Stubborn Attachments
Page 8
The attitude of historical pessimism is therefore one of the most important critiques of my arguments. If historical pessimism holds true, as is suggested by many old-school conservatives, expected rates of return are negative, there are no long-lasting Crusonia plants, and my arguments, even if they hold up logically, do not apply. Historical pessimism is therefore much more than a mood or an attitude; if true, it would shape our substantive views and our practical choices significantly. But for the purposes of Crusonia plant arguments, positive economic growth need only be possible with some probability; this is an argument whose underlying structure recurs throughout these chapters. That positive probability means that the growth calculus will dominate our estimates of the expected returns from different choices. We can therefore reject the final practical stance of the historical pessimists, even as we recognize that they often get the better of the optimists in the substantive arguments about the future course of our world.
The current political opinions of social scientists do not always match up with these conclusions. Many advocates of greater state spending—especially non-economists—seem to like the idea of a very low discount rate. Many of these individuals would like our government to devote more resources to education, to infrastructure, and to improving the environment, all positions associated with the political left overall, at least in the United States. They see a lower discount rate as supporting all of these policies. Yet they also tend to favor redistribution, even when such policies conflict with economic growth. In this sense the political left does not have a consistent attitude toward the importance of the future. Many thinkers on the right suffer from the opposite inconsistency. They often favor market-based discount rates, which are relatively high, but when the topic is redistribution, they worry much more about the longer-term consequences.6
In contrast, I see a deep concern for the distant future as cutting across the political spectrum. A greater orientation toward the future is likely to increase the desirability of policies favoring a market economy, economic growth, and technological innovation. Furthermore, some of the arguments for these institutions may in fact require a deep concern for the more distant future. For instance, positive rates of discount usually imply that we should grant considerable importance to the alleviation of immediate suffering. Market liberalizations, whatever long-run virtues they may have, sometimes increase immediate suffering because they require resource reallocations, namely that many workers must try to find new jobs. Furthermore, market economies often invest their surpluses in long-run growth, rather than redistributing funds to the immediately suffering poor.
Market economies and market reforms look better as more weight is placed on the relatively distant future. A free society is better today than a corrupt and totalitarian one. But one hundred years from now, the difference in human welfare and other relevant values will prove far more pronounced. Over time the United States gained ground on the Soviet Union, rather than allowing convergence.
Who should sacrifice, and when?
Under more normal circumstances, with a long time horizon, a utilitarian or consequentialist framework may still recommend that some individuals sacrifice significant parts of their lives, or risk such sacrifices, for the greater social good. To cite a simple example, Martin Luther King Jr. brought much good to the world with respect to both justice and long-term economic growth. It would be fair to say that King did the right thing in choosing to pursue higher ideals rather than playing golf all day, even though he lost his life in doing so. The same can be said of Gandhi. Nonetheless, such obligations to sacrifice cannot be universal or near-universal. If we all went around sacrificing our own individualistic pursuits to an extreme degree, there would be no civilization left to advance. As we saw earlier, it is more sensible to reject collective sacrificial recommendations that will lower the rate of sustainable economic growth.
In many cases our obligations should be viewed at a collective level. This framework does not pin down a uniquely correct course of action for each individual, so it’s not morally clear which individual is obliged to make the sacrifice. What if there were an innocent girl drowning in a lake, and any one of us could jump in and save her? In these cases the question, “What should I do?” allows for considerable latitude, and the scope of my individual obligation, as a group member, may be indeterminate. It’s good if I risk my safety and jump in after her, but it’s also fine if someone else does it instead. Similarly, it doesn’t have to be my group that protests government injustice, because many other groups could do this as well. This becomes a problem of game theory, and as we know from game theory, the implied obligation of a single individual or group is very often indeterminate.
Imagine a game with payoffs, so that it is better if someone makes a sacrifice to achieve a socially valuable end, but it is worse if everyone sacrifices or tries to sacrifice to achieve that end. The structure of this problem is common to many questions of morality and individual obligation, including the problem of global poverty. Some people should make sacrifices to help out, but, because we must keep our economically advanced civilization up and running, not everyone should make such sacrifices. Arguably this is the paradigmatic payoff structure with which to address questions related to global poverty, sacrifice, and obligation.
A utilitarian standard, in its simplest form, would suggest that the “least cost supplier” should make such a sacrifice. That moves us a little bit closer to common sense morality, namely by stipulating that the sacrificers should be the people most inclined to do so. This would include saints, moral saints, and dedicated agents of social change, as well as individuals who, for whatever reason, don’t find the required sacrifices to be so very daunting.
If several potential sacrificers face the same cost and can produce the same expected social benefits from their sacrifice, then it can be said that one of them (or, more broadly, some subset of them) should sacrifice. In these cases it is morally indeterminate who should sacrifice and who should not, or who should move to Malawi to administer vaccinations to children. This is less oppressive, and less antagonistic to individual life plans, than the typical utilitarian scenarios involving enslaved Western doctors toiling away to suit purposes other than their own.
When confronted with global poverty or other forms of calamity, many (or most) people express something along the lines of, “Well, someone ought to do something.” At the same time, they do not feel strongly that this “someone” ought to be themselves. Such intuitions may seem incoherent or perhaps even selfishly irresponsible. But they match up to one of the realities of this game. Yes, someone ought to do something, but the game itself does not make clear who that someone is. The result is an under-provision of the public good and that many people continue to make the assumption, largely for selfish reasons, that this “someone” is going to be someone else. My point is not that the observed response is socially optimal, only that the best solution to the game corresponds to some underlying features of common sense morality.7
In any case, individuals are more likely to sacrifice too little than too much, and too few individuals are willing to sacrifice much at all. So we can look to a specific recommendation of another kind, even if it cannot always help us calculate which individual should perform a given sacrifice. We should strengthen our consciences, as well as social norms, to increase the probability that the appropriate individuals would be willing to make a needed sacrifice if it turned out to be best, all things considered, that they be the one to step forward. We ought to honor and reward such sacrifices more in order to increase their likelihood. Again, this is not so far from common sense morality.
Should money be redistributed to the rich?
In many cases, purely utilitarian prescriptions will have morally counterintuitive implications, but ones that run counter to the usual concerns about enslaved American doctors serving poor Africans. Namely, utilitarianism may support the transfer of resources from the
poor to the rich. A talented entrepreneur, for instance, can probably earn a higher rate of return on invested resources than can a disabled great-grandmother. Indeed, a common complaint in the literature on inequality is that the rich get richer while the poor get poorer, or at least more or less stay put. If this portrait is to be believed, then the rich earn higher returns on their accumulated wealth, as has been argued by the French economist Thomas Piketty. If we combine the trickle-down effect from the wealth of the wealthy with a zero rate of discount, it is easy to generate scenarios in which utilitarianism would recommend the redistribution of wealth to the wealthy.
For instance, let’s assume, for the sake of argument, that the wealthy earn eight percent on their holdings, annually and on average, while the poor earn one percent. If one-fifth of the gains to the wealthy trickle down to the poor over time, then the poor are better off if the wealthy command more resources. They will receive one-fifth of the eight percent, or 1.6 percent, rather than the one percent they would earn on their own. Usually this trickle won’t reach them right away, but over time the rich will build more factories, buy more products, hire more domestic workers, fund more research and development, push for more immigration, and so on. Sooner or later, many among the poor will benefit. If we have a deep concern for the distant future, it matters less if most of these benefits come later.
The implications of this redistribution to the rich will be anti-egalitarian at first, but over a sufficiently long time horizon the poor will benefit increasingly from the high rate of economic growth. The results need not be anti-egalitarian if we consider the appropriate stretch of time, but they are sure to appear anti-egalitarian by many common metrics, which focus only on the short run or on a single country rather than adopting an appropriately longer, broader, and more global perspective.
I am not suggesting that a good pluralist theory will endorse major systematic redistribution of wealth to the wealthy or the talented. For one thing, this may be one case in which a rights constraint limits the core recommendation of growth maximization. Maybe it’s just not right for a hedge fund manager to seize resources from a bricklayer, no matter how good an investor the manager may be. An alternative concern, one that would also limit redistribution to the wealthy, is that a sufficiently unequal distribution of wealth may lead to lower growth through a number of the channels discussed above.
What’s important, however, is how this reframing shifts the burden of proof by examining the implications of a very low discount rate. Direct, short-term redistribution to today’s poor is no longer the default option for a moral theory that emphasizes individual well-being. Instead, in many cases utilitarianism has to work to avoid the conclusion of redistributing more resources to the wealthy. Once again, it is possible to have a moral theory which focuses on good consequences without requiring everyone to give up eighty percent of their income or to work as a doctor in Africa.8
Finally, we need to think carefully about where the most significant gains of the past have come from, and we should emphasize the extension of those gains rather than redistribution per se. Arguably the most important gifts of the past generation to the current generation come from wise investments, a belief in rules of just conduct, good political institutions, and good values, among other related historical factors. Growth-enhancing institutions do require hard work, but that investment is a positive-sum rather than a zero-sum game across the generations. The resulting moral impulse is one of strengthening good rules conducive to future economic growth, properly understood, and here again we approach a common sense morality.
Our obligations to the elderly
Given the limits on our obligations to the poor, we will have comparable limits on our obligations to the elderly. In fact, we can think of the elderly as individuals who are poor in one particular dimension, namely in their future human capital. The elderly are more likely to die soon than are the young. And while we should do a good deal to help the elderly, the logic of sustainable growth places limits on these obligations, too.
There is a more general question here about how much a consequentialist or utilitarian standard should value the lives of the elderly. I recall a job interview in 1986 in which I was asked by my interviewer, the economist Julius Margolis, “Why don’t we value human lives at replacement cost?” I was caught off guard and didn’t have a good answer for him. Yet his challenge sounded so wrong to me. Don’t the elderly deserve more respect than that? Do they not experience special lives, the value of which cannot be captured by metaphors of replacement? I’ve been thinking about that question for a long time because it challenges a lot of our moral intuitions.
Let’s start with a simple example. If a house is worth $1 million in the marketplace but it can be built anew at a replacement cost of $500,000, then the correct value of the house is $500,000, at least provided we actually take action to replace it. To make that more concrete, we should not spend $800,000 to save the house from destruction when we can replace it with a perfect replica at a cost of $500,000.
So let’s say a human life is worth $4 million, as estimated by standard economic willingness to pay to reduce risk measures, but we can create another human life for about $10,000, say by subsidizing births or by saving another human life in a more economical manner elsewhere. Birth subsidies are probably going to be cheaper than spending $4 million to save a life. So how much should we spend to save or preserve that first human life? Should we spend $4 million? Or is this human life worth only $10,000—in other words, its replacement cost?
Virtually all of us would agree that $10,000—it could be even less—is not, in general, an adequate valuation of a life, including the life of an elderly person. For one thing, the life we would lose and the new life we would create would not be identical. So we would not, strictly speaking, be replacing the life that has been lost. We may also feel a special obligation toward older individuals by virtue of their roles in raising us, building our nation, and defending us in earlier wars. Furthermore, failing to save the first life and investing in helping or creating another life are not usually causally related. Letting an elderly person die rather than spending money to save their life does not automatically activate higher birth subsidies or other life-saving measures elsewhere in the economy.
So no, it is unlikely that measures of replacement cost are the correct way to value a human life. Still, replaceability should not be completely irrelevant to how we think about the value of a life. If one life disappears and another is added, the new life does make up for some of the value lost, at least in utility terms. Argued another way, losing an irreplaceable civilization is a much greater tragedy than losing a civilization in a way which allows for the birth of a new and different one in its place. Replaceability therefore seems to count for something, even if we do not agree for how much.
More practically, the additional wealth that accumulates as a result of economizing on life-saving expenditures does lead people to buy safer cars, to take less risky jobs, and so on. So it can be argued that we will save some number of other lives by investing less in direct life preservation for the elderly. We don’t know whether an increment of wealth saved will in fact rescue or preserve other human lives, but there is some chance that this might be the case. And this possibility lowers the value of spending a lot of money to extend a human life. So, in a contemporary setting, a human life should probably be valued at less than $4 million, or whatever other sum the willingness to pay method, or some other utilitarian calculation, is going to serve up.
We may not know the exact correct valuation of an individual life, but we do know that the possibility of commensurability, the pull of the more distant future, the ongoing replenishment of human civilization, and the value of investing in future lives, when considered as a whole, exert some downward pressure on how much we should invest in extending the lives of the elderly today. My arguments therefore suggest a lower estimate of the value of a life, including
an older life, than most other plausible frameworks, because replacement and replenishment of the civilizational flow are considered as one factor among many. Replacement and replenishment should not be taken as the final word, but yes, they do exert downward pressure on our value of life estimates.
To put it more concretely, today in the United States we are spending too much on the elderly and not enough on the young. And given that the elderly are the ones who vote with greatest frequency, and the young often do not or cannot, that mistake should hardly come as a surprise. Governments should focus on investment, but in the United States, at least, government spending on investment is falling; in recent years, government investment has fallen below its postwar average of about five percent. Similar patterns can be observed in many European national budgets.9 Unfortunately, when government spending needs to be limited or cut, investment is often the first area to go, while entitlements for the elderly remain intact. In this regard, I am suggesting some significant revisions to current trends.
Once-and-for-all changes vs. growth rate changes
Now let’s look at one final building block for deciding the appropriate scope of redistribution, namely the nature of economic growth. At a very general level, beneficial policies fall into one of two categories. First, such policies may yield some benefit in a once-and-for-all fashion; imagine increasing the power of all light bulbs for one year. Second, the new benefits may be ongoing and self-augmenting; imagine scientific policies that speed up the rate at which better light bulbs (or other innovations) are developed. Such policies would permanently increase the rate of economic growth; in other words, they would count as a Crusonia plant, a self-generating and self-refreshing source of ongoing value.