Periodical Revision of L -- Broadly-defined Liquidity
June 10, 2014
Research and Statistics Department
Bank of Japan
The periodical revision of L -- broadly-defined liquidity in the Money Stock Statistics has been conducted.
In general, "L" is revised every three years, in addition to the ad hoc revisions to reflect the recent financial and economic structural changes. Ad hoc revisions are conducted when it is confirmed that the Money Stock Statistics will fully benefit from such revisions. As for the periodical revision, the compilation methods for "L" are reviewed and revised when necessary.
Figures from April 2003 through April 2014 are retroactively revised. This includes revisions following the update of data used for compiling the Money Stock Statistics, such as "International Investment Position of Japan" (end of December 2013). For the revised figures, see the following materials:
- Releases: Money Stock (Preliminary figures for May 2014)
- Long-term time series data:
BOJ Time-Series Data Search at the Bank of Japan website
- Comparison between figures prior to and after revision: Charts 1 to 3 [PDF 115KB]
Details of the revision
1. Government securities
- The method for estimating the amounts outstanding of clearing government bonds, which comprise the majority of those of "Government securities", has been revised.
- Previously, the amounts outstanding of clearing government bonds held by money holders were estimated mainly by using the following two sets of source data: "a survey on the actual holders of clearing government bonds that have been consigned to the participants of the clearing system" for the end of every quarter; and "a summary sheet of the holding balance of clearing government bonds for each business" for the other months. In this method, the portion held by the money holders was calculated by deducting the estimated amount of the portion held by the non-money holders from the source data. This is due to the fact that the available source data includes the amounts outstanding for both money holders, that is, non-financial corporations and local governments (including municipal enterprises), and non-money holders such as financial institutions.
- While this compilation method provided the benefit of timeliness, in its use of monthly data, it also posed a problem. That is, an estimation error of the figures for non-money holders, including financial institutions, can bring about a large impact on the estimated figures for money holders, such as non-financial corporations, since the portion held by the former is much larger than the latter. To address this issue, the revised compilation method looks at the balance sheets of money holders such as non-financial corporations for estimating the amount outstanding of government securities that they hold when possible, while "the survey on the actual holders of clearing government bonds" is used only when otherwise.
- The revised figures are estimated from the same source data as the figures for the Flow of Funds Accounts sectors, which are combined in such a way as to be as close as possible to the concept of money holders in the Money Stock Statistics. These sectors include Households, Private nonfinancial corporations, Public nonfinancial corporations, Local governments, Private nonprofit institutions serving households, Financial auxiliaries, and Finance companies. The figures for government securities in "L" and those in the Flow of Funds Accounts are likely to be consistent, apart from some features such as release schedules, definitions, and compilation methods.
- The figures of "Government securities" from April 2003 -- the beginning of Money Stock Statistics series -- have been revised.
- Note:Some points worth noting when comparing the amounts outstanding of government securities in "L" and those in the Flow of Funds Accounts are: (1) the preliminary figures of "L" are released at the beginning of each following month, while the Flow of Funds Accounts are released three months after the surveyed quarter; (2)"Government securities" in "L" is evaluated at face value and recorded as the average amounts outstanding, while the amounts outstanding of government securities in the Flow of Funds Accounts are evaluated at market value and at end of period; and (3) a part of the amounts outstanding of managed funds on trust accounts in trust banks, which is included in the figures for government securities in the Flow of Funds Accounts, is recorded as the figures for Pecuniary trusts, another component of "L". "Government securities" in this context equals two transaction items in Flow of Funds Accounts: "Central government securities" and "FILP bonds and Treasury discount bills".
2. Investment trusts
- Investment trusts held by money holders are compiled by deducting "the portion held by financial institutions" from "the total of investment trust issued." The total of investment trusts issued is calculated by using the source data of Assets and Liabilities of Domestically Licensed Banks (Trust Accounts) and data from Real Estate Investment Trusts (released by the Investment Trust Association). The portion held by financial institutions is identified from the source data of Assets and Liabilities of Domestically Licensed Banks (Banking Accounts) and the balance sheets of financial institutions.
- In the revised figures, the portion held by financial institutions will include that held by the central bank and tanshi companies (call loan dealers), which was not previously included. This revision is likely to enhance the accuracy of the statistics as a result.
- The figures of "Investment trusts" from April 2003 -- the beginning of Money Stock Statistics series -- have been revised.
Financial Statistics Group, Economic Statistics Division, Research and Statistics Department
E-mail : firstname.lastname@example.org