for this figure, however, is quite vague. It is extremely difficult to pin down the total government expenditure from the fiscal records of the government. The measurement of other concepts encounters similar problems. In this paper, we explore reasons why the fiscal records of the Japanese government are so obscure. We also consider what may be done to generate measures of fiscal activities of the Japanese government that are easily interpretable and usable in assessing their macroeconomic impacts. J. Japan. Int. Econ., June 2002, 16(2) pp. 177–193. University of Pennsylvania, Philadelphia, Pennsylvania 19104, and NBER. © 2002 Elsevier Science (USA).
In the companion paper (this issue), I have discussed statistical problems that
are encountered in using the National Accounts data for analyzing the behavioral
pattern of the household and corporate business sectors in Japan. The data problems
for the private sectors of the National Accounts are serious enough, but they are well
defined and specific within a reasonably well-structured system of accounts, so
that each item can be dealt with individually. Thus, for example, when information
on the amount of employee compensation in the nonfinancial corporate sector
becomes available, this information can be incorporated into the existing system
of accounts, and users can take advantage of this availability immediately.
The problems in the data for the government sector are much more serious,
because these problems originate from the complex and poorly designed administrative
structure of the fiscal operations of the Japanese government. As a result,
the attempt by those responsible for the National Accounts to reconstruct the fiscal
records of the government into a comprehensive and comprehensible system
of accounts in accordance with the SNA93 has been only partially successful. In
this paper, we will focus on the records of government expenditures, leaving the
revenue side of the accounts for another occasion.
At the level of the central government, the government budget is divided into
the general account (Ippan Kaikei), on the one hand, and a collection of 38 special
accounts (Tokubetsu Kaikei) on the other. These accounts are supplemented
by accounts for nine “government related institutions” and a large set of public
enterprises. Government related institutions are public enterprises, wholly
owned by the central government, whose budgets must be submitted to the Diet
for approval. The distinction between these institutions and other public enterprises
wholly owned by the government, such as the Japan Highway Authority
(Nihon Doro Kodan) and the Petroleum Management Authority (Sekiyu Kodan),
appears to be largely administrative and not important from an economic point
of view. Other public enterprises may be more independent of the central government,
and their activities may resemble those of private enterprises. They are
nevertheless still subject to supervision and control by the government at critical
points.
In the introduction to the Fiscal Statistics (Ministry of Finance, 2000a), the
presence of a large number of semi-independent accounts in the national budget
is justified as follows:In principle, it is desirable that the fiscal records of the government should be structured as
a single system, and all revenues and expenditures should be organized in a unified fashion.
However, as the range of the nation’s finances becomes complex and multifaceted, in order to
make the operation of fiscal activities clear, and to improve administrative efficiency, it becomes
necessary to distinguish different types of accounts. Accordingly, the Lawon Fiscal Administration
authorizes the government to establish Special Accounts when (1) the government undertakes a
specific project, (2) the government needs to establish a special fund for some purpose, (3) when
specific revenue is designated to be used for a specific expenditure
