Business Letter

scm41

ContentServer1.pdf

Home >Business & Finance homework help >Management homework help >Business Letter

DO MINIMUM WAGE INCREASES MATTER TO FIRM PROFITABILITY? THE CASE

OF VIETNAM

NGUYEN VIET CUONG*

National Economics University, Hanoi, Vietnam

Abstract: This paper measures the impact of a minimum wage increase in 2005 on profitability of private firms in Vietnam using a difference-in-differences with propensity score matching method. Data used for this analysis are from Vietnam Enterprise Censuses in 2004 and 2006. It is found that the impact estimate of the minimum wage increase in 2005 from 290 to 350 thousand VND on firms’ profit margins is very small and negative and not statistically significant. Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: minimum wages; firm profitability; difference-in-differences; propensity score matching; Vietnam JEL Classification: J31; J25; J21; P42

1 INTRODUCTION

Minimum wages are the lowest hourly, daily or monthly wage that employers are required to pay to employees. Increasing minimum wages often leads to controversial impacts. Possible positive effects of minimum wages are protection of low income laborers, increases in work incentives and productivity, reduction of people covered in subsidy programs, increases in consumption, aggregate demand and generation of multiplier effects (Freeman, 1994; Gunderson, 2005). Because firms can respond to an increase in labor cost by reducing labor demand or increasing the output prices, negative impacts of minimum wage increases can be an increase in unemployment and prices (Hamermesh, 1986; Brown, 1999). The size as well as the sign of the impact of minimum wage increases on employment

and prices is not consistent in empirical studies. For example, negative effects of the minimum wage on employment are found in studies both developed countries (Neumark and Wascher, 2002, 2003; Campolieti et al., 2005) and developing countries (Rama, 2001; Harrison and Scorse, 2005). On the contrary, positive effects on employment of minimum wages are found in studies such as Card and Krueger (1994, 2000), Dickens et al. (1999)

*Correspondence to: Nguyen Viet Cuong, National Economics University, Tran Dai Nghia street, Hanoi, Vietnam. E-mail: [email protected]

Journal of International Development J. Int. Dev. 29, 790–804 (2017) Published online 17 May 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/jid.2920

and Montenegro and Pagés (2004).1 Similarly, strong effects of minimum wages on inflation are found in Card and Krueger (1995) and Macdonald and Aaronson (2000) but not in Frye and Gordon (1981), Katz and Krueger (1992) and Card and Krueger (1995). Minimum wages can affect profitability of firms in two ways. The profit margins of

firms are not affected substantially if firms can pass on higher production cost due to increased minimum wages to consumers or the firms can reduce the production cost by employing fewer workers. On the contrary, the profit margins of the firms will decrease if the higher wage costs are not fully passed or the firms do not reduce their employment (Draca, et al. 2011). The effect of minimum wages on firm profitability is a priori unknown. Although there are a large number of empirical studies on impacts of minimum wages on employment and prices, there are only a few empirical studies on the relationship between minimum wages and firm profitability. Recently, Draca et al. (2011) showed that the minimum wages reduced firm profitability significantly in UK. In Vietnam, there have been several times of increasing the minimum wage since the

year 1993. The real minimum wage increased by around 118 per cent during the period 1994–2009. Increasing minimum wages often leads to debates about impacts of the minimum wage increases. The government states that minimum wages are constructed with consultation from enterprises. Thus minimum wages would have small effects on production, business and employment (Duy, 2009). On the contrary, there are critics that increased minimum wages can increase production costs and lead to burdens to enterprises (Thai, 2009). However, there are no quantitative studies on the impact of minimum wage increases on firm profitability in Vietnam. The main objective of this paper is to measure the impact of minimum wage increases on

firm profitability in Vietnam. The method of impact measurement used in this paper is difference-in-differences with propensity score matching, and the data are from Enterprise Censuses (EC) of Vietnam in years 2002, 2003, 2004, 2005 and 2006. These censuses were conducted by General Statistics Office of Vietnam. The censuses covered all State enterprises, collectives, private and foreign enterprises throughout the country. The number of observations in the 2002, 2003, 2004, 2005 and 2006 ECs is 62 213, 71 751, 91 755, 113 352 and 131 975, respectively. It is interesting that we are able to construct a panel data set of enterprises through these ECs. The EC contains data on the main production and business characteristics of enterprises such as labor, labor cost, investment capital, assets, revenues and profits, taxes and other contributions to State and so on. The remainder of this paper is organised as follows. The second section introduces the

minimum wages in Vietnam. The third section describes the methodology of impact evaluation. The fourth section presents impacts of the minimum wage increase on firm profitability. Finally, the fifth section concludes.

2 MINIMUM WAGE AND ENTERPRISES IN VIETNAM

According to the Labor Law of Vietnam, the minimum wage is set up to cover ‘the basis of the cost of living of an employee who is employed in the most basic job with normal working conditions’. Vietnam has only minimum monthly wage, not minimum daily or hourly wage. The Labor Law of Vietnam also regulates that the government must adjust the minimum wage when ‘the price index increases, resulting in the reduction of the real wages of employees’.

1Neumark and Wascher (2007) present detailed review of studies on the minimum wage and employment.

Impact of Minimum Wages on Firms 791

In addition, minimum wage adjustments are also based on payment capacity of the State budget, since there are a large proportion of employees in the State sector. Since the year 1993, there were nine adjustments of the minimum monthly wage in

Vietnam. The time and the national minimum wages (both nominal and real) after the adjustments are presented in Figure 1. In this paper, we will examine the impact of the increases in the monthly minimum wage

from 210 to 290 thousand VND in 1 January 2003 and from 290 to 350 thousand VND in 1 October 2005 because the data available at the time of writing the paper are Annual Enterprise Censuses of Vietnam from 2002 to 2006. Because of firm-level data, there are no data on wages of individual laborers, thus no data on the number of laborers paid below minimum wages. Instead, we have data on average wages of firms’ laborers. Firms that have the average wage of laborers below the new minimum wage will be affected by the minimum wage increase. In this paper, we assume that firms with the average wage of laborers below the new minimum wages can be affected by the minimum wage increases, and these firms are regarded as a treatment group.

Source: Author’s preparation using data on minimum wages and CPI.

Figure 1. Minimum monthly wage in Vietnam (thousand VND). [Colour figure can be viewed at wileyonlinelibrary.com]

Average month wages of workers (thousand VND)

Ratio of the current minimum wage to the average wage

Note: Wages are in the price of September 2005. Source: Estimation from ECs 2002, 2003, 2004, 2005 and 2006.

Figure 2. Ratio of the current minimum wage to the average wage during 2002–2006. [Colour figure can be viewed at wileyonlinelibrary.com]

792 N. V. Cuong

http://wileyonlinelibrary.com

The average wage increased by around 20 per cent annually (Figure 2). The growth rate of the average wage is higher than that growth rate of the minimum wages. The ratio of the minimum wage to the average wage increased from 0.29 in 2002 to 0.37 in 2003 (Figure 2). This ratio decreased to 0.29 in 2005 but increased to 0.32 in 2006 when there was an increase in the minimum wages in 2006 October. In 2002, there were around 12.9 per cent of firms having the average wage below the

2003 minimum wages. These firms could be affected by the minimum wage increase in 1 January 2003. In 2004, around 9.7 per cent of private enterprises had the average wage below 350 thousand VND, and these firms could be affected by the minimum wage increase in 1 October 2005. A problem in measuring the impact of minimum wage increases is how to define a

control group, which is not affected by minimum wage increases. A large number of state enterprises construct their salary scale for laborers according to minimum wages. It means that as minimum wages increase laborers in State enterprises who have wages above the new minimum wages might also receive higher wages or higher social insurances. For foreign firms, there are different minimum wages, which are higher than the national minimum wages. Thus, in this study, the control group does not include the state firms as well as the foreign firms. Private firms, which have the average wage of laborers above the new minimum wages

can be affected by the minimum wage increase because these firms can have laborers paid below the minimum wages. Control groups, which have the average wage paid to workers above the minimum wage can also include workers below the minimum wages. As a result, firms in these control groups can be affected by the minimum wage, and the impact estimate of the minimum wage can be biased towards zero. In this study, we will use different thresholds to define the control groups to examine the sensitivity of impact estimates of the minimum wage increase to the definition of control groups. More specifically, control groups include private firms that have the average monthly wages of laborers higher than different thresholds: 290 and 600 thousand VND for evaluation of the minimum wage increase in 2003. For evaluation of the minimum wage increase in 2005, control groups include private firms having the average monthly wages of laborers higher than thresholds of 350 and 600 thousand VND.2 It is expected that firms with very high average labor wages tend to have only a few laborers with wages below the new minimum wages; thus, these firms are less likely to be affected by the minimum wage increases. Tables 1 and 2 present the average wages of the treatment and control before and after

the minimum wage adjustment in 2003 and 2005, respectively. It shows that the treatment groups experienced a substantially higher growth rate of wages than control groups. The higher growth rate of the average wage of the treatment group might be because of the minimum wage increase. Profit of the treatment group might be also affected by the minimum wage increase. This issue will be examined in the next sections.

3 IMPACT EVALUATION METHOD

To measure the impact of the minimum wage increase on the firm profitability, we used the methodology of difference-in-indifferences with propensity score matching. Let D be a

2In this study, we also use the higher thresholds including 800, 1000 and 1200 thousand VND. The results from control groups using these thresholds are very similar to results from control groups using others thresholds. Thus, we do not present the findings using thresholds of 800, 1000 and 1200 thousand VND in this paper.

Impact of Minimum Wages on Firms 793

binary variable, which is equal to 1 if a firm has labor wages below the minimum wage, and 0 otherwise. Furthermore, denote Y as the variable of interest, with Yi = Yi1 if firm i has the average wages of workers below the minimum wage and Yi = Yi0 if the same firm i had not had the average wage below the minimum wage. The outcome of interest in this paper is profit margin, which is equal to the ratio of profits to sales. The impact of the minimum wage increase on firm i is then measured by

Δi ¼ Yi1 � Yi0: (1) The most popular parameter in impact evaluation is Average Treatment Effect on the

Treated, which is defined by the following (Heckman et al., 1999):

ATT ¼ E Y1 � Y0 D ¼ 1j Þ ¼ E Y1 D ¼ 1j Þ � E Y0 D ¼ 1j Þ:ððð (2) ATT measures the average effect of the minimum wage increase on firms with the

average wage below the minimum wage. Estimation of ATT is not straightforward, because E(Y0|D = 1) is unobservable. E(Y0|D = 1)

is the counterfactual, which is the expected profit margins of treatment firms if these firm had the average wage above the minimum wage. We use a propensity score matching method to construct a comparison group, which can mimic the treatment group in the absence of the minimum wage increase (Rosenbaum and Rubin, 1983). We start by estimating the probability of being a firm having the average wage for laborers below the minimum wage using a probit model (this is called propensity scores), P(Dit = 1)= F(Xit� 1), where X is a

Table 1. Average monthly wages of treatment and control groups in 2002 and 2003

Treatment group Control groups

Firms with average monthly wage below 290 thousand VND

Firms with average monthly wage above 290 thousand VND

Firms with average monthly wage above 600 thousand VND

Wages in 2002 183.6 1093.3 1253.6 (1.1) (5.1) (6.1)

Wages in 2003 712.7 1250.6 1346.6 (11.2) (6.5) (7.7)

Standard errors in brackets. All variables are in the price of September 2005. Source: Estimation from panel data of ECs 2002 and 2003.

Table 2. Average monthly wages of treatment and control groups in 2004 and 2006

Treatment group Control groups

Firms with average monthly wage below 350 thousand VND

Firms with average monthly wage above 350 thousand VND

Firms with average monthly wage above 600 thousand VND

Wages in 2002 216.1 1439.8 1559.2 [1.6] [7.1] [7.8]

Wages in 2003 737.8 1561.9 1613.0 [22.5] [8.1] [8.9]

Standard errors in brackets. All variables are in the price of September 2005. Source: Estimation from panel data of ECs 2004 and 2006.

794 N. V. Cuong

vector of observed variables before the minimum wage increases. The matching methodology matches each firm having the average wage for laborers below the minimum wage to ‘comparable’ firms having the average wage for laborers above the minimum wage based on the closeness of the predicted propensity scores. The matching estimator is defined as follows:

AT̂ T ¼ X

i2Treatment yi �

X j2Control

G p̂i; p̂j � �

" # ; (3)

where p is predicted propensity scores and G(.) gives the weights on control firm j in forming a comparison with treated firm i. The function G(.) differs for the different matching estimators proposed in the literature. Because we have panel data on enterprises, we can estimate the impact of the minimum

wage increase by the method of difference-in-differences with matching. The difference-in differences method can eliminate differences in outcomes between the treatment and control groups due to unobserved time-invariant effects. The main identification of the difference-in-differences is the parallel growth of outcomes between the treatment and control group in the absence of the treatment. Let Δy be the differences between the variable of interest before and after the minimum wage increase. Then the difference-in- differences estimator is given by the following:

AT̂ T ¼ X

i2Treatment Δyi �

X j2Control

G p̂i; p̂j � �

Δyj

" # : (4)

To estimate the impact of the minimum wage increase in 1 January 2003, the 2002 EC and the 2003 EC are used as the before and after the minimum wage increase, respectively. Regarding the minimum wage increase in 1 October 2005, the 2004 EC and the 2006 EC are used as the before and after the minimum wage increase, respectively. We use different matching estimators including nearest-neighbors and kernel matching

to examine the sensitivity the impact estimates. Standard errors are calculated using bootstrap techniques, which are widely used in empirical studies.3

4 IMPACT ESTIMATION RESULTS

4.1 Performance of Matching

The first step in measuring impact is to predict the propensity score, which is the probability that a firm had the average wages of laborers below the new minimum wages. Because the dependent variable is binary, we used a probit regression. Control variables should affect both the firm profitability and the average wage of firms’ laborers (Dehejia and Wahba, 1998; Smith and Todd, 2005). The control variables should be exogenous to the treatment variable; thus, these variables were measured in 2002 for the evaluation of the 2003 minimum wage increase and measured in 2004 for the evaluation of the 2005 minimum wage increase.4

3According to Abadie and Imbens (2006), bootstrap can produce invalid standard errors for the nearest neighbor matching estimator. However, there is no evidence against standard errors of kernel matching estimators using bootstrap. 4Summary statistics of the explanatory variables are not presented in this paper but can be provided on request.

Impact of Minimum Wages on Firms 795

Table A.1 in Appendix presents the probit regressions of the probability that a firm had the average wages of laborers below 290 thousand VND in 2002, and the probability that a firm had the average wages of laborers below 350 thousand VND in 2004. To examine the common support, we present Figure 3 of the propensity scores. The bars above the horizontal line represent the density distribution of the propensity score of firms with the average wages of laborers below the minimum wage, whereas the bars below the horizontal line represent the density distribution of the propensity score of firms with the average wages of laborers above the minimum wage (control Group 1). The figure shows that the common support is large. This means that for each treated firm, we will be able to find non-treated firms with similar propensity scores. It should be noted that the main aim of the predicted propensity score is to overcome the

multidimensionality problem of matching by covariates. The quality of a constructed comparison group should be assessed by testing whether the distribution of the covariates is similar between the comparison and treatment groups, given the predicted propensity score. We test the equality of means of covariates between treatment and comparison firms using t-tests. The testing results show that for around 90% of the number of covariates, we cannot reject the equality of their means between treatment and comparison groups.5

4.2 Impact Estimates

Table 3 presents impact estimates of the minimum wage increase in 2003 on profit margin using kernel neighbor matching with bandwidth of 0.05.6 It shows that before the minimum wage increase, firms having the average wage of workers below the minimum wage have slightly higher profit margin than firms having the average wage above the minimum wage. In 2002, the profit margin of the firms with the average monthly wage below 290 thousand VND was 2.1 per cent. The matched firms, which had the average monthly wage above 290 thousand VND had the profit margin of 2.0 per cent. In 2003, after the minimum wage increase, the control group still had a higher profit margin than the treatment group. However, these differences are small and not statistically significant. The impact of the minimum wage increase on the profit margin is computed by the overtime difference in differences in the

5Because of the limited length of the paper, we do not present the balancing test in this paper, but it can be provided on request. 6The results from five-nearest neighbors matching and kernel matching with other bandwidths are very similar and not presented in this paper, but it can be provided on request.

Figure 3. Density of predicted propensity scores. [Colour figure can be viewed at wileyonlinelibrary.com]

796 N. V. Cuong

http://wileyonlinelibrary.com

T ab le

3 .

Im p ac t o f th e m in im

u m

w ag e in cr ea se

o n p ro fi t m ar g in

– k er n el

n ei g h b o r m at ch in g w it h b an d w id th

o f 0 .0 5 : P an el

2 0 0 2 – 2 0 0 3

C o n tr o l g ro u p

2 0 0 2

2 0 0 3

D if fe re n ce -i n -d if fe re n ce s

Y 1

Y 0

Y 1 -Y

0 Y 1

Y 0

Y 1 -Y

(1 )

(2 )

(3 )=

(1 )- (2 )

(4 )

(5 )

(6 )=

(4 )- (5 )

(7 )=

(6 )- (3 )

C o n tr o l g ro up

h av in g m o n th ly

w ag e ab o v e 2 90

th o u sa n d V N D

2 .1 1 3 ** *

2 .0 3 2 * * *

0 .0 8 2

2 .0 7 2 ** *

1 .9 7 6 * * *

0 .0 9 6

0 .0 1 5

[0 .1 1 5 ]

[0 .1 4 2]

[0 .1 7 2]

[0 .1 1 1 ]

[0 .1 1 4]

[0 .1 5 5]

[0 .1 53 ]

C o n tr o l g ro up

h av in g m o n th ly

w ag e ab o v e 6 00

th o u sa n d V N D

2 .1 1 3 ** *

2 .0 9 1 * * *

0 .0 2 3

2 .0 7 2 ** *

1 .9 0 8 * * *

0 .1 6 4

0 .1 4 1

[0 .1 1 5 ]

[0 .1 3 9]

[0 .2 3 1]

[0 .1 1 1 ]

[0 .1 0 9]

[0 .1 8 6]

[0 .1 69 ]

T h e o ut co m e v ar ia b le

is th e ra ti o o f n et

p ro fi t to

to ta l sa le s re v en u es

(i n p er

ce nt ). T h e n et

p ro fi t is th e d if fe re n ce

b et w ee n to ta l sa le s re v en u e an d to ta l co st s o f fi rm

s. Y 1 is th e o u tc o m e o f th e tr ea tm

en t g ro up . Y 0 is th e o ut co m e o f th e co n tr o l g ro u p .

* S ig n ifi ca nt

at 1 0%

. * * S ig ni fi ca n t at

5 % .

* * * S ig n ifi ca nt

at 1 % .

S ta n d ar d er ro rs

in b ra ck et

(S ta n d ar d er ro rs

ar e ca lc u la te d u si n g b oo ts tr ap

w it h 2 00

re p li ca ti o ns ).

S o u rc e:

E st im

at io n fr o m

p an el

d at a o f E C s 2 0 0 2 an d 2 00 3 .

Impact of Minimum Wages on Firms 797

T ab le

4 .

Im p ac t o f th e m in im

u m

w ag e in cr ea se

o n p ro fi t m ar g in

– k er n el

n ei g h b o r m at ch in g w it h b an d w id th

o f 0 .0 5 : P an el

2 0 0 4 – 2 0 0 6

C o n tr o l g ro u p

2 0 0 4

2 0 0 6

D if fe re n ce -i n -d if fe re n ce s

Y 1

Y 0

Y 1 -Y

0 Y 1

Y 0

Y 1 -Y

(1 )

(2 )

(3 )=

(1 )- (2 )

(4 )

(5 )

(6 )=

(4 )- (5 )

(7 )=

(6 )- (3 )

C o n tr o l g ro u p h av in g m on th ly

w ag e ab o v e 3 50

th o u sa n d V N D

1 .9 1 2 ** *

1 .8 5 8 * * *

0 .0 5 4

2 .1 2 6 ** * *

2 .0 4 9* * *

0 .0 7 6

0 .0 2 3

[0 .0 98 ]

[0 .1 3 7]

[0 .1 0 7]

[0 .0 8 7 ]

[0 .1 1 4]

[0 .1 1 1]

[0 .1 0 9 ]

C o n tr o l g ro u p h av in g m on th ly

w ag e ab o v e 6 00

th o u sa n d V N D

1 .9 1 2 ** *

1 .8 4 5 * * *

0 .0 6 7

2 .1 2 6 ** *

2 .0 9 9* * *

0 .0 2 7

�0 .0 4 0

[0 .0 98 ]

[0 .1 4 1]

[0 .1 3 2]

[0 .0 8 7 ]

[0 .1 0 9]

[0 .1 0 4]

[0 .1 1 0 ]

T h e o ut co m e v ar ia b le

is th e ra ti o o f n et

p ro fi t to

to ta l sa le s re v en u es

(i n p er

ce nt ). T h e n et

p ro fi t is th e d if fe re n ce

b et w ee n to ta l sa le s re v en u e an d to ta l co st s o f fi rm

s. Y 1 is th e o u tc o m e o f th e tr ea tm

en t g ro up . Y 0 is th e o u tc o m e o f th e co n tr ol

g ro u p .

* S ig n ifi ca nt

at 1 0%

. * * S ig n ifi ca n t at

5 % .

* * * S ig n ifi ca nt

at 1 % .

S ta n d ar d er ro rs

in b ra ck et

(s ta n d ar d er ro rs

ar e ca lc u la te d u si n g b o o ts tr ap

w it h 2 0 0 re p li ca ti o n s) .

S o u rc e:

E st im

at io n fr o m

p an el

d at a o f E C s 2 00 4 an d 2 0 0 6.

798 N. V. Cuong

profit margin between the treatment and control group. The estimates of the impact on the profit margins from the difference-in-differences estimator are not statistically significant. Table 4 presents impact estimates of the minimum wage increase in 2005 on profit

margin also using kernel neighbor matching with bandwidth of 0.05.7 The profit margin of firms in the treatment and control groups is around 2 per cent. Similar to the case of the minimum wage increase in 2002, the impact of the minimum wage increase in 2005 of firms’ profit is not statistically significant. Finally, we examine the effect of the minimum wage increase on the proportion of firms

closing business. Using the panel data of EC 2002 and EC 2003, we can estimate the proportion of firms that appear in EC 2002 but not in EC 2003. Similarly, using the panel data of EC 2004 and EC 2006, we can estimate the proportion of firms that appear in EC 2004 but not in EC 2006. We estimate the difference in the proportion of disappearing firms for the treatment and control groups to examine the effect of minimum wages on business closure. Tables 5 and 6 show that firms with low wages are more likely to close business than firms with high wages. However, these differences are not statistically significant.8

5 CONCLUSIONS

Minimum wages are set up to protect low-wage workers from exploitation. In Vietnam, there have been nine increases of the minimum wage since the year 1993. Increasing minimum wages is sometime to blame for reducing firm profitability. Higher minimum wages implies higher costs and smaller profit margins. This paper is the first attempt to examine the impact of the minimum wage increases on firm profitability and the probability of firms closing business. More specifically, this paper use data from Vietnam Annual Enterprise Censuses from 2002 to 2006 to measure the impact of the minimum wage increase in January 2003 and October 2005 on private firms. The impact measurement method is the difference-in- differences with propensity score matching.

7The results from five-nearest neighbors matching and kernel matching with other bandwidths are very similar and not presented in this paper, but it can be provided on request. 8Another interesting question is whether minimum wages can deter firms’ business entry. However, the difference- in-differences requires data before and after a minimum wage increase. For firm entry, there are no data on wages on the newly established firms. Thus, this issue is out of scope of this study but certainly important for a future study.

Table 5. Impact of the minimum wage increase on the proportion of disappearing firms – kernel neighbor matching with bandwidth of 0.05: Panel 2002–2003

Control group

2003

Y1 (%) Y0 (%) Y1-Y0

Control group having monthly wage above 290 thousand VND 21.27*** 20.04*** 1.23 [0.49] [0.87] [0.94]

Control group having monthly wage above 600 thousand VND 21.27*** 19.46*** 1.81 [0.49] [0.92] [1.27]

The outcome variable is the proportion of firms that appear in EC 2002 but not in EC 2003 over the total number of firms in 2002 (in per cent). Y1 is the outcome of the treatment group. Y0 is the outcome of the control group. *Significant at 10%. **Significant at 5%. ***Significant at 1%. Standard errors in bracket (standard errors are calculated using bootstrap with 500 replications). Source: Estimation from panel data of ECs 2002 and 2003.

Impact of Minimum Wages on Firms 799

We find that the effect of the increase in the minimum wage on firm profitability is very small. The effect of the minimum wage increase on the proportion of firms quitting business is also very small. In addition, the impact estimates are not statistically significant. There can be several explanations for the small and insignificant effect. First, firms are able

to adjust the higher costs because of increased minimum wages so that their profit margins are not affected significantly by the minimum wage increase. It is possible that firms increase working hours or find other ways to increase productivity of workers to offset the increased cost because of minimum wages. Second, the minimum wages might not be effective in Vietnam. The minimum wage is around one-third of the average wages of workers in firms. There are firms that pay their employees below the minimum wage. These firms might not follow the minimum wages, and as a result, increased minimum wages do not affect their profitability. Third, our definition of the treatment variable does not capture well the variation of firms’ exposure to the minimum wage increase. Control groups can include workers below minimum wages and can also be affected by the minimum wages. The impact estimate using these control groups can be biased towards zero. A better indicator of the treatment variable can be the number of laborers with wages below the minimum level within a firm. However, this information is not available in our data sets.

ACKNOWLEDGEMENT

I would like to thank an anonymous reviewer for his/her very helpful comments and suggestions on the article.

REFERENCES

Abadie A, Imbens G. 2006. On the failure of the bootstrap for matching estimators. NBER Technical Working Paper No. 325.

Brown C. 1999. Minimum wages, employment and the distribution of income. Handbook of Labor Economics, Vol. 3. Edited by Ashenfelter O, and Card D. Elsevier Science, Amsterdam, the Netherlands, 1999, pp. 2101–2163.

Campolieti M, Fang T, Gunderson M. 2005. Minimum wage impacts on youth employment transitions, 1993–1999. Canadian Journal of Economics 38(1): 81–104.

Table 6. Impact of the minimum wage increase on the proportion of disappearing firms – kernel neighbor matching with bandwidth of 0.05: Panel 2004–2006

Control group

2006

Y1 (%) Y0 (%) Y1-Y0

Control group having monthly wage above 350 thousand VND 25.15*** 23.84*** 1.31 [0.44] [1.32] [1.47]

Control group having monthly wage above 600 thousand VND 25.15*** 23.49*** 1.66 [0.44] [1.15] [1.53]

The outcome variable is the proportion of firms that appear in EC 2004 but not in EC 2006 over the total number of firms in 2004 (in per cent). Y1 is the outcome of the treatment group. Y0 is the outcome of the control group. *Significant at 10%. **Significant at 5%. ***Significant at 1%. Standard errors in bracket (standard errors are calculated using bootstrap with 500 replications). Source: Estimation from panel data of ECs 2004 and 2006.

800 N. V. Cuong

Card D, Krueger A. 1994. Minimum wages and employment: a case study of the fast- food industry in New Jersey and Pennsylvania. American Economic Review 84(5): 772–93.

Card D, Krueger A. 2000. Minimum wages and employment: a case study of the fast-food industry in New Jersey and Pennsylvania: Reply. American Economic Review 90(5): 1397–1420.

Card DE, Krueger A. 1995. Myth and Measurement: The New Economics of the Minimum Wage. Princeton University Press: Princeton.

Dehejia RH, Wahba S. 1998. Propensity score matching methods for non-experimental causal studies. NBER Working Paper 6829. Mass: Cambridge.

Dickens R, Machin S, Manning A. 1999. The effects of minimum wages on employment: theory and evidence from Britain. Journal of Labor Economics 17(1): 1–22.

Draca M, Machin S, Reenen V. 2011. Minimum wages and firm profitability. American Economic Journal: Applied Economics 3(1): 129–51.

Duy T. 2009. B trư ng LĐTB&XH: Nghe thì c tư ng là tăng lương. (Minister of Labor, Invalid and Social Affairs of Vietnam: Wage Increase), VTC News, dated on 8/4/2009.

Freeman B 1994. Minimum wages – Again! International Journal of Manpower 15(2): 8–25. Frye J, Gordon J. 1981. Government intervention in the inflation process: the econometrics of

“self-inflicted wounds”. American Economic Review 71: 288–294. Gunderson M 2005. Minimum wages In Canada: theory, evidence and policy. The Federal Labour

Standards Review Commission, Minimum Wage No. 512. Hamermesh D 1986. The demand for labor in the long run. in Ashenfelter, O and Layard, R (ed.),

Handbook of Labor Economics, Volume 1, North Holland, Amsterdam, the Netherlands, 429–71. Harrison A, Scorse J. 2005. Moving up or moving out? Anti-sweatshop activists and labor market

outcomes. NBER Working paper No. 10492. Heckman J, Lalonde R, Smith J. 1999. The Economics and Econometrics of Active Labor Market

Programs. Handbook of Labor Economics 3, Ashenfelter A and Card D (eds). Elsevier Science: Amsterdam.

Katz F, and Krueger A. 1992. The effect of the minimum wage on the fast-food industry. Industrial & Labor Relations Review 46: 6–21.

Macdonald J, Aaronson D. 2000. How do retail prices react to minimum wage increases?. Working Paper 2000–20, Federal Reserve Bank of Chicago.

Montenegro E, Pagés C. 2004. Who benefits from labor market regulations? Chile, 1960–1998. In J Heckman and C Pagés, eds., Law and Employment: Lessons from Latin America and the Caribbean, 401–34. Chicago: University of Chicago Press.

Neumark D, Wascher W. 2003. Minimum wages and skill acquisition: another look at schooling effects. Economics of Education Review 22(1): 1–10.

Neumark D, Wascher W. 2002. State-level estimates of minimum wage effects: new evidence and interpretations from disequilibrium models. Journal of Human Resources 37(1): 35–62.

Rama M 2001. The consequences of doubling the minimum wage: the case of Indonesia. Industrial & Labor Relations Review 54(4): 864–81.

Rosenbaum P, Rubin R. 1983. The central role of the propensity score in observational studies for causal effects. Biometrika 70(1): 41–55.

Smith J, Todd P. 2005. Does matching overcome LaLonde’s critique of nonexperimental estimators?. Journal of Econometrics 125(1–2): 305–353.

Thai U. 2009. Tăng lương t i thi u, lương hưu. Báo Thanh Niên (Increasing Minimum Wages and Pension, in Youth Newspaper), dated 06/04/2009, Vietnam.

Impact of Minimum Wages on Firms 801

APPENDIX

Table A1. Probit regression of treatment variables

Explanatory variables

Panel data ECs 2002 and 2003: Control group having monthly wage

above 290 thousand VND

Panel data ECs 2004 and 2006: Control group having monthly wage

above 350 thousand VND

Agriculture 0.5043*** 0.1113 [0.1309] [0.1672]

Fishery n.a. �0.7629*** [0.1116]

Mining �0.4606*** �0.1095 [0.0734] [0.1021]

Manufacture 0.1272*** 0.0233 [0.0315] [0.0375]

Electricity, water 0.1406** �0.2217* [0.0631] [0.1229]

Construction 0.1528* �0.1676*** [0.0875] [0.0402]

Trade 0.0673* �0.2122*** [0.0347] [0.0351]

Hotel 0.2933*** 0.1669** [0.0492] [0.0755]

Transportation �0.1973*** �0.2023*** [0.0266] [0.0518]

Finance 0.1113 �0.3396 [0.2513] [0.2652]

Private enterprises Omitted Limited liability company �0.0386* 0.0345

[0.0198] [0.0276] Joint-stock company 0.0726* 0.2133***

[0.0409] [0.0391] Joint-stock company with less than 50% State capital

�0.6074*** �0.3784***

[0.1125] [0.1416] Fixed assets (billion VND) �0.0011*** �0.0212***

[0.0001] [0.0047] Basic construction capital (billion VND) n.a. 0.0108***

[0.0024] Revenues (billion VND) �0.0146*** �0.0290***

[0.0023] [0.0016] Red River Delta Omitted North East �0.1672*** �0.2103***

[0.0397] [0.0470] North West 0.0995 �0.3534***

[0.0938] [0.1085] North Central Coast �0.2743*** �0.2859***

[0.0399] [0.0498] South Central Coast �0.2808*** �0.2590***

[0.0373] [0.0458] Central Highlands �0.2373*** �0.2743***

[0.0493] [0.0677] South East �0.4716*** �0.5160***

(Continues)

802 N. V. Cuong

Table 0. (Continued)

Explanatory variables

Panel data ECs 2002 and 2003: Control group having monthly wage

above 290 thousand VND

Panel data ECs 2004 and 2006: Control group having monthly wage

above 350 thousand VND

[0.0357] [0.0464] Mekong River Delta �0.5044*** �0.2687***

[0.0327] [0.0445] Urban �0.0714*** �0.2184***

[0.0191] [0.0257] HCM city 0.0361 0.0454

[0.0331] [0.0450] Hanoi �0.2340*** �0.5120***

[0.0314] [0.0367] Constant �0.6574*** �0.5077***

[0.0389] [0.0490] Observations 39597 30379 R2 0.03 0.07

All explanatory variables are measured in the base years. *significant at 10%. **significant at 5%. ***significant at 1%. Robust standard errors in bracket. Source: Estimation from panel data of ECs 2002 and 2003, and panel data of ECs 2004 and 2006.

Table A1. (Continued)

Table A2. Summary statistics of variables used for matching (in 2002)

Variables Type Mean Standard deviation Agriculture Binary 0.0059 0.0763 Fishery Binary 0.0043 0.0654 Mining Binary 0.0095 0.0971 Manufacture Binary 0.2659 0.4418 Electricity, water Binary 0.0225 0.1484 Construction Binary 0.1189 0.3236 Trade Binary 0.4124 0.4923 Hotel Binary 0.0665 0.2491 Transportation Binary 0.0932 0.2907 Finance Binary 0.0011 0.0331 Private enterprises Binary 0.4689 0.4991 Limited liability company Binary 0.4722 0.4993 Joint-stock company Binary 0.0553 0.2286 Joint-stock company with less than 50% State capital Binary 0.0037 0.0604 Number of laborers Continuous 23.4 73.3 Fixed assets (million VND) Continuous 482.6 1870.9 Basic construction capital (million VND) Continuous n.a. n.a. Revenues (million VND) Continuous 1299.4 7357.1 Red River Delta Binary 0.3239 0.4680 North East Binary 0.0716 0.2578 North West Binary 0.0108 0.1034 North Central Coast Binary 0.0650 0.2465 South Central Coast Binary 0.0802 0.2716 Central Highlands Binary 0.0379 0.1910

(Continues)

Impact of Minimum Wages on Firms 803

South East Binary 0.2651 0.4414 Mekong River Delta Binary 0.1456 0.3527 Urban Binary 0.6309 0.4826 HCM city Binary 0.1822 0.3860 Hanoi Binary 0.1736 0.3788 Number of observations 3862

Source: Estimation from panel data of ECs 2002 and 2003.

Table A2. (Continued)

Table A3. Summary statistics of variables used for matching (in 2004)

Variables Type Mean Standard deviation Agriculture Binary 0.0050 0.0705 Fishery Binary 0.0066 0.0807 Mining Binary 0.0122 0.1097 Manufacture Binary 0.2592 0.4383 Electricity, water Binary 0.0072 0.0845 Construction Binary 0.1583 0.3651 Trade Binary 0.3745 0.4841 Hotel Binary 0.0537 0.2255 Transportation Binary 0.0887 0.2843 Finance Binary 0.0012 0.0353 Private enterprises Binary 0.2795 0.4488 Limited liability company Binary 0.5762 0.4942 Joint-stock company Binary 0.1408 0.3479 Joint-stock company with less than 50% State capital Binary 0.0034 0.0585 Number of laborers Continuous 27.1 72.1 Fixed assets (million VND) Continuous 683.6 2673.8 Basic construction capital (million VND) Continuous 190.4 1867.1 Revenues (million VND) Continuous 1820.9 8620.7 Red River Delta Binary 0.3535 0.4781 North East Binary 0.0743 0.2623 North West Binary 0.0094 0.0964 North Central Coast Binary 0.0609 0.2392 South Central Coast Binary 0.0790 0.2698 Central Highlands Binary 0.0262 0.1599 South East Binary 0.2936 0.4555 Mekong River Delta Binary 0.1031 0.3041 Urban Binary 0.6777 0.4674 HCM city Binary 0.2292 0.4204 Hanoi Binary 0.1605 0.3671 Number of observations 3202

Source: Estimation from panel data of ECs 2004 and 2006.

804 N. V. Cuong

Copyright of Journal of International Development is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.