Should we consider the croupiers at Crown Casino to be public sector employees? How about people who file away books at the National Library of Australia? The answers to those questions seem to be yes and no, respectively, according to the Institute of Public Affairs.

The IPA has tried to whip up a bit of a scare about public sector employment, which is rising much, much faster than private sector employment. The only problem with this claim is that it isn’t true. Between 2000 and 2012, public sector employment (federal, state, and local) rose by an average 2.1% a year. Jobs in the private sector rose by 2.0% per year.  Not exactly an eye-popping difference. This means that over the 12 year period, the number of private sector workers per public sector worker went from 5.1 to… 5.1. [fn1]

The number of private sector workers per public sector worker

Ratio from SEE and LFS

Source: My calculations using ABS 6248.0.55.002 and ABS 6202.

That doesn’t make for a dramatic story. If you’re writing about the out-of-control growth of government, these numbers don’t really fit.  The IPA, instead, chooses to focus on a different measure, one based on the industry people are employed in rather than a strict public/private sector divide.

The IPA suggests that if we’re looking at the growth of the state, we need to include people whose employment depends in some way on government – private school teachers, say, or carers in non-government organisations that receive funding from public agencies. They say that this broad measure of public and publicly-subsidised employment “includes government employees, and employees in education, healthcare, social assistance, arts and recreation, and some utilities.”

When you use what appears to be the IPA’s definition of the “public sector,” the story is quite different. Instead of a stable ratio of private to public workers since 2000, the ratio has steadily fallen – this means that “public sector” jobs have come to occupy a larger proportion of the workforce.

The number of “private sector” workers per “public sector” worker – IPA definition

Ratio using IPA definition

Source: My calculations using ABS 6291.0.55.003. [fn2]

I’m pretty sure that the definition of the “public sector” I used to construct the chart above matches the IPA’s, as they say that the ratio of “private” to “public” jobs has declined from 3.2 to 2.5 since 2000.

Under what I believe is the  IPA’s definition of “public employment”, some of the following industries are included:

Some things that don’t appear to count as the public sector, according to the IPA:

It’s pretty clear, to me at least, that the definition of the “public sector” used by the IPA is not particularly useful. Any definition of public employment that includes Gotye but not the person stamping your passport at the airport seems to me to be a poor basis for an op-ed in a national newspaper about the growth of government.

[fn1] I’m using the ABS figures on public sector employment. To calculate private sector employment, I’m subtracting these from the ABS Labour Force data for June of the relevant year. The public sector jobs figures are affected by privatisations, but most of these occurred in the 1990s and earlier.

[fn2] In accordance with what I believe the IPA’s definition to be, I have included in the “public sector” all of the following industries: public administration and safety; education and training; health care and social assistance; arts and recreation services; and electricity, gas, waste and water services. The private sector is the sum of all other industries. If the IPA figures are constructed on some other basis, I would be happy to correct this post.