Thailand effects a significant increase in minimum wage; Relationship to workers productivity?


Workers wage is a cost to business. However, workrs productivity is a gain for business. Thailand just increased its minimum wage significantly. What is the prospect for higher workers productivity?

Wikipedia says:

In a survey of manufacturing growth and performance in Britain, it was found that:

“The factors affecting labour productivity or the performance of individual work roles are of broadly the same type as those that affect the performance of manufacturing firms as a whole. They include: (1) physical-organic, location, and technological factors; (2) cultural belief-value and individual attitudinal, motivational and behavioural factors; (3) international influences – e.g. levels of innovativeness and efficiency on the part of the owners and managers of inward investing foreign companies; (4) managerial-organizational and wider economic and political-legal environments; (5) levels of flexibility in internal labour markets and the organization of work activities – e.g. the presence or absence of traditional craft demarcation lines and barriers to occupational entry; and (6) individual rewards and payment systems, and the effectiveness of personnel managers and others in recruiting, training, communicating with, and performance-motivating employees on the basis of pay and other incentives. The emergence of computers has been noted as a significant factor in increasing labor productivity in the late 1990s, by some, and as an insignificant factor by others, such as R.J. Gordon. Although computers have existed for most of the 20th century, some economic researchers have noted a lag in productivity growth caused by computers that didn’t come until the late 1990s.”[7][1]

  • Thailand‘s government, under Yingluck, again, just increased the Thai minimum wage to 300 baht a day, throughout Thailand; where that “Throughout Thailand” came after the policy was implemented in some parts of Thailand for an extensive study phase.

Before the increase to 300 baht a day, much of Thailand’s minimum wage was at about 200 baht a day, meaning Yingluck just increased the minimum wage by about 30%.

  • Without professional research backing, however, most Thai economist and business people, said that would add between 5% to 10% to a typical Thai firm cost.

From that Yingluck extensive study phase to assess the impact from the minimum wage, the conclusion is that indeed, the higher minimum wage, has an impact, but the impact was not significant, as many Thai firms, were paying higher wages than the new minimum wage already, and in fact, Thailand suffers from workers shortages.

However, that extensive study phase findings, says that Thailand’s Small to Medium Enterprises (SME) would find the higher minimum wage a threat, to their continued operation. And indeed, the Yingluck government, have passed a package to assist SMEs.

  • Macro Picture & ASEAN:

Yet there is a question that comes to the minds of most economist, when higher wage are implemented. And that is the question, of the relationship between higher wages, and workers “Productivity.” ASEAN wise, will this higher Thai minimum wage impact such decisions as Foreign Direct Investment (FDI)?

Vietnam, for example, is now targeting to get more quality FDI, away from FDI that targets to benefit from Vietnam’s cheap labor. If the policy shift succeed, Vietnam, indeed will give its more developed ASEAN partners, such as Thailand and Indonesia, some competition for quality FDI.

However, Vietnam Financial Review says Vietnam’s labor competitiveness is lagging. In the period 2001-2010, the growth rate of labor productivity of Vietnam was 5.13 percent, while China was 2 times that of Vietnam, Thailand 4.5 times, Malaysia 12 times and Korea 23.5 times.

The following is research on the subject of wage and workers productivity:

Do Wages reflect Productivity

Pieter Serneels

Global Poverty Research Group


The support of the Economic and Social Research Council  (ESRC) is gratefully acknowledged. The work was part of the programme of the ESRC Global Poverty Research Group. 1

Do Wages reflect Productivity?

Pieter Serneels

University of Oxford

October 2005


We investigate wage and productivity profiles in the Ghanaian Manufacturing sector using matched firm and worker data. Following Medoff and Abraham (1980,1981), we use performance appraisal as our measure of individual productivity. Controlling for a wide range of human capital variables, including cognitive skills, we find that average wage profiles do reflect productivity profiles. However, wages are steeper in large and unionized firms. This suggests that human capital theory holds for small and non-unionized firms, but that not hold for large and unionized companies.

1. Introduction

Twenty years ago Medoff and Abraham investigated wage and productivity profiles over the life cycle using performance evaluation as a measure of productivity at the individual level (Medoff and Abraham 1980, 1981). They found that wages do not necessarily reflect productivity. Although their findings triggered off substantial discussion, it is hard to find work that reproduces or contradicts the results using a measure of individual productivity. We find only two papers that study productivity at the individual, using a similar method: Bishop (1987) and Flabbi and Ichino (1998).

Nevertheless, as Bartelsman and Doms (2000) state in their review in the Journal of Economic Literature ‘… at the micro level, productivity remains very much a measure of our ignorance.’ (p586).

Other studies analyse productivity profiles on a more aggregate level, namely for groups of workers or indeed for the entire workforce of a firm. A common approach is to predict productivity for each worker using a firm level production function [see for example Hellerstein and Neumark (1993), Hellerstein, Neumark and Troske (1993), Bigsten et al (2000), Jones (2001)]. However, the obtained variable is the average for a group of workers. 1

Although taking the average may reduce the 1, It is the average per category, as entered in the production function. For example if one studies the returns to education, the labour component in the production function is split into different components, one for each level of education. The obtained estimate reflects the average returns for that level of education. A draw back is that one can only enter a limited number of categories in the production function because it becomes quickly too complicated (even more so if one wants to test the translog specification which includes interaction terms – see Berndt and Christensen 1972).

It becomes especially complicated when one wants to simultaneously analyse the effect for different (overlapping) characteristics, like for example education and gender. Another problem with including many categories is that is more likely that a category has zero observations for some cases, which makes the variable undefined when taking the log. The correction that is sometimes applied is to add one, but this 3 measurement error, it also suppresses heterogeneity. The group average may therefore have less variation across firms than productivity measured at the individual level. 

A(second stage) regression of productivity on individual characteristics may then result in coefficients that have lower significance.2 This shows the attraction of a direct measure of individual productivity: It allows a more precise analysis, and should give a more accurate understanding of the determinants of productivity at the individual level. It should also increase our understanding of whether wages reflect productivity, and more general, how labour markets work.3 However, the question is not whether it is useful to investigate productivity at the individual level, but how to measure it.

The ideal measure of individual productivity is a measure of individual output. Even though there seems to be a renewed interest in this approach among economists 4, most of the work in this area has been done by personnel psychology [see Sacket et al (1988)]. The difficulty with measuring individual output is that not all production biases the results, in particular when there are other firms with a small number of observations in the considered category.

2 A second method to obtain a prediction of individual productivity is to include the average of individual worker’s characteristic in a log linear firm level production function (see Hall and Jones (1999), Bils and Klenow (2000), Soderbom and Teal (2002)), for example the years of schooling. [see for example Bigsten et al (2000), Söderbom and Teal (2001).]

The obtained result reflects the effect of the mean years of schooling on firm level productivity. To compare this with the effect of schooling on wages is ambiguous. The first reflects the average effect of a characteristic of the entire labour force, while the second reflects the average effect of an individual characteristic. The two are not necessarily the same. The first one can be generated by different distributions and depends heavily on the composition of the labour force and on whether a characteristic has external effects beyond the individual. Take the example of education. Certain jobs do not need literacy, but only access to literate people. In this case it is the presence of an educated worker that has an effect on firm productivity; and the effect of his education goes beyond the effect on his own productivity.

There is convincing evidence that such externalities exist in households, as shown by Basu and Foster (1998) for India and Valenti (2001) for South Africa. It seems reasonable to expect that such externalities also play a role in a firm. [Further more, this approach does not solve the small sample problem when we want to allow for group specific effects. When for example we want to investigate whether the mean years of schooling has a different effect for female and male workers, the sample will still have missing values for firms without female workers. Not allowing for different age effects for male and female workers will give biased results] The problem with both approaches is that a productivity figure obtained from aggregate data is compared with the wage figure, obtained from individual data 3.

It may for example shy a new light on why a firm hires worker A rather than worker B since we expect that employers will decide based on comparing his wage with his expected productivity. 4 see for example Bandiera, Barankay and Rasul (2005) 4 processes lend themselves to this type of measurement, and if they do, that the measure is costly to obtain. Other methods have therefore been developed.

A popular measure is to consider individual performance using the personal assessment of the worker’s supervisor. Bommer et al. (1995), in their meta study, indicate that individual performance assessment, in spite of its shortcomings, is a successful measure. Medoff and Abraham (1980,1981) were the first to use this type of measure for economic analysis. We replicate their study in a different setting. Using the same evaluation tool – a well-defined ordinal scale for the evaluation of individual performance by the worker’s supervisor – we investigate the determinants of individual performance for a large sample of workers from different firms. We then investigate whether changes in wages reflect changes in individual productivity. If this holds, it is interpreted as supportive evidence for human capital theory: workers accumulate human capital over time and therefore earn more as they grow older.

Medoff and Abraham (1981), as well as Flabbi and Ichino (1998) find that wages have a steeper profile than productivity measured at the individual level, indicating that wages are determined by other factors than human capital. However, both papers use data from one or two firms only. Using a large sample of firms we find that changes in wages do reflect changes in productivity in small and non-unionized firms, but less so in large and unionized ones. This confirms results obtained by Bishop (1987).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s