The Average Earnings Index and Average Weekly Earnings

At the start of February 2008 you asked me to review aspects of the experimental series of Average Weekly Earnings produced by the ONS and I now submit the following report. I recognize that the Review has taken longer than initially hoped. This is because the issues involved were complex and it took some time to come to grips with them despite the fact that the documentation made available by your colleagues was of high quality. Nevertheless, I hope the Report will prove useful to ONS in bringing the Average Weekly Earnings series to a state where it can be put forward for assessment as a National Statistic. have in particular devoted a substantial amount of time to explaining the calculation of Average Weekly Earnings and the way in which this differs from the calculation of the Average Earnings Index. Peter Brodie provided substantial input into section 5.3, Harry Duff provided overall background and the material for Chapter 7, Neil Parkin carried out the calculations needed for Chapter 4 and Paul Smith provided substantial input for section 5.2 and Appendix 3.2 and Appendix 4.1. Peter Brodie and Martin Brand worked on issues raised by Appendix 5. 1 Very helpful comments on and corrections to earlier drafts have been provided by these ONS staff, by Colin Mowl and also by Chris Kelly of HM Treasury and Neal Hatch of the Bank of England. Of course responsibility for both the text of this report and any errors and omissions is entirely mine. The terms of reference for this Review are to investigate: 1. whether the ONS's proposed methodology for Average Weekly Earnings (AWE) is fit for purpose for the principal measure of short-term changes in earnings; 2. whether the reconciliation between AWE and the Average Earnings Index (AEI) provides a satisfactory explanation of the statistical and economic differences between the two measures; 3. whether AWE is a better measure of short-term changes in earnings than the AEI; 4. which AWE measure (fixed weights or current weights) should be used in which circumstance; 5. the nature and extent of supporting information required to supplement AWE; 6. whether, after a transitional period, the AEI can be withdrawn; 7. whether the current sample structure is appropriate, particularly with reference to the financial services sector. In interpreting the meaning of "short-term changes in earnings" it has been assumed that these relate to the growth in earnings …