by Robin Hanson
13These figures on the marginal returns to education (both personal and national) are estimated in Caplan (2017, preprint), drawn from a range of estimates in Pritchett (2001); Islam (1995); Benhabib and Spiegel (1994); Krueger and Lindahl (2001, 1125); Lange and Topel (2006, 462–70); de la Fuente and Doménech (2006). Further complicating the matter is the prospect that there may be some “reverse causation,” where increases in national income trigger more schooling rather than the other way around. See, e.g., Bils and Klenow (2000).
14Spence 1973. Once again, it’s not technically a “Nobel Prize,” but the “Nobel Memorial Prize in Economic Sciences.”
15Actually it’s not just that employers want to directly evaluate each worker’s productivity for themselves. Employers also want show off their employees to outsiders such as customers, suppliers, and investors. When we, Robin and Kevin, visit firms, we often hear them brag (discreetly) about the prestigious degrees of their employees.
16Also note that you’ll probably be agnostic about how, exactly, she managed to get these good grades. She might not be particularly intelligent, for example, but if she’s able to compensate by staying organized and working extra hard, she’s going to bring those same qualities to the job. Or maybe she’s lazy, but brilliant enough that it doesn’t matter. Either way, her grades have proven that she can get things done.
17Or perhaps it was said by Grant Allen. See O’Toole 2010a.
18Thiel 2014.
19On the psychic costs, consider:
Carolyn Walworth, a junior at [high-achieving] Palo Alto High School, recently wrote: “As I sit in my room staring at the list of colleges I’ve resolved to try to get into, trying to determine my odds of getting into each, I can’t help but feel desolate.”
She confessed to panic attacks in class, to menstrual periods missed as a result of exhaustion. “We are not teenagers,” she added. “We are lifeless bodies in a system that breeds competition, hatred, and discourages teamwork and genuine learning.” (Bruni 2015)
20See Carey 2015 for more on online courses; The Thiel Fellowship, “About,” http://thielfellowship.org/about/.
21Macskássy 2013. The analogous figure was 15 percent for high school graduates.
22One recent study (Bruze 2015) suggests that, in Denmark, people are earning “on the order of half of their returns to schooling through improved marital outcomes.”
23Online Etymology Dictionary, s.v. “academy,” http://www.etymonline.com/index.php?allowed_in_frame=0&search=academy.
24Wikipedia, s.v. “Prussian education system,” last modified February 16, 2017, https://en.wikipedia.org/wiki/Prussian_education_system.
25Kirkpatrick 2010; Wikipedia, s.v. “Pledge of Allegiance,” last modified March 3, 2017, https://en.wikipedia.org/wiki/Pledge_of_Allegiance.
26Aghion, Persson, and Rouzet 2012.
27Lott 1999.
28Diamond 1997.
29Bowles and Gintis 1976, 40–1.
30Spring 1973; Braverman 1974; Weber 1976; Brint 2011.
31The effect of school on culture might be just as important as its effect on individual students. It may be, for example, that the high prestige of schools from ancient times helped us all (students, parents, broader society) come to tolerate, and even celebrate, its domesticating influence. That is, if it had been easy to get parents to accept sending their kids off for domestication training, it would have been cheaper and more effective just to send them off to do child labor. But if that was a hard sell, the cover story of “learning from prestigious teachers” might have made school an easier sell. For more on the link between learning and prestige, see Henrich and Gil-White (2001) and Henrich (2015).
32Clark 1987.
33Ibid.
34Almås et al. 2010.
35Weber 1976, 330.
36Ibid., 329.
37In his acceptance speech for New York City’s Teacher of the Year award in 1989, John Gatto said what many teachers surely recognize, but few are willing to state so baldly. “Schools and schooling,” he said, “are increasingly irrelevant to the great enterprises of the planet. No one believes anymore that scientists are trained in science classes or politicians in civics classes or poets in English classes. The truth is that schools don’t really teach anything except how to obey orders” (1990).
38Gaither and Cavazos-Gaither 2008, 313.
CHAPTER 14
1World Bank Open Data, http://data.worldbank.org/.
2The quotation has been edited for clarity. This is the full quote:
In our area of the country, when somebody gets sick that we know or has passed, we take over food. Have you noticed it? We take over food. You can buy that food, you can go to the deli and the grocery store, get something great, hire somebody to bake it. But put it down in the big list of important things for life: you get a lot more credit if you make it yourself. You can put it on your grandmother’s platter, but the women in the kitchen will say, “I know where she got that chicken.” I’m telling you, it works out that way. (Robertson 2017)
3According to de Waal (1996), these helping behaviors extend deep into the prehistoric and even prehuman past. Neanderthals, for example, cared for their injured group members in this way. We know this because adult skeletons have been found with leg bones that were broken during childhood. Even non-primate species have been observed caring for sick and injured group members. These include dolphins, whales, and elephants.
4From Hanson 2008: “Shamans and doctors have long been in demand, even though the common wisdom among medical historians today is that such doctors did very little useful on average until this century (Fuchs 1998).”
5Belofsky 2013, (trepanation) 8, (toothworms) 74–75, (lead shields) 60.
6Ibid., 101–102: “Wielding a scalpel-like seton, doctors would cut into their patients with a ‘sawing motion.’ Foreign objects, usually dried peas or beans, would then be inserted into the gash to promote proper infection and oozing. A doctor would reopen the wound, often every day, for weeks or months afterward, to make sure it didn’t heal.”
7Ibid., 47.
8Szabo 2013.
9Margolick 1990.
10Waldfogel 1993.
11And we will look mainly at the U.S. because that is where we have the best data.
12Skinner and Wennberg 2000.
13Mullan 2004; Cutler et al. 2013.
14Auster, Leveson, and Sarachek 1969.
15Age- and sex-adjusted death rates.
16Fisher et al. 2003. See also Fisher et al. (2000): “Residence in areas of greater hospital capacity is associated with substantially increased use of the hospital, even after controlling for socioeconomic characteristics and illness burden. This increased use provides no detectable mortality benefit.”
17Byrne et al. 2006.
18Skinner and Wennberg 2000. This estimate was not significantly different from zero. The size of this dataset allowed researchers to control for many factors, including patient age, gender, and race; zip code urbanity, education, poverty, income, disability, and marital and employment status; and hospital-area illness rates.
19At the 95 percent confidence interval. Here we’re using a “50 days lost per 1 percent added mortality” rule of thumb.
20Skinner and Wennberg 2000.
21Hadley 1982.
22Brook et al. 2006; Newhouse and Insurance Experiment Group 1993.
23The partially subsidized groups also included a “maximum dollar expenditure.” Once a patient paid the maximum amount in a given year, the rest of his or her care was free.
24Manning et al. 1987.
25Measured in total dollar value of all services covered under the insurance plans. See ibid.
26Unfortunately the RAND study wasn’t large enough to detect effects on death rates, so it tracked only intermediate measures of health.
27Actually there were 23 physiological measures, but we’re omitting one measure (long-distance vision) because the treatment for it—corrective lenses—seems to us more a
matter of physics than medicine.
28Newhouse and Insurance Experiment Group 1993. In fact, the researchers found an “almost significant” result (at the 6 percent significance level) that free medicine actually hurt the subset of patients who started out both poor and healthy.
29Brook et al. 1984. Again we’re omitting the statistically significant (but entirely predictable) improvement in long-distance vision that accompanies a subsidy for eyeglasses.
30Siu et al. 1986; Pauly 1992; Newhouse and Insurance Experiment Group 1993.
31Not every lottery winner ended up enrolling in Medicaid, and not every lottery loser ended up without insurance. There was, however, a meaningful difference between the two groups: in the year following the lottery, winners were 25 percentage points more likely to have insurance than losers.
32Finkelstein et al. 2012.
33Baicker et al. 2013.
34Finkelstein et al. 2012.
35Baicker et al. 2013.
36Such a cutback might be done by raising the price of medicine across the board or by banning the treatments that have the weakest empirical support.
37Tuljapurkar, Li, and Boe 2000; McKinlay and McKinlay 1977; Bunker (2001) estimates. (More at Lewis 2012.) Note, however, than many scientists mistakenly pronounce medicine as responsible for most of our health gains. From Bunker (2001):
The Nobel Laureate and President of Rockefeller University, Joshua Lederberg, wrote that “by the 1960s we could celebrate the conquest of polio and the transformation of formerly lethal infections to easy targets for penicillin and other miracle drugs . . . greater life expectancy—from 47 years in 1900 to 70 in 1960—can be attributed almost entirely to this mastery of infection… .” The Nobel Laureate and former research director of Burroughs Wellcome, the pharmaceutical company, George Hitchings, claimed that “the increase in life expectancy over the last 50 years has been attributed to new medicines.”
38Ioannidis 2005a, 2005b. From Lewis (2012): “The impact of a treatment in a clinical trial is known to be much higher than its effect in everyday clinical practice.”
39Ioannidis 2005a, 2005b.
40Aizenman 2010.
41Getzen 2000.
42Waber et al. 2008.
43Emanuel 2013.
44Periyakoil et al. 2014. Doctors, having witnessed the futility of heroic end-of-life care, are famously keen on avoiding it for themselves, when they become terminally ill.
45Mundinger et al. 2000.
46Schneider and Epstein 1998.
47Mennemeyer, Morrisey, and Howard 1997. In New York City, where patients have their choice among many different hospitals, poor-performing hospitals actually saw an increase in admissions relative to high-performing hospitals. See also Vladeck et al. 1988.
48Mennemeyer et al. 1997.
49Institute of Medicine et al. 1999; Leape 2000.
50Institute of Medicine et al. 1999; National Academy of Sciences 2015.
51Gawande 2007; Jain 2009.
52Lundberg 1998; see also Nichols, Aronica, and Babe 1998, which found that approximately two-thirds of the undiagnosed conditions revealed by autopsy would have been treatable had they been caught earlier.
53Shojania et al. 2002.
54O’Connor 2011.
55Westra, Kronz, and Eisele 2002.
56Staradub et al. 2002 (second opinions caused 8 percent of breast-cancer screenings to result in a different surgical treatment plan).
57Mandatory second-opinion programs: Gertman et al. 1980 (8 percent of elective surgery recommendations were overturned); McCarthy, Finkel, and Ruchlin 1981 (between 12 and 19 percent of elective surgery recommendations were overturned); Althabe et al. 2004 (25 percent of recommended C-sections in Latin America were overturned).
58Lantz et al. 1998.
CHAPTER 15
1Technically, they enter “diapause,” an insect analog of hibernation.
2Pocklington 2013.
3Katz 2013.
4Wikipedia, s.v. “Hajj,” last modified March 6, 2017, https://en.wikipedia.org/wiki/Hajj.
5In fact, Mecca is unbearably hot, reaching average daily highs of 110ºF (43ºC) from June through September. (The Hajj takes place annually by the lunar calendar, and therefore falls on different dates on the solar calendar every year.)
6Wikipedia, s.v. “Ihram,” last modified January 8, 2017, https://en.wikipedia.org/wiki/Ihram.
7As the 16th-century diplomat Ogier Ghiselin de Busbecq said of the Spanish conquistadors, “Religion is the pretext, gold the real object” (Forster 2005, 40).
8Cf. Pascal’s wager.
9This is the approach taken by the New Atheists, for example. And while there’s a lot of insight to be gleaned from it (notably, the idea that religious beliefs are designed to take advantage of our cognitive quirks), it’s largely a distraction from our focus in this book.
10Haidt 2012, 249–50.
11Rappaport 1999.
12Sosis and Kiper 2014.
13Anderson 2006.
14There’s a lot more to be said about the downsides of religion, as the New Atheists (Dennett, Dawkins, Harris, Hitchens) have argued in great detail, but it’s mostly not our goal to tally up the pros and cons and pass judgment on the whole enterprise. Note, importantly, that religion can be useful for adherents without necessarily being good for the entire species. It’s perfectly consistent to believe that religious participation is a selfish individual strategy and that, on net, it’s bad for the world. In this way, religion would be like any other form of clannishness: when everyone else is organizing into clans around you, it may be necessary for you to join one, while at the same time wishing that clans didn’t exist and everyone could just get along.
15In other words, we’re going to provide a functionalist account of religion (Swatos and Kivisto 1998, 193–96). Cf. Haidt: “To resolve [the puzzle of religious participation], either you have to grant that religiosity is (or at least, used to be) beneficial or you have to construct a complicated, multi-step explanation of how humans in all known cultures came to swim against the tide of adaptation and do so much self-destructive religious stuff” (2012, 252).
16Strawbridge et al. 1997.
17Schlegelmilch, Diamantopoulos, and Love 1997: donations. Becker and Dhingra 2001: volunteering. See also Putnam and Campbell: “By many different measures religiously observant Americans are better neighbors and better citizens than secular Americans—they are more generous with their time and money, especially in helping the needy, and they are more active in community life” (2010, 461; as quoted in Haidt 2012, 267).
18Strawbridge et al. 1997.
19Mahoney et al. 2002, 63; Strawbridge et al. 1997; Kenrick 2011, 151.
20Frejka and Westoff 2008; Kenrick 2011, 151.
21McCullough et al. 2000; Hummer et al. 1999; Strawbridge et al. 1997.
22Steen 1996.
23Wink, Dillon, and Larsen 2005.
24Lelkes 2006.
25Haidt 2012, ch. 11.
26This is frequently attributed to Durkheim (who wrote in French), though it may be apocryphal. Nevertheless, it’s a great capsule summary of his views, especially those articulated in Durkheim 1995.
27Few scholars attempt to define religion precisely and unambiguously; there are simply too many boundary cases (like Confucianism) to draw a bright line between religion and non-religion. Most scholars, instead, attempt to associate religion with a cluster of interrelated features, and the more features something has, the more we’re willing to call it a “religion.” Here, for example, are a few “definitions” of religion. Atran and Henrich (2010): “an interwoven complex of rituals, beliefs, and norms.” Rue (2005): “a natural social system comprising a narrative core buttressed by intellectual, aesthetic, experiential, ritual, and institutional strategies.” Sosis and Kiper (2014): “a fuzzy set that comprises (but is not limited to) commitments to supernatural agents, emotionally imbued symbols, altered states of consciousness, ritual performance, myth, and taboo.”
28
Cf. Haidt 2012, 251.
29D. S. Wilson: “Religions exist primarily for people to achieve together what they cannot achieve on their own” (2002, 159).
30Roberts and Iannaccone: “It never makes sense in an economic context for me, a perfectly rational person, to take a resource and just burn it up. But in a group context, strange as it may seem, this can be efficient” (2006).
31Sosis and Alcorta: “Religions often maintain intragroup solidarity by requiring costly behavioral patterns of group members. The performance of these costly behaviors signals commitment and loyalty to the group and the beliefs of its members. Thus, trust is enhanced among group members, which enables them to minimize costly monitoring mechanisms that are otherwise necessary to overcome the free-rider problems that typically plague collective pursuits” (2003).
32Iannaccone: “It can be shown, both formally and empirically, that apparently gratuitous sacrifices can function to mitigate a religion’s free-rider problems by screening out halfhearted members and inducing higher levels of participation among those who remain” (1998).
33Wikipedia, s.v. “Mourning of Muharram,” last modified February 5, 2017, https://en.wikipedia.org/wiki/Mourning_of_Muharram.
34Johnstone 1985.
35Iannaccone 1992, 1998.
36For a smaller sacrifice of fertility, some Christian teenagers wear purity rings as a public commitment to delay sex until marriage.
37This also helps explain why eunuchs have historically held privileged positions.
38On raising commitment as a way to reduce costly monitoring, see Sosis and Alcorta 2003; Iannaccone 1992, 1998. For more general evidence of religious cooperativeness (not necessarily caused by sacrifices), see Tan and Vogel 2008; Ruffle and Sosis 2006; Atran and Henrich 2010. For an overview, see Haidt 2012, 256–57, 265–67.
39On group size in relation to costly rituals, see Roes and Raymond 2003; Johnson 2005 (both via Atran and Henrich 2010). On longevity as a function of costly rituals, see Sosis and Bressler 2003. More broadly, on the longevity of religious vs. secular communities, see Sosis 2000. Sosis and Alcorta 2003.