At some point, I cannot recall just when, I was invited to write a comment on a paper by the famous American economist Edmund Phelps which he was delivering at a philosophy conference. I no longer have his paper, but here is my comment
Socialist Planning and the Life-Cycle Model of
Saving[1]
by Robert Paul Wolff
Professor
Phelps addresses the very difficult problem of intergenerational justice in
taxation. For the most part, philosophers concerned with questions of
distributive justice have confined themselves to the relatively less complex
case of a timeless present in which the members of a social and economic order
confront one another and seek to arrive, in some way, at suitable principles of
just distribution. So long as the only additions to, or subtractions from, the
community occur as a result of voluntary entries or withdrawals by adult men
and women, such principles as are fixed upon can, without special difficulty,
be extended to new members. But troubles arise as soon as we attempt to take
account of the actual human condition, with its endless overlapping life-cycle
of birth, childhood, adulthood, old age, and death.
The
temporal, generational structure of a human community poses three questions of
just distribution which have no natural analogues in the timeless case. First,
what share of the social product shall be allotted to those who are not yet
able to engage in productive work, and according to what principles shall it be
allotted? Which is to say, what shall the children get? Second, what share of
the social product shall be allotted to those who are no longer able to engage
in productive work, and according to what principles shall it be allotted?
Which is to say, what shall the old get? And third, to what extent shall any
given generation of productive workers curtail its present consumption so that
future generations will find it economically possible to engage in productive
work and hence to consume? Which is to say, what is a just rate of social
savings? Professor Phelps' paper deals with one aspect of this complex of
problems, namely the role of taxation in inducing the participants in a
capitalist economy to behave as though they were guided by principles of
intergenerational distributive justice that meet certain criteria of
acceptability. The criterion which Professor Phelps explicitly invokes is
efficiency, in the familiar form of the rule that no principle of
intergenerational distribution will be embodied in a society's fiscal policy to
which some alternative principle is Pareto-preferred. I shall suggest that,
contrary to what would appear to be his intentions, the model employed by
Professor Phelps, and the conclusions which he draws from it, have a readier
and more natural application to the central planning decisions of a socialist
society than to the fiscal decisions of the state apparatus of a capitalist
society. Understood in this way, it seems to me, they constitute an extremely
useful prolegomenon to the determination of the correct socialist principles of
intergenerational distributive justice.
Professor
Phelps employs a two-period model of a capitalist economy. Workers work in the
first period of their life-cycle, consuming part of their income in period one
and saving the rest toward their non-working old age in period two. Their
aggregate savings constitute the capital available to the new generation of
workers, who are in their first, or working period during the first generation’s
second, or old-age period. Thus, the savings policies adopted out of rational
self-interest by the first generation constitute the precondition for the
investment policies adopted by the second generation out of their rational
self-interest. The model assumes that each generation, in the aggregate,
consumes its entire first-period savings (suitably augmented by some social
rate of return) in its second period. Hence, it leaves nothing to its posterity
in the form of bequests.
Intuitively,
it is clear that consumption foregone by worker-savers in their working period
will result both in an increase in the consumption available to them in their
retirement period and also in an increase in the capital available to the next
generation of worker-savers in their working period. Starting with a given level
of aggregate saving by worker-savers in some base period, we may discover that
the utility functions of those worker-savers and of the next generation of
worker-savers are such that some different, lower rate of savings would leave
nobody worse off and someone better off. There may, however, be no rationally
self-interested reason for any individual worker-saver of the present
generation of worker-savers to shift to that lower level of saving. The aim of
Phelpsian fiscal policy is to alter the objective situation, through
appropriate taxation, so that the move to the Pareto-preferred state comes to
be the outcome of rational savings choices.
There
are several rather nice features of the two-generation life-cycle model, if we
look at it from the point of view of the three questions about
intergenerational distributive justice mentioned above. First of all, the
elimination of bequests solves at least one problem of allocation to children.
If there is no inherited wealth, then we need not trouble ourselves over the
vexing issue of the limits on, or justification for, inheritance. Secondly, the
model explicitly answers the question of what the old shall get. In the
aggregate, they will get what they paid for, in the form of savings or pension
plans. Finally, the model offers an answer to the question of a just rate of
savings--or at least it offers a partial answer, since Pareto-preference yields
only a partial ordering.
To
be sure, the model does not, in and of itself, offer any but the slenderest
guide to redistribution, which one might have thought was the heart and
soul of the issue of distributive justice (always remembering that the null
redistribution is a case of redistribution). But as Professor Phelps makes
clear in this and other papers, the model is compatible with whatever
redistributive arrangements we might wish to impose, so it is at least a first step
in the direction of a theory of intergenerational justice.
Nevertheless,
the model employed by Professor Phelps seems to me to suffer from a number of
drawbacks which make it, at one and the same time, an unsatisfactory tool for
the analysis of capitalist fiscal policy and yet a promising tool for the
analysis of socialist long-term planning.
The
first problem is that the human life-cycle consists of three generations or
periods, not two. The work period is preceded by childhood. Now the fact is
that so long as they are charged with the social task of rearing their
children, parents will seek to give those children a competitive advantage in a
capitalist society by buying them better food and clothing, better education,
specialized skill training, and anything else that will offer an edge in the
job market. Some of the consumption in both periods of a worker-saver's life,
therefore, will actually take the form of investment in human capital, with
his/her children being the "firm" in which the capital is invested.
The only way to transform Professor Phelps' model into a true no-bequests model
would be to raise the children in the society communally and deny to natural
parents (or anyone else) the right to spend extra money on the young. That may
indeed be the socially just policy (although even I, from a socialist
perspective, doubt it), but it is certainly not the policy ordinarily
associated with the principles of capitalism. In the absence of such extreme
egalitarian child-rearing procedures, the Phelps model very quickly turns into
what he refers to as a "Barro and Bailey world of dynastic families."
It
might be thought that the no-bequests assumption makes the life-cycle model
irrelevant to capitalism even without the problem of child-rearing, for inherited
wealth manifestly plays a major role in capitalism as we know it. However,
Professor Phelps could get around that problem by positing a capitalism
consisting entirely of joint-stock corporations in which worker-savers
purchased shares as part of their pension arrangements. A worker-saver could
then consume his or her savings in the second period simply by steadily cashing
in his or her portfolio. If there were a one-hundred percent tax on estates,
the result would very shortly be a system of actuarially sound life-trusts
which would produce, as an aggregate result, a net estate of zero for each
generation.
The
problem of child-rearing remains, however. To preserve the structure of the
life-cycle no-bequests model, the state would have to pursue a policy of
active, continual intervention, in order to rectify what would otherwise be a
series of de facto bequests by each generation to its children.
The
second difficulty with the Phelps model is that it has the rather peculiar
consequence of making workers the savers and capitalists the consumers in the
economy.[2]
The classical assumption on which most economic theory has been built is that
capital accumulation in a capitalist economy takes the form of investment of a
portion of profits. Most of the assumptions about capitalists' propensity to
save are based on this conception, and make sense only in terms of it. In the
life-cycle model, however, capital accumulation is essentially a by-product of
old-age savings by workers. It is actually a bit difficult to see, in this
model, who the capitalists are, or why financial institutions would consent to
lend them money for investment purposes. Once again, the model is actually more
appropriate to a socialist economy. If capitalists, as a class, are eliminated
from the model, and capital accumulation is carried on by the state, then in
some sense workers' savings will be the source of investment capital, although
there would be no particular reason to go through the complicated process of
distributing the social surplus ticketed for investment to the workers in the
form of wages and then taking it back in the form of taxes.
The
third difficulty with the Phelps model is that, in keeping with the assumptions
of a private enterprise economy, it limits state direction of the economy to
fiscal policies designed to make it self- interestedly rational for individual
worker-savers to behave in a manner that will produce the outcome independently
conceived as desirable by the authors of the fiscal legislation. Now, the experience
of recent decades very strongly suggests that it is difficult to shape economic
behavior by such indirect means. The effort, as the old saying has it, is
rather like pushing on a string. In the absence of perfect information and
consequent perfect certainty, worker-savers may persist in objectively
irrational savings behavior because of doubts about the aggregate outcome of
their, and their fellow worker-savers', choices. Professor Phelps' fiscal
planners may find themselves reduced to ineffectual jaw-boning, along the lines
of the subway and bus ads during the Eisenhower era that urged Americans to
show their faith in America by buying on credit and going into debt. In a
socialist economy, however, the target vectors of capital and consumption that
are, in Professor Phelps' model, to be the unintended outcomes of individual
choices, could become the values of objective planning functions. Through
democratic methods of collective decision-making, real worker- savers could
choose their collective future, reallocate resources to equalize the
opportunities available to the next generation, and place themselves on the
optimal growth path compatible with their collectively agreed-upon principles
of distributive justice. In such a society, economic planners would, I imagine,
have to engage in calculations very like those which Professor Phelps has shown
us how to carry out.
[1] I should like to thank my
colleague, Professor Robert Costrell of the University of Massachusetts
Economics Department, for his very considerable assistance in the preparation
of this Comment.
[2] I owe this point to Costrell, who
develops it in an unpublished paper entitled “The Destabilizing Distribution
Effect in Life-Cycle Savings.”
No comments:
Post a Comment