Sanjay

Micro­cre­dit may some­times be effec­tive, but per­haps shouldn’t be funded by donations

• Sanjay 16 Feb 2020 18:32 UTC

  5 points

  in reply to: Aaron Gertler's comment on: Clean cook­stoves may be com­pet­i­tive with GiveWell-recom­mended charities

  Thanks for the en­courage­ment. I think that aiming for a “perfect” write-up has been a bar­rier to pub­lish­ing con­tent, so I in­tend for us to pub­lish more shal­low re­views to ad­dress this.

  To an­swer your ques­tion, I think the best fo­cus ar­eas would be the six bul­let points high­lighted near the start of the ar­ti­cle, with a par­tic­u­lar fo­cus on the first two (are the stoves ac­tu­ally used, and are they ac­tu­ally clean?) and the last (what is the best way to fund this work?).

  Also, we would fur­ther in­ves­ti­gate the very use­ful com­ments made by MatthewDahlhausen (which seemed very use­ful and was up­voted by me) and look fur­ther at the GiveWell anal­y­sis as men­tioned by cole_haus (I was aware of this, but had not had the ca­pac­ity to re­view it prop­erly)

• Sanjay 14 Feb 2020 10:55 UTC

  8 points

  on: How do you feel about the main EA face­book group?

  Thank you to JanBrauner for rais­ing this. I have had similar con­cerns.

  How­ever I don’t think stronger mod­er­a­tion is the an­swer, es­pe­cially since it’s hard to mod­er­ate the com­ments, which is where most of the poor qual­ity con­tent arises.

  In­stead, it per­haps needs more con­certed efforts from those who have thought about EA ideas more care­fully to (nicely and con­struc­tively) help to im­prove the qual­ity of the dis­cus­sion.

  This means par­ti­ci­pat­ing in the EA face­book group more ac­tively, rather than spend­ing less time on it and just pay­ing at­ten­tion to the fo­rum (which can­didly, is what I’ve found my­self do­ing)

• Sanjay 13 Feb 2020 15:29 UTC

  3 points

  on: Cli­mate dis­count­ing: How do you value one tonne of CO2eq averted to­day ver­sus (say) 30 years from now?

  Note: this ques­tion was also cross-posted to the Face­book effec­tive en­vi­ron­men­tal­ism group https://​​www.face­book.com/​​groups/​​effec­tiveen­vi­ron­men­tal­ism/​​perma­l­ink/​​2306369839662729/​​

[Question] Cli­mate dis­count­ing: How do you value one tonne of CO2eq averted to­day ver­sus (say) 30 years from now?

Clean cook­stoves may be com­pet­i­tive with GiveWell-recom­mended charities

• Sanjay 4 Feb 2020 21:32 UTC

  2 points

  in reply to: Nathan Young's comment on: Nathan Young’s Shortform

  HaukeHille­brandt has recom­mended sup­port­ing Prof Chris Cham­bers to do this: https://​​lets-fund.org/​​bet­ter-sci­ence/​​

• Sanjay 4 Jan 2020 18:48 UTC

  1 point

  in reply to: kbog's comment on: Con­se­quences of an­i­mal product con­sump­tion (com­bined model)

  I too imag­ined that down­load­ing it and view­ing it MS Ex­cel might be bet­ter, but as far as I can tell there doesn’t seem to be a way of down­load­ing it. Un­less I’ve just not found the way to down­load?

• Sanjay 30 Dec 2019 16:21 UTC

  1 point

  on: Con­se­quences of an­i­mal product con­sump­tion (com­bined model)

  Thank you for shar­ing this. Un­for­tu­nately the model shared ap­pears to be a val­ues-only ver­sion of the model (i.e. when I go to any cell in the spread­sheet, the for­mula bar just shows a num­ber, not a for­mula). This makes it very hard to work out how you came to your con­clu­sions.

  I don’t know whether it’s show­ing this way be­cause of the way One Drive op­er­ates or be­cause you de­liber­ately re­moved the for­mu­lae be­fore up­load­ing the spread­sheet, how­ever I would cer­tainly find it use­ful if you could find a way to share the full model. Thank you.

• Sanjay 30 Dec 2019 16:02 UTC

  2 points

  on: Is be­ing ve­gan the most moral op­tion when it comes to eat­ing an­i­mal prod­ucts?

  Un­der cer­tain as­sump­tions, it is pos­si­ble for eat­ing an­i­mal prod­ucts to be morally bet­ter than eat­ing ve­gan food. In par­tic­u­lar, you would have to be­lieve that pos­i­tive im­pacts on the an­i­mal’s life out­weigh the nega­tive as­pects, also out­perform the net effect of wildlife that would have been on that land no longer be­ing there, and the sus­tain­abil­ity im­pact. I’ve ex­plored this in this blog post, and I sug­gest in that post that it is pos­si­ble for eat­ing meat to be the bet­ter op­tion, but it prob­a­bly isn’t: https://​​in­ter­est­ingth­ing­siveread.blogspot.com/​​2018/​​12/​​ve­g­anism-may-be-net-nega­tive-but-we.html

• Sanjay 30 Dec 2019 14:00 UTC

  5 points

  on: Helping the poor­est with­out man­ag­ing birth con­trol cre­ates vi­cious circle

  As KHor­ton has al­luded to, there is a well-es­tab­lished body of liter­a­ture which in­di­cates that de­vel­op­ment leads lower To­tal Fer­til­ity Rates. Ex­am­ples in­clude the books Com­mon Wealth by Jeffrey Sachs and Fact­ful­ness by Hans Rosling et al. The fol­low­ing on­line re­sources also ex­plore this:

  • This ar­ti­cle in­di­cates that meet­ing the UN’s SDGs would lead to lower pop­u­la­tion growth https://​​www.pnas.org/​​con­tent/​​early/​​2016/​​11/​​28/​​1611386113

  • David Rood­man has ex­plored this point at GiveWell’s en­courage­ment: https://​​davi­drood­man.com/​​blog/​​2014/​​04/​​16/​​the-mor­tal­ity-fer­til­ity-link/​​

  • Giv­ing What We Can has also writ­ten about it: https://​​www.giv­ing­whatwe­can.org/​​post/​​2015/​​09/​​de­vel­op­ment-pop­u­la­tion-growth-and-mor­tal­ity-fer­til­ity-link/​​

Up­date on CATF’s plans for 2020

• Sanjay 14 Dec 2019 14:48 UTC

  10 points

  on: Im­por­tant EA-re­lated ques­tions EA would like to know from gen­eral public

  Great that you’re do­ing this, thanks so much for rais­ing this here!

  Not sure if you’re already aware of this, but you might want to be aware of other stud­ies that have looked at similar ques­tions. In par­tic­u­lar the Money For Good study in the US and the equiv­a­lent in the UK were in­ter­est­ing, albeit some­what dated. (I have the raw data for the UK study). You might want to have a look so that you can use con­sis­tent ques­tion word­ing whether the ques­tions over­lap.

  Some ques­tions that we at SoGive would be in­ter­ested to know more about:

  • The last time you gave to char­ity, what trig­gered you to give? (a) some­one asked me (b) I de­cided my­self [could break (a) down fur­ther, per­haps] [Note, this has been stud­ied be­fore, but to my knowl­edge not re­cently. Also I’ve never seen an anal­y­sis of the cor­re­lates of self-driven giv­ing]

  • For the donor pop­u­la­tion, I’d like to un­der­stand the split be­tween what we at SoGive call Or­gani­sa­tion-loyal, Cause-spe­cific, and Open-minded donors (hope­fully the la­bels are self-ex­plana­tory but if not please ask). We would be in­ter­ested to know speci­fi­cally to what ex­tent are those prefer­ences mod­er­ate prefer­ences (defined as: if the per­son is choos­ing a char­ity to donate to, they will fol­low that prefer­ence) and to what ex­tent are they strong prefer­ences (i.e. if some­one else asks them to donate, they will say no un­less the dona­tion tar­get is con­sis­tent with their prefer­ence). We at SoGive cur­rently have some thoughts on this, which I can ex­pand on if you’re in­ter­ested.

  • In terms of cause ar­eas, it would be use­ful to know whether moral-cir­cle think­ing is a good model for ex­plain­ing cause prefer­ences. E.g. if some­one is happy to sup­port peo­ple in the de­vel­op­ing world, are they more likely to sup­port an­i­mals? Or those in the far fu­ture? (see also the re­search we con­ducted on at­ti­tudes to the far fu­ture)

  • A deeper un­der­stand­ing of why peo­ple never donate would be in­ter­est­ing. Self-re­ported an­swers tell us some­thing, but there is some ev­i­dence that donors (at least some of the time) are non-donors be­cause they just don’t want to give and are look­ing for ex­cuses (“mo­ti­vated rea­son­ing”). I wrote about this here, and refer­enced some stud­ies, e.g. this, this and this (which were men­tioned to me by a con­tact at Re­think Char­ity). Some­thing which tried to quan­tify this (i.e. how many non-donors will never give, and how many non-donors would give if their needs were bet­ter met) would be re­ally in­ter­est­ing, but pos­si­bly too hard for the scope of your study.

  Some re­searchers are do­ing some in­ter­est­ing work on how peo­ple give—the names Beth Breeze and Cat Walker spring to mind, al­though there are oth­ers as well. But you may well be too time-con­strained to wade through all their work, in which case I sug­gest you just take a look at the Money for Good stud­ies men­tioned ear­lier.

  More gen­er­ally, very happy to dis­cuss fur­ther. If you are will­ing to have a chat, let me know: san­jay [at] so­give.org

• Sanjay 28 Nov 2019 14:46 UTC

  1 point

  in reply to: cole_haus's comment on: A cor­rected model sug­gests cli­mate change in­ter­ven­tions may be within a fac­tor of two of di­rect cash transfers

  And is it fair to say that if it ap­plies to the GD im­pact, then ap­ply­ing it differ­en­tially across coun­tries no longer makes sense? (i.e. Hauke’s origi­nal ap­proach does make sense af­ter all?)

• Sanjay 27 Nov 2019 10:50 UTC

  2 points

  in reply to: cole_haus's comment on: A cor­rected model sug­gests cli­mate change in­ter­ven­tions may be within a fac­tor of two of di­rect cash transfers

  Thank you for your re­ply, and sorry that my re­ply is rushed.

  Re: ” Also, I’m not sure I fol­low your claim that no in­come ad­just­ment is hap­pen­ing in the GDvCC model. Even your new tab has a row la­bel­led “in­come ad­just­ment” with the num­ber 1260 which is con­se­quen­tial in the fi­nal re­sult ” my claim isn’t meant to be that there is no in­come ad­just­ment, but rather that there is no in­come ad­just­ment to the SCC figures—there is in­come ad­just­ment to the cash trans­fer figures. (Sorry, can’t re­mem­ber how I phrased this be­fore, so might not have been clear)

• Sanjay 26 Nov 2019 22:36 UTC

  2 points

  on: A cor­rected model sug­gests cli­mate change in­ter­ven­tions may be within a fac­tor of two of di­rect cash transfers

  Thank you very much to cole_haus for do­ing this. I definitely value it when peo­ple take the time to challenge anal­y­sis done by oth­ers.

  As I un­der­stand it, the origi­nal anal­y­sis done by HaukeHille­brandt was cor­rect.

  Here’s my un­der­stand­ing:

  • In his ear­lier post, HaukeHille­brandt set out a calcu­la­tion for the ra­tio of the im­pact­ful­ness of cli­mate change in­ter­ven­tions ver­sus GiveDirectly, which can be found in this spread­sheet.

  • This spread­sheet in­cluded a (po­ten­tially some­what con­fus­ingly la­bel­led) row called “So­cial cost of car­bon per tonne (rel­a­tive to cash trans­fers)”

  • The anal­y­sis done here by cole_haus is, as I un­der­stand it, based on the as­sump­tion that we need to take the so­cial cost of car­bon and ap­ply an in­come ad­just­ment to it

  • How­ever, I don’t be­lieve that this is nec­es­sary—I be­lieve that the ex­ist­ing so­cial cost of car­bon figures are already on a con­sis­tent ba­sis and don’t need to be ad­justed further

  • What’s more, the origi­nal spread­sheet doesn’t seem to have ac­tu­ally been do­ing this (al­though it looked it did). To see this, take a look at this copy of the spread­sheet that I cre­ated. I cre­ated a new tab called ” SJ CC v GD ” which recre­ates the origi­nal calcs, but shows that there is no in­come ad­just­ment to the so­cial cost of car­bon.

  Happy to have my un­der­stand­ing cor­rected if I’ve mi­s­un­der­stood.

• Sanjay 26 Nov 2019 9:29 UTC

  1 point

  in reply to: zdgroff's comment on: The Eco­nomic Lives of the Poor

  +1 to this, I definitely found Port­fo­lios of the Poor a great way of ac­tu­ally un­der­stand­ing the lives of the globally poor

• Sanjay 18 Nov 2019 11:23 UTC

  5 points

  on: What ar­eas of maths are use­ful across dis­ci­plines?

  I would say prob­a­bil­ity and statis­tics.

  If you want to do ev­i­dence-based things, you will want to be able to read an aca­demic pa­per, in­clud­ing the maths. So you will want to be able to not just un­der­stand what is meant by a p-value, but also be able to have thoughts like:

  • this pa­per used a nor­mal dis­tri­bu­tion, but re­ally a stu­dents t /​ logit /​ what­ever dis­tri­bu­tion would have been bet­ter, I won­der how big a differ­ence that makes?

  • they used a nor­mal ap­prox­i­ma­tion for a bino­mial here—and I do/​don’t think that seems like a rea­son­able approximation

  • The pa­per claims this looks like a fairly good fit—they could have used a chi squared test here; I won­der why they didn’t?

  This sort of abil­ity wouldn’t be use­ful if the ex­ist­ing body of re­search con­sis­tently used statis­tics well, but I don’t think that’s the case.

  Fi­nal caveat: an­swer­ing this ques­tion is hard be­cause it’s so broad. I’m ex­trap­o­lat­ing from my own ex­pe­rience, but what’s use­ful for you might be differ­ent.

• Sanjay 18 Nov 2019 11:10 UTC

  4 points

  in reply to: John_Maxwell's comment on: Ap­ply­ing EA to cli­mate change

  “Which is bet­ter from an in­di­vi­d­ual per­spec­tive: stop driv­ing and take the bus to work, or cut food waste from 35% to 0%?”

  The draw­down pro­ject seems to sug­gest that cut­ting food waste is bet­ter, be­cause it’s rated third on its list whereas mass tran­sit is 37th. How­ever I hes­i­tate to sug­gest peo­ple fol­low the guidance of Draw­down. I con­tacted them a few years ago (be­fore the big me­dia splash) with some ques­tions about their method­ol­ogy and got no re­ply. So I don’t feel will­ing to en­dorse (or con­demn) their work.

• Sanjay 18 Nov 2019 11:02 UTC

  3 points

  in reply to: John_Maxwell's comment on: Ap­ply­ing EA to cli­mate change

  “I won­der where most of the waste hap­pens?”

  Most of the busi­ness-driven el­e­ments of the food sup­ply chain are quite effi­cient, I’m told. I.e. each of man­u­fac­ture, trans­port, and re­tail man­age­ment. The waste comes al­most en­tirely from cus­tomers buy­ing things they don’t need and then throw­ing them away.

  If so, the most likely down­side, if any, is the risk of peo­ple con­sum­ing food af­ter its use-by date.

  My source for this claim is a pitch from a food-waste char­ity. I con­sider this to be a slightly bet­ter source than a per­son cho­sen at ran­dom, how­ever I didn’t get the im­pres­sion that the char­ity was rigor­ous about fact-check­ing its claims, so I can’t promise this is cor­rect.