A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | CEA of steady-state grant, EvAc TA costs only; Technical assistance (TA) marginal impact, excluding any additional costs incurred by govt | |||||||||||||||||||||||||
2 | ||||||||||||||||||||||||||
3 | Increase in coverage per year without TA, in percentage points | 3% | Assumption | |||||||||||||||||||||||
4 | Increase in coverage per year with TA, in percentage points | 6% | Assumption | |||||||||||||||||||||||
5 | Implied % of coverage increase each year attributable to TA | 50% | Calculation | |||||||||||||||||||||||
6 | ||||||||||||||||||||||||||
7 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | ||||||||||||||||||||
8 | Coverage with TA | 34% | 40% | 46% | 52% | 58% | 64% | |||||||||||||||||||
9 | Coverage without TA | 31% | 34% | 37% | 40% | 43% | 46% | |||||||||||||||||||
10 | ||||||||||||||||||||||||||
11 | Coverage increase due to TA | 3% | 6% | 9% | 12% | 15% | 18% | |||||||||||||||||||
12 | ||||||||||||||||||||||||||
13 | Cost for TA | $2,401,730 | $2,401,730 | $2,401,730 | $2,401,730 | $2,401,730 | $2,401,730 | |||||||||||||||||||
14 | Population targeted | 28,739,540 | 28,739,540 | 28,739,540 | 28,739,540 | 28,739,540 | 28,739,540 | |||||||||||||||||||
15 | ||||||||||||||||||||||||||
16 | Increase in person-years of coverage due to TA | 862,186 | 1,724,372 | 2,586,559 | 3,448,745 | 4,310,931 | 5,173,117 | |||||||||||||||||||
17 | ||||||||||||||||||||||||||
18 | Cost per person-year of coverage due to TA | $0.80 | ||||||||||||||||||||||||
19 | ||||||||||||||||||||||||||
20 | Implied: Beta's TA is X times more cost-effective than GiveDirectly | 14.0 | ||||||||||||||||||||||||
21 | ||||||||||||||||||||||||||
22 | Cost of TA per person targeted per year | $0.08 | ||||||||||||||||||||||||
23 | ||||||||||||||||||||||||||
24 | Rough estimate to account for leverage/funging | |||||||||||||||||||||||||
25 | Equivalent lives saved by coverage increase | 18,366 | ||||||||||||||||||||||||
26 | ||||||||||||||||||||||||||
27 | Counterfactual use of money, cost per life saved equivalent for government financial costs | $15,000 | ||||||||||||||||||||||||
28 | Additional money spent by government to achieve this coverage increase that's attributable to TA | $5,345,442 | ||||||||||||||||||||||||
29 | Counterfactual lives saved with govt spending | 356 | ||||||||||||||||||||||||
30 | Offsetting lives as % of lives saved by program | 1.9% | ||||||||||||||||||||||||
31 | ||||||||||||||||||||||||||
32 | Factors left out: | |||||||||||||||||||||||||
33 | - Number of years included is very important and I've made an arbitrary assumption. For slightly more information, see https://www.givewell.org/node/3109#Cost-effectiveness | |||||||||||||||||||||||||
34 | - Govt staff time is largely excluded (except for what is implicitly factored into rows 27-30 above). For more information, see https://www.givewell.org/node/3106#Is_the_program_cost-effective | |||||||||||||||||||||||||
35 | - If TA stops, coverage increases may continue without additional investment | |||||||||||||||||||||||||
36 | - Did not discount costs and benefits in future years. Expected impacts to be small and offsetting. | |||||||||||||||||||||||||
37 | - Did not include any benefits from national-level TA | |||||||||||||||||||||||||
38 | - I don't account for declining counterfactual anemia rates over time via interventions aside from IFA properly. I've roughly factored it into the coverage rate above just by adding a percentage point for it, but this isn't the perfect mathematical way to incorporate it into the model. For slightly more information, see https://www.givewell.org/node/3109#Cost-effectiveness | |||||||||||||||||||||||||
39 | - Did not factor in any impacts on birthweight, birth defects | |||||||||||||||||||||||||
40 | ||||||||||||||||||||||||||
41 | ||||||||||||||||||||||||||
42 | CEA of Phase 2 grant, direct impacts only | |||||||||||||||||||||||||
43 | Source | |||||||||||||||||||||||||
44 | Cost of TA for this grant, Rajasthan | $1,510,243 | See https://www.givewell.org/node/3109#Budget | |||||||||||||||||||||||
45 | Cost of TA for this grant, MP | $1,553,193 | See https://www.givewell.org/node/3109#Budget | |||||||||||||||||||||||
46 | Cost of TA for this grant, Total | $4,008,259 | See https://www.givewell.org/node/3109#Budget | |||||||||||||||||||||||
47 | ||||||||||||||||||||||||||
48 | Target population, Rajasthan | 1,570,628 | From an unpublished email from Evidence Action ("Questions about Beta IFA", sent Sept 12, 2018): Rajasthan Statewide target population: 13,350,342; Average target population at district level: 392,657; Average number of districts in a region: 4; Average regional-level target population: 1,570,628 | |||||||||||||||||||||||
49 | Target population, MP | 1,508,740 | From an unpublished email from Evidence Action ("Questions about Beta IFA", sent Sept 12, 2018): MP Statewide target population: 15,389,198; Average target population at district level: 301,748; Average number of districts in a region: 5; Average regional-level target population: 1,508,740 | |||||||||||||||||||||||
50 | Number of kids targeted in relevant region | 3,079,368 | Calculation | |||||||||||||||||||||||
51 | ||||||||||||||||||||||||||
52 | Baseline coverage, Rajasthan | 17.5% | This is a rough best guess. The unpublished "Government of India, presentation on coverage of IFA program" reports 16-19% coverage in Rajasthan. Note: we are unsure how baseline coverage in chosen regions will differ from state-level coverage estimates. | |||||||||||||||||||||||
53 | Baseline coverage, MP | 45% | This is a rough best guess. The unpublished "Government of India, presentation on coverage of IFA program" reports 70% coverage in Madhya Pradesh, but according to Evidence Action's survey in MP (see survey discussed in "Evidence Action, WIFS National Concept Note" Pg. 5), 44% of schools are unaware of the program and only 1/3rd of schools reported implementing it regularly. Note: we are unsure how baseline coverage in chosen regions will differ from state-level coverage estimates. | |||||||||||||||||||||||
54 | Baseline coverage, overall | 31.0% | Calculation | |||||||||||||||||||||||
55 | ||||||||||||||||||||||||||
56 | Increase in coverage per year without TA, in percentage points | 3% | Assumption | |||||||||||||||||||||||
57 | Increase in coverage per year with TA, in percentage points | 7% | Assumption | |||||||||||||||||||||||
58 | Implied % of coverage increase each year attributable to TA | 57% | Calculation | |||||||||||||||||||||||
59 | ||||||||||||||||||||||||||
60 | 2019 | 2020 | 2021 | 2022 | ||||||||||||||||||||||
61 | Coverage with TA | 38% | 45% | 52% | 59% | |||||||||||||||||||||
62 | Coverage without TA | 34% | 37% | 40% | 43% | |||||||||||||||||||||
63 | ||||||||||||||||||||||||||
64 | Coverage increase due to TA | 4% | 8% | 12% | 16% | |||||||||||||||||||||
65 | ||||||||||||||||||||||||||
66 | Cost for TA | $2,401,730 | $2,401,730 | $0 | $0 | |||||||||||||||||||||
67 | Population targeted | 3,079,368 | 3,079,368 | 3,079,368 | 3,079,368 | |||||||||||||||||||||
68 | ||||||||||||||||||||||||||
69 | Increase in person-years of coverage due to TA | 123,175 | 246,349 | 369,524 | 492,699 | |||||||||||||||||||||
70 | ||||||||||||||||||||||||||
71 | Cost per person-year of coverage due to TA | $3.90 | Calculation | |||||||||||||||||||||||
72 | ||||||||||||||||||||||||||
73 | Implied: Beta's TA is X times more cost-effective than GD | 2.9 | See formula | |||||||||||||||||||||||
74 | ||||||||||||||||||||||||||
75 | Cost of TA per person targeted per year | $0.65 | Calculation | |||||||||||||||||||||||
76 | ||||||||||||||||||||||||||
77 | ||||||||||||||||||||||||||
78 | CEA of Phase 2 grant, indirect effects on learning | |||||||||||||||||||||||||
79 | ||||||||||||||||||||||||||
80 | Cost of the grant | $4,328,259 | See formula. Added $320K for the planning grant. We'll likely need to spend more than just this to figure this out. | |||||||||||||||||||||||
81 | Probability of this project ending up as 10x+ cash (at a level of analysis similar to current top charities) | 35% | (Note: this is a conservative assumption. Implies that it's either 10x or not valuable at all.) | |||||||||||||||||||||||
82 | ||||||||||||||||||||||||||
83 | Funding that this project will be able to absorb at 10x+ cash (i.e., amount of funding that would be reallocated to this project if this grant succeeds) | $25,000,000 | Very rough assumption of ~$1m per state per year, and that we'll fund 5 states for 5 years each at that level | |||||||||||||||||||||||
84 | ||||||||||||||||||||||||||
85 | Cost per under-5 life equivalent saved, GiveDirectly | $11,663.00 | Median from https://docs.google.com/spreadsheets/d/13b_qt-G_TQtoYNznNak3_5dzvzgCSUPJnk3l5dMisJo/edit#gid=1034883018 | |||||||||||||||||||||||
86 | Cost per life saved, 10x cash | $1,166.30 | Calculation | |||||||||||||||||||||||
87 | How much more cost-effective than cash will marginal GW recommendations be in the absence of this grant? | 5 | Guess based on our current estimate of cost-effectiveness of marginal dollars to GW top charities. | |||||||||||||||||||||||
88 | Cost per life saved, counterfactual use of marginal GW dollars | $2,332.60 | Calculation | |||||||||||||||||||||||
89 | ||||||||||||||||||||||||||
90 | Lives saved if funding amount inputted above is spent on this project | 21,435 | Assumes this project will end up 10x cash | |||||||||||||||||||||||
91 | Lives saved if funding amount inputted above is spent on counterfactual | 10,718 | Calculation | |||||||||||||||||||||||
92 | ||||||||||||||||||||||||||
93 | Net lives saved | 10,718 | Calculation | |||||||||||||||||||||||
94 | Net lives saved, adjusted for probability of the project succeeding | 3751 | Calculation | |||||||||||||||||||||||
95 | ||||||||||||||||||||||||||
96 | Cost per life saved from this grant | $1,154 | Calculation | |||||||||||||||||||||||
97 | This grant is X times more cost-effective than cash | 10.1 | Calculation | |||||||||||||||||||||||
98 | This grant is X times more cost-effective than marginal spending on GW top charities now | 2.0 | Calculation | |||||||||||||||||||||||
99 | ||||||||||||||||||||||||||
100 |