Brazil: Selected Issues Paper

International Monetary Fund. Western Hemisphere Dept.

doi:10.5089/9781475521320.002.A002

Jump to Content Jump to Main Navigation

Brazil: Selected Issues Paper

Author: International Monetary Fund. Western Hemisphere Dept.

Publication Date:: 12 May 2015

eISBN:: 9781475521320

Language:: English

Keywords:: ISCR; CR; minimum wage; Brazil; wage; bond yield; rate growth; rule in comparison; wage ratio; demand shock; inflation outcome; Wages; Employment; Income inequality; median wage

Download PDF (3.1 MB)

This Selected Issues paper examines infrastructure investment in Brazil. Brazil has inferior overall infrastructure quality relative to almost all its export competitors. Brazil’s infrastructure endowment ranks low by international standards, and its low quality affects productivity, market efficiency, and competitiveness. Areas in which Brazil’s competitiveness has lagged include, but are not limited to, education, innovation, governance, and justice. Brazil’s infrastructure gap has become a major obstacle to growth and filling this gap will entail increasing investment and also stepping up other reforms.

Abstract

Macroeconomic Implications of Minimum Wage Increases in Brazil¹

Over the past decade or so, real increases in the minimum wage have contributed to declining poverty and inequality in Brazil. This paper highlights the potential trade-off between economic performance and the reduction in inequality. While Brazil’s current minimum wage rule is contributing to a reduction in inequality, the empirical and theoretical results show that well-designed alternative minimum wage rules can help to improve competitiveness, job creation, and inflation outcomes. In this context, the current minimum wage rule should be re-evaluated against alternative rules and policy objectives. In addition to re-examining the current minimum wage rule, other policies could be considered to better target inequality, such as well-designed social policies and tax reform.

A. Introduction

1. Minimum wage increases have contributed to a decline in inequality in Brazil over the past decade or so. This paper discusses the macroeconomic implications of the Brazilian minimum wage rule, looking at effects on output, investment, employment, wages, prices, and inequality. The impact of minimum wage increases on macroeconomic aggregates, such as employment for example, is a controversial issue. The largest body of literature exists in the U.S. Two famous debates are Card and Krueger (1994 and 1998) and Neumark and Wascher (1997), and Dube et al. (2010), Allegretto et al. (2011) and Neumark and Wascher (2013). The authors argue for and against a negative employment effect of minimum wage increases arising from a well identified episode of wage increases in a specific region of the U.S., bringing to bear different approaches and different sources of microeconomic data, leading each time to new patterns being revealed in this economic relationship. In this light, while this paper goes a long way in applying and developing a coherent econometric and theoretical framework, it presents a first set of results, which should be tested with different econometric and theoretical models in the future.²

2. Whether the minimum wage is too low or too high matters for employment. If the minimum wage is low, then employers may abuse their bargaining strength in wage negotiations with low-skilled workers and pay less than the marginal product of labor. Employers will be willing to hire more easily at those rates, but the rents they receive tend to increase inequality—which could also rise if part of the rents are distributed to higher skilled workers. On the other hand, if minimum wages are very high, employment growth will be curtailed, other things constant. However, unskilled workers will make a better living, and in sectors with rents these can end up shared with poorer workers, which would reduce inequality. A minimum wage policy should strike a balance between providing appropriate remuneration and encouraging employment.

B. The Minimum Wage and Inequality in Brazil

3. The minimum wage in Brazil currently increases according to a pre-determined formula. Law 12.382 came in force on February 25, 2011, determining the annual minimum wage increase during the period 2012-15 to be equal to inflation over the previous year plus annual real GDP growth from two years before the current year. For example, the minimum wage for 2012 was set at annual inflation during 2011 plus annual real GDP growth during 2010. During 2015 a new bill will be sent to the National Congress, providing for the minimum wage policy for the period 2016-19.

4. During the past decade, wages and the minimum wage have increased above the rate of labor productivity while inequality, as measured by the Gini index, has declined. According to Maurizio (2014), increases in the minimum wage led to wage compression, which helped to reduce inequality. While she focuses exclusively on the effect of minimum wage increases, other factors such as economic growth, increasing terms of trade, and social policies, including Bolsa Família, were arguably equally important for the reduction in inequality.

5. In line with evidence from other countries, wage levels at the lower end of the wage distribution (closest to the minimum wage) have responded most strongly to increases in the minimum wage. The progress of Brazil in reducing inequality is encouraging; the reduction in inequality has been continuous and long-lasting. A comparison of inequality in Brazil with other countries shows that inequality is still an issue, and further improvement is needed (see Arnold and Jalles, 2014). Given that the minimum wage has already increased substantially, the question arises whether further increases in the minimum wage are the right tool to reduce inequality going forward, especially in an environment of low economy growth

6. The minimum wage in Brazil is below the average of major Latin American countries (LA6), and significantly below the OECD average. However, the minimum wage relative to per-capita GDP is very close to the OECD average.³

C. Is Brazil’s Minimum Wage too High?

To assess whether the minimum wage is relatively high or low in Brazil, two approaches are followed. First, Brazil’s minimum wage is benchmarked against other countries. Second, panel regressions are used to determine whether the increase in minimum wages affects employment.

7. Probably the most common concept used to determine whether the minimum wage is high or low is the “Kaitz index”, the ratio of the minimum wage relative to the average or the median wage. In general, because the median wage tends to be below the average wage, the ratio between minimum wage and median wage should be the preferred measure, because it measures to which extent the minimum wage puts pressure from below on the wage distribution. The closer the minimum wage is to the median wage, the bigger will be the knock-on effects of increases in the minimum wage (e.g. if lower skilled workers benefit from wage increases, higher skilled workers will also demand wage increases to maintain a wage differential between differently skilled labor).

Benchmarking

8. The trajectory of the Kaitz index for Brazil is rising over time. This is consistent with the policies in place, which ensure that over time the minimum wage increases faster than average labor productivity in the economy. While this rule does not ensure by itself a rising Kaitz ratio, it makes that very likely. As a starting point for benchmarking, it is reasonable to assume that every country in the world with an active minimum wage policy tries to strike a balance between the pros and cons of minimum wage increases. Given that a large sample of countries exists, countries, on average, should follow a minimum wage policy that finds a good balance between pros and cons. A comparison of Brazil’s Kaitz index with the average of a large set of countries shows that increases in the minimum wage led from a regime of low minimum wages to high minimum wages, relative to the average of other countries.

Panel Regressions

9. The impact of increases of the minimum wage on employment has been debated for decades.⁴ However, more recently, a consensus seems to be emerging, which follows the intuitive explanation provided above: Whether minimum wage increases have a negative employment effect depends on whether the minimum is binding, other things constant. The more the minimum wage is binding—measured for example by an increase in the Kaitz index—the more likely is a negative employment effect. The negative employment effect is more pronounced for the least-skilled group of workers—workers, who earn one or a fixed multiple of the minimum wage. To identify this effect, it is important to control for economic growth, which is a determinant of employment growth for any given level of wages. The most convincing cross-country evidence is based on a sample of 33 OECD countries for the period 1971–2009.⁵ Since this sample includes largely Non-Latin American countries, we run panel regressions with fixed affects on a panel of Latin American countries, including Brazil, Chile, Colombia, Mexico and Peru, for the period 2000–14. The econometric model takes the following form:

E_{i, t} = α + β Y_{i, t} + γ K a i t z_{i, t} + J_{i} + u_{i, t}

where E_i,t, Y_i,t and Kaitz_i,t denote the yearly percentage change in quarterly employment, real output and the Kaitz index. The Kaitz index is computed as the ratio of the real minimum wage to the real average wage, as time series for median wages of certain Latin American countries were not readily available. J_i controls for country fixed effects.⁶

10. The empirical findings suggest that increases in the minimum wage in excess of the average wage increases tend to reduce employment. Specifically, a ten percent year-on-year increase in minimum wages above average wage increases lead to an employment decline of about 0.6-0.85 percent, depending on whether fixed effects are included. The result is robust to different specifications. It also holds for Brazil in isolation, with the coefficient for the Kaitz index being in the same order of magnitude between -0.07 to -0.12 (Appendix A). This suggests that the decline in the unemployment rate observed in Brazil for the last decade has occurred despite the high minimum wage rate growth, and thanks to the country’s record of output growth during the sample period, which has been relatively high by historical standards. This is an important result: high GDP growth can help to offset the drag on employment growth from a rise in the Kaitz index. But the question this result poses is whether, in a lower GDP growth scenario, such as the one faced by Brazil now, employment growth may be strongly affected by the continuation of a policy that tends to increase the Kaitz index (by raising real minimum wages faster than average labor productivity).

Table. Panel Data Regressions

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Table. Panel Data Regressions

	Dependent Variable: Year-on -Year Percentage Change in Employment
	(1)	(2)
Year-on-Year Percentage Change in GDP	0.188***	0.164***
Year-on-Year Percentage Change in GDP	(0.059)	(0.056)
Year-on-Year Percentage Change in the Kaitz Index
	−0.084***	−0.062***
	(0.027)	(0.024)

Constant	2.243***	2.235***
	(0.318)	(0.312)
Country Fixed Effects	Yes	No
Obs	224	224
Periods (average)	55	55
Cross-sections	5	5
r2	0.081	0.108

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Table. Panel Data Regressions

	Dependent Variable: Year-on -Year Percentage Change in Employment
	(1)	(2)
Year-on-Year Percentage Change in GDP	0.188***	0.164***
Year-on-Year Percentage Change in GDP	(0.059)	(0.056)
Year-on-Year Percentage Change in the Kaitz Index
	−0.084***	−0.062***
	(0.027)	(0.024)

Constant	2.243***	2.235***
	(0.318)	(0.312)
Country Fixed Effects	Yes	No
Obs	224	224
Periods (average)	55	55
Cross-sections	5	5
r2	0.081	0.108

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

11. Low-skilled workers tend to be more affected by minimum wage increases than higher skilled workers. The model is estimated at the monthly frequency where GDP changes are replaced with changes in industrial production for Brazil (Appendix B). The results suggest that employment of younger workers is more affected by minimum wage increases than employment of older workers, with an elasticity of -0.3. This result reflects that wages of young workers are more likely to be in the lower half of the wage distribution and closer to the minimum wage and is broadly in line with the literature.⁷

D. Potential Macroeconomic Implications of the Current Minimum Wage Rule

12. Many empirical results suffer from a partial equilibrium critique in the sense that interactions with omitted variables are not captured. Research contributions on the impact of minimum wage increases often focus on one dimension, like the impact of minimum wages on the wage distribution or the impact of minimum wages on employment. Other dimensions like the impact on aggregate demand and consumption are often neglected. According to Maurizio (2014):

“The exercises that arise are of partial equilibrium nature and could be thought of as short-term. […] For example, it is not considered how changes in the minimum wage affect consumption (and, in turn, aggregate demand and employment), especially in those cases where the population directly affected by minimum wage has a high propensity to consume.”

13. To address such criticism, we construct a novel general equilibrium model. A natural starting point is a New-Keynesian General Equilibrium Model, which is the most widely used theoretical model in central banks around the world. The model choice is further motivated and supported by our empirical findings—the negative employment effect of changes in the Kaitz index, is consistent with New-Keynesian Models with an explicit trade-off between higher wages and less employment, other things constant. In the standard model there is only one type of consumer, accordingly we make a minimalistic change to the model and include a second type of consumer with lower productivity and higher propensity to consume, to differentiate between low skilled and high skilled workers. The final model ingredient is the Brazilian minimum wage rule. Low skilled workers earn the minimum wage, which changes in accordance with the minimum wage rule. The model is described in greater detail in the appendix.

14. The current minimum wage rule is assessed by comparing the impulse response functions following demand and supply shocks under two different scenarios:

Scenario 1 – Minimum Wage Rule: The wage of low skilled workers is determined by the current minimum wage formula:

$Δ M W_{t} % = Δ {I n f}_{t - 1} % + Δ {G D P}_{t - 2} %$
Scenario 2 – Standard Wage Rule: The wage of low skilled workers is determined by labor market tightness, e.g. wages compensate workers for the opportunity cost of not producing the home-produced good.

15. The results suggest that the minimum wage rule leads to higher wages and reduces inequality, but it reduces employment and output, and raises inflation relative to the standard wage rule. The results comparing impulse responses following a demand shock and a supply shock are displayed below. The bars in the figure reflect percentage gains from the minimum wage rule relative to the standard wage rule following a demand shock and a supply shock.⁸ Negative numbers reflect losses relative to the standard wage rule. The theoretical model confirms the econometric results—employment of low skilled and high skilled workers is lower under the current minimum wage rule. Specifically, the results show that:

Following a positive demand shock (higher demand, the blue bars), employment of low skilled workers is 1.3 percent lower under the current minimum wage rule in comparison to the standard rule.
Following a positive supply shock—with similar differential impact on low skilled wages—(higher TFP, the red bars) employment of low skilled workers is 0.8 percent lower under the current minimum wage rule.

16. A deeper analysis of the model shows that there are two channels at work that affect inflation, a cost channel and a demand channel. Higher wages increase the marginal cost of firms, but also raise incomes of low-skilled workers that benefit from wage increases, and therefore their demand for goods. Both channels support inflationary pressures, higher interest rates, and lower investment. This suggests that if minimum wages increase with a premium above inflation equal to real GDP growth, then employment, output and investment tend to be lower and inflation tends to be higher. On the other hand, strong minimum wage increases tend to reduce inequality more, posing a potential trade-off between macroeconomic performance and inequality.⁹

E. Policy Implications

17. The results highlight the potential trade-off between economic performance and the reduction in inequality. While Brazil’s current minimum wage rule is contributing to a reduction in inequality, empirical and theoretical results show that well-designed alternative minimum wage rules can improve competitiveness, job creation, and inflation outcomes. In this context, the current minimum wage rule should be re-evaluated against alternative rules and policy objectives.

18. In addition to re-examining the current minimum wage rule, other policies could also be considered to better target inequality. Neumark et al. (2006) found that in environment of low growth and declining terms of trade, minimum wage increases do not have the intended positive distributional effects, pointing towards the tension of higher wages for some and less employment for others. If inequality is a concern, other social policies could be used to more efficiently boost incomes at the lower end of the wage distribution. Moreover, inequality in Brazil also reflects a large disparity in earnings in the upper half of the distribution, evidenced by the large gap between the average and the median wage, suggesting that tax policy could also play a greater role in reducing inequality in the future.

Appendix—Regressions and General Equilibrium Model

A. Quarterly Time Series Regressions

Table. Brazil: Quarterly Time Series Regressions

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Table. Brazil: Quarterly Time Series Regressions

Dependent Variable:	Year-on-Year Percentage Change in Total Employment	Year-on-Year Percentage Change in Total Employment	Year-on-Year Percentage Change in Total Employment	Year-on-Year Percentage Change in Total Employment
	(1)	(2)	(3)	(4)
Dependent variable (t-1)			0.698***	0.689***
Dependent variable (t-1)			(0.078)	(0.079)
Year-on-Year Percentage Change in GDP	0.167	0.058	0.160*	0.138
Year-on-Year Percentage Change in GDP	(0.167)	(0.171)	(0.084)	(0.091)
Dummy for 2014 Months		−3.620***		−0.707
Dummy for 2014 Months		(1.887)		(1.017)
Year-on-Year Percentage Change in the Kaitz Index
	−0.096**	−0.115**	−0.071**	−0.076**
	(0.046)	(0.046)	(0.035)	(0.036)

Constant	3.037***	3.779***	0.653	0.846
Constant	(0.853)	(0.914)	(0.497)	(0.571)
Frequency	Quarterly	Quarterly	Quarterly	Quarterly
Periods	55	55	54	54
r2	0.128	0.195	0.673	0.677

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Table. Brazil: Quarterly Time Series Regressions

Dependent Variable:	Year-on-Year Percentage Change in Total Employment	Year-on-Year Percentage Change in Total Employment	Year-on-Year Percentage Change in Total Employment	Year-on-Year Percentage Change in Total Employment
	(1)	(2)	(3)	(4)
Dependent variable (t-1)			0.698***	0.689***
Dependent variable (t-1)			(0.078)	(0.079)
Year-on-Year Percentage Change in GDP	0.167	0.058	0.160*	0.138
Year-on-Year Percentage Change in GDP	(0.167)	(0.171)	(0.084)	(0.091)
Dummy for 2014 Months		−3.620***		−0.707
Dummy for 2014 Months		(1.887)		(1.017)
Year-on-Year Percentage Change in the Kaitz Index
	−0.096**	−0.115**	−0.071**	−0.076**
	(0.046)	(0.046)	(0.035)	(0.036)

Constant	3.037***	3.779***	0.653	0.846
Constant	(0.853)	(0.914)	(0.497)	(0.571)
Frequency	Quarterly	Quarterly	Quarterly	Quarterly
Periods	55	55	54	54
r2	0.128	0.195	0.673	0.677

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

The figure shows various single country regressions for Brazil. The fit of the regression improves and the Kaitz index coefficient (in absolute terms) becomes larger, if a dummy accounting for the break in participation rates in 2014 is included. The Kaitz index-employment elasticities are in the range of −0.07 to −0.12.

B. Monthly Time Series Regressions

Table. Brazil: Monthly Time Series Regressions

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Table. Brazil: Monthly Time Series Regressions

Dependent Variable:	Year-on-Year Percentage Change in Adult Employment	Year-on-Year Percentage Change in Adult Employment	Year-on-Year Percentage Change in Youth Employment	Year-on-Year Percentage Change in Youth Employment
	(1)	(2)	(3)	(4)
Dependent variable (t-1)		0.770***		0.621***
Dependent variable (t-1)		(0.058)		(0.056)
Dependent variable (t-12)		−0.127*		−0.290***
Dependent variable (t-12)		(0.697)		(0.063)
Year-on-Year Percentage Change in ICB-Br	0.393***	0.121*	0.451*	0.031
Year-on-Year Percentage Change in ICB-Br	(0.007)	(0.0537)	(0.260)	(0.164)
Dummy for 2014 Months	−3.127***	−0.604	−16.990***	−7.957***
Dummy for 2014 Months	(0.604)	(0.647)	(2.565)	(2.502)
Year-on-Year Percentage Change in the Kaitz Index	−0.0839*	−0.0562*	−0.607***	−0.296***
Year-on-Year Percentage Change in the Kaitz Index	(0.042)	(0.033)	(0.220)	(0.112)
Constant	−1.922	−0.402	−1.922	−0.402
Constant	(1.370)	(0.902)	(1.370)	(0.902)
Frequency	Monthly	Monthly	Monthly	Monthly
Periods	128	116	128	116
r2	0.246	0.678	0.276	0.749

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Table. Brazil: Monthly Time Series Regressions

Dependent Variable:	Year-on-Year Percentage Change in Adult Employment	Year-on-Year Percentage Change in Adult Employment	Year-on-Year Percentage Change in Youth Employment	Year-on-Year Percentage Change in Youth Employment
	(1)	(2)	(3)	(4)
Dependent variable (t-1)		0.770***		0.621***
Dependent variable (t-1)		(0.058)		(0.056)
Dependent variable (t-12)		−0.127*		−0.290***
Dependent variable (t-12)		(0.697)		(0.063)
Year-on-Year Percentage Change in ICB-Br	0.393***	0.121*	0.451*	0.031
Year-on-Year Percentage Change in ICB-Br	(0.007)	(0.0537)	(0.260)	(0.164)
Dummy for 2014 Months	−3.127***	−0.604	−16.990***	−7.957***
Dummy for 2014 Months	(0.604)	(0.647)	(2.565)	(2.502)
Year-on-Year Percentage Change in the Kaitz Index	−0.0839*	−0.0562*	−0.607***	−0.296***
Year-on-Year Percentage Change in the Kaitz Index	(0.042)	(0.033)	(0.220)	(0.112)
Constant	−1.922	−0.402	−1.922	−0.402
Constant	(1.370)	(0.902)	(1.370)	(0.902)
Frequency	Monthly	Monthly	Monthly	Monthly
Periods	128	116	128	116
r2	0.246	0.678	0.276	0.749

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

The figure shows various single country regressions for Brazil. The robustness of the results is improved by increasing the number of data points, using monthly data. The lag order of the dependent variable is select to assure stationarity and absence of serial correlation in residuals. The resulting Kaitz index-employment elasticities are -0.06 for adults and -0.3 for youth employment.

C. General Equilibrium Model

Households

There are two representative households, representing high skilled and low skilled workers. High skilled and low skilled workers maximize expected lifetime utility, deriving utility from consumption and home goods. The objective functions of high and low skilled workers are:

E_{0} Σ_{t = 0}^{\infty} β^{t} [u (c_{t}) + u (c_{t}^{H})] and E_{0} Σ_{s = 0}^{\infty} β^{t} [u (c_{t}^{M W}) + u (c_{t}^{H, M W})]

where β < 1 denotes the subjective discount factor, c_t the consumption good, and $c_{t}^{H}$ the home good. The superscript “MW” (like the minimum wage) denotes quantities and prices specific to low skilled workers.

The home good is produced by members of the households that are not participating (np_t) or unemployed (u_t), and spend time at home. Controlling for population growth, assuming a constant household size (s = np_t + u_t + e_t), it becomes clear that consumption of the home good depends on employment:

c_{t}^{H} = {n p}_{t} + u_{t} = s - e_{t} and c_{t}^{H, M W} = n p_{t}^{M W} + u_{t}^{M W} = s^{M W} - e_{t}^{M W}

High skilled and low skilled households differ in the budget constraint. With respect to expenses, high skilled households consume (c_t), and decide how much to invest in bonds (d_t) and capital $(k_{t}^{'})$ . They also pay lump sum taxes (T_t). Next period, they receive interest (R_t = 1 + i_t) on bonds, and the rental rate r_k,t on capital and sell the un-depreciated part of the capital stock at its market price Q_t. For each employed worker, the household receives a wage W_t, and it receives the profits of firms. P_t denotes the overall price level in the economy. The budget constraint of high skilled households is:

P_{t} c_{t} + d_{t} + Q_{t} k_{t}^{'} + T_{t} = (r_{k, t} + (1 - δ) Q_{t}) k_{t - 1}^{'} + R_{t - 1} d_{t - 1} + W_{t} e_{t} + P r o f_{t}

The low skilled households are hand-to-mouth consumers, and consequently have a higher propensity to consume. Each employed member consumes what she/he earns $c_{t}^{'} = W_{t}^{M W}$ . Aggregating over all employed low skilled household members, it follows that $c_{t}^{'} = \frac{P_{t} c_{t}^{M W}}{e_{t}^{M W}}$ , and the budget constraint of low skilled households is:

P_{t} c_{t}^{M W} = W_{t}^{M W} e_{t}^{M W}

Scenario 1 – Minimum Wage Rule: In the first scenario, the wage of low skilled workers follows the current minimum wage rule in Brazil, in which $π_{t} = \frac{P_{t}}{P_{t - 1}}$ denotes inflation.

$\frac{W_{t}^{M W}}{W_{t - 1}^{M W}} = g_{t}^{M W} = \frac{π_{t - 1} Y_{t - 1}^{d}}{Y_{t - 3}^{d}}$
Scenario 2 – Standard Wage Rule: In the second scenario, the wage of low skilled workers increases with market tightness, and compensates low skilled workers for the opportunity cost of not producing the home good. It is equal to the first order condition of low skilled households, with respect to the supplied number of household members, who are seeking employment.

$\frac{W_{t}^{M W}}{P_{t}} = \frac{u^{'} (c_{t}^{H, M W})}{u^{'} (c_{t}^{M W})}$

Firms and Labor Unions

There is a final good producer, a continuum of intermediate goods producers (with index i), an entrepreneur and a capital good producer. The final good producer maximizes the following objective:

P_{t} Y_{t}^{d} - \int P_{i, t} Y_{i, t} d i

subject to the following production function, where ε is the elasticity of substitution between different intermediate goods:

Y_{t}^{d} = {(\int Y_{i, t}^{\frac{ɛ - 1}{ɛ}} d i)}^{\frac{ɛ}{ɛ - 1}}

Intermediate goods producers are subject to Calvo pricing. In each period, only a fraction 1 – θ of producers can change their prices, while all other producers can only index their prices to past inflation. Indexation is controlled by a parameter τ ∈ [0,1], where τ = 0 implies no indexation and τ = 1 implies total indexation. The maximization problem of producer i, who chooses its price $P_{t}^{*}$ is:

E_{t} Σ_{s = 0}^{\infty} {(β θ)}^{s} (λ_{t + s} / λ_{t}) P_{t + s} [\frac{P_{t}^{*} a_{t + s}}{P_{t + s}} - \frac{P_{M, t + s}}{P_{t + s}}] Y_{i, t + s}

subject to the demand function for its product, which results from the optimization problem of the final good producer:

Y_{i, t + s} = {(\frac{P_{t}^{*} a_{t + s}}{P_{t + s}})}^{- ɛ} Y_{t + s}^{d}

where the factor a_t+s(τ) results from indexation, and λ_t is the shadow value of an additional unit of income. The resulting first order condition is often referred to as the New-Keynesian Phillips Curve and determines prices as a mark-up over marginal costs.

Entrepreneurs hire workers and rent capital, and supply the produced output to intermediate goods producers. The objective function of the entrepreneur is:

P_{M, t} Y_{t}^{s} - (W_{t} e_{t} + W_{t}^{M W} e_{t}^{M W} + r_{k, t} k_{t})

The production function is of the standard Cobb Douglas form, with constant returns to scale (α + β_p + γ = 1):

Y_{t}^{s} = z_{t} k_{t}^{α} e_{t}^{β_{p}} {(e_{t}^{M W})}^{γ}

Lower skilled workers have lower productivity (γ = 0.1 < β_p = 0.68).

Capital goods producers supply capital to high skilled households, maximizing the following objective:

Q_{t} i_{t} - (1 - S (\frac{i_{t}}{i_{t - 1}})) i_{t}

The inclusion of capital adjustment costs represented by the function $S (\frac{i_{t}}{i_{t - 1}})$ makes impulse responses of investment more persistent.

There are labor unions, which negotiate wages on behalf of high skilled workers, charging the entrepreneurs a markup over marginal costs. Since minimum wage increases support wage negotiations of labor unions, the mark-up is affected by the minimum wage growth rate:

W_{t} = (1 - s) \frac{u^{'} (c_{t}^{H})}{u^{'} (c_{t})} P_{t} + s g_{t}^{M W} W_{t - 1}

In the first scenario s = 0; in the second scenario s = 0.5. It turns out that magnitude of only affects the magnitude of elasticities, but not the qualitative conclusions. All qualitative results still hold if s = 0 in both scenarios.¹

Government and Monetary Authority

The government issues bonds and finances government purchases g_t with lump sum taxes.²

g_{t} + R_{t - 1} d_{t - 1} = d_{t} + T_{t}

The monetary authority sets interest rates according to a Taylor rule:

\frac{R_{t}}{R} = {(\frac{R_{t - 1}}{R})}^{γ_{R}} {({(\frac{π_{t}}{π})}^{γ π} {(\frac{Y_{t}^{d}}{Y^{d}})}^{γ_{Y}})}^{1 - γ_{R}} \in_{M, t}

Market Clearing, Inequality and Exogenous Processes

The market clearing condition demonstrates how minimum wage increases affect aggregate demand through consumption:

\begin{matrix} \begin{matrix} c_{t} + c_{t}^{M W} + g_{t} + i_{t} = Y_{t}^{d} \\ k_{t} = k_{t - 1}^{'} \end{matrix} \\ k_{t}^{'} = (1 - δ) k_{t - 1}^{'} + (1 - S (\frac{i_{t}}{i_{t - 1}})) i_{t} \end{matrix}

Inequality is measured by the difference in consumption:

i n e q u a l i t y_{t} = c_{t} - c_{t}^{M W}

Demand and supply shocks result from the processes of government expenditure g_t and TFP z_t:

ln(g_t) = (1 – ρ_g)ln(g) + ρ_g ln(g_t–1) + ε_g,t and ln(z_t) = ρ_z ln(z_t–1) + ε_z,t

References

Arnold, Jens M., and João Jalles, 2014, “Dividing the Pie in Brazil: Income Distribution, Social Policies, and the New Middle Class,” OECD Working Paper No. 1105 (Paris: Organisation for Economic Co-operation and Development).
- Search Google Scholar
- Export Citation
Allegretto, Sylvia A., Arindrajit Dube, and Michael Reich, 2011, “Do Minimum Wages Really Reduce Teen Employment? Accounting for Heterogeneity and Selectivity in State Panel Data,” Industrial Relations, Vol. 50, No. 2 (April), pp. 205–40.
- Search Google Scholar
- Export Citation
Card, David, and Alan B. Krueger, 1994, “Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania,” American Economic Review, Vol. 84(4) (September), pp. 772–93.
- Search Google Scholar
- Export Citation
Card, David, and Alan B. Krueger, 1998, “A Re-Analysis of the Effect of the New Jersey Minimum Wage Increase in the Fast-Food Industry with Representative Payroll Data,” NBER Working Paper No. 6386 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Search Google Scholar
- Export Citation
Dolton, Peter, and Chiara Rosazza Bondibene, 2012, “The International Experience of Minimum Wages in an Economic Downturn,” Economic Policy, Vol. 27, Issue 69, pp. 99–142.
- Search Google Scholar
- Export Citation
Dube, Arindrajit, T. William Lester, and Michael Reich, 2010, “Minimum Wage Effects Across State Borders: Estimates Using Contiguous Counties,” Review of Economics and Statistics, Vol. 92, No. 4 (November), pp. 945–64.
- Search Google Scholar
- Export Citation
Maurizio, Roxana, 2014, “El Impacto Distributive del Salario Minimo en la Argentina, el Brasil, Chile y el Uruguay,” CEPAL, Serie Politicas Sociales, No. 194 (LC/L 3825).
- Search Google Scholar
- Export Citation
Neumark, David and William Wascher, 1997, “The New Jersey-Pennsylvania Minimum Wage Experiment: A Re-Evaluation Using Payroll Records,” mimeo, Michigan State University.
- Search Google Scholar
- Export Citation
Neumark, David, Wendy Cunningham, and Lucas Siga, 2006, “The Effects of the Minimum Wage in Brazil on the Distribution of Family Incomes: 1996–2001,” Journal of Development Economics, Issue 80, pp. 136–159.
- Search Google Scholar
- Export Citation
Neumark, David, and William Wascher, 2007, “Minimum Wages and Employment,” IZA, Discussion Paper No. 2570.
- Search Google Scholar
- Export Citation
Neumark, David, Ian J.M. Salas, and William Wascher, 2013, “Revisiting the Minimum Wage-Employment Debate: Throwing out the Baby with the Bathwater?” NBER Working Paper No. 18681, (Cambridge, Massachusetts: National Bureau of Economic Research).
- Search Google Scholar
- Export Citation

Prepared by Fabian Lipinsky and Carlos Goes.

For example, similar to Dolton and Bondibene (2012), further control variables could be included. Another robustness check would be to test for level effects.

In fact, the LAC6 average is above the OECD average, owing to high Minimum Wage-GDP ratios for Argentina, Colombia and Peru.

For a literature review of the impact of changes in the minimum wage on employment, including in Brazil, see Neumark and Wascher (2007).

Dolten and Bondibene 2012.

This specification includes a version of the panel in which the variables would be measured in deviations from the steady state, denoted by a star: $E_{i, t} - E_{t}^{*} = α + β (Y_{i, t} - Y_{t}^{*}) + γ (K a i t z_{i, t} - K a i t z_{i}^{*}) + u_{i, t}$ .

Dolten and Bondibene (2012) reports elasticities of about -0.05 for adult and -0.2 for youth employment.

Specifically, the figure displays the percentage difference in cumulative change of each variable over 10 quarters following each shock.

While the magnitude of the elasticities depends on the calibration of the general equilibrium model, the qualitative direction of the results is very robust and insensitive to changes in the calibration.

The other parameters are close to standard values in the DSGE literature.

An interesting extension could be to replace lump-sum taxes with taxes on labor (that are progressive with earnings) capital and consumption.

Other Publishers

Save
Cite
Email this content

Share Link

Copy this link, or click below to email it to a friend
Email this content
or copy the link directly:

https://www.elibrary.imf.org/view/journals/002/2015/122/article-A002-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Brazil: Selected Issues Paper

Author: International Monetary Fund. Western Hemisphere Dept.

Volume/Issue:: Volume 2015: Issue 122

Publisher:: International Monetary Fund

ISBN:: 9781475521320

ISSN:: 1934-7685

Pages:: 102

DOI:: https://doi.org/10.5089/9781475521320.002

Keywords
Macroeconomic Implications of Minimum Wage Increases in Brazil1
References

View raw image

Brazil: Gini Index, 2004-2013

(0 = minimum inequality; 1 = maximum inequality)

View raw image

Brazil: Wage and Labor Productivity Growth

(Index; 2004=100)

View raw image

Brazil: Real Change in Wages, 2004-2013

(In cumulative real percentage change, by decile)

View raw image

Selected Countries: Hourly Minimum Wage, 2014

(In PPP dollars)

View raw image

Selected Countries: Minimum Wage, 2014

(In percent of GDP per capita)

View raw image

Selected Countries: Minimum Wage to Average Wage Ratio

(Ratios, per year)

View raw image

Model Results: Difference of Minimum Wage Rule vs. Standard Rule

(In percent)

View raw image

Selected Countries: Median Wage to Average Wage Ratios

(Ratio, per year)

Table. Brazil: Monthly Time Series Regressions

Dependent Variable:	Year-on-Year Percentage Change in Adult Employment	Year-on-Year Percentage Change in Adult Employment	Year-on-Year Percentage Change in Youth Employment	Year-on-Year Percentage Change in Youth Employment
	(1)	(2)	(3)	(4)
Dependent variable (t-1)		0.770***		0.621***
Dependent variable (t-1)		(0.058)		(0.056)
Dependent variable (t-12)		−0.127*		−0.290***
Dependent variable (t-12)		(0.697)		(0.063)
Year-on-Year Percentage Change in ICB-Br	0.393***	0.121*	0.451*	0.031
Year-on-Year Percentage Change in ICB-Br	(0.007)	(0.0537)	(0.260)	(0.164)
Dummy for 2014 Months	−3.127***	−0.604	−16.990***	−7.957***
Dummy for 2014 Months	(0.604)	(0.647)	(2.565)	(2.502)
Year-on-Year Percentage Change in the Kaitz Index	−0.0839*	−0.0562*	−0.607***	−0.296***
Year-on-Year Percentage Change in the Kaitz Index	(0.042)	(0.033)	(0.220)	(0.112)
Constant	−1.922	−0.402	−1.922	−0.402
Constant	(1.370)	(0.902)	(1.370)	(0.902)
Frequency	Monthly	Monthly	Monthly	Monthly
Periods	128	116	128	116
r2	0.246	0.678	0.276	0.749

Note: Robust standard errors in parentheses.*** p<0.01, ** p<0.05, * p<0.1

Topics

Countries

Series

About

Resources

Abstract

Macroeconomic Implications of Minimum Wage Increases in Brazil1

A. Introduction

B. The Minimum Wage and Inequality in Brazil

C. Is Brazil’s Minimum Wage too High?

Benchmarking

Panel Regressions

D. Potential Macroeconomic Implications of the Current Minimum Wage Rule

E. Policy Implications

Appendix—Regressions and General Equilibrium Model

A. Quarterly Time Series Regressions

B. Monthly Time Series Regressions

C. General Equilibrium Model

Households

Firms and Labor Unions

Government and Monetary Authority

Market Clearing, Inequality and Exogenous Processes

References

Same Series

Other IMF Content

Other Publishers

Asian Development Bank

Inter-American Development Bank

International Labour Organization

The World Bank

Share Link

Table of Contents

Headings

Figures

Metrics

Macroeconomic Implications of Minimum Wage Increases in Brazil¹