Mobile Money’s next challenge: Enabling the development of a cashless ecosystem

Equitel billboard, Nairobi Kenya (Photo: Niti Bhan Jan 2016)

The latest GSMA State of the Industry report on Mobile Money is out this month and the numbers look great in the developing world.
The report frames the industry’s next challenge as the need to grow the platform beyond the basics of airtime purchase and person to person transfer.
Here are my concerns, starting with the very first sentence – “to convince customers to actively diversify their usage patterns.”

This is where there is a critical need for MNOs to segment their userbase prior to designing fresh approaches to increase adoption and build an ecosystem. According to the report, only a few MNOs have data on urban vs rural, much less on gender.

The report’s fashioning of the data available into the form of an “average user” will hinder the progress more than it will help. Look at the geographic spread across widely varying economies, there’s no such thing as an average user when it comes to a tool closely related to one’s patterns of cash flow and income sources. Usage patterns reflect cash flows – why else would the prepaid business model be dominant in these same locations?

The hard work of disaggregating the information into region specific customer profiles must be done if solutions are to work effectively beyond teh basics of P2P transfer and airtime purchase – mobile money’s equivalent of a phone call and an sms.

Many of the reasons why its important to segment by rural/urban, and the proportion of users in the informal sector and on prepaid subscriptions are covered in my old posts on Google’s BebaPay fiasco – a smartphone app enabled NFC solution for cashless public transport payments introduced in Kenya a few years ago.

Economic ecosystems, particularly those with a heavy dose of the informal sector, and closer links to rural hinterlands, such as those common in sub Sahara, will need to be mapped out and understood before interventions can be designed to lower barriers to adoption. These use cases may not be plug and play components or readymade low hanging fruit, as imagined by the writers of this report. They need grounding in the context of the existing operating environment – formal or informal, urban or rural – and, the characteristics of the informal and rural economies, depending on the segment.

The end of the global middle class: A more frugal world?

The past half decade‘s worth of financial crises and increasing scarcity of resources have led to an increasing equalization in the global water level. Instead of the high tide that would lift all boats, the leveling off of growth is leading to an entirely different equation of purchasing power parity. Tomorrow’s equilibrium seems to imply a more frugal world. ~ Niti Bhan, 2012

I wrote this concluding paragraph just over 3 years ago. Today, I look at research from Pew that informs me the great American middle class has declined by half. An article on the Indian middle class claims they’re actually the world’s poor. And the mythical African middle class emerges, floats and sinks, sometimes all at once in the same article.

Water has found its level, and its barely staying afloat.

If indeed the global demographics are changing such that what was formerly considered the “middle class” by the metrics of the day do not apply anymore, would it not make more sense to rebase and then assess who is in the middle than to go chasing the golden children of the boom years long past?

Or, one could just stop looking for these unicorns everywhere and take the trouble to study the people who are the majority in these markets.

Either way, what was is over and what’s emerging is more frugal world with thinner wallets, fewer bank accounts and propensity to pinch their pennies. The data demonstrates it clearly enough.

Floating Upwards: The Bottom of the Pyramid Segment is No More

Pew Research Center’s latest results on global income distribution show some rather large shifts among the lowest income segments.

The Bottom of the Pyramid or Base of the Pyramid (BoP) segment, defined as those who live on less than $2.50/day has just lost a significant percentage of the population. While one can quibble that $2.50 is greater than $2.01 and thus they are still there in the low income segment, you’ll recall that the BoP segment was originally $2/day in its heyday, estimated to be 4 billion strong in the old days.

The low income segment, as defined by Pew, ranges from $2.01 to $10 a day, which overlaps with the African Development Bank’s ‘floating class’ and the ‘lower middle’ class. Lets look at the ‘floating class’ so as to cover all bases when it comes to the income range of people whose cash flows are irregular and unpredictable. One day might be only $2 but the next could be $10. This daily amount, it floats.

From the Pew report:

This study divides the population in each country into five groups based on a family’s daily per capita consumption or income.³ The five groups are labeled poor, low income, middle income, upper-middle income, and high income. Of the four thresholds that separate these different income groups, two are especially important to keep in mind. The first is $2, the minimum daily per capita income that must be exceeded to exit poverty.⁴ The second is $10, the threshold that must be crossed to attain middle-income status. The thresholds are expressed in 2011 prices and 2011 purchasing power parities.

700 million people exited poverty.

They’re not rich yet nor middle income, but their upward mobility and emergence is undeniable, and their consumption is increasingly visible. They are part of the reason why African statistics on the middle class market are so volatile. Obvious and visible consumption is taking place yet few fit the traditional description of what an emerging middle class segment should look like.

Aspirations change even one rung up the ladder

Income is not the only reason for my confidence in making the claim that the era of the Bottom of the Pyramid, the BoP, is over. Along with the few more shillings or rupees a day, comes a shift in mindset. Suddenly, from worrying non-stop about the next meal, we’re thinking about the possibility of investing in a motorcycle, or a cow. Sending the smart one to high school or taking computer classes.

Some old habits remain as the need for frugal buyer behaviour hasn’t changed with a few more dollars a day, and we’ll still see purchasing patterns influenced by cash flows and informal retail. But the dreams have grown and changed.

Once this happens, even if there might be dips in cash flow, one’s worldview is very different from those who are still struggling to survive. We learn to make do if there’s a tight patch, but we don’t go back to living like “the poor”. And it is this transformation that has broken the back of the BoP forever.

Your product or service for the BoP might now be too small for their newly expanded hopes and dreams and ambitions.

Segmenting the African Middle Class without dollar figures

Continuing the thinking from my previous post on the various attempts to size and value the potential of the emerging African middle classes based primarily in dollar figures, I thought to take a step back from income data to see if I could approach the challenge of segmentation in a different way. Below is the chart estimating the size of the original emerging African middle classes as posited by the African Development Bank back in 2011.

That is, rather than simply segmenting by range of daily expenditure i.e. $2 to $4 or $10 to $20 a day, what if we took a closer look at the reasons behind the spending and segmented by consumer mindset and buyer behaviour. After all, given the size of the informal sector in the majority of African countries and the percentage of population relying on irregular income streams from a variety of sources, few can confidently expect to spend exactly $4 each day. There might be times of abundance when hundreds of dollars may be available, and big ticket items purchased like colour television sets, offset by times of scarcity when one might just be making ends meet. Variability in cash flow is an inherent characteristic of entrepreneurship, regardless of income bracket or revenue sources. Furthermore, we can add geography as a factor, since urban expenditure is of a highly different nature than that in rural regions. Taking all of this (and more, based on years of observations in the field among consumers) here is my version of consumer segmentation of the same demographic as covered in the chart above.

Descriptive segmentation of consumer behaviour

The Middle class – traditional definition, white collar jobs, steady paycheck, education/professional qualifications, closely aligned with “upper middle class” in the AfDB chart.

Emerging “middle” or rather the increasingly visible African consumers – non traditional (OECD cite), rapid upward mobility, primarily based in informal sector trades and services, newly successful entrepreneurs, small businessmen, extremely ambitious

Floating class 1 (“Brass Ring Syndrome“) – seeking status signifiers that are the ‘brass ring’, that is, they are ready to leap upwards, are almost there, focused on investing in future revenue generation opportunities, aspirational, may tend to be seen more in rural areas than urban.

Floating class 2 (“Fragile” or “Newborn”) – seeking footholds to gain enough stability to balance upon so as to make the leap for the brass ring, saving to invest in future revenue generation, hungry for more (not food but a mindset, as in hungrily seeking upward mobility), they may include the youth startups, tech entrepreneurs and all looking for the “something”, maybe more urban, and in the African contextual usage of the word “hustling” for the opportunity.

“Bottom of the Pyramid” – The $2/day demographic made famous by CK Prahalad, they are the pool from which the above three segments are emerging and are critically important in Africa in a way that they aren’t in opposed to India for instance because they don’t think of themselves as permanently poor, just temporarily cash crunched, especially migrant workers. Very, very different consumption behaviour between urban and rural in this segment. They may indeed form the rural version of Floating Class 2. I include them here because AfDB segmentation starts at $2.

Concluding thoughts

Once one’s mindset has evolved into considering oneself as part of a certain lifestyle, even if one’s income is “floating”, there are changes in buying habits that remain as part of this upward mobility. An example is that of the milk ATMs in Kenya. That is why I believe that taking a closer look at shifts in household consumption patterns as indicators of emerging into the so called “middle class” may offer more valuable insights for consumer market analysis than attempting to segment by dollar figures alone.

My 2 shillings worth on the size or value of the emerging African middle classes

There’s been a lot in the news of late about the size and worth of the emerging African middle class subsequent to the release of an as yet unseen report by an economist, Simon Freemantle, at Standard Bank, South Africa. The various headlines conflict each other, some say the middle class isn’t as large as earlier reported, others say its growing at a rapid clip. Their tone seems to depend on which aspect of this alleged report they support.

Instead of simply defining the middle classes by available daily spending power, as the African Development Bank did back in 2011, when they first announced the emergence of these new consumers, the Standard Bank report goes on to assess households by using the South African Living Standards Measure (LSM) as a means to segment them. But because we have yet to find a copy of the actual report itself, only articles referencing it (via a press release, to hazard a guess), there is no clarity on whether the South African LSM segmentation was directly applied to the households under consideration or whether the LSM was adapted for regional, social and cultural differences.

Even the SAARF, the South African body responsible for this evaluation tool has been questioning the validity of the LSM as it is structured at the moment. There are a few different approaches under development, from what I can tell based on a quick search online, including one which seeks to regionalize the LSM so that it can be far more accurately applied across the continent rather than for South African conditions alone. Again, we are not sure which version has been used in this new Standard Bank report.

The bottom line is that the emerging African middle class may indeed be smaller than imagined, though growing rapidly, or, that its as large as the AfDB originally estimated. That is, we still don’t know the size and worth of this consumer market. I suspect the reason for this that we’re trying to measure volatility, the underlying characteristic of the informal sector’s income streams, and that is why the goal posts seem to keep shifting.

The OECD had once said that the global emerging middle classes of today are not the same as those that emerged after the industrial revolution and established the foundations of the highly industrialized nations of the so called ‘first world’. That these new upwardly mobile and aspirational consumers were in fact emerging from the population segment originally designated as the ‘base of the pyramid’ and were less likely to have university degrees or salaried jobs.

This was also the point that Bright Simons made in his HBR article, that while it was undeniable that there was an increasingly visible pattern of conspicuous consumption happening across sub Saharan Africa, it should not be conflated with the concurrent rise of a “middle class” as the term is commonly understood.

I would first ask why are we trying to put numbers on the size of the middle class?

Are we conflating the concept of an educated white collar bourgeoisie with the corporate need for market analysis required to estimate the size and value of a market opportunity before making the decision to invest or enter a new market?

And, if so, then are the two necessarily the same thing?

Market Segmentation in the Informal Economy

This table is from “How to profit from Africa’s different consumer groups” and the research is from NKC Independent Economists group. There is something lacking in understanding the patterns of purchasing power when segmentation methodology from the formal economy are applied ad hoc to markets which are primarily informal.

As mentioned at the end of the previous post, an alternate method of segmenting the mass majority markets across Sub Sahara might be to cluster by volatility of cash flow. Farmers, for example, will tend to have cash 2 or 3 times a year, based on their crop and their geography, and some of these will be earmarked ahead for farm inputs.

Then, depending on which segment one is targeting and the proportion in that bracket earning a living from a variety of sources rather than a fixed salaried job, one can assess how much adaptation would existing business models and payment plans require for reaching the majority of the target audience.
This will differ from product to product, the articles breathlessly divulging all about this suddenly recognized African consumer market are still focusing on the creamy layer at the top of the income pyramid with their mentions of ice cream and caviar.

The old way focused on amounts of periodic cash flow, that is, income, as a means to segment people by disposable income available for consumption. The new way might have to look at their basket of groceries and then decide based on purchasing patterns.

The challenge arises when obsolete methods from a wholly different operating environment are applied out of context and the results interpreted in the same way as though there are no fundamental differences in the population and their mindset. There must be a reason why 96% of the hundreds of millions of mobile phone users across the continent are on prepaid or pay as you go plans.

Sampling uncertainty

This drawing was made by Jeroen Meijer of JAM visual design, Amsterdam, earlier this week during the workshop we held on Monday, November 26th, 2012.

Its a visualization of the chart I use to show how participants were sampled for the prepaid economy project. The axes represent the individual’s ability to accurately predict either timing or amount of their cash flow status, and thus, their ability to plan. By all rights, this should be on the Prepaid Economy blog where this topic has been an ongoing matter for discussion but I wanted to share the communication potential of this format.

Its also a way to segment the undifferentiated masses in the informal economy, where traditional means to segment a population demographic such as income level or education may not be relevant or skew results leading to misinterpretation. What if one could cluster by patterns of cash flow, and thus, consumer behaviour?