Understanding Price Variation Across Stores and Supermarket Chains Stores and Supermarket Chains Some Implications for CPI Aggregation Methods Lorraine Ivancic Kevin J. Fox
Aggregation: Study Motivation � Availability of scanner data: highly detailed information on consumer purchases � Price indexes constructed with scanner data volatile � Price indexes constructed with scanner data volatile � Reinsdorf (1999): Need some aggregation to dampen volatility � Many different ways to aggregate � Very little guidance in the literature about appropriate aggregation methods
Aggregation and Unit Values � Aggregation: calculate average price and total quantity across some unit eg. time, items, stores eg. time, items, stores � Aggregation of prices → ‘unit value price’ � Unit value is appropriate when items within the aggregation unit are ���������� � Question: ‘ when is a commodity (group) – that is, a set of economic transactions , sufficiently homogenous to warrant the use of unit values? ’ (Balk, 1998)
Homogeneity � Focus of this work – test for homogeneity across stores within a chain and across supermarket chains. chains. In particular: � If the same item is found in different stores which belong to the same supermarket chain should we consider the item to be homogenous? And… � If the same item is found in a different supermarket chain should we consider the item to be homogenous?
Homogeneity � Economic theory: higher degree of competition and lower degree of item differentiation → equalisation of prices across sellers � But, price dispersion may exist if different sellers offer � But, price dispersion may exist if different sellers offer different range of auxiliary services to consumers eg. different opening hours, range of items, service � Price of the item now reflects a ‘bundle’ of attributes, including item and service attributes � Consumer not only buying item also buying level of auxiliary service � Same item is NOT homogenous across sellers if bundled with different level of service
Defining Homogeneity Definition Definition The same item sold by different sellers is viewed as homogeneous if the price of the item is found to be consistently the same across sellers in the long term.
Data � Scanner data set provided by ABS � Data collected by A.C. Nielsen � Period covered: 02/02/97 – 26/04/98 (65 weeks) � Period covered: 02/02/97 – 26/04/98 (65 weeks) � 110 stores, 4 supermarket chains � Stores account for approx. 80% of supermarket sales in Brisbane � Additional information: brand name, item weight, description, EANAPN (unique identifier for each item) � Item category: coffee � 436,103 weekly observations
Descriptive Statistics No. of Expd. Share No. coffee No. of No. monthly Stores % items sold weekly obs obs in each chain chain Chain A 26 20 89 89,320 22,381 Chain B 9 4 101 29,155 8,063 Chain C 34 35 123 162,765 41,853 Chain D 41 41 88 154,863 38,953 TOTAL 110 100 157 436,103 111,250
Hedonic Regression Model � Hedonic time dummy regression model: � � ∑ ∑ ∑ ∑ = = β β + + β β + + β β + + ε ε �� �� � � � � � � �� � � � � ��� �� = = � � � � Where: P ti = price of item i in period t D t = time dummy variables, 1…t Z tki = k characteristics of item i in period t � WLS used (expenditure shares) � Monthly observations used
Coffee Characteristics � Product Brand (25 brands, DV) � Decaffeinated (DV) � Additional flavouring (DV) � Additional flavouring (DV) � Bonus (DV) � Espresso (DV) � Freeze Dried (DV) � Product weight (20 weights, spline) Piecewise linear continuous function 7 breakpoints allowing for changes in slope � Stores in each supermarket chain (DV)
Testing for homogeneity across stores within a chain � Test hypothesis of equal prices across stores within a chain chain � Separate regressions run for each chain using relevant store DV’s � Tested store DV coefficients across pairs of stores
Results: Aggregation across stores within a chain Chain A Chain B Chain C Chain D Chain A **** 0.0307 0.0374 0.0323 (0.115) (0.115) (0.053) (0.053) (0.115) (0.115) Chain B **** 0.0067 0.0017 (0.711) (0.931) Chain C **** -0.0050 (0.8005) Chain D **** Chain dummy variable coefficient estimates (P,values in brackets)
Results: No aggregation across stores within a chain Chain A Chain B Chain C Chain D Chain A **** 0.0313* 0.0343* 0.0326* (<0.01) (<0.01) (<0.01) (<0.01) (<0.01) (<0.01) Chain B **** 0.0030 0.0013 (0.585) (0.821) Chain C **** -0.0017 (0.615) Chain D **** Chain dummy variable coefficient estimates (P,values in brackets) * Indicates significance at 1% level
Results: Store Differences within a Chain � Chain A (26 stores): No significant price differences found in the 325 store comparisons found in the 325 store comparisons � Chain B (9 stores): 0 out of 36 comparisons � Chain C (34 stores): 8 out of 561comparisons (1.4%) � Chain D (41 stores): 61 out of 861 comparisons (7.0%)
Results: Implications � Seems reasonable to aggregate across stores in chains A, B and C chains A, B and C � Only two stores in Chain D which seem to have different prices consistently – seems reasonable to aggregate across all other stores in D
Homogeneity Based on Hedonics � The same item is considered homogeneous if in stores in Chain A → unit values for same item in stores in Chain A → unit values for same item in stores in Chain A � Unit values for same item in stores in chains B, C, and D, except for two stores in D. � The two Chain D stores enter index number calculation separately
Index Numbers � 4 different indexes used � Laspeyres, Paasche, Fisher, Törnqvist � 4 different types of aggregation used � Aggregation of items across chains � Aggregation based on hedonics � No aggregation of items across chains � No aggregation of item across stores
Index Number Estimates (Base = 1) Homogeneity Homogeneity No No Over Chains Based on Homogeneity Homogeneity Hedonics Over Chains Over Stores Laspeyres 1.263 1.380 1.518 1.564 Paasche 0.985 0.903 0.808 0.793 Fisher 1.115 1.116 1.107 1.114 Törnqvist 1.115 1.118 1.110 1.116 Over 15 month period
Findings � Aggregation implications: � Hedonic regression may be useful to study the issue of homogeneity and associated issue of how to aggregate homogeneity and associated issue of how to aggregate � This work shows fair amount of aggregation may be justified, leading to less volatile indexes with scanner data � Sampling implications: � Sampling from one store in Chain A or B may be enough to obtain representative prices � Not the case for chains C and D
Recommend
More recommend
