by Dean Karlan
The problem with microcredit is the way it has been pitched: as a one-size-fits-all solution to poverty that can be adopted effectively even without careful impact evaluations, and as something that every poor person should want. For all its merits, it is not that.
I am reminded of something I heard at a gathering on microcredit hosted by the Center for Global Development in early 2010. A group of academics, policymakers, and practitioners had been brought together to discuss the negative spin the media gave to the microcredit evaluations I’ve talked about in this chapter. Someone at the gathering summarized the palpable sense of concern and foreboding among the attendees: “The future of microcredit is at stake.”
What really is at stake here? Microcredit is the means to the end, not the end itself. What’s at stake is an opportunity to improve the lives of the poor. Millions of dollars pour into development aid, but it’s not nearly enough to solve the problems of poverty. My inner economist, the part of me that sees the world in terms of trade-offs, gets frustrated that when we direct so much of our money, efforts, and good intentions toward microcredit, we don’t direct them toward other things—like savings, insurance, education, and health. Some of those other things, many of which you’ll read about in this book, work, and are cheaper and more inclusive ways to get to our ultimate goal of reducing poverty.
So how can we do the best with the resources we have? And can we inspire more people to get involved, by giving them the confidence that there are programs that really work? That’s what is at stake.
The tool is not what matters; reducing poverty is.
5
TO PURSUE HAPPINESS
Having Better Things to Do
Oti pulled up just after six thirty. The equatorial sun had already set and Jake was standing in the cacophony of the dark street. If the car ride had taken a toll on Oti, it didn’t show; he looked fresh and nonchalant and he greeted Jake warmly. But Jake had been standing there for two hours. He was ready to have a serious talk about customer service.
The two had first been introduced by Daniel, the night guard at the compound where Jake was living, in the Labadi Beach neighborhood of Accra. Daniel had suggested that, instead of taking taxis to and from work every day, Jake could hire his friend Oti by the month to chauffeur. They agreed on a price and a timetable. Oti would come at eight o’clock each morning at the house and at four thirty each afternoon at the office. Oti was very friendly and personable, and he had a comfortable and light car that was reasonably functional. When it acted up, at least it was easy to push.
And so began a fine working relationship. But soon it became clear that it was not only Oti’s vehicle that suffered from reliability issues. After a few nights waiting in the hot, sticky dark, it was undeniable: The bloom was off the rose. Rather than fire Oti, though, it seemed a better idea to see whether a little elbow grease on the relationship would suffice, as it often did with his car.
Jake thought the situation might be salvageable because the task in question seemed very simple. There was an element of mystery in the erratic timing; after all, it was barely four miles from Oti’s house to the office. So what did it mean when he missed the appointed time by two hours?
Well, it’s hard to say exactly what it meant. If nothing else, Oti proved that what is true of cat-skinning is also true of arriving late: There’s more than one way to do it. Some ways are ordinary, like waiting until an hour after the pickup time to leave the house because you were watching a movie with your girlfriend, or getting stuck in a horrific traffic jam near the office of the National Lottery Board. Other ways are more involved, like setting out for the office with the tank on empty, then running out of gas on the way and realizing you forgot your jerry can at home, then walking over a mile back home to get the can, then walking to the gas station to fill it, then walking back to find that your car has been pushed out of the shoulder and onto the sidewalk by angry rushhour motorists (much to the dismay of the many pedestrians who are now gathered around it, waiting to give you a talking-to), and finally filling the tank and making the pickup. Yes, there are myriad ways to be two hours late, and Oti knew a lot of them.
Even more remarkable than the breadth and variety of contingencies, though, was the utter equanimity with which Oti endured—or actively authored—them. On the night in question, as on many other nights, he arrived with a smile and a cheery “Oh, Jake! How is it?” and Jake was left to wonder about the rest. If the mistake was his, he was unrepentant; and if the fates had conspired against him, stranding him for hours in an intractable knot of traffic, he was quiescent and no worse for the wear.
It was one thing if Oti didn’t mind wasting his customer’s time. Most of us have met somebody like that. The bizarre thing was that Oti didn’t seem to mind wasting his own time. And so that was the topic of his and Jake’s serious conversation on the ride home, weaving through the lumbering vans and rattling taxis on the pitted roads of the Ghanaian capital.
The Right Place at the Right Time
You’ve heard it a thousand times, that fundamental equivalency between business and real life, stated with appropriate brevity: Time is money. At some point during the dot-com boom this maxim was applied to Bill Gates, with an eye-popping result. Calculating his average hourly earnings over the course of a year, it was determined that should he pass by a hundred-dollar bill sitting on the sidewalk, he should not bend down and pick it up. The reasoning went that it wasn’t worth his time, because he could have made more than a hundred dollars by spending those two seconds at work.
The example is not exactly right, of course (for starters, Bill Gates doesn’t actually get paid by the hour), but the principle behind it is solid. Whenever we spend time doing a particular thing, we can compare it to the other things we could be doing instead. And when we think about the total cost of doing that particular thing, we should factor in the value of the alternatives we pass up in the process. Economists call that factor “opportunity cost.”
Opportunity cost isn’t just for Bill Gates; it’s for everyone. It’s the reason why going to graduate school is more expensive than the mere cost of tuition—because those years could be spent working and earning money. It’s also why you might choose not to work on a Saturday so you can go to the park with your family—presumably you’re passing up a few extra dollars of economic gain, because the fun you would miss out on by spending the day at work is too great. You could make the same argument about the whole question of career choice: People move to lower-paying jobs because they enjoy the work more, or they quit to become stay-at-home parents.
The fact that these choices might leave us with less in our bank accounts at the end of the day doesn’t mean they are uneconomic; they are just a reflection of our priorities, of the things that matter to us. Money is hardly the sole measure of that. Ultimately, opportunity costs are all about satisfaction—about making the choices that make you happiest, and also recognizing what you are not doing and how happy (or unhappy) that would make you.
So even though Oti could have earned money by picking up Jake on time, there might have been mitigating factors that made it a bad choice overall. Suppose the traffic was so horrendous that it would take Oti three hours to make the eight-mile round trip. In this case the frustration and extra time associated with driving could be more costly than the money he would earn, making it a reasonable decision to cancel.
Still, there are plenty of times when, even if we have the big-picture choices right, we can make little errors that cost us. Suppose Oti wanted to watch a movie at some point during the afternoon. He could watch it from noon to two o’clock, when he didn’t have anybody to pick up. In this case the opportunity cost is close to nil (he could, of course, go try to drum up more business, so there is some cost). Or he could watch it from four to six o’clock, and call Jake to cancel the day’s pickup. In this case the opportunity cost of watching the movie is the money he would have earned by driving Jake home.
As far as Jake cou
ld see, Oti was hardly calculating opportunity cost with the cool acumen of an Econ. He did not cancel when traffic was horrible, nor did he make it on time when it clearly would have been easy to do. He said, regarding the movies, “Hey, if I am watching some movies with my girlfriend, what can I do? Should I off it to leave the house? I know I will come to pick you later, or else tomorrow morning at least.” And he said, regarding the traffic, “Oh! If the traffic is too slow, what can I do? I am already sitting inside the car and I have said I would come, so I have to reach and pick you, even if it should take long. I don’t mind if it takes up to two or even three hours.”
Well, there it was, Zen-like in its simplicity, round and smooth as an eggshell. It was complete and, for the moment, irreproachable. It meant that on some days Jake would see Oti, and on other days he would take taxis.
Perhaps you are asking the obvious question: Why didn’t Jake fire Oti and find someone else? Well, sometimes we all let sentiment sway us—Jake isn’t an Econ either. He liked the guy. Maybe the problem was just that Oti knew that all too well.
The Pursuit of Happiness
For the sake of argument, let’s assume that Oti had not struck the ideal balance between work (driving) and leisure (watching movies with his girlfriend). Let’s assume that—if he thought through it clearly—on days without traffic, Oti really would prefer to pause his movie and earn money by picking up Jake. Well, if it’s any consolation, he is not the only one making these mistakes. His counterparts in New York City don’t appear to be doing much better.
A team of behavioral economists, Linda Babcock, Colin Camerer, George Loewenstein, and Richard Thaler, analyzed the records of thousands of New York City cabdrivers to figure out how they chose the number of hours they’d work on a given day. Specifically, they wanted to know: Were they allocating their working hours efficiently?
The basic idea is that, for taxi drivers, there are busy days and slow days. If the weather is bad, or if there is a convention in town, for instance, there are likely to be more fares, and in particular more short fares (which are more profitable per every mile driven). On the other hand, if it’s a beautiful spring day, people are less likely to hail a cab. On busy days, a driver’s average hourly earnings are significantly higher than on slow days. And though they can’t control which days are busy and which are slow, drivers can choose their working hours. Each time a driver takes a car out, he gets to choose how long to work (usually up to a twelve-hour limit on shifts). Meanwhile, we assume that taxi drivers like their leisure time too.
According to standard economics, the solution to the time-allocation problem is simple: Drivers should choose to work longer on busy days—getting in while the getting’s good—and knock off early on slow days—getting out when it is bad. This way they can work the same number of hours per week, but allocate those hours so that they earn the most money possible. The opportunity-cost interpretation is no different: Leisure time is cheaper on slow days than busy days, since drivers are passing up less income with every hour they spend on the couch.
The theory has its place, but perhaps that place is not in New York City’s cabs. (Or Singapore’s, for that matter—a study conducted there by Yuan Chou of the University of Melbourne yielded similar findings.) The economists concluded that drivers were not acting according to standard theory. In fact, they were doing quite the opposite.
On busy days drivers worked less, and on slow days they stayed out longer. This is doubly counterintuitive if one assumes that taxi drivers value their leisure time more on beautiful spring (slow) days than on dreary, rainy (busy) ones. The researchers proposed an alternate explanation: Rather than trying to maximize their average hourly wage, maybe drivers were working toward a specific income goal each day. If this was actually the case, then the data make perfect sense. On busy days drivers reach their targets quickly and close up shop, while on slow days they stay out longer, trolling for those last few fares in an effort to hit their marks.
If we accept the daily targeting theory, we might still ask: What’s the difference? Do a few hours here or there really have a big impact on earnings? Using the drivers’ daily records, the Babcock team estimated how much each driver would have made had he allocated his work hours differently. They found that, on average, drivers could earn about 5 percent more over time by simply working the same number of hours every day; and that they could earn about 10 percent more over time by working more on good days and less on bad days, while keeping the same number of total hours per week. Imagine increasing your salary by 10 percent without clocking any extra hours!
Now, hearing that you’re effectively passing up a 10 percent raise because of your choice of shifts should be enough to prick anyone’s ears. But particularly for the poor, who have the skimpiest of cushions to fall back on, oversights like these can mean significant changes in the conditions of daily life. They are the ones who stand to benefit most from small, piecemeal improvements in their economic decision-making, precisely because they live on the margins.
Oti and Microcredit
Of course, making a mistake to the tune of 10 percent of your income hurts; but it hurts even more when you’re carrying a debt whose interest piles up at six or seven times that rate. That’s where the rubber meets the road with microcredit.
The majority of the world’s poor—and practically all of the world’s microcredit clients—are either self-employed or casual workers, which means they make substantive and complex economic choices on a daily basis. For microentrepreneurs especially, it’s not only about how many hours they will work on a given day, but also about which products to stock and sell; where to operate their businesses; whether to hire other employees and at what wages.
In the United States, sorting through these issues would likely be the work of an executive or strategic consultant. But MBAs are few and far between in the developing world, and in any event you’re not likely to find one selling vegetables or plastic toys off a floor rug on the sidewalk.
The fact is that most microfinance clients are entrepreneurs not by design, but by necessity. Most developing countries lack social safety nets like unemployment checks or food stamps; if you want to eat, you have to work. And since there simply are not enough salaried jobs to accommodate everyone who needs to make a living, people go into business for themselves.
This is a far cry from the American notion of entrepreneurship, which evokes images of people who break out from the crowd because of their energy, independence, creativity, and drive. I do not mean to say that entrepreneurs in developing countries lack these characteristics. Quite the opposite: Were it not for an abundance of determination and ingenuity, a great many of them wouldn’t survive at all. But if you ask them, most would concede that building and running small businesses wasn’t their top career choice. Conversely, entrepreneurs anywhere, who exhibit that extraordinary dynamism and drive we so often imagine, would surely attest that their line of work is not for everyone.
So why does the microcredit movement act like it is? Muhammad Yunus, Nobel Prize–winning founder of the Grameen Bank, articulates it as follows:I firmly believe that all human beings have an innate skill. I call it the survival skill. The fact that the poor are alive is clear proof of their ability. They do not need us to teach them how to survive; they already know. So rather than waste our time teaching them new skills, we try to make maximum use of their existing skills. Giving the poor access to credit allows them to immediately put into practice the skills they already know. . . .
Yunus is a brilliant man, but this romantic point of view is wrong. Survival is one thing; building an enterprise from scratch—especially one profitable enough to sustain borrowing at typical microcredit interest rates—is quite another. In the last chapter we saw evidence from a project in Sri Lanka that showed there were good business opportunities out there, but that not everyone was equally able to take advantage of them.
That really shouldn’t come as much of a shock. Would anyone propose th
at every random man on the street in the United States or Europe, for instance, has the capacity to conceive and manage a thriving small business? More to the point, would anyone propose we start lending money to random men on the street with that in mind? The recent financial crisis has shown that even highly educated individuals in the world’s wealthiest nations sometimes behave irrationally and to everyone’s disadvantage, including their own, when it comes to borrowing. Debt can be the shackles just as much as it can be the key.
But let’s even concede that microcredit can be a valuable tool, at least for some. Still, might it work better if we provided some instructions along with it? If not every poor person is a born entrepreneur, maybe a little guidance would help. That’s the idea behind business training programs, which some microlenders offer to—or, in a few cases, require of—their borrowers.
Now, if Muhammad Yunus were right, and the poor already had all the survival skills they needed, then presumably they would already be operating efficiently, maximizing their profits. And, as he declares in no uncertain terms, providing any kind of skills training (business skills included) would simply be a waste of time. On the other hand, what if business training did prove useful? That would seem to suggest that at least some people are not inherently knockout entrepreneurs. Maybe they’re not inherently microcredit clients either.
After my two years with FINCA in El Salvador, this was one of the persistent questions rattling around my head. It stayed that way well into my last year of graduate school, when I met Martín Valdivia at a conference in Peru. Martín is a researcher at GRADE, a Peruvian think tank full of social scientists engaged in research on poverty. (He is also a great host at all the amazing restaurants in Lima, no doubt an essential ingredient in our collaboration and friendship). Martín and I found that we both had the same quandary, so we teamed up with Iris Lanao, the executive director of the Peruvian microlender FINCA Peru, and sought funding for an RCT on business training.
