ABHIJIT BANERJEE’S rad­i­cal new ap­proach to de­vel­op­ment aid.

Bulletin - - Contents - By Daniel Am­mann and Si­mon Brun­ner


Abhijit Banerjee talks about his rad­i­cal new ap­proach to de­vel­op­ment aid. →

Pro­fes­sor Banerjee, your book “Poor Eco­nomics,” pub­lished in 2011, caused a world­wide stir be­cause of the ex­per­i­ments you are con­duct­ing in the sphere of de­vel­op­ment aid. No­bel Lau­re­ate Amartya Sen called it “mar­velously in­sight­ful” and de­scribed you and Es­ther Du­flo, your wife and co-au­thor, as “two out­stand­ing re­searchers” …

…Thank you for your kind words, but you're ex­ag­ger­at­ing our im­por­tance …

You’re too mod­est. Your ap­proach is vi­sion­ary.

Com­par­a­tive field stud­ies are es­sen­tial for find­ing out what works. Over an ex­tended pe­riod of time, we com­pared pop­u­la­tion groups that were given as­sis­tance with other groups that re­ceived less sup­port or none at all. That was our most im­por­tant con­tri­bu­tion. This method al­lows us to de­ter­mine what mea­sures are ef­fec­tive with­out aban­don­ing the en­tire sys­tem. We don't come in and say, “Here's a magic bul­let that will solve all of your prob­lems.” In­stead, we show how to achieve some­thing use­ful, step by step, within the ex­ist­ing lim­its. We say, “Let's first find out ex­actly what the prob­lem is.” It's quite bor­ing ini­tially, and re­quires a great deal of pa­tience.

What does ef­fec­tive de­vel­op­ment aid look like?

We put to­gether a sup­port pack­age for the very poor, in­clud­ing pro­duc­tive goods, such as live­stock, as well as job train­ing, ac­cess to a sav­ings ac­count and short-term fi­nan­cial as­sis­tance. The re­sult: Af­ter only three years, hunger was less preva­lent than in the con­trol group that re­ceived no sup­port. In­come and sav­ings in­creased, as did con­sump­tion and pros­per­ity. This ap­proach has worked ev­ery­where – in Ethiopia, In­dia and Peru. To­day, ten years later, we can con­clude that it has a last­ing ef­fect.

Ed­u­ca­tion is said to be the key to de­vel­op­ment. Is that true?

If you com­pare in­di­vid­u­als rather than coun­tries, it is cer­tainly true that ed­u­ca­tion leads to higher in­comes and a bet­ter qual­ity of life. But we have also found that learn­ing lev­els in these coun­tries of­ten leave much to be de­sired, de­spite higher rates of school en­roll­ment.

Why is that?

The main prob­lem is ex­ces­sively am­bi­tious, stan­dard­ized and for­mal­is­tic cur­ric­ula, which fail to con­sider the gaps in an in­di­vid­ual stu­dent's learn­ing. So we de­vel­oped a pro­gram we call “Teach­ing at the Right Level,” or TARL. The ba­sic idea is to group chil­dren by per­for­mance level and de­ter­mine where there are gaps in each in­di­vid­ual child's learn­ing.

Could you give us an ex­am­ple?

Let's say a stu­dent is hav­ing dif­fi­culty with sub­trac­tion, even though he is older and should have mas­tered it by now. First that de­fi­ciency has to be rec­og­nized, and then it needs to be fixed. We have tried this at sev­eral sites in Ghana and In­dia. The re­sult: By the end of just 50 days of con­cen­trated in­struc­tion, stu­dents who were ini­tially part of the low­est­per­form­ing group had moved up in one sub­ject to one of the high­est-per­form­ing groups. This is a way to make sure that ev­ery child masters the ba­sic scholas­tic skills.

If the pres­i­dent of a poor coun­try were to ask you how to im­prove the lives of his coun­try’s cit­i­zens, what would you say?

Find out where the prover­bial low-hang­ing fruit is – in other words, find out what can make a big dif­fer­ence at mod­est ex­pense. The an­swer is dif­fer­ent for ev­ery coun­try. Some­times it's the school sys­tem, some­times the health care sys­tem or ac­cess to credit. Then I would try to per­suade him to in­vest enough money in high-qual­ity ser­vices for the poor, in­clud­ing pro­vid­ing af­ford­able ac­cess to good schools, pre­ven­tive med­i­cal care and hospi­tals.

Might an un­con­di­tional ba­sic in­come be a sim­ple so­lu­tion for poor coun­tries?

Yes, if it makes peo­ple feel em­pow­ered and al­lows them to take con­trol of their lives. How­ever, it's also pos­si­ble that re­ceiv­ing money would re­duce their level of ef­fort. We have launched a large-scale ex­per­i­ment in Kenya to test this idea: Over the next 12 years, 6,000 peo­ple in 40 vil­lages will re­ceive a monthly pay­ment of 23 dol­lars. This is roughly equiv­a­lent to the ab­so­lute poverty line. We will com­pare that group with two con­trol groups, one of which will re­ceive the same amount of money for only two years, while the other will re­ceive no sup­port. This will tell us whether an un­con­di­tional ba­sic in­come might be a so­lu­tion.

Crit­ics of­ten ques­tion the ra­tio­nale and pur­pose of de­vel­op­ment aid. This is an in­dus­try that is seek­ing to do good. What is the big­gest mis­take that it is mak­ing?

In many cases peo­ple have too much con­fi­dence in their in­tu­ition. I fre­quently en­counter de­vel­op­ment ex­perts who are con­vinced that they know ex­actly what the root cause of poverty is, and how the prob­lem can be solved. When I ask them how they know that, and what the ev­i­dence is, I of­ten re­ceive lit­tle re­sponse.

What is the great­est mis­con­cep­tion that wealthy coun­tries have about peo­ple in poor coun­tries?

It's that the world's poor­est peo­ple have no choice or that for cul­tural rea­sons they are in­ca­pable of im­prov­ing their lives. Our ex­per­i­ments are also in­tended to prove that given the right help, the poor­est peo­ple can win their bat­tle against poverty.

De­vel­op­ment aid ex­per­i­ment in Kenya: Over a pe­riod of 12 years, 6,000 peo­ple in 40 vil­lages will re­ceive a ba­sic in­come of 23 dol­lars ev­ery month.

Abhijit Banerjee, 57, stud­ied eco­nomics in Kolkata and New Delhi and at Har­vard Uni­ver­sity. He is cur­rently a Ford Foun­da­tion In­ter­na­tional Pro­fes­sor of Eco­nomics at the Mas­sachusetts In­sti­tute of Tech­nol­ogy, where he and his wife, French econ­o­mist Es­ther

Newspapers in English

Newspapers from Switzerland

© PressReader. All rights reserved.