foods that cooked quickly.

In the Philippines, cashless transactions were rarely in the form of goods but tended to involve time or physical labour, primarily as a form of social capital in the community. These complex webs of the rural community’s social networks of trust were obvious in the patterns of sharing and cooperation seen in every country. Groups would invest and save together, for example, the extremely sophisticated cooperative ladies lending circle which had expanded over time to include the services of a local bank in India; or the beekeepers cooperative in Malawi where half the annual profits were saved in a common account while the other half was equally shared.

In addition to the behaviour patterns mentioned above, an external factor was observed to be of great significance in the management of rural household expenses. While it naturally differed in timing and reason from region to region, every household and profession could predict, within reason, the ebb and flow of income based on the seasons of a natural year. In fact, many other observed behaviours were often directly linked to these expected peaks, such as the harvest season, and lows, for example the dry season when fields lay fallow. This pattern of expected ups and downs or seasonality in income flow was seen to affect even those who were not directly involved in agriculture, as the local economies were closely knit and interdependent.

Note: This blog was begun as a way to publicly share my thoughts during fieldwork, so much of the raw data and immediate observations are available under the category “user research” as well as blog posts written during January 2009 to April 2009 as seen in the archives available on the right hand sidebar.