Try It, You’ll Like It: Testing Models for Sustainable Business in Asia and Africa

By David I. Levine
A successful salesman in Cambodia, let’s call him Mr. Bun, had a problem: The owner of his favorite noodle shop would not buy his fuel-efficient cookstove, even though the new stove would save the restaurant money on fuel and reduce unhealthy smoke.

“I don’t have the money to buy it,” the shop owner explained, though Mr. Bun suspected that the owner was also concerned the stove might not save much money or might not cook well.

“No problem,” Mr. Bun replied. “Take the stove for a free trial. My two sons and I will come for noodles each morning. Noodles cost you less than the stove saves you each day. That way you can use the money you save on fuel to pay for the stove over time.”

Mr. Bun’s “noodle contract” meant the noodle shop owner could get the efficient stove without needing a big up-front payment. The shop owner also had the choice to stop the free breakfasts if the stove did not meet his needs or if it broke.

The problems this noodle contract solved—lack of funds, uncertainty about product quality, and concerns about durability—apply to many other new products that can help break the cycle of poverty and increase sustainability in low-income communities. Many of these products ought to be good business opportunities, but they require new business models that address social and financial obstacles.

Getting efficient and safe cookstoves into the hands of more cooks is particularly imperative. The World Health Organization estimates that smoke from unsafe stoves kills a million or more children a year in developing countries. In addition, the high fuel use of inefficient traditional stoves increases poverty, deforestation, and global climate change.

With my colleagues in Uganda, I have been testing a novel sales offer that shares the advantages of Mr. Bun’s noodle contract: a free trial, time payments, and the right to return the stove if it does not work. With this sales offer, the customer pays all or most of the cost of the new stove with money she has already saved on fuel. That removes almost all risk in trying and keeping an improved cookstove.
Ugastove for David Levine post
Figure 1: Ugastove charcoal-burning stove

We have run two randomized trials in Kampala, Uganda, of the novel offer with the charcoal-burning Ugastove, which retails for $7-$11 (Figure 1). We then repeated the test with a wood-burning Envirofit stove in rural Uganda.

Only about five percent of people who were offered a traditional cash-and-carry offer bought the new stove – even though we gave them a week to gather funds or check with other family members. In contrast, among the 355 potential customers, or 47 percent of those who received the novel offer in Kampala, accepted the free trial. And among those customers, only 2 percent returned the stove. While some moved away or defaulted, we received over 97 percent of scheduled payments. We also found that the offer increased sales about twice as much as offering just the free trial or just time payments plus the right to return.

Results were just as strong in rural Uganda. As in the city, about 5% of those given the option to purchase for cash (either that day or a week or so later) purchased a stove. In contrast, 57% of attendees at meetings who were offered a free trial followed by rent-to-own purchased a stove. Overall, we received 99% of the funds owed – a rate even higher than in the urban study.

These results are very promising, but not yet a business model. For example, in the urban study we paid college-educated data collectors to drive to a neighborhood and collect a handful of $2 payments.

The rural study came closer to a business model, as we paid our local promoters (who had organized the marketing meetings) a 10% commission to collect the time payments. Even so, it took a lot of management time to encourage them to collect payments from the more distant customers.

There was also an occasional problem where the promoter collected a time payment and then an emergency arose and the promoter spent the money before turning it over to us. Our supervisor was able to retrieve all collected funds, but these few incidents are related to what can be a bigger problem: Those collecting payments have an incentive to claim the customer defaulted, while actually keeping the money.

We could avoid this problem if vendors were the ones making the loans. The problem is that vendors themselves have limited liquidity. We also found that the vendors we worked with did not like giving credit to customers. Although we found that Ugandans are very trustworthy, they are not very trusting.

We have experimented with several strategies to overcome these challenges. If customers make their time payments with mobile money, then vendors do not need to walk to each home to collect payments. And because the payments flow directly to the distributor, there is no risk that the local promoters might keep the payments. We tried selling Ceramaji water filters in rural and peri-urban western Kenya using this sales offer, coupled with time payments paid over M-PESA, Kenya’s mobile-money platform.
Ceramaji filter for David Levine post
Figure 2: Ceramaji filter

M-PESA is a very widespread mobile payments platform in Kenya. One 2011 study found that over 70% of poor people in Kenya had used it.

As we had seen with stoves, the combination of a free trial followed by time payments increased uptake enormously. No households owned a Cera Maji filter at baseline, although our local NGO had been offering it for some time. By the end of the study, 29% of those to whom we offered the free trial, followed by the MPESA time payments plan, had purchased a filter.

Unfortunately, sending a bill by text message and then waiting for mobile payments did not work well. The customers’ mobile phones were often turned off or lacked charge. In other cases, the husband owned the phone and received the text message, while the wife was responsible for the payments. Even when the message got through, these customers were not used to receiving a bill by text message, walking to the nearby store that served as an MPESA agent, depositing a few dollars in payment, and then sending the payment.

Customers eventually paid for almost all the filters, but it required in-person collection in far too many cases to make this business model financially sustainable. It is plausible that collecting time payments by mobile phone will work in less poor communities or in more urban areas with more familiarity with MPESA. In fact, the same offer might work a few years from now in the same kinds of areas that we worked in.

Our results were more encouraging when we sold water filters in Dhaka, Bangladesh (Figure 3). Again, we compared a lump sum payment (with a week’s notice to gather funds) to time payments and a free trial. Unlike with the cookstoves, the free trial was not necessary to increase demand – perhaps because the filter was a visibly attractive consumer item.
water filter for David levine post
Figure 3: Filter we sold in Dhaka, Bangladesh

As with the cookstove, however, the offer of time payments increased demand a lot. At a price of $28 (roughly the market price), for example, demand was 12% with the one-payment offer and almost half (47%) with time payments.
We compared time payments with a related sales offer, layaway plans. In a layaway, consumers make a series of partial payments and receive the filter when they complete the payments. Layaway is attractive to vendors because they are not lending out a product (so they do not face liquidity constraints) and do not need to trust consumers to complete payments.

We had thought consumers would be less attracted to layaway plans because they wouldn’t receive the filter until they completed payments and would have to trust the vendor to actually deliver the filter.

Much to our surprise, however, the layaway offer increased demand just as much as the offer of time payments!

Some of this high demand may be unique to our distribution partner: icddr,b. They run a well-known cholera hospital in Dhaka, which probably increased customers’ trust. It is important to test how well these encouraging results generalize to other products and settings.
We are continuing to explore sales offers and relations with vendors that can overcome liquidity constraints, uncertainty about product quality and other barriers to adopting products that can save lives and break the cycle of poverty.

Based on our results so far, we think this will be a very fruitful area of research.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s