Tasmania has set an unfortunate record: it’s the first Australian state in which less than half of all adult men are employed full time. In the lead-up to the financial crisis, the proportion of Tasmanian men in work soared, rising faster than the national ratio, but it has since plummeted. In February 2013, just 48.3% of Tasmanian men aged 15 and over were in full-time work; this was 8.3 percentage points below the national figure of 56.6%.
After posting this chart on Twitter I received quite a few responses from people who suspected that this wasn’t really a big deal. I’d like to deal with them in turn.
First of all: what about part time work? If Tasmanian men choose to work part time at a higher rate than mainlanders, then they might have a low full-time employment rate, but a healthy overall employment rate. My first response to this is that this doesn’t explain why full-time work has become so much less common in Tasmania in just the past few years. Did Tasmanian men decide en masse in 2009 to suddenly cut back their hours? That seems implausible. In any case, it’s not what happened.
61% of Tasmanian men are in work of some kind, but the national figure is 67.8%. Tasmanian men are a little more likely to work part-time than mainlanders, but this barely closes the gap between the employment rates. Plus, the overall employment-to-population ratio has plummeted in Tasmania in the past couple of years, just as the full-time rate has.
This isn’t just a story about men, either. In December 2008, 26.9% of Tasmanian women had full-time jobs. Now the figure is down to 23.2%, whereas 30% of Australian adult women are in full-time work. The decline has been steeper for men, which is why I’m focusing on them, but it’s been present for both genders.
One persistent question that people had was about age. Could the difference just be that Tasmania’s population is older than the rest of Australia? . On the mainland, 18.3% of the adult population is aged between 25 and 34; in Tasmania it’s just 13.9%. Those 65 and over account for 19.5% of the population in Tassie, compared to 16.2% in the other states.
So there’s something to this story – Tasmania’s population does skew towards seniors, and this does drag down the employment rate. But it doesn’t account for the big difference in employment rates between Tasmania and the mainland.
Tasmanian men of all ages over 20 are less likely to be in full-time work than their mainland counterparts. For example, 78.4% of mainland men aged 25-34 are in full-time work; in Tassie, it’s 72.9%. They’re a little more likely to be in part time work, as I noted above, but this doesn’t offset the difference in full-time employment rates.
You can put these numbers together and figure out exactly how much of the 7.2 percentage point gap in the male employment rates of Tasmanians and mainlanders is caused by the fact that Tasmanians are older, on average; and how much of the gap is due to the fact that Tasmanians are less likely to be in employment than mainlanders of the same age.
When I do this decomposition analysis, I find that the Tasmanian male employment-to-population ratio is:
- Reduced by 5.7 percentage points due to the fact that Tasmanian men are less likely to be in full-time work than mainlanders of the same age;
- Increased by 1.4 percentage points due to the fact that Tasmanian men are more likely to be in part-time work then mainlanders of the same age; and
- Reduced by 2.9 percentage points by the fact that the Tasmanians are older, on average than mainlanders.
In other words, less than half of the gap between Tasmanian and mainland employment rates (2.9 percentage points out of 7.2) can be explained by the fact that Tasmania’s population is older than their mainland counterparts.
This leaves the question of why, exactly, Tasmanian men are less likely to be in full-time work than mainland men of the same age. That’s a topic for another post.
Note: the data source for everything in this post is the ABS Labour Force survey, here and here. For my decomposition and the final three charts, I am using non-seasonally adjusted data. The decomposition and the charts are therefore based on annual averages for 2012.