Stubborn Attachments From Behind The Veil
Context: What is the Veil of Ignorance?
The Veil of Ignorance is a famous thought experiment from philosopher John Rawls. After spending decades reading and debating about moral and political philosophy, Rawls realized that most moral and political disagreements aren’t about morals at all (X is not about Y), but rather are zero-sum elbowing matches between groups. Southerners didn’t care about political sovereignty, they just wanted to keep their slaves. Rich aristocrats don’t care about meritocracy, they just don’t want to pay taxes. Unions don’t care about worker’s rights, they just want to keep wages high for their members. Since people are unwilling or unable to extricate their particular identity from political debates, all attempts at creating a universal social contract will be mired by interest groups looking to advantage themselves and their kin.
This is where the Veil of Ignorance comes in. The Veil separates a person from their particular characteristics like race, religion, or income. This forces decision makers to consider their identity as random over the sample space of all human beings. Since everyone is faced with the same probabilistic identity, the influence of identity is cancelled out and only universal costs and benefits of the social contract remain.
Rawls’ conclusions
The text in which Rawls introduces this thought experiment is 150,000 words long and many millions more have been written in interpretation, so I will not attempt to produce a summary of his work. Instead, we will focus on a fulcrum argument that is the essential link between the Veil of Ignorance and Rawls’ preferred political conclusions: The assertion that the maximin decision rule is what decision makers would use for creating the social contract under the uncertainty of the Veil.
Maximin is game-theoretic language for making the worst case scenario as good as possible. It is opposed to other decision making strategies such as maximizing expected utility. Expected utility takes into account the values and probabilities of every possible outcome, so increasing the payout for those born into the top 1% increases expected utility, but not maximin. Both strategies have their pros and cons but maximin is extremely risk averse. Making everyone except the very minimum better off seems obviously good, but a world where the minimum has 100 utils and everyone else has 500 is no different from a world where the minimum has a 100 and everyone else has 10,000 in the eyes of maximin. Indeed, according to the maximin decision rule, our world is no better than it was thousands of years ago since there are still people living in hunter-gatherer tribes with the same low endowment of rights and resources.
Rawls uses the maximin decision rule to derive three principles of justice: The Greatest Equal Liberty Principle, The Difference Principle, and The Equal Opportunity Principle. These principles demand complete equality of rights and opportunity, and permit inequality of resource allocations only insofar as they are “to the greatest benefit of the least advantaged members of society.” Rawls uses these principles to advocate for democratic socialism, where personal property and occupational choice are in principle allowed, but are subordinate to state efforts to enforce equality.
Although Rawls admits that maximin is not usually a rational decision rule, he asserts that the singularity of the social contact decision justifies its use. The law of large numbers tells us that the average of a sample of outcomes from a random event approaches the true mean as the number of samples increases. Applied to ‘lifting the Veil’ and assigning identities, the law of large numbers implies that if everyone was randomly assigned an endowment over and over again, eventually the average of each person’s endowment would approach the expected utility calculated at the beginning. However, this assignment happens only once, and there’s no way back. To see how this changes things, imagine you are offered two choices:
1. You flip a coin. On heads you get 10 million dollars, on tails you lose 2 million. Expected value = 4 million
2. You are given 3 million dollars outright
Since you can only play once and you probably don’t have 2 million dollars to lose, most people would choose option 2, even though option 1 has a higher expected value.
Maximin seems better than expected utility in the above hypothetical but it is easy to imagine a scenario where maximin implies ridiculous choices. Option 2 could be “lose 1 million dollars” and maximin would still prefer it over option 1. Expected utility may be willing to gamble too much because of an implicit assumption of repeatability, but maximin over-corrects and is blind to important gains in welfare for those already above the minimum.
Due to the failures of both rules, rational decision makers would choose a middle ground between the two. In actual experiments with the Veil, we find that “Overwhelmingly, the chosen principle is maximizing the average income with a floor constraint: a principle which is a compromise between those proposed by Rawls and Harsanyi. It takes into account not only the position of the worst-off individual but also the potential expected gain for the rest of society.” People are unwilling to gamble with a chance of slavery or starvation, but as long as an acceptable minimum is guaranteed, expected utility makes much more sense than maximin.
This strategy of maximizing expected value with some minimum constraints is the basic conclusion of Tyler Cowen’s Stubborn Attachments. He summarizes “The only non-growth–related values that will bind practical decisions are the absolute side constraints, or the inviolable human rights. In other words…we have a relatively straightforward, exclusive (“worship no other gods”), and practicable formula of "Growth and Human Rights.”
New Conclusions: Floors Without Ceilings
The far more popular and rational decision rule of maximizing expected utility given some minimum constraints leads our veiled decision makers to very different conclusions about which principles of justice society should organize around. First, because they do not know what slice of the pie they will get, they want to maximize its size. More importantly, since they know that the size of the pie is not fixed, they want to maximize its growth so that even in the worst case scenario of getting the smallest slice, exponential economic growth will give them and their descendants high levels of absolute welfare.
This growth maximization is the most important principle of justice derived from our new decision rule for three reasons:
1. Inherent value of economic growth
2. Instrumental value of growth
3. Positive sum games as the source of moral behavior
The first reason is the most obvious. Wealth is good. It feeds people, clothes them, entertains and fulfills them. Growing wealth means feeding, clothing, entertaining, and fulfilling more people.
The second reason leads us to value economic growth even if we are value pluralists and even if the list of things that we value doesn’t include wealth at all. Greater wealth means greater access to education and the arts, more time for leisure and intellectual inquiry, and more resources in general to devote towards any end that you may desire. Applied Divinity Studies in his review of Stubborn Attachments says it best:
“The point is not merely that a rising tide lifts all boats, but that it does so exponentially, such that trying to raise any one boat in particular (pursuing more narrowly conceived ethical values) is self-defeating. Even if economic growth correlates only weakly with plural ethical values, on a sufficiently long time-horizon, it is better to make an investment with compound interest than to engage in the moral equivalent of one-off consumption.
To answer the original question, the ethical basis for Progress Studies is any reasonable ethical theory at all. You can value whatever you want–beauty, utility, hedonic happiness–and it will still make sense for you to operate under the umbrella of Progress.”
The final reason to uphold growth maximization as a guiding principle of justice from behind the Veil is an evolutionary one. In an environment where total wealth does not grow the only way to get ahead is to take from someone else. Therefore, in an environment with a fixed supply of resources, respecting the rights of others is a strictly worse strategy than taking what you can when you can. It is only in a growing economy where two people can interact cooperatively and both come out ahead. We see this truth in comparisons between human society and the societies of other mammals. Male lions brawl with one another; whoever wins devours any children that don’t carry his genetic material. The animal world has no morals. In the zero-sum savannah this is perfectly rational. Any food and territory going to them is food and territory that isn’t going to your bloodline.
Since at least a few centuries ago things have been different in human society. Growth of what were once rival clans becomes mutually beneficial because it creates more vectors for trade, a bigger market for specialized goods and services, and more chances at transformative innovation. The possibility of positive sum interactions is the foundation for all moral behavior. We have not had this possibility for long and we do not yet fully appreciate its importance. If our society ever stops growing, it is only a matter of time before we return to our deep and brutal evolutionary roots. So, even if you still adhere to Rawls’ principles of justice, only a dynamic and growing economy will produce the incentives necessary to uphold them.
Rational decision makers from behind the Veil would see that economic growth is necessary for all other moral values, and so they would uphold growth as the guiding principle of justice for their society. What form of government they would choose, which institutions they would create, and which public goods they would fund is an unanswered question too large to address here, but simply agreeing on this goal to pursue is a necessary first step that most people outside of the sphere of Progress Studies have yet to take.