2020 Top Charity Ideas—Charity Entrepreneurship

This ar­ti­cle was also pub­lished at Char­ity En­trepreneur­ship’s blog.

We’re proud to an­nounce our 2020 Top Char­ity Ideas!

​Each year Char­ity En­trepreneur­ship iden­ti­fies highly effec­tive in­ter­ven­tions in cho­sen cause ar­eas. Our In­cu­ba­tion Pro­gram gives par­ti­ci­pants the skills they need to start high-im­pact non­prof­its based on our top in­ter­ven­tion recom­men­da­tions.

Our 2020 re­search pe­riod fo­cused on four cause ar­eas: men­tal health, an­i­mal ad­vo­cacy, fam­ily plan­ning, and health & de­vel­op­ment policy. We be­gan with sev­eral hun­dred ideas in each cause area. Pro­gres­sive stages of our ex­ten­sive re­search pro­cess whit­tled down to eight recom­mended ideas.

Eighty-hour re­ports linked be­low illus­trate how we came to recom­mend this year’s top in­ter­ven­tions. We also provide In­cu­ba­tion Pro­gram par­ti­ci­pants with im­ple­men­ta­tion re­ports, which provide spe­cific recom­men­da­tions to map a path for­ward for a new char­ity.

Our 2020 top recom­men­da­tions are as fol­lows (in no par­tic­u­lar or­der):


1. Guided self-help – Distribut­ing work­books to en­able in­di­vi­d­u­als to work in­de­pen­dently on their men­tal health, sup­ported by short weekly calls from lay health work­ers.


2. Lead paint reg­u­la­tion – Ad­vo­cat­ing for tighter reg­u­la­tion of lead paint to re­duce the bur­den of lead ex­po­sure on hu­man health and eco­nomic pros­per­ity.
3. Al­co­hol reg­u­la­tion – Ad­vo­cat­ing for in­creased al­co­hol tax­a­tion to miti­gate the harm­ful effects of con­sump­tion.


4. Shrimp welfare – Im­prov­ing the welfare of farmed shrimp, e.g. through col­lab­o­rat­ing with Viet­namese farm­ers to bet­ter oxy­genate the wa­ter, thus re­duc­ing chronic suffer­ing for shrimp.
5. Feed for­tifi­ca­tion – For­tify­ing feed with micronu­tri­ents to com­bat defi­cien­cies and im­prove the health of lay­ing hens.
6. Ask re­search – Helping or­ga­ni­za­tions and policy-mak­ers de­cide what best to ask of the an­i­mal agri­cul­ture in­dus­try. (We ex­plored this in­ter­ven­tion dur­ing our 2019 re­search pe­riod and passed it on to 2020, as de­spite its promise it was not started.)


7. Mass me­dia cam­paigns – Broad­cast­ing in­for­ma­tion about fam­ily plan­ning to re­duce mis­con­cep­tions and em­power women to make de­ci­sions about their fer­til­ity.
8. Post­par­tum fam­ily plan­ning – Pro­vid­ing fam­ily plan­ning guidance to women at pivotal mo­ments for their health and fer­til­ity, such as af­ter giv­ing birth.The above re­ports are time-capped at eighty hours and fol­low the chronol­ogy of our re­search pro­cess. The re­ports be­gin with pre­limi­nary re­search and iden­ti­fy­ing cru­cial con­sid­er­a­tions. Next, we con­sult with ex­perts. We then cre­ate a weighted fac­tor model and a cost-effec­tive­ness anal­y­sis. Th­ese two method­olo­gies al­low us to nu­mer­i­cally quan­tify an in­ter­ven­tion; by in­clud­ing both, we bal­ance out their differ­ent strengths and weak­nesses. Our fi­nal sec­tion brings to­gether in­for­ma­tion gained through­out the re­search pro­cess.

We have cho­sen to or­ga­nize our re­ports in this way to in­crease trans­parency. Read­ers are able to fol­low the re­search as it un­folds and de­vel­ops, and can see how an idea performs from mul­ti­ple per­spec­tives.

For spe­cific ques­tions on the re­search pro­cess, reach out to Karolina Sarek at karolina@char­i­ty­science.com.


Lead re­searcher: Ge­orge Bridg­wa­ter george@char­i­ty­science.com

Our four cause ar­eas achieve im­pact in differ­ent ways, so we tai­lor our met­rics ac­cord­ingly. In this cause area, our cost-effec­tive­ness analy­ses quan­tify im­pact us­ing two met­rics: the satis­fac­tion with life scale (SWLS), and qual­ity-ad­justed life years (QALYs). We mea­sure the ex­pected num­ber of in­cre­men­tal in­creases on the SWLS, and of QALYs per dol­lar spent. We use two met­rics be­cause of how difficult it is to cap­ture sub­jec­tive well-be­ing: al­though we be­lieve that the SWLS may more ac­cu­rately re­flect well-be­ing, QALYs offer the benefit of be­ing widely used and thus more com­pa­rable. Our met­rics re­port (forth­com­ing) pro­vides fur­ther in­for­ma­tion on the var­i­ous met­rics used in this space.

Guided self-help

Distribut­ing self-help work­books en­ables in­di­vi­d­u­als to make progress on their men­tal health ei­ther in­de­pen­dently (known as pure self-help) or with min­i­mal sup­port (guided self-help). Our re­search found strong ev­i­dence that guided self-help in par­tic­u­lar can sig­nifi­cantly im­prove symp­toms of anx­iety, de­pres­sion, and chronic pain – con­di­tions that col­lec­tively af­fect over 2 billion peo­ple wor­ld­wide.

De­mand­ing much less staff time than face-to-face ther­apy, self-help can de­liver highly cost-effec­tive treat­ment for men­tal health is­sues. Ini­tially, we were con­cerned about dropout rates be­cause of the low in­ten­sity of self-help. How­ever, we found ev­i­dence that dropout rates for self-help are similar to those of tra­di­tional ther­apy. For se­vere cases, the strong in­ter­per­sonal re­la­tion­ship be­tween ther­a­pist and pa­tient may be prefer­able. But for those with mild and mod­er­ate men­tal health is­sues, self-help seems a strong al­ter­na­tive to tra­di­tional ther­apy.

We mod­eled the cost-effec­tive­ness of self-help un­der six sce­nar­ios. Th­ese vary ac­cord­ing to the in­come level of the coun­try of op­er­a­tions, sup­port available to benefi­cia­ries, and whether the work­books dis­tributed are new or reused. The most likely sce­nar­ios in­volve dis­tribut­ing new work­books and offer­ing tele­phone sup­port.

In a high-in­come coun­try, mod­el­ing our most likely sce­nario sug­gests a cost-effec­tive­ness of $78 per in­cre­men­tal in­crease in SWLS score, or $20,016 per QALY. In a low-in­come coun­try, cost-effec­tive­ness rises to $20 per SWLS in­crease, and $1,203 per QALY. All of these figures ac­count for the coun­ter­fac­tual im­pact of fund­ing and co-founders.

Our re­search sug­gests that sup­port­ing in­di­vi­d­u­als with short phone calls strikes a good bal­ance be­tween fac­tors like cost and effect size. On­line op­tions such as Skype-based sup­port or the use of avatars seem promis­ing and may be worth ex­plor­ing in fu­ture, but cur­rently the ev­i­dence for these is weaker. We also would like to see more ev­i­dence in low-in­come coun­tries, as most of the ran­dom­ized con­trol­led tri­als on self-help took place in high-in­come coun­tries.

In­volv­ing largely in­de­pen­dent work means that self-help may be a promis­ing way to de­liver ther­apy in con­texts where stigma sur­rounds men­tal health, and where peo­ple are there­fore re­luc­tant to visit a ther­a­pist. Depend­ing on the con­text, work­books may need to be cul­turally adapted, ideally with in­put from the lo­cal com­mu­nity. An ex­pert we spoke to sug­gested this could be done in three months, al­though the time frame will de­pend on var­i­ous fac­tors and could eas­ily ex­tend longer.

A key ad­van­tage of self-help is its flex­i­bil­ity. A new char­ity will be able to ex­plore var­i­ous ther­a­peu­tic tech­niques and tar­get a range of benefi­cia­ries. Based on un­met needs and available ev­i­dence, we think that Over­com­ing Anx­iety, Over­com­ing De­pres­sion, and Liv­ing With Your Pain are promis­ing work­books for a self-help char­ity to dis­tribute, but we ex­pect co-founders of this char­ity to ex­per­i­ment and pivot ac­cord­ing to their find­ings.

Our re­search on self-help has left us con­fi­dent in our recom­men­da­tion, but we still have some re­main­ing ques­tions and space for de­vel­op­ment. We’d like to see fur­ther stud­ies in differ­ent con­texts, and into other types of sup­port (e.g. avatars). We also have re­main­ing con­cerns over how best to re­cruit benefi­cia­ries and over the lo­gis­tics of man­age­ment at large scale (i.e., up­wards of 10,000 benefi­cia­ries).

Read the full report


Lead re­searcher: Ali Ladak ali.ladak@char­i­ty­science.com

For our policy re­search, im­pact flows from two main chan­nels: health out­comes and eco­nomic gain. Health can be mea­sured us­ing dis­abil­ity-ad­justed life years (DALYs), and eco­nomic im­pacts us­ing years of in­come. To com­bine both chan­nels of im­pact in our anal­y­sis, we use GiveWell’s moral weights to roughly con­vert be­tween DALYs and mon­e­tary terms. We ul­ti­mately ex­press cost-effec­tive­ness in benefit-cost ra­tio (con­vert­ing health out­comes into mon­e­tary terms) as well as cost per DALY equiv­a­lent (con­vert­ing eco­nomic gains to DALYs).

Lead paint regulation

As of May 31 2020, less than 40% of coun­tries have con­firmed lead paint reg­u­la­tions. Yet there is no safe level of lead ex­po­sure, whose effects give rise to a range of nega­tive health and eco­nomic con­se­quences. Ad­vo­cat­ing for policy change in coun­tries with­out reg­u­la­tions on lead paint, par­tic­u­larly those with large and grow­ing paint mar­kets, could alle­vi­ate some of these harm­ful con­se­quences and cre­ate pos­i­tive ex­ter­nal­ities. Our cost-effec­tive­ness anal­y­sis found that lead paint reg­u­la­tion could avert a DALY equiv­a­lent for $156; for ev­ery dol­lar spent, we can ex­pect a re­turn of $74.*

Roughly 1% of the global bur­den of dis­ease is due to ex­po­sure to lead, which can cause strokes and heart dis­ease among oth­ers. Through harm­ing brain de­vel­op­ment, lead ex­po­sure can re­sult in lower IQ. This cre­ates in­tel­lec­tual dis­abil­ities in more se­vere cases, and re­duces pro­duc­tivity among those with higher IQs. One study es­ti­mates the pro­duc­tivity costs of lead ex­po­sure in low- and mid­dle-in­come coun­tries to be al­most $1 trillion. The im­pact of lead reg­u­la­tion on IQ alone could cre­ate a range of pos­i­tive out­comes, en­abling peo­ple to in­crease their earn­ings and po­ten­tially spark­ing in­no­va­tion. How­ever, the link from lead ex­po­sure to earn­ings in­volves quite a long and un­cer­tain causal chain: as such, we have con­cerns about the strength of the ev­i­dence here.

Policy in­ter­ven­tions tend to have a lower chance of suc­cess than in­ter­ven­tions in our other cause ar­eas: shap­ing a poli­ti­cal pro­cess in­volves unit­ing a lot more mov­ing parts and com­pet­ing goals than de­liv­er­ing vac­cines or bed nets. How­ever, sev­eral fac­tors sug­gest that lead paint reg­u­la­tion could be rea­son­ably tractable. It’s pretty un­con­tro­ver­sial, and is un­likely to face much op­po­si­tion from the broader in­dus­try and other stake­hold­ers.

There’s also a rea­son­ably clear path for­ward for or­ga­ni­za­tions work­ing on lead paint. Re­sources available in­clude the UNEP toolkit and model law, and the sup­port of the In­ter­na­tional Pol­lu­tants Elimi­na­tion Net­work (IPEN), which has a track record of suc­cess, hav­ing passed lead paint reg­u­la­tion in over 20 coun­tries to date. IPEN flags three el­e­ments as key to effect­ing policy change: pro­vid­ing the data, get­ting me­dia at­ten­tion, and work­ing con­sis­tently.

Our con­ver­sa­tion with IPEN raised sev­eral key points. Firstly, IPEN con­sid­ers that fund­ing is a ma­jor bot­tle­neck among or­ga­ni­za­tions ad­vo­cat­ing lead paint reg­u­la­tion. Rather than found­ing a new or­ga­ni­za­tion, they sug­gest it may thus be more effec­tive to sup­port NGOs cur­rently work­ing in the space. This said, there are promis­ing coun­tries in sub-Sa­haran Africa where IPEN does not yet have a part­ner, and where a new EA or­ga­ni­za­tion could be valuable.

Reg­u­lat­ing lead paint is an im­por­tant first step, but it does not fully solve the prob­lem. Although lead paint is known to be a key source of lead ex­po­sure, it’s un­clear what pro­por­tion of ex­po­sure it causes: we con­ser­va­tively as­sume this figure to be 25%. Ad­di­tion­ally, the pro­posed reg­u­la­tions would not re­move lead from the en­vi­ron­ment, but rather aim to pre­vent more from com­ing in. Once reg­u­la­tions are in place, com­pli­ance may also be an is­sue (as touched on in this ex­pert con­ver­sa­tion).

Fully erad­i­cat­ing the harm­ful effects of lead paint will take time. Even so, lead paint reg­u­la­tion looks to be a promis­ing in­ter­ven­tion for a new char­ity. Mov­ing for­ward, we will con­tinue to dis­cuss with IPEN how best to fit into this space.

Read the full report

​Al­co­hol regulation

​With 4.3% of the global bur­den of dis­ease at­tributable to al­co­hol, ad­vo­cat­ing for reg­u­la­tion has the po­ten­tial to be highly cost-effec­tive. It’s also quite ne­glected, so the coun­ter­fac­tual im­pact of a suc­cess­ful new char­ity is strong. Our anal­y­sis ex­plored sev­eral vari­a­tions of al­co­hol reg­u­la­tion, in­clud­ing availa­bil­ity, pric­ing, and ad­ver­tis­ing. Of the WHO’s best buys, tax­a­tion looks par­tic­u­larly promis­ing based on its cost-effec­tive­ness and the strength of the ev­i­dence. Although 155 coun­tries already tax al­co­hol, less than a quar­ter ad­just these taxes for in­fla­tion, so they be­come less effec­tive with time.

Al­co­hol con­sump­tion causes both short and long-term health prob­lems, as well as nega­tive eco­nomic im­pacts. It is a risk fac­tor for over 200 con­di­tions, in­clud­ing can­cers and car­dio­vas­cu­lar dis­eases; it leads to in­juries (e.g. road ac­ci­dents) and vi­o­lence (e.g. do­mes­tic abuse). Sev­eral meta-analy­ses provide strong ev­i­dence that tax­a­tion low­ers con­sump­tion: it seems that a 10% in­crease in price re­duces con­sump­tion by 5%. Given the wide-rang­ing costs as­so­ci­ated with con­sump­tion, tighter reg­u­la­tion looks highly im­pact­ful.

Our re­search mod­els the cost-effec­tive­ness of an 18% in­crease in price. This figure is roughly in line with WHO’s mod­els and with the high­est rates in the EU and Asia. We ex­pect such a price in­crease to avert a DALY equiv­a­lent for $96. Ex­pressed differ­ently, this sug­gests an ex­pected re­turn of $115 for ev­ery dol­lar in­vested.*

Two key is­sues are that reg­u­la­tion could be pa­ter­nal­is­tic and could hit low-in­come con­sumers hard­est. We can’t neatly re­solve these crit­i­cisms, as at a cer­tain point it boils down to val­ues. But to offer some brief thoughts, the nega­tive im­pact of al­co­hol tax­a­tion on low-in­come con­sumers’ wallets is ar­guably out­weighed by the pos­i­tive im­pacts on their health. Spend­ing rev­enue from the tax on health care could also coun­ter­bal­ance this con­cern. More­over, as a cri­tique, pa­ter­nal­ism im­plies that con­sumer choices are truly free. Such an as­sump­tion ig­nores, for ex­am­ple, the in­fluence of ex­ter­nal fac­tors (e.g. the in­dus­try) in shap­ing de­ci­sions.

The over­ar­ch­ing prob­lem with al­co­hol reg­u­la­tion (and a rea­son it’s so ne­glected), is that it’s hard. Policy in­ter­ven­tions are gen­er­ally tricky, and al­co­hol reg­u­la­tion par­tic­u­larly so. Ad­vo­cates face a pow­er­ful in­dus­try lobby and the weight of pop­u­lar opinion (it­self to an ex­tent crafted by the in­dus­try). A new char­ity ad­vo­cat­ing for al­co­hol reg­u­la­tion would need to be highly at­tuned to the lo­cal con­text and sen­si­tive to cul­tural val­ues.

As this in­ter­ven­tion is quite con­tro­ver­sial, we ex­pect fundrais­ing to be difficult for a new char­ity, par­tic­u­larly out­side the EA move­ment. Nega­tive nar­ra­tives around reg­u­la­tion are per­va­sive: for ex­am­ple, pro­hi­bi­tion in 1920s USA casts a long shadow, even though this in­ter­ven­tion does not seek to ban al­co­hol. Align­ing with es­tab­lished ac­tors in the space (e.g. the WHO, GAPA) could in­crease the chances of suc­cess for a new char­ity. Even so, we’re cau­tious about recom­mend­ing al­co­hol reg­u­la­tion due to con­cerns over fund­ing.

*Please note that these figures de­pend on model as­sump­tions, con­sis­tently es­ti­mated within our health and de­vel­op­ment re­search but not nec­es­sar­ily be­yond.

Read the full report


Lead re­searcher: Vicky Cox vicky@char­i­ty­science.com

To com­pare in­ter­ven­tions in this space, we cre­ated a weighted an­i­mal welfare in­dex. Re­searchers score each an­i­mal’s welfare on this in­dex to ar­rive at an over­all score in welfare points, ad­just­ing for prob­a­bil­ity of sen­tience and ex­pected lifes­pan. Our cost-effec­tive­ness analy­ses quan­tify the num­ber of such welfare points we ex­pect to af­fect per dol­lar spent. To read more about how we cre­ated the welfare in­dex, re­fer to this blog post.

Shrimp welfare

Manag­ing lev­els of dis­solved oxy­gen tack­les chronic suffer­ing for farmed aquatic an­i­mals. With in­suffi­cient oxy­gen in the wa­ter, the an­i­mals strug­gle to breathe. Our 2019 re­search pe­riod iden­ti­fied man­ag­ing dis­solved oxy­gen for farmed fin­fish as a promis­ing in­ter­ven­tion, and led to the found­ing of Fish Welfare Ini­ti­a­tive. This year, we ex­plored whether this could be a promis­ing in­ter­ven­tion for shrimp, an­other ne­glected farmed an­i­mal.

Although our re­port fo­cuses on oxy­gena­tion, we look for­ward to fur­ther work on shrimp welfare, as this is highly ne­glected. For ex­am­ple, con­ver­sa­tions with Daniela R. Wald­horn high­lighted an­other po­ten­tially promis­ing in­ter­ven­tion: pre­vent­ing eye­stalk ab­la­tion, the prac­tice of cut­ting off fe­male shrimp’s eye­stalks. A new char­ity’s pri­mary aim should be to im­prove welfare for shrimp as effec­tively as pos­si­ble, whether this be through man­ag­ing dis­solved oxy­gen, pre­vent­ing eye­stalk ab­la­tion, or an­other well-ev­i­denced in­ter­ven­tion.

In our eighty-hour re­port, we speci­fi­cally model the effect of sub­si­diz­ing aer­a­tion for farm­ers of whiteleg shrimp in Viet­nam. Although we are less sure that shrimp are sen­tient than larger crus­taceans, the im­mense scale at which they are farmed led us to pri­ori­tize their welfare. Our anal­y­sis sug­gests that man­ag­ing dis­solved oxy­gen would cost-effec­tively im­prove their welfare. Through sub­si­diz­ing aer­a­tion, we can af­fect al­most 90 welfare points per dol­lar. This figure takes into ac­count co-founder and fund­ing coun­ter­fac­tu­als.

Con­ver­sa­tions with ex­perts sug­gested that Viet­nam would be the best place for an oxy­gena­tion in­ter­ven­tion for shrimp, due to the scale of the prob­lem and the need for cli­mate change re­silience. As tem­per­a­tures rise, dis­solved oxy­gen lev­els de­crease. In a warm­ing world, aer­a­tion will be­come in­creas­ingly nec­es­sary.

A con­cern with this in­ter­ven­tion is that farm­ers may in­crease stock­ing den­sity, crowd­ing a greater num­ber of shrimp into the same space. This said, higher stock­ing den­sity cre­ates var­i­ous other costs: feed, la­bor, main­te­nance, dis­ease rates, and the size of the shrimp are nega­tively af­fected. As such, it is not in a farmer’s in­ter­est to dras­ti­cally in­crease stock­ing den­sity. How­ever, given the po­ten­tial to harm shrimp welfare, this is­sue is worth ad­dress­ing. If a new char­ity de­cides to sub­si­dize aer­a­tion equip­ment, sub­sidies could be con­di­tional upon a cer­tain max­i­mum stock­ing den­sity.

We ex­pect a new char­ity work­ing on shrimp welfare to face no in­sur­mountable challenges. Cur­rently no es­tab­lished fund­ing land­scape ex­ists, as shrimp welfare is highly ne­glected. But there are promis­ing signs that effec­tive al­tru­ist fund­ing would be available (e.g. this re­quest for pro­pos­als). Depend­ing on the co-founders’ back­grounds, cul­tural bar­ri­ers may pose an is­sue, as most of the top shrimp farm­ing coun­tries are in Asia. How­ever, hiring lo­cal tal­ent could help bridge this gap.

While we were only able to model the im­pact of a dis­solved oxy­gen in­ter­ven­tion, our re­search re­vealed sev­eral promis­ing path­ways for im­prov­ing shrimp’s welfare. A new char­ity may wish to ad­dress wa­ter qual­ity in gen­eral, for ex­am­ple with an ask that en­com­passes pH as well as dis­solved oxy­gen lev­els. They may also choose to work on pre­vent­ing other forms of suffer­ing for farmed shrimp, such as the prac­tice of eye­stalk ab­la­tion men­tioned above.

Over­all, al­though we recom­mend dis­solved oxy­gen, we en­courage a new char­ity to be flex­ible and open-minded in their ap­proach. As they ex­plore the is­sue fur­ther through farm vis­its and micro pi­lots, en­trepreneurs should not lose sight of their first pri­or­ity: im­prov­ing the welfare of shrimp.

Read the full report

​Feed fortification

​Micronu­tri­ent for­tifi­ca­tion can be a highly cost-effec­tive in­ter­ven­tion (e.g. For­tify Health). Re­peat­edly lay­ing eggs takes a sig­nifi­cant toll on farmed hens, de­plet­ing their nu­tri­ent lev­els and lead­ing to frac­tures and bro­ken bones. To counter these and as­so­ci­ated welfare prob­lems among egg-lay­ing hens, a new char­ity could ad­vo­cate for for­tify­ing their feed with cal­cium, phos­pho­rus, and vi­tamin D3.

Strong ev­i­dence sup­ports the po­ten­tial for feed for­tifi­ca­tion to re­duce suffer­ing. We found ten stud­ies on the effect of cal­cium on hen welfare, four­teen on phos­pho­rus, and six on vi­tamin D3.

Our mod­els sug­gest a new char­ity could af­fect roughly 7.5 welfare points per dol­lar when in­clud­ing coun­ter­fac­tual con­sid­er­a­tions for co-founders and fund­ing. To put this an­other way, each dol­lar in­vested in this in­ter­ven­tion spares on ex­pec­ta­tion roughly two hens from bro­ken bones. If we do not ac­count for coun­ter­fac­tu­als in our calcu­la­tion, cost-effec­tive­ness in­creases to al­most thirty welfare points.

A new char­ity should be open to var­i­ous op­tions, and pivot should their own re­search and farm vis­its sug­gest a bet­ter path for­ward than our mod­eled ap­proach. This said, our re­search in­di­cates that work­ing with farm­ers in In­dia is most promis­ing. In­dia is among the top five egg-pro­duc­ing coun­tries, and nu­tri­ent lev­els are be­low par on many farms as its feed stan­dards are only recom­men­da­tions.

Within In­dia, we con­sid­ered a few pos­si­ble strate­gies for this in­ter­ven­tion, in­clud­ing cor­po­rate and gov­ern­ment cam­paigns and part­ner­ing with cer­tifi­ca­tion bod­ies. We be­lieve sub­si­diz­ing for­tifi­ca­tion to be the most promis­ing ap­proach, at least in the short term. It pre­sents the high­est chance of con­vinc­ing farm­ers to raise feed qual­ity, and the low­est risk of ham­per­ing the progress of on­go­ing cor­po­rate cam­paigns such as cage-free com­mit­ments. A new char­ity should col­lab­o­rate with the ma­jor play­ers of these cam­paigns to avoid this con­cern.

A few po­ten­tial challenges are worth touch­ing on. Offer­ing sub­sidies means that the char­ity will need to fundraise well and rapidly, which may thus be a key bot­tle­neck for this in­ter­ven­tion. We have also as­sumed that for­tifi­ca­tion will be as sim­ple as pro­vid­ing farm­ers with a micronu­tri­ent pre­mix; based on our re­search, this seems a rea­son­able as­sump­tion but will re­quire con­fir­ma­tion. Fi­nally, al­though one ex­pert raised hu­mane wash­ing as a pos­si­ble con­cern, we found lit­tle em­piri­cal ev­i­dence to set­tle the ques­tion, and feel that the amount of suffer­ing this in­ter­ven­tion could alle­vi­ate jus­tifies the risk.

Note: Our re­search into feed for lay­ing hens also cov­ered al­ter­nate meth­ods of feed re­stric­tion, which change the qual­ity rather than quan­tity of feed. In prac­tice, this would in­volve re­duc­ing the en­ergy or pro­tein con­tent of feed. How­ever, we con­cluded that these meth­ods do not sub­stan­tially im­prove welfare for hens, and ul­ti­mately could even harm their well-be­ing (e.g. by in­creas­ing feather peck­ing, and dis­cour­ag­ing the switch to slower grow­ing breeds). For this rea­son, we do not recom­mend a new char­ity ad­vo­cat­ing al­ter­nate meth­ods of feed re­stric­tion.

Read the full report

Ask research

Our 2019 re­search in­di­cated that re­search­ing pos­si­ble asks for the an­i­mal move­ment holds promise. A new ask in­sti­tute would con­duct deep re­search to max­i­mize the im­pact of gov­ern­men­tal and cor­po­rate cam­paigns. We es­ti­mate that the best ask could cre­ate three times as much benefit for farmed an­i­mals as the worst.

For fur­ther in­for­ma­tion, see this blog post. ​

Read the full report


Lead re­searcher: Juliette Finetti juliette@char­i­ty­science.com

This cause area is par­tic­u­larly sen­si­tive to one’s eth­i­cal per­spec­tive, so we en­courage read­ers to ex­plore our con­sid­er­a­tions and as­sump­tions in the sup­ple­men­tary re­port. We es­ti­mate cost-effec­tive­ness both in dol­lars per ad­di­tional user of con­tra­cep­tive, and in dol­lars per un­in­tended birth averted. Model­ing un­in­tended births averted al­lows us to more eas­ily cross-com­pare fam­ily plan­ning in­ter­ven­tions, and bet­ter cap­tures the im­pact of an in­ter­ven­tion on out­comes we care about, such as women’s health and well-be­ing.

Post­par­tum fam­ily planning

​After a woman gives birth (i.e. the post­par­tum) is a par­tic­u­larly cru­cial time for fam­ily plan­ning. Short-spaced preg­nan­cies pose a risk for the health of both mother and child, with in­fant mor­tal­ity ris­ing at shorter in­ter­vals. This in­ter­ven­tion would provide post­par­tum women with fam­ily plan­ning in­for­ma­tion and coun­sel­ing as well as con­tra­cep­tive op­tions to miti­gate the risk of un­in­tended preg­nan­cies.

Lack of in­for­ma­tion about fam­ily plan­ning can lead to an un­ex­pected and po­ten­tially dan­ger­ous preg­nancy. For ex­am­ple, a new mother can un­know­ingly be fer­tile be­fore her menses re­turn. As well as pro­vid­ing valuable in­for­ma­tion, open dis­cus­sions about fam­ily plan­ning can help com­bat stigma. Rais­ing the topic when a woman is already at a health care fa­cil­ity also en­sures she can read­ily ac­cess con­tra­cep­tives should she so choose.

Our re­port con­sid­ers sev­eral po­ten­tially op­por­tune mo­ments for dis­cussing fam­ily plan­ning, in­clud­ing dur­ing preg­nancy, af­ter an abor­tion*, shortly af­ter child­birth, and with child­hood vac­ci­na­tion ap­point­ments. The ev­i­dence base for post­par­tum fam­ily plan­ning is stronger than that for any other in­ter­ven­tion we ex­plored in this cause area. Offer­ing ser­vices shortly af­ter child­birth is par­tic­u­larly well ev­i­denced, al­though we can­not be cer­tain when gen­er­al­iz­ing the re­search to other con­texts. The best path for­ward would likely be to broach the topic at more than one mo­ment, so that a woman has enough time to re­flect on her de­ci­sion and dis­cuss op­tions with her part­ner.

Govern­ment buy-in will be im­por­tant for the suc­cess of this in­ter­ven­tion, as it re­quires train­ing health work­ers. Such a part­ner­ship could be difficult for a new char­ity to achieve, es­pe­cially since it would add to health work­ers’ reg­u­lar du­ties. It would be vi­tal to ex­plore ways to sup­port health work­ers so that they are not over­worked. Although there are difficul­ties here, we es­ti­mate a 70% prob­a­bil­ity of suc­cess­fully part­ner­ing with the gov­ern­ment through col­lab­o­rat­ing with ex­ist­ing or­ga­ni­za­tions. Still, lo­gis­tics and ex­e­cu­tion seem challeng­ing for this in­ter­ven­tion, par­tic­u­larly given the difficulty of find­ing good hires – ideally, team mem­bers would be well con­nected as well as tal­ented.

This in­ter­ven­tion may risk over­load­ing new moth­ers with in­for­ma­tion, at a time when there’s already so much for them to fo­cus on. The ev­i­dence we found (e.g. this High Im­pact Prac­tices brief) sug­gests this is un­likely to have a nega­tive im­pact. Even so, we think en­trepreneurs should be mind­ful of this is­sue.

Like our other recom­mended fam­ily plan­ning in­ter­ven­tion, we think Ghana would be a good place for a post­par­tum fam­ily plan­ning char­ity. We could plau­si­bly reach al­most 600,000 post­par­tum women per year in Ghana, so it looks promis­ing in terms of scale. Ad­di­tion­ally, low stock-out rates of con­tra­cep­tion sug­gest that sup­ply is not an is­sue. Our model sug­gests that by train­ing health work­ers in Ghana in part­ner­ship with the gov­ern­ment, we could achieve a cost-effec­tive­ness of $39 per ad­di­tional user of con­tra­cep­tion, or $67 per un­in­tended birth averted.

*Given the difficul­ties of reach­ing women post abor­tion, we ul­ti­mately de­cided to ex­clude this ap­proach from the bulk of our re­search.

Read the full report

Mass me­dia campaigns

​Run­ning mass me­dia cam­paigns would aim to counter mis­in­for­ma­tion and foster dis­cus­sion sur­round­ing con­tra­cep­tion, and in this way sup­port women’s de­ci­sions about their fer­til­ity. Upfront in­vest­ment is needed to cre­ate and tai­lor the cam­paign. But once the ini­tial road­blocks are past, mass me­dia cam­paigns can reach a broad au­di­ence and have the po­ten­tial to be highly cost-effec­tive.

In­tu­itively, me­dia cam­paigns are a use­ful tool. Think of the en­dur­ing power of the iconic 1943 “We Can Do It!” cam­paign: to this day, Rosie the Riveter is a well-known cul­tural sym­bol. But mea­sur­ing im­pact is difficult in this space, and feed­back loops are min­i­mal. Ques­tions about the ev­i­dence base for other mass me­dia cam­paigns (e.g. see GiveWell on DMI’s child sur­vival cam­paigns) meant that our team ini­tially had reser­va­tions. How­ever, a re­cent ran­dom­ized con­trol­led trial of DMI’s fam­ily plan­ning cam­paign in Burk­ina Faso shows en­courag­ing signs. The study found, for ex­am­ple, that con­tra­cep­tives use in­creased by al­most 6 per­centage points. The four ex­perts we spoke to were also pos­i­tive about the promise of mass me­dia cam­paigns.

For this in­ter­ven­tion to achieve im­pact, con­tra­cep­tives must be available. Helping a woman to make an in­formed de­ci­sion about her fer­til­ity is not enough if she then can­not ac­cess those con­tra­cep­tives. Other difficul­ties flagged in our re­search pro­cess re­late to lo­gis­tics, e.g. elec­tric­ity short­ages and fos­ter­ing part­ner­ships with broad­cast­ing com­pa­nies, though limited field op­er­a­tions even at large scale fa­vors the ease of ex­e­cu­tion. Given the skill set needed to run this in­ter­ven­tion (man­age­ment, ex­pe­rience with me­dia, etc.), we also ex­pect find­ing tal­ent to be tricky.

Devel­op­ment Me­dia In­ter­na­tional (DMI) is already do­ing great work on ev­i­dence-based mass me­dia cam­paigns. Their sat­u­ra­tion+ ap­proach in­volves three strands: sat­u­ra­tion (rep­e­ti­tion, reach); sci­ence (ro­bust M&E); and sto­ries (en­gag­ing mes­sag­ing). Of course, that DMI ex­ists raises the ques­tion of a new char­ity’s coun­ter­fac­tual im­pact. We be­lieve that a char­ity could learn from DMI, and add in­de­pen­dent value by filling the gaps rather than rein­vent­ing the wheel with a wholly new strat­egy. Our pre­limi­nary re­search sug­gests that Ghana could be a promis­ing lo­ca­tion: the bar­ri­ers to con­tra­cep­tion seem con­nected to mis­in­for­ma­tion more than to ac­cess, and DMI does not plan to ex­pand to Ghana. See our ex­pert con­ver­sa­tion with Roy Head, CEO of DMI, for fur­ther de­tails on some of these ques­tions.

Our cost-effec­tive­ness anal­y­sis as­sumes that a new char­ity will op­er­ate in Ghana, dis­sem­i­nat­ing ra­dio cam­paigns us­ing DMI’s sat­u­ra­tion+ ap­proach. In this sce­nario, we es­ti­mate the cost per ad­di­tional user of con­tra­cep­tion to be less than $9, and the cost of pre­vent­ing an un­in­tended birth to be $43.

Run­ning this in­ter­ven­tion well re­quires deep en­gage­ment with the lo­cal com­mu­ni­ties served. Un­der­stand­ing the cul­tural val­ues and con­tex­tual bar­ri­ers to con­tra­cep­tion will help en­sure that the cam­paigns in­crease use in a way that re­spects and en­hances women’s de­ci­sion-mak­ing power. Our re­search sug­gests that the key bar­ri­ers re­late to lack of in­for­ma­tion and so­cial per­cep­tions, but a new char­ity should look more closely at such ques­tions within the spe­cific cul­tural con­text.

Read the full report

Cost-effec­tive­ness anal­y­sis is vi­tal as we seek to cre­ate the great­est pos­i­tive im­pact with limited re­sources. But given their re­li­ance on model as­sump­tions, we can­not take such ex­pected value es­ti­mates liter­ally.

Please note that the sce­nar­ios we model in our cost-effec­tive­ness analy­ses do not nec­es­sar­ily re­flect how a char­ity should im­ple­ment that in­ter­ven­tion. Rather, they seek to provide a sense of the pos­si­bil­ities and sketch out plau­si­ble im­pact per dol­lar spent.

The CE team is deeply grate­ful to ev­ery­one who has offered their time and ex­per­tise through­out the 2020 re­search pro­cess. Thank you to all the ex­perts who shared thoughts on po­ten­tial in­ter­ven­tions and pro­vided valuable feed­back on our re­ports. Thank you also to our ded­i­cated team of vol­un­teer in­terns, who helped var­i­ously with re­search, graph­ics, and proofread­ing.


Re­mem­ber that you can help us found more effec­tive char­i­ties by donat­ing to our re­search and In­cu­ba­tion Pro­gram, or donate to one of the new char­i­ties we’ve helped launch. Dona­tions to Char­ity En­trepreneur­ship from tax res­i­dents of the US, UK, and Canada are tax-de­ductible.
A dona­tion of $50,000 would launch an ad­di­tional char­ity from our 2020 batch. DONATE