| 2/2/2007||New Zealand||1. Statistics New Zealand supports the original recommendation proposed by the AEG.|
2. Our support for the compromise is conditional on agreement on the liability recognition criteria being reached before the SNA93 rev. 1 is finalised, i.e. we view this as an interim solution providing a way forward on this issue. We are of the view that the criteria on which to base the classification of government employer pension schemes either as employer schemes (and have the pension liability recorded in the core accounts) or as part of social security schemes (with no liability in the core accounts) must be agreed to before the SNA93 Rev.1 is finalised. We have considerable sympathy with the views put forward by the United Kingdom Office for National Statistics that the present proposal will lead to two different sets of accounts enshrined in the SNA standard, which must be avoided. The recognition of a financial asset/liability is a fundamental concept in the national accounts and has implications beyond this particular issue. An unequivocal position, one way or another, is required.
3. Furthermore, we are somewhat sceptical that the proposed supplementary account will achieve its stated aim, viz. to provide " a complete view of households’ pension entitlements", as pensions provided under social assistance schemes will be excluded. A full accounting of pension entitlements and the "social burden" of these and similar government outlays is probably beyond the scope of national accounts and is a research question best left to alternative expositions.
| 2/1/2007||National Bank of Azerbaijan|| f4cbAzerbaijan2; |
| 1/18/2007||National Bank of Slovenia|| f4cbSlovenia2; |
| 1/11/2007||Bank of Burundi|| f4cbBurundi2; |
| 1/11/2007||Central Bank of Myanmar|| f4cbMyanmar2; |
| 1/11/2007||Myanmar|| f4Myanmar2; |
| 1/11/2007||Central Bank of Bolivia|| f4cbBolivia2; |
| 1/2/2007||Bank of Israel||We support the proposal. Reporting government employee pension schemes and social security schemes liabilities will increase the transparency of government accounts. |
However, we also put a high weight on reporting the parameters for the calculation of liabilities at different countries. We suspect there will be room for a wide heterogeneity relating the items included in the calculation of the liabilities, and in the chosen parameters – the actuarial table being a crucial one. We suggest that as part of this initiative countries will provide an annex with detailed explanations on the included items, and on the chosen parameters for the calculation. Guidelines for performing the calculation would be desirable as well.
| 12/28/2006||Spain||From the point of view of the Spanish Statistical authorities, in principle, we agree with the three main elements of the proposal of your letter dated 10 November 2006. These main elements were included in the compromise text on the treatment of pension schemes in the updated SNA93, text that was fully backed by the Statistical Program Comittee, last November.|
| 12/21/2006||Armenia||I am pleased to inform you that we have no comments.|
| 12/21/2006||Bank of Zambia|| f4cbZambia2; |
| 12/21/2006||Central Bank of Nigeria|| f4cbNigeria2; |
| 12/19/2006||Reserve Bank of Australia||The RBA has no comments to offer on the proposal. |
| 12/19/2006||Czech National Bank|| f4cbCzech2; |
| 12/19/2006||State Bank of Vietnam|| f4cbVietnam2; |
| 12/19/2006||Bank of Colombia|| f4cbColombia2; |
| 12/19/2006||Central Bank of Ecuador|| f4Ecuador2; |
| 12/15/2006||Bank of Korea||We broadly support the recommendations. But it is somewhat difficult to calculate the size of liabilities in practice. So, we support the option of recording unfunded pension schemes which is sponsored by government in supplementary accounts.|
| 12/14/2006||Central Bank of El Slavador|| f4cbElsalvador2; |
| 12/13/2006||Bank of Hungary||We are informed that the European Central Bank also sends its opinion on this issue to the IMF. As the MNB takes part in the work of the Eurostat-ECB Task Force and supports its proposal on the treatment of pension schemes in the new SNA and was able to express its opinion on the topic, we support the ECB's views.|
| 12/13/2006||Central Bank of Chile|| f4cbChile2; f4cbChile2e; |
| 12/13/2006||European Central Bank|| f4ECB2; |
| 12/13/2006||Central Bank of Sri Lanka|| f4cbSriLanka2; |
| 12/13/2006||Mexico|| f4Mexico2; f4Mexico2e; |
| 12/12/2006||India||The compromise on the treatment of pension schemes in 1993 SNA update, suggested by Eurostat-ECB Task Force (six basic principles), is in broad agreement with us.|
It appears that the standard (supplementary) table on all pension schemes providing all flows and stocks of all pension schemes is a very good compromise solution. This Table will provide an overview of pension system in countries. However, pension schemes to be recorded in core accounts (having no direct government liability) may lead to comparability issues across the countries, although the definition appears to be clear (though, in several funded (defined benefit) schemes, we observe government assuming indirect responsibility when the schemes become underfunded). The distinction between government and other pension schemes has disappeared in the compromise recommendations giving way to the distinction of having or not having "direct government responsibility" (with the exception of government sponsored funded schemes). The recommendation under (v)gives a solution that countries should provide explanations on the criteria of choosing which pension schemes are to be recorded in core accounts and which in supplementary table. This will provide the requisite transparency.
| 12/12/2006||Central Bank of Cyprus|| f4cbCyprus2; |
| 12/11/2006||Statistics Canada/Bank of Canada||Statistics Canada is in agreement with the compromise. We have already begun to prepare supplementary tables on all pension schemes for our user community. As we have a borderline social security/employer sponsored program that exists here - it will give us the flexibility to provide the data in supplementary form.|
The compromise has already been discussed with our Central Bank - as Mr Cheung's letter went to them as well, and they are in agreement with Statistics Canada accepting the compromise solution.
| 12/11/2006||Bank of Botswana||The Bank of Botswana notes the proposal made by the Eurostat-ECB task force regarding the distinction between the government employee pension schemes and social security schemes. The Bank supports the ISWGNA in agreeing with the main elements of this proposal which allows for sufficient flexibility to cover differing situations |
| 12/11/2006||Sweden||Statistics Sweden agrees with the proposal on pension schemes.|
| 12/11/2006||Monetary Authority of Singapore||We do not have any comments on the recommendations at this stage. |
| 12/11/2006||Azerbaijan||In our opinion for those cases in which it is not feasible to make a distinction between government employee pension schemes and social security schemes it would be advisable to include in SNA 1993, Revision 1, a new supplementary table that shows the flows and stocks of pension entitlements of all pension schemes. |
| 12/11/2006||United States||The Bureau of Economic Analysis strongly supports this proposal. This proposal provides a modern and accurate accounting for liabilities associated with funded pension plans. This type of treatment is very important to our data users and will be an important step forward for the System of National Accounts. Furthermore, the proposal allows for a degree of flexibility in accounting for unfunded schemes in the core accounts, while still providing for the provision of information on the liabilities of these schemes through a supplementary table. We support this compromise solution and hope that this will help lead to full adoption of this and the other important improvements (such as capitalization of R&D, services of government assets, and the production account) that are part of this update of the SNA.|
| 12/11/2006||Australia||The Australian Bureau of Statistics supports the main elements of the proposal for those cases in which it is not possible to make a distinction between government employee pension schemes and social security schemes. We offer this support on the basis that where it is possible to make this distinction then government employee pensions schemes would be expected to be included in the core accounts, regardless of whether they are funded or not. We see this as an important advance in the updated SNA for countries such as Australia where governments themselves include these schemes in their own accounts. Furthermore, we see that the proposed supplementary table for those schemes that include elements of both government employee|
pension schemes and social security schemes will aid in international comparisons while still enabling countries to adopt a practice suited to their particular circumstances in the core accounts.
We would prefer that the updated SNA provide a principles-based approach for recording government employee pension schemes and that prescription be avoided, recognising that different circumstances exist in different countries. We would also prefer that the proposed supplementary table be kept as simple as possible to avoid onerous reporting requirements.
| 12/11/2006||Slovakia||On behalf of our Director General Mr Balaz I would like to express our support for the principles of the proposed treatment of Employer retirement pension schemes in the System of National Accounts.|
| 12/11/2006||State Bank of Pakistan||In Pakistan, a clear distinction exists between government employee retirement schemes and social security schemes. Both public and private sectors offer occupational saving schemes in the shape of gratuities, provident funds and pensions. The private sector schemes enjoy certain tax advantages, whereas, the public sector gratuity and pension schemes are non-contributory, unfunded and paid on a pay-as-you-go basis. In addition to this, provident fund schemes are contributed by both the employer and employee. The pension and social security schemes operate under a defined set of legislation and a clear distinction exists between the two schemes. Therefore, we agree with the proposal of Eurostat-ECB Task Force on the issue.|
| 12/11/2006||National Bank of Poland||Please be informed, that our opinions on this issue are reflected in decisions of Task Force on Statistical Measurement of the Assets and Liabilities of Pension Schemes in General Government as the representative of NBP participates in it.|
| 12/11/2006||National Bank of Croatia||Despite the fact that separate government employee pension scheme does not exist in the Republic of Croatia, in order to reach statistical harmonization worldwide, we are supporting all initiatives concerning the appropriate treatment of those schemes in revised SNA 1993 framework. Proposed flexible framework which consists of suplementary table not included in the core accounts and additional elaborations done by countries explaining the reasons for not including stocks and flows of gov't employee pension scheme into core accounts could be a right approach in that sense.|
| 12/11/2006||National Bank of Denmark||On behalf of Governor Bernstein I can inform you that Danmarks National-bank supports the proposal. It is a well-balanced compromise between the need for better statistical coverage of pension schemes and the need for in-ternationally comparable core national accounts statistics.|
| 12/11/2006||Bank of Latvia|| f4Latvia2; |
| 12/11/2006||Bank of Germany|| f4cbGermany2; |
| 12/11/2006||Bank of Indonesia|| f4cbIndonesia2; |
| 12/11/2006||Bank of Sweden||Sveriges Riksbank has been involved in the preparations of the new SNA in this respect through the representation in the CMFB (Committe on Monetary, Financial and Balance of Payments statistics), where these issues have been discussed over many years.|
Sveriges Riksbank supports the proposal made by the the Eurostat-European Central Bank Task Force on the Statistical Measurement of the Assets and Liabilities of Pensions Schemes in General Government, as describes in the letter. We find the proposal being a workable compromise, which, though, has to elaborated further.
| 12/11/2006||National Bank of Switzerland|| f4cbSwitzerland2; |
| 12/11/2006||United Kingdom|| f4UK2; |
| 12/7/2006||Central Bank of Iran||In general Central Bank of Iran welcomes and supports the present revision of the 1993 SNA. We would like to make the following brief comments on the issue of employer retirement pension schemes. |
We agree that there are issues to be resolved for some European countries for social security and government employer pension schemes. In Iran such as some other non-European countries, there are no problems in drawing the line between pension schemes offered by government to its employees and social security schemes. In Iran social security fund (organization) as a governmental institution offers pension benefits along with other services to a large part of population and there are also some pension funds for government employees. Statistics and information regarding these pension schemes are available and collected separately.
With regard to point (i) in the proposal, we believe that preparing a new supplementary table can not be generalized to all countries and this may lead to some practical problems or calculation difficulties in the implementation of the SNA. Therefore in this connection some flexibility seems desirable.
Regarding point (ii), we are concerned that the new way of recording will prevent international comparability. To ensure comparability between countries, a set of international agreed criteria is required to explain which schemes should be recorded in the core accounts and which ones can be remained in the supplementary table. In general we think there is a need for more guidelines and clarifications on this issue.
| 12/7/2006||Bank of Thailand|| f4cbThailand2; |
| 12/7/2006||Central Bank of Nicaragua||En el caso de Nicaragua, dentro del gobierno general, existen únicamente 2 ministerios públicos que cuentan con planes de jubilación para sus empleados: La policía Nacional y el Ejército de Nicaragua. Estos planes de jubilación son administrados por los mismos ministerios y de manera separada de la seguridad social; sus empleados públicos no realizan ningún aporte ni reciben ninguna prestación de parte de la seguridad social. |
Estamos de acuerdo en que se incluya en el SCN 1993, Rev.1 un nuevo cuadro estándar en el que se presenten los stocks y flujos de los derechos jubilatorios correspondientes a todos los planes de pensiones, pues esto permitiría que los cuadros resultantes sean comparables entre los países, lo cual no sería posible si se permite la flexibilidad de que los países que tengan estos planes jubilatorios los incluyan o no en las cuentas principales.
| 12/6/2006||Italy||In relation to the ongoing debate on the revision process of System of National Accounts 1993 (SNA 1993) We think that current compromise could be a feasible basis for further discussions.|
In fact, the proposal seems to be able to allow a better representation of national pension systems (with particular emphasis on their long-term sustainability) but, at the same time, does not force National Statistics Institutes to change the current treatment of pension schemes in the core accounts. In fact, the compromise proposal leaves unchanged the recording rules for the schemes already included into the core accounts; in addition, all flows and stocks of all pension schemes (private and public) will be recorded in the new supplementary compulsory table.
Moreover, it seems appreciable the choice of guaranteeing a certain degree of flexibility in the classification of schemes sponsored by government for all employees, taking into account the specific characteristics of each national pension system and the fact that government employer pensions schemes and social security can not be clear-cut divided.
However, in our opinion, two aspects seem to call for a more in-depth analysis.
Firstly, the identification of criteria allowing to classify borderline cases between the two categories of government employers pension schemes and social security schemes.
Secondly, the development of methods for identifying and measuring pension liabilities (and related assets), especially for what it concerns the actuarial techniques to be used in the computations.
We think that the adoption of common criteria about these aspects are essential for allowing to a better international comparability of pension data.
We know that these issues are currently in the agenda of the EUROSTAT/ECB task force on the statistical measurement of the assets and liabilities of pension schemes in General Government.
On account of this, it seems necessary, before giving our final opinion on the revision process of 1993 SNA, to wait for the outcome of the Task Force on these issues.
| 12/6/2006||National Bank of Kazakhstan||The National Bank of Kazakhstan has considered your letter of November 13, 2006, concerning distinction between government employee schemes and social security schemes. |
By the results of consideration, please be advised of the following.
For finding opinions of appropriate experts we have required Statistical Agency, Ministry of Labour and Social Defence of Population, Ministry of Economy and Budget Planning of Republic of Kazakhstan.
Summarizing viewings of above-mentioned experts, as a whole we agree with Eurostat-ECB Task Force proposal and consider it as well-founded and expedient.
At the same time, it is necessarily to inform you that in our country the situation differs from some European countries, in which government employee retirement schemes are entwined with social security schemes. In Kazakhstan, independently from pension schemes the system of obligatory social insurance does work separately, they are not entwined. The system of obligatory social insurance is aimed to reimburse part of revenue, wasted under realizing of insurance cases – loss of job, ability to work, breadwinner. This system does not make differences between citizens with any social categories, so in our country there is no conception “government employee retirement schemes”.
Furthermore, in view of Statistical Agency experts, in order to count flows and stocks of pension allowances and include their into the 1993 SNA in supplementary table, they are needed in appropriate data, which can be get from several relevant agencies. At present they don’t have similar data base.
| 12/6/2006||Denmark||Statistics Denmark welcomes the resent compromise solution on pension schemes. We generally agree with the proposal only to put employer pension-related flows and stocks in the core accounts. However, we do not think that the solution solves the fundamental problems regarding international comparability. Only small formal differences in the way schemes are organised in two countries can still result in different classifications and thereby differences in what is recorded in the core accounts and/or the new supplementary accounts. |
Whether or not the new supplementary table should be compulsory or not for European Union members has little relevance to the upcoming SNA update in our opinion. Therefore we recommend that the last sentence of paragraph (iii) is removed.
Finally we see some problems that need further elaborations in the proposal on pay-as–you-go government sponsored schemes in para. iv:
In the proposal it is unclear how the economic flows are to be recorded in the non-financial accounts. The recognition of a liability of general government and the counterpart asset of households raises the question of how the consistency between the financial and the non-financial accounts are maintained.
If an asset is registered in the supplementary accounts it must be possible to have transactions in the asset. Our concern is how it is ensured that economic flows in the non-financial accounts are consistent with transactions in the financial accounts. In order to ensure the consistency it seems necessary to register an “adjustment for the change in the net equity of households in PAYG schemes” similar to the adjustment in D.8in the present SNA. In order to make this adjustment it is not only necessary to make actuarial assumptions on demographic issues, assumptions on discount rates etc., but also assumptions on issues such as how large a proportion of tax-revenue can be attributed to PAYG schemes in each period i.e. how much shall tax-revenue be reduced and contributions to pension schemes be increased compared to the present treatment in SNA.
| 12/6/2006||Bank of Central African States|| f4cbBAS2f; f4cbBAS2e; |
| 12/5/2006||Germany|| f4Germany2; |
| 12/4/2006||Luxembourg||Luxembourg is favorable to the possible compromise solution presented on the SNA website you mention in your e-mail. This compromise text was also presented to the Eurostat National accounts working party in november 2006 (document CN612 - Annex 2) where Luxembourg was also favorable.|
| 12/4/2006||Belarus||К вопросу пересмотра СНС 1993 о пенсионных системах и учете пенсионных пособий Минстат Республики Беларусь сообщает следующее.|
В настоящее время в Республике Беларусь действует государственная система пенсионного обеспечения. Пенсии выплачиваются домашним хозяйствам из государственного Фонда социальной защиты населения, государственного бюджета и средств предприятий. За счет средств государственного Фонда социальной защиты населения выплачиваются домашним хозяйствам трудовые пенсии. За счет средств государственного бюджета выплачиваются: социальные пенсии; доплаты за особые заслуги, за выслугу лет государственным служащим; пенсии военнослужащим срочной службы и их семьям.
При условии создания в будущем в Республике Беларусь многосистемного пенсионного обеспечения, считаем полезным иметь в пересмотренном варианте СНС 1993 разъяснения по вопросу разграничения между пенсионными системами для государственных служащих и системами социального страхования, предложенные на совещании Рабочей группы Евростата и Европейского центрального банка.
| 12/1/2006||Malawi||The National Statistical Office agrees with the Eurostat-ECB that there should be a distinction between Government employee pension and social security schemes. The reasons for this are outlined below:|
1) The whole proposal is intended to enhance financial management, transparency and accountability by government;
2) The clear distinction between the government employee pension and social security schemes will add clarity in terms of detail of government expenditure on the two items thereby improving the analytical framework of government expenditure;
3) As NSO is improving and expanding its national accounts (Supply and Use Table) and as it is evident that Government will introduce the contributory pension scheme, these items (the government employee pension and other social security schemes) will be easily captured in the Secondary Distribution of Income Account/social contributions;
4) Taking into account that economies are at different stages of development, the impact of these two items would also vary. Therefore, these two should be separated so that if comparison has to be made on government employee pension for several countries, it should be the one and the same thing.
| 12/1/2006||Central Bank of Suriname|| f4Suriname2; |
| 11/24/2006||Samoa||We agree in principle to the proposed changes based on the following reasons.|
1. From the planning perspective, the demarcation of the pension scheme and the social security will make a clear distinction between superannuation (savings for the future) and expenditure pertaining to ensuring that there is social inclusiveness on basic goods and services.
2. The flexibility on the qualification of these expenditures are provided in proposed changes i) and ii) as presented in your letter will ensure that these classifications are more practical and workable.
3. In terms of data collection, there may be some problems however, the innovative and creativity of statisticians is crucial in this matter.
| 11/17/2006||The Netherlands||All in all, I can also agree with the proposed compromise. However, I would like to note two points in relation to the proposed recording of unfunded pension schemes sponsored by government: |
(i) The proposed flexibility may lead to problems regarding international comparability of (the definitions used for the compilation of) government data in the core system. Therefore, I do hope that the additional explanations on the rationale and criteria used for the actual recording inside or outside the core system will provide the way forward to arrive at a set of internationally recognised criteria. Here, I hope/assume that we will be able to arrive at solutions before the actual implementation of the updated SNA.
(ii) The proposal in para. (v) of the compromise proposal relates to schemes sponsored by government for all employees (whether private sector employees or government's own employees). It is unclear to me what this means for the recording of a government sponsored scheme that only relates to government's own employees. Should all pension-related flows and stocks, including pension entitlements, of such a scheme be recorded in the core system, just like the treatment of private schemes according to para. (i) of the compromise proposal?
| 11/17/2006||Ireland||The Central Statistics Office, Ireland agrees with the compromise solution proposed for the treatment of unfunded occupational pension schemes of Government in the update of the SNA.|
| 11/16/2006||Finland||Compromise on the treatment of pension schemes in the updated SNA93 Statistics Finland supports the proposed compromise to include a new standard supplementary table in the SNA. It is very important that all kinds of employment pension arrangements could be seen in the same table and they could be compared. Because institutional arrangements vary between countries it is important that in the core accounts there will be flexibility how to classify pension schemes.|
For instance in Finland all mandatory employment pension schemes are classified into social security funds. Government pension schemes are included, because they are a part of the Finnish collective, mandatory employment pension arrangement, which is organised as separate unfunded systems with buffer funds. It is very important to describe government schemes in the same way than private statutory pension schemes and not to classify them outside government sector.
| 11/16/2006||Portugal|| f4portugal2; |
| 10/11/2006||Armenia||The National Statistical Service of the Republic of Armenia agrees with the most recent AEG recommendations.|
| 10/10/2006||Denmark||At present the provisional recommendation on the treatment of unfunded (or under-funded) employer pension schemes is being reconsidered by a special Task Force, and while we support this process, and agree that there is a need for some clarifications on this issue, we would like to make the following brief comments. Firstly, the proposed treatment not only involves the recognition of the liabilities of the employer, but also involves recording of the associated economic flows in a rather complex way, which only a limited number of highly professional users will be able to understand. Secondly, most of these economic flows are based on imputations, which are in contrast to the principle of keeping the number of imputations in the System to a minimum. Thirdly, the well-known difficulties with the distinction between pension schemes for government employees and social security schemes can – or more precisely will – lead to situations where only small formal differences between the schemes in two countries result in different classifications and thereby different treatments. Considering the size of these schemes this could hamper the international comparability.|
| 10/10/2006||Sweden||Further clarification/analysis required|
| 10/10/2006||Germany||We reject the proposed change.|
We generally doubt whether contingent liabilities for which neither the level nor the modalities of realisation are known should be recorded in the SNA.
Calculating that issue could be done only on a model basis involving many imputations and assumptions. Assumptions would have to be made, for example, regarding future life expectancy, retirement age, retirement claims, and retirement from active service, etc. The future pension claims calculated on a model basis would have to.be discounted to the present value using a suitable interest rate. Choosing different interest rates may lead to considerable differences in value. International comparability would be seriously affected. Results already published would have to be corrected very often in the course of time, which would lead to considerable disturbance and subsequent changes within the core systems.
All the above arguments clearly support calculating pension liabilities outside the core system in supplementary tables or satellite accounts.
| 9/29/2006||Bank of Korea||We broadly support the recommendations. But we support the option of recording unfunded pension schemes in supplementary accounts because it is somewhat difficult to calculate the size of liabilities in practice. |
| 9/28/2006||New Zealand||We agree with the recommendations made by the AEG. In practice we will have trouble measuring any private schemes but in NZ most of the defined benefit schemes are government and actuarial values are available for them. Our current treatment of them in the national accounts already goes part way to meeting these recommendations. |
We support the development of criteria for distinguishing between employer schemes and social security and recognise that this is an issue for some countries.
| 9/27/2006||Iceland||Statistics Iceland is in favour of recognizing and recording as liabilities unfunded employers schemes. In Iceland the private pension schemes are entirely funded. However, the pension schemes for government employees are also funded to a certain extent, but substantially under-funded. Obligations due to this under-funding are recorded in the government accounts in a similar way as other liabilities. This under-funding is also taken into account in estimating government final consumption. In our view this is not a kind of an imputation but rather keeping with the accrual principle of accounting. |
| 9/15/2006||United Kingdom||This is a major issue for users of the accounts and every effort should be made to resolve it during this update. There are new financial accounting standards, implemented in the UK government departmental resource accounts, which demonstrate that this difficult issue can be addressed. The challenge is to bring some equivalent treatment into the integrated national accounting framework. The accounting developments also mean that data availability and reliability is improving rapidly which will ease the implementation of new requirements. (Some countries have already implemented something similar in their national accounts which demonstrates that the practical problems can be overcome.)|
The UK believes that the original package of proposals developed by the IMF taskforce was on the right lines. We support the existing recommendations for private pension schemes. We agree that there are issues to be resolved for some countries for social security and government employer schemes. There may be different estimates for different purposes and we need to make information available for this analysis. However, we should also identify a preferred set of best estimates and ensure that these are in the core accounts. We do not support a set of alternative estimates in supplementary accounts, as this will cause confusion to users.
| 9/15/2006||Latvia||After deep discussions and expert consultations we basically support the
1993 SNA Update Issues.|
| 9/15/2006||Norway||Statistics Norway generally supports that funded and unfunded employers’ pension schemes should be treated equivalently in the accounts. In Norway, private employers’ pension schemes and local govern¬ment employer’s pension schemes are funded, while the central governments employer’s pension scheme is unfunded (pay as you go system). As employees in central government built up claims on their employers in the same way as employees in the private sector, it seems reasonable that the employers’ liabilities and other associated flows should be recorded in an equivalent way independent of whether the employers’ pension schemes are funded or unfunded. |
Principally, a common distinction is between government employers’ pension schemes and social security schemes. In Norway the social security scheme is global, covering the whole population independent of whether you have been employed or not. However, a part of this social security pension is income related. Moreover, there is a clear link and coordination in the Norwegian system between the social security scheme and pension rights you earn as a government employee. In practice therefore, it may be difficult to distinguish between pension schemes for government employees and the social security scheme. This may apply for Norway, as well as for other countries. Thus, we are concerned that the new recording will hamper international comparability. To ensure comparability between countries, it is important that further work is undertaken to find methods so that more or less corresponding system in different countries are classified consistently.
| 8/21/2006||Central Bank of El Salvador|| m4cbElSalvador2; m4cbElSalvador2-english; |
| 8/18/2006||Netherlands||We generally support the recommendations made at the Frankfurt Meeting of the AEG. |
| 8/18/2006||Italy||For what it concerns the statistical treatment of pension schemes organised by governments for their employees, Istat supports the adoption of a criterion for the recognition of liabilities of unfunded public employers’ pension schemes on the Core Accounts based on the intrinsic characteristic of pension obligations.|
More specifically, a liability should be recorded in the core accounts - in order to take account for the obligation of government units to provide future retirement benefits - only when this obligation leads to the existence of an actual liability. An actual liability can be identified when the cumulated rights of civil servants can not be modified unilaterally by Government. On the contrary, when Governments may unilaterally alter the benefits by law, a contingent liability can be identified. In this case, no liability should be recorded in the core accounts but such contingent liabilities should then be recorded, together with social security liabilities, in a set of supplementary accounts.
Eurostat had proposed to set up a task force on the statistical aspects of the measurement of the liabilities and assets of public employee pension schemes and social security schemes. The task force would examine borderline aspects between these schemes and the issue of possible sources and methods for measuring their liabilities and related assets. We are waiting for the Task Force results before give our comments on this point.
| 8/18/2006||USA||We strongly agree with the AEG’s recommendations, especially its key recommendation to recognize the liabilities involved with all employer pension schemes, including unfunded ones, and any associated assets and transactions. The recommended treatment would provide more accurate information on the economic conditions of the employers that provide defined-benefit pensions. The proposed treatment is important not only for the government sector, but also to better understand the liabilities of private-sector employers with under-funded pension plans.|
| 8/2/2006||European Central Bank|| m4ECB2; |
| 8/1/2006||Hong Kong, China||We generally agree to the recommendations for the treatment of autonomous pension funds based on service charges and non-autonomous pension funds measured on a cost basis. |
However, there may be operational difficulty in measuring unfunded pension schemes using the actuarial method as the change of discount rate may affect the estimated liabilities of the funds.
| 8/1/2006||National Bank/ National Bureau for Statistics Moldova||National Bank and National Bureau for Statistics of Moldova agree with the recommendations made at the latest meeting of the AEG.|
| 8/1/2006||Reserve Bank of South Africa||We carefully worked through all the issues and would like to give our general support to the recommendations made by the AEG.|
| 7/31/2006||Israel||We agree that it is important to recognise the liabilities involved with all employer pension schemes, including unfunded ones, and any associated assets and transactions. For government employees the system already includes imputations for the value of such schemes in the compensation of employees, and it seems important to have a fully consistent system of accounts.|
We also think that it would be an improvement to recommend supplementary accounts for social security schemes.
But it is important to have very clear guidelines for accounting for such schemes. Experience has shown that actuarial estimates for unfunded schemes may differ widely depending upon underlying assumptions, and without very clear international guidelines, there will be serious problems of comparability within and between countries.
It seems possible to overcome the problem of separating pension schemes for government employees from social security schemes – if not, they could be shown together, rather than choosing not to show them at all.
| 7/31/2006||Macao SAR||Statistics and Census Service of Macao SAR agrees to the AEG recommendations and has no further comments.|
| 7/31/2006||Spain||El INE propone la siguiente enmienda a la recomendación del AEG sobre el registro de los esquemas de pensiones de empleadores, en la actualización del SNA:|
En el contexto de la recomendación general, que establece que deben registrarse en el marco central del sistema las obligaciones de los empleadores, en los sistemas de pensiones constituidos por ellos, con o sin reservas, debería realizarse una consideración relativa a los esquemas constituidos por las administraciones públicas para sus propios asalariados.
Así, en este caso, solamente se deberían reconocer en el marco central del Sistema de Cuentas Nacionales, aquellos pasivos determinados por las futuras obligaciones de las administraciones públicas en materia de pensiones, cuando los contratos firmados entre dichas administraciones y sus asalariados no puedan modificarse unilateralmente, es decir, cuando la obligación es efectiva o incondicional.
Sin embargo, cuando las administraciones puedan, de forma unilateral, modificar las prestaciones por ley sin posibilidad real de que los asalariados puedan recurrir dichas modificaciones, la obligación en materia de pensiones es contingente y, en este caso, el pasivo correspondiente no se debería registrar en el marco central del Sistema de Cuentas Nacionales, sino en un conjunto suplementario de cuentas.
En ambos casos, no obstante, debe tenerse en cuenta la gran dificultad para estimar (imputar) dichos pasivos de una manera fiable. En este sentido, el Grupo de Expertos que se va a reunir en el ámbito de EUROSTAT y del Banco Central Europeo puede proporcionar las orientaciones necesarias para la estimación armonizada de estos pasivos.
| 7/31/2006||Bank of Indonesia||Indonesia has two retirement pension schemes of unfunded and funded schemes. In this regard, BI considered the solution lies with statisticians who prepare the GFS and the Statistics Indonesia will simply take and put them in national accounts.|
| 7/31/2006||Bosnia and Herzegovina||We agree with AEG recommendations on the update of the 1993 SNA and do not have any further comments.|
| 7/28/2006||France||L'INSEE soutient la présentation équilibrée faite dans le document " Full set of provisional recommendations ".|
Il est en particulier important d'examiner précisément dans quelles conditions il est opportun de reconnaître un engagement envers leurs salariés de certains régimes organizes directement par les administrations publiques en tant qu'employeurs, et dans quelles conditions certains autres de ces régimes doivent être traités comme une partie de la sécurité sociale.
C'est pourquoi l'INSEE approuve le fait que la réflexion doive encore progresser dans de domaine, comme l'indique le document.
S'agissant du transfert " d'obligations " entre régimes, l'INSEE n'est pas d'accord avec le texte proposé, qui propose, lorsqu'un régime est transféré dans un autre, d'imputer systématiquement un transfert d'engagements de retraite, même lorsque le cadre central des comptes ne reconnaît pas de tels engagements. L'INSEE estime qu'il est dangereux d'essayer de résoudre, dans un texte conceptuel, des questions délicates par une proposition de traitement aussi simple sans en avoir évalué avec précision toutes les conséquences possibles.
| 7/28/2006||Japan||In Japan, transactions have taken place in which employee pension funds return the public part of pension reserves to the Welfare Insurances Special Account of the state since 2003. (so-called “Daiko Henjo”)|
These transactions are treated as capital transfers from employee pension funds (private financial corporations) to the Welfare Insurances Special Account of the state (social security fund), because public pension schemes’ liabilities are not recorded in the current 1993SNA.
This treatment has positive impact on net lending/borrowing of the general government.
The amounts of capital transfers concerning these transactions are shown in “Notice on Usage”(See http://www.esri.cao.go.jp/en/sna/h16-kaku/usage.pdf), so that users can calculate the amounts of net lending/borrowing without the impact of “Daiko Henjo” .
| 7/28/2006||National Bank of Slovakia||Concerning the results of the most recent AEG meeting, we fully support the conclusions and recommendations made by the AEG. However, as a member of the ESCB, the National Bank of Slovakia fully supports the comments of the ECB.|
| 7/28/2006||Australia||In Australia the obligation related to government employee pension schemes is considered a liability of governments to households and as such should be recorded on the government and household sector balance sheets. The recording of contributions, imputed returns and benefit flows consistent with the changing liability should also be recorded. In Australia all of these accounting entries are compiled by the relevant government jurisdictions and reported in their public accounts. For national accounts and government finance statistics purposes these data are simply taken from those government records - no estimation is required within the NSO.|
Since a liability is recognised there is no doubt that these amounts should be recorded in the system of national accounts. How the liability is funded is irrelevant. Further, we believe it would be misleading to exclude these liabilities from the government and household accounts as they clearly impact on current and future economic decisions. Where a liability is deemed to exist it must be considered as part of the statistical picture of any economy.
There is no doubt that assumptions are required to estimate the stocks and flows associated with these liabilities but making considered assumptions is a regular part of national accounting. Estimates for depreciation, owner occupied rent and the non-observed economy are just three examples. Since the methods for constructing estimates of the pension liabilities are well understood and used through the general accounting and insurance communities, there should be every confidence that well considered estimates of these liabilities can be constructed. The measures may not be perfect but they are sure to provide a much more realistic economic picture than would be provided if they were omitted completely.
| 7/28/2006||Lithuania||In general we support the recommendations.|
| 7/28/2006||Bank of Norway||The different treatment of funded and unfunded employer pension schemes is an anomaly in the SNA1993. The national accounts should treat employers’ future pension commitments independent of how the retirement scheme is organized. The recommended treatment by the AEG will improve the data quality of the liabilities of the non-financial corporation sector and the assets data of the household sector as well as the net worth of the sectors. We therefore support that the updated SNA should recognise pension commitments as liabilities (and adjust the employers’ contribution to the scheme) regardless of the retirement scheme being funded or not.|
| 7/28/2006||Bank of Poland||We would like to express our reservation about inclusion of liabilities of unfunded employer pension schemes in the core accounts, and not in
supplementary accounts together with liabilities of social security schemes which, in our opinion, are economically similar.|
| 7/28/2006||Bank of Portugal||We support the view that liabilities/assets and associated economic flows of all unfunded pension schemes as well as social security schemes should be recorded in supplementary accounts. The recording in the core accounts should only be considered if there is a guarantee that the results are consistent among countries and through time. In our opinion, in the current stage, this is not the case. For example, the methods for estimating the relevant variables which are used in the computation of pension liabilities (such as the discount factor or the life expectancy) could have a significant impact on the result and should therefore be dealt with in the SNA. Furthermore, it is crucial that different legal arrangements between countries are treated in a consistent way. Moreover, it should be further clarified if when an unfunded social insurance scheme becomes part of social security, there is a liability that simply disappears from the system.|
In our view accumulated benefits and related economic flows for all defined benefit schemes should be calculated using actuarial methods and we recommend that their main implementation guidelines, for the sake of the international statistical comparison, be referred to in the updated SNA.
We agree with AEG in not supporting that the definition of social security pension schemes mentions the existence of "collective multi-employer schemes". We also agree with AEG in not confirming that all unfunded collective multi-employer schemes are to be treated as current transfer schemes. In our view, these recommendations would turn inconsistent with the main goal of the revision i.e., to record the employers pension schemes as saving schemes.
| 7/27/2006||Bank of Sierra Leone||We agree with the recommendations made by the AEG.|
| 7/27/2006||Egypt||We have to consider the large number of practical problems in calculating the liabilities of employer pension schemes. It will be based on a large number of assumptions. It will be better if it is handled in supplementary accounts. |
| 7/25/2006||Bank of Italy||We share the view of many European institutions that treating the obligations of government under unfunded employer pensions schemes in a completely different way from its obligations under social security schemes would result in NA giving a misleading picture of the government’s position. The two types of obligations have similar economic meaning in institutional contexts that are common in Europe. Therefore the inclusion of unfunded schemes in the core accounts appears unwarranted, while a set of supplementary accounts could be introduced in order to represent both types of government obligations in a more accurate way. We appreciate that further thought is currently being given to the issue. We understand that a suggestion is being made that enough flexibility should be allowed in the system, so that different institutional situations can be accommodated. In this context the use of supplementary accounts is being considered, as well as a degree of flexibility concerning the criteria for including employer pension schemes in the core accounts. We take the position that these developments are in the interest of the clarity and meaningfulness of accounts and support them strongly.|
| 7/24/2006||Switzerland|| m4Switzerland2; |
| 7/24/2006||Germany||a) Germany sees a lot of theoretical and practical problems, which emerge by calculating the liabilities of employer pension schemes. The estimates of the liabilities are very sensitive to assumptions and methods. Therefore we do support that a liability should be recorded only when the liability leads to the existence of an actual liability. When the obligation of government units to provide future retirement benefits leads to the existence of a contingent liability, no liability should be recorded in the central framework. Such contingent liabilities could best be estimated in a set of supplementary accounts.|
c) We do not support the use of imputed units and transactions. In the case of non-autonomous pension schemes there is no local kind of activity unit. Therefore no output should be attributed to non-autonomous pension schemes.
d) We are in favour to describe the economic process in monetary and readily observable terms. Therefore we do not support the use of expected values to determine the output of autonomous pension schemes.
e) We recognise that there exists a “price” when the obligation to pay pensions passes from one unit to another. In the case of an actual liability there is no problem and the obligation has to be recorded in the core system. In the case of a contingent liability the value of the claim can alter after the transaction took place and therefore should be recorded in a set of supplementary accounts.
| 7/24/2006||National Bank of the Republic of Azerbaijan||We have analyzed the “Comment on the recommendations of the most recent Advisory Expert Group on National Accounts (AEG) meeting (January 30 – February 8, 2006) in Frankfurt” within the scope of our responsibilities and I am pleased to inform you that we are in agreement with the AEG recommendations. |
| 7/24/2006||Nepal||We are in favor of the idea that the future pension benefits should be accounted as liabilities of the employer and treated (funded and refunded) separately rather than including it in the framework of national accounting as an additional macroeconomic indicators in satellite accounts. Inclusion in the system may have distortion in the accounts and not very right from comparison point of view, since the measurement technique has not been decided and agreed.|
| 7/12/2006||Finland||We are in favour of the approach presented in Europe by ECB and Eurostat to have supplementary accounts for all pension schemes (including social security schemes for employment pensions) and not to record pension li-abilities for all employer (funded and unfunded) schemes in the core accounts. Distinction between pension schemes and social security schemes is not clear at least in the European countries. When it is necessary to com-pare different countries with very different arrangements it is best to start with supplementary accounts and not recording some arrangements in the core system and some still outside of it. It may be quite difficult to develop criteria that would distinguish between different types of schemes. |
Statistics Denmark would prefer liabilities/assets and associated economic flows of all unfunded employer pension schemes to be presented in “satellite accounts”.
| 1/26/2006||France||Please find enclosed a note putting forward the point of view of INSEE on the issue concerning the treatment of pension schemes in the SNA updating process.|
| 12/13/2005||Canada||Canada has already included the liability of so-called unfunded employer pension schemes as a liability of the sponsor and asset of the household sector. Large unfunded defined benefit schemes exist largely for the federal and some provincial government units. This has always been a changing mix, and has caused problems under the old treatment when an ongoing defined benefit plan switched from unfunded to funded status. Since the sponsors were governments whose legislation provided legal status to the pension liabilities and actuarial liability estimates existed in all cases, the distinction between funded and unfunded in these cases seemed artificial. Treating all defined benefit plans on the same basis better reflected the expectations of the contributors whose savings/spending behaviour are affected by participation in the plans. For the pure social security schemes, Canada continues to recognize the pay-as-you-go nature, and has not recognized a liability. One case exists where a regime designed to be pay-as-you-go is now accumulating an invested fund to cover the increased liability related to the large cohort of baby boomers in the system. This fund has not as yet been included as household assets. More work needs to be done to clarify accounting for pensions in the SNA|
| 12/1/2005||Slovenia||We are in favor of the solution that also employers’ unfunded pension schemes and particularly social security schemes will be included into the accounts because of their importance. We also agree with the proposal that they will not be treated in the core accounts but in separate accounts and balance sheets.|
| 11/30/2005||German Bundesbank||Concerning the topic “pensions” the treatment of unfunded employer pension schemes which has been discussed in depth in Europe seems particularly relevant for a large number of European countries where social security constitutes a major part of old age income across the population. With regard to these institutional setups, a distinction between unfunded employer schemes and social security appears to be less relevant than for countries where social security is limited to pure social assistance. The issue also relates to the question of whether civil servants’ pension schemes and the statutory pension system (social security) should be treated differently in national and financial accounts. In this context, it seems important to note that pension promises made by the government, which can be altered unilaterally by changes in legislation, are of a different nature than private pension assurances. This is obvious from recently proposed measures to consolidate the statutory pension system and civil servants’ schemes in Germany, which will result in a substantial cut in accrued entitlements. |
Considering those arguments, we would like to support a mandatory recording of liabilities of unfunded employer pension schemes and social security, but only in a supplementary set of accounts outside the core system. The advantage of this procedure would be a reporting of unfunded employer pension liabilities without drawing a line between social security schemes and unfunded employer schemes. This allows for both cross-country analysis and a country-specific treatment based on the appropriate boundary of pension liabilities in the country concerned.
| 11/29/2005||Estonia||We support the solution recommended by FAWG and ECB’s Statistics Committee: treatment of unfunded schemes in the core accounts remain unchanged and supplementary model estimates are recorded in a separate set of flow accounts and balance sheets.|
| 5/19/2005||Bank of Indonesia||Indonesia has two retirement pension schemes of unfunded and funded schemes. In this regard, BI considered the solution lies with statisticians who prepare the GFS and the Statistics Indonesia will simply take and put them in national accounts.|
| 5/6/2005||Vietnam||It is necessary to consider the cost to employer retirement person schemes and to portray under funding or over funding of defined benefit employer pension schemes as an obligations or a claim of the employers.|
It is possible to collect information in practice.