ABCDEFGHIJKLMNOPQRSTUVWXYZAAABACADAEAFAGAHAIAJAKALAMANAOAPAQARASATAUAVAWAXAYAZBABBBCBDBEBFBGBHBIBJ
1
If you opened this file from a web link, please send an email to causeprioquestions@openphilanthropy.org. We like to track who outside Open Philanthropy is engaging with our research, in part so we can calibrate whether it is worth investing in making spreadsheets like this cleaner and easier to understand.
2
Cells in blue are user inputs
3
You can specify the philanthropic opportunity set (in a piecewise way), and the moral weights assigned to various outcomes under each framework.
4
Here's how these inputs are processed:
5
First, the maroon sections (at far right) use a 2-step approximation to estimate how much money we'd spend on each category, if we obeyed a given moral framework
One opportunity set for LIC health interventions, another for LIC income, etc. Opportunity sets are isoelastic, with an anchor point defined in terms of marginal CE
6
Then the purple sections translate the estimated spending in each category into estimated outcomes (DALYs and income-doublings), if we spent according to a given moral framework.
E.g. the below table shows $1.0B in annual RFMF focused on LIC DALYs at a CE of at least 800x under old OP weights.
7
Finally, the green sections show the total good achieved, according to each moral framework, if we spent $2.5B following a given moral framework.
Opportunity sets take the form: output = x^(1-n)/(1-n) * A*B^n. This yields marginal CE output'=A at anchor point B.
8
9
Inputs: Opportunity Sets
Inputs: Moral Weights ($k)
10
SymbolLIC DALYLIC MIDDW MIDDW DALYFrameworkLIC DALYLIC MIDDW MIDDW DALY
11
B
Anchor Point ($M annual RFMF)
$1,000$1,000$300$200 ... Eta<1$50$20$50$200
12
A*50
CE at Anchor Point (Old Weights)
800x800x800x300x ... Big Diff$100$50$50$325
13
50
Value of outcome ($k, old weights)
$50$50$50$50 ... Compromise$125$50$50$200
14
nEta0.3750.3750.3750.5 ... No Diff$100$50$50$100
15
1/A
Marginal cost at anchor point ($ spent per actual outcome)
$63$63$63$167 ... Low No Diff$80$50$50$80
16
A*B^nscaling factor "K"0.2130.2130.1360.085 ... Old Weights$50$50$50$50
17
[A/(1-n)]*B
Total outcomes achieved at anchor point
2.56E+072.56E+077.68E+062.40E+06
... Weighted Avg
$95$42$50$223
18
2.13E-01
19
Eta<1Big DiffCompromiseNo DiffLow No DiffOld Weights
This is the old way I did it:
20
0.97xProbability weights for avg:25%30%30%15%0%0%
Annual Budget ($M)
$2,5003.13E+07
21
0.95x
22
Note: this section shows CE, in the given moral frameworks's terms, at the anchor pointNote: this section shows what expenditures would need to be in order to equalize marginal CE with LIC DALYsNote: this column shows the cost of one "basket" of interventions with equal marginal CE, per the expenditure numbers at left
23
Good achieved ($B), if proper moral weighting is...
Amount ($M) spent on...
Outcomes (MID, DALY) Achieved
CE at Anchor Point
Expenditure ($M) to equalize MCE with LIC DALYs
Cost of Basket, round 1
Amount spent on LIC DALYs, round 1
CE of LIC DALY, round 1
Expenditure ($M) to equalize MCE with LIC DALYs - round 2
Cost of Basket, round 2
24
Eta<1Big DiffCompromiseNo DiffLow No DiffOld WeightsWeighted Avg
[Calc-check avg]
LIC DALYLIC MIDDW MIDDW DALYLIC DALYLIC MIDDW MIDDW DALYLIC DALYLIC MIDDW MIDDW DALYLIC DALYLIC MIDDW MIDDW DALY
(cost of one "basket" of marginally-balanced interventions)
(cost of one "basket" of marginally-balanced interventions)
25
Spend 2.5B as if...
26
 ... Eta<1$2,984.7$5,273.4$5,544.8$4,353.0$3,645.1$2,583.1$4,644.6$4,644.6$1,380$120$419$5813.13E+076.80E+069.47E+064.09E+06796x318x800x1200x$1,000$87$304$454$1,845$1,355710x$1,355$118$412$571$2,455
27
 ... Big Diff$2,887.8$5,469.6$5,914.9$4,663.2$3,876.9$2,697.4$4,836.8$4,836.8$1,705$269$81$4463.57E+071.13E+073.38E+063.58E+061600x800x800x1950x$1,000$157$47$297$1,502$1,6651322x$1,665$262$79$435$2,441
28
 ... Compromise$2,739.2$5,311.2$6,099.1$4,880.3$4,020.3$2,730.1$4,839.9$4,839.9$2,132$185$56$1274.11E+078.92E+062.68E+061.92E+062000x800x800x1200x$1,000$87$26$72$1,185$2,1101512x$2,110$183$55$126$2,474
29
 ... No Diff$2,660.2$5,190.7$6,041.7$4,926.2$4,104.4$2,871.7$4,773.7$4,773.7$2,035$321$96$483.99E+071.26E+073.77E+061.18E+061600x800x800x600x$1,000$157$47$28$1,233$2,0281227x$2,028$319$96$48$2,491
30
 ... Low No Diff$2,645.8$5,144.0$5,925.3$4,891.9$4,132.4$2,993.2$4,716.0$4,716.0$1,791$512$153$443.69E+071.68E+075.05E+061.12E+061280x800x800x480x$1,000$286$86$28$1,399$1,7871030x$1,787$510$153$43$2,493
31
 ... Old Weights$2,451.4$4,717.3$5,271.0$4,509.3$3,955.5$3,124.7$4,285.7$4,285.7$1,074$1,074$322$302.68E+072.68E+078.03E+069.24E+05800x800x800x300x$1,000$1,000$300$28$2,328$1,074779x$1,074$1,074$322$30$2,499
32
 ... Weighted Avg$2,852.5$5,425.2$6,049.5$4,818.0$3,994.7$2,759.7$4,878.2$4,878.2$1,918$225$104$2533.85E+071.01E+073.96E+062.70E+061519x680x800x1335x$1,000$117$54$154$1,326$1,8861198x$1,886$221$102$249$2,457
33
34
Expected value of good achieved, if we had perfect knowledge
$4,955.7
Calc Check -- outcomes achieved
Calc Check -- CE at adjusted point
Calc Check -- CE at adjusted point
35
LIC DALYLIC MIDDW MIDDW DALYLIC DALYLIC MIDDW MIDDW DALYLIC DALYLIC MIDDW MIDDW DALY
36
Distance from optimal, in terms of CE delta, according to:
37
Eta<1Big DiffCompromiseNo DiffLow No DiffOld WeightsWeighted Avg
Expected Distance From Optimum
3.13E+076.80E+069.47E+064.09E+06796.21x796.21x796.21x796.21xWAHR710.47x710.47x710.47x710.47x
38
Spend 2.5B as if...3.57E+071.13E+073.38E+063.58E+061600.00x1600.00x1600.00x1600.00xWAHR1321.67x1321.67x1321.67x1321.67x
39
 ... Eta<11.00x0.96x0.91x0.88x0.88x0.83x0.95x0.94x4.11E+078.92E+062.68E+061.92E+062000.00x2000.00x2000.00x2000.00xWAHR1511.60x1511.60x1511.60x1511.60x
40
 ... Big Diff0.97x1.00x0.97x0.95x0.94x0.86x0.99x0.97x3.99E+071.26E+073.77E+061.18E+061600.00x1600.00x1600.00x1600.00xWAHR1227.40x1227.40x1227.40x1227.40x
41
 ... Compromise0.92x0.97x1.00x0.99x0.97x0.87x0.99x0.97x3.69E+071.68E+075.05E+061.12E+061280.00x1280.00x1280.00x1280.00xWAHR1029.68x1029.68x1029.68x1029.68x
42
 ... No Diff0.89x0.95x0.99x1.00x0.99x0.92x0.98x0.95x2.68E+072.68E+078.03E+069.24E+05800.00x800.00x800.00x800.00xWAHR778.91x778.91x778.91x778.91x
43
 ... Low No Diff0.89x0.94x0.97x0.99x1.00x0.96x0.97x0.94x3.85E+071.01E+073.96E+062.70E+061519.05x1519.05x1519.05x1519.05xWAHR1197.51x1197.51x1197.51x1197.51x
44
 ... Old Weights0.82x0.86x0.86x0.92x0.96x1.00x0.88x0.86x
45
 ... Weighted Avg0.96x0.99x0.99x0.98x0.97x0.88x1.00x0.98xWAHRWAHRWAHRWAHRWAHRWAHRWAHRWAHR
46
47
1.00
48
49
If best moral weighting is...
Marginal CE of spending on...
Total Gain from spending on...
50
 ... CompromiseLIC DALYLIC MIDDW MIDDW DALYLIC DALYLIC MIDDW MIDDW DALY
51
Spend 2.5B as if...
52
 ... Eta<11773x1773x706x704x$3,913,384$339,922$473,333$818,127
53
 ... Big Diff1637x1310x1310x804x$4,466,616$562,758$168,827$716,726
54
 ... Compromise1506x1506x1506x1504x$5,136,057$446,125$133,837$383,041
55
 ... No Diff1532x1226x1226x2450x$4,989,372$628,621$188,586$235,077
56
 ... Low No Diff1607x1029x1029x2571x$4,606,791$841,896$252,569$224,062
57
 ... Old Weights1947x779x779x3115x$3,346,119$1,338,448$401,534$184,892
58
 ... Weighted Avg1566x1401x1190x1067x$4,808,242$503,349$198,166$539,743
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100