Economic Development Strategy, Openness and Rural Poverty:
1A Framework and China’s Experiences
Justin Yifu Lin
Development Research Center of the State Council, China
This paper argues that both openness and poverty in a country are endogenously determined by the country’s long-term economic development strategy. Development strategies can be broadly divided into two mutually exclusive groups: (i) the comparative advantage-defying (CAD) strategy, which attempts to encourage firms ignoring the existing comparative advantages of the economy in their entry/choice of industry/technology; and (ii) the comparative advantage-following (CAF) strategy, which attempts to facilitate the firms’ entry/choice of industry/technology according to the economy’s existing comparative advantages.
To realize the CAD strategy, many governments of LDCs subsidize the firms in the priority sectors with distorted prices of capital, foreign exchanges, and other inputs, and use an administrative method to allocate the price-distorted resources to the firms. The functions of market will be suppressed. Rent seeking will be widespread. As a result, the performance of the economy will be poor and the equity of income distribution will be worsened. The foreign trade will also be repressed.
Only if the government in a less-developed country follows the economy’s comparative
advantage as the basic principle for promoting the economy’s industrial development, will the economy have an open and well function of market, maintain a high rate of capital accumulation, achieve a rapid upgrading of its endowment structure, and result in more equitable income distribution and less poor people.
JEL Classification: O14_O20_O33_ N65
Key Words: economic development strategy, income distribution, globalization,
1 China mentioned in this paper only refers to Chinese mainland, not including Hong Kong, Macao and Taiwan.
With disintegration of the colonial system in 1950s, many newly independent countries have been striving to develop themselves and to get rid of poverty. But so far, except for a few countries and economies in the East Asia, most of them have not realized these goals. People’s living conditions in most those countries have not been improved substantially. There is even no change of life for people living in sub-Saharan Africa.
thThe trend of globalization in the last decades of 20 century had provided many
opportunities and challenges for most countries in the developing world. Though there is much interest on the relationship between globalization and poverty, a satisfactory analytical framework is wanting. Nor is there consensus on the impact of globalization on poverty from cross-country empirical studies. Just as Nissanke and Thorbecke (2004) pointed out:
Cross-country studies would require precise measurements and definition of the two
key concepts-globalization and poverty. Yet, both concepts are multi-dimensional, and
not easily captured in a composite index to be used in a meaningful manner in
cross-country, comparative studies. Indeed, only detailed case studies are able to
delineate the role of path dependence of multiple factors such as resource endowments,
trade and production structures, policies, and institutions. Such research, if carefully
conducted, should yield high dividends in identifying appropriate policy responses to
globalization in relation to the overriding policy objective of poverty reduction.
China’s experience after implementing reform and open-door policy at the end of 1978 is an interesting case in the study of globalization and poverty. The average annual growth rate of GDP and trade reached respectively 9.4% and 16.0% in
21979-2003. Calculating with current price, shares of primary industry, secondary industry and tertiary industry in GDP changed from 28.1%, 48.2% and 23.7% in 1978 to 14.6%, 52.2% and 33.2%, respectively, in 2003. Similarly, employment structure
3changed from 70.5%, 17.3% and 12.2% in 1978 to 49.1%, 21.6% and 29.3% in 2003.
The life expectancy of Chinese people increased from 66.8 years in 1980 to 70.7 years
2 Data source of GDP and GDP per capita: China Yearbook 2004. Annual growth rate here refers to annual
compound growth rate. 3 Data source of industrial structure and employment structure: China Yearbook 2004. Many people argue that
share of tertiary industry is under-estimated, but that of secondary industry is over-estimated.
4in 2002. The illiteracy decreased from 22.81% in 1982 to 6.72% in 2000. In addition, the share of people with education above high school grew from 7.39% in 1982 to
514.76% in 2000.
In the study of globalization and poverty, the Chinese experience deserves special attention: with an average annual growth rate of 16.0%, China’s foreign trade
dependency ratio has increased dramatically since 1978; and China has also attracted a large amount of foreign direct investments (Table 1 and Table 2). Poverty has been decreasing significantly. According to Ravallion and Chen (2004), in terms of both the
6Chinese official poverty line and the new poverty line established by National
7Statistical Bureau (NSB), there is a significant reduction of head count index (H),
poverty gap index (PG), squared poverty gap index (SPG) in rural areas, urban areas
and the whole country (Table 3). The World Development Index (Table 4) also indicated that China had achieved great success in poverty reduction. In spite of the achievements, in recent years China faces an increasing income disparities and new characteristics of poverty. It is imperative to have an analysis on China’s experiences
of openness and poverty reduction, as well as its achievements and lessons.
Table 1: China’s Foreign Trade Dependence Ratio
Commodities Commodities Commodities Commodities Trade Dependent
Export Import Export/GDP Import/GDP Ratio (%) (%) (%) ;$,Billion； ;$,Billion；
(2) (4) (5)=(2)+(4) (1) (3)
1978 9.75 4.62 10.89 5.17 9.79
1980 18.12 6.00 20.02 6.61 12.61
1985 27.35 9.02 42.25 14.03 23.05
1990 62.09 16.10 53.35 13.88 29.98
1995 148.78 21.29 132.08 18.89 40.18
2000 249.20 23.06 225.09 20.83 43.89
2003 438.228 30.95 412.76 29.16 60.11
Source: NSB: China Statistic Book, each year.
4 Data source: World Bank: World Development Indicators 2004. Life expectancy of 2000 reported in China
Yearbook 2004 of NSB was 71.4 years old, higher than the data reported in the World Bank’s report. But generally
speaking, it is the fact that Chinese people’s life expectancy has greatly increased. 5 Data source: China Yearbook 2004 6 The Chinese official poverty line is 300 yuan/per person per year in rural areas according to the price of 1990; comparable urban poverty line is not available. 7 The new poverty line is 850 yuan/per person per year in rural areas according to the price of 2002 and 1200 yuan/per person per year in urban areas. Consumption bundle of this poverty line can guarantee people’s demand
for 2100 Calorie each day.
Table 2: FDI and China’s investment
FDI FDI/Total Investment in Exchange Rate b ($, Trillion) Fixed Assets (%) (RMB/100$)
1985 1.658 1.91 293.66
1990 3.487 3.69 478.32
1995 37.521 15.65 835.10
2000 40.715 10.24 827.84
2003 53.505 7.97 827.70
a: Gross Capital Formation, which includes the changes in inventories, is derived from
Expenditure-approach National Accounting.
b: Total Investment in Fixed Assets is derived from Investment in Fixed Assets.
Source: NSB: China Statistic Book, each year.
Table 3: Poverty in China(%):1981-2001
Old poverty line New poverty line 1981 1986 1991 1996 2001 1981 1986 1991 1996 2001 1 H28.62 9.85 11.66 4.20 4.75 64.67 23.50 29.72 13.82 12.49 2Rural PG 6.84 1.92 2.84 1.13 0.81 19.99 5.99 8.52 3.55 3.32 3SPG 2.35 0.52 1.17 0.58 0.19 8.44 2.16 3.43 1.50 1.21
H 0.82 0.22 0.00 0.18 0.00 6.01 3.23 1.66 0.61 0.50
Urban PG 0.22 0.00 0.00 0.07 0.00 1.01 0.46 0.53 0.16 0.16
SPG 0.14 0.00 0.00 0.06 0.00 0.35 0.09 0.38 0.09 0.11
H 23.02 7.49 8.52 2.97 2.96 52.84 18.53 22.16 9.79 7.97
National PG 5.51 1.45 2.08 0.81 0.51 16.17 4.63 6.37 2.52 2.13
SPG 1.90 0.40 0.85 0.42 0.12 6.81 1.65 2.61 1.07 0.80
Note: 1. H: head count index
2. PG: poverty gap index
3. SPG: squared poverty gap index
Source: Ravallion, M. and S. Chen (2004)
Table 4:Headcount Index of Chinese Poverty:1996-2001
1996 1998 2001 Series Name
.. .. 3.94 Poverty Gap index at $1 a day (%)
.. .. 18.44 Poverty Gap index at $2 a day (%)
Head account index, national (% of 6.00 4.60 .. population)
Head count index, rural (% of 7.90 4.60 .. population)
Head count index, urban (% of 2.00 2.00 .. population)
Source: World Bank: World Development Indicators 2004.
On the basis of previous studies (Lin, et al., 1996; Lin, 2003), this paper will
develop a general framework to analyze openness and poverty in developing
8countries, followed by an empirical analysis of China’s experience.
8 Obviously, poverty problems not only exist in developing countries, but also exist in developed countries. But
there is a big difference of the living conditions of poverty population in developing countries and in developed
countries. This paper will focus on globalization and poverty problems in developing countries.
Our main arguments are as follows: the volume and structure of international trade, income distribution and poverty of a country are all endogenously determined by its economic development strategy. Countries adopting the comparative advantage-following (CAF) strategy will be open and achieve rapid economic growth, and create more job opportunities for the low-income people, who rely mainly on their labor as a source of income. As a consequence, the income distribution in these countries can be relatively equal and the poverty can be alleviated gradually. On the contrary, countries adopting the comparative advantage-defying (CAD) strategy to promote capital-intensive industries will inevitably have the opposite results. This is because investments in the priority sectors of CAD strategy create limited job opportunities, excluding most poor people from formal labor markets. Moreover, the firms in the priority sectors are nonviable in open, competitive markets. Those firms’
survival relies on government subsidies and protection from competition from abroad. The economy becomes close. As the investment in the priority sectors requires large amounts of capital, only rich people or people with good government connections have the ability to make the investments. The burden of subsidies to these nonviable firms will ultimately carried by those people with relatively less wealth and power. It will necessarily worsen the income distribution and make it difficult for poor people to get rid of poverty.
The structure of this paper is as follows: the following section II is a brief literature review. Section III provides a general framework for analyzing the relationship between economic development strategy, openness and poverty. Part IV discusses briefly China’s experience. Part V tests the hypothesis with China’s panel
data. Part VI concludes.
II. Literature Review
Globalization and Poverty: Theoretical and Cross-country Empirical Studies
The history of studies on globalization, economic development, income distribution and poverty is as long as that of economics itself. There is a huge amount of literature on these topics. Nissanke and Thorbecke (2004) have conducted an extensive literature review. According to their paper, so far economists have not
reached a consensus on the relationship of openness, globalization and poverty. Economic growth and improving income distribution are essential methods to alleviate poverty. At present, there are two opposite views about the impact of openness and trade on economic development and income distribution. Some economists believe that market opening and participation in globalization will contribute to economic development and improve income distribution. But other economists take the opposite position.
The most recent empirical works fail to reach a conclusion on this issue. Milanovic (2002) and Easterly (2003) argued that the relationship between globalization and income distribution (then poverty) could be expressed as an inverted U curve. These conclusions may be useful in describing the phenomena of the past, but have no applicable policy implications for poverty reduction in the future. For instance, we can hardly reach such a policy suggestion on the basis of Milanovic (2002): Economies with low-income should keep themselves away from openness in order to reduce poverty. Nor can we make the following policy suggestion according to the result of Easterly (2003): Economies with relative low openness should not encourage further openness to avoid deterioration of income distribution.
Obviously, openness and trade are treated as exogenous variables in these analyses. But in our opinion, openness, income distribution and poverty are endogenous to a country’s development strategy, which will be discussed in Section III. The effects of openness and globalization on poverty are determined by the following two factors: (i) the intrinsic characteristic of economic structure before openness and (ii) approach to openness and globalization. For instance, will a country make a large investment on international frontier technologies or just purchase technologies suitable for its economic conditions? Will it develop infrastructures or engage in risky R&D with foreign loans? Will it export products produced with its relative abundant resources, or products produced with its relative scarce resources under government subsidies?
Literature on poverty in China
There have been many studies concerning China’s openness and poverty
separately and simultaneously. Anderson (2004) analyzed Chinese rural economy and rural poverty after accession to the WTO with a GTAP model and found that cutting
down agricultural import protections and removing agricultural subsidies will cause rural residents who rely on agriculture income in a relative deteriorated position, compared with that of urban residents. But if the Chinese government reduces the negative nominal protection rate of rice, meat, vegetables and fruit to zero after accession to the WTO, there would be a smaller wage decrease for rural people. The analysis, meanwhile, emphasized that the reduction of agriculture protection in other developed countries would contribute to the reduction of Chinese rural poverty. The analysis divided rural households into three categories: category A includes households whose share of non-farm income was 0%; categories B and C include households whose such income shares are 30% and 60% respectively. After accession to the WTO, income of households in categories A and B will decrease, while those in category C will increase. Of course, the elimination of negative agricultural protection would make households of category A and B better.
Chen (2004) found that the accession to the WTO would have small impact on average household income, household income inequity and poverty in China, but the impact varied greatly among households with different characteristics. Generally speaking, rural households would be harmed in this process, while urban households could benefit from it. The most vulnerable households are those rely heavily on agricultural income. Hertel (2004) proved that after accession to the WTO, in the long run, if the labor force could move freely between agricultural and non-agricultural sectors, both rural households and urban households would benefit from it, though rural households would benefit less. But in the short run, rural households would experience a small loss because of the restriction of labor mobility.
Xue (2004) and Li (2004) analyzed recent urban poverty problems in China and found that laid-off individuals and unemployment with enterprises’ reform increased
urban poverty; poverty of migrants from rural areas is a important element of urban poverty; and poverty were related to educational level and illness. They suggested increasing employment as a way to solve the problem of urban poverty.
Li and Knight (2004) developed a new classification of poverty, based on permanent income hypothesis. According to current income and consumption level, poverty can be divided into three categories: permanent poverty, selective poverty and temporary poverty. Permanent poverty refers to those people, whose current income and consumption are both below the poverty line. Temporary poverty refers to those people, whose current income is below the poverty line, but consumption is above the
poverty line as their permanent income is higher than their current income. Their current consumption is financed either by savings from the past or by new loans. Selective poverty refers to those people, whose current income is above the poverty line but consumption is below the poverty line. They choose to cut down current consumption because they have to pay for the past consumption or have a high expenditure expectation in the future. According to their survey of 1999, Li and Knight (2004) discovered that the overall Chinese urban poverty rate was 9.4%, of which 29% belonged to permanent poverty, 20% belonged to temporary poverty, and
9the remaining 50% belonged to selective poverty. The reason for high saving
propensity and low expenditure propensity of the selective-poverty families was that most of these families did not have members who were ill or children who were still in school. So their current consumption expenditures were very low.
Analysis of Ravallion and Chen (2004) can be referred to as an authoritative analysis of Chinese poverty. They not only evaluated the change of Chinese poverty rate and poverty depth since 1978, but also provided a theoretical explanation for it. According to the definition of poverty line in their analysis, during the 20 years since 1981, proportion of the Chinese population in poverty has decreased from 53% to 8%, but the speed of reduction slowed down since the end of 1990s. At the same time, the degree of inequity in China was increasing. In addition, the analysis pointed out that inequity has already become a common focus for economic development and poverty reduction.
What’s more, their study has another important finding: “the approach for
economic development can affect poverty reduction”. Although we agree with this
finding, we can not agree with their explanation for it: “if the overall growth rate of
10each industry remains constant since 1981, it will take 10 years rather than 20 years
11for poverty incidence to decrease to 8%.” The reason for our objection is simple: if
the growth rate of secondary industry and tertiary industry were the same as that of 1981 during the past 20 years, the growth of income per capita, then the demand for agriculture products would not have been as high as the actual level. In an extreme situation, in a closed economy, as the demand elasticity of agricultural products to
9 If just adopting the income standard or consumption standard, the poverty rate was 4.6% and 7.5%, respectively. 10 The industries mentioned here refer to primary industry, secondary industry and tertiary industry. 11 There is a methodology issue here. Just as Solow (2001, P.283) said when talking about cross-country regressions of economic convergence: “The Causality issue points to a deeper question: Do cross-country
regressions define a meaningful surface along which countries can move back and forth at will?”
both income and price is very low, the increase of agricultural production may result in a decrease, rather than an increase, of farmers’ income. Although this paper
correctly pointed out that Chinese rural poverty reduction in the past was mainly due to the agricultural growth, it can hardly have correct policy implications for the future rural poverty reduction. Given the current population-land endowment structure, China’s agricultural output growth alone can hardly be a sustainable way for poverty reduction.
The literature on Chinese poverty mentioned above has based on detailed econometric analysis and simulation, although the views may be different. This paper does not intend to judge which poverty line more reasonable or which measurement of poverty more precise. What concerns us most is to construct a general framework for analyzing the relationship between poverty and openness. The literature on Chinese poverty mentioned above fails to do so.
III. Economic Development Strategy, Openness and
12Poverty: A Framework
If we do not take poverty arising from loss of working ability due to illness or natural or man-made disasters into account, the majority of the poor in a developing country is the result of interaction of economic growth and income distribution. Obviously, if the increase of income per capita of an economy is so slow that the per capita income remains under the poverty line, most people will live under poverty no matter how equally the income distribution. The analysis of Dollar and Kraay with data of 80 countries in about 40 years showed that: “Income of the poor rises
one-for-one with overall growth” (Dollar and Kraay 2000). Similarly, if income
difference in an economy is growing larger and larger while its income per capita increases, the problem of poverty will also worsen. So the combination of economic growth and improving income distribution is the basis and sustainable way for solving the problem of poverty. Therefore, it is imperative to find a development approach, which can promote economic growth and improve income distribution
Economists attempt to find such a development approach for a long time. They have put forward many suggestions in different historic periods and under different
12 This section draws heavily on Lin (2003).
contexts. However, economists have become less confident when making new suggestions. In a paper entitled “Growth Strategy”, prepared for the Handbook of
Economic Growth, Rodrik (2004) cited two paragraphs by Harberger in 1985 and 2003. The first one praised highly the power of economic principles in designing development policies, but the second one almost lost confidence in economic principles. Rodrik ended his article as follows: “Rule-of-thumb economics, which has
long dominated thinking on growth policies, can be safely discarded.” This remark
reflects the frustration of economists on development problems.
We believe that the key to understand the relationship among openness, growth and poverty is the government’s economic development strategy. According to Lin
(2003), development strategy can be broadly divided into two mutually exclusive groups: (i) the comparative advantage-defying (CAD) strategy, which attempts to encourage firms ignoring the existing comparative advantages of the economy in their entry/choice of industry/technology; and (ii) the comparative advantage-following (CAF) strategy, which attempts to facilitate the firms’ entry/choice of
13industry/technology according to the economy’s existing comparative advantages.
thWhen development economics started to take shape in the mid 20 century, the
dominant view among development economists was to advise the governments in LDCs to ignore their own comparative advantages and to adopt an inward-looking variant of the CAD strategy, such as the heavy-industry-oriented strategy or the import substitution strategy. Most LDCs’ endowment structure is characterized by
relative labor abundance and capital scarcity, whereas the projects of CAD are capital intensive and are not consistent with the LDCs’ comparative advantages. As a
consequence, firms in the governments’ priority sectors are not viable in an open
competitive market. Therefore, the governments must give these firms various supports by intervening the function of markets. Specific intervention measures include (but are not limited to) the following variants: interest rate suppression in order to lower the investment and operation costs of heavy-industry projects, over-valued domestic currency so as to reduce the costs of importing equipments for heavy industry projects, and creating legal monopolies by allowing the projects to
13The concept of comparative advantage is used most often in trade theories. But the term we use here does not restrict to its conventional meaning. Even in a closed economy, an enterprise’ cost-minimizing input structure will
be determined by the economy’s given factor endowment structure and the market demand. If the relative intensity of scarce resource used in the government’s prioritized industry exceeds its optimal cost-minimizing level, then the government’s development strategy is also a CAD strategy (Lin 2003).