By David Hargreaves
Lies, damned lies, and er, unemployment statistics.
Without actually canvassing economists' opinions on the subject, as I confess I haven't done, I would suggest that every economists' least favourite statistics have become those measuring unemployment.
Tomorrow then promises to be major headache day in the economics departments around the country as Statistics New Zealand, for the first time, unloads its three main employment measures all at the same time.
Previously the releases were staggered across a couple of days. But as of 10.45am tomorrow the Stats NZ howitzers will blast out all three sets of figures at once, a move seen by economists as a good one that can encourage what Westpac refers to as a more "holistic" assessment of the overall picture.
The most watched statistics at what the economists like to term the "headline" level come from the Household Labour Force (HLFS) survey, which surveys a sample of the population and then extrapolates that out to produce nationwide statistics for employment and unemployment. The other two sets of figures, which I won't concentrate on here, are the quarterly employment survey of expectations and the Labour cost index expectations.
The "V" word
The HLFS has proven in recent times to be what our economists would refer to as "volatile".
As recently as late last year September quarter figures were produced showing a horrific rise in the unemployment rate from 6.8% to 7.3% - a rise that went against all other evidence showing a strengthening economy. That rise was subsequently reversed for the December quarter and then in March (when the long-suffering economists were on average tentatively suggesting another 6.8% figure) another mega-surprise, a glistening 6.2% figure was produced.
As for the June figures, the market average pick among the economists is for a 6.3% unemployment rate, though as most seem to freely concede, the actual figure produced might not be anything like that.
Westpac senior economist Felix Delbruck's going for the market average pick of 6.3%, a "modest uptick" in the rate after the March quarter plunge.
"Most labour market indicators held up or improved over the June quarter. Unfortunately, when forecasting the Household Labour Force Survey things are never quite that simple," he said.
It's payback time
"On the one hand, there are good reasons to expect some statistical ‘payback’ from a very strong March quarter outturn. The 3.2% jump in hours worked was almost certainly too good to be true.
"And our own Westpac McDermott Miller employment confidence would be consistent with an unemployment rate somewhat north of March’s 6.2%. The March quarter numbers were also flattered by the 2013 Census, which led to more than 7,500 census collectors and other staff being hired in February and March. That’s 0.3% of the labour force – though most of these jobs only lasted for about 6 weeks, so the impact will have been smaller over the quarter as a whole."
However, Delbruck said there are also "upside risks" to the potential figures.
"The HLFS’s estimates of construction sector employment have been suspiciously low. Since March 2011, the HLFS has recorded an increase in construction sector jobs of just 7,000 (4.3%), compared to 12,000 (11%) in the Quarterly Employment Survey, and a 13% rise in construction sector GDP.
"That gap could start shrinking as construction picks up outside Canterbury, or workers move out of temporary accommodation, which the HLFS doesn’t capture. Employment could also get a boost from a rise in labour force participation. Labour force participation among teenage students plunged in the December quarter, and only recovered modestly in March. That could reflect new benefits for teenagers in education and training introduced late last year, but it’s too soon to tell."
Weighing the risks
Delbruck said that "weighing up the risks" Westpac's assumed a small rise in the unemployment rate (to 6.3%); a pull-back in hours worked; and a further modest increase in employment and labour force participation.
ANZ's economists said picking HLFS unemployment rates "has been a lottery of late" given the high volatility in both employment and the participation rate.
"Given population growth, with an unchanged participation rate an employment outturn of less than 0.3% would push up the unemployment rate. However, we predict the labour force participation rate will tick down, meaning our -0.2% employment pick is forecast to result in only a 0.1 increase in the unemployment rate, to 6.3%," they said.
Given recent volatility in the HLFS, an "outlier" result was likely to be treated with scepticism, they said.
"Moderate wage inflation [shown in the labour cost index] is expected to keep [Official Cash Rate] moves off the radar in 2013. However, the RBNZ would not take too kindly to signs of a broadening front in wage pressure emerging. This would up the ante to 2014 rate hikes."
Anything is possible
BNZ economist Doug Steel said that "once again" the BNZ economists have the feeling that "anything is possible" with the HLFS.
"We just hope it holds together in Q2. We anticipate +0.2% quarterly employment growth which, combined with a 67.7% participation rate, would see the unemployment rate move down a tick to 6.1%," he said.
"What the unemployment rate actually prints at will be as much dependent on the participation rate – the change in which has done its best to resemble white noise of late – as an employment outcome.
"Despite positive employment indicators for Q2, from the likes of the PMI and PSI, Q1’s massive +1.7% lift in employment is a hard act to follow. Some technical payback looms as a risk. It will be worth cross referencing the HLFS figures with the Quarterly Employment Survey’s gauges on filled jobs, paid hours and full-time equivalent employees.
"But, barring a very negative outcome across the range of indicators, it will be difficult to get too downbeat on the labour market given employment intentions have leapt to multi-year highs."
Steel said it should be noted that the RBNZ forecast the second-quarter unemployment rate at 6.3%, "so there is some scope for unwind from Q1’s strength without affecting policy thinking".
ASB chief economist Nick Tuffley's going for a 6.4% unemployment rate.
He pointed out that the first quarter of the year is normally a weak one for employment followed by stronger figures in the second quarter. But this year there was actually a strong rebound in employment in the first quarter.
Likely to look weak
"This time around, though, the seasonal weakness in Q1 did not materialise (hence the exceptionally strong seasonally-adjusted result). Without that seasonal effect, the change from Q1 to Q2 is unlikely to be as pronounced as is usual, and so after seasonal adjustment the Q2 result is likely to look weak (i.e. a higher unemployment rate)," he said.
"For these somewhat arcane reasons, a moderately weak result shouldn’t be interpreted as a signal of deteriorating labour market conditions.
"We expect a steady improvement in employment conditions over the rest of the year, underpinned by increased activity in Canterbury in particular. Business surveys suggest a gradual improvement in hiring intentions over recent months."
So, there you have it. The figures tomorrow could be anything. But the overall trend is looking good for lower unemployment figures through the course of the year.