Labor Productivity in Brazil. Fernando de Holanda Barbosa Filho

Transcription

1 Labor Productivity in Brazil Fernando de Holanda Barbosa Filho

2 Labor Productivity Evolution Labor productivity growth is steady but modest over time. 150,0 140,0 130,0 120,0 110,0 100,0 90,0

3 Introduction However, labor productivity is low in Brazil. Low labor productivity is not a composition problem, but a level problem Composition effect is small. Why low labor productivity? Low productivity sectors. Low productivity services. Importance of traditional services and low productivity modern services. Long tail of low productivity firms in all sectors

4 Services Productivity 110,000 90,000 70,000 50,000 30,000 10, Modernos Tradicionais Total Modern Traditional Total Source: FGV/IBRE

5 Labor Productivity Comparison (US$PPP) Agropecuary Industry Services Total Brazil USA Ireland Australia France Japan Great-Britain South Korea Mexico China India Average SEA USA/Brasil 13,9 5,7 5,4 6,1 Average SEA/Brasil 5,3 2,7 3,0 3,2 Fonte: Veloso, Matos, Ferreira e Coelho (2016)

6 Labor Productivity Long tail of low productivity firms in all sectors. Firms in Brazil do not grow overtime. High dispersion over firms productivity. No learning or selection process?

7 Firms Labor Productivity Distribution Brazil and Selected Countries (WBES)

8 Firms Labor Productivity Distribution Brazil and Selected Countries (WBES) Table 2: Labor Productivity per country (LN; US$ PPP de 2009) País Mean Median Variation coefficient BRA 9,8 10,2 2,3 CHL 10,8 10,9 1,2 CHN 10,9 10,8 1,3 COL 10,7 10,8 1,2 MEX 9,9 9,9 1,4 PER 10,4 10,4 1,2 RUS 10,7 10,7 1,4 Total 10,6 10,7 1,6

9 Firms Labor Productivity Distribution Textile

10 Firms Labor productivity Distribution Retail and Wholesale Trade

11 Firms Labor Productivity Distribution Hotels and Restaurant

12 Firms Labor Productivity Distribution Counter factual

13 We interpret this fact as a lack of selection, whereby producers with little growth potential survive because innovative entrepreneurs do not expand enough to force them out of the market. Firmas Dynamics Akcigit, Alp e Peters (2016): Firm dynamics in poor countries show striking differences to those of rich countries. While few firms indeed experience growth as they age, most firms are simply stagnant in that they neither exit nor expand.

14 Firmas Dynamics

15 Firmas Dynamics Firms do not grow overtime. Neither in revenues nor in employees. The same happens with productivity over time. It seems we do not have a selection process.

16 Firms Dynamics over time Number of Employees x Age > BRA CHL CHN COL MEX PER RUS Fonte: Elaboração própria com dados da WBES.

17 Firms Dynamics over time LN (Revenue) X Age > BRA CHL CHN COL MEX PER RUS Fonte: Elaboração própria com dados da WBES.

18 Firms Dynamics over time LN (productivity) X Age > BRA CHL CHN COL MEX PER RUS Fonte: Elaboração própria com dados da WBES.

19 Firmas Dynamics

20 Firmas Dynamics An important next step in this research agenda is to incorporate these findings into the study of industrial policies in developing countries. For instance, many regulations that support and facilitate the survival of small firms might have undesired consequences once the heterogeneity in firms types are taken into account. The quantitative implications of industrial policies for firm selection and firm dynamics are firstorder issues that await future research.

21 Discussion Brazil has several programs that protect small firms. The programs have major tax incentives for small firms. Simplified tax forms (Simples) benefit firms that report revenues above US$1 million (R$3,6 million). These firms are not small. There is also the Lucro Presumido that has a lower tax rate for firms with revenue below US$26 million (R$78 million). The tax incentives lead firms to reduce their size artificially to benefit. This may hurt productivity.

22 Counter Factual Exercises Labor productivity distribution in Brazil has a heavy, low productivity tail. The counter factual exercises aim to reduce its lower tail and project a cuonter factual productivity. First: Level X Composition exercise Second: Ad hoc lower productivity tail reduction Third: Clusters Analysis

23 Level X Composition Exercise Level and composition effect. Level effect explains most of the difference Productivity Differential Counterfactual Original (Composition) Counterfactual (level) CHL 93% 92% -3% CHN 103% 90% 13% COL 89% 71% -10% MEX 3% 11% 9% PER 59% 67% 10% RUS 84% 81% -9%

24 Labor Productivity Comparison Composition Effect Brazil Counterfactual (USA) Counterfactual (Mexico) Contrafactual (South Korea) Productivity Counterfactual Ratio 1,68 1,04 1,62 Observed ratio 6,10 1,73 3,59 Fonte: Veloso, Matos, Ferreira e Coelho (2016)

25 Labor Productivity Comparison Level Effect Brazil Counterfactual (USA) Counterfactual (Mexico) Counterfactual (South Korea) Productivity Counterfactual Ratio 5,30 1,83 2,93 Observed ratio 6,10 1,73 3,59 Fonte: Veloso, Matos, Ferreira e Coelho (2016)

26 Lower tail Reduction Counterfactual Brazilian Productivity Gains (%) Min Mena Median mean Median BRA 1,9 9,8 10,2 CHL 5 10,1 10, CHN 6,4 10,4 10,4 56,6 22,4 COL 7,2 10,5 10,5 71,6 28,8 MEX 4,3 9,9 10,2 4,8 0 PER 6,7 10,4 10,4 59,1 22,4 RUS 3 9,8 10,2 0 0 Source: Author with WBES data.

27 Clusters Analysis Variável de separação Group 1 Group 2 Group 3 Mean Counter Factual Difference (%) LN(revenue) 7,5 10,5 12,3 9,8 10,9 105 LN(Productivity) 6,1 9,8 11,9 9,8 10,7 88 LN (productivity) X LN (Reveneu) Source: Author with WBES data. Productivity by Group 6,2 10,1 12,2 9,8 10,7 90

28 Conclusion Labor productivity grows overtime at a lower rate. Labor productivity level is low in Brazil. Labor productivity is low due to sector low productivity. It is a level effect not a composition effect. Brazil has a long tail of low productivity firms. Lack of selection. Firms do not grow overtime.