The Use Of Scanner Data In The CPI

The consumer price index is an indicator to measure the trend and degree of changes in prices of goods and services purchased by residents in a certain period of time. The indicator is widely used in the following aspects:to measure inflation and deflation, and provide a basis for decision making of macro-control policies; to provide a reduction factor in the GDP accounting; and to provide a basis for indexation adjustment for currency flows such as staff wages, rent, taxes or interest and anything like these. Although the importance of the indicator is there for all to see, that a series of questions like whether the CPI can truly reflect the price level, inflation, the purchasing power of money, real wages, and then as references to the relevant decisions of the government departments is questionable. Scanner data, however, provides a possible solution to these problems; besides, up to now, the countries that have a sound statistical system in the international society have begun to put the use of scanner data in the CPI into practice, but the domestic research status is nearly in the blank in this area. Therefore, in order to improve the quality of CPI, provide better services for decision-makers, and also to coincide with the feeling of residents better, as well as make the CPI in China have the international comparability, this article attempts to put forward the strategies of the use of scanner data in compiling price indices in China based on the attempt to the results of theoretical research and the experience of practical preparation in such countries like Norway, the United States, Netherlands and Sweden.The research idea of this article is to take a series of CPI bias under the traditional data collection as the starting point, leading to a demand for the technology of the use of scanner data in CPI; however, considering the nearly blank state in China, the article turns to analyze the relatively mature countries in the use of scanner data, aiming at providing reference for China to improve the quality of CPI.The thesis consists of five parts. Chapter1is introduction. Chapter2is traditional data collection methods. Chapter3is theories about scanner data. Chapter4is experience of using scanner data in typical countries. Chapter5is China’s reference.Chapter1is an introduction. This chapter describes the relevant background and the practical significance of the use of scanner data in compiling price indices firstly; then, analyzes the current research progress on this issue of foreigners as well as domestic respectively; finally, reveals the structural arrangements and the innovations as well as the inadequacies of the paper.Chapter2introduces the traditional data sources and makes comments about them. The initial data using in the CPI is prices of commodities and expenditure weights. The chapter begins with an overview of the "Consumer Price Index Manual: Theory and Practice (2004)", which represents the general regulations on data sources. Then it summarizes China’ data sources in the two areas as mentioned above. Finally, the chapter has a discussion about a series of deviations under the traditional data collection methods, namely item substitution bias, outlet substitution bias, new goods bias, weight bias, quality adjustment bias and so on, and then puts forward a few measures to address these deviations.Chapter3describes some basic issues about the scanner data. Firstly, it introduces the concept of the scanner data. Then it analyzes the potential advantages and disadvantages using scanner data: the advantages are that scanner data can offer guidelines for data suppliers in decision making and can improve the quality of the calculation of CPI; the disadvantages are that high acquisition costs of scanner data, statistical agencies’low control of data supply, the accuracy of themselves, the fuzzy concepts related to scanner data.Chapter4makes a detailed introduction about the experience from representative countries internationally on the use of scanner data to improve the quality of CPI. First, Statistics Norway takes food and non-alcoholic beverages for example, and has researched the following issues in the framework of scanner data: the choice of items and outlet sample, the basic sub-index formulas and the solution to the drift problem with seasonal products in the calculation of scanner indexes. Second, the U.S. Bureau of Labor Statistics (BLS) has computed a scanner price index of breakfast cereal in the New York consolidated metropolitan statistical area. Third, Statistics Netherlands computes scanner indexes of the following categories of goods:food and non-alcoholic beverages, wine, beer, tools and equipment for house and garden, goods and services for routine household maintenance, not reimbursable medical and pharmaceutical products, pets, pet foods and products for pets, appliances, articles and products for personal care. Finally, describes the methods to use scanner data in CPI in Sweden from the following three aspects:the status quo of everyday commodities in Sweden, the compilation method of scanner data, and the comparison between scanner data and manually collected price data for the CPI.Chapter5envisions a way in which China can use the scanner data into the preparation of CPI program. Firstly, it reveals what type of goods can be collected by the way of scanner data. Then it proposes two possible ways for China’statistics agency to attain scanner data. Again, from the two aspects of the basic indexes and higher-level aggregation proposes the possible index formulas used. Finally, put forward two ways to use the scanner data to make quality adjustment.The innovations of the thesis are reflected in:(1) summarizes the experience of using scanner data in CPI in such countries like Norway, the United States, the Netherlands and Sweden systematically;(2) puts forward the ways to get scanner data in China;(3) brings forward index formulas applied in the basic aggregation and higher aggregation after using scanner data;(4) raises two methods conducting the quality adjustment in CPI using scanner data.The inadequacies of the article is embodied in:(1) because the technology of using scanner data in the CPI is not yet mature, and scholars do research based on the actual situation of their own country, resulting in less common problems in their study, so it is difficult to make a clearer comparative analysis when analyzing the typical countries;(2) when the paper sets about putting the experience into practice specifically in China, in consideration of lacking of scanner data from supermarkets, the study stays at the theoretical level, having not established an experimental scanner price index yet.