Trading the News: Non-Farm Payrolls January 2007
Introduction
The purpose of this article is to show how
a new employment data release (ADP
Non-Farm Employment Change) is building respect amongst market
participants as a factor worth considering, if not a predictive
indicator for the widely anticipated Non-Farm
Payrolls report. In the first week of January 2007 we saw
how the ADP release had an effect on trader sentiment and expectations
prior the Non-Farm Payrolls (NFP) and the potential influence
this had on price action.
What is ADP Non-Farm Employment Change?
Automatic Data Processing (ADP) provides employment
solutions for businesses within but not the entire labour market.
The figures they release are directly comparable to the Non-Farm
Payrolls report as they both measure the number of new jobs created
in the previous month, excluding the farming industry. However,
the ADP data is based on a much smaller sample of the labour
market than official figures. The obvious problem with this is
that a small sample is not necessarily an accurate representation
of the entire labour market. This issue seems to undermine ADP’s
claims that the indicator has ‘predictive value’ in
regards to official statistics.
It should be recognised that the ADP data is a relatively new
fixture in the monthly economic
calendar; it was first released in May of 2006. As you would
imagine it takes traders time to attach any importance to a new
report. However, mention of the ADP release is becoming much more
frequent in news reports, newsletters and economic summaries as
the months pass. In fact news traders (those traders who place
trades based on whether data comes out inline, ahead of or worse
than analyst expectations) are reporting an increasing amount of
activity (buying and selling volume)
at data release times as traders begin to attach more importance
to the report.
The January Release
In January 2007, the ADP release fell on Wednesday
the 3rd at 13:15 GMT (08:15
Eastern Time). There are still no official consensus estimates
for this data but official Non-Farm Payrolls figures were expected
at 115k new jobs created. The ADP release actually came out at –40k,
a surprise negative number given that Payrolls were expected to
show an increase in jobs. This sparked a feeling of pessimism towards
Friday’s news. However this was not reflected by immediate
price action. Poor ADP figures (and the expectation that there
will be poor Payrolls figures to follow) can be seen as Dollar
negative, but this was not enough to halt the Dollar’s technically
based rally against the majors going into Friday.
On the morning of Friday the 5th many traders’ inboxes were
filled with newsletter and news service headlines such as “Traders
Braced For Poor Non-Farm Payrolls Data After Negative ADP Figures”.
Whether these headlines correctly encapsulated market sentiment
or served to influence it remains to be seen but the reaction to
Non-Farm Payrolls number would suggest that traders were pleasantly
surprised. Non-Farm Employment Change came in at +167K, beating
analyst expectations of 115K and contradicting the ADP figure released
earlier in the week.
Click here to enlarge GBPUSD 5min
Conclusion
The price action seen on January 5th at the
time of the Payrolls report is not enough on its own to suggest
that the majority of traders are using the ADP figure for any Payrolls
predictions. Due to the fact that the NFP came out better than
expected, there were no nasty revisions to previous months’ data
and the unemployment rate figure came out exactly as expected it
was inevitable that there would be some significant Dollar buying.
However, the insertion of ADP figures into the newsletter of market
forecasters and news providers proves that the data is being taken
more seriously and this trend will gather steam in the months to
come. Traders will want to monitor the ADP employment change closely
to determine whether there is indeed a positively correlating relationship
with NFP’s. At present it would seem that this relationship
is weak at best but that is hardly surprising given the volatile
nature of Non-Farm employment data and the relative infancy of
ADP’s data.
