Itaú BBA - Could the Brazilian economy grow with high unemployment?

Macro Vision

< Back

Could the Brazilian economy grow with high unemployment?

May 16, 2017

It is possible to see significant economic growth without a significant recovery in the level of employment.

  • The strong decline in employment in the recent recession naturally leads to the following question: is a rebound in economic activity possible without a recovery in employment? Could we experience, as in other countries, a jobless recovery?
  • Our analysis based on past recessions in Brazil shows that a rebound of economic growth does not necessarily depend on a significant decline in the unemployment rate or a strong recovery in consumption.
  • We believe that the high level of unemployment is consistent with the continuity of disinflation, a process that will likely continue throughout 2018.

The Brazilian recession started in June 2014[1] has produced a significant drop in the level of employment. The national unemployment rate has doubled, rising from 6.6% by mid-2014 to 13.2% currently. This strong deterioration in the level of employment in the recent recession naturally leads to the following question: Is it possible to see a recovery in economic activity without a recovery in employment?

The answer to this question calls for labor productivity analysis. If productivity gains drive economic growth, we may see a significant recovery in economic activity taking place before a significant recovery in the level of employment. In other words, if workers already employed increase their level of production, GDP will grow without the need for an increase in employment.

We measured productivity by the ratio between GDP and employment (EP), using both the employed population from IBGE's PNAD Contínua survey[2], and the employed population in the formal market, measured by CAGED (Chart 1). The GDP/EP ratio’s year-over-year indicates how productivity grows over time. Productivity tends to be procyclical in periods of inflection of economic activity, receding during recessions (periods highlighted by the grey columns) and recovering right after the end of such recessions. The data shows a clear deterioration in productivity since the beginning of the current recession (June/14). It is worth mentioning that this declining path already seems to be reversing.      

An important difference appears, however, between the two measures (CAGED and PNAD). Productivity growth as measured by CAGED appears to be procyclical in periods of entry and exit from recessions, but is rarely positive. For much of the past decade, for example, productivity growth measured by CAGED is consistently negative, which seems counterintuitive. The fact is that the economy underwent a noteworthy process of formalization during this period. Thus, CAGED's employed population (formal employment only) grew above GDP, not because productivity was declining, but because formal employment grew consistently above total employment over the period. This phenomenon leads us to focus our analysis on PNAD Contínua data, which aggregates both formal and informal employment, allowing for a more accurate analysis of the dynamics of productivity.

In chart 3, we break down GDP’s year-over-year growth into the employed population growth and the productivity growth. Observe that, despite the recent sharp drop in employment, the decline in productivity was the main reason behind the recession that began in mid-2014. In cumulative terms, from the first quarter of 2014 until the last quarter of 2016, GDP fell by 9.0%, with productivity falling by 6.9% and employment by 2.3%.

Thus, productivity is likely to maintain its procyclical nature, contributing to economic growth as the economy exits the current recession. Therefore, it is possible to see significant economic growth without a significant recovery in the level of employment. Also in chart 3, we have included our growth forecasts for the three variables that make up the chart (GDP, employment and productivity) for 2017 and 2018. Employment tends to resume growth, but a moderate expansion along 2018 will not be enough to compensate for the growth in the workforce, a necessary condition for the unemployment rate to decline.

We expect GDP to grow 1.0% this year and 4.0% next year. On the other hand, we expect the unemployment rate to continue to rise over the coming quarters, peaking at 14.0% by the end of 1Q18 and then slowly drawing back to 13.6% by the end of 2018.

In chart 4, the level of productivity (instead of its growth rate) shows more clearly the significant decline in productivity since the start of the recession. Using our forecasts for 2017 and 2018, productivity will still not recover it previous level.

During economic recoveries, the increase in productivity leads to a reduction in labor cost as a percentage of total production costs, as highlighted in the BCB's latest Inflation Report.[3] Thus, the procyclicality of productivity contributes to the recovery of profits and corporate deleveraging in the aftermath of recessions.

Table 1 shows the cumulative performance of a number of macroeconomic variables in the first year after the end of recessions. We can see that GDP typically grows strongly in the first year after the end of recessions, while job creation, although positive in almost all cases, always underperforms activity growth.

This is all the more evident when we note the sluggish decline in the unemployment rate[4], which shows that the pace of recovery in employment is barely enough to offset the workforce’s expansion. In other words, recoveries in growth soon after recessions are typically strong, but do not produce a significant recovery in the labor market.[5]

Consequently, it is not surprising that household consumption growth also typically falls short of GDP growth immediately after recessions. That is, consumption is not usually a component of aggregate demand that boosts GDP growth upon the end of recessions.[6]

In short, our analysis based on past recessions in Brazil shows that a possible resumption of economic growth following the recession that began in June 2014 depends neither on a significant drop in the unemployment rate nor a strong recovery in consumption. We believe that the high level of unemployment is compatible with the continuity of disinflation, a process that will likely continue throughout 2018.


[1]  In this report, we use dates provided by CODACE to define the beginning and end of recessions in Brazil.

[2]  The PNAD Contínua series begins in January/2012, but was estimated backwards using the Monthly Employment Survey (PME) based on the January/2012 - February/2016 period, during which both PME and continuous PNAD were disclosed by IBGE.

[3]  "Considerations about the evolution of labor productivity and the level of employment in Brazil", Inflation Report for March/2017, pages 35-37.

[4]  In fact, the unemployment rate typically continues to rise for about 4 to 6 months after the end of recessions.

[5]  The slow recovery of the unemployment rate in post-recession periods has been labeled “jobless recovery”, after the recent recessions in developed economies, which have not produced growth resumption coupled with significant decreases in the unemployment rate. The prociclicality of labor productivity suggests that all recoveries have a jobless period, and the persistence of this process distinguishes jobless recoveries from other recoveries.

[6]  Investment and exports are the demand components that typically drive growth upwards during the resumption of activity.


 



 

Fernando M. Gonçalves

Luka Barbosa

André Matcin


 

 



< Back