Poor Economics: Designing robust interventions to fight poverty through randomized experiments

When I hear a professor from MIT saying she knows how to fight poverty, my first reaction is that of skepticism. “Does she even know what poverty is?” That’s how my mind would respond. And yet when I heard the TED talk by Prof. Esther Duflo of MIT titled “Social experiments to fight poverty”, I was thoroughly impressed. It took me some time to understand what exactly in the talk that impressed me so much. Subsequently I bought the book “Poor Economics” which she co-wrote with Abhijit Banerjee. Now, I am slowly beginning to understand why Amartya Sen has said, “A marvelously insightful book by two outstanding researchers on the real nature of poverty”. In this article I want to highlight three things that I find interesting in the book.

Intractable problem, manageable sub-problems: Poverty eradication looks like an unsolvable problem. People from Karl Marx to Mahatma Gandhi have taken a shot at it. However, it is not clear whether we have a handle on it. Silver bullet approach like “give aids” is not helping. Banerjee-Duflo take a different view. They feel that instead of trying to answer the top question, why not look at some sub-problems – each of which might have a definite and practical solution. For example, in the talk, Duflo presents three such sub-problems: How can we have more kids immunized? How can we get more people to use bednets that can reduce malaria affliction? How can we get students to attend school more number of days for a given dollar spent on the cause? The book, like the talk, shows how we can go about systematically addressing these sub-problems.

Randomized experimentation: Let’s take the question of immunization. In Udaipur district of Rajasthan, it was found out that only 1% of the children are fully immunized. The vaccines are there and are available for free. It is not that the parents don’t care about their kids. When their kids get measles parents end up spending thousands of rupees in treatment. So you have empty village sub-centres on one hand, and crowded hospitals on the other hand. So looks like the intention is not translating into action. What do you do? Dulfo & team decided to try random trials in 134 villages in Udaipur district. For one third villages there was no change, another one third villages had immunization monthly camps conducted and the last one third had camp plus a kilo of lentil free for camp participant. As it turned out the immunization percentage jumped by a whopping 37%. Note that this approach is analogous to randomized control trials used in medicine to discover drugs.

Design as if implementation matters: The beauty of Banerjee-Duflo approach is not that it does not make any assumption about the culture, anxiety, aspirations of the poor. The experiments reveal their biases anyway. I call this approach of designing an intervention – design as if implementation matters. Note that the approach does not advocate laboratory experiments – the experiments are performed in-field in actual conditions. I feel that experimentation and immersive research are the heart and the soul of systematic innovation. Banerjee-Duflo approach epitomizes both. Now, I know why the TED talk struck such a chord with me.

I strongly recommend the book for every student of social innovation.

Maganlal Gandhi: Mahatma Gandhi’s innovation partner

"He [Maganlal] was, in my opinion, a genius... He laid the foundation of the science of Khadi by writing his Vanat Shastra” said Mahatma Gandhi at the inaugural ceremony of Magan Museum of Khadi (pdf) at Wardha on Dec 30, 1938². Ever since I read this, I became curious about Maganlal Gandhi. Where did Maganlal learn the science of khadi? Was this science backed up by rigorous experiments? If so, where did Maganlal perform those experiments? I wanted to find out. Finally, I got a peek into the story when my friend Prof. Rishikesha Krishnan connected me to Prof. Shambu Prasad of XIMB Bhubaneswar. Shambu has done extensive research on science of Gandhi. I found answers to some of my questions in an excellent article written by Shambu “Gandhi and Maganlal: Khadi science and the Gandhian scientist”. Here is a short summary.

Maganlal Khushalchand Gandhi (1883-1928) was Gandhi’s nephew (a grandson of his uncle) and 19 years younger to him. Maganlal met Gandhi in 1902, two days before Gandhi was to leave for South Africa. Maganlal was then on the lookout for a job in his native Kathiawar. Gandhi asked Maganlal to come to South Africa offering it as a land of opportunities with ample scope for growth. Maganlal was immediately put into business and was running a family shop of the Gandhis. When Gandhi quit his practice and decided to set-up a farm and take to farming as an occupation, Maganlal was the first to join him unconditionally knowing it involves self-imposed poverty.

In a short time, Maganlal picked up several skills at Phoenix Farm - composing and running the machines in the press, farming, carpentry and tailoring, keeping accounts of the settlement and teaching the children Gujarati and Mathematics. It was Maganlal who suggested the name Sadagraha to Gandhiji when he solicited a better term for “passive resistance” in the local newspaper Indian Opinion. Gandhi later modified Sadagraha to Satyagraha.

Later in India, Maganlal set-up and ran the khadi laboratory at Sabarmati Ashram as he was the head of the Technical Department of the All India Khadi Board (later the All India Spinners Association). Maganlal traveled to Madras Presidency to learn the art. The technical department of the Ashram, tested several samples of yarn (over 300 every month) and gave feedback to the Provincial Congress Committees. These results were widely reported regularly in Young India and Navjivan. As a resource centre in the field of khadi the Ashram used to send its staff, spindles, specimens of yarn, and charts explaining the effect of the wheel to exhibitions all over the country. Gandhi relied on Maganlal to test the various machines and always wanted Maganlal’s opinion on technical developments whether it was Mirabehn’s discovery of the soft spindle, Shankarlal’s Gandiva spinning wheel or the Ramachandra lift pump.

In 1922, a ‘Khaddar Information Bureau’ was constituted to provide or collect information on khadi from the provinces, to inform congress committees and selected workers on reports from the centres. Maganlal edited its ‘Khadi Bulletin’. A syllabus was formulated in 1923 for the weaving school with a regular six-month course. A khaddar service scheme was also instituted under which 600 instructors were to be trained in home carding.

Maganlal passed away while at work in Bihar due to typhoid on April 23, 1928, at the peak of his life and that of the khadi movement. In a moving tribute, titled ‘My Best Comrade Gone’ Gandhi remarked that: ‘The world knows so little of how much my so-called greatness depends upon the incessant toil and drudgery of silent, devoted, able and pure workers, men as well as women. And among them all Maganlal was to me the greatest, the best and the purest.’

Source:

1. Gandhi and Maganlal: Khadi science and the Gandhian scientist” by Shambu Prasad, Presented at the Seminar ‘Gandhi and his Contemporaries’ held at the Indian Institute of Advanced Studies, Shimla, April 13-15, 1999

2. “Mahatma: Life of Mohandas Gandhi” by D. G. Tendulkar, volume 5, page 6.

3. Maganlal’s photo is from "Magan Nivas" at Sabarmati Ashram.

How GE develops innovation leaders through the LIG program

GE launched Leadership, Innovation and Growth (LIG) program in September 2006 and ran it till September 2008, mostly in Crotonville, epicenter of GE’s learning & development. Altogether 2,500 senior managers in 260 teams went through this four day program. The purpose of LIG was to make innovation and growth as much of a religion at GE as Six Sigma had been under Jack Welch. On day four the course wrapped up with a plenary session at which each team had 20 minutes to deliver a presentation to Jeff Immelt, GE’s CEO. When Immelt was asked why he devoted so much time to LIG, he said, “LIG gave me a way to drive change and develop leaders at the same time”. What happened at LIG? Here is an overview on what GE Power senior management team went through at LIG. (source: An excellent HBR article “How GE teaches teams to lead change” by Steven Prokesch, a senior editor of HBR who was invited to attend one of the LIG programs in October 2007).

LIG program is a brainchild of Susan Peters, GE’s VP of executive development and Daniel Henson, then CMO. Before attending the LIG, the senior managers at a business would assess their team’s success in creating a climate supportive for innovation. The assessment would generate an innovation dashboard that would be used during the program. During the program there would be talks by external gurus as well as internal role models. A large amount of time – about 15 to 20 hours – was set aside for breakout sessions. What happens during these breakouts?

During the first breakout on the morning of day one, the Power Gen team guessed and the learned their actual team scores for the 360 degree review of their growth values. This triggered a reassessment of almost every aspect of their business. Some of the questions that got raised by the team were: “We’re not as good at anticipating major trends as we ought to be”, “Is solar a good place to be?” or “Renewable energy, clean coal, nuclear – all are going to be policy dependent. Are we good at this?” The reassessment continued in this manner throughout the four days.

The reflections generated insights – may be the old rules don’t always apply; may be limits on carbon emissions and tax incentives for clean, renewable power matter more. That sparked a conversation about GE’s ability to understand and influence government policies. Managers agreed that it was deficient and that beefing it was therefore a priority. In the next breakout session the Power Gen managers talked soberly about the state of their core. Unless the operations are strengthened it is difficult to free up time for innovative thinking.

At another breakout session the team assessed their innovation portfolio by putting each project in one of the three boxes, a framework created by Prof. Vijay Govindarajan: incremental (aimed at strengthening the core), adjacencies (taking existing technologies to new markets or taking new technologies to its existing markets), nonlinear shifts (discontinuous shifts in technology or markets with radically new products or business models).

The final LIG session involved the reports to Immelt. Power Gen team led by Bolze talked about their biggest takeways from the program, their 10-year projection of revenues (from $13 billion to $40 billion with renewables’ share going from 30% to 50%), and a vision statement – “Powering the world responsibly”. They committed themselves to strengthening the core. They confessed they needed to get better at looking around the corners to spot nonlinear shifts. They listed the capabilities they needed to build: regulatory expertise, faster product development, creating emerging-markets products “in country for country”. They vowed to lighten up a bit and become more playful, a characteristic of innovative companies. As they spoke Immelt asked questions and shared observations.

Within a few weeks after the LIG session Steve Bolze, as required, sent a commitment letter to Immelt, laying out the measures his team would take to increase the pace of organic growth. Such a letter becomes a living contract between Immelt and the team.

image source: ge.com

Managing innovation: story of Tesco India

Last week I got an opportunity to listen to Sandeep Dhar, CEO of Tesco Hindustan Service Centre (HSC) on how Tesco India manages innovation at CII Innovation Forum. Tesco is the third largest retailer in the world by revenue and the second largest in terms of profits. Tesco HSC is responsible for standardizing all back-office processes impacting Tesco business globally and it currently employs 4000 associates. Sandeep demonstrated with examples how Tesco has made innovation an every day practice. How is it different from other places? Let’s see in this article.

Idea qualification: At Tesco, like everywhere else, every idea needs to have a business case which typically means cost saving for an idea from HSC. However, at Tesco, it needs to meet two additional criteria: One, it should improve or maintain customer experience. Two, it should simplify or at least maintain employee work complexity. Several ideas end up getting dropped for not meeting these criteria.

Continuous improvement track: Small ideas are taken seriously at Tesco. In fact, if an idea can reduce average time taken to handle a customer at the TIL by one second, it saves the company 2 million pounds a year in UK alone. How to keep people motivated for doing continuous improvement? At Tesco HSC this is ensured by how the team KPI or SLAs are set. For example, there is a team in India which receives calls when particular equipment like a refrigerator or air-conditioning unit is malfunctioning. Team’s job is to call the relevant vendor of the town and the vendor would send people to the Tesco store for servicing. Typical KPIs of such call-centric team would be percentage of calls serviced in 90 seconds, percentage calls tracked to closure etc. With these metric the improvements would be directed towards improving the efficiency of handling calls. They deliver limited value. Tesco changed the KPI of the team to “reduce the amount of money Tesco spends on maintenance”. With this the orientation of the team changed. They started looking into the reasons for equipment failures, some of the early warning and on preventive maintenance measures. Two years after this change the maintenance budget has come down by 40 percent.

Another team in India is responsible for making payments for the traffic violations of the truck drivers carrying Tesco supplies. Typical KPI would be timely payment of the fine and the accuracy of the payment. This KPI was changed to “figure out ways to reduce the fine”. One of the team members visited the traffic authority web site and found out a mechanism to challenge the fines. He and a couple of his colleagues started looking into the data and started challenging the fines selectively. At times the challenge would be successful, at times it wouldn’t. Then the team went one step further and analyzed the reasons for the traffic violations. They identified some commonly made mistakes and sent the information to the drivers. They were able to identify drivers who were more prone to making traffic violations.

Problem solving track: This is the track where the businesses are asked to share some of their chronic problems. Tesco India sets up cross functional teams to work on them. One such problem was related to Tesco’s online website. Tesco is world’s largest e-grocer. The web site has a favorites list which currently includes all the items a customer had shopped in the past. The drawback of this mechanism was that the favorites list becomes very long and inconvenient for the shopper. A Tesco India team worked with IISc experts and came up with a statistical algorithm that predicts what a shopper might be shopping on a day. For example, if you have purchased 3 gallons of milk on Monday, it would not show milk in favorites on Tuesday. But it may show it on Friday. This has resulted in reducing the favorites list to one third of the original size. This has been piloted and going into production.

In-store work experience: All managers in India go through Tesco Week in Store Together (TWIST) program whenever they visit a country with Tesco stores. For Sandeep a TWIST a year is mandatory. He said if he doesn’t spend a week working the store, his KPI goes down by a notch. When Sandeep spends a week in the store, he may spend a day at check-out counter, another day stacking products, third day he may be doing stock count, on the fourth day he may be going out with the delivery van.

Retail test lab: In 2007, Tesco HSC established retail test lab where recreated all hardware environments that exists in different Tesco stores. For example, sales counters and handheld devices etc. When IT develops a software it is tested in this simulated environment which improves the reliability of the deployment.

Prof. Ulrich Weinberg on Design Thinking

Thanks to Prof. Rajeev and Supriya Dey of IIMB, yesterday I got an opportunity to meet Prof. Weinberg over lunch and subsequently attended his presentation “Design thinking: Looking beyond obvious” at IIMB. Prof. Ulrich Weinberg is the director of the School of Design Thinking at Hasso Plattner Institute, Potsdam, Germany. The school is modeled after D-school of Stanford. According to Prof. Weinberg at HPI D-school you get no credit points, no grades and no certificate worth any official value and yet within four years of its founding it has students from 70 disciplines and 20 nations. Natural question is: What is going on there? And what exactly do these students do that they value learning experience over credits and certificates?

The students participate in 1 hr, 3 hr, 1 day, 3 day, 1 week, 3 week and 6 week workshops / projects which progressively prepares them for the 12 week final project. Every project starts with a cross-functional team of 4 to 6 students and students rotate after every project. With students coming from disciplines as varied as art, history, medicine, engineering, psychology, law, forming a cross-functional team is not a challenge at HPI. The faculty works with industry-Government partners to generate a set of challenges. One such challenge came from DHL: How might we transfer packets from point A to B in the inner cities of future where no cars are allowed? London, Beijing, Mumbai are good candidates. If you feel that the challenge statement is simple looking, mind you it would have taken several days of work for the faculty members to work with DHL to arrive at the crisp looking statement which reflects user needs and pain points.

Out of the 12 weeks, half the time i.e. 6 weeks would be spent in understanding, observing and synthesizing the experience. This means going to the inner cities and observe how things work. Students observe how people walk to bus stops or train stations or cycle. In the DHL project the team was inspired by the workings of Dabbawalas of Mumbai. The most difficult part of this first phase is to switch your solution engine off. This is especially harder for engineers who are so used to solving problems. One of the deliverables is the creation of a detailed persona of the customers whose pain you are addressing. In the DHL case it could be thirty-five year old Chris or a twenty-two years old Sangeeta, a resident of the future city who either works for DHL or partners with DHL in moving the packet forward.

The next phase consists of ideation, prototyping and testing. Since the team has spent enough time in the field, ideation generates hundreds of ideas relevant to the problem. The school has created spaces with movable trolleys and white boards that help these teams in brainstorming. Every day there is a three minute presentation to the rest of the class on your prototype and you get immediate feedback. Weinberg says, “No other feedback can be as powerful as this inputs from your colleagues and faculty”.

The team working on DHL problem came up with an idea of citizen participation in packet delivery. They named it Bring.buddy. It taps all the consumers moving through a city each day, whether via bike, public transport or on foot. Interested participants indicate their travel route for the day using a downloadable smartphone app; a text message then lets them know of any packages needing delivery along the way. When there is such a package, the participant picks it up from the local kiosk where it's waiting and delivers it as they go about their daily business. In exchange for their help, the program rewards them with points that can be redeemed for free train tickets, merchandise coupons or CO2 credits. Check out the YouTube video BringBuddy.

Bring.Buddy created so much buzz in Shanghai trade show and even in Berlin that people started asking DHL when they are implementing it. Finally DHL has decided to pilot it in China along with the student team at HPI. Whether the service reaches market is anybody’s guess. However, can you doubt the learning the students are getting in the process?

If you want to get a glimpse of what design thinking experience might look like, I suggest watching 3-part video of Ideo’s Deep Dive.

Evolution of 3M's Nonwoven technology platform over seven decades

A typical technology platform far outlives the first product which uses it. One of the classic examples is 3M’s Nonwoven technology platform. Originally used in making ribbons for decorating gifts, the platform over seven decades has been used in cleaning pads, surgical tapes, drapes & masks, fasteners, floppy disk liners, absorbent material to combat oil slicks, “metered” paint rollers for home improvement and sound deadners in cars. What does it mean to develop and manage a technology platform like nonwoven? Let’s look at the story in brief.

Al Boese didn’t have a high school diploma and started his career in 3M as a mail boy. In 1938 his boss in 3M’s tape lab, Dick Drew, suggested that he might not be cut out for technical work. Perhaps, Drew counseled, Boese should take time off to find a different job. Boese hung around the lab anyway. One day Drew off handedly mentioned that 3M specifications called for an inexpensive, noncorrosive backing that was fibrous, but not woven, for its popular electrical tape. Rather than hunt for a new job Boese found the best library on fibers at the University of Minnesota’s Home Economics Department and he spent the summer there.

“One day I was walking by the rubber colander in the tape lab,” Boese said, “I stuck a little tuft of acetate fiber in the colander. It heated the surface of the fibers and bonded them together. That was the opening to make nonwovens. Heat and pressure.” Boese started experimenting with this process and set up a small lab called Carfab Lab. Boese’s new process didn’t produce a better backing for electrical tape, but gazing at a department store one day in the mid-1940s, he had an idea. May be, if the new nonwoven material was dyed and sprinkled with color flecks, it could be used in decorative display. Or why not slit the material into strips and make ribbon for decorating gifts?

Boese’s early attempts at ribbon failed as it was structurally weak for wrapping packages and it wasn’t very attractive. “It was obvious to everybody that we had a product failure,” Boese said. In three years the ribbon brought in about $800,000 in revenues and the losses totaled $200,000. Boese was given three months to make it profitable and he did. The new product, 3M Sasheen decorative ribbon, was a hit when it was introduced in 1950s, along with a companion product, Lacelon ribbon.

From ribbon, 3M “married” nonwovens to abrasives in the 1950s to produce Scotch-Brite scrubbing and polishing pads, floor maintenance supplies and industrial polishing materials. A decade later, new dampening sleeves were made from nonwovens that made offset printing much more economical. Disposable surgical face masks and Micropore surgical tape opened the door to other nonwoven medical products.

There were some disappoints along the way, too. 3M never successfully developed nonwovens for book covers, draperies and window displays. A novel product called Skimmit was heralded as the easy way to skim oil off liquids like soups, but consumers never thought so. Early attempts at creating comfortable shoulder pads for clothing fizzled.

Nonwovens had become a part of so many 3M products that a Nonwoven Technology Center was created in 1983 to offer technical knowledge and expertise across the company. By then, about 10 percent of 3M’s business or nearly $1 billion in sales from about 20 divisions represented some form of nonwoven application in products ranging from diapers to diskettes. By the late 1990s that percentage had grown to 15 percent overall and sales of about $2 billion.

Note: I am looking for an example of a technology platform that has evolved in India over a decade or two. In case you have any information, do let me know by either writing a comment on this blog or by sending email to: vinay at catalign dot com.

Source: A century of innovation at 3M (story on pages 50-53, image on page 180).

Gandhi’s failed self-help experiment at Shantiniketan: a lesson in culture-unfriendly intervention

Every organization has a unique culture and any act to bring about a sustainable change needs to be sensitive to the existing culture. Gandhi, in 1915, either didn’t understand this yet or he hadn’t fully internalized it. Gandhi had a four day visit to Shantiniketan in February 1915 a month after his arrival in India. During this visit he conducted a self-help experiment with all the students and teachers. Like his visit, the impact of the experiment was short-lived. What was the experiment? And why did it fail? Let’s explore it here.

In 1904, Gandhi had established an Ashram in South Africa in a town called Phoenix about 14 miles from Durban. In 1915, when Gandhi returned to India many of his Phoenix Farm associates followed him as well. It so happened that the Phoenix party arrived before Gandhi arrived and found a home in Shantiniketan. After visiting Gokhale in Pune and meeting relatives in Rajkot and Porbandar, Gandhi proceeded to Shantiniketan.

The phoenix family had been assigned a separate quarters at Shantiniketan. Maganlal Gandhi, Gandhi’s close associate in Phoenix Farm, was their head, and he had made it his business to see that all the rules of the Phoenix Ashram should be scrupulously observed. Among Gandhi’s friends Andrews and Pearson were also present.

Soon after his arrival Gandhi mixed with the teachers and students and engaged them in a discussion on self-help. He suggested to the teachers that, if they and the boys dispensed with the services of paid cooks and cooked their food themselves, it would enable the teachers to control the kitchen from the point of view of the boys’ physical and moral health. Also it would give lessons on self-help to the students. Some nodded their head tentatively, some appeared more enthusiastic. Gandhi invited Rabindranath to give his opinion. He said he didn’t mind it provided the teachers were favorable. To the boys, he said, “The experiment contains the key to Swaraj”.

Thus started an experiment involving the entire community of 125 boys and their teachers. Boys were running the kitchen, handling the garbage, cleaning the latrine, sweeping the ground and as Gandhi’s biographer Louis Fischer puts it, “forsaking the muse for the monk”. Pearson was looking after the cooking part and Nagenbabu was looking after the sanitary cleaning. It was difficult for the students. Some began to show early fatigue. Cleaning of vessels was especially tedious. A group of students played sitar next to the kitchen so as to make the task feel less cumbersome.

Gandhi had to leave Shantiniketan abruptly to attend Gokhale’s funeral. And the experiment was stopped after some time. As Gandhi notes in his autobiography, “I am of the opinion that the famous institution lost nothing by having conducted the experiment for a brief interval, and some of the experiences gained could not but be of help to the teachers.”

Gandhi and Tagore were united by their love for India and mankind. However, they were very different personalities. In the words of Louis Fischer, “Gandhi was the wheat field and Tagore the rose garden, Gandhi was the working arm and Tagore the singing voice, Gandhi was frugal and Tagore was prodigal”. At Shantiniketan, Tagore’s pupils sang and danced, wove garlands, painted sunrise and made life sweet and beautiful. Gandhi’s experiment essentially turned the place upside down. It is no surprise, it was shortlived. I feel it is a great lesson to understand what culture-unfriendly intervention is like.

Related articles:

Mahatma Gandhi and the heart and soul of systematic innovation

4 types of innovation leaders (Gandhi, Jamsetji Tata, Vikram Sarabhai, George Fernandes)

Sources:

An autobiography, M. K. Gandhi (part V, chapter IV, Shantiniketan)

The life of Mahatma Gandhi, Louis Fischer, HarperCollins

Photo: mkgandhi.org (Gandhi in 1915)