Since June 1950, the Retail Price Survey has been conducted with the specific aim of compiling the Designated Statistics No. 35 in accordance with the Statistics Law in order to gather basic data for the Consumer Price Index (CPI) and other source materials concerning prices. Specifically, retail prices of articles of commerce that are important to peoples' livelihoods, service fees and house/room rents are investigated monthly on a nationwide scale at retail stores, service providers' business facilities, relevant organizations and in households.

  We conduct the Retail Price Survey by dividing survey items into broadly the following three types of survey:

• Price Survey: We investigate the prices of about 500 items of commerce or service fees at approximately 28,000 retail stores and establishments in 167 municipalities (cities, towns and villages) all over Japan.
• House/Room Rents Survey: We check monthly house/room rents paid by about 25,000 households in 167 municipalities all over Japan.
• Hotel Charges Survey: We examine average hotel charges per adult person for a stay of one weekday night and on the day before a holiday at about 530 Japanese-style inns (ryokan) and hotels in 101 municipalities all over Japan.

  The survey results are used as basic data for the CPI calculation. In addition, these data are used to investigate price differences between Japan and other countries for major consumer products and services. Local public organizations also use the results to compile regional CPIs.

  In the Statistics Law, based on which the Retail Price Survey is conducted, there are provisions under which you are obliged to provide the information requested in the Survey as well as penal provisions that will be applicable if you have failed to comply with such request.

  These provisions notwithstanding, the Retail Price Survey can achieve its purpose only if we have your support and understanding. We hope you will cooperate with us in preparing accurate statistics.

*Please refer to the Statistics Act for the provisions under which you are obliged to provide the information requested in the Retail Price Survey.

  The Statistics Bureau of the Ministry of Internal Affairs and Communications draws up a basic plan for the Retail Price Survey and implements it through prefectural governments. The flow of the Survey is as shown below:
   (1) The Statistics Bureau holds a briefing for prefectural staff.
   (2) The prefectural staff offer guidance to the survey field workers (enumerators) on the procedures necessary to conduct an accurate investigation.
   (3) The enumerators receive the data needed for the survey from the Statistics Bureau on their PDAs (personal digital assistants).
   (4) After receiving the data, the enumerators ask questions about prices and rents at retail stores and households and input the data on their PDAs.
   (5) After investigation, the enumerators send the survey date to the Statistics Bureau. 

  The Retail Price Survey is broadly divided into a Price Survey, a House/Room Rent Survey and a Hotel Charges Survey.
   The Price Survey and House/Room Rent Survey are conducted in 167 municipalities across Japan by setting up price survey districts, in which the prices of articles of commerce and service fees are monitored in each municipality. House/Room Rent Survey districts are also set up.
   A specified number of retail stores are selected for each item, based on sales volumes or the number of employees in each survey district, to investigate prices of articles of commerce as well as service fees.
   In addition, a specified number of retail stores that have a greater number of customers in the survey municipalities are chosen when investigating certain items.
   In the House/Room Rent Survey, the competent agencies in the survey municipalities are chosen for the survey on public housing facilities. All houses owned by private sectors in the survey municipalities are chosen for the survey on private rental houses.
   In the Hotel Charges Survey, a specified number of representative inns and hotels in each of the 101 survey municipalities are chosen for survey purposes.

  The enumerators are chosen from among the general public, taking the following requirements into consideration. They are appointed as special local public servants by prefectural governors.

• Those who can prepare the relevant documents properly and perform the necessary office work.
• Those who are, in principle, 20 years of age or older.
• Those who can be entrusted with sensitive information.
• Those who are not directly involved in the election of public office holders.
• Those who are not directly associated with the tax office or police.

   For the retail prices of major items by cities (cities with prefectural government and cities with population of 150,000 or more), in principle, we make the survey results available to the public on the Friday of the week containing 19th day in the following month after the survey.
   For the retail prices of major items for Ku-area of Tokyo and nationwide uniform prices or charges, in principle, we make the survey results available to the public on the Friday of the week containing 26th day in the survey month.
   Moreover, we make the survey results of the annual average prices for all survey municipalities available to the public along with the March results of the retail prices of major items by cities (cities with prefectural government and cities with population of 150,000 or more) every year.

  The Retail Price Survey is conducted according to the Statistics Law and other laws and regulations.
   Those who are involved in the Survey (staff of national and regional public organizations, supervisers and enumerators) are under an obligation to observe the confidentiality of private information that they have obtained in the course of the Survey. Moreover, using the survey records for purposes other than the compilation of statistics is strictly prohibited.
   The information provided by respondents to the Survey is input into Personal Digital Assistants (PDAs) and sent to the Statistics Bureau of the Ministry of Internal Affairs and Communications. Operation of the PDAs and data transmission from these devices is strictly controlled to prevent the dissemination of confidential data.

  When selecting Retail Price Survey subjects, we set up a specified number of survey districts in each municipality, based on the size of the population. Then, we choose representative retail stores (1-3 stores) that have greater sales volumes for each item in each survey district and investigate the retail prices of these stores every month. Moreover, we review, from time to time, the lineup of representative retail stores that have greater sales volumes of each item.
   Therefore, if a discount store in a survey district is considered to be a representative retail stores that has greater sales volumes, it will be included in the survey.
   Recently we have reviewed the method of selection in order to respond more effectively to changes in the purchasing behavior of consumers.

  Bargain sales (articles offered at special discount prices) are, in principle, not covered in the Retail Price Survey. Therefore, if the survey date falls on a day on which bargain sales are being held, we investigate the prices on the nearest non-sale day to the survey date. If bargain sales continue for eight days or longer, we also investigate the prices during such sales.
   The data obtained in the Retail Price Survey are used in the calculation of the Consumer Price Index (CPI). The CPI is designed to capture monthly changes in the price of each representative survey item. Therefore, a short-term bargain sale which happens to take place on a survey date cannot be considered as indicative of the representative prices of relevant items for that month. If such prices were adopted in the Survey, it would be difficult to capture general price trends in the CPI. Since bargain sales are normally limited to specific product brands (goods and services), the usefulness of the investigation into bargain prices of survey items would be questionable from the viewpoint of identifying the true price of such an item.
   Taking into consideration the recent situation in which short-term bargain sales are held routinely, we grasp the actual condition of bargain sale prices in National Survey of Prices conducted every five years. Comparing the result of the 2002 survey with that of the 1997 one, there is no big difference in the movement of change between bargain prices to usual prices, although price level of bargain sales is lower.

  The items covered by the Retail Price Survey are selected according to the results of the Family Income and Expenditure Survey. If household expenditure on a particular item accounts for one-ten thousandth or more of the total household expenditure, it is included in the survey items. Moreover, these items are specified by the Director-General of the Statistics Bureau and investigated in accordance with the so-called ?Basic Specifications?. However, in municipalities in which basic specifications are not distributed in sufficient quantities, alternative items that are distributed in larger quantities are designated as ?Local Specifications?, reflecting the actual market conditions.

• Basic Specifications
     The quality, performance and characteristics (features) are specified for each item and these specifications are established so that the prescribed articles of commerce can be investigated in a uniform way on a nationwide scale. This is designed to accurately monitor genuine changes in prices and exclude such factors as differences in quality and functionality of the survey items.
     If not processed correctly, it will become unclear whether prices have changed as a result of an increase or decrease, or for different reasons.
     The basic specifications are selected by the Director-General of the Statistics Bureau of the Ministry of Internal Affairs and Communications. (Refer to "14. How are the basic specifications established?")
• Local Specifications
     If special local products are consumed on an extensive scale due to differences in climate or traditional dietary culture, those items with the most similar qualities, standards and capacities to the basic specifications that are distributed in large quantities are designated as local specifications in each municipality so that the Survey can be conducted continually.
     A local specification item will be changed if the quantity being distributed drops significantly and as a consequence, it becomes inappropriate as a survey item. Moreover, a local specification item will be discontinued if a survey of the relevant basic specification item becomes possible.

  A total of approximately 500 items are covered by the Retail Price Survey. Based on the Family Income and Expenditure Survey, those items for which household expenditure accounts for one-ten thousandth or more of the total household expenditure are included as survey items. However, if all items for which household expenditure accounts for one-ten thousandth or more of the total household expenditure were to be examined in detail, the number of items would become too large. Moreover, sorting the items by product type, we find that we do not need to check all items to keep track of changes in price of each product group. For example, we investigate 16 items in the fruit category. These 16 items represent more than 80% of the fruit category. Even if we increased the number of survey items, it would only add a number of items that are not important to the results. In this sense, we believe we can keep track of the change in price of the entire fruit category with sufficient accuracy by investigating these 16 items. (Note: the percentage of expenditure on the fruits that are included in the survey items to the total household expenditures on the entire fruit category is referred to as the extent to which these represent the fruit category.)

  Certain survey categories consist of several to several dozen goods. Some are purchased by consumers in large quantities and others in small quantities. As a rule, those goods that are purchased by consumers in the greatest quantities (those goods that show higher percentages of consumption in households) are selected for the Retail Price Survey.
   For example, electric refrigerators are made and sold by several major manufacturers, including refrigerators with and without a freezer. Moreover, there are various refrigerators such as those having different capacities and having two doors to six doors. When we select basic specifications, we take the following factors into consideration:

1. What is the market share of each item?
2. Which items are distributed in large numbers nationally?
3. What specification items represent the price movement of the product category (electric refrigerator)?
4. Which specification items can be investigated continually?

  We determine the survey specification items by taking the manufacturers information, industry data, expert opinions and market distribution conditions into account.

  Some Private brand (PB) commodities that conform to the prescribed basic specifications established by considering their representativeness, marketability, continuity and other factors are surveyed. The result of the Retail Price Survey is used in calculating the Consumer Price Indices. Since these indices are intended to be used for grasping changes in the prices of commodities having the same quality, it is necessary to survey commodities having the same quality continuously in the Retail Price Survey. Therefore, we have established the basic specifications of each item by prescribing quality, performance, standard, capacity, characteristics, etc. so that the survey may be conducted consistently and continuously throughout the country. In conducting the survey, we select commodities that conform to the prescribed basic specifications and are most popular among consumers and have the most stable circulation in the market, at each outlet. Accordingly, any PB commodity falling under this category will be subject to the survey.
  Among the various items of commodities, however, there are some cases where a particular commodity (of a particular trademark) is designated as basic specification of an item. This happens in such a case where, with regard to that item of commodity, a particular trademark occupies high market share or there are no other commodities of the same quality and/or the same standard, etc. So far as such items are concerned, since PB commodities generally occupy only limited market shares and have difficulty in maintaining the representativeness of the items at the present stage, they cannot be subject to the survey.
  (For details, refer to "12. What are basic specifications and local specifications?", "14. How are the basic specifications established?" and the List of Items.)
  PB commodities : Those for which leading supermarkets or similar outlets independently plan, develop and attach their own brands for sale.

  In the Retail Price Survey, a preset number of specification items that are commonly distributed all over Japan are selected for each survey item and their prices are investigated continually, in principle, in a total of 167 municipalities with the aim of accurately monitoring the monthly price movements of these specification items.
   However, there are times when these specification items are not distributed in some survey municipalities. In this instance, we select a set of specification items that are unique to these municipalities as local specifications and keep track of their monthly price movement. Since we are investigating, in these municipalities, specification items that are not commonly distributed on a nationwide scale, the national average price of each item will be meaningless, even if we calculate it, because it includes local specifications and the user cannot determine what constitutes an average price.
   Therefore, such a national average price is not calculated in the Retail Price Survey.