A Journal of Philosophy, Applied to the Real World

Medical Crowdfunding, Political Marginalization, and Government Responsiveness: A Reply to Larry Temkin

Southern Methodist University


Larry Temkin draws on the work of Angus Deaton to argue that countries with poor governance sometimes rely on charitable giving and foreign aid in ways that enable them to avoid relying on their own citizens; this can cause them to be unresponsive to their citizens’ needs and thus prevent the long-term alleviation of poverty and other social problems. I argue that the implications of this “lack of government responsiveness argument” (or LOGRA) are both broader and narrower than they might first appear. I explore how LOGRA applies more broadly to certain types of charitable giving in developed countries, with a focus on medical crowdfunding. I then highlight how LOGRA does not apply to charitable giving aimed at alleviating the suffering of the absolutely politically marginalized, or those especially vulnerable people to whom governments are never responsive.


1. Poverty Alleviation and the Lack of Government Responsiveness Argument

In his challenging and important paper in this volume, Larry Temkin engages with economist Angus Deaton’s argument that foreign aid and other charitable giving to the neediest countries in the world unintentionally does more harm than good (Deaton 2013, Ch. 7.). We can reconstruct one of Deaton’s arguments for this claim as follows:

  1. Extreme poverty and other preventable suffering in a given country cannot be effectively alleviated in the long-term unless that country’s government is fundamentally responsive to its citizens and their needs (e.g., for healthcare, education, infrastructure, rule of law, etc.).
  2. Governments are responsive to their citizens only to the extent that they depend on citizens’ support through taxes, votes, and the like.
  3. Large amounts of aid (whether direct foreign aid from governments or charitable giving) enable a government to remain in power and attain its goals without citizen support; this undermines the government’s reliance on its citizens.
  4. Therefore, large amounts of foreign aid make governments unresponsive to citizens.
  5. Therefore, large amounts of foreign aid prevent long-term alleviation of extreme poverty and other preventable suffering.

Call this the Lack of Government Responsiveness Argument, or LOGRA. LOGRA depends on several controversial empirical claims: (1) that responsive governments are necessary for long-term poverty alleviation, (2) that governments will not be responsive to citizens unless they rely on them for taxes, etc., and (3) that foreign aid enables governments to avoid relying on citizens in these ways. For the sake of argument, assume that these claims are true in at least some cases.

Temkin points out that LOGRA may apply even to demonstrably effective giving of the sort endorsed by Effective Altruism (EA). Givers face what (following Derek Parfit) Temkin calls an Each-We Dilemma. These dilemmas are cases in which an individual can bring about the best consequences by the lights of a certain theory by doing one thing, but the overall consequences will be very bad by those same lights if others also do that same thing. An individual donation to an effective aid organization has a massive positive impact on those who are helped and a tiny negative impact on government responsiveness. If we assume that it is better for burdens to be dispersed among many who each bear only a small cost than it is for burdens to be carried by a few who each bear a large cost, then what we ought to do individually is give (so that everyone bears the miniscule burden of a government made slightly less responsive) rather than refrain from giving (so those few who would otherwise be helped bear huge burdens). But LOGRA implies that many donations to effective aid organizations together risk undermining government responsiveness in a way that inhibits long-term poverty alleviation, which is a massive burden on everyone. Collectively, then, what we all ought to do is refrain from giving.

I will argue that if Temkin is right about this, LOGRA has important implications that go beyond what he addresses in his paper. First, I explore how a version of LOGRA might apply to some forms of domestic charitable giving even in high-income countries with generally well-functioning governments. This means that Each-We Dilemmas concerning charitable giving may be more widespread than Temkin suggests. Second, I suggest that there are groups of especially marginalized people to whom LOGRA does not apply, because governments never rely on them for financial or political support, and therefore have no incentive to be responsive to them even in the best of circumstances. It follows that giving that is narrowly aimed at alleviating the suffering of these groups is not subject to the particular Each-We Dilemma raised by LOGRA.

2. Each-We Dilemmas for Local Giving in Affluent Nations

LOGRA highlights one way in which foreign aid undermines government responsiveness. I am concerned that other forms of charitable giving can similarly undermine government responsiveness. In both affluent and poverty-stricken countries, private individuals or organizations routinely fill gaps in the provision of essential goods and services that can be effectively provided to all in the long term only through state intervention. Although this meets the needs of some individuals in the short term, it risks seriously undermining the government’s ability or motivation to meet all of its people’s needs in the long term. While there are multiple examples of this, I will focus primarily on how crowdfunding for medical expenses can undermine the political will to fix a broken healthcare system. My arguments are conditional, as they depend on controversial empirical assumptions that I am unable to defend here. However, even if these particular assumptions are false, a similarly structured argument should apply in a wide range of analogous situations.

Crowdfunding typically uses web platforms (such as GoFundMe) to solicit direct donations from friends and strangers. Crowdfunding to cover healthcare expenses is widespread and growing, covering everything from cancer treatment to emergency care to experimental treatments to routine expenses for chronic illnesses. There are a number of serious ethical problems with medical crowdfunding. Among other concerns, crowdfunding seems to disproportionately and unfairly benefit those who are tech savvy, have wide social networks, are seen as deserving of help, and whose stories are media friendly; ineffectively distributes aid on the basis of sympathy and luck rather than need; and forces recipients to publicly disclose sensitive health information that they might rather keep private in order to receive funding (see Snyder 2016 and Berliner and Kenworthy 2017 for more on these and other criticisms).

Another major worry is that crowdfunding enables governments to shirk their duties. Campaigns are more frequent in areas with less robust health insurance; for example, a randomized survey of GoFundMe crowdfunding campaigns found that “a much larger proportion of campaigns than expected were based in states that chose not to adopt the Medicaid expansion under the ACA” (Berliner and Kenworthy 2017: 237). Jeremy Snyder points out that “the sites allow individuals to address their need for medical care without addressing the underlying causes of these unmet needs” (Snyder 2016: 39). This is exacerbated by the fact that crowdfunding campaigns routinely ignore structural injustice in their pleas for help. For example, a survey of Canadian campaigns found that they focused on the recipient’s personal relationships, needs, and altruistic characteristics and “almost universally did not appeal to the perceived injustice of having to resort to crowdfunding by Canadians with an existing entitlement to essential medical care, supporting the concern that medical crowdfunding can obscure systemic injustices” (Snyder et al. 2017 p.367).

If we assume for the sake of argument that the following (admittedly controversial) premises are true, we can generate a LOGRA for medical crowdfunding in affluent countries:

    1. Healthcare needs can effectively be met in the long-term only through comprehensive government provision of services.
    2. Governments will provide comprehensive healthcare services to all only if politically pressured by their citizens to do so.
    3. Citizens will politically pressure governments to act only if they perceive a pressing need.
    4. Medical crowdfunding undermines citizens’ perceived need to advocate for comprehensive government provision of healthcare.1
    5. It follows that donors in affluent countries who crowdfund the medical expenses of their friends and neighbors risk undermining the responsiveness of their government in a way that collectively leads to much worse results:

    6. Therefore, medical crowdfunding prevents citizens from politically pressuring their governments to provide comprehensive healthcare coverage.
    7. Therefore, medical crowdfunding removes incentives for the government to provide comprehensive coverage, which prevents the meeting of long-term healthcare needs.

    Snyder rightly notes that “the contribution of any one campaign to these problems is minimal, creating a strong argument that the gain to each user offsets the systemic effects of medical crowdfunding” (Snyder 2016 p.41). He goes on to suggest that “as a result, it is difficult to make the argument that those seeking access to essential medical services through crowdfunding ought not to do so” (ibid).

    But this argument moves too quickly, because individual donors to medical crowdfunding campaigns potentially face Each-We Dilemmas. Individually, donating to a campaign clearly does good (although surely not the most good you can do with your money in EA terms). Collectively, though, donating to medical crowdfunding campaigns risks undermining the only sustainable long-term solution to meeting everyone’s healthcare needs. This is not to say that we should ignore the dire appeals of our family, friends, or strangers for help with their healthcare. With Temkin, I am not ready to “sacrifice the current needy on the altar of need minimization” (Temkin 2019). But we cannot ignore the fact that medical crowdfunding might lead to counterproductive negative effects. We must consider whether our individually good actions are leading to a collectively terrible result, and continue to grapple with the ethics of Each-We Dilemmas in determining whether and how to give to medical crowdfunding campaigns.

    Even if medical crowdfunding does not in fact undermine government responsiveness in the area of healthcare, it is worth exploring whether Each-We Dilemmas of this structure arise for other kinds of giving that risk undermining government responsiveness narrowly in other areas (such as funding for scientific research or the arts). For example, consider U.S. billionaires who make major donations to support K-12 public education, such as Mark Zuckerberg’s $100 million gift to Newark, N.J. public schools,2 or the Gates Foundation’s support of public education aimed at improving outcomes for Black, Latino/a, and low-income students.3 For the sake of argument, assume the (admittedly controversial) claim that the only long-term solution for improving educational outcomes across the board for underrepresented students is state intervention (such as divorcing public school funding from property taxes and providing increased and equitable funding across geographic regions through redistributive taxation). Assume also the (again controversial) claim that well-publicized support of equitable public education initiatives by a handful of billionaires dramatically lessens Americans’ perception of the need for different tax policies, and that this makes the U.S. government less responsive to educational inequality. If these claims are true, these billionaire philanthropists face Each-We Dilemmas, and risk undermining the only feasible long-term solution to educational inequality.

    The worry that private giving might undermine the political will to solve entrenched social problems is not new. For example, J. A. Hobson wrote in 1914 that

    Every act of charity, applied to heal suffering arising from defective arrangements of society, serves to weaken the personal springs of social reform… by the softening influence it exercises on the hearts and heads of those who witness it. It substitutes the idea and the desire of individual reform for those of social reform, and so weakens the capacity for collective self-help in society (Hobson 1914 p.296).

    This echoes socialist and leftist critiques of EA which suggest that it ignores institutional factors and “lets capitalism off the hook” with its tendency to “obscure that the ordinary workings of capitalist markets create and exacerbate poverty” (Gomberg 2002 p.55).4 LOGRA points out a different way in which charitable giving risks undermining institutional effectiveness: not by supporting an exploitative capitalist system, but by preventing capitalist governments from functioning as well as they could.

    3. LOGRA and Absolute Political Marginalization

    In making his case for LOGRA, Deaton writes, “the need to raise funds exists everywhere, and it will often constrain the ruler to pay attention to the demands of at least some of the population” (Deaton 2013 p.295, [my emphasis]). However, neither Deaton nor Temkin pays sufficiently close attention to the fact that governments are responsive only to some of the people residing in their countries. Every government is most responsive to certain constituents (e.g., their wealthy political donors). But there are some groups to whom governments are not responsive at all. Call them the absolutely politically marginalized. Because they are ineligible to vote and/or do not contribute tax dollars or other material support to the state, governments that are focused on self-preservation will have no direct incentive to be responsive to them in any circumstances, even without any financial bolstering from foreign aid. They may have indirect incentives if foreign allies put politically pressure on them, or if politically powerful people mobilize on their behalf. But these indirect incentives will likely not lead to the same degree of responsiveness as would direct incentives.

    Different groups of people are absolutely politically marginalized in different societies, including (but not limited to): undocumented immigrants, refugees, and those denied citizenship (e.g., the Rohingya in Burma); felons in jurisdictions with felony disenfranchisement; people who do not pay taxes and systematically lack political power (e.g., the chronically homeless and unemployed); and people who are enslaved. Non-human animals are absolutely politically marginalized in an even more extreme way: they have few to no legal rights, and are incapable of directly giving financial or political support to the government. Even if LOGRA succeeds in establishing that foreign aid has serious negative consequences for a country as a whole, it does not follow that aid that is narrowly aimed at alleviating the suffering of the absolutely politically marginalized has the same negative consequences for those marginalized people. For we must consider what would have happened had the aid not been given. Most people will be worse off if an otherwise responsive government becomes unresponsive as a result of aid. But the absolutely politically marginalized cannot be made worse off in this way, since the government is not otherwise responsive to them. And so even large amounts of narrowly targeted aid will not harm them, which means that donors who provide such aid do not seem to face Each-We Dilemmas.

    Temkin suggests that one way to avoid LOGRA is focusing aid not on low-income countries with poor governance, but on poor people in middle-income countries with decent governance, such as China and India. We should also consider focusing our aid efforts on supporting the absolutely politically marginalized in any country. And since it is likely that they will remain in the margins unless their governments become more responsive to them, we must also think more carefully about the value of engaging in political action to encourage governments to become more responsive to the absolutely marginalized. This usually happens only insofar as voters and taxpayers advocate on their behalf: undocumented immigrants protesting their own poor treatment will not motivate a self-interested government to change, but politically powerful people protesting this same poor treatment might. Non-human animals are incapable of advocating for themselves, but the activism of their human supporters has led to major gains in animal welfare laws.

    However, we must be cautious that private aid to absolutely marginalized groups does not unintentionally undermine the political will to pressure the government to become responsive to these groups. For if aid groups step in where governments fail in a way that prevents voters from perceiving the dire needs of the absolutely politically marginalized and pressuring their governments in light of this, we risk another version of LOGRA, in which giving to the absolutely marginalized ensures their ongoing marginalization. Ultimately, the implications of LOGRA are potentially both broader (ruling out medical crowdfunding and perhaps other forms of charitable giving in the developed world) and narrower (ruling in giving to the absolutely politically marginalized, unless this itself makes governments less responsive) than it may first appear.


    Thanks to Brian Berkey, Josh Crabill, Alex Dietz, and Jeremy Snyder for helpful feedback on this paper.


    Berkey, B. (2018) “The Institutional Critique of Effective Altruism.” Utilitas 30(2): 143-171.

    Berliner, L. and Kenworthy, N. (2017) “Producing a Worthy Illness: Personal Crowdfunding Amidst Financial Crisis.” Social Science and Medicine 187: 233–243.

    Deaton, A. (2013) The Great Escape: Health, Wealth, and the Origins of Inequality. Princeton: Princeton University Press.

    Gomberg, P. (2002) “The Fallacy of Philanthropy.” Canadian Journal of Philosophy 32(1): 29-66.

    Hobson, J. A. (1914) Work and Wealth: A Human Valuation. New York: MacMillan.

    Snyder, J. (2016) “Crowdfunding For Medical Care: Ethical Issues in an Emerging Health Care Funding Practice.” Hastings Center Report 46(6): 36-42.

    Snyder, J., V. Crooks, A. Mathers, and P. Chow-White. (2017) “Appealing to the Crowd: Ethical Justifications in Canadian Medical Crowdfunding Campaigns.” Journal of Medical Ethics 43: 364-367.

    Temkin, L. (2019) “Being Good in a World of Need: Some Empirical Worries and an Uncomfortable Philosophical Possibility.” Journal of Practical Ethics. 7:1


    1.If crowdfunding primarily benefits those who are most skilled at advocating for themselves, it risks undermining the perceived need to advocate for government provision of services among those who are best situated to do this sort of advocacy in particular.

    2.The impact of Zuckerberg’s gift has been controversial; see https://www.wsj.com/articles/newarks-100-million-education-debate-1441752228 [Accessed 22/5/19]. Thanks to Alex Dietz for suggesting this example.

    4.For further discussion of institutional critiques of EA (and an argument that the only plausible versions of these critiques are consistent with EA principles), see Berkey 2018.