1. Muhanad Sammarmuhanad.sammar@scb.se Statistics Sweden 1

2.  Outlet sample  Product sample  Price collection Manually by price collectors  Collection Period 3 weeks 2

3.  Missing prices  Time-consuming  Monotonically for price collectors 3

4.  “(3) Prices used in the HICP should be purchaser prices actually paid by households to purchase individual goods and services in monetary transactions, including any taxes less subsidies on the products, after reductions for discounts for bulk or off-peak purchases from standard prices or charges…” 4

5.  In the National accounts the transaction prices should be registered  Scanner data measures transaction prices  Better basis than manually collected prices for offered products 5

6.  Identifies uniquely an article, package or service  The EAN-number is international  Article Info - supplier, brand, width, depth, weight, labeling, etc 6

7.  EAN-code - not fruit & vegetables, meat & fish  Three deliveries each month  Variables Turnover per store and EAN-code Number of packages sold Price (=Turnover/Number of packages sold) 7

8.  Replace manually collected prices with scanner data  As auxiliary information  Compute index from a census based on all products  For auditing and quality control 8

9.  Deposits for beverages  Package sizes  Delivery problems  Missing values Small volumes Significant price increase 9

10. Matching categories All (49 600 products) SD and MCP. Equal prices 75 SD and MCP. Unequal prices 11 SD, but not MCP 2 Not SD and not MCP 2 MCP, but not SD 10 10

11. Month MCP Difference S.E.* index SD-MCP January100.24 – 0.09 0.10 February100.73 – 0.14 0.11 March101.02 – 0.14 0.10 April100.94 – 0.17 0.14 May101.14 – 0.17 0.16 June101.67 – 0.11 0.20 July102.11 – 0.12 0.18 August102.12 – 0.13 0.19 September102.35 – 0.18 0.16 October102.54 – 0.17 0.15 November102.68 – 0.22 0.17 * Standard error of MCP (manually collected prices) index 11

12.  Prices were erratic for 15 of the products  For 3 products the store had not updated the shelf price with the latest price information  In 2 cases, unclear descriptions of the products 12

13.  Minimize sources of error Improved education of price collectors Increase the number of products Increase the number of outlets  Average price; a week or a three-week period  Transaction price Substitution effect 13

14.  Today's method is not optimal  Scanner data is compatible with EU’s regulatory framework  Scanner data allows adequate price data  In the long run; may be more cost-effective  As part of quality assurance, Statistics Sweden will continue to engage and develop routines that minimize price collector errors  CPI is an important measure and must be computed on accurate and reliable prices 14

15. THANK YOU FOR YOUR ATTENTION 15