There are some interesting thoughts here, thanks for sharing. I suspect that revealed ratios will vary substantially by context, mostly because decisions also affect many other things, but partly because of a lack of consistency.
One point that is worth stressing is that the ratio for lives is likely to be different (perhaps quite different) from the ratio for dollars -- precisely because the cost of saving lives varies significantly by country.
Thanks, Owen. As I mention below to Ben, I agree that, historically, there has been no consistent ratio. Were the government to act rationally/consistently, however, I think there would be a consistent baseline ratio, and other (e.g., instrumental) considerations could then be added to that ratio to arrive at the proper policy position.
I'm also not sure I understand the point re the difference between the ratio for lives and dollars. Assume I value American lives ten times more than African lives, i.e., my America:Africa ratio is 10:1. Under those circumstances, I'd be willing to spend ten times as much to save an American life. If it costs nine times as much to save an American life as an African life, I'd give to an America-oriented charity, but if it costs eleven times as much, I'd shift to an Africa-oriented charity.
Now suppose there are nine Africans on one railroad track and one American on another; the train has to take one of the two tracks, and which one is up to me. . . . (you know where this is going.) Let me know if I'm misunderstanding your point, but I think the ratio should be same no matter whether we're talking life exchanges or dollars for aid.
My point is that there is a trade-off ratio you'd use for dollars (as in your first example, between you and your sister), and another ratio you'd use for lives, and that these will come apart.
Suppose that it costs $10,000 dollars to save a life at the margin in Country A, and $100,000 in Country B. If you valued (lives in Country A):(lives in Country B) at 1:20, then you should value (dollars in healthcare in Country A):(dollars in healthcare in Country B) at 1:2.
Actually, I think your concept of provincialism is not crass enough.
I suspect that any "concern" the US government has for the lives of non-Americans is almost entirely instrumental. The Army doesn't care about minimizing enemy civilian casualties as an end in itself--it cares about those casualties only to the extent that they influence popular and international support for the war effort. Similarly, the government cares about foreign aid mostly in order to further its political objectives. (For instance, compare the list of top US foreign aid recipients with the list of countries by PPP GDP per capita.)
If you were to calculate politicians' "provincialism ratio" based on their past actions, my guess is that you would come up with wildly inconsistent results. That's because it's not the number of casualties that matters, it's how the casualties would be perceived.
Thanks, Ben. I agree that a nation has instrumental motives that are different in kind from those an individual might have, and those motives might create the misimpression that the nation cares about foreign lives more than it actually does. Disentangling these motives would be important so we're sure to begin with an accurate baseline ratio.
I disagree that nations' concerns are entirely (noting that you say "almost entirely") instrumental, precisely because--in theory--a democratically elected government represents the electorate, and we, the electorate, are not indifferent to foreign lives. A president who had no qualms eradicating an entire nation when it is in America's strategic interests would not (I hope) be reelected.
I also agree that America does not have a consistent provincialism ratio, for many reasons. One reason is that where casualties won't come to light, the government is less likely to make sacrifices to prevent them. That, however, is simply a failure of the democratic process. In a perfect world, the entire electorate would be aware of every decision its representatives made, and their consequences. When our government touts its military accomplishments but hides those accomplishments' true costs--knowing that their constituencies would not be okay with those costs--they're stepping outside their proper role as our representatives.
I would think that a simple way to do this would be to compare international charitable donations with domestic donations.
One nitpick regarding drone strikes: Do they really weigh "the number of civilians that would likely die through the use of drones"? If the risk of harming any civilians in a specific drone strike is not negligible, and the military proceeds anyway, aren't they intentionally violating international human rights law and/or international humanitarian law? See, for example, "The Truth about the United States Drone Program".
Interesting article, thanks for writing it.
I believe this ratio goes to the heart of what it means to be American and, therefore, should be determined by the electorate. Although how that ratio is translated into foreign/military policy requires expertise, selecting the ratio does not. ... I would like to see the candidates—as well as coworkers, friends, and family—openly debate this ratio.
Reminded me a bit of Robin Hanson's Vote on Values, bet on beliefs - the electorate could vote on a provincialism ratio, and then its implementation be up to economists, etc.
Disclaimer: I am new to the EA world (although not necessarily its methods), which means I have no idea which of my half-baked ideas are worth sharing (i.e., insightful) and which are not (i.e., well-trodden ground). So consider this a test-run.
Nearly all of our actions (not just our giving patterns) are the product of our personal ratios. If someone gives me the choice between a dollar for me and two dollars for my sister, I have to determine whether my Me:Sister ratio is greater or less than 2:1. If someone gives me the choice between a dollar for my sister and 100 dollars for a nameless child in Africa, I have to determine whether my Sister:African Child ratio is greater or less than 100:1. Etc.
We discuss these personal ratios (if not quite in these terms) all the time, but I have seen less attention paid to national ratios.
It is the official policy of the American government to value American lives over non-American lives. (Were it otherwise, the concept of a nation-state would be mostly meaningless.) But it is not the policy of the American government to place no value on non-American lives, as evidenced by, among other things, its concern (if minimal) for civilian casualties in foreign countries. Accordingly, the American government has what could be termed a provincialism ratio that guides (in a considered or unconsidered manner) its foreign and military policies.
One example of this ratio in action is American drone strikes.[1] The trade-off when deciding whether to destroy a military target through the use of drones or ground troops is, in part, American soldiers’ lives for non-American civilian lives. When a drone strike destroys a school full of children, many Americans look askance at their government’s decision not to use ground troops. But of course, the other half of the decision is how many of the ground troops likely would have died in pursuit of the target.
I expect—indeed I hope—that before deciding whether to use drones or soldiers, the military provides an estimate of (i) the number of civilians that would likely die through the use of drones; and (ii) the number of American soldiers that would likely die through the use of ground troops. In other words, the military provides the decision-maker with a ratio, and the decision-maker decides whether that ratio is acceptable.
Although more difficult to determine, one could also calculate a provincialism ratio based on how much foreign aid America provides relative to those foreign programs' domestic counterparts. If the same amount of American tax dollars could save the life of one Washington DC resident or 1,000 individuals in southeast Asia, what would the American government do?
I believe this ratio goes to the heart of what it means to be American and, therefore, should be determined by the electorate. Although how that ratio is translated into foreign/military policy requires expertise, selecting the ratio does not. Indeed, we could create calculators that would roughly determine for individuals, based on their own behaviors, what their personal ratio is, and then they could learn how that ratio might translate to foreign/military policy. I would like to see the 2016 presidential candidates on a spectrum based on their provincialism ratio: how does Hillary Clinton compare to Jeb Bush? I would like to see the candidates—as well as coworkers, friends, and family—openly debate this ratio.
The very concept of a provincialism ratio will strike many as crass, and it is. But it exists, and perhaps by being forced to acknowledge that it exists, we will at least have a considered ratio and, overtime, one slightly less crass.
[1] This example is imperfect because most people would agree that—holding national origin constant—the death of an innocent civilian is more tragic than the death of a soldier. For this reason, the ratio animating drone-strike-related decisions likely overestimates the American government’s value of non-American lives.
Related here is the fact that the US military compensates victims generally up to $2500, at least if they are not directly related to a military strike: link.