Journal Issue
Share
Article

Uruguay: Report on the Observance of Standards and Codes (ROSC)—Data Module Volume I

Author(s):
International Monetary Fund. Statistics Dept.
Published Date:
February 2014
Share
  • ShareShare
Show Summary Details

Overall Assessment

1. This Report on the Observance of Standards and Codes (ROSC)—Data Module is a reassessment of the exercise conducted in 1999–mid-2000, which was based on information collected during a November 1999 mission and used the dimensional structure of the first July 2000 draft version of IMF’s Data Quality Assessment Framework (DQAF). This reassessment covers national accounts, and consumer (CPI) and producer price indices (PPI). The Uruguayan agencies that compile and disseminate these datasets are: the Central Bank of Uruguay (CBU) for national accounts and the National Institute of Statistics (NIS) for the CPI and PPI. This Report is the first data ROSC based on the 2012 version of the DQAF, which assesses data quality against the relevant statistical standards current in 2012, including the System of National Accounts 2008 (2008 SNA), the Consumer Price Index Manual 2004 (CPI Manual), and the Producer Price Index Manual 2004 (PPI Manual).

2. The Report contains the following main observations. Following the 1999 data ROSC mission,1 Uruguay made significant improvements in statistical compilation and dissemination of the statistics covered by this ROSC—Data Module. The progress achieved allowed Uruguay to subscribe to the Special Data Dissemination Standard (SDDS) on February 12, 2004. Since its SDDS subscription, Uruguay has been in observance of the SDDS, meeting the specifications for coverage, periodicity, timeliness, and the dissemination of advance release calendars. In the last (2011) SDDS annual report on observance, Uruguay met the SDDS requirements for timeliness with episodic exceptions for some data categories. Uruguay exceeds the SDDS timeliness requirements for labor market (employment, unemployment, and wages/earnings), price (consumer prices and producer prices), and international investment position data. Currently, Uruguay is using two regular timeliness flexibility options for general government operations and central government operations; additionally, it is using an “as relevant” timeliness provision for analytical accounts of the banking sector for countries with extensive branch banking systems. No flexibility options are being used regarding real sector statistics. Appendix I provides an overview of Uruguay’s dissemination practices for real sector statistics compared to the SDDS.

3. In applying the DQAF, the remainder of this section presents the mission’s main conclusions. The presentation is done at the level of the DQAF’s quality dimensions, by agency for the first two dimensions and across datasets for the remaining four. Uruguay’s macroeconomic statistics are generally of good quality and adequately meet users’ needs. Since 1999, Uruguay has made tangible improvements on the methodological and dissemination aspects of data quality in national accounts and price statistics, as the coverage, weights, and base period of both macroeconomic statistics have been recently revised and updated. One important testament to the improvement of statistics is the subscription to the IMF’s SDDS after the 1999 ROSC mission.

4. The legal and institutional basis for Uruguayan statistics is sound and follows international good practice. The Central Bank of Uruguay (CBU) is a government agency endowed with technical, administrative, and financial autonomy. Quality awareness and quality management processes are included in the CBU’s 2010–2014 Strategic Plan and 2012 Strategic Initiatives. However, no explicit legal provision provides the responsibility to the CBU for compiling national accounts statistics. Staffing allocated to the Economic Statistics Area (ESA) is also limited. The National Institute of Statistics (NIS) is a public body responsible for the development, monitoring, and coordination of national statistics. Resources are very limited to perform its broad functions, although quality awareness is in place. The NIS in the process of implementing a quality management system. The process of recruiting economic statistics professionals into the Civil Service has slowed staffing at both the NIS and the CBU, and retention of qualified staff is a challenge for the NIS. A mechanism is not in place to collect feedback from data users. Under maintaining the relevance of statistics, both agencies could conduct more outreach to data users and give them more advance notice of pending methodological changes.

5. Under institutional integrity, based on the mission’s meetings with data users, the CBU and NIS possess a high level of “trust capital” to deliver impartially-compiled and technically-sound statistics within their resource envelopes. This said, staff training and support to staff for conducting research could be improved. Neither the CBU nor the NIS comment on erroneous interpretation and misuse of statistics. Both agencies follow sound practices to ensure the transparency of their methodologies as well as compilation and dissemination practices, the bedrock of statistical trust capital. However, both the CBU and NIS could give users more advance notice of changes in methodology. Ethical standards meet international norms and good international practice.

6. Under methodological soundness, the national accounts conform with the System of National Accounts 1993 (1993 SNA) and the CPI and PPI broadly conform with international methodological guidance from the Consumer Price Index Manual (2004), the Practical Guide to Producing Consumer Price Indices (2009), and the Producer Price Index Manual (2004). The principal exceptions are that the CPI does not cover either the implicit rent or the net acquisitions of owner-occupied dwellings, and the PPI does not cover services and exported output. The industrial activity classification is the International Standard Classification of All Industrial Activities, Revision 3 (ISIC, Rev. 3). The DQAF 2012 of this ROSC assesses Uruguay’s practices against the latest methodological standards; thus for the national accounts there is a need to plan adoption of the 2008 SNA and ISIC, Rev. 4. The product classification follows a national extension of the ISIC, Rev. 3 rather than the international standard Central Product Classification. The basis for recording follows the 1993 and 2008 SNA with the exception of components of government revenue and expenditure that are recorded on a cash and obligation basis, respectively, which deviate in degrees from the SNA accrual requirement.

7. Under accuracy and reliability, Uruguay’s national accounts are built on a good core system, but there are areas for improvement in source data. Business register maintenance could be better in identifying and deleting out of business units and reclassifying units changing principal industrial activity. Household expenditures are surveyed about once a decade. Coverage of the financial statements of financial corporations, public nonfinancial corporations, and public-private partnerships is robust, but not of private nonfinancial enterprises. About 60 percent of the GDP calculation is based on fixed input-output ratios from 1997. Household consumption is not independently derived and total changes in inventories and changes in inventories for most products are obtained as residuals. Uruguay disseminates annual but not quarterly GDP by the expenditure approach at current prices and does not compile annual integrated economic accounts by institutional sector, in particular, the generation of income account. For the CPI, reselection of the sample of detailed products has not been done for an extended period, and for both the CPI and PPI, statistical outputs/intermediate results are not validated with available information from alternative sources.

8. The CPI and PPI get good marks for source data on the basis of well designed surveys and advanced data capture technologies. The national accounts need improvement in the coverage and validation of source data. Both the CPI and PPI need to strengthen assessment and validation of intermediate data and statistical outputs. The staffing issues noted under prerequisites for quality have had their impact on the ability of the NIS, the principal supplier of source data, to maintain good data validation practices for secondary source data used in the PPI. Major revision studies are made for the national accounts when changing the base year, but regular revisions are not analyzed to detect bias in the series. The impacts of weight updates for the CPI and PPI are not analyzed.

9. Under serviceability, periodicity, and timeliness meet SDDS standards and the national accounts and price statistics get good marks for consistency with one another, the exception being that coverage of the CPI shelter component should cover not only renters, but also owner occupants. Detailed national accounts data are only available up to 2008. Long-time series are not available on the CBU website. There is no regular schedule for updating the base year of the national accounts. The quarterly national accounts get good scores on the elements of revision policy and practice, but the causes of current revisions are not explained to users. A monthly index of economic activity is not disseminated. The CPI and PPI would benefit from a more regular and frequent schedule of weight updates.

10. Both the CBU and the NIS get good scores on user accessibility of data and descriptive information (metadata) on the sources and methods through which real sector statistics are compiled, and both do well in providing assistance to users. The CBU and NIS could also undertake more regular revision studies and both could engage users more fully to improve statistical products and services.

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

Assessment by Agency and Dataset

12. Assessment of the quality of three macroeconomic datasets—national accounts, consumer price index, and producer price index—were conducted using the DQAF 2012. In this section, the results are presented at the level of the DQAF elements and using a four-level rating scale (Table 1). Assessments of the prerequisites of data quality and the assurances of integrity (Dimensions “0” and “1” of the DQAF) are presented in Tables 2ab. For each dataset, the assessment of methodological soundness, accuracy and reliability, serviceability, and accessibility (Dimensions “2” to “5” of the DQAF) are shown in Tables 3ac.

Table 1.Data Quality Assessment Framework 2012—Summary Results
Key to symbols: O = Practice Observed; LO = Practice Largely Observed; LNO =Practice Largely Not Observed; NO = Practice Not Observed; NA = Not Applicable
Dimensions/ElementsDatasetsNational AccountsConsumer Price IndexProducer Price Index
0. Prerequisites of quality
0.1 Legal and institutional environmentLOOO
0.2 ResourcesLNOLNOLNO
0.3 RelevanceLNOLNOLNO
0.4 Other quality managementOLOLO
1. Assurances of integrity
1.1 Institutional integrityLOLOLO
1.2 TransparencyLOLOLO
1.3 Ethical standardsOOO
2. Methodological soundness
2.1 Concepts and definitionsLOOLO
2.2 ScopeLOLOLNO
2.3 Classification/sectorizationLOOO
2.4 Basis for recordingLOOO
3. Accuracy and reliability
3.1 Source dataLNOLOO
3.2 Assessment of source dataLNOOLO
3.3 Statistical techniquesLOOO
3.4 Assessment and validation of intermediate data and statistical outputsOLNOLNO
3.5 Revision studiesLONONO
4. Serviceability
4.1 Periodicity and timelinessOOO
4.2 ConsistencyOOO
4.3 Revision policy and practiceLOLNOLNO
5. Accessibility
5.1 Data accessibilityLOOO
5.2 Metadata accessibilityOLOLO
5.3 Assistance to usersLOOO
Practice observed: Current practices generally meet or achieve the objectives of DQAF internationally accepted statistical practices without any significant deficiencies. Practice largely observed: Some departures, but these are not seen as sufficient to raise doubts about the authorities’ ability to observe the DQAF practices. Practice largely not observed: Significant departures and the authorities will need to take significant action to achieve observance. Practice not observed: Most DQAF practices are not met. Not applicable: Used only exceptionally when statistical practices do not apply to a country’s circumstances.
Table 2a.Assessment of Data Quality—Dimensions 0 and 1—Central Bank of Uruguay
0. Prerequisites of quality1. Assurances of integrity
Legal and institutional environment. The Central Bank of Uruguay’s (CBU) Charter does not clearly establish its responsibility for the compilation of national accounts statistics. The National Statistical System Act (16.616) regulates integration (Article 1), principles (Article 3), and statistical confidentiality (secrecy obligation, obligation to provide information) (Article 3) (Article 14). CBU’s Charter: Law 16,696 (March 1995) and modifications Law 18,401 (October 2008) regulate, among other things, the obligation of secrecy (Article 21) and the power to require information (Article 55). The CBU may request for statistical purposes, of any individual or legal entity, whether public or private, all information needed to duly fulfill its functions and duties. It shall be covered by the administrative secrecy and will be strictly confidential. The CBU may impose fines on any person or entity not supplying the information lawfully required from it, or submitting incomplete or inaccurate information. The amount of the fine will range between 10,000 IU (ten thousand indexed units) and 20,000 IU (twenty thousand indexed units) in the case of legal persons, and will be 4,000 IU (four thousand indexed units) in case of natural persons. Payment of the fine does not exempt from the requirement to submit the requested information. However, fines are not used to reduce nonresponse on national accounts collections owing to their administrative cost for the CBU. In practice, There are coordination and communication issues inside statistical agencies and among them, as well as some duplication of effort in the collection of source data among data producing agencies. So far, the CBU has been excluded from access to individual tax records for statistical purposes under a tributary secrecy provision. However, there is a framework agreement NIS-CBU for access to microdata released by the National Institute of Statistics (NIS). CBU officials are subject to the Rules of Discipline, which sets penalties for noncompliance of the duties including the secrecy of data. The rule of thumb to follow is not to disaggregate groups containing less than three individuals, unless by consent of the informants. The filling of the questionnaires is done online. Respondents are consulted in case of inconsistencies in reported data.



Resources. Staff and financing at the bank are not adequate. The Economic Statistics Area (ESA) currently has 10 vacancies, which represent over 50 percent of its analysts. Overall, staff has been retained, but the ESA cannot manage rotation, because it is at the minimum of staff numbers. Recruiting staff into the Civil Service takes a long time and has slowed the recruitment of additional national accounts staff. The information technology is generally adequate. The ESA structure, although recent, could be updated to better reflect the current flow of work.



Relevance. Users are not consulted on the relevance of the CBU’s statistics in a regular manner. There is no an advisory group on statistics.



Other quality management. The mission and vision of the CBU and the ESA in particular stresses data quality. The institution is in the process of improving the organizational infrastructure to take into account economies of scale and process optimization through the Strategic Planning. Independent assessments from international organizations have been sought to examine the quality of national accounts. The CBU has received technical advice from ECLAC.
Institutional integrity. The CBU, created by the Article 196 of the Constitution of the Republic, is an agency endowed with technical, administrative, and financial autonomy. Staff positions are technical. The promotions are made on a competitive basis taking account of background evaluation and merit. However, staff training and support to staff for conducting research could be improved. The CBU rarely comments on erroneous interpretation and misuse of statistics.



Transparency. The administrative rules of the CBU establish transparency as one of the principles in the general rules of administrative actions. Law 18381 from October 17, 2008 seeks to promote transparency of the administrative function of any public body while ensuring the right of people to access to public information. The CBU, by resolution of its Board No. 201 dated June 29, 2011, has developed administrative procedure regulations implementing the Act, which is available on the intranet of the institution. Disclosure outside the CBU is done at the same time for all users. Users are not given advance notice of major methodological changes.



Ethical standards. The CBU Bylaws establish ethical standards that are followed by CBU staff regarding hiring, behavior, obligations, and prohibitions. The institution has a code of ethics (Board Resolution No. 485 of December 29, 2010) which is available on the intranet. It also has a Disciplinary Regulation (Board Resolution No. 153 of March 22, 1994) which is also available on the intranet.
Table 2b.Assessment of Data Quality—Dimensions 0 and 1—National Institute of Statistics
0. Prerequisites of quality1. Assurances of integrity
Legal and institutional environment. The NIS is a public body which aims at the development, monitoring and coordination of national statistics. The Statistics Law No. 16,616 enacted on October 20, 1994, states that the National Statistical System (NSS) is formed by the NIS as the governing body, the Sector Coordinating Units (one for each relevant subject area) and other offices producing statistics. Among NIS functions are compiling statistical, demographic, economic and social information; coordinating and supervising the NSS; disseminating information produced by the NIS, or other bodies of the NSS; promoting statistical research, developing knowledge in the field of statistics and economics; providing training to staff of the Statistical Offices of the system; establishing technical standards to be implemented in the development of official statistics; approving plans and statistical programs of the NSS; and monitoring their implementation. Article 3 of the Statistics Law states that the statistical confidentiality must be observed concerning the individual data provided by the source of information, so as not to reveal the identity of these sources. Article 13 states that a source of information may be any person or entity that is, permanently or temporarily, in the country. Article 14 states that all natural or legal persons, nonstate public persons and public bodies are required to provide data that may be required for statistical purposes by members of the NSS and within the period prescribed. Noncompliance with information request is penalized with a fine. The amount of the fine is determined by the NIS, between a minimum of 20 IU (twenty indexed units) and a maximum of 50 IU (fifty units adjustment). Article 7.D requires that NIS compiles, publishes and disseminates data within the area of its competence, which includes the consumer price index (CPI) and producer price index (PPI).



Resources. The NIS’s challenge in hiring and retaining qualified specialists in economic statistics that is made more acute under current civil service arrangements.



Relevance. Users are not consulted on a regular basis about the relevance of statistics produced by the NIS. Although, feedback is not discouraged, there is no formal process in place for users to express their opinions about their statistical needs.



Other quality management. The NIS has in place a quality policy. A System of Quality Management was created a few years ago for which the scope is to improve statistical data. Consequently, the success of this endeavor will require additional staff training about this framework. In 2008, the Construction Cost Index was certified according to the ISO 20252:2006 standard, as a pilot. During the next five years the NIS will implement the system in other areas of the Institute, according to specific work plans that will be defined annually. The quality management system is still at the implementation stage.
Institutional Integrity. Article 3 of the Statistics Law states that the statistical rigor that NIS should observe is the systematic application of the principles, methods and procedures generally accepted in the statistical art and science. Technical autonomy is observed in the development of statistical activities with independence and objectivity, and based exclusively on statistical principles. The CBU does not comment on erroneous interpretation and misuse of statistics.



Transparency. Transparency is a valued quality at the NIS. It is the right of the provider of information to know the objectives of the statistical activity for which they are providing data for and to be informed that by law, their information is confidential. No ministerial commentary is added to the data released. Although in theory there is nothing that prevents the NIS to inform users of upcoming methodological changes, the practice has been to inform users of the CPI and PPI of any such changes only at release time.



Ethical standards. NIS staff follow established ethical standards in the performance of their duties.
Table 3a.Assessment of Data Quality—Dimensions 2 to 5—National Accounts
2. Methodological soundness3. Accuracy and reliability4. Serviceability5. Accessibility
Concepts and definitions. Follow the System of National Accounts 1993 (1993 SNA). A program for changing the base year and implementing the 2008 SNA is under development, but has not reached the approval stage.



Scope. Data published on GDP cover the entire Uruguayan economy. In the base year 2005, efforts were made to cover both recorded and non-recorded activities. No annual integrated economic accounts by institutional sector and quarterly GDP by the expenditure approach at current prices are compiled. Free zones/bonded warehouses/factories operated by offshore enterprises under customs control are included in GDP. Contraband is estimated.



Classification/ sectorization. The activity classifier is ISIC Rev.3 and a national classifier of products. Central Product Classification (CPC), Classification of Individual Consumption by Purpose (COICOP), and Classification of Functions of Government (COFOG) are not used.



Basis for recording. Market prices are used. Transactions and flows are recorded on accrual basis except for government revenues. Government expenditure is on an obligation basis rather than on an accrual basis. Transactions between establishments of the same enterprise are recorded on a net basis because the data are collected at enterprise level, rather than on the recommended gross basis.
Source data. The business register is partially updated. Elimination of enterprises that close or reclassification of enterprises that change activity is not timely. Annual enterprise statistics are collected through a regular survey program that is not always consistent over time or timely. Enterprise data are sometimes not accurate enough to be usable. Household surveys are conducted by the NIS on a regular basis, but only on a 10-year interval. Data for construction, services, and quarterly estimates are limited in scope. Comprehensive government finance statistics are available regularly. Financial statements and supplementary data are available only for public corporations and public-private partnerships. There is no access to financial statements for private nonfinancial corporations. Source data on free zones are incomplete. The coverage of total value added by all data sources is around 56 percent.



Assessment of source data. Accuracy of the data from surveys, administrative records, and other sources is not routinely assessed. Data from the economic survey are not properly validated due to lack of resources at the NIS and some collected data are thus only partially used. There is no information on non-sampling errors for most surveys.



Statistical techniques. Output estimates are compiled at the group and class level of ISIC, Rev.3 and Rev.2, respectively. Fixed ratios from 1997 are used to obtain 60 percent of GDP. Although employment data are used to make adjustments for nonobserved activities, available household data on salaries, and mixed income are not used. GDP estimates by expenditure were derived independently for the base year but quarterly household consumption expenditure is not independently derived. Changes in inventories are residually obtained. Proper techniques are used to compile net acquisitions of owner-occupied dwellings and work in progress, but no inventory valuation adjustment is made. Proper procedures are followed for compiling volume measures of GDP. Quarterly data are benchmarked and seasonally adjusted. Chain indices are not used.



Assessment and validation of intermediate data and statistical outputs. Intermediate results are validated and checked against other independent data sources. The supply and use framework is used to investigate discrepancies and make the statistical outputs consistent. Last available SUT is from 2008. Revision studies. Major revision studies are made when changing the base year, but regular revisions are not analyzed to detect bias in the series.
Periodicity and timeliness. GDP estimates are compiled quarterly and annually. Quarterly GDP estimates are disseminated 75 days after the end of the reference quarter although sometimes with short delays. The annual GDP estimates are disseminated within three months after the end of the reference year.



Consistency. Annual national accounts are internally consistent. Quarterly GDP estimates are consistent with annual estimates. The national accounts statistics are consistent with balance of payments, and government finance statistics. The statistical series are consistent over time.



Revision policy and practice. Quarterly data are preliminary when first released. The quarterly revision takes place every quarter during the current year and once annual data are released.



Annual preliminary data for the previous year are first released in March of the following year and are preliminary until a supply and use table is compiled. Data for the two previous years are usually revised. There is no regular schedule for updating the base year. Current revisions have a regular schedule. Major revisions are analyzed and published, but specific causes of regular revisions are not disseminated.
Data accessibility. Quarterly and annual data are disseminated for GDP in current and constant 2005 prices in pesos, volume indices, and implicit deflators, broken down by sectors of economic activity. The CBU website disseminates quarterly and annual data with different base periods from 1988. Linked time series are not available on the CBU website. Detailed data are only available up to 2008. Data at 3-digit level are only available for some activities. Aggregated data are available from 2009 onwards. Quarterly GDP by expenditure is disseminated only in constant 2005 pesos and volume indices. The official estimates are published on the CBU website (http://www.bcu.gub.uy) where annual and quarterly reports and tables can be found. Consolidated accounts are annually disseminated. Charts and briefings are disseminated along with the data. An advance release calendar that provides a next quarter precise release date is posted on the CBU website. Statistics are usually released on a preannounced schedule. Data are disseminated simultaneously to all users on CBU website. Statistics not routinely disseminated are made available to users upon request to the Institutional Communication Center: info@bcu.gub.uy.



Metadata accessibility. The current methodological basis for the National Accounts is described in the CBU publication “Revisión Integral de las Cuentas Nacionales 1997–2008” (National Accounts Integral Revision, 1997–2008) and is disseminated through the CBU website http://www.bcu.gub.uy/Estadisticas-e-Indicadores/Paginas/Metodologias.aspx. The historical data of 1983 Base (1988 Revision) can also be found at the CBU website. Different levels of data detail are made available to meet users’ requirements.



Assistance to users. There is no established procedure to consult CBU’s visitors on their data gathering experience. There is an e-mail address on the CBU website to address consultations. Assistance to users is not monitored and revised periodically.
Table 3b.Assessment of Data Quality—Dimensions 2 to 5—Consumer Price Index (CPI)
2. Methodological soundness3. Accuracy and reliability4. Serviceability5. Accessibility
Concepts and definitions. The CPI is compiled following the Consumer Price Index Manual 2004. The level of detail of goods and services is sufficient to perform detailed analysis of price changes.



Scope. The CPI covers a set of aggregates that are consistent with the final consumption expenditure of households for 375 products. It includes all resident urban households at the national level (urban Montevideo city and other urban areas), families of all sizes and income levels. It excludes the imputed value of owner-occupied housing. Illegal goods are not included in expenditures in view of the difficulties involved in data collection. Purchase/sale prices of used goods are not collected.



Classification/sectorization. The COICOP is used to classify CPI data.



Basis for recording. Following international best practice, CPI weights are based on consumption expenditure valued at purchasers’ prices, including indirect taxes and excluding interest on credit purchases. Prices are those recorded at the time of purchase. The weights are based on net purchases, e.g., for insurance, it is the difference between gross premiums and claims paid.
Source data. The base period for the CPI is December 2010 = 100. Weights are from the 2005–2006 Household Expenditure and Income Survey, supplemented with data from the national accounts. The price survey includes electronic point of sale (scanner) data. Press articles are sometimes used to validate price change for some items such as gasoline, but validation is not extensive across the index. The periodicity and timeliness of the price survey complies with international best practices. The sample of detailed products is not updated to allow new products to enter the index.



Assessment of source data. Surveys are audited to monitor the work of enumerators. Outliers are confirmed with respondents.



Statistical techniques. The CPI is a Laspeyres price index. Automation of the compilation procedures minimizes processing errors. Following international practice, prices of temporarily missing, seasonal, and new products are imputed using price movement of products in the same group, based on cell sufficiency criteria to determine the level of aggregation of the imputation. Elementary aggregates of individual price movements use the geometric mean formula.



Assessment and validation of intermediate data and statistical outputs. Unusual changes in the index resulting from potential problems in price data are investigated and corrected if warranted. Intermediate results and final outputs are not validated against comparable data from other sources.



Revision studies. No such studies are carried out by the NIS for the CPI to assess the impact of weight and basket updates.
Periodicity and timeliness. The CPI is produced monthly and released on the second business day of the month following the reference month.



Consistency. The statistical series are internally consistent. “Cifras” publishes monthly data for the reference year and for the three previous years. NIS’s website disseminates “Cifras” and monthly time series data since March 1997 with similar breakdowns of those published in “Cifras.” NIS’s database, which can be accessed through the Internet, disseminates monthly time series as follows: (i) by major CPI groups since 1985; and (ii) the general index since July 1937.



Revision policy and practice. CPI data are deemed final and are therefore not subject to revision. Weight and basket updates are not on a regular schedule.
Data accessibility. Datasets are published with different levels of detail. E-mail subscription in Uruguay is available on request from difusión@ine.gub.uy or by fax. Data can be viewed on the NIS website at http://www.ine.gub.uy and on Uruguay’s National Summary Data Page (NSDP). The CPI release calendar for the year is posted on the NIS website. Statistics are promptly disseminated at 2 p.m. on the posted release dates. Data are released simultaneously to all users in a press release “Indice de Precios al Consumo” and a monthly publication “Cifras.” Certain institutional users can access the data up to two hours prior to the official release schedule. At the same time, they are disseminated to the banking system via an interbank computer network, and published in the Diario Oficial de la República (Official Gazette). Data not routinely disseminated are available upon request.



Metadata accessibility. A methodological note is published in “Indice de Precios del Consumo, Cambio de Base–Diciembre 2010, Nota Metodológica” This note explains the concepts and methods used, in addition to the changes were introduced during the last update. Deviations from international standards are not noted.



Assistance to users. Support is available to users of statistics. The NIS website contains a list of publications and documents available to users. The NIS also has a library where users can search for old publications, not available in electronic format. For special requests of unpublished statistical series, the applicable price changes are calculated upon request.
Table 3c.Assessment of Data Quality—Dimensions 2 to 5—Producer Price Index (PPI)
2. Methodological soundness3. Accuracy and reliability4. Serviceability5. Accessibility
Concepts and definitions. The producer price index (PPI) of Uruguay is compiled by NIS following the Producer Price Index Manual 2004. There are no price indices for intermediate consumption.



Scope. The PPI covers the gross output of agriculture, livestock, hunting, forestry, fishing, mining and quarrying, and manufacturing. All resident enterprises in the formal market that produce for the domestic market are considered in scope for the PPI. It would however be desirable to expand the coverage of the PPI to include statistics for other sectors such as utilities, services and construction, as well as exports of goods and services of all types.



Classification/sectorization. The classification of institutional units and transactions follows 1993 SNA guidelines. For classes of activities, ISIC Rev.3 is used, but with slight modifications that account for the unique environment of the country. The NIS uses the CPA product classification.



Basis for recording. Price is defined as the value per unit of good or service traded in a purchase/sale operation carried out between a seller and a buyer. In light of the objectives of the index, producer prices are collected. The prices collected represent the amount received by the producer, in the case of Agriculture, Livestock, and Forestry, and Manufacturing sections; landed prices for fisheries, and for Mining and Quarrying, the quarry price.
Source data. Information for the index base weights is drawn from the supply-and-use tables compiled from the CBU. Product selection is based on a sample survey of specifications. The data support PPI statistics at the 4-digit level. Annual surveys are conducted on industrial activity. The periodicity and timeliness of price data are adequate. Prices reference the tenth of the month, but should be spread through the month for volatile items.



Assessment of source data. Increases of + or – 5 percent over the last reported prices are examined for accuracy. The proportion of output of sections A–D of ISIC Rev.3 the PPI covers is at least 90 percent. Data from secondary sources (government ministries) are not validated.



Statistical techniques. The PPI uses a Laspeyres formula with base and reference period March 2010=100. Use of e-forms and automatic data capture minimize data-entry errors. The calculation procedure is standardized. Temporarily missing and seasonal prices are imputed from the price changes of similar products. Product replacements are handled using a previous month price supplied by the producer or targeted imputation.



Assessment and validation of intermediate data and statistical outputs. Unusual price relatives are investigated and if erroneous corrected. Checks and validations with external data sources are seldom performed at this stage.



Revision studies. The base and weights are updated with no preset timeframe. No revision studies are performed since the index is not revised. There are also no studies that analyze the effects of updating the weights.
Periodicity and timeliness. The PPI is produced monthly and released on the day before the last day of the reference month, exceeding the SDDS timeliness requirement. However, the unusually high degree of timeliness does not allow sufficient time analyze and validate the results.



Consistency. Aggregates are presented at the class, group, division, and industry level of Uruguay’s adaptation of ISIC, Rev. 3. The General (all-items) PPI can be obtained by aggregating from any of the levels of the stratification structure. Consistent time series are available over a period longer than five years. Time series are reconstructed, to the extent possible, when there are changes to source data, methodology, or statistical techniques. Methodological changes in indices are announced when there are important discontinuities and interruptions in the time series. Producer price statistics are broadly consistent with other price statistics of a similar nature; otherwise, discrepancies are seldom explored or analyzed.



Revision policy and practice. Revisions in weights and baskets do not follow a regular schedule. Methodological changes are sometimes posted on the NIS website but is not common practice to do so.
Data accessibility. PPI data are published with supplementary tables and charts to facilitate analysis. Estimates are presented in detail on the website. http://www.ine.gub.uy. No seasonally-adjusted data are produced. Tables, press releases, and the release calendar are available on the website. Data are disseminated to the public as well as the authorities on the same day. An e-mail is sent to the media with the press releases after they are made publicly available on the website. Certain institutional users can access the data up to two hours prior to the official release schedule. Upon request, special arrangements might be made to meet specific needs, provided that confidentiality conditions are respected.



Metadata accessibility. A methodological document, Indice de precios al productor de productos nacionales (IPPN) is available on the NIS’s website and is updated at the same time as the basket that includes changes that have occurred with the most recent update. SDDS metadata are sent to the Central Bank as changes occur. Short updates are published on the website.



Assistance to users. Contact points are listed in each questionnaire and on the NIS website. In addition to service catalogs, a number of documents prepared by the NIS are available on its website. Information that is not available on the website may be obtained from the institutional library. Special requests are handled by the dissemination unit, where users are informed whether their requests are feasible and referred to the appropriate unit.

Staff’s Recommendations

13. Based on the review of Uruguay’s statistical practices, discussions with the data producing agencies, and responses from data users, the mission has developed a set of recommendations. They are designed to increase further Uruguay’s adherence to internationally accepted statistical practices and would, in the mission’s view, enhance the analytical usefulness of Uruguay’s statistics. Some additional technical suggestions are included in the Detailed Assessments volume.

Cross-cutting recommendations

High priority

  • Ensure that the NIS has adequate financial resources, staff, facilities, and training, and take further steps to increase retention of qualified staff.
  • The legal framework to hire/contract staff by the NIS and the CBU need to be more flexible to accelerate filling vacancies with technical qualifications in economic statistics.
  • Sign an agreement between the CBU and the Ministry of Economy and Finance (MEF) so that the CBU can have access to income tax records by economic activity in order to improve the coverage of national accounts and compile integrated economic accounts by institutional sector.
  • Expedite the filling of the ten vacancies in the CBU Economic Statistics Area. Additional staff will be needed for changing the base of the national accounts and implementing the 2008 SNA.
  • Conduct a new household Income and Expenditure Survey every five years to strengthen estimates for household final consumption expenditure and update the CPI weights.

Other recommendations

  • Initiate regular consultations with public and private sector users, including through fostering users’ groups and an advisory committee to improve the usefulness of statistics and advise on statistical program priorities. The advisory committee should include participation of academics, private sector analysts, and producers’ associations and meet on a regular basis (e.g., twice a year).
  • Announce in advance any planned changes in concepts and methodology. With major changes, users should be invited to provide feedback before they are implemented.
  • Analyze data requests and comments from users periodically in order to improve the service provided to them.
  • Initiate an employee rotation program to enhance the versatility of professionals and the redundancy of skills of statistical units and teams in view of limited staff complements.

National Accounts

High Priority

  • Update the base year of the national accounts and develop plan for implementing the 2008 SNA as soon as possible with a clear timetable.
  • Update the business register regularly.

Other recommendations

  • The NIS should review and update the classification by economic activity of the business register obtained from income tax records and provide an improved version to the MEF, so that the MEF and the CBU can use the income tax records data with an updated and improved classification.
  • Adapt the functions of the CBU ESA’s departments to an integrated organization by institutional sector (non-financial sector, financial sector, public sector, and external sector), so that the supply and use table by economic activity and product by sector and the economic integrated accounts by sector are compiled and analyzed by the same staff with coordinators by type of statistic.
  • Apply the standard ISIC Rev. 4 and CPC Ver. 2 for classifying activities and products, respectively, the COICOP for classifying household financial consumption expenditure and the COFOG to classify government final consumption expenditure in order to facilitate international comparisons.
  • Increase the level of detail of disseminated national accounts data to three digits of the ISIC.
  • Improve estimates of nonobserved activities by using available income data from the permanent household surveys.
  • Consider the use of chain indices for calculating volume measures.
  • Reconstruct historical series as far back as reasonably possible when changing the base year. Although linked time series are available only on GDP by the production approach for the period 1997–2011, longer-time series of main national accounts aggregates are not available.
  • Apply the economic activity classification of the economic survey (ISIC, Rev. 4) to the permanent household survey in order to reduce the volatility of employment data by economic activity which is preventing greater use of the data in the national accounts. Include an expenditure module every two to three years to monitor changes in consumption patterns and improve estimates on household consumption.
  • Improve the coverage of the economic survey, the surveys on construction, domestic trade and services (annual, quarterly, and monthly) to 80 percent of the economic activity in the country.
  • Use monthly value-added tax (VAT) data by economic activity to improve the coverage of quarterly national accounts estimates.
  • Conduct a quarterly survey on sales of services and on inventories of inputs and finished and resale products.
  • Improve the coverage of the monthly index of economic activity in order to disseminate it and use it in all its potential for policy decision making.
  • Review the Economic Survey to collect data of inputs and outputs of the establishments that integrate an enterprise at least for the new base year of the national accounts. Apply a standard product coding in the survey, preferably CPC. Conduct an integrated enterprise/establishment survey that includes the information of the establishments that belong to each enterprise when changing the base year of the national accounts.
  • Use available data on prices and volume to apply the double deflation/inflation method to main inputs and outputs on a quarterly basis.
  • Use data on inventories collected in the economic survey and from income tax records in order to improve the coverage of changes in inventories.
  • Explain current data revisions to users.

Consumer Price Index

High priority

  • Incorporate owner-occupied housing in the CPI.
  • A regular cycle for updating the CPI basket on a timely basis should be established. The advantages of this decision are the following:
    • Improve the relevance of the CPI for current economic conditions.
    • Improve the planning and organization of the resource requirements for updating the basket.

Other recommendations

  • A basket update exercise is an opportunity, in addition to updating the weights, for a comprehensive review of the CPI procedures, concepts, and methods. For example, the choice of the sample of representative products should be refreshed when updating the basket.
  • The NIS could undertake a study of the difference between the household expenditure survey and comparable components of national accounts household consumption expenditure and consider using the national accounts data as the source of, or as a control for, the weights of the CPI. The advantages of this option are the following:
    • Improved coherence of macroeconomic statistics between the CPI and the national accounts.
    • Weighting information that incorporates not only the most recent household expenditure survey, but also retail sales and other current information that has better coverage of certain components of household expenditure.
    • More timely CPI weights, which reduce the lag between the weight reference year and the price reference month.
    • The possibility of updating the basket weights more often and at lower cost.
  • Improve the analytical capacity of the NIS CPI unit.
  • Institute more data confrontation-type analysis for validation of results using alternative data sources.

Producer Price Index

High priority

  • A regular calendar for updating the PPI should be established. The calendar should specify an update cycle of no more than five years.
  • Add additional economists/statisticians to the staff.
  • Institute more data confrontation-type analysis and the analysis of the third-party data used in the PPI.

Other recommendations

  • Expand the scope of the PPI to include first utilities, and subsequently services and construction.
  • Expand the scope of the PPI to also include products destined for export, which is the international standard.
  • Consider delaying the release of the PPI until one or two weeks after the reference month to provide more time to analyze the results and the data sources; this is not currently done given the limited resources.
Appendix Table. Practices Compared to the SDDS Coverage, Periodicity, and Timeliness of Data
SDDS Data CategoryCoverage

(meets SDDS

requirement)
PeriodicityTimelinessComments 1/
SDDSUruguaySDDSUruguay
Real Sector
National accountsYesQQQ3M
Production index/indicesYesMM6W

(1M encouraged)
48D
Forward-looking indicators(encouraged

data category)
M or QM or Q
EmploymentYesQMQ5W
UnemploymentYesQMQ5W
Wages/earningsYesQMQ5W
Consumer price indexYesMMMNLT 5D
Producer price indexYesMMM1D
Addendum: PopulationYesAA
Note: Periodicity and timeliness: (D) daily; (W) weekly or with a lag of no more than one week from the reference data or the closing of the reference week; (M) monthly or with a lag of no more than one month; (Q) quarterly or with a lag of no more than one quarter; (A) annually; and (…) not applicable.Italics indicate encouraged categories.

No flexibility options are taken for national accounts, CPI, or PPI. However, Uruguay does take flexibility options for timeliness of central and general government operations releases and an “as relevant” timeliness flexibility option on the analytical accounts of the banking sector for countries with extensive branch banking systems.

Note: Periodicity and timeliness: (D) daily; (W) weekly or with a lag of no more than one week from the reference data or the closing of the reference week; (M) monthly or with a lag of no more than one month; (Q) quarterly or with a lag of no more than one quarter; (A) annually; and (…) not applicable.Italics indicate encouraged categories.

No flexibility options are taken for national accounts, CPI, or PPI. However, Uruguay does take flexibility options for timeliness of central and general government operations releases and an “as relevant” timeliness flexibility option on the analytical accounts of the banking sector for countries with extensive branch banking systems.

1The 1999 assessment was based on a preliminary version of the DQAF, which was updated in 2001, 2003, and most recently 2012, as international standards have evolved. As noted below, the current assessment thus is based on broadly higher accomplishment thresholds than the 1999 ROSC.

Other Resources Citing This Publication