I have been looking at the latest participation rate charts. The past few months have seen a dramatic collapse in the participation rate.
This collapse in the participation rate has made our unemployment rate look benign. If we adjust the unemployment rate for this collapse, we can see things are not in as good shape. In effect, we are in far worse shape now than we were at the height of the great recession (or global financial crisis if you prefer the Australian nomenclature). Furthermore, the current trajectory is not good.
Another metric, a favourite for Ross Garnaut, is the hours worked per month per civilian population count. In this chart the post great recession bounce has disappeared.
The participation rate can only soak up exits from the labour market for so long. On this trajectory, it is only a matter of time before the unemployment rate starts moving quickly.
I have been asked how I calculate the constant participation adjusted unemployment rate. I start with an R data frame that merges three ABS excel spreadsheets: the national trend, seasonally adjusted and original series (tables 1 through 3). This data frame is named tsao in the following code. I use the ABS series identifiers as the column names in the data frame (which I have commented).
# A183819T ==> Participation rate ; Persons ; Percent : Trend # A163163R ==> Civilian population ; Persons ; 000 : Original # A183810W ==> Employed - total ; Persons ; 000 : Trend benchmark <- max(tsao$A183819T) tsao$adjLabourForce <- tsao$A163163R * benchmark / 100 tsao$adjTUnemployed <- tsao$adjLabourForce - tsao$A183810W tsao$adjTUERate <- tsao$adjTUnemployed / tsao$adjLabourForce * 100