Australia
Report on the Observance of Standards and Codes-Data Module

Australia’s statistics are of a high quality. Based on the review of its statistical practices, a set of recommendations is presented designed to increase its adherence to internationally accepted statistical practices. The aim of the review is to enhance the usefulness of Australia’s statistics in terms of cross-cutting recommendations, making national accounts series available; a national consumer price index; producer price index; and government finance statistics to include a breakdown of stocks and flows of financial assets and liabilities, monetary statistics, balance of payments, and international investment position statistics.

Abstract

Australia’s statistics are of a high quality. Based on the review of its statistical practices, a set of recommendations is presented designed to increase its adherence to internationally accepted statistical practices. The aim of the review is to enhance the usefulness of Australia’s statistics in terms of cross-cutting recommendations, making national accounts series available; a national consumer price index; producer price index; and government finance statistics to include a breakdown of stocks and flows of financial assets and liabilities, monetary statistics, balance of payments, and international investment position statistics.

I. Overall Assessment

1. Australia’s statistics are generally of a high quality. The Australian Bureau of Statistics (ABS) has been among the pace-setters in the development and adoption of cutting-edge methodologies. The institutional framework for the provision of statistics is long standing, with the Census and Statistics Act, 1905 providing the legal basis for statistical compilation and the Australian Bureau of Statistics Act, 1975 establishing the institutional arrangements. The ABS is the central statistical agency. Its responsibilities include the compilation and dissemination of national accounts, the consumer price index (CPI), the producer price index (PPI), government finance statistics (GFS), and balance of payments and international investment position statistics (BOP/IIP). Commonwealth financial statements are compiled by the Department of Finance and Deregulation (DoFD), while the Reserve Bank of Australia (RBA) is responsible for monetary statistics.

2. Australia has been a subscriber to the Special Data Dissemination Standard (SDDS) since April 1996, and started posting its metadata on the Dissemination Standards Bulletin Board (DSBB) in September 1998, meeting the SDDS specifications in July 2001. Australia is in observance of the SDDS, meeting the specifications for coverage, periodicity, timeliness, and the dissemination of advance release calendars.

3. Australia is availing itself of flexibility options for:

  • The periodicity of the production index data category (Australia compiles and disseminates quarterly data and not monthly data as required);

  • The timeliness of the production index category (Australia disseminates the index no later than one quarter after the reference quarter and not within six weeks as required);

  • The periodicity of the consumer and producer prices data categories (Australia compiles and disseminates quarterly data and not monthly data as required).

4. Australia is also availing itself of a targeted flexibility option for the timeliness of central government operations, where a three-month lag is allowed for the last month of the old fiscal year and a two-month lag for the first month of the new fiscal year, provided that quarterly general government data are disseminated with a one-quarter lag.

5. As Australia has migrated to the newest versions of the statistical manuals—the System of National Accounts, 2008 (2008 SNA), the sixth edition of the Balance of Payments and International Investment Position Manual (BPM6), as well as the more established Government Finance Statistics Manual, 2001 (GFSM 2001), and the Monetary and Financial Statistics Manual, 2000 (MFSM 2000)—this Data Module of the Report on the Observance of Standards and Codes (Data ROSC) assesses Australia against the prescriptions of these methodologies. As a result, comparisons of this Data ROSC against such exercises for other countries in the past need to be treated with caution. The 2003 Data Quality Assessment Framework (DQAF), which underpins the Data ROSC, permits assessment against the prior international methodology in several of these topical areas. Thus Australia has arguably set itself a higher bar for this assessment than have other countries. On the other hand, as additional countries migrate to the new methodologies, this Data ROSC for Australia may come to be the new benchmark.

6. The Data ROSC assessment is taking place at a time when the ABS in particular is facing challenges. The migration to the new methodologies has been a multi-year undertaking, and has been largely successfully achieved. It has involved substantial efforts to introduce fundamental changes across much of the institution’s macroeconomic statistical activities. All Commonwealth agencies are subject to a 1.5 percent annual efficiency dividend and the ABS is no different. However, in terms of its funding relative to other agencies, the ABS’ position has improved as it is exempt from the additional whole-of-government departmental efficiencies savings (which apply to most other agencies) over the four years FY 2010 to FY 2013; the ABS has also suffered an unusual level of loss of expertise, with the retirement of a cohort of highly experienced staff and a salary scale that is below the median for public institutions in Canberra. Added to this has been the global financial crisis (GFC), which seems to have led to statistical outcomes that are difficult to interpret. To achieve cost savings, the ABS introduced, among other measures, a 24 percent cut in the sample size for the labor force survey. User concern to accurately track the impact of the GFC on employment subsequently led to the restoration of the sample. After an external review by the DoFD, the ABS was given some replenishment of funding, but continued to have to find savings through reducing staffing at all levels to provide resources to increase ABS salaries in line with public service wage movements. The September 2009 Quarterly National Accounts (QNA) had to be rereleased to correct compilation errors, with some loss of credibility amongst the user community. At this stage the worst of the transitional problems may be behind: the December QNA was issued successfully. However, a number of issues continue to defy analytical explanation—for instance, the divergent paths of the various estimates for GDP. Some users have also suggested a disconnect between growth in employment and that in the production based estimate of GDP; however growth in hours worked in this period is in line with growth in production and seems to reconcile the two series. Meanwhile, wage and other cost pressures are expected to continue, as the ABS strives to maintain its reputation for statistical excellence. While some of this is outside its control, careful management of its staff resources—including succession planning—will be important for its work program for the future.

7. As noted above, Australia’s statistics overall are of a high quality. Insofar as issues are identified in this Report, they seem in large part related to one or more of three factors: first, the short-term legacies of the recent methodological transitions, where a number of “loose ends” remain to be tidied up; second, the continuing resource pressures, under which the ABS feels itself unable to introduce any initiatives without being provided with additional funding; and third, an emphasis on meeting domestic information requirements where the view is that domestic requirements demand data that are on bases that differ from the relevant international standards. This last factor is naturally a particular focus of this ROSC exercise. Deviations from international standards, where material, may cause problems, not so much because after careful consideration Australian statisticians adopt an approach different from that mandated by the international methodologies, but because in such an event lower priority seems to be given to mapping, or providing other means to ensure that Australia’s figures can be interpreted on the same basis as those of the rest of the world. Also, with Australia’s statistics at the forefront of methodological innovation, the wider statistical community would hope to be able to draw on the lessons of Australia’s statistical transition; this would be assisted if any deviations were explained and quantified. Studies of the GFC have recognized the importance of data, and the need for harmonized statistics in order to interpret international economic developments. Achieving international comparability is made more difficult by Australia implementing the 2008 SNA and BPM6 standards ahead of other countries. It is therefore problematic if Australia does not release figures in accordance with international standards as still applied elsewhere, and also does not provide the means to convert or link its data to the 2003 SNA and BPM5 basis. As regards monetary statistics, the RBA recently made the move towards the adoption of the MFSM and implemented the underlying standardized report forms developed by the IMF. As a result, monetary statistics are now posted on the RBA website in formats that allow for easy international comparisons. Nonetheless, there is still room for improvement in the accuracy of source data and institutional coverage.

8. Section III of this report provides summary recommendations. Some of the recommendations can be implemented very quickly, without needing significant additional resources, while others may require staff and financial resources.

9. In applying the IMF’s DQAF, the remainder of this section presents the mission’s main findings. The findings are presented at the level of the DQAF’s quality dimensions, by agency for the first two dimensions, and across datasets for the remaining four.

10 Pre-requisites of quality and assurances of integrity.

  • National accounts, CPI, PPI, GFS, and BOP/IIP statistics are compiled by the Australian Bureau of Statistics in accordance with the Census and Statistics Act, 1905, which states (Part III, paragraph 9), that the Statistician … “may from time to time collect such statistical information…as he or she considers appropriate.” The Australian Bureau of Statistics Act, 1975 (ABSA) establishes the ABS as the central statistical agency in the country, and provides the Statistician, as the head of the ABS, with substantial independence, as well as discretion over all aspects of the collection and dissemination of statistics. The Statistician cannot be dismissed except on grounds of misbehavior, incapacity, or bankruptcy, and even then only on the vote in the same session of both houses of parliament. The Statistics Determination Act, 1983 sets out confidentiality provisions, and conditions for disclosure of information. There is limited prior access to data in a lock-up, and any prior access is fully disclosed to the public. The ABS has traditionally been at the cutting edge of methodological development, and at the present time is the only major statistical agency in the world that has essentially fully migrated to the new methodologies in all the main macroeconomic datasets. It has traditionally been well resourced but has not been immune from budgetary pressures faced by all Commonwealth agencies. However, despite receiving some additional funding in the past year, the ABS undertook some staff downsizing, and has also lost significant expertise through retirements. The existence of diverse statistical interests and objectives has made it difficult for the ABS to undertake short-term reprioritization of statistical programs to achieve economies; it is dependent on securing earmarked funding to take on additional statistical commitments.

  • Commonwealth monthly and annual financial statements are compiled by the Department of Finance and Deregulation, under provisions of the Financial Management and Accountability Act, 1997. The integrity of GFS compilation is further strengthened by the Charter of Budget Honesty Act, 1998, which requires among others things, regular fiscal reporting based on external reporting standards, defined as Australian GFS, and public sector accounting standards developed by the Australian Accounting Standards Board. From 2008–2009 onwards the Australian government also adopted the Australian Accounting Standards Board 1049 which requires the adoption of ABS GFS across the face of financial statements for the whole of the government and general government sectors, where it does not conflict with Australian accounting standards. Where differences exist, the reconciliation of accounting with key GFS aggregates is required.

  • Monetary statistics are disseminated by the Reserve Bank of Australia—albeit the Reserve Bank of Australia Act, 1959 does not include specific provision in this regard-with data from the banking sector collected by the Australian Prudential Regulation Agency under the Australian Prudential Regulation Authority Act, 1998. The RBA does not carry out users’ surveys beyond the circle of its correspondents in the financial sector.

11. All agencies seek to maintain a high level of professionalism, and transparency. All follow a Code of Conduct, which staff are made aware of at induction and periodically thereafter. Staff in all agencies follow the precepts of the Australian Public Service, which stresses professionalism and integrity. In all cases, choice of techniques is based solely on statistical considerations. There is no pre-release of information, except for limited lock-up access by the ABS that is explained to the public.

12. The methodological soundness of Australia’s statistics has been for many years amongst the highest in the world. Australia continues in this lead position, being the first country to essentially fully adopt the 2008 SNA and the BPM6, as well as the more established GFS 2001 and the MFSM 2000. Most practices are in line with these new, and hence demanding, standards, but there are some deviations, no doubt at least in part because the full transition is not yet complete. Annual and quarterly national accounts are essentially on the new basis, but a few categories of data are not yet worked out: these include the value of growing crops, databases, valuables, and illegal activities. As regards the CPI, the treatment of deposits and loans is an issue, but international guidelines in this area are not prescriptive and the practical problems of compilation faced by the ABS are common to other countries. Coverage embraces household final monetary consumption, with minor departures, such as the exclusion of second hand cars. However, for the CPI, the target population is restricted to the six State capitals plus Darwin and Canberra, which together cover only 64 percent of the population, far fewer than in other developed countries. The Household Expenditure Survey (HES) covers all households, but use of expenditure data for the weights is restricted to the main cities. The same geographic restriction applies to price collection. Published PPIs cover the minimum requirements for manufacturing, construction and mining, but not for services. The use of ANZSIC06 for the PPI allows a closer concordance with International Standard Industrial Classification (ISIC) than its predecessor and there are no conceptual differences. The classifications are broadly consistent at higher levels of aggregation. The latter is also the case, but to a more limited extent, for the Consumer Price Index Commodity Classification (CPICC) used for the CPI. Despite differences between the CPICC and Classification of Individual Consumption According to Purpose (COICOP), including conceptual differences, there is a reasonable concordance at a higher level which, with some manipulation of the lower-level data facilitates international comparisons on a comparable basis. The GFS are produced on nationally derived concepts, definitions, and presentational framework; where differences exist from GFSM 2001, source data are sufficiently detailed to allow linkage to GFSM 2001. The scope of GFS is consistent with GFSM 2001, but a stocks-and-flows data breakdown between domestic and foreign financial assets and liabilities, and between banks and nonbanks, is lacking. Recently, the RBA adopted the MFSM methodology for the compilation of monetary statistics. Nevertheless, the institutional coverage of monetary statistics—currently central bank, banks, credit unions, and building societies—could be further improved to include money-market corporations and cash management trusts as they already are in other financial aggregates disseminated by the RBA. In the BOP/IIP statistics the recording of repos is not in line with the recommendations of the BPM6 and 2008 SNA, as is the use of the creditor approach for interest on debt securities, as well as the use of fair value for debt other than securities.

13. Australia’s statistics generally demonstrate accuracy and reliability, with sound practices regarding source data, assessment of source data, statistical techniques, and revision studies. All agencies conduct periodic reviews on the various elements of accuracy and reliability. However, there are some areas, particularly as regards source data, where enhancements should be considered. As regards national accounts, many source statistics are good, but the Annual Integrated Collection suffers from a small sample size and limited product detail. The HES is now conducted only every six years, as against an international guideline of five years; past empirical evidence suggests that this may not be a big issue in terms of measured inflation. The ABS is reviewing the frequency of weights updating as part of its ongoing 16th Series review and is considering a number of options for updating weights more frequently. For monetary statistics, source data lack accuracy as the split between transactions in national currency and transactions in foreign currency is not available at the requested level of detail. Also, holdings of certificates of deposits issued by banks are not sectorized; consequently the total outstanding amount is allocated to broad money.

14. The serviceability of Australia’s statistics is in many cases excellent, as a result of the Australian agencies making considerable efforts to meet users’ needs. As noted above, the ABS is currently undertaking a major review of the CPI—formally the 16th Series review—soliciting users’ views. On the other hand, Australia maintains five flexibility options within the SDDS—more than most countries—and has made no progress in addressing the country’s need for them since Australia first subscribed to the SDDS more than a decade ago. As regards periodicity, for a number of series Australia disseminates on a quarterly basis while the SDDS mandate is for monthly. This seems particularly problematic as regards the CPI, which was by some distance the category for which the users in the informal survey expressed most dissatisfaction. The RBA and some sector users contend that a monthly CPI would serve to enhance the central bank’s ability to operate monetary policy because of better timeliness; also greater availability of data points would enable better analysis on, for instance, turning points, and could thus help avoid mistakes in economic management. The ABS is one of only two OECD country statistical institutions that do not compile a monthly index: this is one of the issues being considered in the 16th Series review. Overall, timeliness of Australia’s statistics is good, in some cases significantly exceeding SDDS requirements. However, as noted above, Australia still maintains flexibility options on timeliness. The production index is disseminated not later than one quarter after the reference period, as against the six-weekly requirement with one month encouraged. Also, Australia uses the option of less timely dissemination of central government operations at year end that is available to countries using the GFSM 2001 methodology. Consistency checks are not always applied appropriately across datasets.

15. Australia’s statistics have generally exemplary accessibility. All the disseminating agencies have put considerable efforts into developing their websites and making them user friendly. The ABS provides substantial amounts of free data on its website for the areas that it covers. Requests for customized additional data are charged. Publications also include abundant metadata, although these seem not always well customized for different types of user, and not for all data categories. Data releases also include analytic descriptions of developments. National accounts data are available on the website with series available back to September 1959. The CPI release covers a range of indices, with explanation of the factors contributing to inflation. The PPI release covers the headline figure and a range of sub-indices. Contact persons for assistance are identified at the front of most ABS data releases. For GFS, consistent data series on an accrual basis are available from 1998–1999, while some cash-based data are available from the ABS from 1961–1962. The DoFD provides a contact point for further assistance. The RBA ceased its hard-copy statistical publication in January 2010. It provides a wide range of monetary data covering a long time period on its website. Metadata are provided in an accompanying methodological note. The RBA does not identify a contact person for further assistance, but has a general contact point for users. All agencies provide a catalog of publications concerning the data. All disseminate information about future data release dates, although the DoFD on occasion has not released data on the specified date as administrative and internal clearance procedures determine the actual release date.

16. Section II provides a summary assessment by agency and dataset based on a four-part scale. This is followed by staff recommendations in Section III. Practices compared to the SDDS are summarized in Appendix I. The authorities’ response to this report and a volume of detailed assessments are presented in separate documents.

II. Assessment by Agency and Dataset

17. An assessment of the quality of six macroeconomic datasets—national accounts, the consumer price index, the producer price index, and government finance, monetary, and BOP/IIP statistics—was conducted using the DQAF, July 2003. In this section, the results are presented at the level of the DQAF elements and using a four-point rating scale (Table 1). Assessments of the prerequisites of data quality and the assurances of integrity (Dimensions “0” and “1” of the DQAF) are presented in Tables 2ac. For the various datasets, the assessment of methodological soundness, accuracy and reliability, serviceability, and accessibility (Dimensions “2” to “5” of the DQAF) is shown in Tables 3af. The analysis was carried out on the basis of a self-assessment by the compiling agencies, discussions during the mission, and an informal users’ survey.

Table 1.

Data Quality Assessment Framework July 2003—Summary Results

Key to symbols: O = Practice Observed; LO = Practice Largely Observed; LNO =Practice Largely Not Observed; NO = Practice Not Observed; NA = Not Applicable

article image
Practice observed: Current practices generally meet or achieve the objectives of DQAF internationally accepted statistical practices without any significant deficiencies. Practice largely observed: Some departures, but these are not seen as sufficient to raise doubts about the authorities’ ability to observe the DQAF practices. Practice largely not observed: Significant departures and the authorities will need to take significant action to achieve observance. Practice not observed: Most DQAF practices are not met. Not applicable: Used only exceptionally when statistical practices do not apply to a country’s circumstances.
Table 2a.

Assessment of Data Quality—Dimensions 0 and 1—Australian Bureau of Statistics

article image
Table 2b.

Assessment of Data Quality—Dimensions 0 and 1—Department of Finance and Deregulation

article image
Table 2c.

Assessment of Data Quality—Dimensions 0 and 1—Reserve Bank of Australia

article image
Table 3a.

Assessment of Data Quality—Dimensions 2 to 5—National Accounts

article image
Table 3b.

Assessment of Data Quality—Dimensions 2 to 5—Consumer Price Index

article image
Table 3c.

Assessment of Data Quality—Dimensions 2 to 5—Producer Price Index

article image
Table 3d.

Assessment of Data Quality—Dimensions 2 to 5—Government Finance Statistics

article image
Table 3e.

Assessment of Data Quality—Dimensions 2 to 5—Monetary Statistics

article image
Table 3f.

Assessment of Data Quality—Dimensions 2 to 5—Balance of Payments and International Investment Position Statistics

article image

III. Staffs Recommendations

18. Based on the review of Australia’s statistical practices, discussions with the data-producing agencies, and responses from data users (see Appendix III of the Detailed Assessments volume), a set of recommendations is presented below. They are designed to increase further Australia’s adherence to internationally accepted statistical practices and would, in the mission’s view, enhance the analytical usefulness of Australia’s statistics. Some additional technical suggestions are included in the Detailed Assessments volume.

Cross-cutting recommendations

  • Review deviations from current international standards, recognizing the broader benefits of international comparability, with a view at least to providing mapping to enable internationally comparable estimates to be derived.

  • Review the periodicity of data categories, particularly those where flexibility options have been taken under the SDDS. Especially, with the uncertainties in economic developments since the outbreak of the GFC, early discernment of turning points and of changes in trends could significantly help economic management.

  • Ensure staff management and succession planning so that the agencies providing official statistics can respond effectively to the needs of a rapidly changing external environment.

National accounts

  • Make (as an alternative presentation) national accounts series available that strictly follow the 2008 SNA in matters as the accrual of interest and valuation of loans.

  • Improve the sample size and product detail of the Annual Integrated Collection. Update the HES more frequently and review the PPI with the objective to improve its usability for national accounts compilation (see also below).

Consumer Price Index

  • Examine the costs and benefits of producing a national consumer price index and associated sub-indices on a monthly basis.

  • Examine the costs and benefits of conducting a HES on a more regular basis to facilitate the more frequent updating of CPI weights. This would include re-examining the potential bias in the CPI from out-of-date weights and how the bias would be mitigated with increases in the frequency of updating weights.

  • Assess the impact on the CPI of restricting the expenditure weights and CPI price collection to the six State capital cities plus Darwin and Canberra, and examine the options for compiling a CPI covering all the domestic territory of Australia.

  • Consider the costs and benefits of replacing the Australian expenditure classification CPICC with the internationally standard COICOP. Investigate options including a move to COICOP, further correspondence between CPICC and COICOP, or a classification system which is structured to deliver both. The implications for the coding of the HES also need to be considered.

Producer Price Index

  • Assess user needs for monthly data. Examine the costs and benefits of producing a limited number of key producer price indices on a monthly basis to meet the ROSC requirement for monthly indices and the needs of users.

  • Examine the costs and benefits of producing the outstanding PPIs for services, taking into account the added value from the gain in coverage, the methodological and data issues that would arise from developing PPIS for each service industry, and user needs.

  • Review the compilation of deflators in the national accounts. This review should cover both conceptual issues and issues of practical measurement. The aim would be to make better use of the data collected by Prices Branch.

Government finance statistics

  • Extend the transactional coverage of GFS data to include a breakdown of domestic/foreign and bank/nonbank stocks and flows of financial assets and liabilities, available in the quarterly financial accounts dataset.

  • Improve the timeliness of the monthly GFS data (Central Government Operations) dissemination, so as to eliminate the need for the targeted flexibility option.

  • Review deviations from GFSM 2001, and consider adopting the classifications and concepts fully so as to further increase consistency with other macroeconomic datasets, and attain international comparability.

  • Consider the costs and benefits of additional work to re-disseminate quarterly data for general government after benchmarking so as to attain a publicly disseminated time series consistent with annual data.

Monetary statistics

  • Obtain, from APRA, source data presenting systematically the split between transactions in national currency and those in foreign currency.

  • Collect data for a proper sectorization of holders of certificates of deposit issued by banks. Such sectorization can be obtained through a survey of money market operators. A first step would be to identify all holdings by financial corporations reporting to APRA.

  • Conduct consultations outside the financial sector with the view to better identify the needs of academics, media, international investors, etc.

  • Expand the institutional coverage of monetary statistics to money market corporations and cash management trusts.

Balance of payments and international investment position statistics

  • Consider publishing additional datasets that fully comply with the international standards for classification/sectorization as well as data that allow conversion to the BPM5 presentation format to serve international users.

  • Enhance consultations between the ABS and RBA to avoid differences in classification. Consider bringing the reserves data in the BOP/IIP and the reserve template in line with international standards regarding the recording of repos.

  • Publish or reclassify data that are currently available but not published or erroneously classified in the prescribed manner of BPM6.

  • Investigate the assumption that all portfolio equity investment abroad is through investment funds, with a view to showing separately dividends on equity excluding investment fund shares and obtaining a better estimation of reinvested earnings attributable to investment fund shareholders.

  • Consider using the sign convention as required under the BPM6 standards.

  • Even though this is not a requirement in the international data standards, it is recommended that the ABS start publishing remaining maturity of external debt by sector in view of the high international demand for these data.

Appendix Table 4.

Practices Compared to the SDDS Coverage, Periodicity, and Timeliness of Data

article image
Note: Periodicity and timeliness: (D) daily; (W) weekly or with a lag of ## week(s) from the reference date; (WD) working days, or business days; (M) monthly or with a lag of ## month(s); (NLT) not later than; (Q) quarterly or with a lag of ## quarter(s); (A) annually; (SA) semiannual; and (…) not applicable.Italics indicate encouraged categories.

Given that the data are broadly disseminated by private means, the timeliness with which official data are disseminated is not time critical.