Lies, Damn Lies and AFR Editorials on statistics
I write to express my strongest disappointment that a fine paper like the Australian Financial Review resorts to a conspiracy theory to seek to explain my decisions over the past 12 months on changes to the statistical program of the ABS ("Cynical look at the stats on data", AFR 18 December 2008). I am also concerned that the AFR published this opinion without paying me the courtesy of contacting me to get my views on the situation.
The real facts are as follows. In late 2007 I had drawn to the attention of the incoming government, the increasing input cost pressures I was facing. I explained that this would require me to reduce the planned ABS work program for 2008/09 to remain within my appropriation. The situation was subsequently compounded by the government announcement in early 2008 of an additional efficiency dividend to apply to all government departments and agencies in 2008/09.
In these circumstances, and as you rightly point out in your editorial, it was inevitable that some trimming of the most expensive of our statistical collections would be necessary. However these cuts were initiated well before the emergence of the global economic crisis, and the impact on economic indicators such as the labour force survey and the retail trade survey was small compared to reductions that I had to make in the 2011 Population Census program and in some other statistical programs. Also, I was of the view that it would be inappropriate to make cuts to important economic indicators such as the CPI and National Accounts.
Furthermore, I have not been provided any evidence that the minor loss of precision in national employment data or the marginal increase in the volatility in already "noisy" monthly retail sales statistics has made any appreciable difference to policy decisions, and it does the AFR no credit to speculate that it has. My subsequent decision to reinstate the cuts in the Retail Trade survey was made in response to the emerging impacts of the financial crisis supported by the assurance of key policy agencies that in these changed circumstances even a marginal reduction in statistical uncertainty would be of value to them.
The decision to reinstate these cuts exposes the ABS to renewed budget pressures, but in the rapidly changing economic circumstances, I considered it was more important to reinstate the cuts quickly and to deal with the budget issues subsequently.
Brian Pink
Australian Statistician
ABS response to 'Cynical look at the stats on data' (Australian Financial Review 18 December, pg54)