Page 407 - Invited Paper Session (IPS) - Volume 1

IPS173 Yuki U.
investment, and direct investment income) will be answered by an expanded
number of enterprises, including non-exporting enterprises.
These revisions allow the Bank to more accurately capture corporate
behaviour and business environment surrounding enterprises in Japan,
enabling users of the Tankan to analyze the impact of expanding overseas
business activities on changes in domestic business activities in a more
6
accurate and timely manner.

3. Further improvements
To reflect changes in the business trends of enterprises in a timely manner,
the Bank considers further improvements to its survey framework in addition
to revisions to survey items. Currently, in order to conduct the survey as
frequently as practicable, the Bank uses a sample survey framework for the
Tankan, extracting sample enterprises from the population. This is done by
fixing a population at a certain point of time given the technical limitation that
population data is only updated every 4 years (Chart 6), while selected sample
enterprises based on the population are basically fixed for that period.
7
However, the population data Chart 6 : Population size
number of enterprises in thousands
changes over time from the time when 225
fixing the population at a specific time,
making estimated figures based on 220 Fixing population

selected enterprises of the Tankan less 215
representative of the population. In 210
fact, a fairly large portion of the
samples change, every time the Bank 205
updates population data and reselects 200
samples about every 4 years.
The Bank is considering in the near
future, conducting sample stratification and extracting sample enterprises
more frequently than the current 4 yearly basis. This is in line with the
Government of Japan’s work on statistical revisions to more accurately reflect
economic structures in a timely manner.

6 For every survey, the Tankan collects responses from enterprises during about one month
survey period, and releases results on the day after the end of the survey period.
7 The Bank examines statistical accuracy regularly (once a year, in principle) and if necessary,
adds new sample enterprises, to prevent the lowering of statistical accuracy caused by a
decrease in the number of sample enterprises due to bankruptcies, mergers, and other factors.
The recent examination showed that the statistical accuracy after the December 2018 survey
satisfied the criteria, such as standard error ratio of sales within the criteria 3% or less for
manufacturing and 5% or less for nonmanufacturing for each enterprise size. For details, see
"Regular Examination of the Statistical Accuracy of the Tankan" (February 2019).
396 | I S I W S C 2 0 1 9