TED2010
Esther Duflo: Social experiments to fight poverty
Filmed:
Readability: 2.8
1,298,863 views
Alleviating poverty is more guesswork than science, and lack of data on aid's impact raises questions about how to provide it. But Clark Medal-winner Esther Duflo says it's possible to know which development efforts help and which hurt -- by testing solutions with randomized trials.
Esther Duflo - Development economist
Esther Duflo takes economics out of the lab and into the field to discover the causes of poverty and means to eradicate it. Full bio
Esther Duflo takes economics out of the lab and into the field to discover the causes of poverty and means to eradicate it. Full bio
Double-click the English transcript below to play the video.
00:18
So here it is. You can check: I am short, I'm French,
0
3000
2000
00:20
I have a pretty strong French accent,
1
5000
2000
00:22
so that's going to be clear in a moment.
2
7000
3000
00:26
Maybe a sobering thought
3
11000
2000
00:28
and something you all know about.
4
13000
2000
00:30
And I suspect many of you gave
5
15000
2000
00:32
something to the people of Haiti this year.
6
17000
3000
00:35
And there is something else
7
20000
2000
00:37
I believe in the back of your mind
8
22000
2000
00:39
you also know.
9
24000
2000
00:41
That is, every day,
10
26000
2000
00:43
25,000 children die
11
28000
2000
00:45
of entirely preventable causes.
12
30000
3000
00:48
That's a Haiti earthquake every eight days.
13
33000
3000
00:51
And I suspect many of you probably gave something
14
36000
2000
00:53
towards that problem as well,
15
38000
2000
00:55
but somehow it doesn't happen
16
40000
2000
00:57
with the same intensity.
17
42000
2000
00:59
So why is that?
18
44000
2000
01:02
Well, here is a thought experiment for you.
19
47000
3000
01:05
Imagine you have a few million dollars that you've raised --
20
50000
2000
01:07
maybe you're a politician in a developing country
21
52000
3000
01:10
and you have a budget to spend. You want to spend it on the poor:
22
55000
3000
01:13
How do you go about it?
23
58000
3000
01:16
Do you believe the people who tell you
24
61000
2000
01:18
that all we need to do is to spend money?
25
63000
2000
01:20
That we know how to eradicate poverty,
26
65000
3000
01:23
we just need to do more?
27
68000
2000
01:25
Or do you believe the people who tell you that
28
70000
2000
01:27
aid is not going to help, on the contrary it might hurt,
29
72000
3000
01:30
it might exacerbate corruption, dependence, etc.?
30
75000
3000
01:34
Or maybe you turn to the past.
31
79000
2000
01:36
After all, we have spent billions of dollars on aid.
32
81000
3000
01:39
Maybe you look at the past and see.
33
84000
2000
01:41
Has it done any good?
34
86000
2000
01:43
And, sadly, we don't know.
35
88000
2000
01:45
And worst of all, we will never know.
36
90000
3000
01:49
And the reason is that -- take Africa for example.
37
94000
2000
01:51
Africans have already got a lot of aid.
38
96000
2000
01:53
These are the blue bars.
39
98000
2000
01:55
And the GDP in Africa is not making much progress.
40
100000
3000
01:58
Okay, fine. How do you know what
41
103000
2000
02:00
would have happened without the aid?
42
105000
2000
02:02
Maybe it would have been much worse,
43
107000
3000
02:05
or maybe it would have been better.
44
110000
2000
02:07
We have no idea. We don't know what the counterfactual is.
45
112000
3000
02:10
There's only one Africa.
46
115000
2000
02:12
So what do you do?
47
117000
2000
02:14
To give the aid, and hope and pray that something comes out of it?
48
119000
3000
02:18
Or do you focus on your everyday life
49
123000
3000
02:21
and let the earthquake every eight days
50
126000
2000
02:23
continue to happen?
51
128000
2000
02:25
The thing is, if we don't know
52
130000
2000
02:27
whether we are doing any good,
53
132000
2000
02:29
we are not any better
54
134000
2000
02:31
than the Medieval doctors and their leeches.
55
136000
3000
02:34
Sometimes the patient gets better, sometimes the patient dies.
56
139000
3000
02:37
Is it the leeches? Is it something else?
57
142000
3000
02:40
We don't know.
58
145000
2000
02:42
So here are some other questions.
59
147000
2000
02:44
They're smaller questions,
60
149000
2000
02:46
but they are not that small.
61
151000
2000
02:48
Immunization, that's the cheapest way
62
153000
3000
02:51
to save a child's life.
63
156000
2000
02:53
And the world has spent a lot of money on it:
64
158000
2000
02:55
The GAVI and the Gates Foundations
65
160000
2000
02:57
are each pledging a lot of money towards it,
66
162000
2000
02:59
and developing countries themselves have been doing a lot of effort.
67
164000
3000
03:02
And yet, every year
68
167000
2000
03:04
at least 25 million children
69
169000
2000
03:06
do not get the immunization they should get.
70
171000
3000
03:09
So this is what you call a "last mile problem."
71
174000
3000
03:12
The technology is there,
72
177000
2000
03:14
the infrastructure is there,
73
179000
2000
03:16
and yet it doesn't happen.
74
181000
2000
03:18
So you have your million.
75
183000
2000
03:20
How do you use your million
76
185000
2000
03:22
to solve this last mile problem?
77
187000
2000
03:24
And here's another question:
78
189000
2000
03:26
Malaria. Malaria kills almost
79
191000
3000
03:29
900,000 people every year,
80
194000
3000
03:32
most of them in Sub-Saharan Africa,
81
197000
2000
03:34
most of them under five.
82
199000
2000
03:36
In fact, that is the leading cause of under-five mortality.
83
201000
3000
03:39
We already know how to kill malaria,
84
204000
3000
03:42
but some people come to you and say,
85
207000
2000
03:44
"You have your millions. How about bed nets?"
86
209000
3000
03:47
Bed nets are very cheap.
87
212000
2000
03:49
For 10 dollars, you can manufacture and ship
88
214000
3000
03:52
an insecticide treated bed net
89
217000
2000
03:54
and you can teach someone to use them.
90
219000
2000
03:56
And, not only do they protect the people who sleep under them,
91
221000
3000
03:59
but they have these great contagion benefits.
92
224000
2000
04:01
If half of a community sleeps under a net,
93
226000
3000
04:04
the other half also benefits
94
229000
2000
04:06
because the contagion of the disease spread.
95
231000
3000
04:09
And yet, only a quarter of kids at risk sleep under a net.
96
234000
3000
04:12
Societies should be willing to go out
97
237000
2000
04:14
and subsidize the net, give them for free,
98
239000
2000
04:16
or, for that matter, pay people to use them
99
241000
2000
04:18
because of those contagion benefits.
100
243000
2000
04:20
"Not so fast," say other people.
101
245000
2000
04:22
"If you give the nets for free,
102
247000
2000
04:24
people are not going to value them.
103
249000
2000
04:26
They're not going to use them,
104
251000
2000
04:28
or at least they're not going to use them as bed nets,
105
253000
2000
04:30
maybe as fishing nets."
106
255000
2000
04:32
So, what do you do?
107
257000
2000
04:34
Do you give the nets for free to maximize coverage,
108
259000
2000
04:36
or do you make people pay
109
261000
2000
04:38
in order to make sure that they really value them?
110
263000
2000
04:40
How do you know?
111
265000
2000
04:42
And a third question: Education.
112
267000
2000
04:44
Maybe that's the solution, maybe we should send kids to school.
113
269000
2000
04:46
But how do you do that?
114
271000
2000
04:48
Do you hire teachers? Do you build more schools?
115
273000
2000
04:50
Do you provide school lunch?
116
275000
2000
04:52
How do you know?
117
277000
2000
04:54
So here is the thing.
118
279000
2000
04:56
I cannot answer the big question,
119
281000
2000
04:58
whether aid did any good or not.
120
283000
2000
05:00
But these three questions, I can answer them.
121
285000
3000
05:04
It's not the Middle Ages anymore,
122
289000
2000
05:06
it's the 21st century.
123
291000
2000
05:08
And in the 20th century,
124
293000
2000
05:10
randomized, controlled trials
125
295000
2000
05:12
have revolutionized medicine
126
297000
2000
05:14
by allowing us to distinguish
127
299000
2000
05:16
between drugs that work
128
301000
2000
05:18
and drugs that don't work.
129
303000
2000
05:20
And you can do the same
130
305000
2000
05:22
randomized, controlled trial for social policy.
131
307000
3000
05:25
You can put social innovation to the same
132
310000
2000
05:27
rigorous, scientific tests
133
312000
2000
05:29
that we use for drugs.
134
314000
2000
05:31
And in this way, you can take the guesswork
135
316000
3000
05:34
out of policy-making
136
319000
2000
05:36
by knowing what works,
137
321000
2000
05:38
what doesn't work and why.
138
323000
2000
05:40
And I'll give you some examples with those three questions.
139
325000
3000
05:44
So I start with immunization.
140
329000
2000
05:46
Here's Udaipur District, Rajasthan. Beautiful.
141
331000
3000
05:49
Well, when I started working there,
142
334000
2000
05:51
about one percent of children
143
336000
2000
05:53
were fully immunized.
144
338000
2000
05:55
That's bad, but there are places like that.
145
340000
3000
05:58
Now, it's not because the vaccines are not there --
146
343000
2000
06:00
they are there and they are free --
147
345000
2000
06:02
and it's not because parents do not care about their kids.
148
347000
3000
06:05
The same child that is not immunized against measles,
149
350000
3000
06:08
if they do get measles, parents will spend
150
353000
2000
06:10
thousands of rupees to help them.
151
355000
2000
06:12
So you get these empty village subcenters
152
357000
2000
06:14
and crowded hospitals.
153
359000
2000
06:16
So what is the problem?
154
361000
2000
06:18
Well, part of the problem, surely, is people do not fully understand.
155
363000
3000
06:21
After all, in this country as well,
156
366000
2000
06:23
all sorts of myths and misconceptions
157
368000
2000
06:25
go around immunization.
158
370000
3000
06:28
So if that's the case, that's difficult,
159
373000
2000
06:30
because persuasion is really difficult.
160
375000
3000
06:33
But maybe there is another problem as well.
161
378000
2000
06:35
It's going from intention to action.
162
380000
3000
06:38
Imagine you are a mother
163
383000
2000
06:40
in Udaipur District, Rajasthan.
164
385000
2000
06:42
You have to walk a few kilometers to get your kids immunized.
165
387000
3000
06:45
And maybe when you get there, what you find is this:
166
390000
2000
06:47
The subcenter is closed. Ao you have to come back,
167
392000
2000
06:49
and you are so busy and you have so many other things to do,
168
394000
3000
06:52
you will always tend to postpone and postpone,
169
397000
2000
06:54
and eventually it gets too late.
170
399000
2000
06:56
Well, if that's the problem, then that's much easier.
171
401000
3000
06:59
Because A, we can make it easy,
172
404000
3000
07:02
and B, we can maybe
173
407000
2000
07:04
give people a reason to act today,
174
409000
2000
07:06
rather than wait till tomorrow.
175
411000
2000
07:08
So these are simple ideas, but we didn't know.
176
413000
2000
07:10
So let's try them.
177
415000
2000
07:12
So what we did is we did a randomized, controlled trial
178
417000
3000
07:15
in 134 villages in Udaipur Districts.
179
420000
2000
07:17
So the blue dots
180
422000
2000
07:19
are selected randomly.
181
424000
2000
07:21
We made it easy -- I'll tell you how in a moment.
182
426000
3000
07:24
In the red dots, we made it easy
183
429000
2000
07:26
and gave people a reason to act now.
184
431000
2000
07:28
The white dots are comparisons, nothing changed.
185
433000
3000
07:31
So we make it easy by organizing
186
436000
2000
07:33
this monthly camp where people can
187
438000
2000
07:35
get their kids immunized.
188
440000
2000
07:37
And then you make it easy
189
442000
2000
07:39
and give a reason to act now
190
444000
2000
07:41
by adding a kilo of lentils for each immunization.
191
446000
3000
07:44
Now, a kilo of lentils is tiny.
192
449000
3000
07:47
It's never going to convince anybody
193
452000
2000
07:49
to do something that they don't want to do.
194
454000
2000
07:51
On the other hand, if your problem is you tend to postpone,
195
456000
3000
07:54
then it might give you a reason to act today
196
459000
2000
07:56
rather than later.
197
461000
2000
07:58
So what do we find?
198
463000
2000
08:00
Well, beforehand, everything is the same.
199
465000
2000
08:02
That's the beauty of randomization.
200
467000
2000
08:04
Afterwards,
201
469000
2000
08:06
the camp -- just having the camp --
202
471000
2000
08:08
increases immunization from six percent to 17 percent.
203
473000
2000
08:10
That's full immunization.
204
475000
2000
08:12
That's not bad, that's a good improvement.
205
477000
2000
08:14
Add the lentils and you reach to 38 percent.
206
479000
3000
08:17
So here you've got your answer.
207
482000
2000
08:19
Make it easy and give a kilo of lentils,
208
484000
2000
08:21
you multiply immunization rate by six.
209
486000
3000
08:24
Now, you might say, "Well, but it's not sustainable.
210
489000
2000
08:26
We cannot keep giving lentils to people."
211
491000
2000
08:28
Well, it turns out it's wrong economics,
212
493000
2000
08:30
because it is cheaper
213
495000
2000
08:32
to give lentils than not to give them.
214
497000
2000
08:34
Since you have to pay for the nurse anyway,
215
499000
2000
08:36
the cost per immunization
216
501000
2000
08:38
ends up being cheaper if you give incentives than if you don't.
217
503000
3000
08:42
How about bed nets?
218
507000
2000
08:44
Should you give them for free, or should you ask people to pay for them?
219
509000
3000
08:47
So the answer hinges
220
512000
2000
08:49
on the answer to three simple questions.
221
514000
2000
08:51
One is: If people must pay for a bed net,
222
516000
3000
08:54
are they going to purchase them?
223
519000
2000
08:56
The second one is:
224
521000
2000
08:58
If I give bed nets for free,
225
523000
2000
09:00
are people going to use them?
226
525000
2000
09:02
And the third one is:
227
527000
2000
09:04
Do free bed nets discourage future purchase?
228
529000
2000
09:06
The third one is important
229
531000
2000
09:08
because if we think people get used to handouts,
230
533000
3000
09:11
it might destroy markets to distribute free bed nets.
231
536000
3000
09:14
Now this is a debate that has generated
232
539000
2000
09:16
a lot of emotion and angry rhetoric.
233
541000
3000
09:19
It's more ideological than practical,
234
544000
2000
09:21
but it turns out it's an easy question.
235
546000
2000
09:23
We can know the answer to this question.
236
548000
2000
09:25
We can just run an experiment.
237
550000
2000
09:27
And many experiments have been run, and they all have the same results,
238
552000
2000
09:29
so I'm just going to talk to you about one.
239
554000
3000
09:32
And this one that was in Kenya,
240
557000
2000
09:34
they went around and distributed to people
241
559000
2000
09:36
vouchers, discount vouchers.
242
561000
2000
09:38
So people with their voucher
243
563000
2000
09:40
could get the bed net in the local pharmacy.
244
565000
2000
09:42
And some people get 100 percent discount,
245
567000
3000
09:45
and some people get 20 percent discounts,
246
570000
2000
09:47
and some people get 50 percent discount, etc.
247
572000
3000
09:50
And now we can see what happens.
248
575000
2000
09:52
So, how about the purchasing?
249
577000
2000
09:54
Well, what you can see is that
250
579000
2000
09:56
when people have to pay for their bed nets,
251
581000
2000
09:58
the coverage rate really falls down a lot.
252
583000
3000
10:01
So even with partial subsidy,
253
586000
2000
10:03
three dollars is still not the full cost of a bed net,
254
588000
3000
10:06
and now you only have 20 percent of the people with the bed nets,
255
591000
2000
10:08
you lose the health immunity, that's not great.
256
593000
3000
10:11
Second thing is, how about the use?
257
596000
2000
10:13
Well, the good news is, people, if they have the bed nets,
258
598000
2000
10:15
will use the bed nets regardless of how they got it.
259
600000
3000
10:18
If they get it for free, they use it.
260
603000
2000
10:20
If they have to pay for it, they use it.
261
605000
2000
10:22
How about the long term?
262
607000
2000
10:24
In the long term,
263
609000
2000
10:26
people who got the free bed nets,
264
611000
2000
10:28
one year later, were offered the option
265
613000
2000
10:30
to purchase a bed net at two dollars.
266
615000
3000
10:33
And people who got the free one
267
618000
2000
10:35
were actually more likely to purchase the second one
268
620000
3000
10:38
than people who didn't get a free one.
269
623000
2000
10:40
So people do not get used to handouts; they get used to nets.
270
625000
3000
10:43
Maybe we need to give them a little bit more credit.
271
628000
3000
10:48
So, that's for bed nets. So you will think, "That's great.
272
633000
2000
10:50
You know how to immunize kids, you know how to give bed nets."
273
635000
3000
10:53
But what politicians need is a range of options.
274
638000
3000
10:56
They need to know: Out of all the things I could do,
275
641000
3000
10:59
what is the best way to achieve my goals?
276
644000
3000
11:02
So suppose your goal is to get kids into school.
277
647000
3000
11:05
There are so many things you could do. You could pay for uniforms,
278
650000
3000
11:08
you could eliminate fees, you could build latrines,
279
653000
2000
11:10
you could give girls sanitary pads, etc., etc.
280
655000
3000
11:13
So what's the best?
281
658000
2000
11:15
Well, at some level, we think
282
660000
2000
11:17
all of these things should work.
283
662000
2000
11:19
So, is that sufficient? If we think they should work intuitively,
284
664000
2000
11:21
should we go for them?
285
666000
2000
11:23
Well, in business, that's certainly not the way we would go about it.
286
668000
3000
11:27
Consider for example
287
672000
2000
11:29
transporting goods.
288
674000
2000
11:31
Before the canals were invented
289
676000
2000
11:33
in Britain before the Industrial Revolution,
290
678000
3000
11:36
goods used to go on horse carts.
291
681000
2000
11:38
And then canals were built,
292
683000
2000
11:40
and with the same horseman and the same horse,
293
685000
3000
11:43
you could carry ten times as much cargo.
294
688000
3000
11:46
So should they have continued
295
691000
2000
11:48
to carry the goods on the horse carts, on the ground,
296
693000
3000
11:51
that they would eventually get there?
297
696000
2000
11:53
Well, if that had been the case,
298
698000
2000
11:55
there would have been no Industrial Revolution.
299
700000
2000
11:57
So why shouldn't we do the same with social policy?
300
702000
3000
12:00
In technology, we spend so much time
301
705000
2000
12:02
experimenting, fine-tuning,
302
707000
2000
12:04
getting the absolute cheapest way to do something,
303
709000
2000
12:06
so why aren't we doing that with social policy?
304
711000
3000
12:09
Well, with experiments, what you can do
305
714000
2000
12:11
is answer a simple question.
306
716000
2000
12:13
Suppose you have 100 dollars to spend
307
718000
2000
12:15
on various interventions.
308
720000
2000
12:17
How many extra years of education
309
722000
2000
12:19
do you get for your hundred dollars?
310
724000
2000
12:21
Now I'm going to show you
311
726000
2000
12:23
what we get with various education interventions.
312
728000
3000
12:27
So the first ones are if you want the usual suspects,
313
732000
2000
12:29
hire teachers, school meals,
314
734000
2000
12:31
school uniforms, scholarships.
315
736000
2000
12:33
And that's not bad. For your hundred dollars,
316
738000
2000
12:35
you get between one and three extra years of education.
317
740000
3000
12:39
Things that don't work so well is bribing parents,
318
744000
2000
12:41
just because so many kids are already going to school
319
746000
3000
12:44
that you end up spending a lot of money.
320
749000
2000
12:46
And here are the most surprising results.
321
751000
3000
12:49
Tell people the benefits of education,
322
754000
3000
12:52
that's very cheap to do.
323
757000
2000
12:54
So for every hundred dollars you spend doing that,
324
759000
3000
12:57
you get 40 extra years of education.
325
762000
3000
13:00
And, in places where there are worms,
326
765000
2000
13:02
intestinal worms,
327
767000
2000
13:04
cure the kids of their worms.
328
769000
2000
13:06
And for every hundred dollars,
329
771000
2000
13:08
you get almost 30 extra years of education.
330
773000
3000
13:11
So this is not your intuition,
331
776000
2000
13:13
this is not what people would have gone for,
332
778000
2000
13:15
and yet, these are the programs that work.
333
780000
3000
13:18
We need that kind of information, we need more of it,
334
783000
2000
13:20
and then we need to guide policy.
335
785000
2000
13:24
So now, I started from the big problem, and I couldn't answer it.
336
789000
3000
13:27
And I cut it into smaller questions,
337
792000
3000
13:30
and I have the answer to these smaller questions.
338
795000
2000
13:32
And they are good, scientific, robust answers.
339
797000
3000
13:36
So let's go back to Haiti for a moment.
340
801000
2000
13:39
In Haiti, about 200,000 people died --
341
804000
3000
13:43
actually, a bit more by the latest estimate.
342
808000
2000
13:45
And the response of the world was great:
343
810000
2000
13:47
Two billion dollars got pledged just last month,
344
812000
3000
13:51
so that's about 10,000 dollars per death.
345
816000
2000
13:53
That doesn't sound like that much when you think about it.
346
818000
3000
13:56
But if we were willing to spend 10,000 dollars
347
821000
3000
13:59
for every child under five who dies,
348
824000
3000
14:02
that would be 90 billion per year
349
827000
3000
14:05
just for that problem.
350
830000
2000
14:07
And yet it doesn't happen.
351
832000
2000
14:09
So, why is that?
352
834000
2000
14:11
Well, I think what part of the problem is that,
353
836000
2000
14:13
in Haiti, although the problem is huge,
354
838000
2000
14:15
somehow we understand it, it's localized.
355
840000
2000
14:17
You give your money to Doctors Without Borders,
356
842000
2000
14:19
you give your money to Partners In Health,
357
844000
2000
14:21
and they'll send in the doctors, and they'll send in the lumber,
358
846000
3000
14:24
and they'll helicopter things out and in.
359
849000
3000
14:27
And the problem of poverty is not like that.
360
852000
2000
14:29
So, first, it's mostly invisible;
361
854000
2000
14:31
second, it's huge;
362
856000
2000
14:33
and third, we don't know whether we are doing the right thing.
363
858000
3000
14:36
There's no silver bullet.
364
861000
2000
14:38
You cannot helicopter people out of poverty.
365
863000
2000
14:40
And that's very frustrating.
366
865000
2000
14:42
But look what we just did today.
367
867000
3000
14:45
I gave you three simple answers to three questions:
368
870000
3000
14:48
Give lentils to immunize people,
369
873000
2000
14:50
provide free bed nets, deworm children.
370
875000
2000
14:52
With immunization or bed nets,
371
877000
2000
14:54
you can save a life for 300 dollars per life saved.
372
879000
3000
14:57
With deworming, you can get
373
882000
2000
14:59
an extra year of education for three dollars.
374
884000
2000
15:02
So we cannot eradicate poverty just yet,
375
887000
3000
15:05
but we can get started.
376
890000
2000
15:07
And maybe we can get started small
377
892000
2000
15:09
with things that we know are effective.
378
894000
3000
15:12
Here's an example of how this can be powerful.
379
897000
2000
15:14
Deworming.
380
899000
2000
15:16
Worms have a little bit of a problem grabbing the headlines.
381
901000
2000
15:18
They are not beautiful and don't kill anybody.
382
903000
3000
15:21
And yet, when the young global leader in Davos
383
906000
2000
15:23
showed the numbers I gave you,
384
908000
2000
15:25
they started Deworm the World.
385
910000
2000
15:27
And thanks to Deworm the World,
386
912000
2000
15:29
and the effort of many country governments and foundations,
387
914000
2000
15:31
20 million school-aged children got dewormed in 2009.
388
916000
3000
15:34
So this evidence is powerful.
389
919000
2000
15:36
It can prompt action.
390
921000
2000
15:38
So we should get started now.
391
923000
2000
15:40
It's not going to be easy.
392
925000
2000
15:42
It's a very slow process.
393
927000
2000
15:44
You have to keep experimenting, and sometimes ideology
394
929000
2000
15:46
has to be trumped by practicality.
395
931000
2000
15:48
And sometimes what works somewhere doesn't work elsewhere.
396
933000
3000
15:51
So it's a slow process,
397
936000
2000
15:53
but there is no other way.
398
938000
2000
15:55
These economics I'm proposing,
399
940000
2000
15:57
it's like 20th century medicine.
400
942000
2000
15:59
It's a slow, deliberative process
401
944000
2000
16:01
of discovery.
402
946000
2000
16:03
There is no miracle cure,
403
948000
2000
16:05
but modern medicine is saving
404
950000
2000
16:07
millions of lives every year,
405
952000
2000
16:09
and we can do the same thing.
406
954000
2000
16:13
And now, maybe, we can go back to the bigger question
407
958000
3000
16:16
that I started with at the beginning.
408
961000
3000
16:19
I cannot tell you
409
964000
2000
16:21
whether the aid we have spent in the past has made a difference,
410
966000
3000
16:24
but can we come back here in 30 years
411
969000
3000
16:27
and say, "What we have done,
412
972000
3000
16:30
it really prompted a change for the better."
413
975000
3000
16:33
I believe we can and I hope we will.
414
978000
2000
16:35
Thank you.
415
980000
2000
16:37
(Applause)
416
982000
2000
ABOUT THE SPEAKER
Esther Duflo - Development economistEsther Duflo takes economics out of the lab and into the field to discover the causes of poverty and means to eradicate it.
Why you should listen
2009 MacArthur fellow Esther Duflo is pushing the field of development economics by studying specific causal relationships that lead to or perpetuate poverty. She looks at close-to-home issues: household behavior, education, access to finance and health.
At MIT, she's the founder and director of the Jameel Poverty Action Lab (J-PAL), a research network that evaluates social programs. It's concerned less with wide-ranging policy than with specific questions. Sample: If schoolkids could get their uniforms for free, would attendance go up? What's an effective way to reward mothers for immunizing their babies? Randomized trials offer new insights toward creating global equity and prosperity. Her work may blur the lines between economics and activism, but it's a role Duflo not only considers comfortable but vital.
Her new book is Poor Economics, with Abhijit W. Banerjee.
Esther Duflo | Speaker | TED.com