The GiveWell Blog


We haven’t written much about mega-charities: extremely large international charities (budgets of $250+ million per year) carrying out a very wide range of activities, and commonly recognized as household names. We’re thinking of groups like UNICEF, Oxfam, Mercy Corps, Catholic Relief Services, Save the Children, World Vision, and CARE.

The main reason we haven’t written much about these groups is that we still know very little about them. They tend to publish a great deal of web content aimed at fundraising, but very little of interest for impact-oriented donors. On the occasions when we’ve engaged with these groups, we’ve come away with the feeling that they engage in a wide variety of activities, and we can’t get a concrete sense of (a) the specifics of the activities; (b) the organization-wide track record; (c) likely uses of additional funding. (We wrote in 2007 about our inability to put together bird’s-eye views of their activities).

Below are general impressions from our limited information on, and interactions with, these groups. Note that in preparing this post, we examined the websites of the 7 organizations named in the first paragraph, looking for whatever information we could find on specific projects (as opposed to broad characterizations of activities), results (technical writeups, not narratives), and financial information (any budget breakdown by project or program type, or revenue source – we tabulated our findings in this spreadsheet).

  • Mega-charities tend to provide only very broad-brush information on their activities, and next to no information on their results. Budget breakdowns are extremely broad (see our spreadsheet). Even the “papers” or “publications” sections of these websites tend to focus on papers giving general advice, rather than on details of past executed programs. (Examples: UNICEF, World Vision.) We have seen evaluation databases from Oxfam and CARE, and wrote about our impressions on the latter (similar to our impressions on the former) in 2009.

    CARE is, as far as we can tell, the only one of these organizations that publishes descriptions of its many specific projects, though these descriptions are still relatively broad and do not include budgets.

  • Mega-charities tend to have extremely diverse activities, with relatively small amounts spent on health and large amounts spent on disaster relief (see our spreadsheet). We’ve reviewed what we can of these groups’ disaster relief work, and in general we view it unfavorably from a transparency/accountability standpoint. The relatively low amount of resources going to health is a negative for us, due to our view that this is the most promising area for individual donors. UNICEF appears to be more health-focused than the others discussed here.
  • Mega-charities tend to get large amounts of money from governments, particularly the U.S. government. While the “percentage of support coming from the U.S. government” is not always clear and depends on how one counts cash vs. in-kind contributions, government support seems to generally be at least 20% and sometimes closer to 70% of total revenues. An exception is Oxfam; we couldn’t find a specific statement of revenue sources to address this, but in a recent conversation Oxfam representatives stated to us that Oxfam takes very limited funding from governments and no funding from the U.S. government.
  • Mega-charities appear to us to often act as “contractors” for governments and other mega-donors. Scanning CARE’s projects provides some illustration of this idea, which is largely based on informal conversations we’ve had about these organizations (as well as the significance of government funding to these organizations). While these organizations often have substantial unrestricted funding as well, it isn’t clear to us whether this funding is used to supplement big-funder contracts or to run projects at mega-charities’ discretion. One thing we generally don’t see when examining these mega-charities is any indication of an overarching strategic plan, another sign that they may see themselves as essentially contractors (Oxfam is again an exception.)

Overall, our impression is that your donation to these organizations is very hard to trace, but will likely supplement an agenda of extremely diverse programming, driven largely by governments and other very large funders. We feel that when donating to these groups, you’re unlikely to get the sort of impact-per-dollar that you can with our top charities, which focus on some of the most proven and cost-effective interventions.

One organization we haven’t mentioned in this post is Doctors Without Borders. We perceive this group as more strategic, more transparent/accountable, more focused on health, and more promising overall than the organizations discussed here, and we intend to investigate it further in the coming year. We will also likely be investigating Oxfam, which we perceive as being the most independent and strategic of the organizations listed above, largely due to its lack of reliance on government funding.


  • Eric Friedman on December 29, 2011 at 2:15 pm said:

    This post is interesting and thoughtful, but I think it is incomplete by not discussing some of the potential advantages inherent in being a mega-charity.

    For example, the post notes that mega-charities work in very diverse areas, some of which are not among the most cost effective. This should be acceptable if it is because of restricted funding that limits their ability to choose what they do. Importantly, being forced to work in diverse areas could give them broader knowledge of what works best, which could be an advantage in allocating unrestricted gifts.

    Further, if a mega-charity developed a successful innovation, its larger budget (of unrestricted funding) would allow it to scale it up faster, and its reputations would enhance its ability to spread the innovation to other organizations.

    Also, it wouldn’t surprise me if the larger organizations have an advantage in being able to hire the most experienced and capable employees to run their programs. This comment is speculative, as I’m not an expert in these types of human resource patterns.

    In addition, the mega-charities are less likely to run in to barriers due to a lack of room for more funding. Their scale and infrastructure gives them a greater ability to absorb and deploy mega-donations of, say, $10 million and greater. Although a donor of $10 million could split such a donation among smaller organizations, my understanding of GiveWell’s experience is that it would be very hard to find a lot of these really good organizations.

    There are a lot of potential advantages inherent to being a mega charity. It is a very important question to ask whether they actually exploit these advantages. I would at least hope that the mega-charities could articulate the process by which they decide how to spend unrestricted gifts. I agree that GiveWell can’t recommend an organization until it addresses these issues adequately. Maybe some of them are reading your blog and will step up to the plate.

  • Eric, I agree with the specifics of what you say, but I’m not sure I agree with the framing. In brief, each of the advantages you list seems to come inherently with a disadvantage. So I wouldn’t put it as “There are a lot of potential advantages inherent to being a mega charity. It is a very important question to ask whether they actually exploit these advantages.” – I’d put it as “Being a mega charity has structural advantages and disadvantages, and it isn’t clear whether the pros end up outweighing the cons.”

    A bit more detail:

    • I think what most of your points reflect is that megacharities are well-funded and powerful, and to me there’s a natural flipside of that, pertaining to accountability. I’d say the same advantages you listed mostly apply to governments, multilateral agencies and foundations as well (and to a greater degree). For me, much of the appeal of giving to a charity as opposed to a government/foundation/multilateral is the idea that the organization is less powerful, and can’t – for example – scale up its innovations without some sort of market test.
    • Your first and last points also reflect the idea that a small number of people at the top are responsible for a large amount of programming and funding. We aren’t actually sure to what extent this is true – we think the different projects at mega-charities may have very little practical connection to each other and very little central oversight/management – but if it is, this too comes with inherent advantages and disadvantages. Under this model, the people at the top will likely have a wider view, but they probably won’t have as deep a view (relative to the people at the top of smaller charities). (While a bednets-only organization can be run by someone who has years of experience just with bednets, the CEO of a mega-charity can’t have years of experience with each of the different kinds of programming being carried out.) The people at the top may have more options when thinking about how to spend unrestricted funds, but they also will likely have less connection to and understanding of each of these options (again relative to the people at the top of a smaller charity).

    When I think about whether a mega-charity has natural advantages over a smaller charity, the question seems analogous to asking whether it’s a good idea to combine many small charities under one central management structure (which I think is probably a reasonable way of thinking about what a mega-charity is). The advantages and disadvantages go hand in hand.

    I do believe these are interesting questions and concede that we didn’t discuss them in the above post. The intent of the post was more to give our observations (and summarize why we don’t recommend mega-charities) than to discuss structural pros and cons.

  • jascollins on October 1, 2013 at 12:05 pm said:

    Holden, I wonder whether the thinking tool of considering mega-charities as umbrellas over many smaller charities might help OPEN them up to transparency; I know that most (for example) church organizations have separate accounting for their local and their umbrella organizations – The local congregations each keep THEIR OWN records, and the parent organization only really announces percentages of the aggregate.

    This is even reflected in their giving structure. Congregations will often have a fund for general giving, but when they need something special, they might make additional requests; “Please consider also giving to the Missionary Fund for our members who are out spreading the good word,” or, “Remember also, the fund to replace the sacristy roof needs help.” I know similarly that the general fund for one congregation might be poor or flush with respect to the average or to the parent organization.

    While I have a NUMBER of problems with the way churches are traditionally run, many of them have at least BEEN running for very long times, and what works for them has been shown to work. Would it be possible for you to request in-depth info on individual projects WITHIN Oxfam or CARE? I imagine they’d be happy to help, as this might allow them to better allocate their resources.

    I realize these numbers can be rough in the extreme; having been involved on the ground with local projects, I know that even the most dedicated chapters can get sloppy numbers, especially for a service organization – The few dollars they get are frequently just plowed back into materials for performing more service – and things like people donating gas, or billboard space, or coming to a fix-up day with their own materials just don’t get measured. But at least you can get an idea whether the individual group is applying their moneys to the need or not.

    I also rather suspect that some few mega-charities would be far less happy about this sort of thing, given how different their administration/giving ratios compare to the average local group’s giving structure, and that reluctance would also be telling, and useful to know about.

    I also realize that the parent organizations’ separate business of building out the aggregate numbers is then duplicated, but frankly a separate accounting might be the best thing to make some of them more careful about their numbers.

  • Alexander on October 8, 2013 at 3:04 pm said:

    jascollins – we’ve had a pretty significant number of conversations over the years with divisions within “mega” charities, and we’ve made some progress on understanding better how they typically function. (See, for instance, our writeup on the landscape of vaccination organizations, though there are a number of other potential examples, some not public.) We haven’t done as much in terms of interfacing with individual local projects, in part because of our perception that large international charities have a reasonable interest in coordinating funding to their constituent projects, and that it would be unhelpful to try to circumvent those processes.

    More generally, trying to collect thorough data from a number of local affiliates or projects of a major international charity seems likely to be time consuming and unlikely to carry sufficiently large benefits to be worthwhile from our perspective.

Comments are closed.