What's the Best Way to "Sponsor" Baby Girls?

A reader named Gunjan Aggarwal writes:

I came to the U.S. 7 years ago, worked in U.K./Switzerland/Netherlands/India prior to that. I work in human resources and have been fortunate to have been successful thus far in my career. We are moving on to a new location and a new job this year but this year will also perhaps give me an opportunity to invest some time/leadership on a cause that I have been very keen to “do something about”: contribute towards improving the lot of the girl child in India.
I have always thought of crowd-sourcing an incentive scheme by which we will “adopt” a few girls in their womb and give the parents a small amount every month, $50, to give birth to their girl child, to educate her till the age of 21. I was even more determined to do this in the wake of all the news about crimes against women in India — but then I heard your podcast on the “Cobra Effect.”

I would love to connect and get your thoughts on “scheming” this incentive forward!

I applaud Gunjan’s initiative, admire his creative idea — but yes, am concerned that it might be easily gamed (and, perhaps even more easily and more importantly, simply wouldn’t work). So let’s help him out: please use the comments section to brainstorm the best way to set up the program he’s thinking about, replete with potential pitfalls. Bonus points for anyone who knows of a similar program that’s working well (neither Gunjan nor I could come up with anything).

Addendum: The NYU economist and development scholar William Easterly e-mails us some cold water:

Stephen and Steve, can I volunteer my services to save you from embarrassment on the blog post today on sponsoring infant girls? [Bill: it would seem too late for that, no? 🙂] It’s been known in aid and development for decades that child sponsorship does not work (unless you by “work” you mean attract donations). The NGOs that originally did it (most notoriously Save the Children) have been forced by critics to abandon it, and no reputable NGO promotes child sponsorship today.  The reason it never worked (and in fact Save the Children was also forced to admit that they really never even actually did it) is simple: the administrative costs of tracking small donations from an individual donor to an individual recipient child are enormous, so that the administrative costs would eat up all of the donation and then some. So there’s no need to crowdsource this question: just ask any development economist or NGO veteran.

All the best,


Easterly’s note makes me wonder if indeed transactions costs are still too high in this digital heyday to accommodate Gunjan’s idea, or one like it.

Also: how about investing in a different kind of person — young high achievers who need capital to fulfill their dreams — in Upstart.com?

Valerie Ross

I work for an organization that for the past 3 years has been sponsoring girls in rural villages in China where we find the highest rates of gendercide (and locals who are willing to help) by providing a monthly stipend to mothers who are pregnant with or have recently given birth to girls in an effort to encourage the family to keep the girl. So far our program only supports through the first year of the girls life and then re-evaluates on a case-by-case basis (we are a tiny organization). We have some data and testimonial evidence that suggests the program is working, and we also found some great work by Avraham Ebenstein in which he theorizes that a monetary incentive would actually help solve the gender imbalance. Check out his paper: http://www.allgirlsallowed.org/estimating-dynamic-model-sex-selection-china

Mike B

If left to their own devices wouldn't the increasing scarcity of female children eventually increase their realitive value? This is a problem that basic supply and demand forces are wonderful at solving.

Seminymous Coward

That's true; however, the currently "surplus" supply is composed of real, living human beings, so I don't think a wait-it-out approach is very moral.


I applaud Mr. Aggarwal on both his desire to help girls as well as his seeking guidance on unintended consequences before moving forward.

While the Cobra Effect is a definite concern especially in terms of creating a "growth industry", my concern is the use of extrinsic motivation (money) for this endeavor. The research shows that the external motivator may actually cause a decrease in the care these children may receive, or, at the very least, will not keep them from receiving the best possible (for their given situation) care and education.

I would urge Mr. Aggarwal to read both "Drive" and the mis-named yet perfectly named "To Sell Is Human" by Daniel H. Pink. Both would help him understand my stated concerns as well as perhaps offer him a different approach to the dilemma. Good luck, sir.

John Borneman


Mike B

Why do you want to increase the reproductive capacity of countries that are already suffering from severe overpopulation? Male preference is actually a good way for these countries to reduce their long term birthrates. You can't easily change culture and subsidizing female children will only create more people that will more than likely wind up abused, oppressed and possibly set on fire. If cultures can't respect women by all means let them not have any.

Seminymous Coward

How exactly do you think these "excess" female children are exiting the population now?

Timothy Ogden

Mr. Easterly in this case is flat out wrong on views of child sponsorship in general.

Child sponsorship is far from abandoned. In fact it is still accounts for a significant portion of American charitable dollars leaving the country.

Save the Children continues to run a large child sponsorship program (https://sponsor.savethechildren.org/?msource=spxgpspn0712&gclid=CN3CyNrV4LQCFUOK4AodJlUAIw), as does World Vision (http://donate.worldvision.org/OA_HTML/xxwv2DoChildSearch_B.jsp?xxwvLocation=0000&xxwvSearchType=ALL), Feed the Children (http://www.feedthechildren.org/site/PageServer?pagename=org_child_sponsorship) and Compassion (http://www.compassion.com/), the four largest American NGOs focused internationally.

Enter your name...

You might pick a village and pay for certain milestones, like part of the school fees for girls, or free school uniforms, or a bigger gift upon graduation. If having and educating girls is visibly rewarded in the community, then parents will have less incentive to kill them.

On a side note, as a means of increasing graduation rates, I've wondered why we don't make senior year free (or much, much cheaper). Scholarships seem skewed towards people just beginning school, but the people who reach the final year are the ones who have the best chance of finishing.

kris hildebrand

As a coach (but forward so you have an idea of preception of reality), 1) are you not trying to set up a system that people game, isn't that the point (ie: do you really want more female babies or do you want better conditions for females; these are different things; seems that setting up a system to promote more female babies will get you that)
2) what is the real reasons that females are treated poorly? Is it that women do not have enough money so they only spend it on boys, or is that having a boy is more finacially lucritive, so that is what they should spend their resouces on. If i was smart i, would figure out how to more resources from elsewhere to put into my son.…


I think there is a very important point made here.
If families did not have to pay huge amounts of money to get their girls married off (often resulting in increased poverty for the family), then they would be having more baby girls. Work to eliminate 'bride price' and, more generally, to increase respect and consideration for women in society, and families will have baby girls again.

Enter your name...

In terms of increasing school attendance, I've read that free food (school breakfast or lunch) is efficient.

Timothy Ogden

Took me a while to track this down, but here is the only rigorous attempt that I know of to evaluate child sponsorship programs. It uses a discontinuity (varying age cutoffs and varying times of implementation) in participation in six different countries and finds positive effects on adult life outcomes for children who had been sponsored:


Now this shouldn't convince anyone that child sponsorship "works" or is a good idea. But it should enter your Bayesian filter.


If the primary challenge is the cost to administer, why not reduce it by monitoring less? By this I mean give a larger stipend less frequently.

Perhaps at birth and then every few years the sponsor girls are administered a test - pass the test and the family receives the next stipend. If the girl's education becomes viewed as an asset to produce income, the family is likely incentivized to see she succeeds. To me this is a lot like a parent paying for grades (but paid in larger, less frequent lump sums).

I believe this to be a better approach because the incentive structure passes most of the cost of monitoring on to the parents of the sponsored child.

Seminymous Coward

I think there's merit to this idea if you throw in free medical checkups at the same time, which can also screen for abuse.

Alexandre Castro

Brazil has a successful program to increase school attendance that started in the 90’s called “Bolsa Escola”. The goal of the program is to reduce child labor, and families receive a small monthly amount for keeping their child at school instead of working. It has shown interesting results, increasing school attendance but not reducing child labor, since deprived families prefer to combine both incomes. Here’s an article about this effect: http://ideas.repec.org/p/van/wpaper/0407.html

Nevertheless, school attendance did increase. So I wonder what would happen if the value of the payments to families in similar programs were skewed by gender. Maybe significantly increasing the amount for female students might make them more valuable to their families, and cause a positive result. Administrative costs would also be lower if you use the resources through existing programs, such as “Bolsa Escola”.

Is important to say that I don't know if this was ever tried before and that Brazil doesn't face the same gender discrimination problem.



I would have to respectfully disagree with William Easterly: .there are child sponsorship programs that work. Of course no organization tracks the individual donations from all individual donors to particular recipients. He is correct in suggesting that such an effort would be ludicrously expensive. However, NGOs do know on average how much it costs to provide the basics for children that their parents' can't afford (nutrition, education, health care, etc.). By asking individual donors to pay for the average cost of providing those things for one child the tracking challenge Easterly mentions is avoided. Nobody is really sold by the idea that their particular dollars are going to provide for the needs of a particular child. Money is fungible. However, many people are attracted to the idea that they can form a friendship with a particular child through correspondence and see how their money is being used. This personalizes the connection between first and 2nd/3rd world at the expense of a few translators on either end (hardly an onerous expense). By offering educational trips to the countries where the NGOs work and offering to provide an opportunity to meet the sponsored child, this provides another level of optional personal contact and assistance verification to donors. This personal connection and monthly giving also makes it easier for average families to budget and maintain their support long term, lowering fundraising costs.

My wife and I have been sponsoring a girl for the last 12 years. The girl's father was a widower day laborer with two kids who couldn't afford enough food, let alone to send his daughter to school since even primary schools in many countries are exorbitantly expensive for laborers. Instead of having to drop out of school to help provide food for the family, the girl will graduate this year from college with a degree in education. We get several letters each year from her noting her progress, and it has been simply amazing to us to watch this child grow up from a distance.

I would like to know specifically why Mr. Easterly says child sponsorship doesn't work? I must admit to being incredibly skeptical of his claims because my family has seen the results first hand through the letters of a future teacher who would never have even graduated from grade school (let alone had enough food, clothes and access to basic healthcare) were it not for the work of the NGO we donate to (the Christian Foundation for Children and Aging). From my perspective, I truly doubt my wife and I would have given every month for over a decade if we hadn't made a personal connection with a child in another country who was being benefited through our donations. $30 a month is a small price to pay for ensuring that one child's basic needs are met and for being able to maintain a long distance pen pal friendship with a child who needs encouragement and hope to overcome the challenges of being born into poverty in a developing country. Given that only 6% of donations go to CFCA for fundraising and administration, it would seem that my theory about the personal connections forged by child sponsorship lowering fundraising costs also makes sense and likely makes up for the increased overhead of hiring a few translators.

Before Mr. Easterly goes off half-cocked, I would encourage him to carefully look into some child sponsorship organizations and see if what they are doing "isn't working." I look forward to either hearing him present evidence beyond the mere assertion or taking back what he wrote.




So is Easterly saying that most of the 86% of donations going to "programs" actually just goes to administration costs, as opposed to the 5% allocated to "management and general"? Or that they are just lying?

I think that the sponsoring a specific child is just to give you more of an emotional connection and feel like you are making a tangible difference, to incentivise people to donate, not that every dollar you donate always goes solely to that individual.

"World Vision is a Christian relief and development organization dedicated to helping children, families, and communities worldwide reach their full potential by tackling the causes of poverty and injustice.

We help transform the lives of the world’s poorest children and families in nearly 100 countries, including the United States, through interventions such as nutrition and safe drinking water programs, poverty relief and community development programs, disaster assistance, and more. Our non-profit work extends assistance to all people, regardless of their religious beliefs, gender, race, or ethnic background."



Regarding William Easterly's comment about the failure of child sponsorship:

I did some volunteer teaching work with the Salesian Sister of St. Don Bosco/ Madreselva in Ethiopia. I know that they successfully arranged for sponsors to pay for the schooling of individual children. While this model may or may not work without a staff of nuns who consider it their holy duty to dedicate their lives to service I think it's important to note that not EVERY sponsorship program is untenable.