New Incentives — General Support (2014)

[Added December 19, 2016: GiveWell's experimental work is now known as GiveWell Incubation Grants.]

New Incentives' response to this page is available here.

Note: This page summarizes the rationale and details behind a grant to New Incentives made by Good Ventures (in partnership with GiveWell according to the framework laid out previously). The page also discusses New Incentives' progress since the grant was made. Because of the relatively small size of the grant to New Incentives, we did not spend as much time vetting New Incentives and writing up the details of our investigation as we would for a top charity.

Much of the content on this page came from conversations with New Incentives; rather than citing each statement, we asked New Incentives to approve the content on the page in a process similar to approving conversation notes. See the process section for more details.

This page reflects our understanding of New Incentives as of December 2014. New Incentives has shared several major updates since then, which we have briefly noted below but not fully incorporated into the page.

Summary

Grant to New Incentives: In January 2014, Good Ventures (with input from GiveWell) made a $100,000 grant to New Incentives. The grant was the first of a set of grants that aims to increase the number of contenders for GiveWell's top charity status.1 New Incentives was selected for the grant because it runs a priority program, is committed to transparency, and is at an early stage of development (more).

What do they do? In June 2014, New Incentives began a conditional cash transfer (CCT) program in Akwa Ibom State, Nigeria. The program aims to incentivize mothers to prevent mother-to-child transmission (PMTCT) of HIV by giving them cash transfers for collecting antiretroviral drugs (ARVs), delivering in a clinic, and getting their newborns tested for HIV (more).

Does it work? There appears to be strong independent evidence that CCTs and PMTCT are effective programs. New Incentives does not yet have a track record of successfully implementing its program. Women seem to have dropped out of the program at a high rate, and it is not clear why (more).

Cost-efficiency and cost-effectiveness: We estimate that 76% of New Incentives' expenses will fund cash transfers. New Incentives' PMTCT CCT program could potentially be in the same range of cost-effectiveness as our top charities, though we are very uncertain about many of the inputs into our analysis (more).

Funding situation: The grants made by Good Ventures and the Lampert Family Foundation seem to have been crucial to allowing New Incentives' CEO and Founder Svetha Janumpalli to move to Nigeria to start New Incentives' recent work. New Incentives is seeking a larger amount of funding in 2015 to expand its PMTCT work and start a new program (more).

Bottom line: We believe that in 2014, New Incentives has made substantial progress towards possibly becoming a GiveWell top charity in the future. However, the program is still at a very early stage, and we are unsure how effectively New Incentives will retain participants, how much monitoring data it will be able to collect, and how cost-effective its program is. We plan to continue following New Incentives, to consider it for future grants, and to eventually consider it for a top charity recommendation.

Published: July 2015

Background on the grant to New Incentives

In January 2014, Good Ventures (with input from GiveWell) made a $100,000 grant to New Incentives.

GiveWell, Good Ventures, and the Lampert Family Foundation collaborated on the process of vetting New Incentives. The Lampert Family Foundation also made a $100,000 grant to New Incentives in 2014.

As noted in this blog post, GiveWell recommended that Good Ventures make the grant to New Incentives primarily because (a) it runs a program, conditional cash transfers (CCTs) that has a significant evidence base and (b) it is a startup organization with a founder, Svetha Janumpalli, who we believe is committed to allowing us to thoroughly vet New Incentives' progress. Due to these factors, we believe it is plausible that New Incentives will eventually become a GiveWell top charity. We are open to recommending grants to other groups meeting similar criteria.

What do they do?

In June 2014, New Incentives began a CCT program in Akwa Ibom State, Nigeria to incentivize mothers to seek treatment to prevent mother-to-child transmission (PMTCT) of HIV. The program gives women cash transfers for collecting antiretroviral drugs (ARVs), delivering in a clinic, and getting their newborns tested for HIV. In the future, New Incentives plans to conduct a randomized controlled trial (RCT) of its PMTCT CCT program, and it is considering starting an additional CCT program.

New Incentives' history

New Incentives' CEO Svetha Janumpalli founded the organization in 2011 with funding from the Lampert Family Foundation.2 To test a range of possible CCT programs, New Incentives provided funding to four NGOs in different countries to run small pilot CCTs incentivizing various behaviors.

Ms. Janumpalli worked on New Incentives full time from July 2011 to June 2012. From June 2012 to early 2014, Ms. Janumpalli worked on New Incentives part time. The pilot CCTs continued to run on a very small scale.

In March 2014, Ms. Janumpalli moved to Nigeria to work full-time to implement New Incentives' PMTCT program there. New Incentives selected Nigeria because it believed that the country had high rates of extreme poverty and maternal HIV and that it was underserved by charitable organizations.

New Incentives began to run its PMTCT program in Akwa Ibom State in June 2014. New Incentives chose Akwa Ibom because New Incentives believes it to be an area with high poverty, high prevalence of HIV, and limited use of PMTCT services. New Incentives told us that before it started its program, many women visited clinics once to register their pregnancies, but few of them returned to receive follow-up care or to deliver in the clinic.

Program structure

New Incentives focuses on the last trimester of pregnancy, delivery, and the first months of breastfeeding.3

According to New Incentives, about 40% of cases of mother-to-child transmission of HIV occur during pregnancy, 40% occur during birth, and 20% occur during breastfeeding.4

The program in Akwa Ibom includes three cash transfers to participants:

1. First transfer: After HIV-positive pregnant women register with New Incentives, they receive a transfer of 6,000 Naira (about $30).5
2. Second transfer: If they give birth in the clinic where they registered with New Incentives (where clinic staff give PMTCT care), participants are eligible for another transfer. The transfer is 6,000 Naira (about $30) plus 2,000 Naira (about $10) for each month during pregnancy that the participant picked up her ARVs, for a total of up to 14,000 Naira (about $70). Participants cannot receive the reward for picking up ARVs if they do not also deliver at the clinic.
3. Third transfer: Participants are eligible for the third and final transfer if they get an early-infant HIV diagnosis test for their newborn, which is given about six weeks after birth. The transfer is 6,000 Naira (about $30) plus 2,000 Naira (about $10) for each month after giving birth that the participant picked up ARVs for herself and her baby. The total for this transfer is up to 10,000 Naira (about $50).

In total, participants can receive 30,000 Naira (about $150) over the course of the program.

New Incentives' process

In this section, we describe the process New Incentives has told us it uses to determine eligibility for each cash transfer. Below, we discuss in more detail what evidence we have seen that New Incentives has carried out the process as described.

Registration and first transfer

As of October 2014, New Incentives worked in three clinics (hospitals and primary health centers) serving impoverished populations with high rates of HIV.6 The clinics hold weekly Booking Days on which women come in to register their pregnancies, and the clinics give them antenatal care, including HIV testing. New Incentives trains nurses to refer HIV-positive pregnant women to New Incentives for the CCT program.

A New Incentives field officer explains the terms of the program to each woman and enrolls those who agree to participate. Terms of the program include:

• New Incentives keeps women's participation in the program private.
• Women are eligible for New Incentives' program only once in their lives, to avoid creating an incentive for women to become pregnant again to be able to participate. New Incentives takes a picture of each new participant's face, then uses biometric software to compare the new photo to the existing database to make sure that it does not match any of the previously enrolled participants.
• Participants must allow New Incentives to take photos of their patient cards (their personal medical records) and of their abdomens. New Incentives uses this information to verify the participants' pregnancy, HIV status, and use of ARVs.
• Some participants are asked to participate in further verification checks. Field officers randomly select at least 1 out of every 10 participants and ask nurses to give her another HIV test in front of the field officer. Field officers also randomly give pregnancy tests to at least 1 out of every 10 participants who are not visibly pregnant.

New Incentives distributes cash through payment codes redeemable at branches of a major Nigerian bank, which New Incentives told us are common in Akwa Ibom. After New Incentives checks the information that new participants gave at registration, participants call New Incentives to request the payment code. To receive the code, the participant must verify her identity based on her personal information and a password given to her during registration. Occasionally, participants call New Incentives to report that their codes are not working, and in these cases, New Incentives verifies the issue and then issues a new code.

Second transfer for delivery at the clinic and collecting ARVs

When they enroll in New Incentives' program on Booking Day, participants are told that they should call New Incentives' hotline after they deliver in the clinic to receive the second cash transfer. When a participant calls in, New Incentives' field officers ask questions to verify her identity, ask about the date and location of her delivery to check against clinic records, and administer a short survey about how she spent the first transfer. New Incentives checks information about her delivery and use of ARVs against at least two sets of health center records. Each set of records is maintained by a different staff member at the clinic. If the participant has met the conditions for the second transfer, New Incentives then gives her the payment code.

Third transfer for HIV testing and ARVs

The process for the third transfer is similar to that for the second. Participants are asked to call New Incentives two months after delivery if they have had their children tested for HIV. New Incentives asks them to take a survey about how they spent the second transfer. New Incentives' staff use clinic documents to verify that participants have picked up ARVs and had their children tested. If they qualify, participants then receive the code for the third and final transfer.

Preventing fraud

Here, we describe New Incentives’ processes as the organization described them to us.7 Below, we discuss in more detail what evidence we have seen that New Incentives has carried out the process as described.

Clinic staff

• As mentioned above, New Incentives checks information used to determine eligibility for transfers (such as whether a woman is pregnant, has HIV, and complies with conditions) against clinic records maintained by multiple clinic staff and often by multiple departments within a clinic or hospital. The goal of these checks is to make it more difficult for clinic staff to extort bribes by enrolling women who are not eligible for the program or threatening to record inaccurate information about women who are eligible.
• New Incentives asks participants how much clinics charge them for care in order to monitor whether participants are asked to pay more for care than other women who use the clinic.
• If at some point in the future New Incentives believes that clinic staff are engaging in fraud, it may offer rewards for whistleblowers who expose fraud or employ mystery clients to visit clinics on New Incentives' behalf to monitor fraud.

Field officers

New Incentives told us that headquarters staff randomly conduct in-person audits of field officers' enrollment sessions. Field officers' calls with participants are recorded and headquarters staff audit a random portion of these calls. Headquarters staff also verify key enrollment records for all participants.

Partnerships

New Incentives partners with the State Ministry of Health. A Ministry staff member attended each New Incentives enrollment day to monitor New Incentives' work during the first few months of the program.8

There is a large PMTCT program in parts of Nigeria including Akwa Ibom, run by FHI 360. The program seems to be focused on improving the availability of PMTCT, with some funding for demand creation but not for CCTs.9 FHI 360 has shared data and worked closely with New Incentives. New Incentives has hired two of FHI 360's former volunteers as field officers.

Expenses

New Incentives' projected expenses for its PMTCT CCT program for 2014-2015 are:10

Category	Expense	Percentage of total expenses
Transfers	$331,266	76%
Transfer admin	$12,773	3%
Field staff	$11,040	3%
HQ staff	$46,000	11%
HQ costs	$32,960	8%
Total	$434,040	100%

These expenses are projected for a total of 1,781 participants, for a cost of $244 per participant, of which $186 (76%) is transferred to the participant.11 This budget assumes that all participants enrolled in the program will fulfill all of the conditions of the transfers. We would guess that compliance will be imperfect, which could reduce the total spending and the percentage spent on transfers.

Update: In January 2015, New Incentives sent us an updated budget for 2014-2015, but we have not yet reviewed this new information.

Future plans

• In 2015, New Incentives plans to spend $314,540 and enroll about 1,400 women in its PMTCT program.12 To do so, New Incentives plans to scale from the 3 clinics where it worked in October 2014 to 8 clinics in 2015. New Incentives does not anticipate that it will need to hire additional field staff for this expansion.13
• New Incentives plans to conduct a randomized controlled trial (RCT) of its PMTCT program with a total of about 500 women.14 Women in the treatment group will participate in New Incentives' usual program; women in the control group will be first contacted 8-10 weeks after they deliver and offered an incentive to take the endline survey. The primary outcome of interest will be study participants' use of PMTCT services.15 New Incentives plans to rely on administrative data on use of PMTCT services.16 As of January 2015, New Incentives had identified a research team and planned to conduct the study in 2015.17 New Incentives told us that the study will be pre-registered.18
• In late 2014, New Incentives was considering a new CCT program to incentivize mothers to vaccinate their infants.19 New Incentives believed that it could enroll many more women per clinic in CCTs for vaccinations than in CCTs for PMTCT, because all pregnant women, not just women who have HIV, would be eligible.20 Adding CCTs for vaccinations would allow New Incentives to expand its programs, including CCTs for PMTCT, to new clinics because there would be enough women eligible for CCTs for vaccinations to make it feasible to operate in facilities with a lower volume of HIV-positive patients. This would allow New Incentives to expand further in Akwa Ibom rather than expanding to new states or to clinics with a very low volume of eligible women. New Incentives estimated that the CCTs for vaccination would be somewhat less cost effective than CCTs for PMTCT.21

  Update: As of February 2015, New Incentives has finalized its assessment process for a second CCT program and decided against a CCT for vaccinations. Instead, it is planning to start a CCT program to incentivize women with high risk pregnancies to deliver in clinics.22 We have not yet reviewed details of this plan.

Does it work?

This section addresses the following questions:

• What is the independent evidence of program effectiveness? We believe the academic evidence bases for CCTs and for PMTCT are fairly strong, though we have not recently undertaken in-depth research on these interventions.
• How poor are the program participants? We do not have a very strong sense of how poor participants are. The vast majority said they had permanent floors and metal roofs. Participants reported that 28% of the children they had previously given birth to had died.
• What do participants buy with the transfers? Participants report using transfers to buy baby supplies, food, medicine, and goods to sell.
• What proportion of women follow the conditions and are retained in the program? Only about one quarter (8 out of 33) of women who enrolled in the program had received the second transfer by the time that they were expected to. Our understanding of why women dropped out of the program is very limited.
• Is there any evidence of fraud or coercion? New Incentives has looked for evidence of fraud or coercion. About 5% (9 out of 166) of participants reported that they paid a bribe to bank staff to withdraw their transfers.
• What are the possible negative and offsetting impacts of New Incentives' program? It may be hard to maintain the confidentiality of some women's participation in the program, and the program may create perverse incentives or crowd out other reasons for seeking PMTCT treatment.

We note that New Incentives' current program is very new. The data summarized below cover participants who enrolled in the program between June and November 2014.23

What is the independent evidence of program effectiveness?

Cash transfers are one of our priority programs, though we have focused on unconditional cash transfers and have not researched how effective conditional cash transfers are at changing behavior. We reviewed PMTCT in 2009 and found that it was a promising intervention, though we have changed our process for evaluating interventions since then and our research from 2009 is now out of date. To more thoroughly research New Incentives' effectiveness, we would review the academic evidence to update our views both on conditional cash transfers and on PMTCT.

How poor are the program participants?

New Incentives shared monitoring data it had collected through November 2014.24 Data for some participants is missing for some questions because New Incentives did not ask certain questions when it first started the program and because some questions apply to stages of the program that not all participants have yet reached.

New Incentives collects data at registration on participants' level of poverty:

• 9% (15 of 168) of participants reported that they had earthen or straw floors (the rest had permanent floors made of concrete, wood, or tile or had "other" types of floors), and 9% (14 of 155) reported that they had mud or thatch roofs (the rest had aluminum, zinc, or "other" types of roofs).25
• Participants' reports suggest that 28% (105 of 379) of the children that they had given birth to over the course of their lives had died by the time the participants enrolled in New Incentives' program.26
• Participants reported that they make a median of 2250 Naira (about $11) per month.27 However, we are unsure whether women accurately reported their earnings. Data on their husbands' earnings are difficult to interpret.28

What do participants buy with the transfers?

Participants generally reported that they bought or planned to buy baby supplies, food, medicine, or goods to sell.29 It is hard to say from these self-reports how the transfers affected participants' overall spending, since participants may use the transfer money to buy things that they would have bought without the transfers, while reallocating other income to other expenses or savings.

What proportion of women are retained in the program and comply with the conditions?

As of November 27, 2014, New Incentives had enrolled 227 women.

• 142 of 176 (about 81%) of women who registered with New Incentives before November 14 picked up the first transfer.30
• 8 of 33 women (24%) who registered with New Incentives and who the clinic projected to have delivered by November 1 had picked up the second transfer by the end of November.31 New Incentives told us that some participants who did not receive the second transfer failed to qualify because they did not give birth in the clinic.32 In other cases, New Incentives said that it lost touch with women before they delivered.

New Incentives believes that retention through the second transfer will be higher in the future because New Incentives' field officers are becoming more experienced, are fluent in local languages, and are better able to explain the program to participants than Ms. Janumpalli was when she was running the program on her own.33

We are unsure how many women stopped communicating with New Incentives relative to how many did not give birth in a clinic.

Few participants have reached the third stage of the program at this time, so we have limited information on what portion of women are retained through all stages of the program.

New Incentives' response to this page discusses the issue of retention and the steps New Incentives is taking to address it.

Is there evidence of fraud or coercion?

• Are women who are not pregnant signing up for the program?
  • New Incentives takes a picture of each participant to record whether she looks visibly pregnant.34
  • New Incentives told us that it randomly selects at least 1 of every 10 participants who does not clearly appear to be pregnant to receive a repeat pregnancy test. We have not yet asked to see the results from these tests.
  • New Incentives shared with us photos of each participant's patient card. We do not know how to interpret the patient cards. On the few we checked, clinic staff had recorded medical information from an antenatal examination, including what we interpret to be the position of the fetus.
• Are HIV-negative women signing up for the program?
  • We believe that the patient cards show that clinic staff recorded that each participant has HIV based on a diagnostic test.35
  • New Incentives randomly chooses at least 1 of every 10 participants to receive a repeat HIV test. So far, 32 of 227 women (14%), have been retested. New Incentives staff take a photo of the test results. New Incentives has reported that all tests so far have been positive.36 We have not asked to see the photos.
• Do New Incentives' field officers, clinic staff, or bank staff take a cut of the transfers?
  • 9 of 166 participants (5%) reported paying a bribe to bank staff to collect the first transfer. (New Incentives did not report the size of the bribes.)37
  • We have not yet asked to see New Incentives' data on clinic costs for participants versus other women who use the clinics. We believe this analysis would be helpful, but not sufficient, to know whether clinic staff are taking a cut of transfers.
  • We have not yet asked to see the results of the audits of field officers conducted by New Incentives headquarters.38
• Do participants control how the transfers are spent? We have limited information on this question, because the data we have seen includes only 9 women’s responses to questions about whether other people influenced how the participant spent her first transfer or forced her to use it in a certain way. Those 9 participants reported that no one influenced or forced their spending.39

What are the possible negative and offsetting impacts of New Incentives' program?

• Privacy concerns. HIV-positive women may face stigma if other people learn their HIV status. One participant called New Incentives to say that her husband found the consent form that New Incentives gave her. He erroneously believed that the form indicated that New Incentives would publish pictures of his wife on the Internet. He claimed that he would leave her if she did not drop out of New Incentives' program, so she asked to be removed from the program. In response to that incident, New Incentives clarified the language on the consent form.40

  As more women move through the program and collect cash transfers, bank staff may be able to recognize that women receiving transfers of a certain amount are likely program participants.

• Incentivizing contracting HIV or getting pregnant. Because New Incentives provides transfers only to HIV-positive pregnant women, women might intentionally contract HIV or become pregnant in order to be eligible for the transfers. Based on a survey among local stakeholders, New Incentives believes that the transfers are not large enough to encourage women to contract HIV or become pregnant. In addition, New Incentives allows each woman to enroll in the program only once in her lifetime. If this rule is enforced consistently, it should eliminate the incentive for women who have already participated to get pregnant again.
• Crowding out other reasons to seek PMTCT care. Women who participate in or learn about the program may come to believe that the main reason for them to seek PMTCT care is to earn cash transfers rather than to protect their babies from HIV. If so, they may not seek PMTCT care if they are unable to enroll in New Incentives’ CCT program.

Cost-efficiency and cost-effectiveness

What percentage of New Incentives' expenses are given as transfers to recipients?

We have not yet seen actual spending data from New Incentives. Based on its budget and expected number of participants for May 2014 to December 2015, we estimate that 76% of its expenses on its PMTCT CCT program would be transfers to participants.41

What is the impact per dollar of New Incentives' program?

Based on a very rough cost-effectiveness calculation, we believe that New Incentives' PMTCT CCT program could potentially be in the same range of cost-effectiveness as our top charities. However, we have limited information about many of the parameters that go into the analysis. To refine our understanding of New Incentives' cost-effectiveness, we would continue to observe New Incentives' progress, ask New Incentives about some of the parameters in the analysis, and conduct further research on the academic evidence on CCTs and PMTCT.

Note that our calculations linked above do not take into account the benefits to participants and their families of cash transfers themselves but do include the costs of the transfers. The only benefit we include in the analysis is PMTCT. A more complete analysis of New Incentives' cost-effectiveness would seek to also include the benefits of the cash transfers and the benefits of PMTCT care to the health of participants themselves.

Funding situation

Funds received

Before receiving the grants from Good Ventures and the Lampert Family Foundation in early 2014, New Incentives had received $40,000 from the Lampert Family Foundation in a previous grant and about $30,000 from other sources. New Incentives tried unsuccessfully to raise funds from multiple sources. Accordingly, we believe that the grants from Good Ventures and the Lampert Family Foundation were likely key in allowing Ms. Janumpalli to move to Nigeria to work full-time on New Incentives.

New Incentives expects that the grants from Good Ventures and the Lampert Family Foundation will cover conditional cash transfers for about 1,000 participants, sustaining New Incentives' operations through early 2015.42 New Incentives had about $186,000 on hand on September 29, 2014.43

In November 2014, New Incentives announced that it had received a $100,000 Gates Foundation Grand Challenges Exploration Grant, which it will use to fund the PMTCT RCT.44

Room for more funding

New Incentives projected that it would need about $286,000 in 2015 to expand CCTs for PMTCT and an additional $529,000 for CCTs for vaccinations to run that program at full capacity.45

Update: In January 2015, New Incentives sent us an updated budget for 2014-2015 proposing a high-risk pregnancy program instead of the vaccination program and requesting $378,464 for that program. We have not yet carefully reviewed the new budget.

Ms. Janumpalli believes that a number of funders interested in HIV and CCTs might fund New Incentives once it has scaled up and has RCT results to demonstrate the effectiveness of its programs. These funders may include the MacArthur Foundation, the Clinton Health Access Initiative, Ashoka, the Omidyar Network, and the Mulago Foundation.

New Incentives as an organization

New Incentives is run by CEO and Founder Svetha Janumpalli and Chief Strategy Officer Patrick Stadler. New Incentives currently employs two field officers in Nigeria to run its program.46

To date, New Incentives has been very willing to share detailed information and documentation about its program with us.

In September 2014, New Incentives told us that it had lost its tax exempt status in the US because it had failed to file its Form 990 with the IRS.47 New Incentives had thought that it had two more years to file its first Form 990. New Incentives noted that it did not receive advance warning from the IRS. New Incentives regained its status by December 2014,48 and we do not believe that this issue will have any significant negative impact on New Incentives. New Incentives has retained a lawyer to avoid similar problems in the future.

We feel that this issue reflects somewhat negatively on New Incentives' ability to administer the organization. On the other hand, New Incentives seems to have responded promptly and thoroughly to the problem, and sharing the incident with us reflects well on New Incentives' transparency.

Major questions for further investigation

• How strong is the academic evidence base for CCTs and PMTCT?
• How has New Incentives spent funds to date?
• Why has retention in New Incentives' programs been low? Is it improving over time?
• How much of the monitoring data that New Incentives intends to collect will it be able to collect?
• How cost-effective is New Incentives' program?
  • How much does it cost clinics to provide PMTCT services?
  • What are the baseline rates of mother-to-child transmission of HIV and utilization of PMTCT services in Akwa Ibom? (New Incentives has provided us with some approximate figures, but we have not seen the full details.)

Our research process

New Incentives founder Svetha Janumpalli reached out to Good Ventures in 2012.49 GiveWell then had four conversations with New Incentives' staff and partners to discuss New Incentives' history, plans, and funding situation in order to evaluate New Incentives for a potential grant. Good Ventures and the Lampert Family Foundation worked with GiveWell to vet New Incentives and then made grants in January 2014.

Since then, GiveWell has had four follow-up conversations with New Incentives and has reviewed documents sent by New Incentives. Rather than publishing notes on the individual conversations, we sent this page to New Incentives for approval. This review is less detailed than a review of a GiveWell top charity because the grant to New Incentives was relatively small, and New Incentives has not yet established the track record it would need to become a top charity.

We plan to continue to speak with New Incentives about once every three months to keep updated on its progress. For notes from our conversations with New Incentives, see our conversations page.

Sources

Document	Source
Binagwaho et al. 2013	Source (archive)
Conversation with New Incentives on January 20, 2015	Unpublished
Conversation with New Incentives on October 28, 2014	Unpublished
Conversation with New Incentives on September 19, 2013	Unpublished
Conversation with New Incentives on September 29, 2014	Unpublished
FHI 360 2013	Source (archive)
GiveWell's non-verbatim summary of a conversation with New Incentives on Dec 13, 2012	Source
Google currency converter	Source (archive)
New Incentives beneficiary data June-November 2014	Unpublished
New Incentives budget 2014-15 (with GiveWell's analysis), October 2014	Source
New Incentives draft concept note for RCT, October 2014	Source
New Incentives draft overview of CCTs for high risk pregnancies, January 2015	Source
New Incentives draft overview of CCTs for vaccinations, October 2014	Source
New Incentives draft RCT timeline, October 2014	Source
New Incentives fraud detection and prevention strategy, October 2014	Source
New Incentives Gates Foundation Grand Challenges Press Release, November 2014	Source (archive)
New Incentives operational overview, May 2014	Source
New Incentives, update for GiveWell, August 2014	Source
Siegfried et al. 2011	Source (archive)
Svetha Janumpalli, email to GiveWell, December 6, 2014	Unpublished

• 1

  Good Ventures has also made grants to IDinsight and Evidence Action as part of its effort to increase the number of contenders for GiveWell's top charity status.

• 2

  Conversation with New Incentives on September 19, 2013.

• 3

  In addition to various conversations with New Incentives, the following subsections of the “What do they do?” section are informed by New Incentives operational overview, May 2014.

• 4

  We have not asked for data to support these claims.

• 5

  Currency conversions from Google currency converter; as of February 25, 2015, 1 NGN = 0.0050 USD.

• 6

  New Incentives budget 2014-15 (with GiveWell's analysis), October 2014, "CCT implementation expenses" sheet, "Nr of active clinics" column says that in "Oct 2014" there were 2.5, which we have assumed means 2 at the start of the month and 3 at the end of the month. We have not asked for evidence of the level of poverty and HIV prevalence in these health facilities.

• 7

  This section on preventing fraud relies mostly on New Incentives fraud detection and prevention strategy, October 2014.

• 8

  "Ministry of Health: We have developed a close working relationship with the Ministry of Health. They send a staff member on each day we go to clinics to ensure we are adhering to the agreements we set regarding which types of records we can document and the type of data we can collect. The presence of the Ministry has helped us make several improvements based on the local context. The Ministry takes our program seriously and has invited us to join all statewide meetings and research initiatives regarding PMTCT. This will help eventually get all the data we need on clinics and gives us the opportunity to actively promote our agenda." New Incentives, update for GiveWell, August 2014, Pg 3.

• 9

  FHI 360 2013, available on FHI 360's website, is a plan for an $88 million PMTCT program in Akwa Ibom with the State Ministry of Health from 2013-2015. We are not sure if the program is actually being implemented as laid out in the budget. According to the budget:

  • About $300,000 will be spent on demand creation for PMTCT, none for CCTs. "PMTCT demand creation systems sub-total," Pg 43.
  • About $34 million will be spent on strengthening PMTCT service provision: "PMTCT service supply systems sub-total" Pg 41 and "Health Care Commodities sub-total" Pg 43. $14,149,295 + 20,211,748 = $34,361,043.
  • Much of the rest of the $88 million will be spent on activities less directly related to PMTCT treatment, such as prevention of HIV among women of reproductive age, prevention of unintended pregnancies among women living with HIV, HIV diagnosis for HIV-exposed infants, and life-long antiretroviral therapy for women living with HIV. Pgs 34-49.

• 10

  @New Incentives Budget 2014-15 (with GiveWell's analysis), October 2014@, "GiveWell Analysis" sheet.

• 11

  The figure in US dollars has changed over time because New Incentives has continued to give out the same amount in Naira, and foreign exchange rates have changed. As mentioned above, the total amount transferred to participants is about $150 as of February 25, 2015.

• 12

  New Incentives budget 2014-15 (with GiveWell's analysis), October 2014, "2014-2015 Total Budget" and "CCT Implementation Expenses" sheets.

• 13

  New Incentives budget 2014-15 (with GiveWell's analysis), October 2014, "CCT implementation expenses" sheet. New Incentives budgets that each field worker could manage up to 1000 beneficiaries.

• 14

  New Incentives draft concept note for RCT, October 2014. See also New Incentives draft RCT timeline, October 2014.

• 15

  New Incentives expects that the sample size may not be large enough to show a significant difference in HIV status.

• 16

  New Incentives budget 2014-15 (with GiveWell's analysis), October 2014, "RCT budget 2015-2016" and "GiveWell Analysis" sheets.

• 17

  Conversation with New Incentives on January 20, 2015

• 18

  Conversation with New Incentives on January 20, 2015

• 19

  New Incentives draft overview of CCTs for vaccinations, October 2014.

• 20

  For example, in one General Hospital, New Incentives expects it could enroll 240 women per month for CCTs for vaccinations vs. 40 for PMTCT. New Incentives budget 2014-15 (with GiveWell's analysis), October 2014, "CCT implementation expenses" and "CCTs for Vaccinations" sheets.

• 21

  New Incentives estimates $3,000-3,500 per life saved for vaccinations compared to about $1,900 for PMTCT. (Note that a large portion of this cost is money transferred to participants.) Conversation with New Incentives on October 28, 2014. We have not vetted either of these estimates.

• 22

  New Incentives draft overview of CCTs for high risk pregnancies, January 2015

• 23

  New Incentives beneficiary data June-November 2014, “Booking_Date” column.

• 24

  New Incentives beneficiary data June-November 2014.

• 25

  New Incentives beneficiary data June-November 2014, "House_Floor" and "House_Roof" columns. There are 227 records in total, and we have excluded N/A and blank data in these calculations.

• 26

  New Incentives beneficiary data June-November 2014, "Children_Delivered" and "Children_Alive" columns. We interpreted blanks as 0 and 99 as N/A (based on context).

• 27

  New Incentives beneficiary data June-November 2014, "Mother_Earnings" and "Husband_Earnings" columns. The 2250 Naira/month figure for women’s earnings is based on 227 records; it includes women reported to have no income (76 women) but excludes women for whom data do not appear valid (13 women with a 1 or 99 reported, which seem to flag invalid data). Excluding the women with no income, the median is 5000 Naira (about $25).

• 28

  Some husbands who are listed as having a job are reported as earning $0 per month, so the data husbands' earnings seems inconsistent and we do not know how to analyze it.

• 29

  New Incentives beneficiary data June-November 2014, "Intended_Use_T1Hotline," "Other_Intended_Use_T1Hotline," "Actual_Use_T2Hotline," "Intended_Use_T2Hotline," and "Other_Use_T2Hotline" columns.

• 30

  New Incentives beneficiary data June-November 2014, "Amount_Received_T1Collection" column. Based on the "Booking_Date" column, we excluded from the count women enrolled on or after November 14, who may have not yet gotten around to picking up their first transfer by November 27 (the end of the time frame that the data cover).

• 31

  New Incentives beneficiary data June-November 2014, "Reported_T2Collection" and "Estimated_Due_Date_ANC" columns. "Estimated_Due_Date_ANC," an estimate of due date obtained by adding months of pregnancy left (based on 9 - "Month_of_Pregnancy") to "Booking_Date," and an estimate of due date obtained by adding three weeks to "Phone_Reminder_Date_Delivery" are occasionally inconsistent. (There is a difference of at least two months between at least two of these dates in about 5% of records). Nevertheless, we simply use "Estimated_Due_Date_ANC" for this analysis.

• 32

  Some women misled New Incentives about where they gave birth because they did not think that New Incentives would be able to determine where they delivered. New Incentives believes that it is now better able to communicate to women that it will be able to check whether they follow the conditions, particularly now that New Incentives' field staff can explain the program to participants in their native language.

• 33

  Conversation with New Incentives on September 29, 2014

• 34

  New Incentives beneficiary data June-November 2014, "Photo_of_Belly" column.

• 35

  New Incentives beneficiary data June-November 2014, "Photo_Patient_Card_Inside" column.

• 36

  New Incentives beneficiary data June-November 2014, "Retest" column.

• 37

  New Incentives beneficiary data June-November 2014, "Dash_T1Collection" column.

• 38

  New Incentives fraud detection and prevention strategy, October 2014

• 39

  New Incentives beneficiary data June-November 2014, "Influence_Use_T2Hotline" and "Forced_Use_T2Hotline" columns. It seems that so few participants were asked because few participants had received their second transfer, and the question is asked during the survey given before the second transfer.

• 40

  Conversation with New Incentives on January 20, 2015

• 41

  See above.

• 42

  New Incentives budget 2014-15 (with GiveWell's analysis), October 2014, "CCT Implementation Expenses" sheet.

• 43

  Conversation with New Incentives on September 29, 2014

• 44

  New Incentives Gates Foundation Grand Challenges Press Release, November 2014

• 45

  @New Incentives Budget 2014-15 (with GiveWell's analysis), October 2014@, "2014-2015 Total Budget" sheet.

• 46

  As mentioned above, these field officers were formerly volunteers with FHI.

• 47

  Conversation with New Incentives on September 29, 2014

• 48

  Svetha Janumpalli, email to GiveWell, December 6, 2014

• 49

  GiveWell's non-verbatim summary of a conversation with New Incentives on Dec 13, 2012