Tuesday, 8 November 2011

'Do conservative governments make people want to die?"

I recently reviewed 'Why Some Politicians Are More Dangerous Than Others' for the LSE Politics and Policy blog, a new book by James Gilligan, Clinical Professor of Psychiatry at New York University. In the book, through observations covering over 100 years of US data, Gilligan outlines distinct trends in patterns of homicide and suicide ('violent death'). Namely, he observes that when a Republican is president rates of violent death increase and when a Democrat is president rates of violent death decrease. He puts this down to the much higher levels of socio-economic stress which are caused by Republican policies, such as increased unemployment and income inequality. Quite literally, it seems that conservatism kills.

The question which naturally arises for the British reader is whether this pattern holds in the UK. In other words, is increased mortality correlated with Conservative governments? Does conservatism kill?

This 2002 paper from Shaw, Dorling and Davey Smith sets out to answer this question with regards to suicide (with the rather morbid sub-heading 'Do conservative governments make people want to die?'). In the paper, the authors compare the suicide rate (per million) for each five-year period over the 20th-century. They then assign a prime minister to each five-year period (where there is more than one, the PM in power for the longest period during the five years is chosen).

The results are pretty stark. For example, the five most 'suicidal' five-year periods are as follows:

  • 1961-1965 (Macmillan, Conservative: 137 suicides per million)
  • 1931-1935 (MacDonald, Conservative/Liberal coalition: 135)
  • 1936-1940 (Chamberlain, Conservative: 124)
  • 1926-1930 (Baldwin, Conservative: 123)
  • 1986-1990 (Thatcher, Conservative: 121)
All in all, the authors state that the mean suicide rate during period of Labour and Liberal rule was 103 suicides per million per year. When the Conservatives were in power, the suicide rate was nearly always on average 1.17 times greater than this. In total, the paper argues that the 17% excess suicide attributable to Conservative governments is equivalent to 35,000 extra suicides over the past century. Or, to put it more bluntly, 2 suicides for each day that the Tories are in power.

The paper by Shaw and colleagues was a response to an Australian study which also found that higher rates of suicide were associated with conservative governments. So from just these three examples, we have evidence of a link between conservatism and death which spans the US, Britain and Australia. The reasons for this are unclear and contested, ranging from psychosocial explanations (there is more hope under centre-left governments) to material ones (there are better public services and lower poverty rates).


  1. Or it could be that conservative governments are elected at times of social change, creating feelings of anomie which increases the chance of suicide. Also, "the average labour" figure is rather dubious, you'd want the individual values of labour goverments in comparison with conservative ones, and look at trends as a whole.

  2. Hi Ashley, thanks for your comment.

    Yeah I'm sure there are lots of alternative theories you could explore. As you say, it could hypothetically be that centre-left governments cause social and economic problems which are then inherited by centre-right governments. If that's the case, you would have to say it is in fact the actions of social democratic parties that is in fact linked to suicide.

    I admit too that the paper I link to is a bit dubious. Even the authors themselves admit it is a simplistic descriptive analysis and a topic which needs more research (not aware of anything more robust though - this is the paper cited in Gilligan's book).

  3. What were the figures under a Labour government?

  4. Check the full paper. The authors found that suicide rates under Conservative governments tended to be around 15% higher than the average for Labour/Liberal in the C20.