"Improving Economic Statistics in order to Improve Economic Policy and Research: (2) IO Tables, SNA (GDP) Estimates, GDP Deflators, and Productivity Indexes" (in Japanese)

This is the second article of my series, "Improving Economic Statistics in order to Improve Economic Policy and Research." It fleshes out some of the issues presented in the first article, Miwa [2014]. Focusing on SNA (GDP) estimates, this study investigates the availability and usefulness of several major series of economic indices: e.g., input-output (IO) tables, GDP deflators, CSPI (and CGPI), and productivity (growth) indexes. In particular, it examines IO tables and SNA base-year annual estimates. It concludes that the nominal GDP, GDP deflator and real GDP, real GDP growth rate, inflation rate, and real productivity growth rate (both for the overall economy and the individual sectors) depend on fatally unclear generation methods (both statistical information based on and estimation methods). Relevant information simply is not disclosed. Two points are important. First, the government has failed to disclose the information necessary to understand the quality of the data on the service sector. With its steady growth, the service sector now constitutes an overwhelming share in the economy. Yet statistical information about this sector has long been insufficient. Unfortunately, this makes the IO tables and SNA estimation problematic, yet the government has failed to provide information necessary for understanding the consequences posed. Second, the government has provided only badly flawed deflators. IO tables and SNA estimation and various statistical materials such as production values are all in nominal values. We need deflators in order to estimate from nominal values real GDP, its growth rate, and the level and growth rate of productivity in individual sectors. Unfortunately, the basic information for estimating appropriate deflators remains unavailable. Relevant information on the estimation methods (therefore, the specific nature and substance) of deflators are unclear and little disclosed, so that it is not easy even to understand the reality of the situation. As a consequence, there remains serious ambiguity in the published real statistical estimates, including real GDP and its growth rates.