I wrote this article in response to Roko's request for an article about efficient charity. As a disclosure of a possible conflict of interest I'll note that I have served as a volunteer for GiveWell. Last edited 12/06/10.
Charitable giving is widely considered to be virtuous and admirable. If statistical behavior is any guide, most people regard charitable donations to be worthwhile expenditures. In 2001 a full 89% of American households donated money to charity and during 2009 Americans donated $303.75 billion to charity .
A heart-breaking fact about modern human experience is that there's little connection between such generosity and positive social impact. The reason why humans evolved charitable tendencies is because such tendencies served as marker to nearby humans that a given individual is a dependable ally. Those who expend their resources to help others are more likely than others to care about people in general and are therefore more likely than others to care about their companions. But one can tell that people care based exclusively on their willingness to make sacrifices independently of whether these sacrifices actually help anybody.
Modern human society is very far removed from our ancestral environment. Technological and social innovations have made it possible for us to influence people on the other side of the globe and potentially to have a profound impact on the long term survival of the human race. The current population of New York is ten times the human population of the entire world in our ancestral environment. In view of these radical changes it should be no surprise that the impact of a typical charitable donation falls staggeringly short of the impact of donation optimized to help people as much as possible.
While this may not be a problem for donors who are unconcerned about their donations helping people, it's a huge problem for donors who want their donations to help people as much as possible and it's a huge problem for the people who lose out on assistance because of inefficiency in the philanthropic world. Picking out charities that have high positive impact per dollar is a task no less difficult than picking good financial investments and one that requires heavy use of critical and quantitative reasoning. Donors who wish for their donations to help people as much as possible should engage in such reasoning and/or rely on the recommendations of trusted parties who have done so.
A commonly used statistic for charity evaluation which has a thin veneer of analytical rigor is a charity's “overhead ratio”: that is, the relative amounts of money spent on programs vs. administration. According to a press release issued in December 2009 by Philanthropy Action, Charity Navigator, GiveWell, Great Nonprofits, Guidestar and Philanthropedia :
For years, people have turned to the overhead ratio—a measure of how much of each donation is spent on “programs” versus administrative and fundraising costs—to guide their choice of charity. But overhead ratios and executive salaries are useless for evaluating a nonprofit’s impact.
While the idea of sending money “straight to the beneficiaries” is tempting, nonprofit experts agree that judging charities by how much of their money goes to “programs” is counterproductive. “Achieving a low overhead ratio drives many charities to behaviors that make them less effective and means more, not less, wasted dollars,” says Paul Brest, President of the Hewlett Foundation, and co-author of Money Well Spent.
The common focus on low overhead ratio has produced perverse incentives; pressuring some charities to skimp on administrative costs that would improve the efficacy of their programs. More importantly, cost-effectiveness of different charities' activities varies so dramatically as to totally eclipse any usefulness that the overhead ratio might have in a world of charities performing homogeneous activities.
A Comparison of Cost-Effectiveness
A well-known and well-funded charity is the Make-A-Wish Foundation, “a 501(c)(3) non-profit organization in the United States that grants wishes to children (2.5 years to 18 years old) who have life-threatening medical conditions.” According to the website's Managing Our Funds page:
The Make-A-Wish Foundation® is proud of the way it manages and safeguards the generous contributions it receives from individual donors, corporations and other organizations.
Seventy-six percent of the revenue the Make-A-Wish Foundation receives is allotted to program services. This percentage well exceeds the standard upheld by organizations that monitor the work of charities.
And indeed, the percentage allotted to program services is sufficiently high in juxtaposition with other financial statistics so that Charity Navigator grants the Make-A-Wish Foundation its highest rating. But how cost-effective are the charity's programs?
The Make-A-Wish Foundation 2009 Annual Report states that “A record-breaking 13,471 children had their wishes come true in FY09.” The annual report gives a break down of wishes by type: for example, 40.3% of the wishes were trips to the Walt Disney World Resort, 11.7% of them were shopping sprees, 7.1% of them were celebrity meetings and 5.5% of them were cruises.
The annual report claims that in 2009 the charity's “total program and support services” amounted a figure of $203,865,550. Thus, the Make-A-Wish Foundation implicitly reports to spending an average of $15,134 for each wish that it grants.
A charity that helps children in the United States far more efficiently is Nurse-Family Partnership which provides an approximately three year long program of weekly nurse visits to inexperienced expectant and early mothers for at a cost of $11,200 yielding improved prenatal health, fewer childhood injuries and improved school readiness. A deeper appreciation of how little good per dollar the Make-A-Wish Foundation does relative to what is possible requires a digression.
In November 2010 the United Nations released its 2010 Human Development Report ranking the world's countries according to a "Human Development Index" based on data concerning life expectancy, education and per-capita GDP. One of the lowest ranked countries on this list is Mozambique which has an infant mortality rate around 10%. This contrasts dramatically with the infant mortality rate in the United States which is less than 1%. Every tenth pregnancy in Mozambique is followed by the grief of losing a child within several years. A child in sub-Saharan Africa who survives past the age of five is more likely than not to live a full life extending past the age of 60 .
Why is the infant mortality rate in Mozambique so high? A major cause of death is infectious disease. Around a third of infants in Mozambique do not have the opportunity to receive the standard vaccinations for polio, measles, tentanus, tuberculosis, diphtheria and other fatal diseases because of the poverty of their surroundings and some of them will die as a result.
An organization called VillageReach is working to improve Mozambique's health logistics. Between 2002 and 2008 VillageReach ran a pilot program in the Mozambique province of Cabo Delgado designed to improve the province's health logistics. This program was dramatically successful. One tangible indicator of impact is that VillageReach increased the percentage of Cabo Delgado infants who received the third and final dose of the diphtheria-tetanus-pertussis vaccine from 68.9% to 95.4%, yielding a final percentage higher than that of the average in any sub-Saharan African country. When one looks at the available evidence in juxtaposition with the cost of the program and runs through cost-effectiveness calculations one finds that under conservative assumptions VillageReach saved an infant's life for every $545 donated to VillageReach.
Now VillageReach is in the process of expanding its operations to more provinces of Mozambique, hoping to expand its pilot project into seven more of Mozambique's eleven provinces over the next six years. VillageReach requires an additional ~ $1.5 million  to implement its proposal as fast as possible. In light of the fact that VillageReach has so far received only about 20-25% of this funding, it's plausible that additional donations will have a cost-effectiveness similar to that of those used for the pilot project.
Thus we see that while a $15,134 donation to the Make-A-Wish Foundation can be expected to grant an average of one wish to an ill child (a good thing all else being equal), a donation to VillageReach can 27 infants lives! With this framing it becomes clear that the amount of good per dollar that the Make-A-Wish Foundation is doing is negligible relative to that of VillageReach . No parent would prefer to send a child to Disney World over preventing even a single one of his or her children from contracting a life threatening illness!
Nor is this phenomenon of badly suboptimal giving specific to Make-A-Wish Foundation donors. Even if one restricts one's attention to the cause of health in the developing world , many donors donate to charities pursuing health interventions in the developing world that do a thousand times less good per dollar than the most cost-effective health interventions.
A hypothetical charity running programs like VillageReach's which embezzled 95% of its budget and had correspondingly greatly reduced cost-effectiveness would still be doing far more good per dollar than the Make-A-Wish Foundation or the least effective developing world charities do. This example makes it clear how profoundly useless the overhead ratio is for assessing the relative quality of a charity.
Holding Charities Accountable
Donors should be aware that charities frequently cite misleading cost-effectiveness figures in their promotional materials. And just because a charity claims to be performing activities of very high value doesn't mean that the charity is performing the activity as reported. William Easterly recently commented on Peter Singer's child in a pond metaphor  saying:
In our situation trying to help a poor person, what we're actually doing is we're not physically able to rush in ourselves and save the child. In fact, we are not even able to observe whether the child is saved or not. What we are doing is we're sending money off to someone else on the other side of the world...and we're counting on them to save the child. And so I guess to put the metaphor another way, if your person who was saving a child was in a situation where they were physically unable to help and they knew they had to delegate it to someone else, then it would also be morally reprehensible if they did not find a person who was reliable who they were sure was going to save the child. And it would be morally reprehensible if they did not in fact check up to make sure that the child was saved. That would be just as morally objectionable as your situation of yourself directly failing to rush to the aid of the child.
Of course, for a donor with limited time and energy it is frequently not possible to personally check that a charity is performing its stated function. As such, it is useful to have independent charity evaluators that evaluate charities for impact. The only such organization that I'm familiar with is GiveWell which has reviewed 409 charities working in the areas of equality of opportunity in the United States, health in the developing world, and economic empowerment in the developing world and has highlighted those charities with the strongest evidence of positive impact. VillageReach is currently GiveWell's top ranked charity in the cause of health in the developing world.
There are many causes that GiveWell has not yet covered and there may be charities working in them that absorb donations substantially more cost-effectively than VillageReach does. GiveWell has prepared a Do-it-Yourself Charity Evaluation Guide as an aid to donors who are interested in personally investigating charities working in causes that GiveWell has not yet covered.
Volunteering, Nonprofit Work and Cost-Effectiveness
So far I've restricted my discussion to charitable giving. Giving is not the only philanthropic activity that people engage in; some people volunteer their time to benefit others and some people choose to forgo income to work at a lower paying nonprofit job that they deem to have greater social value than the job that they would otherwise take. There are many instances in which such philanthropic activities are the best way to help people, but one should consider such activities against the backdrop of there being huge variability in the cost-effectiveness of philanthropic activities. GiveWell's recommended charities have set a concrete minimal standard for optimizing cost-effectiveness of philanthropic activities.
To determine whether or not volunteering or taking a nonprofit job is a good way of helping people, one should compare additional positive impact that one would have by switching jobs with the positive impact that one would have by donating all of one's forgone income to the most efficient charity that one can find. For those with low earning potential and skills that are useful and rare in the philanthropic world, the most efficient way of helping people will typically be volunteering and/or non-profit work. For those who have high earning potential and lack skills that are especially rare in the philanthropic world the most efficient way of helping people will typically be taking a high paying job and donating one's income to an efficient charity. 
Of course, many people who volunteer or forgo income to work at a non-profit do so not only with a view toward helping people but also because they want to experience the visceral sense of helping people directly or of working directly on a cause that they feel passionate about. This latter factor can be a good reason to engage in such activities. Humans are not automatons capable of persistently adopting the most efficient course possible. Fulfilling our own very substantial personal needs and desires is important to maintaining good health and energy. At the same time, in view of the great variability of cost-effectiveness of various philanthropic activities, if one doesn't devote some resources toward helping people as efficiently as possible, one will probably accomplish very little of one's potential capacity to make the world a better place. 
People often have good intentions and frequently fail to direct them to create the substantial positive impact that they could if they thought carefully about how to do as much good as possible. I have already mentioned GiveWell as a useful resource for donors who interested in accomplishing the most good for their dollar. Such donors may also find it useful to visit Giving What We Can which is a society whose members pledge to donate a portion of their income “to whichever organizations can most effectively use it to fight poverty in developing countries” and whose members “share advice on the most effective ways to give.” By thinking critically and making use of available resources, one can reasonably expect to be able to have a much greater positive social impact than one otherwise would be able to.
 Figures taken from a survey by Independent Sector and The Annual Report on Philanthropy for the Year 2009.
 See the section of GiveWell's review of VillageReach titled Room For More Funds?
 For an indication of the relative cost-effectiveness of health interventions in the U.S. refer to a 1995 academic journal article from titled Five-Hundred Life-Saving Interventions and Their Cost-Effectiveness.
 Alan Dawrst's essay titled Why Activists Should Consider Making Lots of Money gives more on this topic.
 Eliezer Yudkowsky's Purchase Fuzzies and Utilons Separately gives a nice discussion of this theme.