The Chinese comparative advantage : where did it come from, and will it sustain?

The Chinese comparative advantage: “Where did it come

from, and will it sustain?”

Camiel Zandbergen

10250689

Bachelor Thesis

Table of contents:

1. Introduction

-Relevance -Motivation

2. Literature Review

- Understanding China’s growth: Past, Present and Future. - The end of Chinese cheap labor.

- Rising wages: Has China lost its global labor advantage?

- The Rise of China and East Asian Export Performance: Is the Crowding-Out Fear

Warranted?

- Does comparative advantage explains export patterns in China?

3. Methodology

-Model

-Explanation & intuition variables

4. Data

5. Results

6. Conclusion

7. Appendix

8. Reference list

Introduction:

China has experienced great economic growth over the last decades. The average GDP growth is 9,98% between 1987 and 2011, resulting in a very powerful economic force. One of the main reasons for these developments is the low labor costs in China. This made it possible to produce and offer labour at a very low and attractive price.

Obviously this created a comparative advantage towards other exporting countries with higher wages, and helped China to become the largest exporter in the world. With an average of 22,44% export growth over the last 25 years, its export value is now almost 100 times larger than in 1987(Appendix A).

Around 1978, the average annual wage of a Chinese worker was approximately 1000 U.S Dollars, which is 3% of the average annual wage of American workers. China was one of the poorest countries in the world at this time. Through the years, the average wage of a Chinese worker has risen to almost 5500 U.S Dollars. That is a 550% increase in 35 years, which is quite agressive. The wage level in China is expected to be 20.000 U.S Dollars in 2020(Hongbin Li, et al. 2012). When the wage level rises as much as the productivity, there will be no labor productivity loss. In that case the comparative advantage will sustain. When the production of an employee rises relatively as much as the wage rises, there will be no loss. But since 1997, the wage is rising at a faster pace than productivity. This results in worsening of the labor productivity level, and therefore a worsening of the competitive position towards other exporting countries. This brings up the question if the comparative advantage of China will sustain, vanish or converge into another advantage?

To answer this question, five papers were reviewed. In these papers, different perspectives on the Chinese comparative advantage are described. After the literature review, a regression was formed. This regression was an adjusted version of the regression in the article of Changjun & Ping(2002). The findings were that the model was significant, and the regressors were generally in line with the intuition. But the regressors were not significant enough, because of missing data.

Literature Review

The relevance of the article of Xiaodong Zhu, 2012.

It is always very important to investigate different sides of a story. As for this article, it also stipulates that the migration from rural workers to the urban sector had a great influence on the overall productivity. So not only the low wages were an important source of China’s economic growth, also the reformation of the labor market. And next to that, the reason why the current high export level is possible is because of the reduction of legal barriers. The importance of these developments for my thesis is obvious, because the thesis is trying to determine the relationship between exports and wages.

In the article of Xiaodong Zhu(2012), it is shown that there has been a huge development in China over the last few decades. “In 1978, China was one of the poorest countries in the world.” One can use the real per capita GDP level as an estimate,

assuming that this level was about one-fortieth of the U.S. level in 1978. From that year, the real per capita GDP of China has grown approximately 8 percent per year.

Nowadays, the Chinese level is at almost 20% of the U.S. level. One of the reasons is the high labor participation rate growth in the period 1978-2007. This rate was almost 520% of the average rate in 1952-1978. The reason for this is probably the one-child policy in China, which ensured that the parents were still able to work. This increased the labor force, as expected. This labor participation increase accounts for 7% of the per capita GDP growth. The most important source was the Total Factor Productivity

increase, which accounts for almost 80% of the growth in GDP per capita.

The rising labor force, and the heavily increasing productivity, caused a reform of the Chinese labor market. In 1978, 69 percent of all workers were active in the

agricultural sector. This employment share moved to 62 percent active workers in the non-agricultural sector, in 2007. The growth in TFP in the agricultural sector accounts for 91% of the reduction in employment in that sector. This makes sense because agricultural productivity rised, so there were less workers needed in this sector than before. Therefore these workers were reallocated in the non-agricultural sector, where the average labor productivity is generally higher than in the other sector. So this reallocation contributes positively to aggregate productivity growth.

At the beginning of this particular time interval, the labor productivity was 6 times higher in the non-agricultural sector than in the agricultural sector. That is why one could say that the most important source of China’s growth during the first two decades of economic reform were caused by the reforms in the agricultural sector. After 20 years of productivity growth, the gain from reallocation naturally declined in the later years. The growth continued due to the reduction of legal barriers in 1997. This resulted in an aggressive growth of private enterprises.

The relevance of the article of Hongbin Li, Lei Li, Binzhen Wu and Yanyan Xiong, 2012.

The increase of the wage level in China is one of the main motivations of the thesis. The fast increase during the last three decades is very interesting, and the central argument for assuming that the comparative advantage of China is in danger. Especially in the last decade there have been huge export increases, and in this article it is also told what else changed. Before 1998, the labor productivity growth was usually higher than the wage level growth. After this breaking point, the opposite was true. Labor is becoming more expensive, and this could affect the attractiveness of China as a workshop.

In the article of Hongbin Li et al (2012), it is shown that there has been a huge development in China’s wages over the last few decades. In 1978, the Chinese average annual wage was only 3% of the U.S. wage level. The wage was lower than China’s neighbour Asian countries, and also low relative to productivity. Decades later, in 2010, the wage grew to almost 550% of the level in 1978. One could regard this as similar to countries like Thailand and the Philippines. It is even higher than India and Indonesia. Especially the last decade has shown huge increases, with an average wage growth of 13,8% per year between 1998-2010. If the wage growth is lower than labor productivity growth, then labor becomes less expensive.

In this article they calculate the growth of gross labor productivity by deducting the growth rate of the labor force from the growth rate of real GDP. Between 1982-1997 the real GDP grew with 5,5% annually, and the labor force with only 1,9% per year. This implies a labor productivity growth of 3,6% a year, exceeding the real wage growth of 1,3%. So labor was becoming less expensive during this time period. This is very different for 1998-2010. The real GDP grew with 12,7% a year on average, and labor force growth decreased to only 1,4% on yearly basis. This results in a labor productivity growth of 11,3 percent, which is very high. But because the real wage growth was 13,8% in this period, labor was becoming more expensive.

It is stated that due to the labor market reforms, the return to education is increasing. This suggests that the link between wage and productivity is becoming stronger, which is proved by the results given earlier. The returns to additional years of schooling increased from 2% to 10% between 1988-2009, and the return to college education increased from 7% to 49% during the same period.

In 1970, the total fertility rate in China was 6 births per woman. This rate has decreased to 1.4 in 2010, as a result of the one-child policy. As one can imagine, the large amount of children in 1970 were not able to have more than one child. Obviously this created a bigger labor force, a labor force that almost doubled in the last 31 years to 1 billion in 2011. The working age population in 1980 was 59,3%, and this increased to 74,4% in 2011. For now, less children means less potential workers. So because of the low fertility, the Chinese labor force growth is slowing down. It is mentioned that this could also explain the rising wages since 1997, which is accompanied by the decline in labor force growth.

China is already experiencing labor shortages. The ratio position/seeker is rising from 0.65(2001) to 1.08(2012). This means that the number of vacancies is becoming higher relative to the number of job seekers. Labor is becoming scarce, which usually causes rising wages. The migration of rural workers to urban areas has helped to keep the wages low in the past decades, but this pattern seems to change. The number of migrant workers increased from 25 million in 1985, to 159 million in 2011. But this process is slowing down; the breaking point was in 1997. Pre-1997 the growth rate of migrant workers was 10,8%, and post-1997 the growth rate is only 4,6%.

Major barriers of growing migration are the fact that those who can migrate at the lowest costs are already doing so, and another barrier is the hukou system. This is a household registration, and every person has to be registered. This specifies the

rural/urban station and is based on place of birth. This hukou status has negative side effects for migrants, because rural migrants in urban environments can’t enjoy the public welfare benefits. This restricts the internal mobility in urban and rural areas, and a rising marginal cost of migration is a consequence. In 1980 China was cheap, and the ratio of wage to productivity was a huge comparative advantage towards other nations. This resulted in sourcing of labor, fast employment growth and rural-urban migration.

Now because of the changes in the late 1990’s, the underpricing of Chinese labor seems to come to an end. Wages are rising faster than labor productivity, particularly in labor-intensive exporting industries such as apparel and electronics, which will likely move out of China to countries like India and Vietnam. China is becoming a middle-wage country, and when the current growth rate is maintained, the average real wage in urban China will be 20.000 U.S. dollars in 2020. This level was reached by the US in 1980, by Japan in 1986 and by Korea in 1995.

The article suggests that China needs to move to higher value-added industries, export or domestic. China is ready to move up the technological ladder. As a results of the heavy investments in R&D by the manufacturing firms, the labor productivity in China has been increasing at 11.3% a year over the last decade. The expenditure on R&D per worker is increasing at an annual rate of 16,9% in the past 20 years. Next to that, capital deepening has also increased and is expected to have a huge impact.

But probably more important, the human capital has risen dramatically. In 1999, the Chinese government started an aggressive “College Expansion” movement that increased the college entry class enrollment from 1.1 million in 1998 to 6.6 million in 2011. It is predicted that in 2050, approximately 40% of the Chinese labor force will hold a college degree. This is similar to the Japanese labor force today. The comparative advantage of China is shifting from labor to productivity and education. If China can improve the quality of education and develop institutions that help to foster innovation and entrepreneurship, then going forward, China may join a place alongside Korea and Japan as a formidable force in high value-added manufacturing and innovation.

The relevance of the article of Dennis Tao Yang, Vivian Weijia Chen and Ryan Monarch, 2010.

It is very important to take into account that the average wages in China can be divided into different categories. There are skilled labor and un-skilled labor, and a skilled worker is highly educated. So a un-skilled worker did not follow any higher education, at a

university or community college. In the past years, the average real wage in China

increased. In the article it is stated that this increase is mostly in the skilled labor sector. As for the effect on export, we must look at the un-skilled labor sector. This article gives insights in the fact that the average overall wages do increase, but that this does not really appeal to the wages related to export.

China is becoming familiar as the world’s workshop, and as the biggest platform for FDI’s. It has deeply affected global trade systems and the competitive environment. The main drive for these developments is China’s comparative advantage, low wages. But these wages have increased with globalization, average real wage almost tripled between 1997-2007. The Lewisian turning point is almost there, which implicates the end of cheaply supported labour for the continuing industrialization. So wages will become too high in the urban sector. Both studies forecasted severe labour shortages and an acceleration of wage rises in China, starting at 2009/2010.

Recent studies have warned that China is losing its lustre as a manufacturing base for exports. The margins of multinational corporations tighten due to the accelerated wage growth, and therefore the circumstances in China for FDI are worsening relative to other emerging economies in Asia. Reformation of the labor market has led to growing unemployment in the urban sector. From 6,1% in 1996 to 11,1% in 2002. A big gap between regions was caused, with respect to wage growth. A number of government policies and strategies have contributed to unbalanced growth between regions.

This includes urban-related fiscal and credit programs as well as development strategies aimed at the growth of coastal regions, especially when combined with restrictions on factor mobility.

The Chinese government was aware that this could cause instability and other social conflicts. They founded the Western Development Strategy, aimed to retrieve the balance between coastal and in-land regions. This worked, soon the gap stopped

increasing and by 2006/2007 it started declining. This article assumes that the wages for un-skilled labor are still very low, and these wages are important for exports. Un-skilled labor is construction, manufacturing and basic service sectors. The wages in these sectors are very low, and this also holds compared to other developed Asian economies. The wage of skilled labor is increasing very fast, and also the proportion of skilled individuals in the labor force is increasing. So these two developments result in a fast increase of the average wages in China.

Since the WTO accession in 2001, the GDP growth has been 10% a year and the export growth has ben 30% a year. With exports and GDP growth likely cooling to more sustainable levels, the demand for labor will lessen accordingly. Despite the high export growth in manufactured goods, the total employment in manufacturing only increased by 3% in recent years. Because the manufacturing sector employs predominantly workers with low education levels, it’s moderate pace of growing demand is not putting additional pressure on labour demand.

This could also be explained by improved productivity coming from continued enterprise privatization and intense global competition. So higher exports don’t result in enormous labor demand increases. And also the current wage growth level in skilled labor will not sustain, because the rate of return to education has already tripled over the last two decades and reached a conventional level. It is expected that the wages of skilled labor will stabilize and remain constant. Another reason is that there is a huge supply of college graduates over the past several years. So the supply grows, and the quality of schooling declines. This should result in a downward pressure on the wage level.

The relevance of the article of Prema-Chandra Athukorala, 2008.

The importance of the wage level for the exports is also depending on the relevance of labour in exports. When the usage of labour rises as much as exports, the wage level will be relevant in many ways. But when the relative usage of labour declines, meaning that the share of labor in total export is declining, the relevance is in question. When the export industry in China is moving from labour intensive goods to capital intensive goods, the whole development of the higher wages has less impact.

The share of manufactures in China’s total merchandise exports increased from less than 40% in late 1970’s to 92% in 2005/2006. Until about the early 1990’s,

traditional labour-intensive manufacturers were the prime movers of export expansion. In particular apparel, footwear, toys and sport goods. Since then, there has been a notable shift in the export composition away from conventional labour-intensive product lines and towards more sophisticated product lines. The sophisticated goods should be those within the broader category of machinery and transport equipment. This caused the traditional labour-intensive product share in total export to decline from 49% to 31%. The share of machinery increased from 17% to 44%, so there is a clear development observed here. Over the past decade the share of final goods in machinery exports has remained around 75%, with only minor year-to-year changes.

Given the fact that the production of parts and component is generally more capital and technology intensive than final assembly, these numbers clearly suggest that China’s export success has so far been underpinned largely by its comparative

advantage in international production arising from labour abundance. When

components are netted out, more than 80% of total Chinese manufacturing exports from China can still be treated as labour-intensive products.

The relevance of the article of Y.U.E. Changjun and H.U.A. Ping, 2002.

The development of China’s exports started years back in 1978. It is very important to take the reforms between 1978 and 2000 into account. The comparative advantage was

exploited exactly the way it should be, and this resulted in huge export growth.

Since the liberalization of the foreign trade system and the exchange rate regime at the end of 1978, China has experienced remarkable export growth, averaging 13,6% per year from 1980 to 2000, far higher than the world average of 5,9% over the same period. In this particular period, China’s share of world exports became 4 times larger than in 1980. As a labor-abundant underdeveloped country, China would appear to have a comparative advantage in labor-intensive products and thus tend to export them to world markets.

China’s reforms and its open-door policies have allowed its economy to move from a heavy industry-oriented development strategy in a capital-scarce economy to a comparative advantage strategy in a labor-intensive economy. There was better

exploitation of China’s comparative advantage in labor-intensive manufacturing. Second and most important, comparative advantages turn out to explain the level and trend of export patterns in China during its economic transition. The provinces whose export patterns are congruent with the comparative advantage principal exported more goods and their exports increased more quickly.

Methodology

The model used is as followed:

Exp=a0+a1Rwage+a2Grad+a3Lp+ a4Rcpi

The variables express the change with respect to the previous year. So they measure growth.

Example: (Rwage1991-Rwage1990)

(Rwage1990)  gives growth of Rwage in 1991 with respect to (wrt) 1990

Dependent variable:

Exp=The export change wrt to the previous year. Independent variables:

1. Rwage= The real average annual wage change wrt to the previous year.

The intuition is that when the rwage goes up, this will mean that the costs of producing will increase. This should result in less export demand, because of the higher prices. So the intuition is that the coefficient of this variable is negative.

2. Grad= The change in the amount of college graduates wrt the previous year. It captures the relative value of graduates wrt the whole population.

The intuition is that when there are more individuals with a college degree, the value of human capital increases. This could lead to better technology and innovation, which could increase total factor productivity. As a result of this, the export demand could increase. Perhaps this is due to lower prices or better quality.

3. Lp= The change in labor productivity wrt the previous year.

The intuition is that when lp increases, labor becomes less expensive. And when labor becomes less expensive, this could result in lower prices. Obviously, this will lead to a higher export demand.

4. Rcpi= The change of (CPICHINA_/CPIUS_{) wrt the previous year.}

The intuition is that when the prices in the US rise faster than the prices in China, the export demand of Chinese products will increase. So in this particular case, it is expected that when the ratio goes up, the export demand goes down.

This model is an adjusted version of the model that was used by Changjun & Ping(2002). The model explains the effect of the independent variables on the dependent variable. The hypothesis contains the relationship between wage and export. As is also discussed in the article of Honbin Li et al.(2012), the effect of real wages on export is not clear. One could say that if the wages go up, the costs of the exported products go up, resulting in a higher price. This will result in less demand. But it is also a possibility that higher wages will create an increase in labor productivity.

Data

The attained data is yearly data between 1987-2011. Almost all data was available, only data on labor productivity was partly unavailable. For this variable, there are only 16 observations. The complete data is in Appendix A, in the table below there is summary. Variable Mean =Average growth

between 1987-2011(%) Std. Dev. Observations

Exp .224384 = +22,44% .212051 25

Rwage .150404 = +15,04% .0567723 25

Grad .112292 = +11,23% .1277266 25

Lpgr .127450 = +12,75% .1190857 16

Rcpi .029476 = +2,95% .0651149 25

1. Exp=The export change wrt to the previous year.

2. Rwage= The real average annual wage change wrt to the previous year. 3. Grad= The change in the amount of college graduates wrt the previous year. It

captures the relative value of graduates wrt the whole population. 4. Lpgr= The change in labor productivity wrt the previous year.

5. Rcpi= The change of (CPICHINA_/CPIUS_{) wrt the previous year.}

-The data on export, annual wages and the CPI of China are published on the website:

http://www.stats.gov.cn/english, which is the National Bureau of Statistics China. Obviously, the annual wage divided by the CPI of China approximates the real wage. -The data on the CPI of the US are published on the website:

http://www.usinflationcalculator.com/, which contains information from 1913 about the CPI in the United States. The CPI of China and the CPI of the US contain the

information needed to calculate the Rcpi variable.

-The information about the number of graduates in China and the labour productivity was given on the website: http://tongji.cnki.net, which is the China Statistical Yearbooks Database.

Explanation of particular trends in Appendix A:

-Low export growth in 2009: This is a result of the lower world market demand, caused by the financial crisis.

-High growth in number of graduates since 2000: This is a result of the aggressive “College Expansion” movement of the Chinese government, which started in 1999. -Negative/lower growth of (CPICHINA_/CPIUS_{) since 1997: This is probably a result of the} reduction of legal barriers in 1997, which resulted in a aggressive growth of private enterprises. The market was more open for international trade, and this brought the price levels closer to each other.

Results

Because there was not enough data on labor productivity for the entire period, I first tried to regress export on the other variables. The results were not significant in all cases, probably due to omitted variable bias. After labor productivity was plugged into the regression, the following results were estimated:

regress exp rwage grad lp rcpi

exp rwage -1.418129 (0.206)* grad 0.6556851 (0.046)** lpgr 1.025375 (0.082)* rcpi 1.418164 (0.250) _cons .1985132 (0.304) Prob > F 0.0073 R2 0.6930 ***=Significant at 1% level **=Significant at 5% level *=Significant at 10% level

The overall significance of this model is very high. This is showed by Prob > F = 0.0073 and R2 _{of 0.6930.}

Rwage:

Just as the intuition, there is a negative relation between rwage and exp. When the real wage growth goes up, the export growth goes down. The coefficient is -1.418, and this means that the relation is increasing. The t-value of this estimator is -1.34, and even though it is a one-sided test, at a 5% significance level the coefficient of rwage is not proven to be different than zero. At a 10% significance level, the coefficient is valid and the hypothesis that it is zero is rejected.

Grad:

Also intuitively, there is a positive relation between grad and exp. The relation is

decreasing, meaning that when grad goes up by 1%, exp will go up by less than 1%. This is explained by the value of the coefficient: 0,656. The t-value of the coefficient is 2.25, meaning that the coefficient of this variable is non-zero at a 5% significance level.

Lpgr:

As expected, there is a positive relation between lp and exp. This is almost with constant returns to scale, meaning that when lp goes up by 1%, exp will also go up by 1%. This is explained by the value of the coefficient: 1,025. The t-value of this coefficient is 1.91, meaning that it is not significant at a 5% significance level. At a 10% significance level, this coefficient is nonzero.

Rcpi:

Against our intuition, there is a positive relation between rcpi and exp. The relation is with increasing returns to scale, meaning that when rcpi goes up with 1, exp will go up with more than 1. This is explained by the value of the coefficient: 1,42. The t-value of the coefficient is 1.22, meaning that the estimator is not different from zero. This is the case for a 5% significance level, and also for the 10% significance level.

I think that the problem is that there are not enough observations. Almost all the coefficients were as the intuition, but it is hard to make a good model with only 16 observations.

Conclusion

It is very hard to conclude if there is a relation between export growth and real wage growth. My intuition tells me that there is a clear relationship, but I cannot verify this with the results. I think that, as is also explained in one of the articles, China will experience a shift in the comparative advantage. The wages increased the most in the skilled labour sector. When wages rise faster than productivity, this will cause a loss in labour productivity. That is undeniable, and this will most definitely affect the

comparative advantage.

But this does not mean that the competitive position of China is in danger,

because they invested in other potential comparative advantages. Investments in human capital, more sophisticated technologies and higher productivity. Next to that, some argue that there is no significant relationship between the current wage growth and the future developments in export. The reason for this is that the wages and the labour demand did not rise abnormally in the export sector, but only in the skilled labour sector. I think that it is inevitable that the wages in the export sector will also rise in the nearby future, but this negative influence on the competitive position will be

Appendix A

This table contains the retrieved data:

exp rwage grad lpgr rcpi

1987 35,85% 8,96% 33,20% 3,57% 1988 20,18% 8,15% 2,39% 14,12% 1989 10,71% 11,51% 2,53% 12,60% 1990 52,65% 26,58% 4,93% -2,18% 1991 28,18% 9,03% -1,25% -0,77% 1992 22,19% 12,59% -2,83% 3,30% 1993 13,01% 15,35% -6,58% 20,64% 11,36% 1994 97,20% 24,42% 10,30% 47,03% 20,96% 1995 19,48% 28,44% 25,01% 21,83% 13,91% 1996 1,00% 22,08% 3,14% 1,77% 5,15% 1997 20,55% 9,76% -2,20% 6,63% 0,49% 1998 0,47% 19,79% -0,84% 10,43% -2,36% 1999 6,09% 12,27% 1,24% 8,25% -3,52% 2000 27,69% 10,27% 11,18% 10,23% -2,90% 2001 6,74% 15,65% 8,32% 10,62% -2,04% 2002 22,35% 16,01% 28,18% -10,82% -2,36% 2003 34,66% 10,79% 39,49% 11,21% -1,08% 2004 35,32% 11,17% 26,64% 19,52% 1,17% 2005 27,58% 16,97% 27,55% 12,16% -1,55% 2006 23,86% 14,70% 22,36% 9,88% -1,65% 2007 20,44% 14,98% 18,00% 12,09% 1,95% 2008 7,43% 16,02% 13,74% 12,45% 2,02% 2009 -18,29% 19,44% 3,22% -0,30% 2010 30,47% 9,08% 7,82% 1,67% 2011 15,15% 12,00% 5,19% 2,13%

References:

Athukorala, P.C., (2009). The Rise of China and East Asian Export Performance: Is the Crowding-Out Fear Warranted? The World Economy(2009), doi:

10.1111/j.1467-9701.2008.01151.x

Changjun, Y.U.E., Ping, H.U.A., (2002). Does comparative advantage explains export patterns in China? China Economic Review 13 (2002), pp. 276-296

Hongbin, L., Lei, L., Binzhen, W., Yanyan, X., (2012). The end of Chinese cheap labor.

Journal of Economic Perspectives, 26:4(2012), pp. 57-74

Xiaodong, Z. (2012). Understanding China’s growth: Past, Present and Future. Journal of

Economic Perspectives, 26:4(2012), pp. 103-124

Yang, D.T., Chen, V.W., Monarch, R., (2010). Rising wages: Has China lost its global labor advantage? Pacific Economic Review, 15:4(2010), pp. 482-504