Via Craig Newmark, I saw a column by John Lott summarizing his 1999 paper with Lawrence Kenny, “Did women’s suffrage change the size and scope of government?” Lott and Kenny conclude Yes, by comparing the spending and revenue patterns of state governments before and after women were allowed to vote. I haven’t looked at the analysis carefully and would need a little more convincing that it’s not just a story of coinciding time trends (they have a little bit of leverage because women were given the vote sooner in some states than others), but the story is plausible, at least from the perspective of voting patterns nowadays.
On the other hand . . .
poll data appear to show that the gender gap in voting between men and women is relatively recent–if anything, women used to vote more Republican than men did–so it’s not clear if the effect Lott seems to be finding is occurring from women actually voting for conservative candidates or from some indirect effect of legislators trying to adapt to what they perceive as the preferences of women.
Different views of what is authoritative
I hadn’t heard of Lott’s journal article but it seems to be well-known, with over 100 citations on Google scholar. I’m actually surprised Lott didn’t mention it in his Fox News column, instead only linking to his recent book. I guess that indicates the difference between an academic reader such as myself who is generally more persuaded by peer-reviewed articles (although not always!) rather than general readers who might feel that a book is more authoritative.
Beware the status-quo bias
Finally, I have a couple of comments on Lott’s column. He writes, “it seems that the policies adopted are much more important than who puts them into action, and the evidence indicates that women have long gotten their way” regarding government spending. But I don’t see how he can make this claim. Accepting his claim that women’s suffrage has increased spending, this is still compared to an all-male electorate. This isn’t the same as women “having long gotten their way.” A more reasonable conclusion is that outcomes are somewhere in between what women want and what men want. In fact, before 1980, men turned out to vote at a higher rate than women did, and so it would be more natural to assume that outcomes were closer to men’s preferences than women’s. I think Lott is making a sort of status-quo bias here and perceiving any change from all-male voting as women “getting their way.”
Finally, I think that numerical representation is a real issue. This is not a matter of discrimination, but just the simple fact that certain groups such as women are underrepresented in public office relative to their population. Currently, women are 16% of congressmembers and 24% of state legislators, so they still have a ways to go before they catch up.
P.S. Lott also writes, “without the women’s vote, Republicans would have swept every presidential race but one between 1968 and 2004.” Poll data show a pretty close race in 1996 among men (in the Gallup poll, Clinton led Dole by 1% among men); it’s possible that with an electoral vote calculation, Dole was the clear winner in that group, but my guess is that polling uncertainty is large enough that we can’t really know how the men’s vote would’ve gone in that year.
P.P.S. To summarize: this is an interesting topic. I’d like to hear from Adam Berinksy and other experts on early-20th-century public opinion: what are their thoughts on the political opinions and voting patterns of men and women back then, and how do they interpret the fact that increases in government spending coincided with women’s suffrage?
With respect to the P.S. – since Republicans won all but three of the presidential races between 1968 and 2004, that means women's votes made the difference in all of… two, or perhaps one, presidential races over the last 40 years.
"without the women's vote, Republicans would have swept every presidential race but one [instead of, as John says, every race but three] between 1968 and 2004."
But this is simply another way of saying that without the men's vote, the Republicans would have lost more often. Republicans, the special interest group of a narrow minority!
Adding onto Derek's comment – this is like the similar comment encountered, that the Democratic Party only wins because of black votes. It clearly shows that certain people have a warped idea of what constitutes 'normal', and what is 'other'.
Given Lott's past performances (see http://timlambert.org/lott/), I'd want to see the raw data before accepting his analysis. It also seems a bit simplistic to measure women "getting their way" as government spending without looking at what the spending was for.
Thanks very much for the email and the posting. Not that it matters, but the Fox News piece did actually link to this research. The forth link in the piece goes here: http://ssrn.com/abstract=160530
"A more reasonable conclusion is that outcomes are somewhere in between what women want and what men want."
I don't know who got how much, but I can say that the size of government more than doubled within 10 years of women being given the right to vote and that is well before they made up anything near a majority of voters. The paper looks at a range of laws from prohibition to divorce rules and it is clear that those went decidedly towards positions that women generally favored as soon as they were given the right to vote.
"In fact, before 1980, men turned out to vote at a higher rate than women did,"
I would have to check the paper to see about that date (I am sure that it is in there someplace), but as noted above women were getting positions closer to what they wanted as a group well before they represented a majority of voters.
"Currently, women are 16% of congressmembers and 24% of state legislators, so they still have a ways to go before they catch up."
The piece was arguing that the actual policies seem as least as important as who is in there to implement them.
"Poll data show a pretty close race in 1996 among men (in the Gallup poll, Clinton led Dole by 1% among men); it's possible that with an electoral vote calculation,"
It is precisely in counting up the electoral votes that you see Clinton winning. Of course, if women hadn't been able to vote, Clinton wouldn't have won in 1992 and thus wouldn't have probably won in 1996.
Dear Barry":
There are some problems with the source that you reference. A professor at the University of Maryland compiled these notes.
http://doubletap.cs.umd.edu/WikipediaStudy/nameca…
http://doubletap.cs.umd.edu/WikipediaStudy/lamber…
http://doubletap.cs.umd.edu/WikipediaStudy/Lott1/
On-topic:
– 'the size of government more than doubled'
This is an n of 1. Do we observe this pattern in New Zealand after 1893? Do we observe the opporite in Utah after the Edwards-Tucker Act? Do we observe this since 1984 in Liechtenstein? We do not. Mentioning this solitary datum smacks of agitprop. Readers of stats blogs tend not to be interested in agitprop.
– 'prohibition'
Was there a 1932 disenfranchisement I missed?
Off-topic:
Anyone publicly caught cherrypicking, miscoding, sock-puppeting and treating data and analysis as purest political football needs to learn to take his lumps and move on. Please, lump it, big man.
John,
1. Thanks for pointing out the link. I just checked link #1 and didn't notice the others.
2. I still don't see how you can say that women have "gotten their way." Even setting aside issues of time series collinearity, you have data where men were 100% of the voters, data where men were most of the voters, and data where women are slightly more than half of the voters. There's no data where women are 100% of the voters, and the most reasonable assumption still seems to me to be that the current outcome is something roughly halfway between what the two sexes want.
I think you're on stronger ground arguing that there was an effect than in arguing that the outcome is that women have gotten their way.
3. I understand your point about issue preferences, but I suspect that numerical representation has effects as well.
4. You say "if women hadn't been able to vote, Clinton wouldn't have won in 1992." Assuming the simple counterfactual of only counting men's votes: Gallup had Clinton beating Bush 41-37 among men, so I doubt that Bush would've won in the Electoral College either.
How many counter examples to this are there? Here in the UK, David Lloyd George's welfare reforms (considered the founding of the welfare state) preceded the enfranchisement of women.
It may be that for every issue that was already on the table at the time women got enfranchised, very soon the side that women tended to favor got enacted. However, this doesn't mean that for every issue, the side that women tended to favor got enacted. It seems very plausible that there are many issues that women have preferences on that aren't even on the table now, let alone back in 1920.