The Indonesian authorities—the Ministry of Finance (MoF), BPS-Statistics Indonesia, and Bank Indonesia (BI)—express their appreciation for the findings presented in the Data Module Report on the Observance of Standards and Codes (ROSC). The mission had made a number of recommendations regarding compilation, dissemination, and methodology for statistics, but it was gratifying to note the mission’s appreciation of the hard work of the authorities to improve the statistical system. These findings represent a “snapshot” as of March 2005, and this document elaborates on the response by the Indonesian authorities to the staff’s recommendations made during the ROSC mission on March 28–April 11, 2005.
Harmonize macroeconomic statistics across all sectors, in particular, by addressing intersectoral discrepancies between monetary statistics and GFS, as well as discrepancies between government finance and BOP statistics in BOP and national accounts.
Efforts have been pursued to build institutional cooperation between BI and the MoF by arranging institutional meetings to harmonize monetary statistics and the GFS post ROSC assessment, and especially to discuss the use of government financing data from BI rather than from the MoF.
II. Comments from the Ministry of Finance
A. Government Finance Statistics
|0.3 Relevance||The MoF is open to feedback from government finance statistics (GFS) users, but no systems are in place to monitor user needs.||Currently, the MoF is open to feedback from GFS users and has made contact persons available to accommodate user needs on the SDDS website. No contact persons are listed on the MoF websites at www.depkeu.go.id and www.diapk.depkeu.go.id. The MoF will arrange for the contact person and also schedule an annual meeting with users.|
|0.4 Other quality management||GFS compilers lack a user perspective on the outputs of the GFS compilation process. No process is in place to monitor the quality of the GFS program, nor to obtain information from users on new and emerging data requirements.||The MoF compiles only fiscal data into the GFS program. No specific unit in the MoF is responsible for monitoring the GFS program activities.|
|1.2 Transparency||No information on terms and conditions under which GFS are compiled and disseminated is made public. Internal government access to GFS prior to their release is not available to the public.||The MoF already has published the information on its websites at www.depkeu.go.id and www.djapk.depkeu.go.id. but because the MoF has just completed its reorganization, the data are not complete yet. GFS data are produced by the Directorate General of Treasury Affairs, but there is no access to modify the data, since all data are subject to audit by the Supreme Audit Board (BPK).|
|2.1 Concept and definitions||The central government is on GFSM 1986 basis. The local government is on a nonstandard basis.||The central government still uses GFSM 1986 and is in the process of converting to GFSM 2001 by 2008.|
At present, the local government still is not in not compliance with either GFSM 1986 nor GFSM 2001. However, local government statistics are now in the development phase. The MoF now is drafting a Government Regulation on Local Government Financial Management in compliance with GFSM 2001, such as in the budget classifications and accounting rules.
|2.2 Scope||The GFS covers only budget transactions.||Besides budget transactions, the GFS also covers some, but not all, extrabudgetary transactions.|
|2.4 Basis of recording||Historical prices are used to value nonfinancial assets. Recording is on a cash basis.||The MoF uses historical prices to value the assets, but the MoF will perform asset revaluations based on market prices. For the time being, MoF transaction recording uses the modified cash basis (cash toward accrual) and plans are under way for the implementation of accrual basis accounting in 2008.|
|3.1 Source data||No data are obtained for government activities outside the budget process.||The government, in this case the MoF, has data on activity outside the budget process.|
|3.5 Revision studies||Regular revision studies are not carried out.||Regular revision studies will be carried out soon.|
|5.2 Metadata accessibility||No published documentation describes the concepts and methodology of GFS in Indonesia, apart from material included on the SDDS website.||The MoF has the GFS concept and methodology in Indonesian version, but it is not published yet due to the need for copyright arrangements.|
The MoF is going to translate the metadata published on the SDDS website.
|New GAS will be implemented as soon as the legal process is complete. At this time, the Law on GAS is still in parliament waiting for approval.|
|The MoF has made some efforts to strengthen the management system in past years, and now it has just finished its organizational structure reform. A treasury unit in the new structure is responsible for accrual basis recording of transactions.|
|The MoF is already moving gradually toward an accrual accounting system.|
|For fiscal year 2005, the MoF has set up a comprehensive register of all sector units by organization and function classification.|
|The amendment of accounting regulations is fully dependent on the legal process for GAS.|
|The compilation and dissemination of GFS for the central government sector can be completed within six months after the reference period, in line with Law No. 17 of 2003 concerning State Finance. However, for local governments, the MoF is considering the development of certain techniques to overcome the time lag problem.|
|Indonesia consists of more than 450 autonomous local governments, some located in remote areas. Moreover, most of them have limited human resources capacity. It is difficult to get timely preliminary data, but the MoF will use estimates as preliminary figures.|
|The central government already has standards for compilation procedures. In addition, the drafting of a government regulation for local government reporting standard is still in progress.|
|The operating statement, statement of sources and use of cash, and (partial) balance sheets are under development and will be implemented for fiscal year 2005.|
III. Comments from BPS-Statistics Indonesia
A. National Accounts
|1.3 Ethical standards||There is no written guidance on ethical standards. However, all staff are regularly reminded of the need to keep data confidential. Also, all civil servants are subject to the application of general ethical standards as set out in the government regulation 30/1980.||A set of ethical standards has been written to guide BPS employees for outlining correct behavior in collecting, processing, analyzing, and disseminating information with reference to the Fundamental Principles of Official Statistics by the United Nations Statistical Commission. All BPS employees are subject to the application of the general ethical standards as set out in government regulations, including all regulations determined by BPS-Indonesia.|
|2.1 Concepts and definitions||The national accounts are broadly in line with the 1968 System of National Accounts (SNA), but are in the process of being updated to the 1993 SNA.||There are two main problems in implementing the 1993 SNA, i.e.: (1) lack of data availability; and (2) low demand for the 1993 SNA because the 1968 SNA is still sufficient for data users. However, BPS insists on gradual implementation of the 1993 SNA—as recommended by the United Nations—by filling in data gaps and educating the data users.|
|2.2 Scope||The published data cover annual and quarterly GDP, from both the production and expenditure approaches, and at current and constant prices. Annual supply and use tables (SUTs) are not produced. The production and asset boundaries generally conform to the 1968 SNA, but some 1993 SNA changes have been implemented. Free zones are covered in the production figures, but not in imports.||By conducting special surveys, BPS is able to compile a five-yearly SUT (detailed input-output (I-O) table); however, BPS is not yet able to produce an annual SUT due to the lack of detailed, complete, and periodical data.|
|Through sourcing of various data from BPS and other institutions, in the near future, BPS will gradually develop concise an annual SUT with limited sectors, depending on availability of data. Sectors will be extended gradually as data become available.|
|BPS adheres to the Special Trade System in recording foreign trade statistics where the free area of Batam is treated as Abroad, so that it is not included. Unfortunately, BPS is not able to estimate free zone import data from the insufficient available data.|
|BPS definitely plans to include figures for free import and export ports from various data sources such as BI. Over time, BPS will assume responsibility for collecting its own data.|
|2.4 Basis for recording||The valuation rules used for recording flows and stocks are generally in accordance with the 1968 SNA. However, all government data are recorded on a cash, rather than accrual, basis. Also, average monthly, rather than daily, exchange rates are used.||In calculating government consumption, BPS relies heavily on data provided by the MoF, where cash basis is applied. When the MoF applies the accrual basis, BPS will calculate the component of government consumption based on the accrual basis.|
Daily exchange rates have been used to replace average monthly exchange rates.
|3.1 Source data||There are annual censuses of production establishments with 20 or more employees. However, there is no survey of nonfinancial services. An economic census of businesses is undertaken every 10 years, but is not updated in the intervening periods. The household budget survey is weak because it does not cover higher-income households.||BPS has, in fact, been collecting data on nonfinancial services, such as transportation, hotel, tourism, and others. In addition, to capture unincorporated small-scale businesses, BPS conducts the Integrated Small-Scale Business Survey (SUSI).|
|BPS is extremely concerned with its existing data performance, therefore, efforts will be pursued for a gradual improvement in the quality of all data collected by BPS, including the SUSI.|
|With its present limited resources, BPS is not yet able to update the business registers adopted from the economic census in the intervening periods. Nevertheless, BPS performs updates in several important sectors.|
|Improvement is expected in the quality of the Household Budget Survey (NESS - Susenas) through expanded coverage of certain categories of respondents, such as the high-income households, beginning in 2005.|
|3.2 Assessment of source data||BPS routinely assesses source data. Coverage of source data is not routinely accessed.||BPS has been holding regular meetings with data producers before estimating the GDP figures. In the near future, BPS will finalize a plan to regularly evaluate all techniques and methodologies of data collection from different sources.|
|3.3 Statistical techniques||Extensive use is made of benchmark data from the five-yearly I-O tables. Constant price techniques and coverage of informal activities could be improved.||By applying SUT, the problem can be gradually solved (see answer to 2.2). The SUSI results eventually will be utilized to improve the estimation of informal activities.|
|3.4 Assessment and validation of intermediate data and statistical outputs||The five-yearly I-O tables remove the statistical discrepancy, but this is not done for intervening years. The I-O tables introduce large revisions to the annual GDP series.||Statistical differences between GDP and the I-O figures will be eliminated through backcasting program. Since the year 2001, the differences have been taken into account in the GDP calculation of subsequent reference periods pending the production of the new I-O table.|
|3.5 Revision studies||Only ad hoc studies of specific activities are undertaken.||BPS will construct a general system for undertaking studies for the revision of published GDP estimates.|
|4.2 Consistency||The quarterly estimates are fully consistent with the annual figures. The current and constant price figures are totally comparable. For the periods following the 2000 I-O tables, the statistical discrepancy has been shown separately in the publications. The policy is only to publish figures from the last I-O year. This means that revisions made to earlier years are not available to users other than BI. There are inconsistencies with BI in imports and exports and, possibly, with the MoF in respect of local government figures, but these have not been investigated.||BPS is compiling a backcasting of the new GDP series (2000-based year) from 1990 to 1999. The results will be published formally by 2006.|
|BPS affirms that it performs reconciliation activities aimed at harmonizing import and export data which are used by BPS and BI.|
|4.3 Revision policy and practice||There is an established revision policy, which is always kept and fully explained to users. Preliminary data are clearly indicated in the publications. The results of the ad hoc revision studies are not published.||In general, the data status will be changed based on incoming data. However, the system of data revision is not yet fully documented, so that the results of the revision studies cannot be published at this time. Therefore, BPS will gradually develop and publish the results of the ad hoc revision policy system.|
|5.1 Data accessibility||The GDP figures are published in a clear manner with different levels of detail, as||The seasonally adjusted quarterly estimates are available, but BPS is not yet disseminating the estimates to avoid misinterpretation among data users.|
|appropriate to the specific publication. An analysis of the developments in the current period is also given. Seasonally adjusted quarterly estimates are not published, even though they are available. An advance release calendar is published. The data are available to all users at the same time via a press release and on the BPS website. Some additional breakdowns can be supplied as long as they do not breach the confidentiality rules.||Therefore, BPS plans to create special programs to educate data users in the interpretation of these data.|
|Update census lists of enterprises continuously with registration of new enterprises and exclusion of nonoperating expenses.||At present, BPS can continuously update the list of the medium- and large-scale manufacturing enterprises, but not for small-scale establishments due to the huge number of business enterprises operating in Indonesia.|
|Introduce comprehensive annual establishment surveys for nonfinancial services industries.||BPS intends to gather data on activities in nonfinancial services by introducing a new annual establishment survey.|
BPS will resume the collection of information on nonfinancial services through the quarterly establishment survey no later than the year 2006/2007.
|Publish annual GDP estimates with a lengthy time series (e.g., 20 years).||BPS has already planned to publish lengthy GDP series data in a wide range of dissemination media. One of these is a special publication titled The 60 Years of Indonesian Independence, which presents the long-time series data for various kinds of statistics, including GDP figures.|
|Develop a set of annual SUTs starting from 2000.||BPS will use the SUT framework in the annual GDP compilation in the year 2007 in order to assure consistency of the GDP estimation with the demand and supply approach.|
|Expedite the conversion to the 1993 SNA.||BPS will gradually apply the 1993 SNA, including all accounts and tables recommended by the United Nations. Therefore, the special agenda will be developed to replace the 1968 SNA with the 1993 SNA in the period 2005 to 2008. BPS thus expects the 1993 SNA to be fully applied by 2008.|
|Publish quarterly seasonally adjusted GDP data.||BPS will publish quarterly seasonally adjusted GDP data by the year 2007; however, prior to the publication, BPS will educate data users on the use of the adjusted GDP in order to avoid misinterpretation of the data.|
IV. Comments from Bank Indonesia
The report is a fair assessment of standards and codes implemented by BI in its compilation of the monetary and balance of payments statistics. The shortcomings highlighted by the IMF ROSC mission are recognized as challenges. Resources permitting, the authorities are committed to make improvements in the near future in conformity with the internationally accepted standards and codes.
With respect to monetary statistics, staff assessments indicate mostly “observed.” This result was only achieved because of the tireless efforts of BI in developing monetary statistics since 2002, with support from technical assistance teams from the IMF’s Statistics Department in September 2002 and June 2003. Notwithstanding the results of the assessments, some improvements are still needed in some areas of monetary statistics.
For balance of payment statistics, the staff assessments indicate, for the most part, “observed” and “largely observed.” A possible explanation for this result is the intensive work undertaken by BI to improve quality and compilation of data in accordance with BPM5.
A. Monetary Statistics
|2.2 Scope||2.2.1 Mutual funds (REKSADANA) that issue deposit-like liabilities are excluded from the scope of monetary statistics.||This issue is currently the concern of the Monetary Statistics Division and has already been scheduled in the blueprint for the expansion of monetary statistics coverage. The work has started through the development of institutional cooperation between BI and the MoF, c.q., Directorate General of Financial Institution (DGFI) as the supervisory agency for nonbank financial institutions in Indonesia (including mutual funds).|
The Memorandum of Understanding (MoU) regarding data collection cooperation between BI and DGFI was issued in mid-2005. Data collection cooperation on finance companies data has begun (2005) and will be followed by insurance companies data (2006), pension funds data (2007), and mutual funds data (2007).
|3.1 Source data||3.1.2 Source data collected on REKSADANA do not meet statistical requirements.||Data collection efforts from the above-mentioned institutions is aimed not only at developing a reporting format that meets both statistical and supervisory requirements, but also in designing an online reporting system from each financial institution to DGFI using the BI extranet server. With this reporting system, data submitted by each institution will be received at the same time by DGFI (for supervisory purposes) and BI (for statistical purposes).|
|3.4 Assessment and validation of intermediate data and statistical outputs||3.4.2 Reported interbank positions are not consistent.||Some interbank positions between BI and commercial banks are acknowledged to be inconsistent in terms of the data sources. These inconsistencies occur due to the time lag of reporting between BI and commercial banks.|
However, as recommended in the report of the monetary and financial statistics mission (May 21-June 4, 2003), BI has already eliminated this interbank inconsistency in the dissemination of monetary statistics by using only data from BI, which are assumed to be more accurate.
|4.2 Consistency||4.2.3 GFS domestic financing data are not reconcilable with the monetary statistics.||Institutional cooperation with the MoF has also been taken by arranging institutional meetings to harmonize monetary statistics and GFS post the ROSC assessment.|
|The development of an online reporting system from mutual funds to DGFI by using the BI extranet server will be initiated in 2007. With this reporting system, data submitted by mutual funds will be received at the same time by DGFI (for supervisory purposes) and BI (for statistical purposes).|
|According to the blueprint for the expanded coverage of monetary statistics, mutual funds data will be included from 2007.|
|BI has already eliminated this interbank inconsistency in the dissemination of monetary statistics by using only data from BI, which are assumed to be more accurate.|
B. Balance of Payments Statistics
|1.2 Transparency||1.2.4 When BI makes changes to BOP methodology or source data, the public is usually informed concurrently when the change is effected.||BI has the same policy of informing users of the publication of major changes in methodology, source data, and statistical techniques, i.e., by giving notes or explanation in the Indonesian Financial Statistics (IFS) whenever such changes take place. No advance notice of such changes is given to the public in the form of articles in bulletins, briefings, or news releases or on the website. In the future, advance notice of such changes will be posted on the BI website and/or the IFS.|
|2.1 Concepts and definitions||2.1.1 Financial accounts do not provide for a separate recording of transactions in assets and liabilities. It is planned to publish these data for 2004 in Q2, 2005.||Financial accounts do record gross transactions in liabilities. The publication of transactions in assets is delayed to mid-2005 due to a lack of adequate source data. To date, the International Transactions Reporting System (ITRS) is the best available data source. However, recent changes in the program used to produce BOP statistics from the ITRS (from Infobeacon to Cognos) has contributed to the delay in this publication.|
|2.2 Scope||2.2.1 Not all resident-nonresident transactions are covered, e.g., goods for processing, goods procured in ports by carriers, repairs on goods, computer services, oil and gas exploration, e-commerce transactions, reinvested earnings, direct investment abroad, etc. It is planned to publish these data for 2004 in Q2, 2005.||In mid-2005, BI plans to publish data on goods for processing, goods procured in ports by carriers, DI abroad, PI assets, OI assets, trade credits, and the oil and gas sector in DI statistics.|
|2.3 Classification/sectorization||2.3.1 The classification of institutional units to their relevant sectors does not completely follow BPM5, e.g., public enterprises are classified under the government instead of other sectors.||Reclassification of public enterprises will be made.|
|2.4 Basis for recording||2.4.1 The principle of market valuation specified in BPM5 is not followed completely. The value of transactions derived from stocks data includes valuation and other changes.||Revision on valuation of debt securities liabilities transactions derived from stock data has been made. The value of transactions now does not include exchange rate changes.|
|2.4.2 Recording of transactions is not completely done on an accrual basis.||The ITRS records the settlement of a transaction; hence, the services data compiled through the system is not recorded when rendered. In this case, the transactions are not recorded on the accrual basis principle. However, for the time being, this represents the best available data, especially in the coverage.|
|2.4.3 Financial account transactions in assets and liabilities are not published separately.||The publication of transactions in assets is delayed to mid-2005 due to a lack of adequate source data. To date, ITRS is the best available data source. However, recent changes in the program used to produce BOP statistics from ITRS (from Infobeacon to Cognos) has contributed to the delay of this publication.|
|3.1 Source data||3.1.3 Results of ITRS are satisfactory for the banks, but the BI segment was not collected. The results are not incorporated in BOP compilation.||Attempts have been made to include the BI segment in ITRS even though the progress is slower than expected. The delay is due to technical problems in converting accounts to ITRS transaction codes.|
|3.3 Statistical techniques||3.3.1 Information from surveys for some respondents needs to be adjusted for missing data. Methods of adjustment need to be improved.||DI survey results for position data are adjusted for missing data. The previous position data are used to adjust the current position.|
|3.3.2 Estimation technique needs to be improved. Data on a gross basis are not collected for some items, e.g., telecommunications services. Shuttle trade is not estimated.||According to regulations, ITRS requires all transactions to be reported on a gross basis. Net transactions should be reported by using a specified code. It is still possible for banks to report transactions on a net basis for the simple reason that this is the best information provided by their customers. Therefore, it is not our intention to collect data on a net basis.|
A study on shuttle trade will be conducted in 2006.
|4.2 Consistency||4.2.1 Substantial magnitude of errors and omissions (E&Os) over the long run. Financial transactions are not consistent with changes in IIP.||We believe that unrecorded financial asset transactions contribute to the substantial E&Os over the long run and the inclusion of those data will leave E&Os purely as unidentified transactions and statistical discrepancy. According to our BOP Blue Print 2004-2006, the financial asset data will be finalized in mid-2005 and incorporated in the BOP table starting with the Ql 2005 data.|
To this time, BI has not reconciled the IIP and BOP, i.e., compile the IIP statistics consisting of the beginning and end-of-period positions, and changes in position reflecting transactions and other changes. However, not all financial transactions are inconsistent with the changes in IIP. For example, transactions in reserves assets, equity securities liabilities, and external debt can be reconciled with changes in IIP. Furthermore, transactions in debt securities liabilities derived from stock data can still be reconciled with changes in IIP after removing other changes.
|4.2.2 Revisions to historical trade data, after the introduction in 2004 of imports and exports online reporting, go back to 2003.||The implementation of an online reporting system in 2004 represents a cornerstone in reconciling trade data with the BPS. Revisions to data following the online reporting system are conducted in cooperation with the Indonesia Trade Customs and BPS, and for the time being have been reconstructed only as far back as 2003. Accordingly, it should be recognized that the effort to make backdated revisions is not solely under BI’s control.|
|4.2.3 Discrepancies with the BPS trade data have not been reconciled.||Although the process of data reconciliation is in place, the acceleration of the process to get rid of the discrepancies very much relies on the BPS. For example, the discrepancy in the import data occurs because BPS does not cover imports from the Bonded Zone Area (including the Batam zone) while BOP covers this data. We have asked BPS to collect the bonded zone imports data for a long time and this year BPS will start to collect the data.|
|Internally, BI is continuing the study of E&Os. In the near term, with the inclusion of the asset side of the financial account, we will find out whether the lack of those data is responsible for the greatest contribution to the E&Os.|
Meanwhile, BI has also initiated the study of undocumented transactions in cooperation with BPS. These transactions, primarily those related to illegal trading, are believed to contribute to the substantial magnitude of E&O.
|In the near future, added explanation of BOP statistics is planned for the IFS. Plans also call for the construction in 2006 of more detailed metadata.|
|BI usually organizes a public information forum before conducting a survey. Business lunches are used as a forum to explain the survey results to the survey respondents and the importance of the survey results in compiling statistics. Formal meetings with business associations are also used to explain BOP statistics and the data that we need from these associations.|
|A study of shuttle trade is scheduled to be conducted in 2006. However, a preliminary discussion has been conducted with BPS for the study of undocumented transactions.|
|In the near future, added explanation for BOP statistics is planned for the IFS. Plans also call for the construction in 23006 of more detailed metadata.|