Pay for Success: Investing in What Works Part I


Jitinder Kholi:
Hi, everyone. My
name is Jitinder. I work for the Center
for American Progress. The first thing you
will notice about me is I am not American. I come from a different country
and the work I do at the Center focuses on government reform,
how you make government more innovative and how you look
for ways to make government more effective. So that is kind of my
work at the Center. We first wrote about this
concept sort of a year and a half ago and we published a
report which talked about it. And since then a lot of people
in Washington and elsewhere have gotten excited about this idea,
we are very pleased about that. Okay. So, now you all know
a little bit about me. I know some of you in
this room but some of you, many of you I don’t know
so I want to know a little bit about you. So tell me a little bit
about who you are and where you come from. How many of you come
from government? Oh, wow! Lots of people come
from government. Great. That’s good. Because I come from government. I spent 15 years in the British
government and therefore I sort of look at this very much from
the perspective of a government. And, you know, I’ve worked in
the nonprofit occasionally. And I have done other things. But my sort of slant tends to be
sort of toward the government. I feel good many of you
come from government. How many of you know a
lot about this concept? Okay. How many of you know very
little about this concept? This session is for you guys. (laughter) This session is not for those
of you who put your hand up the first time around, you know, it
is very unlikely I am going to say anything you don’t know. If you know a lot about this
concept indeed it is likely you are going to be sitting there
going it’s not exactly right the way he put it. Because I am simplifying
this on purpose. I am trying to get
the concept across, it’s a very complex concept, I
am just trying to get it across so that people
understand what it means. Through the day,
a lot of nuances; a lot of the
complexities will come out. But we wanted by 10:00 to make
sure that everybody had at least a basic understanding
of what it means, otherwise you will be sitting
there for the rest of the day going I’m not exactly
what is going on. What is that language
that they’re talking? And as you know government
is good at jargon. And there will be lots
of people today who will speak a lot of jargon. So trying to get a
lot of people sort of socialized into the space. I have got some slides. We have a tiny, tiny screen. This is going to be hard work. There is one particular
slide which is going to be particularly challenging. It has lots of little type. When we get to that we are going
to try to hand it out so people in the room have access it to. But for the moment let’s try
and work with that screen. Okay. So for those of you from
government you will be used to the way government
traditionally finances. For those of you from the
nonprofit sector you would also be used to
the way government traditionally finances. If we move to the next slide you
would be used to something that looks a bit like this. So, you know, we in government
create funding streams. Funding streams might focus on
different policy areas, so on the slide behind me you’ve got
one on helping ex-offenders get back into work. There might be a
funding stream on that. There might be a funding stream
which is like a drug treatment program to help reduce
drug dependency. There might be one
on housing support. And there might be another
one on restorative justice. Are people familiar
with that concept? Restorative Justice? So there might be
another one on that. The funding streams almost
certainly have a maximum money allocated if indicate and clicks
will get to them so I have made up some numbers for this example
of 20 million all the way down to five and these funding
streams basically will give money to providers. Okay. And almost certainly these
funding streams will come from different places. So, you know, in this particular
example you have got four government agencies, you have
got four funding streams. Each of these government
agencies give money through their funding
streams to providers. Does that sound familiar? Not calling people so far skit? Okay. So I think there are
various problems with this model and I probably want to
explore — in the room, first, before I tell you what
my prejudices are, what some of the problems
are with this model. What is kind of wrong with
the of way we do stuff here? Audience Member:
Siloed. Jitinder Kholi:
Siloed? Yup. Tell me a little
bit more with that. Audience Member:
Well, if you’re trying to take a
developmental approach like the children in need — people are
concerned about doing that in a very siloed way doesn’t meet
their developmental concerns. Jitinder Kholi:
Right, right, right, yes. Audience Member:
(inaudible) Jitinder Kholi: Interconnection,
Silo(ness), you know, agencies often
don’t work together. I think is a pretty
common problem. Funding streams to have
different constraints on them. It’s a very common
problem as well. Other thought in the room? And I want people who are not
so familiar with the concept of Pay for Success because
you all know the answers. Yes, sir? Audience Member:
The selection of providers is
sometimes political in nature or a hazard of some other, not
necessarily type of (inaudible). Jitinder Kholi:
Did you say the selection
of the providers? Speaker:
Yeah. Jitinder Kholi:
Absolutely. So the way money
flows to providers can be done in this sort of politicized
manner, is a very common problem. Yes? Audience Member:
Leave a little room for
flexibility (inaudible). Jitinder Kholi:
Right, right. Little room for flexibility. And often that’s because there
are very strong input controls each funding stream
has it’s on condition. In order to get money out of
this funding stream you have got to contort yourself into this
particular shape and then to get money out of the next funding
stream you have to contort yourself into a different shape. And that can be a pretty
unpleasant experience. But it can also make you pretty
inflexible in terms of what you can do. Yeah, absolutely. Other thoughts? Audience Member:
(inaudible) Jitinder Kholi:
Right, right, wonderful. Okay, any other thoughts? Yes? Audience Member:
(inaudible) Jitinder Kholi:
Right, if you’re lucky. Right. It might be the input, right? It might be, you know, if you
can contort yourself into a particular shape. If you promise to do all
these things we will give you the money. So another thing I would
raise is around, you know, sort of responsiveness to
performance which is very similar to your point which
is actually if something works really well so if the third
provider in the little green box there is doing an amazing job,
it’s actually quite hard to expand the money that
is going to them. Partly because you know there is
only 8 million sitting in that drug treatment program, it is
quite hard to move money from another funding stream
into that program. And often actually the dates
on whether that program is succeeding is not easily
accessible so there is sort of another set of issues there. Other issues? Audience Member:
(inaudible) Jitinder Kholi:
Inefficiencies. Yes.
Yes. Say a bit more? Audience Member:
There may be many things or
training or applying for the RTs or whatever, where as far as
justice and helping ex-offenders and drug treatment that
are all doing those things, but not doing them together. Jitinder Kholi:
Right, right, right. And often individuals, you know,
like what is missing from this diagram are the beneficiaries,
the real people who benefit, often those individuals actually
are on a wraparound service which sort of takes care
of all of their needs. What government does is is it
generates that kind of concept to individualize projects
providing one part of one the, you know, one part of, I mean,
one ingredient and it helps them but probably to different
people so, you know, person 1 gets ingredient one;
person 2 gets ingredient two. Actually, they both need
ingredients one and two. Government is not really
good at making that reality. Any other thoughts? I am very pleased at everything
you guys have said because if we go to the next slide we’ll
see some of the stuff I would have predicted. So, you know, I think,
you know, to summarize, I think some of the issues
are, you know, we fund inputs. The point you were making. Sometimes outputs but
very rarely outcomes. It’s hard to know what’s
actually working because the way we release money
is really focused on your ability to persuade us you’re
going to do what we need you to persuade yourself in
order to get our money. It is very rarely focused on
whether what you do has worked. That is not something
we think about. It is very hard to shift
money between programs. We didn’t talk
about that so much. But one particular issue is
these funding streams are set in stone often through
appropriations. If the drug treatment
program is very effective, the housing program
is ineffective, how likely are you to be able to
move the money from the housing program to the drug
treatment program? My sense of that is it is harder
than we would like it to be. Do you people agree
with that statement? And there are lots of
bureaucratic reporting. Actually you are sending reports
back and if you are in the nonprofit sector it can be
deeply frustrating to have to fill in all these government
forms and all this government paperwork all the time to
persuade them that you are doing the right thing. And then another problem is
this: If something really works, it doesn’t scale. The amazing drug treatment
program in one part of the country, you know,
you are from Virginia, how likely is it that the
person sitting over here from Connecticut is going to know
about the program that you’re funding works so well the fact
that it travels to Connecticut. You know, my sense is that
this is a really big challenge. When we at the Center got into
the subject it was because of that challenge. You know, that was the first
thing we were really worried about is how do you scale stuff. Okay. So that is a
world we all recognize. It’s a world that
we all live in. It’s a world that we occupy. And it’s a world that I sense
most of us would think things could be different. It’s not easy to work at how it
could be different because it’s the way it is for a reason. For lots of reasons. Historic and actually the world
we have just talked about is the world of governments around
the world find themselves in. It is not a particularly
an American issue; it’s an issue that applies all
around the world because of the way that governments operate. Okay. So imagine a world
which was different. And let’s call it
Pay for Success funds. And imagine a world where you
have got — keep going — okay, let’s imagine we’ve
got two people. I am going to call one of
them a government agency. And I am going to assume, Larry,
you are government agency for the purpose of this
conversation and. And I’m going to assume
someone else is what I call the dealmaker and I want a deal
maker in the room who might be a more realistic. Who wants to be a dealmaker? A cepio, if you want
to call that it jargon. Do you want to
volunteer to be a cepio? Oh, sorry, I didn’t
see you, George. It’s because I am looking over
there; you are right over here. So George and Larry. George is what I
call the dealmaker. This is a term that
no one else uses. I invented the term for
the purposes of today. I didn’t want to use the term
cepio because I thought it was kind of complex and I would
have to explain what it meant. But if you hear the term
cepio, if you hear the term intermediary, or when we
hear the term dealmaker, it means the same thing. So you are a dealmaker
in this world. And you, Larry, are
the government agency. And I want you — I kind of
imagine a world where you are going to do a deal, a
deal where basically — Audience Member:
It’s hard to imagine that too. Jitinder Kholi:
Oh, that’s right. (laughter) Do you want to
imagine that world? Audience Member:
Amen. Jitinder Kholi:
This is good. This is good. Okay. So I want a world
where you do a deal. And the deal has kind of a
particular characteristic in it. You know, in this deal George is
making a promise to deliver an outcome that you want,
that you, government, want. And you are promising George
that you are going to give George some money when he
delivers the outcome for you. Now make up an outcome. Audience Member:
Supportive Housing. Jitinder Kholi:
Okay. Larry:
Supportive housing
for homelessness. Jitinder Kholi:
Say a little more
about the outcome. Larry:
Specifically we quantify that
we have X amount of people that are homeless. We calibrate if we are were
to provide a certain number of support housing units it
would reduce it by a specific percentage which would in turn
save us X million dollars. Jitinder Kholi:
Maybe I can help out a bit and
say the outcome you want is a reduction in the number of
people who are homeless in a particular part of your state. Sort of go with
that for a minute. And George says, you know what? Why doesn’t George
answer this question? There you are, sir. George:
I think I know some people who
could do that for you, Larry! Jitinder Kholi:
Okay. So George says he has got a way
of making that a reality and, in fact, he is so confident he
has got a way of making that a reality that he is not going to
ask you for any money until he has made it a reality which
is kind of funny if you are government, right? You know, you have to pay
for it until it happens. Which another good thing about
something like homelessness it is probably costing you
something to have people who are homeless. And in deciding how much to give
him he might just be able to work out how much money he might
save and do some kind of deal with him to give him some
proportion of this savings. George:
What’s it worth to you, Larry? Larry:
I’d say maybe a third of
what we’re already paying. Jitinder Kholi::
Whoa! George:
Let’s talk. Jitinder Kholi:
Time is tight. The government is
so much tighter. Fiscally tight in Virginia? Wow! Larry:
Forty percent. (laughter) Jitinder Kholi:
You are a hard negotiator. Okay. So let’s imagine this deal:
let’s imagine from a government perspective the deal you have
done is about an outcome. It is not about a set of inputs. You have not told him what he
needs to do and the deal he has done is about a payment
at the point at which the outcome is achieved. Okay? George:
Right. Jitinder Kholi:
So that is the deal that
we are talking about. And at the heart pay for
success just means that. You know, there is a lot the
other stuff we’re going to talk about in a few minutes. But at the heart
that is all it is. It’s a relationship
between government and an outside organization. Such that, the government
promises the outside organization money in return
for the outside organization promising to achieve an outcome. Okay. Now it’s a little bit more
complicated than that. The thing about George is what
you said I know some people who can — what did you
say, you remember? George:
— who can do that for you. Jitinder Kholi:
Right. George has some people
who can do that for you. So we are gonna get
those people on the screen? Those people are providers. That is not different
than the normal world. Actually, you know, George with
all of his talents probably isn’t the best provider
of homelessness services, I am guessing? (laughter) But George is confident that
he can deliver people who, you know, who can deliver
the outcome you want. He can make that happen for you. So we have some providers
and then there is a set of relationships between the
dealmaker and the providers which are going to
flash on the screen. Keep going. And, you know, there is
two aspects of those ratios. One is, between providers and
the dealmaker there is one set of relationships which are
that the dealmaker funds and manages providers. And both of those
terms are important. Funding, because you need
funding to flow but manages, if we go back to where we
were in the earlier slide, actually that is something
government finds challenging. Managing in real-time
service delivery, you know, from third parties who are
often funded through grants, can be a real challenge. George, on the other hand, is
massively incentivized to manage these guys really well, right? Because he is not going to
give you the money, right? He is only giving you 30%. Okay. Providers however, then
providers then work with beneficiary populations and
they report progress back pretty regularly to the
dealmaker in this model. Now, the slight problem
with all of this — anyone spotted a BROT? Yes? Audience Member:
I don’t know if it’s part of it
but I think this idea of getting better results with less money. I would even question one of the
initial challenges is achieving the results within the
populations that were listing within those programs is
a difficult proposition. So I would question even
whether or not the current investment level is
proportionate to what the needs are. So, that’s part of
the chance I see. Jitinder Kholi:
Right. Right. I agree. We will come on to that. One of the things in how in
a fiscally tight world — you know, where are you from? Audience Member:
Los Angeles. Jitinder Kholi:
So how in a fiscally
tight world — well, you know your budgets are
probably going to be tighter in the coming years, right? How do you make sure that the
amount of money that flows into the services you need, you
know, provided to your people, how do you make sure that
amount of money is adequate. And one way of doing that might
be doing these kind of invest to save type deals and so I
was going to come on to that in a bit. But, yeah, absolutely. Okay. So, you know,
what is missing? Yes? Audience Member:
Money. Jitinder Kholi:
Yes, right, there is no money. No money flowing here, right? There’s no money from Larry to
George until George has achieved the outcome. I am making Kippy a provider
for the purpose of this. Kippy, on the other hand, is
actually working with homeless people to get the outcome we’re
interested in and Kippy is, like, I need to pay
salaries for my staff. And so there is no money. A bit of a problem. How might we solve that problem? Any thoughts? Audience Member:
Investors. Jitinder Kholi:
Investors, right. So we’re going to flash
up some investors. So effectively we do a deal. Effectively what George does is
he says to a group of investors, look, I am so confident I can
do this because, you know what? I have actually been up and down
the country and I found the five things that really work, you
know, in achieving this outcome. And, you know, Larry is a nice
guy but he doesn’t really have the time to go around the
country and find these people. And I can find these
approaches and I can, you know, make them a reality in the
Commonweal of Virginia. And if you investors
pay me for this, not only can I make
this a reality, but actually when
I have done it, Larry is going to pay me and
I can give it back to you. That kind of relationship
between investors who effectively pump prime, pay
for the operational capital, you might call it, they pay for
the work to be done in return for the promise that
over some length of time, they would get their money back
because they have confidence that George is a good dealmaker. And the reason I chose that term
dealmaker is because it is — the dealmaker is
right in the middle. You know, they’re other person
is who everything flows through. And they’re the person who
really needs to be engaged with this idea at a level of
detail that is, you know, quite astonishing so it’s a hard
job for them which is why I use that term dealmaker
and why, you know, focus is very much on George
in this kind of example. Okay. Thoughts on this model? Audience Member:
(inaudible)? Jitinder Kholi:
Interesting question. Evaluation. Now, when I think of evaluation,
there’s two ways to think of the word evaluation. I think one way of thinking of
the word evaluation is to think about it as, you know, those
kind of long-term analyses that are done of whether
the thing worked. And, you know, you probably need
some concept of evaluation for something like this done just
like you do for anything else. But there is another way of
thinking of that term evaluation which is, you know,
how often, you know, which is as you do something do
you learn about how to make it better in real-time? And one thing that is
interesting if you go back to that traditional model actually
we find that quite hard to do that in governments. But George, on the other hand,
would be heavily incentivized to do this because when provider
2 is not doing a great job, George would be incentivized
to make, you know, manage, provide the two who could do a
good job or shift resources from the work of provider
2 to provider 1. It could be provider 2 is just
not doing a good job with the work, you know, the process of
doing it is not very effective. It is not that they
are not very effective. You know, George was really
confident that his approach would work. But, you know, he got
that slightly wrong. So he has got incentives
to real-time manage. In away that —
for me that is a form of evaluation because it is evaluating what is happening and
working out what to do about it. But if there was an academic in
the room — is there an academic in the room? — they would tell me
it is not an evaluation. I am not an academic. Okay. That in a nutshell is pay for
success; that is what it means. Now, it can be different. You know, I have sort of
given you a stylized example. It’s not that stylized. This is actually the only
real example I have given you. In this real example which is a
prison in the UK in a town about 50 miles north of London
called Petrobras. I think you pronounce it. How do you pronounce it? Audience Member:
Petrobras. Jitinder Kholi:
Petrobras — sorry —
Petrobras — Kippy used to live in England isn’t that
hilarious — in this town called Petrobras
or Petrobras, about 50 miles north of London,
you know, the British
government, my colleagues in London, have
done a deal with an external organization basically to say if
you can deliver a reduction in the recidivism rate of the
people that come out of that prison, I am going to
give you $20 million. You have a certain amount of
time to do it and we need good measures of whether you
actually achieve it. But we’re not going
to manage what you do. We’re only going to manage
whether you achieve it on whether we give you
the money, you know, that is basically the
deal that they have done. You know, in that case that
organization is social finance, the dealmaker of social
finance is in the room, yes, social financier in the room. Is that a reasonable assumption,
a reasonable assumption of the deal you have done
in Petrobras? Audience Member:
For the U.S., yes. Jitinder Kholi:
Okay. So, you know, that particular
example the deal being done that’s being done it’s between
the social financier in the room — and they are going to be
talking on the panels this afternoon — and with the
British government who are not speaking. You know, I happened to
used to work there but I don’t right now. And there are investors, some of
who are also in the room today. Kippy, I think you are an
investor in that pilot. The Rockefeller, The Rockefeller
Foundation is an investor. (laughter) And there are providers. So that is the example. Now, it doesn’t
need to be that way. You could imagine it
being done differently. You could imagine that
actually, you know, the dealmaker is
actually a provider. You can imagine a provider
coming along saying I am so confident that I can make this
work that I am going to be the dealmaker and the provider. Yes, sir? Audience Member:
(inaudible) Jitinder Kholi:
Yeah. Audience Member:
(inaudible) Jitinder Kholi:
So I didn’t hear the first one
— I didn’t hear the second one. Audience Member:
(inaudible) That’s a program that
incentivizes the construction of affordable housing. Jitinder Kholi:
Right. Okay. So this is very different
to a tax credit. A tax credit is basically
government saying we would like to incentivize external
organizations to do certain things and if they do those
things we’ll give you money. You know, enormous deadweight
in that all tax credits have deadweight in them for
obvious reasons, you know, many of those people would do
those things anyway and there is no real concept
of outcomes there. You know, it is often
for the activity. You know, tax credit is
basically the same as a spending program just delivered
in a different way. But it is, but a performance
contract is pretty similar, too, and, you know, a traditional
pay of a performance contract has very significant
similarities with what Larry and George were
talking about earlier. There are some
differences, though. I would say one difference is
that often pay for performance contracts have a very strong
element of payment to do the operation and then a
bonus to do, you know, for achieving the outcome. In this model the bonus is like
a hundred percent and, you know, the operational
payment is like zero. Now, it doesn’t
need to be that way. It could be that there is
just some number, you know, somewhere between naught and
a hundred where it still Pay for Success. But once it gets to, like, 80
or 90 it means to me we are in a pretty standard
performance contract. The other thing
is that, you know, if you think about the way we
talked about this example this is very much around
sort of, you know, you almost came at it from
the nonprofit angle a bit. You know, it’s the
nonprofit who finds that getting money out of the
government involving grants etcetera and so you could
imagine this being something that could be of particular
benefit to nonprofits partly because of the role
that the investors play. And who are the investors? To date, they are largely
foundations who have a particular interest
in that kind of work. They don’t always have to
be foundations but to date they are foundations. Any other questions
about the concept? Yes, sir. Audience Member:
(inaudible) — federal
and local government. But there is also diminishing
funds within philanthropy, they’re also not
just diminishing, but most of these funds are
programmatically committed. So how do you create the
compelling case for philanthropy to invest not just the case, but
the case that leads to, well, that leads to
ultimate investment? Jitinder Kholi:
Right. If I may, I’m actually going to
ask you to — well, maybe I will ask Kippy to answer that
question in a little bit. Why does philanthropy care– I know you want to cover
this later in the day. I know Kippy is going to cover
this but let me just give you one quick answer to that which
is you were changing the terms of the deal for lots of
different people here. So the terms of the deal
for philanthropy is– traditionally you have given
grants and here you are giving investments and if they work
then government funds you back. The terms of the deal for
government are traditionally you have paid for inputs and,
you know, you have made those payments regardless of whether
they have been effective or not. And here what we’re
saying is you’re funding, you are only funding
when the outcome happens. So actually you are shifting a
lot of the responsibility that traditionally has been in
philanthropy or a government out to this dealmaker. And that is sort of relying on
the dealmaker taking a lot of the responsibility. Yes, sir? Audience Member:
I wonder if you could clarify
that this is two different pots of money from foundations. The traditional grants come
from the 5% minimum required by the IRS. These investments come out
of the 95% of the endowment, the investment
capital portfolio. Jitinder Kholi:
That certainly can be true. And I don’t want to say
foundation for each of them. You know, but one thing that is
different is that because these are investments rather than
grants you can imagine a world where they come out of
that endowment, the 95%, rather than the 5% which is
funded by the kind of return on that endowment. So that is a really big
difference and that, to some extent, helps to
answer your question, I hope. Yes? Audience Member:
Of the UK the $20 million in
the UK what was the time period? Jitinder Kholi:
Someone might correct me
but three to five years. Is that right? Audiene Member:
Eight for whole program. Jitinder Kholi:
Eight. Okay. Any other thoughts
on what the model is? Just explanatory stuff. We’re going to come on to
(inaudible) in a moment. Yes? Audiene Member:
I find it fascinating. This is not unlike the more
tangible hybrid public/private partnerships that are being done
in the infrastructure world. I find this, you know,
very, very similar to that. And instead of kind of being
on that tangible infrastructure side, this is really
on the social side. But the concept and the
fundamentals, you know, in terms of consortiums,
investors, performance based, sharing risk are all very,
very similar to that model. Jitinder Kholi:
Right, really, really important. You know, this is familiar
territory in a whole lot of things that government does. It is unfamiliar territory in
the things that are social to do with people. The kind of soft stuff that
government does and yet that stuff is just as important
for society, I believe, and that stuff has just as real
returns in terms of what you are trying to achieve. Okay. So if I we go to the next slide
— let me tell you what I think some of the challenges are. First of all, I think
this is complex. There are a lot of people here
who need to kind of, you know, be looking in roughly the same
direction before this happens. So it is kind of not worth it
if the numbers are very small. You know, if we are talking
about a $1 million transaction maybe that is too small. That is a kind of
question there. It is complex a relatively high
transaction cost here and so, you know, you need it to be
kind of worthwhile for it to be a reality. There is a lot of different —
the other thing I say is there a kind of, you know, I
sometimes describe, you know, we were having the kind of
conversation between George and Larry I kind of describe
that as a bit like a dance. You know, these people who
are at the center of the relationship basically need
to do a deal about what they feel is right. And that’s not an easy dance to
do and it requires high skilled, particularly at
the government end, also at the kind of dealmaker
end, to do the deal right in, — particularly in a
world where, you know, we’re worried the procurement
officers in government are left longer than they need to be. That can be a
particular challenge. You know, Larry, you need to
know how much is the right amount of money to promise. If you mis-promise, you will
lose money on this deal. George needs to know what is the
right amount of money to demand. If he mis-asks, he’ll
loses money on this deal. This is a difficult deal to do. It almost needs
to work perfectly. It needs to be measurable. This outcome needs to be
measurable in a very clear way. In the UK example there
is a control group. Every prison, not coming
out of pre-prison, and then there is the group of
people coming out of prison. And you have quite good data
about the likely chance of recidivism of the kind of
control group and therefore you can work out whether the
dealmaker is delivering the outcome that you want. Audience Member:
The control group are
the people coming out? Jitinder Kholi:
Yeah, so the control group
in the UK is every prisoner who does not come out of these prisons have
similar characteristics comes out of every other prison. Yes. Larry:
How do you measure recidivism
by those who are still in? Jitinder Kholi:
Okay.
What we’re talking about– The people by the particular
type of prisoner, you know, short-term prisoners, basically,
typically government has not provided services for
that group of people. Now it is saying it is doing
this deal thing if you can deliver a reduction in
recidivism for that community. You know, but how do
you know it is a real reduction in recidivism? And the way you know that is
by looking at other prisoners coming out of other prisons, not
beneficiaries of the service, so measurability is
really important and often a real challenge. And what’s real
measurability in this world? That’s a kind of question. Another sort of challenge
is there is nothing, in this particular model,
there is nothing stopping the dealmaker walking. Actually, you know, for year
in, you know, George is like, you know, 4 million or maybe 8
because you were sitting at the front end, you promised him 8
million and said I’m never going to get this 8 million. You know, there is no way I’m
going to achieve this outcome; he can walk. And so it’s kind of hard to
imagine this pure model for essential services, you
know, where you have a legal requirement to
deliver that service, it’s hard to use this model. It is going to work better for
other kind of things at the end of the day we’re going to
ask you to help us, you know, sort of brainstorm on where
it is most appropriate and where it isn’t. So that is where we are
going to kind of end the conversation today. And outcomes need to flow
in a sensible timeframe. Investors are not
going to wait forever. You know, it is kind
of an investor deal. This is the look, isn’t it? (laughter) It also requires enormous
political courage is another point to make. First of all, you know, actually
in the conversation you guys had, you didn’t say you
must do it this way. And that actually — and there
is a reason why government tells providers to do things in
particular ways because if they do them in different ways you
know we in government worry what if the media finds out and
then it makes big news. So actually, if George concludes
the best way to, you know, if in the petry example social
finance concludes that the best way of reducing recidivism
is giving prisoners Play Stations, that I know
would be a front page story in the national newspapers. In the UK we have a very
active tabloid press, as I am sure you all know. Actually, you know, actually the
British government can’t stop him from doing social
finance doing that. And when and if that is a
front page story in a national newspaper the secretary of the
relevant ministry doesn’t have the power to say, what do we do? How do we stop these guys? We’ve got to write them a little
thing saying you mustn’t do this because they’re like, well,
it’s working or maybe it is not working and we are stupid and
we’re going to do it anyway. That is one issue. You know, there is a real
probability of private sector return here so there is a real
possibility that people in the private sector will
make money out of this. Now whoever, it probably was
someone over there who was saying actually this is familiar
territory in other fields. And I will be willing to
make it familiar territory in this field. That is the kind of
question that I would ask. And then the other thing is it
potentially can go, you know, and this is the model we were
talking about a minute ago. We were talking about a
government agency doing the deal when we were talking,
we were saying, Larry, you are doing the detail,
but actually the could be more than one government agency. Maybe government could act
together and that requires a level of fiscal courage
that goes, you know, beyond the normal in many
government agencies, you know, many government environments. So this is not easy. There are significant challenges
to make this a reality. But there are also advantages. And I would say, you know, one
and we have talked about them briefly, one is that it allows
social — successful innovations to scale, you know, removing
input controls to providers can be an amazing thing. I know as someone who has tried
to get grants out of people the nonprofit sector
and the contracts, you know, the constraints
that people place on you are really challenging. Public money is the only
allocated approach to work that is in a particularly, you
know, fiscally tight world, in a world where, you
know, we are tighter on the money we have. That can be particularly useful. And potentially could leverage
not only philanthropic capital but also private capital. And we know there is a lot more
money in the private capital space than there is in the
philanthropic capital space. Okay. So that is “Pay for Success.” I think it is incredibly
exciting as an idea. If we look at the very
final slide, you know, I think questions that we might
explore during the day are, you know, what is the
potential of this idea? Is this idea potentially
useful to you in your world? And how would you need
to vary it in your world. Many of you are from government. Many of you have come because
you are interested in this idea. Some of you are already
implementing it. And so actually this is an
opportunity today to both learn from the people who are doing
it but also to kind of walk away with a sense of, well, can
I do something with this. There are lots of dealmakers in
the room today who can help you think through this issue so
today is an opportunity to get into that space. And I think I would encourage
you to kind of give it, you know, give it the space and
respect it deserves as an idea. I think what is really exciting
about the idea is, you know, I cannot imagine of changing
the way we do every bit of traditional
government financing. I do not imagine that world. This is not an alternative for
every dollar that flows through the traditional model. But I can imagine a proportion
of government money, particularly for services that
fit well in this kind of world are being used in this way. And I can imagine that
being potentially, you know, probably at the most
exciting thing in public financing in years. But this could be amazing if we
can start applying these kind of approaches for a proportion of
the money we spend on social issues that would be amazing. And it could lead to better
outcome for government; it could lead to better outcomes
for individuals on the ground, and even better outcomes for
investors and foundations. And why do I care about that? The reason I came into public
service is because I care about people on the ground. I care about, you know, making
the world a better place. And every time I see the bit
of money spent the wrong way in government, every time I
see something in government that isn’t quite working or
a contract is mis-managed, I kind of think surely
there is a better way. And so I am excited about this
because I think this could be a better way. So, there we are. Yes? Audience Member:
I have a question
about — I love this scale(ability) potential. I mean the incentives
to scale things up. My question is in the long view,
if something works and the cost incentives for government are
proven — we can assume that the dealmaker comes out of it
and that government pays for this new way of doing business
at scale level or what’s the tenure down the road if
everything works well? Jitinder Kholi:
I don’t know the answer to
that question and I think people aren’t thinking about
that question yet because they’re pretty excited by the
idea of trying to get it going. And I think that is the
really important question. I think sort of my
instincts there are two answers to that question. One answer to that question
is that actually it is the dealmaker who has the ability
to manage the providers so effectively that if you revert
to a model where you remove the dealmaker because these
things have been proven to be successful, actually
you will lose that. You know, it is not just who
government gives money to; it is also how government gives
money and the dealmaker does it in a different way to the
way we do it in government. That in one way
of looking at it. The other way of looking at it
is that some of those kinds of kind of ways of
working in, you know, that the dealmaker is starting
to make a reality are actually applicable and trying to mention
in the public sector I don’t know the answer. There is no, you know, I think
that is something that people need to work through. I think we have
got an end thing. My watch says it is
like 50; is that right? I am going to take two
more minutes so people can do some questions. Yes? Audience Member:
My question, going back
to the evaluation issue, whose standards of
evidence prevail? Clearly the evaluators will
have different standards. There will be disagreements
over what constitutes reliable credible evidence. How does that get worked out
among the various parties? Jitinder Kholi:
I think the answer is that is
part of the dance, you know, if we go back to the
deal that is being made, the dance between, you
know, Larry and George, they are going to have to
decide the answer to that question in advance. They cannot renegotiate the
answer to that question every sort of six months. And so it happens in the UK that
this is not a very hard question to answer because there is lot
of quite good data available. But often that won’t be true. Often data will be dirty,
data will have lags, and so actually working out that
question early is important. You know, you may well
need a third party. Audience Member:
— is in the investor’s
perspective on that pretty critical or is
it mostly between — Jitinder Kholi:
Well, I would say the central
deal is between these two guys. Now, the reality is George can
only do the deal because he has got some investor sitting behind
him and the investor will have a stronger view about what
is acceptable and they need persuading. But I am hoping Larry doesn’t
have to talk to the investor. And the question that you are
asking is really about under what circumstances would
Larry release the money. Yes? I think you might be
the last question. Audience Member:
I have a question about payback
really from two points of view. One, if you are
taking the UK example, eight years down the track there
is a political risk in terms of whether the government or
the time is going to be in a position to fund it. And secondly, the payback could
be calculated on the basis of savings over many years the
government would be making so that the sum could be quite a
significant amount to pay at any given moment. So how would
government manage that? Jitinder Kholi:
I think the political
risk is real. I am really pleased
you raised that. You know, I think the
dealmaker is taking a risk that the future Larry
will probably try and renege on the deal. Now actually, I think that
deal exists in lots and lots of contracting between governments. You know, when you are doing
negotiating to build a road or a bridge or Olympic stadium,
as what we are doing in the UK, that is an issue in
that kind of contract. So there is a kind of series of
issues around governmental change and that is a risk you
are taking into account. You know, I suppose, you know,
public budgeting has to take into account in long-term cost
savings and I think there is a question whether the
appropriation system is comfortable with that. Now, while we are talking
about north point north, north, north, north — I don’t
know how many north point one percent of public finance,
then this is like a small issue. When the numbers get bigger,
this would be an issue and I think people need
to get into it. Okay, I think I have some funny
faces saying why don’t you stop that I think the time
to stop has come. (laughter) In about seven minutes,
I think the next bit of the session starts. Oh, no, now, okay, not at 10:00. Those people coming at
10:00 are going to be late. Okay. So I am here all day. Anybody have any questions,
come and grab me. And good luck! Enjoy. (applause)




Leave a Reply

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