The United Nations Development Program (2005) gave the following statistics for 2003.
Japan has a population of 127.7 million people and a per capita income of $27 967. Average life expectancy is 82 years, highest in the world.
Earth has a population of 6.3 billion people and a per capita income of $8229. Average life expectancy is 67.1 years.
Yemen has a population of 19.7 million people and a per capita income of $889. Average life expectancy is 60.6 years.
Zambia has a population of 11.3 million people and a per capita income of $877. Average life expectancy is 37.5, fifth lowest in the world.
International inequalities in life expectancy are simply staggering. Ten countries, all in sub-Saharan Africa, have average life expectancies at birth that are, like Zambia’s, 25 years or more below the global average (and 40 years or more below that for Japan). In 33 countries, all but two in sub-Saharan Africa, life expectancy at birth is 15 years or more below the global average. Intuitively, these inequalities in basic life prospects seem plainly unjust. But can this intuitive conviction be vindicated by an argument? If present international inequalities in life expectancy are unjust, what obligations do rich nations – or their individual citizens – have to remedy the injustice? This chapter offers answers to both questions.
For simplicity, we can divide competing views of the obligations owed to the global poor into three categories: the global rich owe them a lot, a little, or nothing. The traditional utilitarian answer is that the global rich owe a lot. In particular, they (i.e., we) are obligated to give as much of our personal income to the global poor as is required to bring them up to our level of well-being. Peter Singer (1972), the Australian philosopher, famously argued for this answer on the basis of what is, in effect, a Good Samaritan principle: if one can prevent a very grave harm from befalling another person, at no more than a trivial cost to oneself, then one is obligated to prevent the harm, that is, to help the other person. However, many people reject the utilitarian answer because the obligation it imposes is “too demanding,” and Singer’s argument for it is controversial, not least because it depends on a rather controversial interpretation of which costs are properly discounted as “trivial.”
Despite these controversies, the Good Samaritan principle itself remains eminently plausible. This suggests a simple argument for the second kind of view, on which the global rich at least owe the global poor a little. Let us consider a more specific version of this answer: the richest nations minimally have an obligation to transfer 1% of their gross domestic product (GDP) to the poorest nations. For added concreteness, imagine this as an obligation incumbent on the “major seven” (G7) countries of the Organisation for Economic Co-operation and Development (OECD). In that case, for 2004, we are considering an obligation to transfer some $241.5 billion (OECD, 2005, p. 13). By contrast, in 2004, official development assistance (ODA) from the G7 was a mere 0.22% of GDP or $56.686 billion (OECD, 2005, p. 65). So even a 1% transfer would represent a clear improvement over the status quo.
From the standpoint of a rich nation (or individual), 1% of annual income is a trivial cost. I take it that this holds true on a completely straightforward and uncontroversial interpretation of “trivial cost.” Hence, according to the Good Samaritan principle, if transferring 1% of GDP can prevent very grave harms from befalling the inhabitants of poor countries, the G7 has an obligation to make the transfer. Unlike the utilitarian obligation, a 1% obligation cannot be rejected as “too demanding.” If that is right, then it seems there is little to be said – at least, not at the level of principle – in favor of the third view, on which the global poor are owed nothing.
But is 1% enough to prevent very grave harms from befalling (very many of) the global poor? To invoke the Good Samaritan principle, we must be able to supply an affirmative answer on this point. Perhaps any obligation that imposes only a trivial burden on the rich would likewise produce only a trivial benefit for the poor. Fortunately, this is decidedly not the case. To see how much good 1% of G7 GDP might do, as well as to bring out the connection between global health and global justice, let us examine what might reasonably be expected from spending the 1% on improving the health of the globally worst off.
We should begin by reviewing the fundamental determinants of health in developing countries, which is what efforts to improve the health of the globally worst off would need to target.
As this brief review suggests, to improve the health of the globally worst off, resource transfers should be targeted at (i) primary health care and public health, (ii) basic nutrition and income support, and (iii) education (especially for girls and women). If we allocated 0.25% of G7 GDP to each of these fundamental determinants, that would still leave 0.25% to cover existing development commitments (presently, recall, at 0.22%). A transfer of 0.75% of GDP from the G7 would fund a per capita package of $144 for 1.26 billion people, which covers the world’s poorest quintile.
Suppose, then, that $144 were spent annually per capita on the three fundamental determinants of health in jurisdictions where life expectancy is 15 years or more below the global average: this includes not only many countries of sub-Saharan Africa but also the worst-off Indian states and Chinese provinces (Gwatkin et al., 1999). What kind of improvement in life expectancy might one reasonably expect? To judge by the historical record, the answer is “a very significant improvement.” I have in mind the experience of those developing countries that have achieved exceptional life expectancy despite a very low GDP. Among “open societies,” they include Sri Lanka (life expectancy, 71 years), Kerala (71 years; an Indian state, but with a population of 30 million), and Costa Rica (77 years). Among “closed” societies, they include China (71 years), Cuba (77 years), and Vietnam (71 years). In all of these societies, life expectancy is notably higher than the global average (67.1), and this was achieved precisely by following the path of concerted investment in (i) primary healthcare and public health; (ii) the provision of a nutritional floor; and (iii) basic education, including for girls (and, thus, a high degree of literacy among women) (Caldwell, 1986; Mehrotra and Jolly, 1997). As Caldwell concluded (1986, p. 209): “These findings … show that low mortality is indeed within the reach of all countries.”
Equally important for our purposes is the fact that the absolute cost of following this path to higher life expectancy is quite low. In fact, 0.75% of G7 GDP is enough to fund the described per capita package for the world’s poorest quintile at levels comparable to those actually employed by the high-achieving developing countries. That is partly because $144 per capita is a real dollar figure, whereas cross-national comparisons should be made in purchasing power parity (PPP) equivalents. Since the relevant PPP multiplier can be conservatively set at three, $144 (PPP) can be spent per capita on each of the three fundamental determinants. By way of comparison, in Sri Lanka for example, total health expenditure in 2002 was $131 (PPP) per capita (UN Development Program, 2005) and public educational expenditure in 1995–7 was $100 (PPP) per capita (UN Development Program, 2001) (recently, it has been less).
Now this is not to claim that any developing country that manages to spend $432 (PPP) annually per capita on the fundamental determinants of health will succeed in lifting the life expectancy of its inhabitants above (or even, to) the global average. However, the evidence does make it reasonable to expect substantial progress in that direction. To fix ideas, let us say that life expectancy in the worst-off jurisdictions could thereby be raised 10 years. Since a 10 year loss in life expectancy certainly qualifies as a “very grave harm,” the opportunity to avert this outcome at a trivial cost, therefore, satisfies the terms of a Good Samaritan obligation. On this basis, we may conclude that the G7 are obligated to transfer 1% of their GDP to the world’s worst-off jurisdictions.
If the global rich owe the global poor at least a little, it follows that they do not owe them nothing. But it does not follow that the global rich do not owe more than just a little. It may still be, for example, that they owe 5 or 10% of GDP, rather than simply 1%. On what basis can we say that the global rich do not owe more than 1%? As far as the ideal theory of justice is concerned, we cannot say it – how much more, if anything, the rich owe remains an open question.
Yet if we adopt the perspective of non-ideal theory, things are different. In non-ideal theory, the aim is not to settle the ideal requirements of justice, finally and completely. Rather, it is (among other things) to define interim targets for practical action, toward which progress can be made before a complete ideal has been settled in theory. Non-ideal theory proceeds here by anticipating the minimum requirements that any plausible and complete ideal theory of justice will include, and it does so by identifying a common core of requirements on which plausible contending ideal theories agree. Hence, for practical purposes, non-ideal theory simply sets the question of what else justice may require aside. Theoretical disagreements about whether the G7 owe more than 1% of GDP need not, and should not, obstruct efforts to act on the 1% obligation in the here and now.
In fact, for similar reasons, we need not regard the Good Samaritan argument as the decisive basis of the obligation either. Agreement on the 1% obligation can be secured among a significant coalition of rival moral theories. Each can endorse the obligation for its own reasons. Arguably, this coalition includes utilitarians, global egalitarians and prioritarians of various kinds, decent humanitarians, as well as many decent ordinary people. More specifically, potential members include such diverse theorists as Singer himself (2002, p. 192) and Thomas Pogge (2002, Ch. 8), both of whom explicitly endorse a 1% minimum, and, outside of philosophy, (economist) Jeffrey Sachs (2005, Ch. 15) and (musician–activist) Bono, who endorse the UN’s Pearson target of 0.7% of GDP.
Moreover, this coalition can be made broader still. For example, Rawls (1999, pp. 114–18) rejected the idea of permanent obligations of international distributive justice. Nevertheless, he did acknowledge certain obligations of transitional justice, which aim to assist “burdened societies” to become “well-ordered,” in part through meeting the basic needs of their inhabitants. Transitional obligations are distinguished from permanent ones by their built-in “cut-off point.” To bring Rawls and his followers on board, then, it suffices to add a suitable cut-off point to the 1% obligation. Let us say that the obligation cuts off when no country – better still, or Indian state or Chinese province – has an average life expectancy of 10 years or more below the global average. When that point has been reached, the question of whether the G7’s obligation to transfer 1% of GDP annually to the worst-off jurisdictions is a permanent obligation or not will acquire practical purchase. But until then, non-ideal theory can safely ignore it.
Likewise, libertarians may reject the idea that the 1% transfer is an obligation of justice, or perhaps even that it is in any way obligatory, preferring instead to regard it as a humanitarian act of charity. Often this stance is driven as much by the conviction that humanitarian relief is always discretionary and supererogatory as it is by opposition to claims of (international) distributive justice. Now, as a philosophical matter, it is actually a mistake to think that charity is never strictly obligatory or morally mandatory (Buchanan, 1987). Again, in non-ideal theory, this does not matter so much. The crucial qualification for membership in our coalition is willingness – willingness to transfer 1% of income to improve the well-being of the globally worst off – including, if need be, willingness accompanied by insistence on the description “discretionary transfer.” Consequently, libertarian scruples need be no impediment, either to a G7 nation’s transferring 1% of its GDP to the worst-off jurisdictions or to a moral theory’s endorsing the transfer.
Since clinicians are professionally committed to improving the health of those in need, and likely sensitive to the urgent claims of the worst off, many may feel that the most pressing concerns about the 1% obligation are practical, rather than theoretical. While there is a whole level of important detail that we do not have space to address here – concerning the issue of aid effectiveness, for example, or subdivisions of responsibility between global and local actors – I do want briefly to address a pair of strategic practical objections. I thereby hope to allow the proposal at least to keep its foot in the doorway to action.
In the real world, of course, not everyone discharges his or her obligations adequately. When some actors shirk their obligations within a given distributive scheme, and so do less than their fair share, what does justice require of the remaining actors – the compliant ones, who have already done their (original) fair share? Does justice consequently require the compliant to do yet more, and to pick up the slack created by the non-compliant? Certainly not, as this would be extremely unfair (Murphy, 2000).
Accommodating this concern belongs to a different branch of non-ideal theory, which functions to calibrate the requirements of justice under conditions of partial compliance. In the present instance, we should regard the 1% obligation assigned to each G7 nation as invariant under non-compliance by other G7 nations. That is to say, if one of the G7 were to transfer 1% of its GDP annually to developing nations (at present, none does), then the non-ideal theory of justice would require nothing further of it. Its fair share is limited to 1% – even if no other G7 nation complies at all with its obligation, with the result that life expectancy remains 15 years or more below the global average in many more jurisdictions than would reasonably be expected under full compliance by the G7.
A slight twist on this point also helps to answer a second practical objection to the 1% proposal, which has to do with the prevalence of corruption and waste. Even someone who accepts, in principle, an obligation to improve the well-being of the globally worst off, and who is also prepared to act on it, may reasonably baulk if it turns out that, in fact, this 1% will simply disappear down a black hole of corruption. There is no obligation to line the pockets of the corrupt, no matter how many destitute there may be, unreached, in the background.
The objection is well-taken up to a point, but it still does not defeat the proposal. Invariance under non-compliance cuts both ways: while each G7 nation is only responsible for its fair share, it remains responsible for its fair share even if none of the others comply. Let me illustrate and then explain. I shall take Italy as my example, since in 2004 Italian ODA (0.15%) was actually the furthest from 1% in the G7 (OECD, 2005, p. 64). But in 2004, 0.85% of Italian GDP (1% minus existing ODA) was $13.687 billion (OECD, 2005, p. 12). Under full compliance (by the G7), Italy would be responsible for transferring that $13.687 billion at a rate of $144 per capita. In other words, Italy’s fair share of improvements in the fundamental determinants of health among the world’s poorest quintile would cover a population of 95.05 million people.
Non-compliance by the rest of the G7 does not relieve Italy of its obligation to transfer 1% of GDP. Moreover, 1% is also small enough that no G7 nation can – and hence, Italy cannot – plausibly claim that solitary compliance will put it at any serious relative disadvantage within its peer group (i.e., the G7). If need be, then, Italy should simply go it alone and transfer $13.687 billion annually to the globally worst off at the full compliance rate of $144 per capita. To do so, Italy would have to choose a mix of jurisdictions where life expectancy is 10 years or more below the global average (our cut-off point, recall), up to a total population of 95.05 million people (e.g., roughly a seventh of the population of sub-Saharan Africa).
Transferring its 1% on this basis would enable Italy to cover the same fair share of the globally worst off as it would cover under full compliance by the G7 – neither more nor less. But in that case, corruption and waste are only relevant to Italy’s action if they are so prevalent that insufficient non-corrupt (and badly off) jurisdictions exist for Italy to reach its fair share of 95.05 million people effectively. To put it the other way round, if there are enough non-corrupt (and badly off) jurisdictions – as I believe there are – that Italy can still reach its fair share effectively, by simply avoiding corrupt jurisdictions altogether, then, corruption and waste are no impediment to Italy’s action on its 1% obligation. At worst, they are an impediment to later full compliers, which is not Italy’s problem.
Of course, Italy only serves here as an example. The analysis could be repeated, with somewhat different numbers, for whichever of the G7 cared to increase its ODA to 1% first. Since each of the G7 is in the same boat as Italy, the corruption and waste objection at least begins by being irrelevant.
According to the non-ideal theory of justice we have discussed, which of our opening cases represent examples of global distributive injustice? What are rich nations obligated to do about it?
In Yemen, life expectancy (60.6 years) is less than 10 years below the global average (67.1 years). It, therefore, exceeds the cut-off point we added to the 1% obligation, in order to bring Rawls (1999) and his followers into our coalition of the willing. This means that, in non-ideal theory, Yemen is not a case of global distributive injustice. It also means that worse-off jurisdictions (those below the cut-off point) have priority over Yemen for receiving transfers from the G7.
In Zambia, life expectancy (37.5 years) is almost 30 years below the global average (67.1), well below the cut-off point. Zambia is, therefore, a case of global distributive injustice. Any nation in the G7 that has not already expended its 1% elsewhere – that is, every G7 nation (including Japan), since none has reached 1% – can make progress in discharging its obligations by transferring $1.627 billion annually to Zambia (i.e., $144 per capita) to improve the fundamental determinants of its population’s health. Zambia is one of the countries with priority over Yemen.
This chapter condenses and simplifies material from Sreenivasan (2002 and 2007), which interested readers may wish to consult. For more on current debates about international distributive justice, see Caney (2005, Ch. 4). For more on the social determinants of health, see Marmot and Wilkinson (1999). For more on global health and development, see Mehrotra and Jolly (1997) and Leon and Walt (2001). Life expectancy figures given are for 2003 and have been taken from the 2005 Human Development Report (UN Development Program, 2005), except in the case of Kerala, where they are for 1988–91 (Krishnan, 1997).