Details
Discussed by jeff shumway New York State Social Impact Bond - Anti-Recidivism 2013
Discussed by jeff shumway Philadelphia City Council Hearing On Social Impact Bonds 2016
Start Date 2016-03-02
Notes COUNCILMAN GREEN: Thank you, 14 Madam Chair. 15 I just wanted to spend some 16 time to kind of walk through the New York 17 workforce reentry transaction and get 18 some numbers in reference to that 19 transaction. 20 So what was the term of that 21 model or that project? 22 MR. SHUMWAY: So in that 23 particular model, the full project is 5.5 24 years. There are two overlapping cohorts 25 of four years each. Each one has 1,000 Page 70  STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 participants. 3 COUNCILMAN GREEN: And what 4 were the total amount of dollars? 5 MR. SHUMWAY: So 13.5 million 6 was raised from the high net worth 7 individuals and went to the Center for 8 Employment Opportunities to fund its 9 operations, its expanded operations. 10 COUNCILMAN GREEN: And then 11 what was the -- I guess for the 12 investors, what was the rate of return 13 for them? 14 MR. SHUMWAY: So there has been 15 no rate of return yet. We're still in 16 the -- 17 COUNCILMAN GREEN: Well, if the 18 metrics are made, what would be the rate 19 of return? 20 MR. SHUMWAY: Right. So the 21 baseline estimate is between 3 and 5 22 percent. There's a cap on the return 23 that can be achieved, and that's quite 24 common across these type of projects. 25 The cap in our case is 12.5 percent. So Page 71 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 if you had outcomes that were much better 3 than we had seen in the prior evidence, 4 the investors can't reap beyond that. 5 And then they can actually lose up to 90 6 percent of all the money they put in if 7 the outcomes don't meet certain 8 thresholds. 9 COUNCILMAN GREEN: So in the -- 10 and this is a question now for the entire 11 panel. For the Pay for Success bonds or 12 social impact bonds, how does the 13 borrowing costs of this type of 14 investment differ from a traditional 15 general obligation or GO bond? Page 72 16 MR. SHUMWAY: 17 should answer that one. 18 MR. RACHLIN: 19 returns to investors or 20 costs have ranged anywhere from sort of 3 21 to 6 or 7 percent in most circumstances, 22 with the possibility for larger returns 23 if things go significantly better than 24 expected, but the sort of baseline 25 borrowing cost has been in that kind of I think Andy Sure. So the the borrowing STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 range, which makes it a couple of 3 percentage points more expensive than Page 73 4 borrowing 5 6 important 7 is the question of what the risk-adjusted 8 return is on these investments. General 9 obligation bonds that are backed by the for general obligation bonds. I think that there's an sort of context for that, which 10 full faith and credit of the body issuing 11 them basically say you're going to get 12 paid no matter what. We're going to 13 borrow this money to build a highway, and 14 whether or not we build it, we're going 15 to pay you. 16 These are a much higher risk 17 investment. They're saying we're only 18 going to pay you if a whole bunch of very 19 complicated things happen, which makes 20 them significantly riskier. If you were 21 to try to calculate, that would be very 22 difficult, if you were trying to 23 calculate a risk-adjusted return. 24 Investors are basically accepting a below 25 market rate return for these investments STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 in exchange for the social good that they 3 expect them to do. Page 74 4 5 that cost 6 process? 7 8 That's an 9 COUNCILMAN GREEN: And does also include the evaluation MR. RACHLIN: Yes. Yeah. all-in net return. MR. SHUMWAY: If I can make the 10 same point from a slightly different 11 perspective. I spent many years advising 12 non-profits. The number of pilot 13 programs and non-profit programs which 14 have not shown outcomes when scaled up or 15 expanded is tremendous. Almost all 16 non-profit programs when scaled up do not 17 find the outcomes that you would expect. 18 So we're actually asking the investors to 19 take on a significant risk, a risk that 20 we have not as a society figured out how 21 to handle yet, and they're doing it at 3 22 to 7 percent. So this is a massive 23 undertaking for us as a society, and to 24 be able to shift that risk is actually a 25 major part of what we're doing. STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 I think often we have trouble 3 as the intermediary at Social Finance 4 explaining to government how massive that 5 risk transfer is. But that risk transfer 6 is a big part of what we're doing, and as 7 Andy said, we're not getting 8 risk-adjusted returns given what risk has 9 been taken on. 10 COUNCILMAN GREEN: And this 11 next question is more in reference to 12 more prospective. Looking at this model 13 and just looking at some of the attempts 14 that have been made regarding either 15 recidivism or education or homelessness, 16 if I put my former banker or securities 17 lawyer hat on, from an investor 18 perspective, I may be -- if I may be 19 interested in these types of bonds going 20 forward, I may be more likely to do maybe 21 recidivism as opposed to homelessness or 22 education or reading attainment, because 23 in the homelessness world or educational 24 attainment, there are other variables 25 that may impact the success. Whereas, Page 75 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 recidivism, it's a different perspective. 3 So are there any concerns about 4 investors going forward may be less 5 willing to invest in those type of 6 programs as opposed to like recidivism? 7 MR. SHUMWAY: I think the 8 reason the recidivism is where this all 9 started, so if you go back to the UK 10 where the social impact bond was 11 pioneered by our Social Finance 12 colleagues there, it's because there's a 13 clear cost of remediation, as Noelle 14 mentioned, and it's a heavy cost and it's 15 concentrated in a single agency. Much of 16 the cost of putting together one of these 17 is in negotiating with the government. 18 Andsotobe able to negotiate with a 19 single payor and a single agency is much, 20 much easier. 21 I don't see anything standing 22 in the way of doing this with many other 23 issue areas, except that and if you think 24 of homelessness -- and Andy can speak 25 more to this in some of the other Page 76 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 deals -- you're talking about Medicaid 3 savings, which is obviously split between 4 state and federal funding streams. 5 You're potentially talking about 6 emergency department, jail, police, court 7 costs. So you have a wrong pockets 8 problem, which gets fairly complex. 9 So that's one of the things 10 that we think through. We think through 11 that, is there interest from the payor. 12 We also think through is there an 13 intervention we believe in based on the 14 evidence, are there scalable service 15 providers, is there a target population 16 that wants this service, because, 17 frankly, there are needs that are also 18 not translating into demand for services, 19 which is problematic for us. There are a Page 77 20 number of 21 22 any major 23 be applied. 24 MR. RACHLIN: I agree. If you 25 look at -- there's, give or take, criteria here. But on its face, I don't see issue area where this could not STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 $300-plus million worth of Pay for 3 Success transactions being contemplated 4 across the country as we speak, and if 5 you look at the policy areas in which 6 they are seeking to intervene, they are 7 actually trending toward early childhood 8 education as the largest share of that 9 pie. And I think that's indicative of 10 where people's interests are. We 11 certainly were eager to invest in 12 homelessness-related and the foster 13 care-related projects that we did, and 14 other investors were as well. So I think 15 for well-constructed projects that are -- 16 that represent real opportunity, there's 17 going to be interest across a lot of 18 policy areas. 19 COUNCILMAN GREEN: Ms. Baldini, 20 you talked about an article you put in 21 Cascade, which, from my recollection, 22 that's the Federal Reserve's publication, 23 and you talked about risk and rewards. 24 For the bonds that have -- I believe one 25 or two may have already closed, and have Page 78 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 there been any concerns that if metrics 3 were not met, the impact on social impact 4 bonds going forward? 5 MS. BALDINI: Examples of ones 6 that have not met their performance 7 metrics? 8 COUNCILMAN GREEN: Right. For 9 the ones that have closed -- and I can't 10 recall if they met their metrics or not. 11 MS. BALDINI: There was, and 12 that was a significant part of the 13 meeting that we held here locally to 14 increase knowledge of this topic. On 15 Rikers Island in New York, there was an 16 example where the intervention did not 17 meet the set performance benchmarks, so 18 the city did not pay for those services. 19 And that's why right now it's really 20 important that philanthropy is also 21 interested in becoming involved. It's 22 very collaborative currently, because 23 there is quite a bit of risk and most 24 investors aren't willing to fully take on 25 that risk. So at this point, we see Page 79 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 philanthropic institutions coming in to 3 provide some risk capital. And so that 4 was certainly the case in New York with 5 Bloomberg Philanthropies. But they came 6 to speak with us and other Philadelphia 7 stakeholders at that meeting just about 8 really what did that failure mean, and in 9 a way, it proved that this tool does 10 shift the risk to the investors or in the 11 philanthropic institutions versus being 12 on the taxpayer when you're seeking those 13 outcomes. 14 MR. RACHLIN: I was just going 15 to offer that that makes us in the early 16 stages of Pay for Success or social 17 impact bonds very much like other tools 18 that have been created to achieve similar 19 outcomes over time. I think about the 20 early days of the Low Income Housing Tax 21 Credit when there were people -- it was 22 less well understood how to execute it 23 the right way. There were more failures. 24 And so investors were paying 80 or 82 25 cents on the dollar for those tax Page 80 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 credits. At this point, the Low Income 3 Housing Tax Credit world has figured 4 out -- has become very sophisticated. 5 There are very, very, very few failures, 6 and so investors are paying 98 or 99 7 cents on the dollar because they see much 8 less risk. So it's not -- it's certainly 9 not an unprecedented stage we're at. 10 MS. BALDINI: And we're also 11 interested in the opportunity to learn 12 from those failures, so how can the 13 service providers in New York share the 14 lessons that they learned. And we've had 15 that evaluation component, which in most 16 funding scenarios we wouldn't have. We 17 wouldn't be able to clearly track it as 18 well as we can through a social impact 19 bond. 20 COUNCILMAN GREEN: Madam Chair, 21 I have one last follow-up question. 22 In reference to -- Mr. Shumway, 23 you talked about negotiations with the 24 public sector and cities and getting to 25 explain the shift of risk, and I'm Page 81 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 assuming -- and I hate to assume, but 3 part of that conversation may be from the 4 public sector perspective why are we 5 paying this type of rate of return, and 6 giving examples, you said 3.5 on one 7 project up to 12.5. 8 What has been the pushback or 9 conversations from the public sector in 10 reference to that and how were you able 11 to overcome them? 12 MR. SHUMWAY: Right. 13 Absolutely. So the immediate question 14 that comes up when you explain this 15 concept is, smart folks will say why 16 don't we just pay for this service if 17 it's so good. And that's absolutely 18 possible. If government wants to 19 contract in an outcomes-based way through 20 a performance-based contract and simply 21 put some of the payment to the non-profit 22 at risk, it can do that. What you're 23 getting here is the ability to put it all 24 at risk and to shift that risk onto 25 investors. Page 82 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. 2 So the idea of paying for risk 3 is not something that government, for 4 performance risk, is not something that 5 government is particularly familiar with. 6 There are some instances where 7 performance-based contracting is used at 8 different levels of government, but it's 9 for things like -- the performance 10 outcome is cleanliness of the building or 11 uptime of the IT system. It's not 12 something that has been done in human 13 services before. So that's a bit of a 14 leap. 15 Having that conversation about 16 paying for the risk transfer is tough. 17 There are a bunch of legal things that 18 come along with that as well. In other 19 words, how you write the termination 20 clauses in the contract. So, for 21 instance, if government makes some shift 22 in policy that obviates the need for this 23 service moving forward, who is 24 responsible? How are investors repaid 25 and, if so, at what rate? How is the Page 83 STREHLOW & ASSOCIATES, INC. (215) 504-4622 Committee on Public Health and Human Services March 2, 2016 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 3/2/16 - PUBLIC HEALTH - BILL 160018, etc. non-profit made whole? How do we wind down services in a way that's responsible for the participants? That all has to be thought through in a way which is not typical of the normal contract. So that just requires education and more thought than the typical contract.