Paul Niehaus: Giving away money to people living in extreme poverty

Paul Niehaus: Giving away money to people living in extreme poverty


Hi, it’s great to be with you. Thank you
Sam for that very charitable introduction. So on a note of humility
I’ll start by talking about one of my bigger failures – this is one of my bigger
failures as a husband, and I’ll explain in a moment why I want to start here so
over the course of my been married for 10 years to Lauren who’s wonderful and
one of the things that I think I’ve generally done pretty well at is buying
presents for her and so I take a lot of pride in that but this is an example of
a time that I really struck out so I was convinced that you know we live in San
Diego and we live near some of these beautiful canyons and I thought it would
be great if we got a drone and we can take this out and fly it around with the
kids and Lauren opened the package and took one look at it and sent it back to
the store so for the record she has since I come around on this and admits
that it actually would have been fun so I think in the long run I’m gonna win
but the look though the reason I tell the story is partly because I’m I don’t
want to start with this sort of list of achievements which i think is a bit
overly grandiose but also because I want to remind us that when we try to help
other people even if they’re people that we know really really well right the
people that you know best in your life it can actually be quite hard it can be
hard to figure out what is going to be the right thing for them and so I want
us all to approach this question today of how to help people are living in
extreme poverty with sort of an appropriate stance or perspective of
humility right kind of recognizing that you know for most of us this they’re
living lives that are very different from anything that any of us have
experienced and so that we want to approach it with that from that stance
so let me start with some the substance with a couple of numbers and to kind of
motivate what I’ve been working on for much of the last ten years which is
trying to enable the flow of money to poor people so this is a set of numbers
that I think makes me feel very optimistic about the world this shows
you the trend over time as estimated by folks at Brookings in the global poverty
gap so what is the global poverty gap that means if we take everybody in the
world who’s living on less than some poverty line say a dollar ninety a day
which is where the world bank currently pegs things and say how much money would
it take to get them up to that line and then we add that number up for everybody
in the world right so what this graph tells us is that that number the
poverty gap has been falling pretty substantially over time and that’s
really exciting right and maybe some of that is because of our own efforts and
kind of things we’ve done to try to accelerate the process of development
but a lot of that is because people are sort of figuring things out right and
they’re finding ways to improve their lives and that’s something to feel
really good about another really interesting thing to do with these
numbers or to compare them to what we spend on foreign aid so this shows you
what foreign aid looks like over the same time period official development
assistance right from governments to governments and so in fact in around
2005 for the first time official development assistance exceeded the
global poverty gap this doesn’t mean that then some mechanical sense we could
end poverty instantly if we wanted to right because finding all of these
people and figuring out exactly how much money it would take to get them to the
line is an impossible job right we can’t do that and it’s not even clear that
that would be the best way to spend the money if we wanted to do that but I want
to give you some sense that you know in terms of ballpark figures right the
kinds of money that we’re currently spending on trying to help poor people
and the kinds of money it would take to get everybody out of extreme poverty are
in the same range right which maybe also should be a reason to be optimistic and
also to kind of ask the question which is going to be at the core of our
conversation today which is should we actually give more of that money
directly to poor people and let them decide what to do with it as opposed to
deciding that ourselves so I’m the other important thing that happened in 2005
was I went to grad school I didn’t get into MIT
unfortunately so I had to go to Harvard down the road and obviously I would have
had a much more successful career and be given a much more important talk if I
had gotten into MIT I’m sure but but it was actually a very exciting time to be
in grad school even at Harvard and part of that was because of the boom in
randomized controlled trials kind of an experimental program evaluation that
really started to take off around that point in time so one thing I found that
a lot of people don’t know about sort of anti-poverty work is that you know aid
and foreign assistance could really got going in a serious way in the in the
1960s after independence in many of the former colonies but despite the fact
that we’ve been working on this problem for the last 50 years or so for a very
long time there was very little kind of rigorous experimental testing
of our ideas and so I think we sort of have an idea that we’ve learned a lot
about what works but I actually think that on many important questions we
actually learned very little up until around the early 2000s but in the early
2000s that started to explode and so it was a very exciting time to be in grad
school we were learning lots and lots of things really fast there were some
things that you know I think we thought should work that turned out to work less
well than we thought they would and that’s disappointing but still very
valuable but there were also some things that seemed to work consistently a lot
better than I think most people expected and so one of the patterns that I
noticed and my co-founder does it give directly noticed was that giving money
to people and letting them decide what to do with it seemed to work better than
a lot of people had expected and so even at the time and this is a somewhat dated
figure that just highlights some of the studies that we sort of looked at when
we started give directly I think there was this very strong fact pattern
emerging from lots of lots of studies in different contexts different places
around the world and you know it’s such a diverse literature that it’s
impossible to summarize and I won’t attempt to do that here but I think that
three sort of broad generalizations that are mostly true and useful for you to
have in mind the first is that in general people sort of we’re doing
useful reasonable things with money and we could certainly debate whether
they’re doing the best possible things with money but they’re doing things that
you know seamless sensible and like they would improve their lives you know
sending their kids to school buying assets buying food investing in housing
these kinds of things so that was one a second one is that one of the things
that people were really worried about and that I had heard a lot is that if
you give money to people that they’re not going to work as hard to improve
their own lives right they may become dependent on the transfers or you know
economists jargon they may reduce their labor supply and so that certainly could
happen and in fact you know when you think about some of the kinds of work
that extremely poor people do some of it is very painful it’s very tedious and I
personally think it might actually be a great thing if they didn’t have to do
quite so much of that but in any case in the sort of surveys that we have done so
far across a number of these experimental tests that doesn’t seem to
be the case if anything it seems like more often people will do a bit more
work sometimes that may be because they’re able to purchase assets that
they needed so you know if you were trained as a welder but you can’t afford
welding equipment you can’t use those skills so I think
there’s a little bit of that going on but I think that was sort of you know
big concern number one that people have generally had about giving money to poor
people which at least for these populations in these countries on
average hasn’t been the case who knows what big concern number two is this is big concern number two right
this is my favorite local brew so there’s a big concern and you hear this
all the time that if we give money to poor people they’re gonna waste it on
alcohol drugs things that would be self harmful to them right and you know it is
certainly true I mean I’ve met people who have gone out to get a drink and
celebrate right so there’s no doubt at all that sort of people who you know
I’ve received transfers you know some people have spent some of that money on
alcohol but again when we’ve done systematic reviews and said on average
what are the impacts to the best that we can measure these things the answer has
been not that people have spent more money but if anything that on average
people have spent less money on these things and I actually think there’s some
really interesting thinking to be done about the relationship between substance
abuse and poverty if it is perhaps a little bit more nuanced than the way
we’ve thought about it up until now but that’s kind of via the overall average
fact pattern in the data so far so in grad school we sort of looked at all of
these things which I think already at that time we’re pretty well established
facts and we said this seems kind of exciting and you know we’d like to
potentially be able to give away our own money to poor people if nothing else
because we were you know hopeful that we would get jobs and have some money to
give and you know so we started thinking about how you would do this and then
realized sort of the second thing that was happening in a lot of emerging
markets was that poor people are getting plugged into the financial system in a
way that they hadn’t been previously and I remember I had the first conversation
I had that really got me thinking about this was with someone that was building
sort of mobile ATMs in India so sending somebody around with a motorcycle to a
bunch of villages to process transactions and then come back at the
end of the day and so you know we were talking to him about how we might use
this to try to combat fraud and corruption in some of the major public
sector programs but you know at some point it hits you like this guy has
created bank accounts for some of the poorest people on the planet and if
you’ve ever now and talked to these people you understand that the
opportunity to put money into those accounts is really exciting right and
it’s sort of our world is fundamentally different when we’re connect
to people financially in that way than it was before so we started looking
around the world and asking where would be the best place to do this where can
you find a lot of poor people who have access to digital payments and not
surprisingly that was at the time it probably still is Kenya because of the
success of safari comes mobile money service there but outside of Kenya as
well right in most countries in the world more and more it’s possible to
send payments to the poorest people on the planet securely and very cheaply so
we looked at these two things and said you know the infrastructure is there for
the first time the evidence is a lot better than I think we had been brought
to expect and so this ought to exist and it doesn’t and we did talk to some of
the big existing NGOs to ask whether they’d be interested in letting us do
this you know would you let us send money straight to poor people and it was
really interesting because a lot of these organizations do run a lot of cash
transfer programming and they increasingly do but I think it’s also
you know when you think about it so nobody said yes to us right no one said
sure we’ll take your money and give it to poor people and I think that it is to
be honest it’s a threat to the business model right because fundamentally the
aid infrastructure the NGO system it exists to allocate capital on behalf of
poor people and there’s a real question about whether it does it better than
poor people themselves right and I personally think that in some cases it
can and does and that’s great we should support that but I think what we were
describing to people was in fact an existential threat to the business model
for a lot of these organizations that was the conclusion that we came to so we
decided we’re gonna have to create this and do it ourselves and so that was the
genesis of give directly the numbers when we started and the numbers now are
still very small so to give you some sense of where aid goes these numbers
are very hard to get by the way that nobody publishes this stuff and so you
sort of have to ask around and get it from people and some of these things I
probably can’t even source so maybe you should discredit it but you know for for
the British government which I’d say is one of the more cash friendly aid
agencies they support a lot of the cash transfer programs that national
governments run around the world it’s maybe a couple of percentage points of
their spending for humanitarian aids so aid to refugees to victims of natural
disasters this is probably the most cash friendly sector in foreign aid it’s
maybe seven or eight percent right now and for charitable giving for Americans
giving overseas it’s basically nothing I know it’s basically give directly
which is about fifty million dollars a year out of about twenty billion dollars
a year but Americans give each year to International Development so the numbers
are and were very very small so you know what we looked at the kind of existing
system and sort of thought about you know what we liked and didn’t like about
it I think there were a couple of things that we felt were problematic that
motivated us to create gif directly and so I want you to sort of look at this
example from a gift catalog you see these these sort of charitable gift
catalogs you can kind of go and pick out a gift for a family they’re kind of fun
right they’re kind of cute so I’m gonna pick on this a little bit I actually
don’t think these are the end of the world but you know some of the issues
with this sort of way of thinking about giving are that you know it’s a bit
made-up like the fifty dollar figure nobody actually knows kind of what the
cost is if you kind of dig in and ask people where that came from it’s
somebody in the communications department came up with a number if you
know if you read the fine print you don’t actually get to pick whether
someone gets a goat or not it says that you know the kind of the organization is
going to decide what they think makes most sense but sort of at a more
philosophical level the more I’ve thought about these gift catalogs it
raises this very basic question which is does it really make sense for me sitting
in Boston to decide whether some farmer on the other side of the world should
get a goat or a chicken or a cow or some other thing you know it seems
implausible to me that that’s the best way to allocate capital it’s about as
possible as saying that the farmers should decide what I put in my 401 K or
that they should pick stocks for me right because I just don’t know anything
at all about these people’s lives so that felt very strange to us and it also
felt very disconcerting that when you dig into the numbers and try to
understand you know where does the money actually go right and what is the impact
of it that’s very very hard to do so the aid sector features lots and lots of
layers of subcontracting you’ll often be described as sort of local partners
which could be a good thing but it creates a lot of financial opacity it’s
very difficult to figure out what the ultimate cost structure is so for
example if I raise some money from you and I spend some of it on management and
some on fundraising and then I send 95% of it to a local contractor and they
have their own cost structure I’ll call all of that program services right it
doesn’t really tell us it’s not a measure of value created for poor people
right I think I put up here as an example the Haiti earthquake which is
probably one of the more egregious examples of just not only sort
of lack of impact but just inability to figure out what was actually done with
nine billion dollars worth of money and that for reference that’s about a
hundred percent of Haitian GDP at the time right so it’s like raising
everybody doubling everybody’s income in the country for a year and not really
being able to say kind of what we did with it or what we got for it so these
are some of the things that sort of bothered us and led us to think that it
would be important to have a benchmark something some things you have compared
to compete with all of this so what give directly actually does what we set up
was sort of very nuts and bolts see it’s largely sort of anti-fraud work that we
do we’re gonna go find poor people in sort of poor parts of the country we’re
gonna do that by first locating poor areas of the countries that we work in
we’re in seven countries in Africa right now and so you know the details are
gonna differ from place to place then within these communities we’re
gonna find poor people a lot of cases we’ll just enroll everybody who lives in
a poor village sometimes we’ll use fairly simple proxies like what their
house is built out of that are pretty predictive of poverty and then we’ll
sign people up for a mobile payment system like in Pisa send money to them
monitor the payments so historically we’ve been in touch by phone with around
98% of the people that we’ve sent money to and that’s very valuable as a way of
generating feedback for us on what we’re doing well and poorly and whether people
understand the process and these sorts of things and then also collaborate on a
lot of impact evaluation so we currently have I think 13 RSC T’s that are
completed or in some stage of progression and so continuing to work
with the academic community with organizations like JPL and IPA has been
really important to us in part because what we’re doing is so unorthodox and
strange to a lot of the world I think it’s been critical to have the
most rigorous possible data to back up what we’re doing initially I think the
reaction was generally you guys are crazy
so this was probably one of the first significant pieces of coverage that we
got I remember the first six-figure check that we got from anybody there the
investment committee was reviewing this and the thing had the file had basically
gone into the trash and someone said this is crazy and then the principal
said oh well I kind of like crazy things let me have a look at that one and it
came back out of the trash and we ended up getting the check the first seven
figure check that we got they took it to the investment committee and the woman
in charge said they were days old I told her program
they were smoking crack this is an absolutely terrible idea but you know we
got a chance to meet and talk about what the data said and eventually they
decided to smoke the crack with us so I think the narrative has evolved and the
headlines kind of reflect that we’re sort of you see a more positive
narrative and more openness to the idea that this could actually not be a crazy
idea that it’s actually something worth considering and it’s something we want
to sort of compare our existing efforts to and ask whether we’re actually doing
better so this will give you some sense and I think some sense of the role that
the sort of the rigorous impact evaluation has played in the evolution
of our story as an organization I think that endorsement has been big so getting
endorsements like the one that Sam mentioned from giveWell and funding from
some other organizations that are thoughtful in their approach and kind of
analytical in their approach to giving has played a big role so these have been
some key funders or endorsers for us and so the growth I think reflects that so
we’re now about 200 people as I said we’re in seven countries across Africa
historically we’ve delivered around 90% of each dollar that we raised to people
and spend about 10% on the cost of delivery and so feel really good about
that number so give you some sense of operational performance so let me kind
of give you a sense of where that leaves us I’m gonna skip over these guys here’s
the way I think about the world today so take this quote as an example I told you
earlier that the humanitarian sector is kind of one of the more cash friendly
sectors here’s the head of the Norwegian Refugee Council writing in the sort of
yearly review I think as he says here there’s really no longer serious dispute
among people who have looked harder the data about whether cash transfers can
significantly improve humanitarian aid or aid in general but I think it
challenges established ways as he says of conducting business right in the in
the aid sector or the manatorian sector so the way we sort of think of the
problem is you know how do we as an organization fit into this broader
sector and what role can additional evidence play in trying to unblock some
of what are kind of now largely political as opposed to I think
evidentiary obstacles to using cash transfers more so I’ll give you a couple
examples of things that we’re doing along those lines
and this is one that’s largely operational so one of the concerns that
we often have about giving money to people is that it just won’t work in
very remote corners of the world right either the money won’t get to them or it
will get to them and then they won’t be able to use it because there aren’t any
markets you know it’s too expensive to travel just gonna bid up prices things
like that and so in this project which we ran in one of the most remote parts
of northern Uganda we we found that kind of the hardest place to reach that we
could imagine and to give you a sense this is a place where sort of people
live you know five miles up a mountain or five hours hike up a mountain and in
order to get Sall connectivity you have to walk around to the other side to get
text messages and then come back to the community so sort of really really hard
to reach places and so went up there and kind of dropped money on people to see
if we give these people mobile money can they get it and can they use it and so
in this case it worked out actually very well so just about everybody was able to
get their money a lot of them traveled a couple of hours to a town or a day to a
larger town and were able to spend the money and so you know we have some sense
of what the costs associated with that are but I see projects like this is very
important to kind of challenge what I think are often convenient excuses for
not trying something new and different which are you know it’s gonna be too
hard in these kinds of places and I think that’s one role that we can play
and so this is largely about generating operational evidence on what is feasible
I think I lost the slide that I wanted to show you let me tell you then quickly
about one other project which I’m really excited about and this is our ongoing
collaboration with US aid so US aid as I mentioned before spends very very little
of its money on cash transfers and US aid is also politically a very
risk-averse organization and for good reasons so they are aggressively
micromanage by Congress right they get money that is line-item allocated to
each mission saying use this money for nutrition in Rwanda and they’re going to
get lots of visits from congressmen and Senators who want to check in and see
how things are being used so they have good reason to be cautious in adopting
something new and potentially controversial like this we spent a long
time working with US aid first to create a legal framework within which they
could actually use taxpayer dollars for cash transfers and the the loophole that
we found was that at US aid if you bring in money from an external partner which
in this case was Google they can then try things that are a bit more
innovative they don’t have to go through the normal
procurement process so we use that process we worked with the general
counsel to come up with a legal framework for doing this and the
framework says things like we have to call everybody and ask them what they
did with the money and if anybody says that they used it for terrorism or for
alcohol or for birth control then Google pays for those guys and then USA and US
taxpayers don’t have to worry so kind of crazy things so you know it’s two years
of work to get that set up but we got to the point where US aid is now not only
giving money to poor people and not only running experimental tests to evaluate
the impact of giving money to poor people but actually doing that in some
cases head-to-head with their own conventional programming so we can learn
in the same sample in the same population how does what the program
that the experts have drawn up compared with what poor people do themselves with
the money so we’re early days on that journey I think of collecting and
conducting evaluations like that with US aid in several countries across Africa
the first one we got back I’d say you know the cash transfer has had roughly
the expects that I would have expected sort of fairly reasonable and and you
know largely positive as it turns out and to be honest I think this is bad for
the partnership the program that the US aid staff had designed and spent about
thirty million dollars a year on had no discernable impacts on any of the
primary metrics that it was meant to move and so that’s obviously incredibly
valuable learning for us to know that what we’re doing doesn’t work
it was optically disastrous for us aid they sort of put their heads in the sand
and refused to be interviewed by anybody and talk about this and internally some
of the people associated with the project I’ve obviously gotten a lot of
blowback because of that I would like to see a more nuanced set of results that
are perhaps a bit more favorable to the status quo so that there’s a less
concern and trepidation about continuing to do this but for whatever it’s worth
you know I think that’s an incredibly valuable lesson and I think that’s an
impact evaluation and a lesson that would not have been learned if we had
not asked the simple question of you know is it is it actually better to do
this than to just give the money to the people that we’re trying to help so that
that’s the second approach that I’m pretty excited about the third big thing
that I want to talk about and this brings us to the topic of basic income
is to consider cash transfers as a sort of long-term commitment which is there
different approach by the way from what we’ve been doing up until now so
everything that I’ve spoken about up until now we’ve given money to people
once and then walked away and there are various reasons why we thought that
might make sense there’s a lot of evidence that poor people have pretty
high returns to capital and so we sort of want to get capital out the door as
quickly as we can and not keep it in our bank account of Bank of America or it’s
losing a bit of money after inflation but there’s obviously enormous
excitement and interest right now in this issue of basic income around the
world whether it’s here in the US where we’re worried about job loss to robots
and automation and their pilots starting to be run or overseas in developing
countries in India there’s a real conversation right now about basic
income in the one of the recent economic surveys Arvind subrahmanyam the former
chief economist devoted a whole chapter to thinking about what’s the incidence
of benefits that our current social Support Scheme generates and how would
that compare to a basic income scheme and how do we pay for a basic income
scheme so we sort of thought of this in two ways I think one is that simply is a
way of getting people who haven’t thought about this crazy idea of giving
money to poor people this is a hook right this is a conversation that is
that is happening and generating interest and so it’s a way of getting
people to think about that question that they perhaps wouldn’t otherwise have and
so for that reason alone we felt like it was probably worth doing something on
basic income but the second is that when you think about basic income there
actually a lot of really interesting questions about the impacts of giving
people and potentially entire communities of basic income that we
haven’t addressed yet despite the fact that we’ve learned such enormous amounts
over the years about cash transfers of one form or the other and so a couple of
years back we made a decision to do a significant basic income evaluation of
our own and that’s what this this story is about so the evaluation is going to
take place in Kenya we’re going to be randomizing communities villages some of
them are we getting getting basic income some of them are getting getting other
cash transfer structures and some are going to be getting nothing at all and
we’re gonna be following these people for at least 12 years so the income
commitment that we make to them and the funding that we’ve raised it’s about a
thirty five million dollar project is for 12 years so both in terms of
duration and also in terms of scope this is going to be the largest basic income
study that’s been conducted to date and I want to give you a sense of the kind
of main questions that I hope answer by doing this through this
evaluation the four big things on my mind which I think in spite of all the
studies we don’t yet know enough about our first what happens not just to the
people who receive the money but to a local economy when a huge amount of
money flows in and so I want to talk a bit about that second does it matter
not just that you have more money now but that you anticipate having more
money in the future right does the commitment to a basic level of income
support matter third does it matter that individual adults which is what’s going
to happen in this study get the money as opposed to households so in almost all
anti-poverty schemes in developing countries the household is the unit of
analysis and the unit that is targeted for transfers but here we’re going to be
giving money to individuals and so you should think about things like you know
if there’s a family that’s not working is one spouse one partner more likely to
leave or two or do they have more leverage because they have the ability
to take their transfer with them and then the final one which is a really
philosophical and incredibly hard question is what do we as social
scientists actually go out and measure and say is the good when we study
something that is sort of as fundamental as economic security right what does
success look like what is a better life and how can we measure it and think
about this so I have collaborators on this project including tab neat whose
was here Deveney got me the tea and abbe Ajit
who’s a we’re both Tomita faculty here at Sloan and Abhijit in the Economics
Department in MIT these are a lot of the questions that
we’re grappling with right now so when I kind of talk you through how we’re
thinking about them and then open it up and have some discussion about basic
income or more broadly so on that first question what happens to local economies
there’s actually very little that we know about this right now we’re just
getting back results from an earlier study I’d give directly that is I think
one of the first large-scale attempts to answer this question so this is a study
where we dropped about fifteen percent of GDP on this region here in western
Kenya and looked at the impacts has had on prices on enterprise activity so not
just the people who got the money but the economy as a whole so we have a
first set of results on this that are super interesting we’re not seeing for
example big impacts on prices so we shocked GDP by about 15 percent and
consumer prices go up by about point one three percent on average we are seeing
big in on enterprise revenue as you’d expect so
people get a lot of money and they spend a lot of money and then you can trace
that money back and see how it filters through the economy to other folks it
turns out it generates an increase in labor earnings for a lot of the people
that didn’t get money as they work a bit more and earn more for their labor and
so a lot of the money ends up flowing to other workers in the economy so lots and
lots of really interesting stuff that we’re seeing from that it certainly
raises kind of in my mind the fundamental issue that we need to be
doing these evaluations at a large enough scale to be able to see effects
like this because they’re really first-order right if I want to think
about the impacts of this intervention it’s not going to be enough to randomize
something like this at the household level and ask how is my life different
from my neighbours life if I miss out on sort of spillover effects of this
magnitude so that’s one big lesson and one thing we have to be thoughtful about
so the second is this point about duration and so here I’ve sort of shown
you the three experimental arms in the basic income study the first one here is
lump sum transfers which you get right at the beginning and then they stop
right so that’s closest to what give directly historically has been doing
with most of its transfers so that you should think of as like a wealth
transfer right or a capital transfer the next two arms both give you a stream of
monthly payments that are smaller and the difference here is that in one arm
those lasts for two years and then stop and in the other the kind of long term
basic income arm they’re gonna last for twelve and so what’s really nice about
that is that when we go back to the field which we plan to do this summer
and conduct our first end line we’re gonna be able to look at two families
they’ve both received the same amount of money they both have the same amount of
additional money in their pocket but they have different expectations about
the future right and about the extent to which their livelihood their economic
security is guaranteed so this will let us isolate some of these basic questions
about how does the the prospect of economic security in the future
affect your behavior or the kind of life that you live and the way you feel about
it today which i think is one of the big open questions that we haven’t really
been able to get at yet question three is this question about family and family
structure and whether it matters that you give money to individual adults as
opposed to the household you know it’s funny because there’s a lot of received
wisdom and development that you should give money to women and to be clear I
don’t think that’s a bad idea at all I think it’s perfectly reasonable there’s
actually very little evidence on the differences between giving money to the
female or the male head of the household when we started give directly we thought
there would be lots of evidence on this and that it would be an easy decision to
make based on the data and we found there’s actually quite little there’s a
little bit that suggests that giving money to women is likely to lead to
bigger impacts for kids and a little bit that suggests giving money to men is
likely to lead to more investment and enterprise profit and things like that
and so I think there’s still actually a lot to be learned about this question
part of the reason is that because so many people were sure that giving the
money to the women is the right thing that’s what almost all programs do and
so there’s just very little variation for researchers to study experimental or
otherwise but this will be one of a handful of programs that I know of that
give money to both adults in the household and allow them to keep those
transfers even if the household breaks up so you could think of that as
something to be very excited about if women are able to get out of abusive
relationships for example so intimate partner violence is a real issue in some
of these areas you could worry a lot about what’s going to happen to kids if
families split up these are all things like very important things for us to be
thinking about and then there’s last point about how to think about value I
want to close with a story this is one of the guys who participated in the
basic income pilot his name is Ed wine so in the first village where we tested
the logistics of this we obviously don’t have systematic data we don’t have a
control group here but we went and talked to a bunch of people to try to
get some sense of what things changed in their lives and what things should be
measure more systematically when we go back to the field and so this was one of
the stories that really surprised me and the reason is I had the expectation that
if we give people a basic income they’re more likely to migrate and look for work
elsewhere right so one of the classic ideas and development is that over time
people must leave the rural areas and go to urban areas as they transition out of
Agriculture and into manufacturing and so forth
and so you know we expected to see that and we expected that getting a basic
income might make you more willing to take that risk because going to the city
to look for work is a risk right you don’t know how long it’s gonna take to
find a job or whether you’re gonna find one that you like and so the thing that
surprised me about this guy I’d mine is that he did the opposite so ed wyan had
actually already left he was working in a nearby city as a security guard as
family was back in the village and when they started to receive a basic income
he came home and the reason he did this was a
sensible once you understand the story so when he came back he worked as a
charcoal maker in the village and he made less money than he did as a
security guard but he was still able to put his kids through school because with
the basic income they have enough to pay school fees and he was able to see his
kids every night and so the challenge to us as researcher is if I go out and
measure people’s income I’m gonna say we reduced this guy’s income right I think
that all of us would say on a more human level that we’ve improved his life and
the life of his kids and we have to find some way to try to capture that and then
all the other varied forms of this right all the other there any complicated and
variegated ways in which people’s lives may be different when they’re receiving
a basic income and whether or not that don’t really boil down simply to dollars
and cents right so I think this is the hardest challenge in conducting a study
like this which is not meant to kind of narrowly improve people’s economic
security their productivity or their output or anything like that but it’s
meant to speak to sort of very basic questions about what is the good life
and do people live a better life when their Economic Security is guaranteed so
that’s what we’re working on and I love your thoughts on that but more generally
just want to stop and make time for questions and discussion so thank you by
the way outside of today if you want to follow along and read more these are
places to follow us on Twitter Annie Lowry’s book on basic income I think is
really excellent she’s very thoughtful she’s a brilliant writer she spent some
time in the field with our pilot in Kenya so you can also learn a bit more
about it there if you’re interested thanks Paul so we’ll open it up to
questions now I will I think Paul let you take the lead if anyone has a
question okay great so then I will cede my time and just let the audience yes just to clarify how do you decide how
much money each person gets is it the same all around no matter how many kids
they have it cetera yeah so we’ve done different things in different projects
you know if we’re working with a partner they may have a specific question or
goal so they want to answer and it’s going to be driven by that when when
it’s been left entirely up to us we’ve largely given the same amount to every
household and that’s typically been about one year’s budget for the average
recipient so think about it you know I’m going to come in and give you one year
of income and then leave and you figure out what to do with that I think that it
would make a ton of sense to tailor it and have it reflect the number of people
in the household if possible the number of people in the household is a hard
thing to measure reliably because if I ask you how many people are in your
family and the transfer you get is going to be a function of that you have an
obvious incentive to be generous in your description of who is in your family and
so that’s the main reason why we haven’t attempted to do that but it would be
interesting thing to try that and see you know how accurately we can you know
how much bigger is family size when I ask you and money as is riding on it
than when I don’t very interesting what’s the procedure
for sustained give directly and government cash transfer programs which
are huge and scale and also the difference between our remittances going
into poor families from from here and and give directly and the others yeah
that’s a that’s a wonderful question so I mean I think the two big differences
between what give directly does and what governments do are one is scale right as
you point out so give directly sort of at the scale of a maybe a typical you
know cash transfer program that like one African country might run but you know
there are there are programs that are much bigger and collectively government
programs are much bigger and then the flipside is probably the kind of
flexibility to innovate and to try new things so government programs are
subject understandably to kind of really tight political constraints if they want
to try something new the optics of that the way it’s going to be perceived and
written about is something we have to worry a lot about we have a lot more
flexibility to try things that we think might make sense because we’re
philanthropic lis funded right so you know a lot of lots of ways in which they
differ I think that we’re better at delivering cash transfers than the
average government program as well but I think those are kind of the two axes
that are I would I would focus on your second question was yeah yeah yeah so at first I mean it’s a very
important capital flow right so as you say it’s probably at least 600 billion
it’s very hard to estimate the kind of gray part of the remittance economy and
so that in itself dwarfs official development assistance in many years
it’s going to be bigger than foreign direct investment right that’s an
enormous capital flow to poor countries you know one obvious difference is that
that’s going to flow to people who are you know families that are at least
wealthy enough to support a migrant and so it probably isn’t going to reach the
very very bottom of the period another thing that and actually I’ve
been involved in starting a FinTech company which was now spun out a
remittance company and so we’re actually in the remittance business and this is
something that I’m very excited about I just discovered I had no idea I didn’t
know this because we’re our goal is as a company is 3% pricing on all these
quarters from Europe into Africa and it turns out that 3% pricing is actually a
sustainable development goal which I had no idea and probably tells you that we
have too many sustainable development goals but it’s nice so you know I think
the other thing that’s really interesting about remittances is that
there is at least some evidence that people sending them would like to have
some control over how they are used right because they don’t want to be
totally unconditional and so that’s an area which I think could be a really
exciting one for some innovation right so for example if you could create like
a gift register where you know if we had Amazon in Africa right if logistics
worked I could create like an Amazon gift register for you and buy you the
things that we both feel good about me sending to you so I think that would be
a really interesting frontier yeah hi I’m Alex I’ve been interested in
basic income for a number of years and I host a weekly basic income discussion
group here in Cambridge thanks thanks yeah so I’m not really a data guy I tend
to tend to think more about the theory I like to tell people you know your data
is only good as the model through which you interpret it and that kind of thing
so I was wondering about kind of the the economic theory or thinking about with
basic income and kind of what problem you’re hoping it for it to solve because
we know that you know the economy doesn’t work unless consumers have money
to spend so are you thinking of it specifically as a solution to poverty or
you think or do you ever think of it more broadly as what is the mechanism by
which the economy or we can efficiently get money into the hands of consumers in
the economy instead of you know like making up work for people to do that
kind of thing yeah great so I have a whole bunch of things I’d love to say
and I want to try to limit and make sure we have time for other people I say a
few things so first in terms of how we conceptualize cash transfers in general
whether it’s basic income or some other form of cash transfer I’ve talked a bit
here about impact and kind of comparative impacts relative to other
things we could do so I think that’s a reasonable and fairly data-driven way to
think about it if I wanted to be more theoretical I think there’s sort of an
ecosystem way to think about it which is the current decision-making model in
international development is say you’re a social entrepreneur you have an idea
maybe it’s you know providing people with solar paneled a solar solar driven
electric lights so their kids can read and do homework or something there’s
something that sounds good and you know you want to get this thing funded you’re
gonna go to San Francisco to New York to wherever you’re going to pitch it and
try to convince people this would be a great thing for poor people and
hopefully you succeed and then you’re able to fund it and deliver these things
to poor people an alternative ecosystem would be that the money has already gone
from the rich people to the poor people you have the same thing but in order to
succeed you need to market it to them you need to convince them that this
would be worth their money to buy it and so you know you still have a persuasion
problem on your hands but the kind of incentive hurdle you have to clear the
market incentive is very different and so kind of from a conceptual lens I
think that’s the way I like most of think about cash transfers it’s a
different model for decision-making which you know may be better in some
cases and worse than others but I think that’s better than thinking
of it as like should I give people cows or money that makes sense the second
thing I wanted to say is that when you think about something like basic income
as a national policy I think you can think of it either as a form of
redistribution it is for the most the simplest way to think about it and then
you can also ask might it also in addition have beneficial effects on
growth I think the first question you know can you design something that’s
basic income like and that leads to a more equitable distribution of social
resources is a fairly straightforward one and the answer is probably yes
the second question it’s a bit more open right so I mean it’s possible that you
invest more when you have this money because you weren’t able to get money
you know – from a bank because the credit system doesn’t work right it’s
possible that you didn’t invest because life is scary and risky and you weren’t
sure that you’d be able to get anywhere in life and that you feel a bit more
confident and we’re willing to take risk when you have a basic income those I
think of a sort of more open areas as more open conjectures for testing but I
think that kind of basic framework of redistribution and growth as kind of two
important objectives and thinking about how basic income fits into both of them
is helpful for me anyway conceptually thanks Paul this is the exact sort of
work that I think a lot of people a lot of us had Jay Paul find so inspiring and
you know really excites us about just working at an evidence organisation like
Jay Paul I have a question for you I think you know for the longest time as
you said the trajectory was that where’s the evidence show me the evidence that I
should be doing basic income as you saw from the elections in India both the
major political parties have now promised some version of targeted basic
income with one party going as far as saying if we come to power will give
32,000 or some really fairly high amount so this thing you know and I think the
reality is that now you’re going to be the political economy is going to
overtake the evidence bit and so this is now in three months India is going to do
some targeted basic income in you know we’ve already been doing things like
Narita but there at least were linked to something so I have a question for you
in terms of public finance advice right this is very soon going to become less
of an evidence question and more of a public finance question if I’m the
finance minister of India what should I cut so as to fund targeted basic income
and these people will come to you for advice because there not many other
people who know anything about basic income so how does one even begin to
think about that you know that that that trade-off
what should I cut you know there’s what there are obvious simple things that you
know if I’ve been transferring old-age pensions and other things those are
financial transfers I should rollover but that’s not going to fund the whole
thing absolutely so I mean first the political situation in India is
fascinating and you probably know more about it than I do I think that the my
sense is that the sort of the Congress proposal was largely a sort of desperate
bid to get back into the conversation and it wasn’t something they had really
thought through and much depth at all and now they’re being forced that they
get through in a bit more depth but I’m you know in terms of how to pay for it
it’s a great question because people often say oh this is a sort of a policy
with bipartisan support there are a lot of people on both the left and the right
who are open to it and I think that’s sort of true from the 30,000 feet and
once you get into the question of how to pay for it quickly becomes very untrue
right because what people on the right typically think would mean by support is
that they would be very happy to replace a lot of the existing safety net a lot
of the in-kind transfers and the bureaucracy that they perceive to be
very costly to society with something much simpler right like a basic income
whereas what people on the Left tend to mean is that they would like an
incremental basic income funded out of additional revenue right they don’t want
to cut Medicare or cut the old-age pension they want to add to it right and
so those are obviously fundamentally very different policy proposals right in
India you know we’ve arvind has a view on this and karthik and Sandeep and i
who kind of work a bunch in india have discussed and different people have
different perspectives and so I think it’s a live debate but you know I’ll
tell you what Karthik and Sandeep and I have kind of come to think is that it’s
very very hard politically to cut anything at least initially so our view
is start with something that is small but that you demonstrate you can do in a
reliable way so we wrote an op-ed proposing something you know start with
say 1% of GDP which would still be a lot right or 1% inclusive growth dividend is
what we called it but don’t try to cut something else initially set it up show
that it works and then over time it makes it easier to ask the question of
well we’re spending a bunch of money on this you know fertilizer subsidy or
whatever it is which is very harmful we have this direct transfer program which
is working very well why not fold it in right but I
wouldn’t want the birthing of the direct transfer program to be contingent on
cutting the distortionary subsidy say the fertilizer scheme because that means
you have avid enemies on day one and I think that makes it very very hard so
that’s kind of our current conjecture but it’s you know there are a lot of
different points of view on this and I think they all have some merit well could I say actually about one more
thing on this so the other thing that I think is actually very important if we
do switch from any sort of in-kind transfer to a basic income transfer and
this we’ve thought a lot about in the context of food subsidized food is
indexing so one thing for example that I think the current government in India
would be very happy to do is to set up a cash transfer in lieu of some existing
benefit like food and now let it lose its value over time right because I
think that would be a sort of an indirect way of cutting the welfare
state and in fact the pensions in India which you mentioned the old a pension
hasn’t increased in something like you know 20 years and every every year I get
an email from John drehs this petition they try to increase the pension and it
never goes up right so you know I think you want to be very thoughtful about
whether you index these things to changes in the cost of living or give
people some out if the value falls overtime you know perhaps the ability to
go back to getting food instead so yeah I wanted to make sure that we flagged
that thing so yeah yeah I have a question from the web audience one from
Mike is there any concern that providing large amounts of cash to villages
undermines the ability of governments in those areas yeah this is a great and I’d
say a pretty open question so I’m we we as part of the study the sort of a
larger village level randomized study I mentioned earlier looked at local Public
Finance and whether local governments in Kenya were doing anything differently
and so I totally expected to see some sort of reaction possibly people you
know government taxing some of these transfers and using it to pay for things
and we really just saw nothing like very precise zeros across the board like no
change at all in sort of the amount of revenue that local governments are
collecting or the amount that they were spending and so that surprised me and I
guess we’re still trying to figure out kind of understanding what what we
learned from that so in a sort of purely descriptive and empirical sense I say
that sort of some of what we’re learning so far I think there’s kind of a broader
question about governance in a world of cash transfers which is part of the role
of the state is redistribution part of the role of the state is also to
provide public goods right to build infrastructure and things that the
private sector might not build on its own and you want to be careful that you
know doing redistribution more effectively say with cash transfers
doesn’t crowd out the important public investments that have to be made so I
think this is something people often worry about I mean mind sort of personal
perspective on this is that it could in a way be very helpful discipline to give
people some option of receiving money directly you know if you ask people and
say what you want the government to do a lot of times they don’t say I want the
government to give me money they say I want the government to provide health
care or kind of public services or things like that but I think the issue
is that in many cases public service provision is just as a stris ly bad
right the schools are so bad that parents don’t send their kids there they
send them to the private sector cool even though the public sector school is
free so I think there ought to be ways of sort of using the direct transfers as
an option as a lever as a way of kind of creating pressure on the state to
perform more effectively in providing the public goods that should be a big
part of its role the question here are those scenarios you’ve worked with in
which you or whomever eventually stopped giving basic income what happens to
those individuals and communities so that specifically hasn’t happened for us
any the transfers that we’ve given out as I said are largely one-time so in
that sense you know they all stop immediately
we’ll have this study that I mentioned where after two years some transfers
will stop and we’ll get to learn something about the consequences of that
I guess the other thing that’s been sort of interesting which I’ll mention is
there some transfer is that in fact never even started because there are
some people who chose not to receive them and you know 98 percent I think or
so of people that we’ve offered free money to have said yes but in a couple
of areas where we’ve worked there have been a sizeable minority who have said
no and so that’s been very interesting I think that one of the things we’ve
realized is that having somebody come to you and offer free money is so strange
it is so out of the norm right it would be as if but you know Publishers
Clearing House showed up at your door with the giant check right even weirder
than that and so it actually creates a lot of space for negative rumors to
spread right so we’ve had kind of opposition politicians who want to
discredit this thing tell people that were the Illuminati that we’re gonna
come back in a year and take people’s kids and you know these things sound
insane but I think that you know what we’re doing is so insane that it doesn’t
always seem totally crazy to people that’s been an interesting lesson so you
talked about having differing levels of success in different countries so I’m
interested in which country context it’s worked and which it hasn’t and what
aspects of the country context you think lend to a more successful evaluation and
which country context it hasn’t worked in it boy so I think there are two ways
to think about success one is sort of operational performance so we know what
it costs us to deliver a dollar in western Kenya in northern Uganda in cote
de bois in malawi in sort of different places like this and so there you know
there’s just sort of performance metrics right how much does it cost us to
deliver a dollar how long does it take things like that those tend to follow
kind of predictable patterns right the places that we think of as hard to work
it costs more it might be like 20 cents per dollar delivered as opposed to ten
or five right in terms of impact I guess I wouldn’t say quite what you said which
is that it has worked in some places and not in others I’d say that when you look
across studies you see different impacts everywhere and those probably reflect a
lot of this sort of variation in context none of them are sort of obviously right
or wrong at least to me I guess you need to look at the impacts and say what’s
you feel is good or not good and you know my expectation is that it should be
very different because people are different and places are different so
you know I think that’s a topic on which it’s very hard to generalize I guess one
thing I would say is that I’m I’m interested in and excited about studying
populations who for fairly random reasons just have very little capital at
the moment so refugees is a good example I think of people who typically come
with you know the skills that they’ve acquired but largely just the shirts on
their back right it’s very difficult for them to bring any assets so we’re doing
work in Uganda for example Uganda has absorbed more refugees over the last few
years than the entire European Union has they get them from all sides and so we
think this is a population that is just a very low capital to labour ratio and
so intuitively the returns to capital there should be very high and so we
don’t have the results back from that one yet but that’s sort of my intuition
that if we want to look for places where they’re really outsized returns those
might be the kinds of places to look so I think we have time for one or two more
short questions thank you yeah thank you so much my
question is about the outcome level impact measurement that you do I
understand that on output level you are very efficient but I wonder whether you
can explain how you think about outcome level impact measurement and whether you
can use that to substantiate your claim that you might be a much more effective
alternative to the traditional NGO kind of UN system right yeah this is I mean
this is a deep question so let me imagine a hypothetical so I mean across
different studies people have measured hosts of different things right they’ve
measured impacts on schooling on earnings on education they’ve missed
there’s a paper that looks at the impacts of cash transfers on suicide
rates so I mean pretty much any outcome you can imagine there at least one or
two studies that look at it but you know let’s say to your point and of thinking
practically about the way the aid system works say we do a study and we see that
people spend you know twenty percent of the money we give them goes in to
spending on education and we see some increases in enrollment things like that
and eighty percent goes into spending on other things the interesting thing about
that is that the aid system is structured around these silos so if
you’re an education officer and you look at this the 80% of the money that poor
people are spending on other stuff is sort of a tax from your point of view
because your objective is education so if you compare giving people money to
something that you could you know an education and intervention that you
could run it’s kind of a low bar right there’s a pretty good chance you should
be able to find something that does more per dollar for education outcomes than
cash transfers right I think sometimes that low bar is actually very helpful
because the things that we do the status quo things we do or so disastrously
ineffective that even that 20% bar might weed out some of the bad stuff but you
could imagine a more sophisticated way of thinking about it which is if you’re
an education guy and I’m the health guy and people spend you know thirty percent
of the money on health maybe usually should tranche it up maybe
you should pay for twenty percent of the cash transfer and I should pay for
thirty percent of it and collectively we’d all be very happy with this even if
none of us individually would have paid for it right is something that I’ve been
interested in when you look at the way kind of the organization of the east or
the way the silos are built I guess building on that last question
on comparing cash transfers to other kinds of programs do you I mean I know
that’s not a whole heap of evidence on like direct comparisons between the two
but you have a sense of what the like next step so that analysis will be in
what the bigger answer questions are yeah it’s interesting I think there are
sort of two instincts people have on this one is like let’s find the very
best thing so it’s sort of like cash transfers seems to work pretty well
sometimes and graduation seems pretty good and bed nets are pretty good and
like let’s take the very best things we can find and horserace them and find the
very best thing so you know there’s obviously some interest in some appeal
to that I think my view is much more sort of clean up things at the bottom
end of the distribution so I think there’s just an enormous amount of silly
stuff that we waste money on that I there hasn’t been tested at all or has
been tested pretty systematically doesn’t seem to work and yet we still do
it and so the question that we’ve been asking is can we use cash transfers as a
comparison to those sorts of things to kind of force a bit more accountability
for not continuing to do those things right if you can’t do better than just
naively giving out money to poor people than like what are we here for right why
are we paying program officers we sort of want to force that question if you
would not because I think the kind of top end of the distribution question
isn’t interesting but and so I really like for example the way giveWell
publishes its reviews right so Sam mentioned that we’re one of their
recommended charities what they actually do each year is they now say here are
the five things that we think you better then give directly and here is the
multiplier like give directly x’ like 1x and we think that bed nets are like 2.5
X or something like that that’s kind of exactly the role that we’d like to play
in the broader ecosystem here and online Thanks again, all right

Leave a Reply

Your email address will not be published. Required fields are marked *