According to Yobbo in comments yesterday:
The modern workforce is a lot more flexible and a lot of people change jobs every 2-3 years.
It’s certainly true, as the labour mobility statistics show, that lots of people change jobs – in 2004, more than 40% of workers had been in their jobs for three years or less. But to imply that this is a trend, as Yobbo’s remark does, is yet another example of our poor ability to compare over time. In an article labour market economist Mark Wooden wrote in 1999, he has a table showing that in 1975 36.1% of workers had been in their jobs for two years or less. In 2004, 34.5% of workers had been in their jobs for two years or less, a slight decline in short-term job holding.
Another curious – because it does not conform with our perceptions – feature of the 1975 statistics is that the ‘jobs for life’ that we supposedly use to have are rather hard to find. 8% of the workforce had been in their jobs for 20 years or more, slightly less than the 8.6% in that category in 2004. In the same period, the proportion of workers in their jobs for 10-20 years grew from 12.4% to 15%.
One likely explanation is the ageing of the workforce, with the average worker nearly three years older than 20 years ago. As people get older they tend to stay in their jobs for longer periods of time. Having turned 40 I’m noticing this in my peer group (and saving time in updating my contacts) and myself, having now spent 6.5 years with the same employers. And with young people a shrinking proportion of the total population, and delaying their entry to the workforce by studying, the proportion of workers with a propensity to be job fickle has decreased.
Andrew Norton 
Aging and getting involved in bringing up kids reduce mobility. I think governments could improve mobility by eliminating or reducing transactions costs of selling the family home (especially stamp duty). Immobility has less impact with more temporary and permanent skilled migration.
LikeLike
When you control for gender, you do get a different story for men and women though (as you usually do with labour force trends). Even over the last 10 years, there has been a slow decline in average tenure for men and a slow increase for women, which have tended to cancel each other out. So I imagine that if you looked back thirty years it would be much more dramatic.
I for one am in favour of anything that promotes/illustrates greater gender equality, but for those people (who unfortunately are not few enough in number) who believe that the only real labour market is the male labour market I guess the story is not as good.
LikeLike