At the beginning of the new year, the Coalition was keen to spruik its economic credentials, using the latest employment figures to do so.
In a doorstop interview, Treasurer Scott Morrison said that "2017 was a year of extraordinary jobs growth in Australia, over 400,000 jobs created in the year".
"That was the strongest calendar growth in jobs we have seen on record," he said.
Did 2017 see the strongest growth in jobs in a calendar year on record? RMIT ABC Fact Check gets to work on the numbers.
Mr Morrison's claim is oversimplified.
The Treasurer accurately quoted the increase in the number of jobs in 2017, and it was indeed the largest increase in the raw number in a calendar year since records began, but by concentrating on the raw figure, Mr Morrison ignores the size of the labour market in previous calendar years.
Experts told Fact Check that while the raw figure of over 400,000 seems impressive, measuring the proportional increase yields a much better comparison between different years, as the numbers naturally increase over time as the economy and population grow.
This, they said, would be a much better measure of the "strongest" growth.
In 2017, Australia recorded an increase in employed persons of 3.35 per cent, but other calendar years recorded much larger proportional increases, even if the raw number was smaller.
Mr Morrison's claim also ignores other important underlying data which can demonstrate the "strength" of the labour market.
Had the Treasurer delved deeper than the raw headline jobs figures, he would be more on steady ground.
Australia experienced record full-time job growth in 2017, in both raw and proportional terms.
Last year also saw the equal largest increase in the participation rate — a jump of 0.9 percentage points.
But when it came to the employment to population ratio, which accounts for population growth, 2017's increase of 1.64 percentage points is beaten by a number of other years, notably 1994, which recorded an increase of 2.50 percentage points.
Whilst these figures show that 2017 was undoubtedly a strong year for jobs growth, the idea that it was the "strongest calendar growth in jobs we have seen on record" is not borne out by all of the data.
The Labour Force Survey
As well as releasing raw figures, the ABS releases trend and seasonally adjusted figures.
Trend figures are defined as "the 'long term' movement in a time series without calendar related and irregular effects", whilst seasonally adjusted figures remove "the seasonal effects from the original time series", including the changing date of easter, for example, and the different compositions of trading days in a given month in different years.
For assistance in assessing Mr Morrison's claim, Fact Check contacted Garry Barrett, a professor in the School of Economics at the University of Sydney; Marc Chan, an associate professor in the Department of Economics at the University of Melbourne; and Alberto Posso, an associate professor in the School of Economics, Finance and Marketing at RMIT University.
All three experts recommended the use of seasonally adjusted figures in this instance.
The Labour Force Survey has raw figures for persons employed going back to 1969, but its seasonally adjusted series does not begin until February 1978, which means that the first calendar year change that can be measured is 1979.
The seasonally adjusted figures show 12,037,700 people employed in December 2016, and 12,440,800 employed in December 2017 — an increase of 403,100. This appears to be the figure that Mr Morrison is quoting.
The graph below shows the 12-month increase in employment for each month going back to February 1979.
The largest figure, 409,300, was recorded in the year to August 2005, not December 2017.
However, in making his claim, Mr Morrison referred to the "calendar" year, which indicates a comparison between December figures for each year.
Experts consulted by Fact Check differed in their opinions on whether 12-month changes ending with other months of the year could be compared with December figures reliably.
Associate Professor Chan said that this approach was sound due to the numbers being seasonally adjusted, and noted that the numbers show employment growth has accelerated recently: "3.3 percent from December 2016 to December 2017, as opposed to 2.9 percent from October 2016 to October 2017."
Associate Professor Posso expressed a preference for keeping finishing month constant, even when presented with seasonally adjusted numbers: "June 2017 to June 2018 is potentially different to December 2017 to December 2018."
And Professor Barrett noted that different end months in 12 month comparisons "will reflect long-term trends or cyclical factors at play".
As Mr Morrison mentioned "calendar" years, Fact Check accepts that the fairest manner to assess Mr Morrison's claim is comparing calendar year growth, that is the 12 months to December for each year.
Fact Check has calculated and graphed the calendar year growth in raw jobs numbers for every year going back to 1979 below, using data from the ABS Labour Force Survey.
When comparing figures to the year ending in December for each year to 1979, December 2017's seasonally adjusted figure is the largest recorded.
But raw employment figures do not tell the full story.
Employment figures in context
Professor Barrett told Fact Check that the raw figures were "meaningful" but that "as an economy grows over time (along with the population) the absolute numbers will naturally be increasing".
"A number like 400,000 seems large, and it is important to provide context (such as the size of the labour force)," he said.
Both Associate Professor Posso and Associate Professor Chan agreed with this concept.
Below, Fact Check has calculated and graphed the proportional change in employed persons for every year to December going back to 1979, using the numbers in the ABS Labour Force Survey.
When expressed as a percentage of employed persons for that year, there are a number of years that beat 2017, at 3.35 per cent in seasonally adjusted terms, for employment growth, notably 1985 and 1989, which both recorded 4.1 per cent growth.
Once again, however, these numbers don't tell the full story.
All three experts recommended other measures to supplement the assessment of the "strength" of the job market over the calendar year, including measuring full-time employment growth, the growth in the participation rate, and the growth in the employment to population ratio, all of which are published in the Labour Force Survey.
"Labour markets are complex, so a full story requires that we look at a bunch of data," Associate Professor Posso told Fact Check.
Professor Barrett said that "longer-hour, and more secure, jobs reflect a better performing labour market".
Associate Professor Chan agreed, saying that strong full-time employment growth "supports the notion that the labour market is improving".
Below, Fact Check has graphed the proportional change in full-time employment growth for each calendar year of the Labour Force Survey back to 1979.
In December 2017, an increase of 303,300 full-time jobs was recorded, or 3.69 per cent.
Both the raw figure and the proportional figure are the largest of any calendar year.
The participation rate
The ABS defines the labour force participation rate as "the labour force (persons employed or unemployed) expressed as a percentage of the population".
Associate Professor Chan told Fact Check that this was an important indicator in measuring the strength of the jobs market, because "if unemployed people stop looking for jobs, they will not be counted as being unemployed but instead as being detached from the labour market."
He said this would show up as a low participation rate, which he noted happened in the United States after the global financial crisis.
Fact Check has graphed the percentage point change in the labour force participation rate for each calendar year back to 1979.
In 2017, a positive change of 0.9 percentage points left the participation rate at 65.7 per cent at the end of the year.
This percentage change is equal highest with the 0.9 percentage point increase recorded in 1986.
The participation rate ended at 61.9 per cent that year, which is far lower than 2017. This means that although growth was equally as strong in both years, the year 2017 ended with a far greater proportion of people engaged in the labour force.
Employment as a percentage of population
Experts contacted by Fact Check all noted the importance of the growth in population for assessing the strength of the labour market.
"This indicator tells us the proportion of economically active individuals in a country. It can be an indicator of the health of the economy," Associate Professor Posso told Fact Check..
The employment to population ratio measures the number of employed people aged 15 years and over as a percentage of the civilian population aged 15 years and over.
Fact Check has graphed the percentage point change in the employment to population ratio for each calendar year back to 1979.
In 2017, the ratio increased by one percentage point, from 61 per cent to 62 per cent.
However, there are a number of years with higher growth than 2017. The year 1994 recorded a percentage point increase of 1.4 per cent over the year — from 56.1 per cent the previous year, to 57.5 per cent.
Can governments claim credit for employment growth?
In making his claim, Mr Morrison did not claim credit for "creating" the 403,100 jobs, as politicians sometimes imply.
But his announcement of the jobs figures heavily implicated the Government's economic management as a primary factor.
Associate Professor Chan said that the question of how much credit governments can claim for jobs growth was an ever-present one among academics.
In relation to 2017, he said that Australia "benefited at least partially from the global recovery because the Australian economy is highly connected with the global economy".
Associate Professor Posso called it the "million dollar question", and pointed out that there are a lot of "multiplier effects that stem from government policies".
"For example, if growth is in an export-intensive industry that is exporting goods to China, many would argue that the government had very little to do with it. However, in reality, export-intensive industries require a bunch of friendly policies that allow them to grow."
Professor Barrett said that it was "hard to identify and separate out the impact of one factor — here the Federal Government — on an aggregate outcome such as employment growth".
"It is clear that many other factors affected the employment growth figures than the Federal Government — for example the Reserve Bank of Australia has played an in important role in stabilising the performance of the aggregate economy," Professor Barrett told Fact Check.
"When there is a downturn governments of all persuasions are quick to point out market forces at play (e.g. adverse exchange rate movements, worsening terms of trade, global uncertainty, end of the resources boom) and blame external factors.
In good times, governments like to bask in the sunshine and claim responsibility."
Lead researcher: Matt Martino, Online Editor, RMIT ABC Fact Check
- Scott Morrison, Press conference, January 21, 2018
- Australian Bureau of Statistics, Labour Force Survey, December 2017
- ABS, Time series analysis: the basics, March 2017
- ABS, Feature article - Labour force participation: international comparison, January 2003
- ABS, Australian labour market statistics, employment measures, July 2013