September 2011 Employment: Unemployment Ticking Up

After the employment losses of August, the labour market rebounded slightly in September, adding a marginal 7k jobs (‘000):

Since the labour force expanded by 26k, that meant an uptick in the unemployment rate to nearly 3.3%:

As callous as it sounds, this levels probably more sustainable, as it’s much closer to Malaysia’s long term average.

Of more immediate interest is that structural break in employment that I noted last month. Taken at face value, the jump in the labour force and employment numbers around the end of last year also resulted in some fantastic numbers in the labour force growth numbers (log annual change):

That’s an average of over 8% per annum in log terms, which looks implausible. The average of around 2% seen in 2010 looks closer to the truth. One of my suspicions is that this jump came from the illegal workers amnesty exercise that came at the end of 2010, resulting in a jump in both official employment and the labour force numbers.

I had some hopes that we could get confirmation this month, as another exercise was being carried out. I’m still in the dark however, as the most recent exercise involved deportation rather than legitimising the illegal workers coming forward.

So I did a little regressing, and took out the jump via a dummy variable:

LOG(LF) = α + β*@TREND + γ*D1

…where LF is the labour force and D1 the dummy variable with value zero up to December 2010 and value one thereafter. The results look good:

LOG(LF) = 9.13 + 0.00185*@TREND + 0.056*D1

All the coefficients are significant at the 99% confidence interval. The trend coefficient suggests an annualised increase (in log terms) for the labour force of around 2.245%, which is close to the observable average for 2010.

If you think this exercise is a little out of left field, I more than half suspect that DOS themselves are estimating the underlying “official” numbers (you’re NOT going to be tracking the individual employment record of 19.6 million people, which is the estimated number of people both in and out of the labour force).

So while the labour force increase remains a bit of a mystery, at least we have a plausible estimate for labour force growth. And that’s one important factoid, as the NEM projections are based on a population growth of around 1.3%, yet we have a labour force and employment growth running 80% faster.

Technical Notes:

September 2011 Employment Report from the Department of Statistics (warning: pdf link)

P.S. I wish DOS could at least take the initiative to get the report edited professionally – the English is awful.