DISCUSSÃO - "NAÇÕES, GERAÇÕES E JUSTIÇA CLIMÁTICA", DE AXEL GOSSERIES: COMENTÁRIOS E RESPOSTA AOS CRÍTICOS

Gosseries on intergenerational savings

 

Nicholas Vrousalis*

*Assistant Professor in Political Philosophy Leiden University Institute of Political Science Wassenaarseweg 52 2333 AK. Leiden. The Netherlands.

n.vrousalis@fsw.leidenuniv.nl

 

According to John Rawls, intergenerational justice is part and parcel of the more general theory he calls ‘justice as fairness’. Justice as fairness, in its final form, involves two principles:

First Principle: Each person is to have an equal right to the most extensive total system of equal basic liberties compatible with a similar system of liberties for others.
Second Principle: Social and economic inequalities are to be arranged so that they are both (a) to the greatest benefit of the least advantaged consistent with the just savings principle, (b) attached offi and positions open to all under conditions of fair equality of opportunity. (Rawls 1999, p. 266, emphasis added)

Part of Axel Gosseries’ aim is to make sense of Rawls’ theory of intergenerational justice. In what follows, I shall first raise some doubts about the cogency of Gosseries’ interpretation of Rawls, and then argue that that interpretation is implausible in its own right.^[1]

1. Gosseries’ interpretation of Rawls’ just savings principle takes its cue from Gaspart and Gosseries (2007). On this interpretation, intergenerational justice has a dualist structure:^[2]

During the accumulation stage, each generation has to transfer more to the next one than it has inherited from the previous one, in order to reach, after a few generations, the threshold level beyond which a society is be able to be well-ordered. Once this level is reached, we enter steady state stage, where no further savings are required. (p. 4)

Gosseries’ dualism offers a structural interpretation of Rawls’ view that ‘[o]nce just institutions are firmly established and all the basic liberties effectively realised, the net accumulation asked for falls to zero. At this point society meets its duty of justice by maintaining just institutions and preserving their material base.’ (Rawls 1999, p. 255) Gosseries draws two conclusions pertaining to the steady state. The first involves a prohibition on net dissavings and the second on net savings. The first conclusion is based on the following argument:

Let us assume that the stock required to support just institutions amounts to 10 units per head and that the stock effectively reached at the current stage of history is 100 per head, mostly constituted of renewable resources. Imagine that Gx destroys part of this stock and decides to transfer to Gx+1 a stock very significantly inferior to 100 units per head, while still superior to 10 units per head. Rawls would argue that there is no violation of the demands of intergenerational justice. (p. 5)^[3]

Gosseries’ second conclusion involves a prohibition on savings:

When Rawls criticizes utilitarianism because it calls for excessive savings, he is totally right. The implication of the very same idea is that we should prohibit any generational savings because of the opportunity cost they impose on the least well off that are assumed to be members of our generation rather than the next one. In that sense, allowing our generation to continue to save once the sufficiency level has been reached is unfair. It is unfair towards the least well off within our generation. (p. 6)

I shall first question the textual basis for a Gosseries-like dualist interpretation of Rawls, and then argue that the prohibition on savings, whether or not asserted by Rawls, is implausible in its own right.^[4]

2. Gosseries’ defence of the accumulation stage draws its inspiration from the putative priority of liberty: ‘accumulation can be justified in the name of reaching a level of wealth enabling a society to minimally guarantee the protection of basic liberties.’ (Gaspart and Gosseries 2007, p. 198) In this connection, we must look more closely at what Rawls says about the relevant priority rules:

The principles of justice are to be ranked in lexical order and therefore liberty can be restricted only for the sake of liberty. There are two cases: (a) a less extensive liberty must strengthen the total system of liberty shared by all, and (b) a less than equal liberty must be acceptable to those citizens with the lesser liberty. (Rawls 1999, p. 220)

Assume there are two non-overlapping generations, G1 and G2, each consisting of one person. Take, first, condition (a). Note that, by assumption, G2 cannot share his (more advanced) institutions with G1, just as he cannot ‘share’ a cake or a hug (although G1 can, perhaps, share his cakes – but not his hugs – with G2 by leaving them on the ground for G2 to take). It follows that, while G1’s increased net saving might strengthen the ‘system of liberty’ of G1 and G2 taken together as individuals, that system is not ‘shared by all’. For G1 does not benefit from the sacrifice he makes, and the institutions he lives under do not improve at the bar of justice. Indeed, they get worse than they would have been if he had abstained from transferring resources to G2 (Gosseries acknowledges this in p. 5). Now take condition (b). It looks as if, prima facie, (b) contradicts any categorical duty^[5] of G1 to save for G2. For the reduction in G1’s liberty for the sake of G2 ‘must be acceptable’ to G1. This acceptability condition figures prominently in Rawls’ theory: ‘It is only when social circumstances do not allow the effective establishment of these basic [liberty] rights that one can concede their limitation.’ (Rawls 1999, p. 132)

Why, then, are we entitled to assume that such acceptance is likely to be forthcoming, in a way that justifies accumulation? There is clearly some tension here, for on the one hand, Rawls’ just savings principle seems to make intergenerational transfers obligatory, and on the other, Rawls’ priority rules, which are intended as constraints on that principle, make them merely permissible. It is therefore not obvious that, barring G1’s choice to save, an accumulation stage is defensible on Rawlsian grounds.^[6] This conclusion only follows on the assumption that the less well-off generation G1 suffers ‘less than equal liberty’ compared with G2 (in part because it lacks the material wherewithal to construct liberty-enhancing institutions). This assumption is implicit in Rawls (1999, p. 220 and passim). The case for accumulation is therefore undermined by Rawls’ own priority rules.

3. I turn, now, to Gosseries’ steady state, which is said to prohibit both savings and dissavings. The prohibition on savings is incompatible with Rawls’ claim, cited by Gosseries, to the effect that savings may stop once just (or decent) basic institutions have been established. (…) a society may, of course, continue to save after this point, but it is no longer a duty of justice to do so. (Rawls 2001, p. 107)

It follows that net per capita savings are permissible, but not obligatory. Gosseries infers that Rawls thereby ‘abandons’, or ‘drops’ the difference principle (p. 4). Now, it is true that Rawls says that the difference principle ‘does not hold for the question of justice between generations’ (Rawls 1999, pp. 253-4). And Gosseries rightly points out that this is too strong: the suspension of the principle of freedom of movement may, in some cases, be justified, but is not tantamount to abandonment of the principle. That Rawls does not abandon, but merely suspends, the difference principle is further vindicated by his insistence (see quote provided in the opening paragraph of this paper) that the just savings principle merely introduces a constraint on the (operation of) the difference principle.

Exegesis aside, the interesting question is whether Gosseries’ substantive conclusion is true, that is, whether steady-state net savings are impermissible. He takes such a proscription to follow from leximin, which Rawls does not associate with the steady state, but – Gosseries thinks – should have (see the quote beginning with ‘When Rawls’ in section 1). The proscription is, I think, implausible. For surely G1 should be allowed to pass on some portion of his wealth to G2, at least as long as he does so for the right reasons.^[7] A norm that makes room for such voluntary transfers is not prohibitive (a law that says: ‘don’t walk on the grass unless you want to’ is not prohibitive, in the relevant sense). Gosseries’ account of the steady state does not make room for such transfers, and is, therefore, implausibly proscriptive. Here’s a relevant example.

G1 consists of two people, A and B, and G2 of C and D. Their respective levels of well-being (in the relevant metric) in the absence of any direct intergenerational transfers are depicted in column I:

 

		I	II
G1	A	6	5
	B	5	4
G2	C	7	7
	D	5	7

 

One day A and B jointly decide to pass on to D one unit of well-being each (they do this by passing on means of production, which are readily transformable into well-being). The resulting intergenerational distribution is depicted in column II.

On the leximin view, any transfer like the one resulting in distribution II is strictly prohibited, no matter what the motivating reasons of the agents producing it, since II renders B worse off. I take this conclusion to be counterintuitive, and Gosseries does too: ‘the solution may consist in rendering one’s leximin egalitarian view responsibility-sensitive.’ (p. 6) Note that this is an important concession: it implies that steady-state savings are not prohibited.^[8] It follows, further, that ‘positive net savings first, zero thereafter’ is not the appropriate slogan for any defensible intergenerational dualism à la Gosseries. Indeed, if the arguments of sections 2 and 3 are sound, both parts of that slogan are misconceived. More importantly, perhaps, the argument of this comment suggests that discussions of intergenerational justice might benefit from less focus on purely distributive matters, and more focus on the ethical nature of the actions, motives, or relationships that such distributions engender.

 

Notas

^[1]A minor terminological point: the term ‘sufficientarianism’ has two importantly distinct senses. There is the institutional sufficientarianism of Rawls’ theory of intergenerational justice, which is sometimes associated with an ‘accumulation stage’ (see below). And then there is distributive sufficientarianism, which is the view that we should aim that people have enough of what matters. Institutional sufficientarianism does not entail distributive sufficientarianism, and vice versa. On occasion Gosseries runs the two together (see, for example, the discussion of Brundtland, p. 5) in a way that obscures the thrust of his argument.

^[2]Unless indicated otherwise, page numbers refer to Gosseries (2013).

^[3]I have been unable to find textual evidence corroborating this attribution in Rawls 1999.

^[4]The prohibition on dissavings is compatible with Rawls’ argument and has some measure of plausibility (Gosseries points out that Rawls allows for intergenerational dissavings, but only subject to attainment of stable just institutions). The prohibition entails that, ceteris paribus, if members of G1 choose to grow the population such that population(G2) > population(G1), Rawlsian justice requires that they save more in absolute terms. What we are concerned with here is per capita saving.

^[5]By which I mean: a duty not conditional on the level of freedom or well-being enjoyed by G1 relative to G2.

^[6]Gaspart and Gosseries (2007, pp. 207-8) offer an empirical argument as to why G1 would in fact be very likely to veto such transfers.

^[7]As far as I know, Rawls never denies that the provision of aid to ‘non-burdened’ nations is permissible. Any such denial would be independently implausible. And insofar as it is legitimate to extrapolate from Rawls’ theory of international justice to his theory of intergenerational justice – Gosseries thinks it is – this counts against proscribing savings.

^[8]Rawls briefly discusses gift-giving in Rawls (1999, p. 245), but not in a way that illuminates the idiosyncracies of the intergenerational dimension.