It’s Raining Men! Hallelujah?
Pauline Grosjean and Rose Khattar
We document the implications of missing women in the short and long run. We exploit a natural historical experiment, which sent large numbers of male convicts and far fewer female convicts to Australia in the 18th and 19th century. In areas with higher gender imbalance, women historically married more, worked less, and were less likely to occupy high-rank occupations. Today, people living in those areas have more conservative attitudes towards women working and women are still less likely to have high-ranking occupations. We document the role of vertical cultural transmission and of homogamy in the marriage market in sustaining cultural persistence. Conservative gender norms may have been beneficial historically, but are no longer necessarily so. Historical gender imbalance is associated with an aggregate income loss estimated at $800 per year, per person. Our results are robust to a wide array of geographic, historical and present-day controls, including migration and state fixed effects, and to instrumenting the overall sex ratio by the sex ratio among convicts.
Keywords: Culture, gender roles, sex ratio, natural experiment, Australia
JEL: I31 N37 J16
I didn’t go to this presentation, but its a great title for a paper. The key question with papers like this is what actually caused the gender imbalance in various areas in those historical times. People didn’t just settle anywhere, and even prisoners were not just sent anywhere. There were reasons one must understand if one is to know what factors to ‘control for’.
Take the basic hypothesis: too many men 2 centuries ago means conservatism now. So too few men for a generation should then mean one gets girl power? Think about the more recent cases of lots of men dying and the subsequent cultural shifts. Does the logic go through for Vienna, which lost a lot of men in the first world war, or Berlin in the second? Does the logic hold for France’s losses in the first World War, or Japan’s and Russia’s in the second, both with losses heavily gender skewed? And these losses were much more recent than merely the 18th and 19th century. Thinking even more recently, do you get girl power in the areas of Afghanistan where the men have killed each other, and did you get a wave of feminism in Iran after the very bloody war with Iraq in the 80s? I don’t think so. So one has to be ex ante skeptical of such historical determinism.
The idea that a particular generation’s experiences were pivotal for all the rest, independent of what came before or after, and would not have been washed out by the huge migration streams after the 18th and 19th century, is very suspect. It would be good to have some informed historian comment on what actually determined where people went in those days.
Could it be that the areas with the greater gender imbalance were the more isolated areas, and still are isolated places i.e fairly inward looking?
No idea, John. You really need a good historian to keep you honest with this kind of question, otherwise one is in real danger of falling in love with the story one wishes to see.
I think the hypothesis is more along the lines ‘many generations of male domination lead to cultural outcome x’. It was a long term issue – not just a point in time.
SMH did a piece on it – excerpt below. http://www.smh.com.au/comment/its-reigning-men-how-our-convict-past-explains-our-glass-ceiling-20140628-zspo9.html
“Unusual is an understatement. For the best part of a century from 1788 to 1868 a total of 157,000 prisoners were sent from Britain to Australia. Only 25,000 were women. By 1833 male convicts accounted for 80 per cent of the recorded east Australian population. Among convicts the ratio of men to women was 8 to 1. Over time the ratio in the general population settled down to 3 to 1.”
Germany, Japan, France, Russia – not quite the same situation. Perhaps if the male populations were wiped out for 4 or 5 generations running? I reckon ‘girl power’ is a pretty reasonable expectation in that scenario.
typical. The results of the authors are not based on the average ratio, but on the spatial distribution of the ratio at particular moments. So the results of the authors are not based on a continuous measure. And what do you think happened to those superfluous men and their culture (hint: they had no progeny, so the majority of men each of those generations had to come from some other culture and only a few of those procreated). And of course the major population waves came much later, from other cultures, unlike Europe post 1900. Furthermore, even taking the period given, you have the niggly issue of which spatial distribution you take and why. If the spatial gender distribution remained the same, the question is why it would be so particular (rural/urban?). If the spatial distribution changed in this period, the question is which snapshot you presume to be pivotal.
As I said, you truly need a real historian to illuminate what went on. Otherwise, a fool and his history are soon parted.
Your first argument is a straw man. It’s not like Japan et al and is not presented as a single generational issue. You haven’t responded to this.
Your argument about progeny is also a straw man.
You don’t think the lucky bloke/s fending off between 2 and 8 other suitors over generations is going to affect the culture of marriage and work, because the unsuccessful suitors don’t have children?
Hint- that there are between 2 and 8 unsuccessful suitors, and the effect that has on behavioural norms of the successful suitors and their wives or loved ones, might be the pertinent issue here.
While I haven’t read the paper either, I’m fairly comfortable the hypothesis is worth looking at. Its an interesting question to explore.
You seem to have dismissed it out of hand and used fallacious arguments to do so.
Imperfect study design sure. But how can you do an rct on this stuff?
Dahl,
you dont understand what “The results of the authors are not based on the average ratio, but on the spatial distribution of the ratio at particular moments.” means, do you? The results are not driven by a high ratio for generations in the whole country so what you quote and argue is not in fact related to their results. You always have to watch the rabbits as they go into the statistical hat and not get sidetracked by the romance of the storyline.
John,
yep, the story doesn’t seem to add up. But it is a great title for a paper.
Now now. You’re clearly avoiding any discussion of your principal arguments and instead you suggest others don’t understand the terribly complicated statistics around location and time.
I haven’t looked at the paper – and I suspect you haven’t either.
The arguments you present against it are the wrong basis to dismiss it out of hand. And you have avoided any discussion on those points.
Something doesn’t gel: Australia was the first country in the world where women could both vote and stand for parliament, and there were many other things going on that were quite advanced. In contrast Australia , after the losses of WW1 was a more conservative place.
Fair enough. Although NZ did beat us to the vote.
However, were there any other countries establishing their consitutions and becoming a nation at that time? Timing and ease of making the change would have to be a factor to consider.
Interestingly, while women could (and did) stand for election from 1903 in Aus, it took until 1942 for the first to be elected.
The UK had their first women representatives in 1918. NZ, 1933. US, 1917. Canada, 1921.
Were we really as progressive as we like to make out due to the federation changes around voting?
I don’t know. But I don’t think the voting argument is as strong as the credit we give it. On stated preference, we may win the race. On revealed preference, we appear to be some decades behind.
Paul
“There was no such thing as the Scientific Revolution, and this Book is about it.”
:-)
John,
yes, there is lots of weird stuff in this paper, though it has clearly been a lot of work. For instance, the authors find that attitudes to women in politics are less conservative if the historical male ratio was high and effects on labour force participation (which jars with the other results). The found effects on attitudes disappear when you put in enough controls. Indeed, the labour force outcomes in the 1933 census are almost equally dependent on historical male ratios as the 2011 census, despite 80 years (3 generations) of equal gender ratios. There is also lots of niggly stuff going on with changing boundaries to regions, dropping this or that region in various regressions, and whether or not clustered errors are accounted for. Some effects are reverse, such as the effect of gender ratios in the 1933 census on current outcomes. Some analyses are done in a weird way, such as when the IV-first stage results dont include historical controls but the second stage does. And even at best, the historical gender ration explains a measily 1.7% of current variation in attitudes. It thus really is a bit of a Hodgepodge of results from which my take-home message is that we probably are looking at some form of regional selectivity, ie that the driving element in both historical ratios and current circumstances is something particular to the region. The paper doesn’t really say much about the allocation of either convicts or the others across regions, so, as I have kept saying, to go beyond the beauty pageant of slogans, one needs to find a competent historian to step in and explain the driving forces of gender ratios at the time.
Paul re ” people living in those areas ” does it say what areas/places?
One would like to think the Government would have set a good example in its team of top end decision makers, but no, only one woman in the Cabinet.
Sad decision making for that to happen. Do I believe there was only one woman available with the required talent for the Cabinet ? No.
Cartoon refers . . . . .
http://cartoonmick.wordpress.com/editorial-political/#jp-carousel-774
Cheers
Mick
Just for the record, there’s what I recall as a very detailed treatment of the Australian gender imbalance and its repercussions in Geoffrey Blainey’s The Tyranny of Distance, a classic which is ageing extremely well.