The Central Bank of Costa Rica (BCCR) coordinated the responses to recommendations of all three institutions whose statistics were evaluated. Responses concerning the cross-cutting recommendations were prepared by the BCCR. Response to national accounts (NA), producer price index (IPP), monetary (MS), and balance of payments (BOP) statistics issues were prepared by the BCCR; responses on consumer price (CPI) statistics were prepared by the National Institute of Statistics and Censuses (INEC); and on government finance statistics (GFS) by the Ministry of Finance (MOF).

A. Cross-cutting Recommendations

The Macroeconomic Statistics Department (DEM) of the BCCR is entirely in agreement with the assessments and recommendations made in the DQAF. To facilitate recognition of shifting statistical conditions, challenges, and opportunities, these responses from the authorities put forward observations, clarifications, and plans for addressing the recommendations made by the April 2009 ROSC mission, including with respect to the measures adopted since the mission’s visit.

• Strengthen the legal framework to authorize the statistical compiling agencies, particularly the INEC and the BCCR, to collect statistical data from the nonfinancial private sector and to impose sanctions for noncompliance.

To implement a recommendation of this nature, coordination among various entities is required, in addition to legal opinions and legal procedures. Accordingly, the BCCR stands ready to coordinate the necessary arrangements with the relevant legal authorities, in an effort to obtain the desired legal framework.

• Expedite the implementation of the agreement between the BCCR and the INEC for collecting basic statistical data for macroeconomic statistics.

Currently, there is a “Framework Agreement for cooperation between the BCCR and the INEC” to support the management of basic information and statistics required by the BCCR to prepare macroeconomic statistics. In July 2009, the INEC delivered the directory of institutional units and establishments, which includes 47,424 records and which constitutes the first in a series of outputs expected to materialize under the Agreement. The directory will be updated by the Economic Surveys Area of the Economic Division of the BCCR. In addition, it is expected that in October 2009, the INEC will initiate the pilot test of the quarterly survey of the economically active population. Furthermore, the agreement for conducting the survey for the monthly monitoring of agro-industry is expected to be in place by the last quarter of the year.

In tandem with the above, the document entitled “Letter on the Project for Changing the Base Year for the Macroeconomic Accounts” specifies other outputs envisaged in the agreement between the BCCR and the INEC, which will facilitate efforts to compile the basic statistics needed to implement the change in the base year.

• Give priority to the on-going automation projects in all institutions.

The BCCR’s strategic projects include the project for automating the management of economic statistics (AGEE). Progress has been made in those automation efforts that will capture information on financial intermediaries. In addition, the agreements on information exchange currently being negotiated with public entities now include clauses that will simplify future automation efforts. For 2010, there are plans to build the technology platform for automatic data capture, beginning with the General Superintendency of Financial Institutions (SUGEF) and the Office of the Comptroller General of the Republic, in order to obtain accounting information on financial intermediaries and on the budgets of public entities, respectively.

The AGEE project is designed to be completed in 2013 and has the approval of the Board of Directors of the BCCR.

• Establish a formal and regular consultation process with data users.

The BCCR has various mechanisms for handling queries made by data users. Each table published on the website includes e-mail and telephone information for data managers. In addition, queries are received through the Information Center, the Central Bank Webmaster, and the administrator of the Economic Data Dissemination System (SDDE). Even though the wide range of channels for receiving queries facilitates user access, queries may sometimes take an inordinately long time to reach the right person. Moreover, once a query is received, there is no standard procedure or adequate record for handling the transaction. Options for complying with this recommendation are being evaluated, including getting the Customer Service Center’s call center involved in receiving and channeling queries from data users.

• Disseminate the statistical revision policy and identify and explain the revisions.

The Economic Division—which is committed to the principle of transparency—is for the medium term considering the possibility of including on the BCCR’s website a note identifying and explaining the main changes occurring in the revision of annual and subannual historical data. In January and July of each year, the BCCR publishes the document entitled “Programa Macroeconómico” (Macroeconomic Program), which includes the main determinants of changes in the forecasts of the main key variables monitored by the BCCR.

• Give advance notice of changes in methodology, source data, and statistical techniques used in compiling the macroeconomic statistics.

The Economic Division has maintained good dissemination practices with respect to giving advance notice of significant changes in methodologies, less so in terms of disseminating the use of new statistical techniques; the Economic Division expects to take action on this matter in the medium term.

B. National Accounts

• Firmly establish, by law or other formal provisions, the BCCR’s responsibility for compiling and disseminating the NA.

To implement a recommendation of this nature, coordination among various entities is required, in addition to legal opinions and procedures. Accordingly, the BCCR stands ready to coordinate the necessary arrangements with the relevant legal authorities, in an effort to obtain the desired legal framework.

• Provide access by the BCCR to the data on individual declarations available at the MOF. The BCCR should provide the tax authority with formal assurances that individual data will be handled with strict confidentiality.

In conjunction with the MOF, the BCCR is preparing a framework agreement to govern information exchanges, as well as rules and regulations pertaining to management of confidential information. This agreement is scheduled to be formalized by end-2009.

• Give high priority to the updating of the reference year of the NA.

One of the strategic projects approved by the BCCR Board of Directors involves changing the base year for the macroeconomic accounts; this project is expected to be completed in 2014. The purpose of the project is to enhance the quality, quantity, and timeliness of Costa Rica’s statistical output. Accordingly, the project has been designed to supply the inputs and improvements needed to ensure the adequate development and subsequent updating of the base year. However, the success of this project will largely depend on the extent to which the INEC can generate the basic statistics required within a reasonable period of time. Any delay on the part of the INEC would have an adverse impact on the project’s delivery date.

• Take advantage of the opportunity to incorporate the new features of the 2008 SNA that are particularly important for Costa Rica.

Currently, the Macroeconomic Statistics Department (DEM) is reviewing new features of the Manual of National Accounts (2008 SNA) to be incorporated in the project for changing the base year for the national accounts. Accordingly, research is being carried out to determine consistency and compatibility with the overall integrity of the national accounts system. Specifically, documents are being prepared discussing the implications of adopting the new SNA 2008 features. Ongoing research activities include: the treatment of goods for processing, the treatment of repurchase agreements, new classifications for financial assets and liabilities, the treatment of insurance, and the impact on measuring GDP resulting from the apportionment of financial intermediation services indirectly measured (FISIM).

• Update the ISIC classification and harmonize its use by all relevant institutions. Classify government final consumption expenditure by COFOG. Classify households’ final consumption expenditure by COICOP.

The BCCR has worked jointly with the INEC to define national versions of ISIC 4 and CPC 2.0. In addition, the BCCR is serving on a committee whose objectives include updating ISIC across government institutions.

One of the deliverables from the first phase of the project for changing the base year consists in an official paper containing the classifications to be used for industries (ISIC), goods and services (CPC), classification of individual consumption by purpose (COICOP), and classification of the functions of government (COFOG).

• Improve the estimation of quarterly GDP for activities currently estimated without indicators by conducting quarterly needed surveys and avoiding the use of the Chow-Lin method for benchmarking (the Denton method is preferred).

There are plans to carry out quarterly surveys of specific industries (restaurants, services rendered to businesses, public administration services, community services, social services, and personal services) with a view to preparing a quarterly indicator reflecting pertinent trends.

In the case of agro-industry, there is an agreement with the INEC which will provide monthly statistics (area and production), which will be used for benchmarking in the case of those activities that currently lack monthly indicators.

In addition, one of the projects envisaged as part of the exercise for changing the base year, and which is currently at the design stage, will involve setting specific requirements for a quarterly survey on the status of private construction projects. This will assist efforts to prepare a program for the scheduling of construction costs, which in turn will help to improve the monthly indicator for construction and real estate activities.

In 2006, the Macroeconomic Statistics Department carried out a benchmarking study to assess and compare the advantages of the Chow-Lin and Denton methods in order to establish national accounts on a quarterly basis; however, this experiment did not come down conclusively on either side. The study in question indicates that if the indicator is the right one, both methods will tend to yield quite similar results.

• Use data from both household surveys and any other available source, to generate employment and income matrices, by crossing data on employment by economic activity, size, employment category, and income components to better capture the activities of the nonobserved economy.

The framework agreement for cooperation between the INEC and the BCCR makes provision for conducting the quarterly survey of the economically active population. The results of this survey are expected to help prepare an employment matrix. This survey is geared toward households and is designed to compile information from employees and unemployed persons, and to gather information on hours worked, jobs, and remuneration of wage and salary earners. The object of the exercise is to ensure that the BCCR can calculate labor market indicators such as labor costs, unit labor costs, labor price indices, household incomes, total unemployment, indices of productivity, economic activity, etc.

• Improve the estimation of rentals for own-occupied dwellings by taking into account types of dwelling, geographical location, and other characteristics. The imputed rentals for own-occupied dwellings should be based on rentals paid for non-furnished dwellings.

The BCCR is arranging with the INEC to have the forms for the quarterly survey of the economically active population include a module connected with housing details. The goal is for the information provided by this module to generate data on the type of housing, its geographic location, ownership status, square meters, and actual and imputed rent. This information should in turn generate data on residential rents by type of housing, which is of vital importance in estimating residential rent (actual and imputed).

In addition, the same information is expected to be generated through the population and housing census which the INEC plans to carry out in 2011.

• Disseminate the most disaggregated supply and use table possible in terms of the imported products classified by CPC, as well as the intermediate uses of each domestic product allocated to the corresponding user activities.

Issues to be addressed when implementing the project for changing the base year include a matrix of imports. In addition, as part of the same project, there are plans to conduct surveys of enterprises in an effort to allocate the components of intermediate consumption to each of the corresponding user activities.

• Collect basic data on inventories of main inputs, products, and merchandise for wholesale and retail trade to directly estimate changes in inventories.

Since early 2009, the BCCR’s Economic Division has been engaged in a Business Outlook and Performance Survey which is carrying out regular (fortnightly) monitoring of inventory flows and of the outlook for the use of or demand for such inventories. This survey is applied to over 1200 selected firms based on the currently available framework of institutional units and establishments. This survey is expected to serve as a basis for designing a consultation tool for compiling data that can be used to directly estimate changes in inventories.

C. Consumer Price Index

• Establish a formal CPI advisory group comprised of important index users from the government sector, the business sector, labor unions, the press, and academia.

We do not deem necessary to establish a formal advisory group for the CPI with the participation of various sectors to obtain specific advice or assistance for the index. However, we feel this group should be formed when the methodological revision is initiated for each base change. On this recommendation, we are studying the possibility of establishing a site for receiving comments or advice from the various users, where not only doubts but also concerns for improving the indicator can be expressed.

• Include a rental equivalence component in the CPI market basket to represent the cost of owner-occupied housing.

At the moment, the CPI does not include either equivalent or imputed rentals corresponding to measurement of the price of the housing services consumed by homeowners. Indeed, we would point out that according to the international manuals there must be a clear relationship between a good or service and the price paid for it. In this case, the valuation of the benefit it provides to the homeowner is subjective, which is why this price was not included. The INEC feels that research needs to be started on this matter—bibliographical reviews, consultation of experts, and assessments of experiences in other countries that have incorporated this price in the CPI basket—so that the implications of including it in the index can be more clearly surveyed.

• Closely monitor markets to identify the need to incorporate quality changes in the CPI.

The methods used to incorporate quality changes in the CPI are implicit. The explicit or direct methods that provide information on the value of quality changes are generally not used because of the difficulty of obtaining adequate information and the need for additional resources to apply those methods, even though we know they can lead to improvements in the measurement of changes in the index. For the next change in the indicator base, we plan to conduct research on the methodology and procedure of hedonic regression and consider the possibility of applying it to articles where necessary. This is subject, of course, to the availability of appropriate information and resources.

• Initiate procedures to compare, on a regular basis, the behavior of consumer price indices with the behavior of similar series from the PPI, the agricultural price indices, and implicit deflators from the national accounts.

We shall begin with a revision of the PPI and then study how this indicator can help in assessing the behavior of the CPI and how to proceed with exogenous information. In future, studies will be done of additional indices of the other economic sectors.

• Expand the geographic coverage of the CPI to include all urban areas of the country. If it is not possible to collect prices in all urban areas, the market basket weights should reflect household expenditures from all urban areas.

This recommendation will be taken into account when the base of the index is being changed, considering the analysis of prices and household expenditure in various areas, provided that the necessary resources are available to carry out the pertinent studies and if the contribution to the representativeness of the indicator is significant. We must clarify that the INEC has no plans to create regional offices. In 2010 regional work will begin for the purposes of the quarterly employment survey. This does not imply that offices will be established in the areas furthest from the INEC head office to facilitate price collection for the CPI or other statistics.

• Change the national version of COICOP used for the CPI so that it fully corresponds with the international version of COICOP being used for the harmonization project for Central America, Panama, and the Dominican Republic.

The classifier of the articles used in the CPI corresponds to the Classification of Individual Consumption by Purpose (COICOP) in 12 groups. However, it was adapted to make it comparable with the classification of the previous basket and to take account of the purposes for which the articles in question are used in our circumstances. It is feasible to reconstruct the COICOP on the basis of the available information and the current classification, if this is of interest for comparing the CPI with those of other countries, readjusting the groups according to the corresponding articles and in the order of the 12 groups following the original version. In future, the use of the COICOP as established must be assessed, as there are various factors that would have to be considered.

Further to these comments on the mission’s recommendations, we want to add some observations to the content of the report:

1. The INEC currently has the personnel, computational resources, and adequate physical premises for calculating the consumer price index (CPI). This indicator has the attributes of quality and timeliness consistent with the availability of resources such as currently allocated to the institution. These and other matters covered in Law 7839 on the National Statistics System (SEN) could be improved to the extent that the government of the Republic is able to provide greater budgetary resources to the institution. This situation represents a tough challenge that the INEC is facing, and we hope it will continue to improve in the coming years.

2. The INEC is considering the possibility of conducting the next National Survey of Income and Expenditure (ENIG) in 2012, given that it is not feasible to do this for previous years for a number of reasons, including: the burden of the work that the INEC has planned for 2010 and 2011 (improved household survey, the quarterly employment survey, the national population and housing censuses, the agriculture and livestock surveys, the launch of the SEN, etc.), obtaining the financing necessary for designing the ENIG, and the development of the phases preliminary to implementation of the survey, for which various months are required before the start of data collection.

3. The 2006-base general index series is the only one available from 1976 to date. Linked series are not available for the large groups and products. However, we are certain that the specifications or groups do not vary substantially from one series to the other (1995 base and 2006 base), and the series can therefore be generated for a product or any aggregation.

4. As regards the remark on the sample of establishments at which the prices of 389 varieties are collected for 292 goods and services, we would clarify that the sample contains 1,840 commercial establishments and services and 210 households at which prices of housing rentals are collected.

5. We must emphasize that during its initial contacts with price reporting units, the INEC gives them the following information: an explanatory note on the CPI and its uses, a formal request for information, and a bulletin with the principal data from the indicator for the previous month, not a brochure summarizing CPI methodology.

6. The INEC periodically gives talks on the CPI to the National Commission on Salaries, but we would point out that it is not a part of the Commission. This Commission meets twice a year to assess public sector wage adjustments and consists of representatives of the Ministry of Labor and Social Security, the unions, the civil service, and the MOF.

7. We want to clarify that when it is indicated that a thorough evaluation of data sources is underway and that it is being used to update the PPI and national accounts data, this is a reference to the Directory of Establishments and does not apply to the CPI section.

8. We wish to stress that within the framework of the Support for Implementing the National Statistical Development Strategy (ENDE) project that is being implemented with World Bank support, the service of an expert in Administrative Law was contracted to revise the regulations of Law 7839. At the end of 2009, this expert will hand in a report and recommendations (including on the collection of statistical data from the nonfinancial private sector), which can be revised in the current standards to improve the country’s statistical production.

9. With regard to the formulation of the National Statistics Plan, we would point out that this is a part of the activities and products of the ENDE project that is being implemented with World Bank support. The timetable for defining the Plan was reprogrammed for the purpose of taking into account the results of the Inventory of the Principal Statistical Operations, which is already under implementation. Consequently, it is planned that in the first half of 2010 the National Statistics Plan (for the short and medium terms) will be available, along with the Strategic Plan of the SEN for the long term.

D. Producer Price Index

• Assign full-time staff to the compilation of the PPI.

At the BCCR’s Economic Division, the Index Number Area has plans (for 2010) to update the basket for the Producer Price Index (PPI) and will have full-time staff to keep it updated.

• Update the weights, the establishment sample, and the product sample for the PPI.

The preparation of the new PPI includes updating for the weights and the basket of products.

• Expand the coverage of the PPI to include export goods.

The information necessary to carry out this recommendation is currently available.

• Compile the PPI by product and by economic activity.

This recommendation is being considered for the current calculations for the PPI. The PPI for services has been classified on the basis of ISIC3 and CPC. The components of the new PPI will also be classified on the basis of ISIC3 and CPC.

• Establish a policy for revising the PPI at least every seven years.

The operations plan for the Economic Division will include this recommendation.

• Make compilations of the PPI for product aggregation at a greater level of detail than the four-digit ISIC [Rev. 3].

The calculations for the new PPI will use this recommendation.

E. Government Finance Statistics

• Other quality management. Quality is a main concern in the GFS production. However, no formal processes are in place to monitor the quality of the statistics compiled and disseminated, or to deal with quality issues in the statistical program.

Regarding consolidation of public sector data, before starting the process we conduct a thorough revision entity by entity, looking for inconsistencies in the recording of revenues and expenditures. Additionally, there is a procedure in place that requires each analyst of the Technical Secretariat of the Budgetary Authority (STAP) to validate the data inputted by the entities before accepting the information to be uploaded in the System of Data Consolidation (SICCNET).

• Disseminate on a monthly basis the budgetary central government revenues and expenditures table on a “recognized” basis.

• Include financing data on the table. Disseminate on a monthly basis the revenues, expenditure, and financing table for the MOF’s “reduced” nonfinancial public sector. When dissemination begins, discontinue publishing the BCCR’s table for the “reduced” public sector.

With respect to the first recommendation, data on “recognized” revenues and expenditures are currently being compiled on a monthly basis at the time financing data are closed. The Directorate of Public Credit is responsible for this closing. These data are sent to the Western Hemisphere Department of the IMF on a monthly basis, as required for the monitoring of the stand-by agreement signed by our country.

For 2010, the “recognized” figures for the month of January will be published on the website of the MOF and disseminated to the press. We are also considering publishing the operations table on a monthly basis, and not only annually as is the current practice.

F. Monetary Statistics

• Expand the coverage of the ODCs survey through the inclusion of the accounts of short-term investment funds.

Currently, the General Superintendency of Securities (SUGEVAL) periodically reports information on short-term investment funds, which is necessary if this information is to be included in the sectoral balance sheets, surveys, and related statistics on ODCs. This recommendation will be implemented in early 2010.

• Include the social security system accounts within central government.

Joint efforts are currently being made with the General Superintendency of Financial Institutions (SUGEF) to develop an information system that will enable us to sectorize the accounts in the financial institutions balance sheet with a greater degree of detail.

• Include all short-term securities issued by the BCCR and by ODCs in the measure of broad money disseminated by the BCCR. Exclude all long-term securities issued by the BCCR and by ODCs from the measure of broad money disseminated by the BCCR.

This recommendation has already been addressed.

• Review the consistency between monetary and government finance data on a monthly basis, and eliminate any found discrepancies.

Efforts are being made to coordinate with the pertinent office to review the consistency of the monthly information.

• Start compiling an OFC survey with data for pension funds, insurance corporations, long-term investment funds, and the Housing Mortgage Bank.

This is one of the projects scheduled to begin in the second part of 2010.

G. Balance of Payments Statistics

• Revise the legal framework to give the BCCR authority to get primary data from nonfinancial private sector units in a timely fashion.

To implement a recommendation of this type, coordination is required among various entities, in addition to legal opinions and legal procedures. Accordingly, the BCCR stands ready to coordinate the necessary arrangements with the appropriate legal authorities in an effort to obtain the desired legal framework.

• Set up a BCCR statistics users group to provide a permanent forum for discussing ways of better meeting users’ needs.

The Statistics Area of the External Sector (AESE) will coordinate with the Economic Surveys Area to prepare a survey aimed at specialized users, with the aim of determining the relevance and practical usefulness of the existing statistics, as well as to assess possible information needs that are currently going unmet.

• The AESE, in coordination with AEE, should identify through the business directory construction activities in Costa Rica that have a component of direct investment, and include them in the financial account. Regarding the compilation of data on health-related activities and compensation of seasonal workers, the AESE will start implementation of specific surveys in these areas.

At present, work is in progress on determining the requirements for a quarterly survey on monitoring the status of construction projects. This survey is expected to include questions that will make it possible to identify FDI-related components. Action will also be taken to coordinate with the Costa Rican Tourism Institute to ensure that tourism concession contracts can incorporate key BOP information.

With respect to measuring health-related activities, efforts will be made to coordinate with the Services and Construction Statistics Area to assess coverage of the current indicator, and relevant additional actions will be carried out.

Issues pertaining to seasonal workers are being addressed in a project to strengthen procedures for compiling data on special activities. Research will be conducted into work done in this field in other countries (Mexico, for example) and action will be taken to coordinate with the Economic Surveys Area (AEE) in order to ensure that data-collection surveys take into account country-specific conditions.

• The BCCR should exclude the capital contribution to FOCEM and the claim on Honduras’ securities from official reserve assets since these are to be classified in other portfolio assets of the monetary authorities.

The capital contribution to FOCEM ($4.0 million) and the questioned claim on Honduras ($2.7 million) fail to meet all the criteria to be regarded as reserve assets. The Macroeconomic Statistics Department agrees with the IMF’s recommendation and considers that these items should be reclassified under other assets of the monetary authorities.

• Estimate flows and stocks of foreign direct investment equity at market prices. Estimate market prices of the transactions in goods of the high technology industry in Costa Rica.

The Statistics Area of the External Sector (AESE) will take steps to gain access to the financial statements of firms falling within the purview of FDI, which are listed on the major stock markets at the international level. The value of the shares of these enterprises will also be tracked, in order to make this a benchmark and to make the requisite adjustments in BOP and IIP statistics.

• Record transactions related to government external debt liabilities on an accrual basis.

Action will be taken to coordinate with the Public Credit Directorate of the MOF to generate external debt statistics on an accrual basis, and to make sure that this information is available in the SIGADE (Debt Management and Analysis System).

• Develop sound statistical techniques to estimate nonrecorded trade flows and include them in goods.

Research will be carried out to investigate the procedures used in other data-collecting economies to record transactions of this nature, and the feasibility of implementing such procedures within Costa Rica will be analyzed.