HomeSNAISWGNAKnowledge BaseData Publications
You are here:   ISWGNA >> Updating the SNA >> Towards the 2008 SNA >> 1993 SNA Update Information >> List of Issues

1993 SNA Update Information - Country comments for issue:
Government owned assets

Issue description
Issue description in [English] | [French] | [Russian] | [Spanish]
When summing costs to measure non-market output, the 1993 SNA recommends that the value of the services provided by a producer’s own non-financial assets should be measured as consumption of fixed capital. This means that neither a return on capital to these assets nor, equivalently, an opportunity cost of capital is recognized. This leads to an inconsistency with the rental that would have to be paid if the assets were rented. Should the SNA recommendation be changed and the cost of consumption of fixed capital be replaced with capital services (consumption of fixed capital, expected holding gains/ losses and the capital or interest costs)?
Country comments
Number of country comments for selected issue:127
  Date postedSourceComment
 11/27/2006AustraliaWe agree with the recommendations made by AEG
 10/11/2006ArmeniaThe National Statistical Service of the Republic of Armenia agrees with the most recent AEG recommendations.
 10/10/2006DenmarkWe are opposed to include yet another imputation which will in many cases be based on insufficient statistical information and arbitrary assumptions. This issue represents a resumption of a discussion that took place in connection with the 1993 SNA, without any new arguments being put forward.
The basis for the proposed imputation is the value of the fixed assets owned by the non-market producers. It is obvious from table 20 in the ESA 95 data transmission program that only very few of the 15 old Member States of the EU have this information. Since the statistical systems of these 15 countries are rather well developed, compared with many countries in the rest of the world, we would expect the situation to be even worse on a worldwide level.
Even if there were full information on the fixed assets a rate of return still have to be decided. The recommendation gives some indicative guidelines regarding this decision, but in the end it will be a decision of a somewhat arbitrary nature, and it would not even create “comparability” with market producers, as major parts of the productive assets are left out of this imputation.
 10/10/2006SwedenDisagreement with proposal
 10/10/2006GermanyIn the context of issues 15 and 16, it is demanded that, for measuring the output of non-market producers, calculatory interest on capital should be included in addition to existing expenditure elements (compensation of employees, consumption of fixed capital, intermediate consumption, taxes less subsidies).
We reject both changes.
That issue, too, was discussed and rejected already on the occasion of the 1993 SNA revision. Since that time, no new knowledge has been gained, and the relevance of those elements has not changed either. In major parts, determining that aggregate is based on imputations, especially on the interest rate to be used. Consequently, the results would not be free from arbitrary decisions. International comparability would be affected negatively.
 9/29/2006Bank of KoreaWe generally agree to the recommendation that the estimate of consumption of fixed capital for assets owned by non-market producers is replaced with an estimate of cost of capital services, including consumption of fixed capital as well as a return to capital, viewed as an opportunity cost. We also agree that the expected real rate of return on government bonds could be an appropriate indicator.
And we agree to the recommendation that land should be excluded from assets to applying a return to capital
We think, however, it is difficult to determine appropriate indicators for each asset because each asset has different characteristics. Because of those difficulties, more detailed information about assets is essential to apply this method.
 9/28/2006New ZealandConceptually we support the inclusion of the cost of capital services in the valuation of non-market output, particularly due to the consistency that this introduces with the market sector.
However, practically this is a significant change to introduce, and we have some concerns relating to its implementation. In particular, the scope of assets and the choice of parameters in the user cost equation. For example the exclusion of land and natural resources which can be very significant with regards the services provided by government. Does their exclusion undermine the objective that we are trying to achieve?
 9/27/2006IcelandAlthough Statistics Iceland is in general opposed to increased imputations in the current SNA update, issue 16 is an exception. From a theoretical point of view this kind of imputation is a definite step forward and can be implemented in Icelandic national accounts with an acceptable degree of accuracy.
 9/15/2006United KingdomWe agree with the compromise now being recommended by the AEG. There is strong logic for extending the costs recognised in the accounts and the change will also meet the needs of users of the accounts.
 9/15/2006LatviaAfter deep discussions and expert consultations we basically support the 1993 SNA Update Issues.
 9/15/2006NorwayFrom a theoretical point of view, Statistics Norway has sympathy with the proposal to include a return to capital in the measurement of non-market output. In this way, assets used in market production and non-market production would be treated in a consistent way. However, we are worried regarding the possibilities - i.e. the empirical basis - to actually implement the ideas. In particular, one will lack data for the rate of return to be used in the estimates. Even though the recommendations include some guidelines with respect to choice of rate of return, we are afraid one would end up with rather arbitrary choices, making international comparisons difficult.
 9/6/2006Monetary Agency of Saudi ArabiaAs regards to non-market producers, consider for instance someone who fixes his car by himself or provide instruction assistance to his kids for school, how can national accounting treat or estimate these 'intangible' services? Similarly, it is cumbersome to identify and correctly treat other services such as those provided by social non-profit organizations.
 8/21/2006Central Bank of El Salvador m4cbElSalvador16; m4cbElSalvador16-english;
 8/18/2006NetherlandsWe generally support the recommendations made at the Frankfurt Meeting of the AEG.
 8/18/2006ItalyWe are not in favour of including a return on capital for assets used in non-market production. Capital services in general are not well treated in the current system of national accounts. Their production is not appropriately described within the sequence of accounts. Including a return on capital would bring distinctive effects on the estimation of aggregates, and the case for instance of mixed income which is affected by the capital/labour breakdown is not clearly outlined. The global impact on GDP has not been sufficiently studied. Finally, there is a risk of adopting arbitrary criteria for the measurement of all aggregates.
 8/18/2006USAWe strongly endorse the recommendations of the AEG regarding the inclusion of the cost of capital services for government-owned and assets. This treatment is conceptually correct. Based on the recommendations approved by the AEG, statistical offices should be able to prepare high quality estimates that are consistent with the concept and internationally comparable. Implementation of this recommendation will provide users of the accounts with more appropriate information on which to base their decisions.
 8/2/2006European Central BankThe ECB generally supports the recommendations made by the AEG.
 8/1/2006National Bank/ National Bureau for Statistics MoldovaNational Bank and National Bureau for Statistics of Moldova agree with the recommendations made at the latest meeting of the AEG.
 8/1/2006Bank of BotswanaIt seems that the AEG remains determined to proceed with this proposal despite the fact that, even after several rounds of discussions, there continues to be widespread disagreement at both the theoretical and practical levels. The discussion at the February AEG meeting, does not appear to have allayed these concerns, and the project manager’s note seems not have taken into account the many comments that were made after the end of November 2005. In these circumstances, great caution in moving forward appears to be warranted. The level of disagreement supports the view that this should be considered as a radical change and, as such should be avoided.
 8/1/2006Reserve Bank of South AfricaWe carefully worked through all the issues and would like to give our general support to the recommendations made by the AEG.
 7/31/2006Macao SARStatistics and Census Service of Macao SAR agrees to the AEG recommendations and has no further comments.
 7/31/2006SpainEl INE no está de acuerdo en incluir en el marco central del Sistema de Cuentas Nacionales una tasa de rendimiento del stock de activos fijos de las Administraciones Públicas y las ISFLSH.
La valoración de la producción de no mercado se obtiene por agregación de los costes incurridos, hecho que la diferencia de la correspondiente a las ramas de mercado. Así, la inclusión de estos servicios de capital determinaría un excedente de explotación neto no nulo para las ramas de no mercado, hecho que debería evitarse, dada su difícil interpretación.
Además, deben tenerse en cuenta las grandes dificultades para poner en práctica este cambio, al exigir la disponibilidad de una estimación fiable y comparable del stock de activos fijos en poder de las administraciones públicas, stock del cual deberían excluirse aquellos activos utilizados en la producción de mercado de dichas administraciones.
Finalmente, pensamos que la introducción de estos servicios de capital en las administraciones públicas es un cambio fundamental en el Sistema, hecho que debería evitarse, tal y como se acordó en la Comisión Estadística de las Naciones Unidas, con respecto a los criterios sobre la actualización del SNA-93.
 7/31/2006Bosnia and HerzegovinaWe agree with AEG recommendations on the update of the 1993 SNA and do not have any further comments.
 7/28/2006VietnamWe agree with the outcomes of AEG.
 7/28/2006National Bank of SlovakiaConcerning the results of the most recent AEG meeting, we fully support the conclusions and recommendations made by the AEG.
 7/28/2006Lithuaniae are not in favour to record a cost of capital of own assets for government and non-market producers.
 7/28/2006SlovakiaWithin the outcome (a) we would welcome more specification of an appropriate indicator for the case where it is not possible to use the expected real rate of return on government bonds.
 7/28/2006Bank of PortugalBanco de Portugal would like to express general support for the recommendations made in the Frankfurt meeting of the Advisory Expert Group on National Accounts (AEG).
 7/28/2006Bank of PolandPlease find our general support for the AEG recommendations made during its recent meeting in Frankfurt.
 7/27/2006Bank of Sierra LeoneWe agree with the recommendations made by the AEG.
 7/27/2006EgyptFor practical reasons, we are not in favor of including the imputed capital services for assets owned by the government and other non-market producers.
 7/25/2006Central Bank of Costa RicaIn principle and from a theoretical point of view, we agree with the recommendation to estimate capital services for government owned fixed assets, nevertheless, based on implementation feasibility we prefer not to extent the measurement to city parks and historical monuments.  We recognize users’ need of this type of data, but we must express our concern regarding its implementation.
 7/25/2006Bank of ItalyWe broadly support the conclusions.
 7/24/2006Sweden m4Sweden16;
 7/24/2006Switzerland m4Switzerland16;
 7/24/2006GermanyThese specific recommendations imply the fundamental decision to include a capital ser-vice for non-market producers’ owned assets. As mentioned already in March 2005, we are not in favour of the including an imputed capital service for assets owned by general government and private non-profit institutions in the core system of the SNA. There are both, conceptual and practical reasons:
1. The recommendation ignores the difference between non-market producers and mar-ket producers (with regard to the valuation of GVA or output). In reality, however, there is a difference and that should be reflected in national accounts as well.
2. The creation of a virtual surplus for non-market producers should be avoided. This in-creases the risk of misinterpretation and non-comparability of NA data. How do we explain for instance to the public, a net operating surplus in the case of non-market producers.
3. Danger for business cycle analysis: How do we explain (in the worst case) an eco-nomic growth, which results only from the introduction of the virtual capital service.
4. There are important practical problems in addition:
a) Are capital stocks available and are they in comparable internationally? If the com-parability of capital consumption is doubted in para. 2.177 in the present SNA-93, why should the capital stock i.e. the net fixed assets after deduction of capital con-sumption be more reliable and comparable?
b) The accuracy of the allocation of the net fixed capital stock to non-market produc-ers (for instance in the case of outsourcing and public private partnerships) or the problem of excluding assets used for market production in the government sector.
c) The question of calculating the appropriate rate of return in a comparable way or the problem of calculating different rates of return over long periods (time series).
5. Finally, the recommendation is not in line with the fundamental criteria adopted by the UN Statistical Commission updating the SNA-93, because it is a fundamental change which should be avoided. In addition, it is an old issue that has been discussed and rejected before in the 1993 revision process, and we doubt that meanwhile there has been an increase in economic significance or an advancement in methodological re-search in this respect.
 7/24/2006National Bank of the Republic of AzerbaijanWe 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.
 7/10/2006DenmarkDisagreement
Statistics Denmark does not support including a return to capital in the estimation of government output. Including a return to capital is a major change to the System and such radical changes should be avoided.

However, if a majority support this major change to the System, Statistics Denmark supports the proposed calculation method. We believe it is important that a return to capital appear for all fixed asset and only fixed assets. Among the countries which have estimates for some of the assets from the balance sheets, it is most common that they have estimates for fixed assets but not other classes of assets such as inventories and land. It would make implementation of a return to capital in the System much easier if inventories and land are excluded.
 3/17/2006Cuba m3cuba;
 2/23/2006South African Reserve BankWe would like to confirm our earlier statement that we are not in favour of this recommendation. However, since it is in principle accepted by the AEG, we feel strongly that only assets that have an equivalent use in market production should be included.
 1/27/2006Central Bank of El SalvadorNo estamos de acuerdo en su incorporación debido a que su producción está definida típicamente como no de mercado. La imputación de los retornos sobre el uso de los activos del gobierno en la práctica presenta dos limitaciones fuertes. (i) la contabilidad gubernamental no está en todos los casos con base devengado, sino caja, lo cual repercute en las valoraciones bajo el esquema del SCN93 (ii) la valoración de bienes inmuebles públicos de uso colectivo, no tiene necesariamente una rentabilidad de mercado sino una rentabilidad social cuya medición no es fácil y no siempre permite una comparabilidad internacional.
 1/17/2006Central Bank of West African States cb_WesternAfricanStates16;
 1/17/2006KuwaitKuwait agrees to include a return to capital on non-financial assets used in non-market production.
 1/3/2006CubaConsideramos que los servicios de capital deben registrarse como parte del superávit bruto de operación y formar parte del VAB y considerarlo dentro del marco central
Consideramos en el caso de la producción del gobierno deben incluirse también los servicios de capital
 1/3/2006SingaporeWe agree that we should in principle include a return to capital on non-financial assets used in non-market production. The feasibility of implementing the new conceptual treatment remains a concern, however.
 12/22/2005Serbia and MontenegroWe agree with the suggested changes.
 12/15/2005RussiaEstimation of non-marketed services cost including assessed return on capital value looks quite impracticable. Including the return on capital in non-marketed services will increase their cost and the government expenditures on collective consumption. Besides, the estimation of return on capital should be based on reliable data on market value of capital assets in this sector, and it is very hard to accomplish.
Also, the application of various approaches to different assets types looks quite problematic for practical reasons.
 12/14/2005FranceL'INSEE soutient le principe consistant à retenir un excédent net d'exploitation non nul pour mesurer la production non marchande des admnistrations publiques. Ce résultat sera obtenu en appliquant un taux de rendement au stock d'actifs fixes détenus par les branches non marchandes.
Ce changement est de nature à mieux traduire le calcul économique de ces unités. Le taux choisi doit refléter la particularité du comportement des administrations.
L'INSEE souligne cependant les difficultés pratiques de mise en œuvre de ce changement, qui suppose la disponibilité d'une évaluation des stocks d'actifs fixes détenus par les administrations.
 12/13/2005CanadaCanada approves the addition of adding a return to capital to measurement of the cost of government services. This would bring the valuation of non-market output in line with that of market output. In terms of the scope of assets included in the AEG discussion certainly a rate of return should be added to the assets used by civil servants to produce government services should be included. Assets owned by the government but provided to the economy for general use should be included but care should be taken not to double count returns already included in other marketed outputs produced by using those assets. Imputing a rate of return to inventories, land, and natural resources held by the government has not yet been considered in the Canadian context.
 12/12/2005Bank of KoreaWe agree with the recommendations made at the July 2005 meeting of the AEG, viewed as an opportunity cost and the definition of an asset.
However, we think that it might be necessary to draw a broader consensus on the coverage of assets and the rate of return to be applied to assets. It is somewhat difficult to calculate the value of capital services on assets such as roads, other infra structures and city parks serving the economy and community at large. Moreover, due to insufficient data on government and other non-market producers' owned assets, it doesn't seem easily practicable to choose the rate of return.
 12/12/2005AustriaWe are not in favour of the recommendations made by the AEG because of conceptual and practical reasons:
The difference between market and non-market producers regarding their role in the production process is ignored when giving up the convention of zero net operating surplus.
Another imputation to the system would be added without empirical foundation.
The difficulties in the selection of the rate of return that would lead to major distortions between countries.
 12/12/2005MexicoLas mediciones de servicios de capital para complementar la cuenta de producción mejoran la información para hacer análisis de productividad, por el hecho de que contribuyen a tener mediciones más exhaustivas de los componentes de la función de producción. Sin embargo, considerar estas recomendaciones como una obligación en la estructura central de las cuentas podría traer problemas en países que aún están haciendo esfuerzos por mejorar las mediciones de sus stocks de capital, o que están construyendo los mismos. Sería recomendable introducir las mediciones de servicios de capital como opcionales.
Adicional a la anterior opinión que ya ha sido expuesta anteriormente, se reconoce que si el tópico 15 Costo de los servicios de capital se decidió incluirlo en la actualización del SCN en cuadros complementarios, éste debería también ser considerado como tal. En cuanto al alcance de la medición se considera que la medición de los servicios de capital de la tierra, los activos fijos y los del subsuelo serían los aspectos más importantes de la propuesta y los que recompondrían las mediciones de los productores no de mercado de manera más sustantiva.
 12/9/2005European Central BankIn its previous letter the ECB expressed its strong reservations regarding the proposals made by the AEG on this item in December 2004. They are still valid, even if the results of a world-wide consultation will have to be awaited; its follow-up on a one-on-one basis as agreed during the July 2005 meeting will be brought forward to the forthcoming AEG meeting in January 2006. The ECB still assumes that the implementation issues are difficult to tackle (coverage of assets, specification of rate of return) - with negative effects On the quality of compiling GDP and GNI
 12/5/2005Denmark m3Denmark16;
 12/2/2005NetherlandsWe generally support the recommendations of the AEG.
 12/2/2005AustraliaAustralia agrees with the AEG in principle recommendation and supports the follow-up of comments from global and country consultations on this issue. However, while we appreciate that practical questions such as data availability need to be considered, these should not override the underlying principles.
Australia believes that, on balance, capital services should only be generated in respect of those assets which 'contribute to production'. Assets which provide services to the community at large (eg parks) should not be included as they do not generally contribute to production as it is currently defined.
 12/2/2005TurkeyWe agree with the recommendations made at the July 2005 meeting of the Advisory Expert Group on National Accounts.
 12/1/2005GermanyWe are not in favour of the recommendations because of general, conceptual and practical reasons:
1. The recommendation is not in line with the criteria of ISWGNA for updating the SNA be-cause it is a fundamental change which should be avoided. In addition, it is an old issue that has been discussed and rejected before in the 1993 revision process, and mean-while there has been no increase in economic significance nor an advancement in methodological research in this respect.
2. The recommendation ignores the difference between non-market producers and market producers (with regard to the valuation of GVA or output). In reality, however, there is a difference and that should be reflected in NA.
3. The recommendation seems inconsistent, since the AEG has favoured not to implement the concept of capital services for market producers (as an “of which position” of sur-plus). Why should this concept be implemented for non-market producers (NMP)?
4. The creation of a (virtual) surplus for non-market producers should be avoided. This in-creases the risk of misinterpretation and non-comparability of NA data.
5. More generally: the recommendation would produce more imputed elements in NA and fewer market transactions, i.e. NA would be less suitable for important questions like business cycle analyses or analyses of income or tax burden.
6. There are important practical problems in addition:
a) The problem of calculating the different assets used in the local kind of activity units in different branches within the government sector; exclude assets used for market pro-duction in the government sector.
b) The problem of calculating the appropriate rates of return, differentiated by type of as-set used in different branches.
c) The problem of calculating different rates of return over long periods (time series).
 12/1/2005NorwayOn this subject we should like to repeat our former comments:
"We agree with the proposal to include a return to capital in the measurement of non-market output. To ensure international comparability, the SNA should include specific guidelines on how to calculate the rate of return to be used in the estimates. The range of assets covered should be those that are covered by consumption of fixed capital in the current estimates of non-market output, i.e. including infrastructure assets and construction work, land improvements etc. on assets such as city parks."
 12/1/2005BrazilThis issue poses a greater question: how many countries are able to implement it. The CONAC (Brazilian National Accounts Coordination) are concerned about the number of issues that can not be implemented in, even long term.
Despite this first question we believe the recommendation has a good theoretical background but some questions that need clarification. We do not agree with the inclusion of some assets like city parks, monuments etc. We would accepted assets used by the employees in their work or bringing benefits. It is important the borders be better clarified.
We point also the definition tax of return to be used is unclear.
Brazil does not plan to implement this recommendation. In the moment, ours priorities are others and we do not the data available to develop a estimation like that.
 12/1/2005Central Bank of NicaraguaNo se discute el principio teórico de que cualquier activo debe producir un retorno económico por su uso, aunque no se considera plausible calcular el mismo, para el sector institucional de gobierno general en Nicaragua, debido a que en el corto plazo:
• No se dispondría de una base de datos sobre stock de capital de productores no de mercado.
• Además de la complejidad que representa estimar una tasa de retorno por tipo de activo.
 12/1/2005IranAlthough calculating the cost of capital services will furnish analysts and policymakers with a suitable tool, they face some problems due to inaccessibility to the appropriate data, It is the same case with the consumption of fixed capital formation.
 12/1/2005MozambiqueWe are pleased to express our agreement on the AEG recommendations.
 12/1/2005Bank of BotswanaIt is very striking that while the expert group continue to strongly support the introduction of this change, at least in some form, this is the only issue among those considered at the July AEG meeting where there is significant dissent shown in the comments from countries. This is to the extent that comments fully in favour of the proposal were in the minority. Moreover, this dissent is spread across the full range of countries (i.e. there are several among the most developed countries included in the group that have doubts), while the range of criticisms is also extensive, covering both theoretical and practical issues. The proposed in-depth consultations with national agencies is welcome, but the range and extent of criticisms that have been made already should serve to temper the continued enthusiasm at the expert level to proceed with considerable caution.
 12/1/2005SloveniaWe are not in favor of the proposal to include costs of capital services into the measurement of non-market producers’ output. From the theoretical point of view we basically agree with the proposal although we think that measurement of government and other non-market producers output is an agreement and we do not think that inclusion of capital services as a part of operating surplus would substantially improve the comparability of market and non-market production.
We are not in favor of the inclusion of capital services mainly because of practical reasons: we think that it is very difficult to define appropriate rates of return which can be used for calculating capital services for individual categories of assets. Due to this reason we think that calculations of capital services would be in many cases too artificial and not very appropriate for economic analysis.
 12/1/2005United KingdomWe agree with the AEG decision that the cost of capital of own assets for both Government and other non-market producers also be reflected in the accounts. However, on scope, we feel that, only assets that have an equivalent use in market production should be included.
 11/30/2005Slovak RepublicGenerally it appears, that the number of imputations and its share in national accounts system is gradulally extended. This fact may contribute to international comparability. It should be suitable to include it to supplementary and not core accounts. Amounts related to including a return to capital on non-financial assets used in non-market production have significant influence on non-market output also from the viewpoint of nature of this output (some imputations have a market nature), which is benefit from the viewpoint of international comparability, but this procedure is hardly accepted by domestic experts from the viewpoint of actual economic efficiency. Therefore we welcome further discusion in this issue.
 11/30/2005Palestinian Central Bureau of StatisticsThe new recommendation need to be clarified more and we hope to have more elaboration an the logic and idea behind the new treatment.
 11/30/2005FinlandStatistics Finland sticks to the earlier opinion on reservations (below) about the proposed new treatment. Though we can see that there are certain theoretical grounds to change the present treatment, we would like to emphasise many practical problems relating to the issue: determining rate of return, uncertainties of calculating capital stock models versus the important role of government sector.
If the proposed change will be executed we hope that it will only cover the assets under present calculations of consumption of fixed capital but no other asset types like land.
Our earlier comment:
Statistics Finland have large reservations about introducing capital services of government owned assets. We also consider that in the country comment by Sweden there are good aspects concerning the concepts 'non-market'-'non-profit'.
We would like to emphasise the following points:
If the return to capital is to be viewed as a opportunity cost, as AEG has proposed, then what is the "opportunity" in the case of government. When a private corporation is planning an investment, it, in theory, compares the estimated net present value of the investment to its target rate (e.g. ROI, ROE) which should always exceed the risk-free interest rate. If the NPV rate of return does not exceed the target rate, the company should forego the investment and return the cash to the owners.
A democratic, elected government, however, cannot just stop investing and providing basic services, even if the estimated NPV of its investments falls below risk-free interest rate. Therefore the rate of return for government owned assets can differ from private sector. Since government institutions are non-profit organisations their implicit "target rate" for investments is probably much lower than in the private sector. It would be interesting to hear why exactly should we assume a larger than zero real rate of return for government owned assets and how can we know what the actual rate is.
One solution could be dropping the concept of opportunity cost and instead focusing on the cost of capital. In this case the rate of return for government owned assets could be set equal to the coupon rate of government bonds, because that is the price the government has to pay in order to raise cash for new investments. However, as Netherlands has pointed out, the return on government bonds is higher in countries with high country risk and thus these countries would have higher level of public sector production and operating surplus. This is probably not desirable.
Governments also have other sources of cash besides borrowing, like taxes, which implies that the cost of capital is actually lower than the return on bonds. Finally, it can be argued that resorting to cost of capital measures is evading the real problem; that we do not know the rate of return for government owned assets.
The whole issue crucially depends on the choice of the measure for the rate of return to capital. If we cannot find a satisfactory measure, we should stick with the current SNA.
 11/30/2005ItalyIstat fully agrees with the recommendations made at the July 2005 meeting of the Advisory Expert Group on National Accounts.
 11/30/2005SwitzerlandWe support the AEG proposal to include the return to capital in the costs of using non-financial assets also for non market producer. In effect this return to capital would reflect the opportunity cost of capital which essentially exists for market as well as for non market producers.
About the range of assets which should be covered, we agree on the principle of including all types of non financial assets. However it is important to point out the fact that the data availability on capital stock and the choice of the appropriate rate of return are major weaknesses of this proposal.
A clear analysis of the feasibility of this recommendation should be undertaken in order to find a pragmatic solution for the implementation.
 11/30/2005State Bank of PakistanWe have gone through recommendations made by Advisory Expert Group (AEG) and fully agree with them.
 11/30/2005State Bank of VietnamWe don't agree with the recommendations suggested that including an estimate of capital services instead of consumption of fixed capital will increase the value of gross output and net value added for non market—production when it is estimated as the sum of costs incurred.
This recommendation is infeasible in practice, Because:
First: Public non- financial assets such as: buildings, computers, vehicles, roads, bridges, dam, city parks and historical monuments are very difficult to collect information and can not measure the sum of cost incurred, especially roads, bridges, city parks and historical monuments.
Second: If we change method of measure, to make GDP to be increased is not significant in real value of GDP.
Services from Government — owned assets should be treated the same of 1993 SNA treatment.
 11/29/2005TadjikistanМы не согласны с предложением включать прибыль на капитал в измерение нерыночных услуг по следующим причинам.
Во-первых, на наш взгляд, неясен экономический смысл такой условной оценки, который приведет к увеличению ВВП, и пользователям данных будет трудно объяснить, что такое прибыль в счетах государственного управления от использования государственных административных зданий для предоставления нерыночных услуг. Это потребует условных оценок в финансовом счете (F.5 – Акции и другие виды участия в капитале) и в балансе активов и пассивов (та же категория).
Во-вторых, с более практической точки зрения, для вышеупомянутых условных оценок требуются данные о запасах основных фондов в секторе государственного управления в рыночных ценах, которые во многих странах (по крайней мере, во всех странах СНГ) в настоящее время отсутствуют или плохого качества. Неясно, какой процент на капитал можно использовать для этих условных оценок. В общем, преимущества условных оценок неясны, и их введение может осложнить международные сопоставления ВВП.
 11/29/2005EstoniaAgainst proposals with respect to international comparability of GDP. Also, the specific guidelines how to calculate the rate of return should be determined and the scope of fixed assets still needs a clarification.
 11/29/2005People's Bank of ChinaI agree with your improvements and have no other suggestions.
 11/21/2005USAWe again strongly endorse the principle to include a return to capital on non-financial assets used in non-market production.
 6/30/2005Slovak RepublicWe do not support counting of capital services influence on production mainly due to the reason, that non market production is in the question and using of designed approach artificially increases its use. Furthermore, from the practical viewpoint, availability of relevant information is questionable.
 6/30/2005Serbia and MontenegroWe agree with the recommendations made at the December 2004 meeting of the AEG.
 6/22/2005Bank of GhanaThe 1993 SNA implies that by convention, there is a zero net operating surplus for non-market producers, which means that there is no return to capital on their assets. During the development of the 1993 SNA it was proposed to change this but the proposal has come up again, in light of work to integrate the capital services approach to the contribution of non-financial assets to production. We generally agree with the Canberra II Group on the recommendations for change.
 6/3/2005Central Bank of VenezuelaWe do not agree with assigning a single value to assets that are similar in all respects, notwithstanding the institutional unit that owns them. We believe that the use of fixed assets in productive processes has different purposes depending on the type of producer. Market producers make their investment decisions to obtain a yield, while nonmarket producers make investments primarily to produce nonmarket services for the benefit of households or the general welfare.
The determination of a return on capital for nonmarket producers is incompatible with the primary nature for which such institutional units were created. In this regard, the net operating surplus of these units should always be equal to zero, as recommended by the SNA93.
In countries where interest rates and inflation are high, the impact of the incorporation of capital services on Gross Domestic Product (GDP) levels could be significant, generating a change in the productive structure by increasing the Gross Aggregate Value (GA V) of nonmarket producers.
We recommend treating the issue of capital services for nonmarket producers as a satellite account where these sectors could be evaluated and, if they are not owners of nonfinancial assets, as productive units generating a surplus.
 5/27/2005Pakistani. Agrees with proposal to include a return to capital and which is an opportunity cost, in the measurement of non-market output. However, data availability on capital stock and selection of rate of return will be a problem.
ii. In terms of the range of assets, which could be covered, agrees including those assets in the generation of government output similar to those assets used in market production.
 5/23/2005IsraelWe are strongly in favor of including capital services from all government-owned assets, and also for all assets owned by other non-market producers. Including such capital services will improve comparability between sectors within countries, and between countries, where the institutional structure differs. We also think that the decisions on purchasing or producing certain assets in government (or the collective that elected the government) are made taking into account opportunity costs, similarly to decisions in private units, although some of the government-owned assets have a longer service life and may render collective services. Government has the possibility to choose between investment of the money or spending the money on various current purchases. However, we do think that adding clear guidelines for measurement and recommendations for coverage is extremely important.
 5/20/2005National Bank of PolandWe are of the opinion that such government owned assets as road and other infrastructure assets including city parks etc. do not contribute to the production of government services. Information - if available - on such assets are of interest of its own but should be shown separately from other government owned assets. Detailed recommendations could be prepared after general agreement is reached.
 5/19/2005Bank of IndonesiaBI supports the AEG recommendation conceptually, but the value of this item is not significant to be recorded.
 5/18/2005Central Bank of ChileThis SNA recommendation, also contemplated in the IMP Government Finance Manual, causes countries to try to update their asset statistics. It is the entities producing government data that should discuss and examine the feasibility of recording capital services so as not to diminish the consistency of the Government Finance Manual with the National Accounts.
We agree that capital services should be included in the measurement of nonmarket output, particularly for services that directly benefit the people for which there is no market price.
The valuation of capital services at market prices is debatable since the measurement of nonmarket output presents a noncash surplus. Furthermore, it would not be symmetrical with the measurement of the output of other services provided by general government, for example, public health and public education. Despite there being a market price for such services, they are measured by the sum of their costs.
We certainly agree on the need to measure and incorporate capital services in the measurement of government output, provided that this does not alter the principles of the SNA, that is to say, that services produced by nonmarket producers cannot be greater than the costs involved, including in such costs the consumption of fixed capital.
In order to implement this definition, however, you come up against the restriction of availability of detailed series on the stock of a...sets. In this regard, it is not sufficient to perform studies for certain years. Rather, the availability of statistical sources is required, ensuring the ongoing recording of statistics over time.
Finally, if on this issue, as with other issues, no consensus is reached on international application, it is always possible to progress on these issues through recommendations for satellite accounts or supplementary measurements.
 5/16/2005JordanDos is not in favor of including a return to capital, viewed as an opportunity cost, in the measurement of non-market output because there is no reasoning to treat the non-market producers differently than the market producers. If the concept of capital service should apply in the non-market producers it should be applied for market producers also. If it is decided to consider it as part of output then the DOS recommend to calculate return on the fixed assets only which are similar to those used by other sectors such as buildings, machinery and transport equipments. Also it should be clear that calculating the capital stock of other assets is very difficult in practice beside the difficulties in calculating a proper rate of return on such assets which may lead to comparability problems between countries.
 5/16/2005Eastern Caribbean Central BankWe agree in principal to impute a rate of return on capital used in non-market production. However, we have problems with the availability of data in our region, on capital stock for public administration. In terms of the range of assets that should we covered, we are in favor of assets in the generation of government output similar to those assets used in market production. We do not agree to include assets used by the community at large such as city parks and monuments.
 5/13/2005State Bank of PakistanWe agree with the proposal to include a return to capital, which is an opportunity cost, in the measurement of non-market output. However, data availability on capital stock and selection of rate of return would be a potential problem. In terms of range of assets which could be covered, we also favour including those assts in the generation of government output similar to those assets used in market production.
 5/12/2005Czech National BankI support the opinion expressed by Mr. Heller (the Czech Statistical Office expert).
 5/12/2005European Central BankThe ECB has strong reservations regarding the proposals made on item 16: The ECB shares the hesitation of several AEG members on this topic. The proposal has clear advantages as it avoids that the contribution to GDP of an activity run by a non-market entity would be lower - all other things equal - than the same activity run by a market producer. However, the interpretation to be given to the imputed rate of return may not be straightforward. Furthermore, the difficulties to fix the exact coverage of involved assets and the rate of return to be applied would adversely affect the transparency and reliability of GDP and GDI estimates. Against this background. we would favour a complete definition and evaluation of the issue at stake before a final decision is taken by the AEG.
 5/11/2005IndiaIndia agrees with the AEG decisions to include return of capital for those of type of assets used by civil servants in course of their work e.g. computer, vehicle, building, and roads and other types of infrastructure. However, it may be difficult to include the assets of monuments, land, etc. for measuring returns to capital, in view of lack of data and imputation methods.
 5/10/2005USAThe U.S. Bureau of Economic Analysis endorses the “strong support in principle for including a return to capital… in the measurement of non-market output.” Use of depreciation as a measure of the value of services of government fixed assets is a partial measure of the total value. In theory, the service value of an asset should equal the reduction in the value of the asset due to its use during the current period (depreciation) plus a return equal to the current value the asset could earn if invested elsewhere (net return). A report by outside experts convened by the National Research Council also endorsed this principle (see Courtenay M. Slater and Martin H. David, eds., Measuring the Government Sector of the U.S. Economic Accounts, Committee on National Statistics, National Research Council, National Academy Press, Washington, DC, 1998.)
 5/10/2005Bank of BotswanaThis proposal may appear fine in theory; but it cannot be assessed sensibly without specific proposals for practical application, and there may be serious problems at this stage. The current measure has the obvious drawback of inconsistency between sectors, but at least the consumption of fixed capital measure has the advantage of being less controversial than some imagined rate of return on government assets, which could be open to manipulation. A change which will significantly affect the level of GDP may require cross-country coordination and, since it is based on an imposed assumption, is likely to be controversial. To replace one artificial measure of output (consumption of fixed capital) with another (assumed rate of return) is limited progress, at best.
 5/9/2005AustraliaAustralia agrees with the AEG in principle recommendation. We have not yet reached a position on the scope of government owned assets on which a return to capital should be included in government output.
 5/9/2005MaldivesSince government is producer of non market goods and services, then it is but logical that there should be no service of capital. In general the decision of government for capital formation is not based on return to asset nor opportunity cost since government is a non- market producer. Capital service on government owned assets is important component for productivity analysis and is of interest to calculate. But the lack of reliable data hinders this attempt. We wish to agree with the AEG on the issue of data availability.
 5/9/2005United KingdomWe strongly support the recommendation to impute a rate of return on capital used in non-market production. However, issues have to be resolved on what rate of return to be used on one hand and, on the other, the scope of assets to be included. And, that any decision reached must take into account the feasibility of implementation. Further than this, it should be noted that the cost of capital services should be imputed for all non-market production, and not only limited to government owned assets.
 5/9/2005Central Bank of IranWe agree with the principle to include a return to capital for non- market producers, although this approach is a significant change to the SNA and it will lead to a significant increase in both GDP and NDP. This treatment ensures that assets owned and used in non-market production are valued in the same way as assets owned and used in market production and implies internal consistency in the valuation of assets and recognition of their role in production. However in terms of the range of assets which could be covered, we are in favor of assets in the generation of government output similar to those assets used in market production. We disagree to include other assets due to different nature of government services with private services.
 5/9/2005ItalyWe agree from a theoretical point of view on including a return to capital, viewed as opportunity cost, in the measurement of non-market output. We have no problems with regard to the availability of data on capital stock of Public Administration.
About the types of assets for which it is possible to estimate a return to capital, we agree on including the assets used by Government employees in the course of their work (e.g. computers, vehicles, building) and assets bringing benefits to the economy at large (e.g. roads and other infrastructure). We do not agree on including assets used by the community at large, such as city parks and monuments and land (they are non-produced assets). Anyway, it would be better not to introduce significant changes in Public Accounts.
 5/6/2005Commonwealth of Independent States We do not agree with suggestion to include a return on capital in the measurement of non-market services. The reasons for that are as follows.
Firstly, in our view economic meaning of such an imputation which would inflate GDP is not clear and it would be difficult to explain to the users of data the operating surplus in government accounts due to use of administrative government buildings employed for provision of non-market services. This would seem to require imputation in the financial account (F.5- shares and other equity) and in balance sheet ( the same category).Secondly from a more practical point of view, the imputation in question requires data on stocks of fixed assets in government sectors in market prices which in many countries ( at least in all CIS countries) are not immediately available or of poor quality. It is not clear what rate of return could be used for the imputation. All in all, the merits of the imputation are not clear and its introduction may complicate the international comparisons of GDP.
 5/6/2005FinlandStatistics Finland has large reservations about introducing capital services of government owned assets. We also consider that in the country comment by Sweden there are good aspects concerning the concepts 'non-market'-'non-profit'.
We would like to emphasise the following points:
If the return to capital is to be viewed as a opportunity cost, as AEG has proposed, then what is the "opportunity" in the case of government. When a private corporation is planning an investment, it, in theory, compares the estimated net present value of the investment to its target rate (e.g. ROI, ROE) which should always exceed the risk-free interest rate. If the NPV rate of return does not exceed the target rate, the company should forego the investment and return the cash to the owners.
A democratic, elected government, however, cannot just stop investing and providing basic services, even if the estimated NPV of its investments falls below risk-free interest rate. Therefore the rate of return for government owned assets can differ from private sector. Since government institutions are non-profit organisations their implicit "target rate" for investments is probably much lower than in the private sector. It would be interesting to hear why exactly should we assume a larger than zero real rate of return for government owned assets and how can we know what the actual rate is.
One solution could be dropping the concept of opportunity cost and instead focusing on the cost of capital. In this case the rate of return for government owned assets could be set equal to the coupon rate of government bonds, because that is the price the government has to pay in order to raise cash for new investments. However, as Netherlands has pointed out, the return on government bonds is higher in countries with high country risk and thus these countries would have higher level of public sector production and operating surplus. This is probably not desirable.
Governments also have other sources of cash besides borrowing, like taxes, which implies that the cost of capital is actually lower than the return on bonds. Finally, it can be argued that resorting to cost of capital measures is evading the real problem; that we do not know the rate of return for government owned assets.
The whole issue crucially depends on the choice of the measure for the rate of return to capital. If we cannot find a satisfactory measure, we should stick with the current SNA.
 5/6/2005Macao, SAR ChinaIn fact, it is infeasible to estimate the value of capital services on assets such as roads, city parks, monuments,…etc. It will be helpful to provide us methods for estimating the value of the assets and the rate of return.
 5/6/2005VietnamWe don’t agree with the recommendations suggested that including an estimate of capital services instead of consumption of fixed capital will increase the value of gross output and net value added for non market production when it is estimated as the sum of costs incurred.
This recommendation is infeasible in practice, Because:
First: Puplic non- financial assets such as: buildings, computers, vehicles, roads, bridges, dam; city parks and historical monuments are very difficult to collect information and can not measure the sum of cost incurred, especially roads, bridges, city parks and historical monuments.
Second: If we change method of measure, to make GDP to be increased is not significant in real value of GDP.
 5/6/2005TurkeyWe agree with the recommendations made by the Advisory Expert Group on National Accounts at its second meeting in December 2004.
 5/4/2005DenmarkStatistics Denmark accepts that in principle the proposed changes to the SNA are in accordance with economic theory and are in principle a better way to estimate government output.
However, practical reasons do that we oppose all the changes.
First, estimation of capital services requires full estimation of capital stocks, which is only done in 7 (of which only two countries with breakdowns) out of the 15 old countries in the European Union. In the rest of the world, the number of countries that estimate capital stocks are much lower. Since the required input data to estimation of capital services is not available in most countries, it is unlikely that a large number of countries would be able to estimate capital service in a satisfactory manner.
Second, we agree with AEG that the rate of return for government owned asset is not well defined with respect to estimation method. A range of estimates could be used.
Third, some unsolved problems regarding estimation of capital stocks remain. Most countries use the PIM method – written down acquisition prices – for valuation of capital stocks. But these estimations do not reflect changes in market prices in the market for used assets. This is especially important for buildings – which is the most important type of asset for most governments.
Since the PIM-estimated value of the assets does not necessarily reflect the market value of the asset, adding an opportunity cost – in principle reflecting the market value – for holding the assets is problematic.
 4/29/2005NorwayStatistics Norway agrees with the proposal to include a return to capital in the measurement of non-market output. We are, however, worried about data quality and the possibility to find an acceptable rate of return to be used in the estimates. Thus, to ensure international comparability, the SNA should include specific guidelines on how to calculate the rate of return. By our view, the range of assets covered should be those that are covered by consumption of fixed capital in the current estimates of non-market output, i.e. including infrastructure assets and construction work, land improvements etc. on assets such as city parks.
 4/29/2005Central Bank of ColombiaEl Banco de la República de Colombia considera que el registro de los retornos sobre el uso de los activos del gobierno es recomendable, bajo una contabilidad completa de causación (devengado), es decir donde estén armonizadas las cuentas del SNA93 con los otros sistemas de información. Sin embargo, en la mayoría de los países como en Colombia, aún no se tienen las cuentas del gobierno, en base devengado, como lo recomienda el nuevo manual de las finanzas públicas (GFSM01). Esto implica que haya serios limitantes de la información sobre el valor de mercado de los activos del gobierno y, por supuesto, sobre sus retornos. La restricción es mayor cuando se pretenden valorar activos como los parques públicos, museos, etc, cuyo consumo es no rival, en cuyo caso la tasa social de retorno, que sería útil para valorar su coste de oportunidad, no es fácil de determinar.
 4/28/2005Trinidad and TobagoWe agree with the recommendations of the AEG.
 4/25/2005Bank of Sierra LeoneWith reference to the above subject we agree with the recommendations of the Expert Group on National Accounts (AEG) especially where there is harmonization of definitions and other concepts between the Balance of Payments (BOP) and system of National Accounts (SNA).
 4/14/2005Bank of FranceS’agissant de l’inclusion éventuelle dans la production des administrations publiques du coût des services du capital fournis par les actifs (point 16), les services de la Banque de France partagent les interrogations de plusieurs membres du Groupe d’experts. En effet, une telle approche aurait l’inconvénient de conduire à l’inclusion d’éléments purement estimatifs dans le Produit intérieur brut et les revenus des administrations publiques.
 4/13/2005Bank of KoreaWe agree with the proposed recommendation in principle, viewed as an opportunity cost and the definition of an asset.
However, we think that it might be necessary to have an additional discussion on the coverage of assets and the rate of return to be applied to assets. It is somewhat difficult to calculate the value of capital services on assets such as roads, other infra structures and city parks serving the economy and community at large. Moreover, due to insufficient data on government owned assets, it is not easily practicable to choose the rate of return.
 4/12/2005Central Bank of HondurasEl cambio propuesto es un cambio significativo al SCN 93, con ventajas en su admisión al obtener una mayor consistencia en la valoración de activos en el sistema y un reconocimiento de su papel en la producción.
 4/12/2005GreeceWe agree with the recommendations of the AEG on the issues for which a decision has been taken, at the December 2004 meeting of the group.
 4/11/2005MalawiI fully endorse the recommendations of the Expert Group on National Accounts.
 4/11/2005Central Bank of Kuwait- Kuwait agreed to include a return to capital in the measurement of non-market output.
- Kuwait favored to include those assets in the generation of government output similar to those assets used in market production.
 4/11/2005Bank of Tanzania/National Bureau of StatisticsComputation of government output in Tanzania using National accounts system has been not easy mainly due to problems emanating to categorization of items to be included. Thus concerns about the rate of return to be chosen and availability of data for capital stock are relevant and should be taken into consideration when reviewing SNA 1993.
 4/11/2005Hong Kong, ChinaWe agree in principle that a return of capital for non-market production should be estimated. Regarding the range of assets to be included, we support that in addition to assets used directly in the production process, e.g. machinery and vehicles, assets providing benefits to the economy and the community such as buildings and structures should be covered on conceptual grounds. Nevertheless, we share the AEG concern on rate of return to be chosen and availability of estimate of capital stock. Clear guidance on these issues should be provided.
 4/11/2005The NetherlandsIn relation to the issue of government owned assets (cost of capital services), we would like to respond to the questions raised in the excellent issue paper by Anne Harrison:
a) With regard to the selection of assets that should be subject to a return to capi-tal we think that a rate of return should be included for fixed assets only (and thus not for non-produced assets). This is a pragmatic and transparent choice which will not lead to an undesirable expansion of data sources required for measuring government output.
b) We do not see why a rise in the value of government capital (such as cultural heritage) should lead to an increase in the rate of return for these assets. We would argue that such a value increase would simply lead to a revaluation of the asset.
c) We agree that the capital service corresponding to these returns to capital should be used in the valuation of government output instead of consumption of fixed capital. Yet, the user cost model indicates that (foreseen) changes in asset prices should be reflected in the costs of capital as well. We do not plea for doing so in the valuation of government output. However, we think it should be indi-cated that including a return to capital does not necessarily mean that the full costs of capital are taken into consideration. We have some concern about the choice of an appropriate rate of return. The most obvious choice is probably a of return to government bonds. This return may differ between governments as a result of differences in risk premiums. It seems perfectly logical that a higher risk premium leads to higher costs of capital. However, it is intuitively less obvious that lesser solvent governments have hig-her production compared to higher solvent governments.
d) Further clarification of the benefit criterion (benefits for the entire community versus the owner of the asset) would indeed be appropriate.
 4/11/2005IranWe will encounter some problems in the calculation due to suitable data inaccessibility though calculation of cost of capital services provides the analysts and policymakers with a useful means, as it is the case of consumption of fixed capital formation.
 4/11/2005Germany m2(c)de16;
 4/11/2005Philippines Estimating cost of capital services for government owned asset in the Philippines is faced with a number of problems such as coverage (inclusion of roads, bridges, parks, and other infrastructures), data on the service life of the asset covered, use of an appropriate rate of return, and availability of data on capital stock. Can we seek clarification on the use of the discounted method for estimating cost of capital services? Can you cite actual examples from countries already using the suggested method?
 4/11/2005Czech RepublicAs far as Government owned assets – cost of capital concerns, we do not recommend to agree with so significant change in the revision of SNA 1993. In the Czech Republic, treatment and valuation of fixed assets is the same for market and non-market production. The starting point is the assessment of gross current market price or replacement cost of the asset. Then net value is estimated according to the share of remaining service-life in the full service-life of the asset. The result should be compared with actual prices of comparable second-hand assets if they are sold/purchased. We don`t think that anything should change such our attitude be “the capital service” calculated by whatever way.
It is not clear why the net operating surplus should be created only by non-financial capital and not by human work.
We don`t support extending of weakly founded imputations in national accounts because they may make easier unrealistic adulteration of so important aggregates as GDP, national income, savings etc.
 4/11/2005Russian FederationEconomic meaning of imputing operating surplus as a component of the value of non-market service is not clear. Besides estimation of operating surplus in question will require reliable data on the market value of fixed assets which are not immediately available; implementation of the IMF Manual on GFS will facilitate obtaining these figures. The proposal to treat different various types of government assets in the context of imputation of operating surplus raises some doubts on both conceptual and practical grounds.
 4/11/2005South African Reserve Bank We are not in favour of this recommendation. The valuation of the range of asset classes could become rather subjective, the implementation for developing countries difficult and international comparability a cause of concern.
 4/11/2005South AfricaSouth Africa disagree with the recommendations and suggest that it is used separately for productivity and other kind of analysis.
 4/11/2005Sweden m2(c)swe16;
 4/1/2005PolandIn the scope of government owned assets we are against including return to capital into measurement of the non-market output.
Navigation Options
*Back to Issues
*See all issues/subissues with country comments
*See all AEG recommendations for this issue
*See all expert comments for this issue

About  |  Sitemap  |  Contact Us
Copyright © United Nations, 2024