National Accounts - Quarterlies September 2017

#Reference period# We collected information for this release for the period July–September 2017. #Revisions# We made several revisions to previously published data in this release. The revisions come from three sources: • Quarterly series reconciled to new annual benchmark data • Methodology improvements and updated data sources • Updated indicator data. In general, methodology improvements affected the whole time series; reconciliation to new annual benchmarks primarily caused revisions to recent years (2015 onwards); and updated indicator data caused revisions to recent quarters. We discuss the key revisions in the section below. Quarterly series reconciled to new annual benchmark data In the September quarter of each year, we reconcile quarterly constant price series to new annual national accounts data which have been through the process of supply-use balancing. The reconciliation is essential to maintain the quality of the quarterly series over time. The new annual data comes from National Accounts (Industry Production and Investment): Year ended March 2016 and National Accounts (Income and Expenditure): Year ended March 2017 published on 24 November 2017. Not all industries are reconciled to the annual national accounts data. The methods we use to calculate the annual contribution to GDP by industry vary – some industries use annual national accounts data, while others use other data sources. However, all industries and components are affected by an update in the weights used for chain-linking, which allows the relative size of each lower-level series to be reflected in the industry and component aggregates. More information on the methods used for calculating each industry of GDP, and each component of the expenditure measure of GDP, is in the Quarterly gross domestic product: Sources and methods (fourth edition) . Methodology changes and updated data sources In addition to the annual benchmark updates, we implemented several methodology changes and updated data sources that led to revisions to GDP series. These changes are outlined in the section below. Preview of 2017 national accounts improvements provides more detail about some of the changes made this quarter. Removing fixed weights We regularly make methodology improvements to improve the quality of our quarterly volumes. This year, we introduced chain-linking to replace fixed 2010 weights for lower-level series to some parts of the production and expenditure measures of GDP. Weights are updated annually when producing chain-volume series. But some lower-level weights are not updated annually. Removing fixed weights through introducing chain linking allows more up-to-date current price weights at lower levels of our series to be brought in every year. These better capture shifts in the structure of the economy. Since rebasing quarterly GDP estimates from being expressed in 1996 prices to 2010 prices in 2014, we have been reviewing which industries in the production and expenditure measures of GDP have fixed lower-level weights, and are actively working to remove them. As we get further away from the base-year, the less relevant March 2010 weights become when there are fixed weights. This year, we have removed fixed weights for lower-level series on the production measure of GDP for: • Retail trade and accommodation • Local government and • Unallocated taxes. On the expenditure measure of GDP, we removed lower-level fixed weights for: • Gross fixed capital formation – plant, machinery, and equipment; and transfer costs; and • Imports and exports. The removal of fixed weights caused small revisions within the impacted industries across the time-series. The impacts were minor at aggregate GDP level. We will continue our program of removing lower-level fixed weights, and will implement more changes in coming years. Retail trade survey update We incorporated new data from changes to the Retail Trade Survey (RTS). Methodology changes were made to the RTS to make better use of administrative data, reduce the number of respondents, and implement consistent methodology across economic statistics. Incorporating the new retail trade survey data caused revisions to levels of retail trade activity across the time-series, and revisions to movements from June 2006 quarter onwards. It also caused revisions to household consumption expenditure from June 2016 quarter onwards. Household economic survey incorporation We incorporated new household economic survey (HES) data into our household consumption expenditure estimates. The HES is a direct measure of spending by New Zealand households. Every three years, annual HES data becomes available from a more extensive survey, while the years in between are modelled from other indicators. The new HCE data improved information about the mix of goods and services consumed by households for the March years 2014 to 2016. Agriculture production Statistics incorporation We incorporated data from the Agricultural production statistics: June 2017 (provisional). These stock levels for June 2017 replaced modelled estimates. The incorporation of this data led to revisions to agricultural activity on the production measure of GDP and to agricultural inventories on the expenditure measure of GDP. Local Authority Statistics incorporation: We incorporated changes from the Local Authority Statistics: September 2017 quarter release, which improved the quality of estimates for previous quarters. The improvements included both changes to previous reported data for some of the larger local authorities, and the correction of processing errors. The coherence of individual council data across time, and between the Local authority survey, Local authority census, and the annual reports of the local councils has improved. Updated indicator data We made other revisions due to updated source data. Doing this as updates become available is a normal procedure, as it better reflects what is going on in the economy. Other revision to the production measure of GDP • We received updated respondent information and other source data, which caused revisions to agriculture, forestry and fishing; manufacturing; electricity, gas, water and waste services; telecommunications; finance and insurance; rental, hiring, and real-estate services; and health care and social assistance. Other revisions to the expenditure measure of GDP • Gross fixed capital formation revised due to updated overseas trade data and updated transfer costs data • Household consumption expenditure revised due to updated telecommunications data • Imports and exports revised due to updated overseas merchandise trade and updated overseas trade indexes data. Revisions table The following table shows the previously published and revised quarterly movements for the June 2017 quarter GDP and expenditure on GDP (GDE). Previously published and revised quarterly movements - Percentage change from previous quarter GDP GDE Quarter Previously published Revised Previously published Revised Jun-13 1.1 0.9 -0.2 -0.2 Sep-13 0.4 0.7 0.7 0.8 Dec-13 0.0 0.2 0.5 0.5 Mar-14 1.4 1.4 1.0 1.1 Jun-14 0.8 0.5 0.5 0.5 Sep-14 1.1 1.3 1.1 1.2 Dec-14 1.1 1.5 1.0 1.2 Mar-15 0.1 0.3 0.3 0.7 Jun-15 0.1 0.7 0.8 1.0 Sep-15 0.9 1.0 1.2 1.5 Dec-15 1.1 1.0 0.8 1.1 Mar-16 0.7 1.1 1.1 1.2 Jun-16 0.8 1.1 0.7 1.0 Sep-16 0.7 0.7 0.7 0.6 Dec-16 0.4 0.4 0.3 0.3 Mar-17 0.6 0.8 0.5 0.3 Jun-17 0.8 1.0 1.1 1.4 Source: Stats NZ en-NZ