Women and children first! Not!


In a recently published working paper (see here), Swedish economists Elinder & Erixson revisit the issue of gender, social norms, and survival in maritime disasters. This topic has attracted some attention through the controversial work of Frey et al. (e.g., here; see also here) which made much of the alleged bravery of male passengers on the Titanic who allegedly sacrificed themselves so that women and children might be saved, seemingly providing further evidence for a “women and children first” social norm. Specifically, Frey et al. argued that “even though the two vessels and the composition of their passengers were quite similar, the behavior of the individuals on board was dramatically different. On the Lusitania, selfish behavior dominated (which corresponds to the classical homo economicus); on the Titanic, social norms and social status (class) dominated, which contradicts standard economics. This difference could be attributed to the fact that the Lusitania sank in 18 min, creating a situation in which the short-run flight impulse dominated behavior. On the slowly sinkingTitanic (2 h, 40 min), there was time for socially determined behavioral patterns to reemerge.“ (from their abstract)

There was much wrong with the Frey et al paper (an earlier version of which I reviewed for another journal). The authors’ assumption that these two “treatments” differed only in the speed of sinking was unwarranted, as speed of sinking, and time pressure, were confounded along multiple dimensions. For example, the knowledge conditions were quite different across the two “treatments”. The Titanic sank three years earlier than the Lusitania; it also had a reputation for being ‘unsinkable’ that  made it easy to seemingly accept the “women-and-children-first” social norm. Indeed, the first life boat was launched only one hour after the Titanic struck an iceberg and was populated only by a fraction of the passengers it could carry. In essence, chivalry seemed rather inexpensive on the Titanic. At least during the first couple of hours. In contrast, the passengers on the Lusitania almost certainly knew about the sinking of the Titanic. They also knew that their ship was likely to be targeted by the German Imperial fleet. Thus the passengers on the Lusitania surely had no doubt what happened, and what their likely fate would be, when the Lusitania was hit by a torpedo. That very moment chivalry had become rather expensive.

Elinder & Erixson analyse a database of 18 maritime disasters spanning three centuries, and covering the fate of over 15,000 individuals of more than 30 nationalities. Taking their cue from the work of Frey and his collaborators, they test six hypotheses.

Their results suggests that there is little place for chivalry during maritime disasters: Women have a substantially lower survival rate than men. Contradicting the adage that captains (and crews) go last in these kind of situations, Elinder & Erixson find that crews, and captains, do significantly better than passengers (whether female or male); see Figure 1 in their paper. (Yes, Francesco Schettino, you are not alone.) Predictably, children fare worst. Importantly, the authors also find that duration of a disaster has no effect on the survival probability of social norms. The one thing that seems to give the “women-and-children-first” social norm a chance is an explicit policy of the captain. Essentially this finding demonstrates that the potential for punishment has some sway. Elinder & Erixson (2012, p. 8) also show “that women fare worse, rather than better, in maritime disasters involving British ships. This contrasts with the notion of British men being more gallant than men of other nationalities.”

9 Responses to "Women and children first! Not!"
  1. Another aspect about Frey’s work that you should mention is his highly unethical fourfold simultaneous submission of that study. See the Economic Logic blog for much detail about it.

  2. Yeap, I know … (the Economics Intelligence blog, too, which documents further transgressions) but I do not want to go there here and now.

  3. “… crews, and captains, do significantly better than passengers …”

    But surely it would be amazing if they didn’t. Compared with the women and children they’d all be able-bodied, they’d be more familiar with the hazards and so less prone to panic, they’d know how to launch and get aboard a lifeboat, they’d know how to keep it afloat in a storm, they’d even be more likely to know how to swim.

    Expertise, as well as behaviour, figures in the odds of survival at sea. Note that’s even true in plane crashes today – if a plane ditches in the water, or catches fire after a crash landing, the crew consistently gets out much more easily than the pasengers.

  4. bit of schadenfreude here, Andreas.

    As to the content, for a social norm like ‘women and children first’ to be possibly implemented you need the unusual situation of there to be time and opportunity for something like group decision making to be even possible. Large ships going down slowly with some lifeboats fit that bill, other situations dont. Ships that break up on cliffs in big storms for instance or that are too small or full of sick patients and children who cant swim dont really qualify as useful examples.
    Without knowing what these other 18 disasters are hence, it is very hard to say whether or not Frey et al’s story is bust or not. I have my suspicions that disasters spanning three centuries must include a lot of situations not conducive to any form of decision making.
    You say chivalry was cheap on the Titanic, but that would clearly not have been true of the senior crew who knew the severity of the situation. Indeed, it sounds unlikely that it would have held for anyone onboard: you dont go out in a small lifeboat in a freezing ocean for a picknick and you dont look at one being launched and say ‘ooh, I wonder if we have a probem?’.

  5. Impressed by your German language skills, Paul.

    Although not quite sure what the schadenfreude refers to.

    The Elinder – Erixson paper is 10 pages long and comes with an 68 page appendix (single-spaced) that documents all the 18 disasters. Interesting reading that might answer your questions.

    As to the Titanic: if senior crew really knew early that they were headed for serious trouble, why would they have launched the first lifeboats more than an hour after the Titanic struck the iceberg? Apparently it took quite a while for the situation to become bad enough for people to realize that something was really going wrong: http://www.spiegel.de/flash/0,5532,28319,00.html (in German but with pictures ;-))

%d bloggers like this: