Itaú BBA - Itaú Activity Index for Brazilian States

Macro Vision

< Back

Itaú Activity Index for Brazilian States

March 16, 2018

Results point to a heterogeneous recovery so far

For the version with all charts, please open the attached pdf file 
 

This report launches an economic activity index for each Brazilian state (Itaú State Activity Indexes - ISAI). Series are available on a monthly basis, until December 2017.

With these indexes, analysis of the level of economic activity can be broken down by state after 2015 (the most recent year for which official data for states were published) and with greater frequency than that provided by the official indicator calculated by census bureau IBGE.

The recovery in activity in 2017 was more pronounced in Mato Grosso, Amazonas, Pará, Roraima and Pernambuco. Five states have not yet signaled a recovery: Alagoas, Paraíba, Mato Grosso do Sul, Rio de Janeiro and Espírito Santo.

Eight states show activity levels above those observed before the recession: Roraima, Mato Grosso, Acre, Tocantins, Pará, Paraná, Piauí and Amazonas.

Purpose: Track activity in states after 2015

IBGE breaks down Gross Aggregated Value by state, which is virtually the same as GDP. However, there are lags in the data. As of the middle of 1Q18, there is only data until 2015. Furthermore, these are only available on an annual basis, while aggregate GDP is available as far as 4Q17, and there are monthly activity indicators related to January 2018. 

With several monthly indicators referring to each state and available for the recent period, we were able to get proxies for GDP growth in each state, based on the historical relationship between these monthly figures and official GDP.

These proxies can contribute to answer many questions, such as:

- How is the ongoing gradual recovery distributed across states?

- What states experience economic activity levels above the peak seen before the recession?

Creating the series

From a supply standpoint, GDP may be understood as the sum of final output of all three economic sectors: agriculture, industry and services. Based on this definition, proxies for state GDP are created from monthly data referring to each state’s economic activity. These are the series used: 

- Agricultural production proxy: calculated internally, based on state agricultural production data gathered by the following IBGE surveys: Agriculture and Livestock Census (2006), Municipal Agricultural Survey (PAM), and Systematic Survey of Agricultural Production (LSPA).

- Monthly Industrial Survey: IBGE Physical Production (PIM-PF): provides monthly data referring to industrial output in each state. Importantly, there are figures available for only 14 states — thus, a limiting factor in the calculation of GDP for some states.

- Monthly Retail Survey IBGE (PMC): tracks retail sales in each state. Although the indicator tracks sales (not production), the series captures key information about retail-related services. 

- Net formal job creation as per the General Registry of the Employed and Unemployed (CAGED – Labor Ministry): creation of formal jobs, incorporating relevant information about the level of economic activity in each state, not reached by the surveys mentioned above (as other activities in the service sector).

We considered using data in the IBGE’s Monthly Service Survey to capture the effects of changes in sectors covered by the survey. However, the series is short (started in 2012) and its adoption would limit our study to 2012-2017. Thus, we chose not to use it. 

Using these data, the official series of Gross Aggregate Value for each state (available through 2015), and econometric models, we created preliminary monthly series for the level of economic activity in each state. Finally, we applied an adjustment factor in these series so that, aggregated in a moving base by the weight of each state in GDP in the previous year, they become compatible with our Itaú Unibanco Monthly GDP (please refer to Revision of Itaú Unibanco's Monthly GDP Series). The resulting post-adjustment series generate the economic activity indexes for each state (ISAI). (please refer to the technical note for details).

Results point to a heterogeneous recovery

The table below shows estimates for the accumulated performance in each state (and region) between (i) 4Q17 and 2Q14, and (ii) 4Q17 and 4Q16. These intervals enable an assessment of the recovery in 2017 and the performance accumulated since 2Q14, respectively.

Highlights: 

  • The recovery in activity in 2017 was more pronounced in Mato Grosso, Amazonas, Pará, Roraima and Pernambuco. Five states have not yet signaled a recovery: Alagoas, Paraíba, Mato Grosso do Sul, Rio de Janeiro and Espírito Santo (the latter two were negatively affected by developments in the oil & gas industry).
  • Eight states show activity levels above those observed before the recession: Roraima, Mato Grosso, Acre, Tocantins, Pará, Paraná, Piauí and Amazonas.

 

Artur Manoel Passos

Alexandre Gomes Cunha



< Back