The Leading Indicator of Employment (LIEmp) aims to anticipate the main labor market movements in Brazil, based on data extracted from the following questions of the Business Survey and Consumer Expectations Survey produced by FGV/IBRE:
• Expectations regarding the labor market (Consumer Survey);
• Expected employment (Industry Survey);
• Current business situation (Industry Survey);
• Business trend (Industry Survey);
• Expected employment (Services Survey);
• Current business situation (Services Survey);
• Business trend (Services Survey).
The Consumer Expectations Survey is a monthly survey that generates indicators based on the feeling of the Brazilian resident in the seven main capitals of the country on topics such as the local economic situation, families’ financial situation, employment, intention of spending on durable goods, etc. The topic "employment" covers questions about the present and future labor market situation. The LIEmp includes results obtained from the question in which the consumer reveals their expectations for the coming months regarding the labor market in the city they reside.
The Consumer Survey indicator, obtained from the question about the future situation of the labor market, is constructed from the disaggregated data of the seven main capital cities in Brazil, as listed below:
• Belo Horizonte;
• Porto Alegre;
• Rio de Janeiro;
• São Paulo.
The indicators for each of the capitals are adjusted seasonally and are expressed as deviations from the long-term trend through the extraction of the estimated trend. The indicators are aggregated using a weighting generated by Principal Component Analysis (PCA) in order to form the Future Local Employment Index of the Consumer Expectations Survey.
The PCA is a multivariate statistical technique that consists of transforming a set of variables into another set of variables of the same dimension called principal components. It is associated with the idea of reducing data mass with the least possible loss of information. Each principal component is a linear combination of all the original variables. They are independent of each other and estimated for the purpose of retaining the maximum information - in terms of the total variation contained in the data. The method separates important information from the redundant and random ones.
The Leading Indicator of Employment also includes six series from the Business Surveys: three from the Industry Survey and three series from the Services Survey.
The Manufacturing Industry Survey provides monthly information on the general state of the economy and its trends, as it can guide business and economic policy decisions. The survey covers issues that deal with the present and the near future. The data is aggregated from 21 industrial kinds. In the studies, aiming the elaboration of the LIEmp, different forms of aggregation were tested from 14 subsectors, according to the following discrimination:
• Non-Metallic Mineral Products;
• Electrical and Communications Material;
• Transportation Material;
• Pulp, Paper and Cardboard;
• Pharmaceutical and Veterinary Products;
• Plastic Material Products;
• Textile industry;
• Clothing, Footwear and Fabric Artifacts;
• Food Products;
• Other products.
Analogously to the exercise of aggregating the future labor market situation indicator, from the Consumer Expectations Survey, the indicators for each of the sub-sectors are adjusted seasonally and are expressed as deviations from the long-term trend through the extraction of the estimated trend. These indicators are then aggregated, using a weighting generated by PCA in order to form the following series of the Industry Survey, used in the calculation of LIEmp:
• Forecast of employment level for 3 months - Series of 14 sub-sectors aggregate by Principal Component Analysis;
• Current business situation - Series of 14 sub-sectors aggregate by Principal Component Analysis;
The item of Business trend for 6 months is incorporated into LIEmp in its officially disclosed format.
The Services Sector Survey is a monthly study that seeks to monitor the current situation and the short-term trends of this sector. In the Services Sector Survey, companies that operate in the national territory are surveyed. The sectoral coverage of the survey is similar to that of the Annual Survey of Services (PAS) by the IBGE, broadly covering all service sectors except government, continuing education and health activities, the financial sector, and trade. The sectoral segments are determined by the National Classification of Economic Activities, version 1.0 (CNAE 1.0) by IBGE. The most disaggregate results are obtained for twenty-nine Groups (three digits) of CNAE 1.0, and aggregated into the seven Segments (four-digit sum) of CNAE 1.0 of the IBGE statistics for the Services sector presented below:
• Services provided to families;
• Information services;
• Services provided to companies;
• Transport, ancillary services to transport and mail;
• Real estate activities and rental of properties;
• Maintenance and repair services;
• Other service activities.
Analogously to the exercises performed in the Industry and Consumer surveys, the indicators for each of the seven segments are adjusted seasonally and expressed in deviations from the long-term trend through the extraction of the estimated trend. These are aggregated using weighting generated by PCA in order to form two series of the Services Sector Survey used in the LIEmp calculation:
• 3-month employment level forecast - Series of seven segments aggregated by Principal Component Analysis;
• Current business situation - Series of seven segments aggregated by Principal Component Analysis.
Again, in the case of item Business Trend for 6 months, the performance of the officially disclosed series exceeded that of the aggregations made with weightings generated by the Principal Component Analysis and, therefore, the way in which the indicator of this item is incorporated into LIEmp.
For the seven indicators described above to be aggregated, the weightings are given by Principal Component Analysis. The indicator starts in June 2008, the date from which the survey data used is available, and is based on June 2008 = 100.
For the selection of the best combination of survey series, it was applied as reference the Employed Population, from the Monthly Employment Survey (PME) by the IBGE was used as reference. At no point, however, does the indicator propose to reproduce - or even predict - any specific statistical series.
The LIEmp is an indicator created from statistical data produced by IBRE/FGV. The choice of series and weights was performed by statistical methods, and may undergo periodic revisions. On this occasion, any changes will not cause changes in the previously disclosed series.