The quarterly employment survey for the three months ended June, out this morning, will take on unusual importance for financial markets with the measure of average weekly pay likely to be soft again.
The survey gave the first hint that something odd was happening with the labour market last quarter when it reported 1% growth in full-time equivalent employment and 1.1% growth in hours worked - both seasonally adjusted.
Westpac chief economist Brendan O'Donovan said the Statistics New Zealand survey was a "simple average" of the wages being made to workers.
Much of the growth in employment was from low-skilled workers and young people and their re-entry into the workforce tended to depress the average wage.
Also due out this morning was the labour cost index which was expected to show another quarter of slow wage inflation, he said.
"Although unemployment is falling, employers still have the upper hand and June tends to be one of the softer quarters for wage growth. The upturn in wage inflation is due later this year, by our reckoning."
On Thursday, the household labour force survey might reveal the truth behind the March quarter's fall in unemployment and whether or not it was an aberration, Mr O'Donovan said.
The last quarter's data was one of the most surprising pieces of economic news ever seen.
"All and sundry were expecting only the slightest growth in employment and the unemployment rate was tipped to rise above 7.3%. Instead, the survey suggested that employment had rocketed ahead by 1% and unemployment had fallen to 6%."
The details were just as strong as the headline, he said.
Employment growth was entirely full-time.
Hours worked rose 1.7% and underemployment fell.
The labour market picture was seemingly so inconsistent with the slow emergence from recession suggested by other data that Mr O'Donovan began to question its validity.
When the June data is released on Thursday there might be some clarity on whether the labour market is really improving at the rate suggested in March.
Westpac was forecasting June employment growth was 0.4%.
If the participation rate stayed flat, as expected, that would see unemployment correct to 6.2%.
"We'd expect full-time work to account for most of the new jobs and we'd expect to see a further lengthening of the average work week as some employers offer more overtime."
The risks to the forecasts depended on the underlying cause of the March data.
One possibility was that the household labour force survey led economists astray through sampling errors or problems with seasonal adjustments.
The other possibility was that the forecasters got it badly wrong while the survey correctly reported developments, Mr O'Donovan said.
Forecasters consistently overstated the likely employment decline during the recession and might now be understating employment growth in recovery, he said.
"Perhaps employers are finding it easier to find workers now whereas employment growth was hamstrung by a shortage of workers last decade."
The 90-day probation rules might have removed some risk for employers, encouraging them to employ people despite the uncertain environment.
Markets would get a major shock if the unemployment rate came in below forecasts while there would be little reaction if the number printed as high as 6.4%, Mr O'Donovan said.
