Working Paper No. 40: Quheng Deng: Estimating the Effect of Minimum Wage on Firm Profitability in China

Deng, Quheng
Published: 2015/11/7 15:45:30    Updated time: 2015/11/12 16:09:15

Abstract: Utilizing the firm-level panel datasets and hand-collected data on county level minimum wage, this paper estimates the effect of minimum wage on firm profitability. As firms may take time to adjust in response to changes in minimum wage, this paper estimates a dynamic panel model with lagged minimum wage. To capture the heterogeneous effect of minimum wage on profitability, this paper further estimates a quantile regression dynamic panel model. Estimation results suggest that the effect of minimum wage in the current year is negative across the whole conditional distribution of profitability and it exhibits an inverted-U shape across conditional quantiles. Instead, the effect of lagged minimum wage is positive at the 5th, 10th, 15th quantiles, negative at the 90th and 95th quantiles, and not significant at other quantiles. Turning to the overall effect of minimum wage, we can find that minimum wage exerts significantly negative effect on profitability at the 5th quantile and quantiles higher than 40th and the absolute value of the effect of minimum wage increases with these quantiles. For other quantiles, the overall effect of minimum wage on profitability is negligible.
Keywords: Minimum wage, firm profitability, quantile regression dynamic panel model

(version: 2015/11/11)

Author:

        Deng, Quheng ---- Institute of Economics, Chinese Academy of Social Sciences

 

1. Introduction

Minimum wage policy is a way that government intervenes the labor market by prescribing a minimum wage level for salaried workers. Despite its clear policy target, the real effect of minimum wage policy is highly controversial. The inconclusiveness of the impact of minimum wage policy lies in two aspects. On the one hand, minimum wage policy has not only direct effects on wage level but also indirect effects on employment, wages, and firm profits. On the other hand, due to the data and methodology issues, even the direct effect on wage level of minimum wage policy is hard to measure with sound precision, not to say the indirect effect, the measurement of which turns out to be more difficult in general.

Although no consensus has yet been reached, there is a voluminous literature examining the impact of minimum wage policy on wage and employment. As minimum wage tends to increase the wage level of those salaried workers whose wage was below the minimum wage standard, it is expected that low wage workers will benefit most from the minimum wage policy. This conjecture has been confirmed in many studies. For instance, using the CPS data spanning from 1979 to 1997, Neumark et al. (2004) find that low-wage workers are most strongly affected, while higher-wage workers are little affected. However, to investigate the overall effect of the minimum wage policy on employees, the disemployment effect of the minimum wage policy has to be gauged since the increase of wages may decrease labor demand.

As discussed by the competitive labor market model, labor demand plays the key role in determining the quantity of employment in the presence of a binding price floor set by the minimum wage level (Boeri and van Ours, 2008; Card and Krueger, 1995). It follows naturally that an increase in the minimum wage leads to a decrease in employment. However, the empirical findings of Card and Krueger (1994, 1995) runs counter to the theoretical prediction abovementioned. They find instead that an increase in the minimum wage does not have a negative but has a small positive effect on employment. The findings of Card and Krueger are consistent with the labor monopsony model, under which firms face an upward sloping labor supply curve. Thus it is possible that an increase in the minimum wage results in an increase in employment.

Card and Krueger (1994, 1995) reignited the debate on the effect of minimum wage on wage and employment, joined by numerous later studies. Despite extensive researches on the effects of minimum wage on employees, there are very few studies examining the effect of minimum wage on firms. As wage and employment are labor market outcomes co-determined by the labor supply side (workers) and the labor demand side (firms), it is vital to analyze how firms respond to changes in minimum wage. Still, there is no consensus on the impact of minimum wage on firm profitability, the focus of this paper. Ceteris paribus, an increase in minimum wage leads to higher wage costs, to which firms may adjust by lowering profits. However, firms may have other alternatives to cope with the increase in minimum wage without suffering from loss of profits. Firstly, firms can simply choose not to comply with minimum wage policy if the cost of noncompliance is not too high. Secondly, firms can also pass on the higher wage costs to consumers by increasing prices of their products. Thirdly, higher wage costs may induce firms to substitute away from labor with capital. Finally, rising wage costs may push firms to enhance efficiency (Card and Krueger, 1995, pp. 313, 353).

Empirical studies present mixed evidence on the impact of minimum wage on firm profitability. Card and Krueger (1995) utilized the stock market event study methodology to evaluate the impact of the 1989 minimum wage hike on shareholder wealth. They calculated excess returns as the abnormal returns to stockholders of each firm in the sample, which is the predicted return deducted from the daily return. After examining the changes in excess returns following the news about minimum-wage hikes, they find that the evidence is mixed that news about a minimum-wage hike induces investors to adjust their valuation of firms downward (Card and Krueger, 1995, pp. 347). The same methodology was applied to the data from New Zealand by Pacheco and Naiker (2006), who find that there is an insignificant impact of a significant reform to the youth minimum wage on profit expectations for low wage employers by investors. Lin (2012) improves the stock market event study methodology by decomposing excess returns. Based on the decomposition results, he finds that there was a significant negative impact on firms that was neutralized by positive market performance, yielding a neutral aggregate effect of the 1989 minimum wage hike. Utilizing the introduction of a UK national minimum wage in 1999 as a natural experiment, Draca et al. (2011) examines the effect of minimum wage on firm profitability with a difference-in-differences approach. Their findings show that minimum wages reduce firm profitability significantly.

There are also studies on the effects of minimum wage on firm profitability in developing and transition countries. For instance, Harasztosi and Lindner (2015) exploits the large increase in minimum wage in Hungary in 2001 and finds that the rise minimum wage does not lower profitability among low-wage employers. Cuong (2013) investigates the effect of minimum wage increase in 2005 on profitability in Vietnam using a difference-in-differences with propensity score matching method and finds that the increase in minimum wage has no effect on firm profitability.

Minimum wage standard has increased dramatically in China in recent years, starting from 2004. The sharp increase in minimum wage in China has aroused wide debate on its effects on wage, employment, and income distribution among scholars, policy makers, and the general public. As China is approaching the end of cheap labor, some argue that increases in minimum wage will induce more labor supply. However, others argue that increases in minimum wage may raise firms’ cost and then reduce labor demand. Both arguments are supported by empirical studies in China and the impact of minimum on employment remains an open question. However, little research has been done to investigate the effect of minimum wage policy on the employer side. Among exceptions, Sun et al. (2013) examines the effect of minimum wage on firm’s export and finds that the effect of minimum wage on export exhibits a U-shape. Ma and Gan (2013) finds that minimum wage tends to decrease firms’ tendency of providing on-the-job training. Mayneris et al. (2015) finds that minimum wage reduce the survival probability of firms, increases incumbents’ productivity, maintains incumbents’ employment and profits.

This paper attempts to add to the literature by analyzing the effect of minimum wage on firm profitability. Utilizing the firm-level panel data and the hand-collected county-level minimum wage data, this paper estimates the effect of minimum wage on firm profitability. Compared with previous studies applying impact evaluation methods (mainly, difference-in-difference) to very few (usually two) time points, this paper examines the effect of minimum wage on profitability over a much longer time horizon and thus makes use of much more information from the data.

Since firms may take time to adjust in response to changes in minimum wage, this paper estimates a dynamic panel model allowing for lagged effect of minimum wage by the use of the differenced GMM method. However, as the linear dynamic panel model cannot reveal differences of the effects of minimum wage on profitability, this paper estimates a quantile regression dynamic panel model, which is also less susceptible to outliers.

Estimation results of that quantile regression dynamic panel model suggest that the effect of minimum wage in the current year is negative across the whole conditional distribution of profitability and it exhibits an inverted-U shape across conditional quantiles. Instead, the effect of lagged minimum wage is positive at the 5th, 10 th, 15th quantiles, negative at the 90 th and 95th quantiles, and not significant at other quantiles. Turning to the overall effect of minimum wage, we can find that minimum wage exerts significantly negative effect on profitability at the 5th quantile and quantiles higher than 40 th and the absolute value of the effect of minimum wage increases with these quantiles. For other quantiles, the overall effect of minimum wage on profitability is negligible.