National Institute of Statistics and Geography’s Response

Response on National Accounts Recomendations

3. In the annual accounts for agriculture, to develop estimates of carry-over work-in-progress for short-lived crops with the help of improved annual survey procedures. This will allow agriculture to be included in the benchmarking of the quarterly to the annual data.

National agricultural surveys (ENA) 2012 and 2014 have covered the production and characteristics of the agricultural and livestock production units, providing information on major crops and livestock species, production costs and machinery and related equipment, for a period covering two main agricultural crops short cycle. The results of both surveys will be evaluated and taken into account for agricultural crops and animal species. With this information, complementary to the related administrative records, the feasibility of offering a broader picture of the supply and use of 33 products subject to the ENA, from the Base year change to 2013 (CAB 2013), will be assessed.

4. Collecting and use of data on inventories in quarterly or monthly surveys of the wholesale and retail trade industries.

In the annual wholesale and retail survey (EAC), good’s inventories for resale information is available, according to the economic activity, given that in each event the annual survey the holdings information is requested at the end of the year of reference and the immediately preceding year, thus allowing to calculate changes of inventories. This information is available from the new statistical design of the EAC and will be evaluated jointly with the valuation methods commonly used in commercial activities surveyed, to incorporate them in annual calculations of national accounts, from the CAB 2013.

Thus, the possibility of incorporating these questions in the monthly wholesale and retail survey of businesses should be considered carefully in order to avoid affecting timeliness and quality of responses, and the positive attitude of respondents will be evaluated. Recognizing that measurement of inventories on a monthly or quarterly basis is a complex task, due to the difficulty of timely compiling and the quality of the response by all companies considered.

5. Estimating changes in inventories for raw materials and the trade industries, and collecting quarterly data on changes in inventories. Incorporating inventory valuation adjustments into procedures for making current price estimates of production or expenditures, as appropriate.

As mentioned in the previous recommendation, best practices will be explored to assess, in economical manner and without burdening our informants, the possibility of incorporating a module of questions as recommended for the monthly wholesale and retail survey of businesses.

6. In calculating FISIM, include the opening balance in the average stock of loans or deposits. In the case of subsidized loans from the Development Banks, measure the output of FISIM based on arms-length rates charged on similar loans to calculate subsidy on the interest rate and record a current transfer from the Development Bank to the government to cover the cost of the subsidy to the borrower.

This recommendation is being studied in order to analyze the impact of the subsidy and the transfer generated by subsidized loans from development banks, in order to evaluate its possible instrumentation from the CAB 2013, taking into account the changes that the Bank of Mexico has planned with the adoption of the sixth edition of the Manual of Balance of Payments and the International Investment Position.

7. Use the BI ratio of the fourth quarter of the previous year to avoid a step problem in the first quarter.

The adopted procedure use one of the three methods of setting recommended by the IMF Manual of Quarterly National Accounts and ensures consistency in all short-term indicators, so the recommendation is considered as not applicable.

The results of the assessment in Institute were presented to the IMF mission, highlighting that in the case of Mexico, the instrumentation of the recommended method would generate distortions or breaks in the series between monthly and quarterly indicators, as well as between national and regional statistics, so BI annual average adjustment factor will still be used to minimize these differences. The aforementioned assessment covered more than 736 series observed between the original series and the Benchmarking ones, for quarterly GDP and monthly national indicators and by State.

8. Apply either volume indexes for each type of good or import subject to taxes and subsidies on products to separate base year estimates of the net tax from each type of good or import, or estimate volume measures of taxes/subsidies on products by applying base-year-tax rates to the volume of transactions subject to a specific tax/subsidy.

This recommendation is adopted to review the calculation procedures used for the determination of taxes and subsidies on products at constant prices. The analysis of this change in the method of calculation it is being evaluated to incorporate it from the CAB 2013.

9. In the quarterly accounts, developing methods to measure quarterly government consumption on an accrual basis and to guarantee that estimates of government investment reflect changes in the work-in-progress.

This recommendation is closely linked to the first cross-cutting recommendation, so it will be proposed to address the issue within the working group formed in the framework of the SNIEG.

10. Developing complete estimates consumption of fixed capital (CFC) for the general government. Include the CFC expense in the measure of government output and the measure of output of nonmarket producers whose output is based on expenses.

To meet the recommendation, series of collections of capital and the fixed capital consumtion of the General Government will be compiled, for their evaluation and incorporation, as soon as possible.

11. Besides, investigating sources of individual revisions that are especially large, investigate the presence of any systematic biases in the preliminary estimates in analytical studies of revision histories.

The information needed to analyze and evaluate the nature of these differences will be collected, in order to document the results of the assessment and present them with the corresponding statistics to CAB 2013, both in short-term and annual series.

12. Calculate balanced supply and use tables that show uses by industries of inputs produced by other industries on a yearly basis to verify the internal consistency of the accounts.

Best practices and international recommendations will be revised, in order to obtain and study, sources and methods regarding annual calculation of the inter-sectoral transactions. Its implementation from the Base year change to 2013 shall be assessed.

13. Investigate inconsistencies between the capital accounts and the financial accounts to spot their specific causes and identify ways to reduce their magnitude.

This recommendation will be adopted considering a review of major interrelationships between current accounts and capital of institutional sectors, as well as those referring to accounts of accumulation and funding, subject to the availability of information. It will be incorporated in the CAB 2013.

14. To provide a specific explanation of the difference between general government saving and investment in the national accounts and the deficit and investment published by the Ministry of Finance, a reconciliation showing the sources of differences should be prepared jointly by INEGI and the Ministry of Finance.

This recommendation will be part of the proposals that will be brought up to the group that is being proposed with State Units responsible for accounting harmonization and the generation of statistics of public finance, in accordance with the first recommendation of the cross-section.

15. To publish more detailed information on products and on expenditures of households as estimated in the framework of the national accounts.

Starting with the next CAB 2013, there are around 200 strategic products scheduled to be released, which we will disseminate the supply and intermediate and final uses, in order to publish them as a complement of supply and use tables. This action is considered that will attend the recommendation.

16. Providing detailed information on calculation procedures, either in the metadata documents or in analytical background papers or working papers will help users to understand and interpret the data.

International best practices will be reviewed to assess its implementation, so that users may better understand and interpret the results of the national accounts.

Response on Other Statistics and Web page Recomendations

17. Assign the compilation of seasonally adjusted series of national accounts to the DGACN as they already perform this duty in order to avoid delays in the publication of these series:

INEGI is working on the proper procedure to strengthen the capacities of national accounts staff in short-term series seasonal adjustments. Capacity building activity will be led by the staff who is currently in charge of the seasonal adjustment.

Subsequently, the alternatives will be evaluated to ensure the disappearance of existing delays and increase the level of breakdown of series which are seasonally adjusted. It should be noted that, for the main aggregates, original and seasonally adjusted figures are disseminated at the same time.

18. Design seasonal adjustment procedures to minimize inconsistencies between seasonally adjusted estimates. The identification of outliers and other assumptions and options used in estimating seasonal factors should be consistent in the aggregate and the time series that it comprises. Important decisions made in estimating the seasonally adjusted series and economic meaningfulness of the results from the seasonal adjustment process should be reviewed for reasonableness by an economist with expertise in national accounts. Improving communication with key users of the accounts concerning seasonal adjustment procedures and outcomes should be a priority.

Regarding monthly, quarterly and annual indicators consistency, seasonal adjustment is only applied to the first two periodicities. The possible inconsistency among the indicators could be a result of the calendar effects correction (Easter week, trading days and leap year), jointly with the moving seasonality that indicators show along the years. International recommendations suggest not to force the sum (or average) of seasonal adjusted (s. a.) data to equal the sum (or average) of the original data from a theoretical point of view, since forcing can derive in: a) a bias in the seasonally adjusted data, especially where calendar and other non-linear effects are relevant, and b) a not optimal final seasonal adjustment. However, the best choice will be evaluated, taking aim to the best seasonal adjustment and communication with data users.

Regarding the inconsistency between monthly and quarterly s. a. series, the inter-institutional working group on seasonal adjustment in Mexico, comprised Bank of Mexico, Ministry of Finances, Ministry of Economics, and the INEGI, has been working closely on this issue for the last three years. As a result, a substantial improvement in such consistency has been achieved and the new approach has been applied to short-term indicators produced in the System of National Accounts since 2013. In 2014 this improvement was recognized within the Committee for Macroeconomic and National Accounts Statistics, to which the working group on seasonal adjustment reports.

Having in mind the aforementioned consistency, the analysis of different types of calendar effects requires greater detail in models revision and therefore, more time. However, since the X-12 ARIMA procedure involves the calendar effects correction and the use of filters, it is not reasonable to expect consistency between monthly and quarterly series for an indicator at 100 percent.

Regarding the inconsistency between the aggregates and the components, INEGI uses the direct approach. This approach is accepted by the international guidelines on seasonal adjustment of OECD and Eurostat. Also the IMF Manual of Quarterly National Accounts states that “Conceptually, neither the direct approach nor the indirect approach is optimal. There are arguments in favor of both approaches. It is convenient, and for some uses crucial, that accounting and aggregation relationships are preserved. Studies, however, have shown that the quality of the seasonally adjusted series, and especially estimates of the trend-cycle component, may improve, sometimes significantly, by seasonally adjusting aggregates directly or at least at a more aggregated level. Practice has shown that seasonally adjusting the data at a detailed level may leave residual seasonality in the aggregates, result in less smooth seasonally adjusted series, and provide series more subject to revisions.”

In the case of Mexican series, the indirect approach for the aggregates frequently presents trend-cycle components with high volatility, provides more revisions, shows residual in the adjustment of calendar effects, and sometimes residuary seasonality. That is the main reason for preferring the direct approach, which provides better results on the behavior of the aggregates.

19. Establish and announce a schedule for publishing the seasonally adjusted data.

Recently, on last February, the dissemination of the products of short-term approach is now based on analysis of seasonally adjusted series. For this reason, dissemination of these is now associated with the official 2015 data release calendar.

20. Include a module on monthly/quarterly costs for the agricultural cycle (sowing, growth, and harvest) in the 2014 Agricultural Survey for important crops, in order to obtain the data to calculate annually and quarterly agricultural work-in-progress.

The national agricultural survey (ENA) is a survey that is currently carried out in Mexico in biannual basis, so it is not possible to obtain a monthly or quarterly indicator. Specifically, the part dealing with costs has not been captured in previous agricultural census. The first agricultural national survey (ENA) 2012 became a first approach to obtain expenses exclusively from large producers, whereas in the survey of 2014, obtaining the expenditure was extended to the whole production units, and it even sought to capture some specific expenditures for crops in order to provide a more accurate picture on the conditions in which the agricultural activities are made.

Data captured in both polls, 2012 and 2014, will be reviewed and utilized nationally and by State within the framework of the Base year change to 2013, with the aim of assessing the feasibility of calculating costs of harvested and in progress agricultural production, for the calendar year. With the conclusions that are to be obtained, it will be possible to propose improvements to the modules of expenses or costs in agricultural statistical projects.

21. Conduct a survey in the most important states and local governments in order to obtain timely and representative data on these institutional units.

State and municipal public finance statistics are not responsibility of the Institute, so the work is focused on the capture of information on income and expenditure of local governments on annual and monthly basis. This information has been used to complement the public finances of the General Government estimates. The figures generated by the Institute are scattered in different periods, taking into account the process of review and approval of the respective local congresses.

The absence of total accounting harmonization of a section of the federal entities and their municipalities strongly condition the recommended sample design, so with the compiled information from public accounts for the fiscal year of 2014, that will start in April this year and records of timeliness and completeness in the response showed by the entities over time, the best option will be assessed and we will propose it within the group to be formed on harmonization and generation of public finance statistics.

22. Study the results of the household surveys and the ENIGH in order to determine the extent of income underreporting and, if necessary, adjust output, compensation of employees and mixed income of the household sector to improve coverage of the informal sector and to reduce overall discrepancies between the biennial surveys and the annual national accounts.

The best practices and international recommendations for this type of settings will be explored, as well as those ones arising from groups of experts of the international organizations such as the OECD and the ILO, and through exchanging experiences in the interior of the Institute, in order to develop methods of adjustment that will reduce the discrepancies between the various forms of cash income that capture the occupation and employment surveys and the national surveys of income and expenditure of the households, among themselves and regarding primary income and gross disposable income in households accounts by institutional sectors.

23. Undertake research on the need for under-reporting adjustments for data from household surveys on income from informal activities and some kinds of expenditures to reduce disparities.

This recommendation is complementary of No. 22 and will be attending jointly, assessing the feasibility of using the results of the research on the subject and if that is the case, apply them in the next Base Year Change to 2013.

24. Include questions on intermediate inputs in monthly surveys of manufacturing industries to improve the quarterly and preliminary annual estimates of intermediate inputs.

The Institute redesigned the monthly survey of the manufacturing industry to capture establishments mainly serving the domestic market, as well as those that form part or belong to a manufacturing company of export within the “Program for Manufacturing, Maquiladora and Export Services (IMMEX).” In the case of IMMEX establishments, we are already questioning on expenditures on intermediate inputs of national and imported origin (raw materials, containers and packaging, freight charges and tolls, professional services, rentals, among others), which facilitates calculations of value added export of the manufactured global industries in the national accounts. (VAEMG).

Taking into account the foregoing, we will evaluate the possibility to extend to the rest of the establishments in the sample, the module of expenditures on intermediate inputs, which we hope will strengthen quarterly and annual national accounts estimates.

25. Standardize the imputation methods acrossall surveys and apply the cluster analysis technique (the percentage change of the data from an establishment with similar size andtechnology in the same economic activity and geographical area, if available; if not, the percentage change of the activity class for the current period).

In the framework of the Integrated System of Economic Units Surveys, a set of statistical methods for imputing missing data has been defined, which consider the specific characteristics of the sector of activity, the sample design and economic variables.

Missing data imputation methods use the technique of cluster analysis for establishments of the same economic activity of the North America Industrial Classification System (NAICS), and the evaluation of the method to extend its application is underway, considering also the size of the establishment based on employed personnel and income, as well as the geographical area to which it belongs, for all surveys in specific variables.

26. A survey of users of the accounts conducted by INEGI and discussions with users showed a strong demand for specific explanations for revisions. Include a discussion of the sources of the largest or most important revisions in press releases presenting revised estimates. If a longer investigation is needed to determine the cause for a large revision, the research findings should be reported when they become available.

This recommendation will be attended soon. It will be proposed to the Government Board of the Institute to supplement the guidelines currently governing policy of figures revisions, through incorporating a section dealing with the elaboration of explanatory notes on major changes in the figures when these are carried out.

27. The website functionality should be improved, and an index by keyword of the available SCNM series should be available on the website. Significant newly available data should receive additional publicity.

Regarding an index by keyword, communication has been established with the office in charge of Data Communication, to assess the possibility of this index to the GDP and National Accounts web portal, considering the current policies and guidelines to do it. Also, significant newly available data will receive additional publicity.

28. Post responses to the most common questions on a national accounts “frequently asked questions” (FAQ) web page.

This and the previous recommendation are complementary and they shall be adopted considering the current publishing policies of the Institute.