Improving the Investment Climate in China

Transcription

1 Improving the Investment Climate in China David Dollar, Mary Hallward-Driemeier, Anqing Shi, Scott Wallsten, Shuilin Wang, and Lixin Colin Xu March 2003 This report is based on an investment climate survey conducted in 2002 in five Chinese cities (Beijing, Chengdu, Guangzhou, Shanghai, and Tianjin). We gratefully acknowledge financial support from the United Kingdom s Department for International Development (DFID). A grant from DFID supported the collaboration of the Enterprise Survey Organization in this survey and fellowships for two ESO staff, Mr. Yang Yumin and Ms. Li Hui, to visit Washington, DC, for analysis and preparation of this report. This project is part of a larger effort in the World Bank Group to help countries assess their investment climates and to identify reforms that will lead to higher productivity, more efficient investment, and ultimately more job creation and growth.

2 Table of Contents Chapter 1. Investment Climate Matters 1.1 What do we mean by investment climate 1.2 and why does it matter Chapter 2. China s Investment Climate in International Perspective 2.1 International Integration 2.2 Infrastructure 2.3 Regulatory burden, governance, and corruption 2.4 Entry and Exit 2.5 Human Resources, Skills, and Technology Endowment 2.6 Access to Finance 2.7 Conclusion Chapter 3. Measuring the Investment Climate and its Consequences in China 3.1 Introduction 3.2 Regional Variation In Investment Climate International Integration Private Sector Development Domestic Entry and Exit Barrier Labor Flexibility Skills and Technology Financial Services Government Effectiveness A brief Summary: A Score Card for the Cities 3.3 Impact of the Investment Climate Methodology Investment Rates Sales Rates Productivity 3.4 Policy Implications 2

3 ANNEXES 1. Technical appendix 2. References 3

4 Investment Climate at a glance China, India, and Thailand Macro environment China India Thailand / / /1 GNI per capita (US$, PPP) Population, mid year (millions) GDP growth ( and ,avg %) Openness (Imports+Exports/GDP) Private Investment (% GDP) Public Investment (% GDP) FDI inflows (net, % GDP) Micro environment Inputs Labor force education (avg yrs educ, manufact.) Excess labor force, % Suppliers availability (used for main input), median Stock of inventories of inputs (days of production) R&D (% sales) GNI per cap, PPP $ India Thailand China Governance Control of corruption Rule of law Political Stability Number of visits by gvnt officials, avg per year % of senior manager time with gvnt officials Infrastructure Share of firms with own generator,% Excess cost of private electricity, % of public Days to clear imports, longest in last year Cost of shipping 4, % Telephone lines in largest city (per 1000 people) Personal computers (per 1,000 people) Paved roads, % of total Finance Cost of capital (lending interest rate, %) Share of credit from financial institutions, % Credit to private sector (stock, % GDP) PC per 1,000 people India Thailand China Credit to Priv. Sector (% gdp) Entry/Exit and Operation Cost of labor Import duty on capital equipment Median number of days to start a business Number of permits to start business Bankruptcy rate, % of total firms Uncertainty of expectations in sales 7, % India Thailand China Source: WDI, ICU firm surveys 1/ or most recent available year 2 Scale of -2.5 to 2.5. Higher values correspond to better outcomes 3 Average for Thailand 4 Transport cost as share of value of export to US, textiles, Ratio of average wage to average value added, median value 6 On all imports for Thailand 7 % variation of avg expectations 4

5 Chapter 1. Investment Climate Matters During the last decade, major developing countries including China have begun to integrate much more with the global economy. The countries that are aggressively integrating have grown significantly faster than those that are not. In the 1990s, the more rapidly globalizing developing countries (measured in terms of increased trade participation) grew at 5.0 percent per capita, while the rest of the developing world posted negative growth of 1.1 percent. 1 Among the more aggressive globalizers were Brazil, China, Mexico, Philippines, Thailand, and India. That globalizing developing countries are doing well on average is good news. But these averages disguise considerable variation in performance within this group. China has done spectacularly well, and is the unchallenged leader of the pack. The country has doubled its ratio of trade to GDP over the past two decades (to 41 percent of GDP in 1999), and has had per capita GDP growth of nearly 8 percent during Malaysia was another winner: in spite of the temporary income compression due to the Asian crisis, it could still enjoy per capita GDP growth of 3.8 percent during the 1990s. Again, despite the crisis, Thailand s per capita GDP growth in the 1990s averaged 3.8 percent. However, the per capita GDP growth of another relatively aggressive globalizer, Brazil, has only been around 1 percent for ; and growth in the Philippines was only 0.4 percent. India, with per capita GDP growth of 3.3 percent during is in the middle of the pack (figure 1.1). 1 During the same period, the rich countries grew at about 2 percent per capita. 5

6 The implication of these variations is striking. Such differences in growth rates sustained for one or two decades make a huge difference in living standards and the extent of poverty. While China and India had comparable levels of GDP per capita (measured at purchasing power parity) in 1990 (approximately $1400), over the following decade India s per capita income nearly doubled, while China s nearly tripled. Thus, today, China s per capita income is about 50 percent higher than that of India. Together with its faster growth, China has also had significantly faster poverty reduction (figure 1.2). 8 Figure 1.1 Per capita GDP growth rates in globalizing developing countries (average ) China Malaysia India Thailand Mexico Brazil Philippines 6

7 Figure 1.2 Poverty reduction in India and China is closely related to the growth rate Percent per annum ( ) * GDP per capita growth rate India Poverty reduction * India poverty reduction figure is for China The purpose of our paper is to examine some of the reasons for such performance variations across countries. We argue that openness to foreign trade and investment is an important, but not sufficient, condition for sustained GDP growth. For China and other developing countries to do well, good macro and trade policies need to be complemented with a host of other institutional factors and policies that can be classified under the broad heading investment climate. In the next section we define in more detail what we mean by investment climate. Section 2 then briefly reviews some of the macro evidence that shows the importance of investment climate for sustained growth and poverty reduction. While illuminating about the importance of the investment climate, the macro literature does not really provide much specific guidance about what aspects of the investment climate are important and what specific reforms are needed in particular countries. For this reason 7

8 we go down to a more micro level in chapters 2 and 3. Chapter two looks at China in international perspective. The source of information includes surveys conducted by the World Bank with the Enterprise Survey Organization of China s National Statistical Bureau, comparable surveys conducted in other countries, and various cross-country databases. In general, China compares favorably in areas such as macro and political stability, integration into the world market, and infrastructure. Abundance of cheap labor associated with rural-urban migration has been and continues to be a comparative advantage of China. Not everything is rosy, however. The financial sector is not operating efficiently the vast majority of credit has been provided to state-owned enterprises, which often cannot service their debts, and small- and medium enterprises have to rely mainly on retained earning and personal wealth (or parent company financing) to finance their investment. Moreover, China also lags its more developed East Asian neighbors in terms of education level and infrastructure. Another important caveat is that our survey covers five major cities Beijing, Tianjin, Shanghai, Guangzhou, and Chengdu. So, the results should be interpreted as showing something about the investment climate in these cities. There remains an important question about the investment climate in smaller cities and in more interior locations, which we plan to take up in future surveys and analysis. Chapter three compares the investment climate in the five cities (Beijing, Tianjin, Shanghai, Guangzhou, and Chengdu), using the ESO-WB survey of 1500 firms. The main findings are the following: First, investment climate shows large variations across the five cities. In particular, Shanghai is characterized by the best international integration, financial services, good entry and exit conditions and labor market flexibility, 8

9 but has the lowest level of domestic private sector participation. Guangzhou is the best in labor market flexibility, entry and exit fluidity, government efficiency, and private sector participation; good in international integration and financial services; but relatively low in skills and technology. Beijing is largely in the middle of the pack, with no especially strong advantage or disadvantage. Tianjin is good in private sector participation, in the middle of the pack in terms of entry and exit fluidity and labor market flexibility, poor in international integration and government efficiency, and worst in skills and technology and financial services. Chengdu, the only inland city we have surveyed, lags the farthest in almost everything, with the following exceptions: a good level of private sector participation, with reasonable skills and especially technology. Second, the growth potential from improving the investment climate could be quite large. For instance, we consider what gain Chengdu would get from attaining the level of investment climate indicators that we observe in the better cities of Guangzhou or Shanghai. We estimate that in this scenario firm productivity could be increased by about 30 percent and that the investment rate of the typical firm would increase from the 14 percent that we actually observe in the Chengdu sample to about 19 percent. The specific point estimates inevitably have some uncertainty around them, but the general point is that firm productivity, investment, and growth are related to aspects of the investment climate and that addressing weaknesses identified in this analysis should lead to substantially better firm performance. Third, within the categories of investment climate we have found the extent of international integration and the extent of development of the private sector to be especially important. Other issues such as entry 9

10 and exit barriers, labor market flexibility, and financial services are important as well. Each city has specific areas that could be improved. A. What do we mean by investment climate? The quantity and quality of investment flowing into China or any specific region depend upon the returns that investors expect and the uncertainties around those returns. These expectations can be usefully categorized as the following broad yet interrelated components: First, there are a set of macro or country-level issues concerning economic and political stability and national policy towards foreign trade and investment. By these, we generally refer to macroeconomic, fiscal, monetary, and exchange rate policies as well as political stability. As far as these macro indicators go, China performs quite well, as will be documented in chapter two. Second, there is the issue of efficacy of a country s regulatory framework. As far as firms are concerned, these relate to the issues of entry and exit, labor relations and flexibility in labor use, efficiency and transparency of financing and taxation, and efficiency of regulations concerning the environment, safety, health, and other legitimate public interests. The question is not whether to regulate or not, but whether such regulations are designed in incentive compatible ways, avoid adverse selection and moral hazard, serve the public interest, are implemented expeditiously without harassment and corruption, and facilitate efficient outcomes. While such variables are hard to measure, our surveys clearly suggest that regulatory efficacy varies widely across countries and, as far as China is concerned, across provinces. 10

11 Third, and no less important, is the quality and quantity of available physical and financial infrastructure, such as power, transport, telecommunications, and banking and finance; and given the imperfect mobility of skilled workers and the clustering of technology, the endowment of skills and technology. When one surveys entrepreneurs about their problems and bottlenecks, they will often cite infrastructure issues such as power reliability, transport time and cost, and access and efficiency of finance, along with the lack of skilled workers and the difficulty of access to advanced technologies as key determinants of competitiveness and profitability. China s success in the 1990s suggests that it has many positive features in its investment climate, and one objective of our study is to understand what has contributed to China s success which can provide useful lessons to other developing countries. The evidence also shows that there is still room for significant improvement in the investment climate. To be sure, China has been excellent in the first dimension of the investment climate (i.e., macro environment), as characterized by political and macro policy stability. However, some recent changes in the structure of the Chinese economy require further structural reforms. The WTO accession, for instance, requires China to shift from a discretion-based governance system to a rule-based one, which requires the reduction of the role of the government in how firms operate. Financing also should be less favorable to one particular type of ownership. Another challenge has been the migration of rural residents to cities, and the pressure of job creation due to both migration and Xia-Gang workers (i.e., laid-off workers from SOEs). This challenge also imposes demand for regulatory reforms to reduce entry barriers for new, and small- and medium-sized enterprises, which have been shown to be the most important force behind job creation; 11

12 and it also pushes for reforms in financial institutions in order to allow SMEs more ready access for credit. This paper hopes to shed light on what would be some fruitful areas for further reforms. Two caveats are in order at this stage. First, we are not interested in the quantity of investment per se. Indeed, recent work on economic growth [Easterly 1999] has shown that there is surprisingly little relationship between the quantity of investment and the rate of economic growth. In many instances, this is due to a distorted and dysfunctional institutional and policy milieu where neither public nor private investments produce the benefits that they should. Our focus, therefore, is not on the quantity of investment, but on the overall institutional and policy environment the investment climate that determines whether or not investments pay off in terms of greater competitiveness of firms and sustained growth. Second, while we recognize that social infrastructure is no less important than its physical and financial counterparts, we have chosen to exclude from our definition of the investment climate such issues as the provision of basic education and health services. It is a deliberate choice. The reforms needed to improve social services are quite different from the issues of infrastructure and regulation of industry on which we concentrate. B and why does it matter Spurred by the endogenous growth theories of Romer (1986) and Lucas (1986), there is now a vast empirical literature that investigates the determinants of growth. Some of the empirical results are fairly robust and provide macro evidence about the importance of 12

13 the investment climate. Fischer (1993), for example, found that high inflation is bad for growth. This commonsense result is hard to dispute. To an extent, inflation reflects exogenous shocks that are beyond the government s control. But truly high inflation typically reflects serious monetary mismanagement. There is also a clear negative relationship between government consumption and growth, which was first noted by Easterly and Rebelo (1993). No doubt, some government expenditures are socially productive, but developing countries with very high government spending usually have inefficient bureaucracies and high levels of corruption. A number of studies, most recently Frankel and Romer (1999) and Dollar and Kraay (2001), find that openness to trade and direct foreign investment accelerates growth. These findings are in the spirit of the new growth models, and emphasize the importance of market size for creating a finer division of labor and stronger incentives to innovate. In addition to macro and trade policies, financial development is also a catalyst for growth [Levine, Loayza and Beck (2000)]. All else being controlled for, countries that have more developed stock markets and/or deeper banking systems tend to grow faster. Investment climate measures such as the strength of property rights, rule of law, and level of corruption are also well correlated with growth [Kaufmann, Kraay, and Zoido- Lobatón (1999); Knack and Keefer (1995)]. These studies typically use data generated from surveys of private businesses, and reflect the extent to which investors and/or firms perceive problems with harassment, corruption, and inefficient regulation. A problem of these measures, however, is that they are often based on a small sample of very large entrepreneurs and hence do not provide a robust assessment of how rule of law and 13

14 corruption are experienced by small and medium enterprises, which form the backbone of the economy. Thus, the empirical cross-country literature provides evidence that growth and poverty reduction are promoted by a good investment climate an appropriate policy package of private property rights, sound rule of law, macroeconomic stability, government spending that is not excessive and well focused on public goods, and openness to foreign trade and investment. However, most of the macro-indicators of policy and investment climate used in these studies are quite crude, and are of little help to countries in identifying what specifically needs to be done to create a better climate. For instance, the existing cross-country macroeconomic measures are quite similar for China and India (e.g. rankings on rule of law, corruption, or overall infrastructure quality from different international sources) (ICU, World Bank, 2002). Both countries fit the empirical growth studies in that both have done relatively well. India has grown at about twice the rate of the OECD countries in the 1990s. Yet, China has grown much faster and had much greater poverty reduction. Macro-indices fail to explain such differences. Thus, while the macro evidence is useful as background and motivation for the rest of our work, it suggests the need to delve at a much more micro level, and to survey large numbers of producers, including SMEs, to understand the rich differential relationship between investment climate and growth. 14

15 Chapter 2: China s Investment Climate in International Perspective China s solid macroeconomic performance over the past decade is unquestionable. Moreover, this growth remained relatively stable even during the East Asia crisis, when other countries in the region saw large GDP decreases (figure 2.1). Figure 2.1 Growth remains strong through the East Asian crisis Percent GDP growth 15 China 10 Korea, Rep. 5 Indonesia Philippines Malaysia Thailand Other macroeconomic indicators are also consistent with this long-running economic growth. Low inflation, for example, tends to be an important factor in sustainable economic growth, though moderate inflation may not be harmful (Barro, 1997). 2 Inflation in China was relatively low through the 1990s. Moreover, it remained 2 Spurred by the endogenous growth theories of Romer (1986) and Lucas (1986), there is now a vast empirical literature that investigates the relationship between inflation and growth. Fischer (1993), for example, found that high inflation is bad for growth. This commonsense result is hard to dispute. To an extent, inflation reflects exogenous shocks beyond the government s control. But truly high inflation, however, typically reflects serious monetary mismanagement. 15

16 low during the East Asian crisis, unlike other Asian countries which saw higher inflation (figure 2.2). 3 Figure 2.2 Low inflation, even during the East Asian crisis CPI percent change Average China India S. Korea Malaysia Indonesia Thailand Philippines Strong macroeconomic performance, however, masks the underlying components of the investment climate, described in chapter 1, that affect (and are affected by) the macroeconomy. While chapter 3 will explore the microeconomic foundations of China s investment climate, this chapter will compare China s aggregate performance in certain investment climate measures to the rest of the world. Broadly speaking, these factors include international integration, governance, infrastructure, entry and exit of businesses, human resources, and finance. While macroeconomic indicators present only good news, as we will see the investment climate analysis reveals both strengths and serious weaknesses. 3 Some debate these inflation figures, suggesting that enterprises have systematically understated inflation (Young, 2000). 16

17 1. International Integration The strong macroeconomic environment has brought and has been encouraged by China s increased integration in the world economy. Indeed, China s entry into the WTO is a major signal of China s international outlook. The important role that trade plays in promoting productive investment and growth has long been recognized. Using different measures of openness to trade, including both its relative size (as measured by import and export shares) and degree of distortion (as measured by average tariff rates and dispersion), research strongly suggests that greater openness is associated with higher growth in both industrialized and developing nations. Sachs and Warner (1995) find that openness is a highly significant determinant of growth, and combined with property rights may even represent sufficient conditions for growth in poor economies. Kang and Sawada (2000) find a similar effect of openness on growth, arguing that combined with financial development it increases growth rates in developing economies by decreasing the cost of human capital investment. 4 Foreign direct investment (FDI) and trade are good indicators of this integration. 5 Net FDI increased dramatically in the last decade, from $2.7 billion in 1990 to $37 billion in This growth is impressive, and as a share of GDP, net FDI into China is higher than many other Asian countries, though not as high as some in Latin America (figure 4 Rodriguez and Rodrik (1999) argue that some indicators of openness are highly correlated with other indicators of economic performance, including macroeconomic policy, or that they imperfectly reflect a country s trade policy regime. The high correlation of components of the Sachs and Warner index with policy and institutional variables yields an upward bias in the estimation of trade restriction effects, while tariff and non-tariff barriers, the two variables that directly measure trade openness, have little explanatory power when considered separately in cross-country regression studies. 5 A number of studies, most recently Frankel and Romer (1999) and Dollar and Kraay (2002), find that openness to trade and direct foreign investment accelerates growth. These findings are in the spirit of the new growth models, and emphasize the importance of market size for creating a finer division of labor and stronger incentives to innovate. 17

18 2.3). FDI as a share of GDP has come down somewhat in the second half of the 1990s (figure 2.4), but in absolute terms China is still the largest recipient in the developing world. In regards to trade, meanwhile, China has reduced tariff rates to about one-third of what they were two decades ago: from 49.5 percent in 1982 to 16.8 percent by 1998 (World Bank 2002). Partly as a result, trade increased from 15 percent of GDP in 1980 to nearly 50 percent of (a much larger) GDP by Imports increased from about $US 36 billion in 1980 to $US 192 billion in 2000 (in constant 1995 US dollars). Likewise, exports increased from $US 27 billion in 1980 to $US 239 billion in Figure 2.3 FDI as a share of GDP, China India S. Korea Malaysia Thailand Philippines Brazil Argentina -2 Indonesia 18

19 Figure 2.4 FDI as a share of GDP in China is high, but declining Infrastructure One component of the investment climate is a country s infrastructure. Businesses in countries with poor infrastructure must spend more effort to, for example, acquire information, receive inputs, and get their products to market. The costs of poor infrastructure can easily undermine any cost-competitiveness that a business would otherwise have, at best making it more costly for a given firm to operate and at worst deterring entry in the first place by making it uneconomic. Infrastructure in China appears to do relatively well compared to other developing countries. Two separate surveys suggest that businesses do not perceive infrastructure to be a tremendous obstacle. Figure 2.5 reports the share of firms in the World Business Environment Survey reporting that infrastructure posed no obstacle for the operation and growth of [their] business. Approximately 42 percent of Chinese firms responded that infrastructure posed no obstacle, much better than Brazil or India, but about average 19

20 for East and South Asia. Figure 2.6, meanwhile, reports the average responses of business executives surveyed in the World Economic Forum s Global Competitiveness Report (GCR), which asks respondents to rate the overall infrastructure quality of their country on a scale of 1 to 7, with one being the worst ranking and seven the best. 6 While China ranks slightly ahead of India, it falls below Brazil, Malaysia, and Thailand. Figure 2.5 Share of firms reporting that infrastructure was no obstacle to doing business Thailand India Brazil East Asia China South Asia Malaysia OECD Source: World Business Environment Survey 6 There are serious problems with simply averaging responses. For example, it assumes that a response of 2, for example, means the same thing for each respondent, a notion that research on surveys contradicts (see, for example (Recanatini, et al., 2000). At least one study has criticized the Global Competitiveness Report on the grounds that it is methodologically flawed and theoretically weak (Lall, 2001). 20

21 Figure 2.6 Overall quality of infrastructure, average response (1=poorly developed and inefficient, 7= among the best in the world) India China Brazil Thailand Malaysia OECD Source: World Economic Forum, 2002 These two figures are illustrative, but also show the limitations of such surveys: for example, the GCR suggests that infrastructure, overall, is better in Brazil than in China, while the WBES suggests that fewer firms in China believe that infrastructure is an obstacle to their doing business. Several factors could be driving this seemingly anomalous result. First, neither survey is random or representative, making it difficult to compare one country to another. Second, firms surveyed in Brazil could be more technologically advanced than those in China, meaning that Brazilian firms might demand higher quality infrastructure. If so, infrastructure quality in Brazil could be higher than in China, but still pose a relatively greater obstacles to firms there. Because of the imprecision inherent in such surveys, it is worth exploring more factual indicators. Below we compare data on shipping costs and ports, telecommunications development, and power provision. 21

22 Shipping and ports China is relatively well integrated into the world market through efficient shipping and relatively well functioning ports. Data from U.S. customs concerning all of the containerized shipments into U.S. ports allow for a comparison of the shipping costs for a particular product such as textiles from different ports around the world. China has a large shipping cost advantage over India and significant cost advantages over Thailand, Indonesia, and Brazil (figure 2.7). Figure 2.7 Shipping cost advantages Cost advantage compared to India 50% 40% Textiles E. Coast W. Coast USA 30% 20% 10% 0% -10% Thailand Indonesia China S. Korea Brazil This reflects the large volume of exports from China as well as relatively efficiency of its ports. On the related issue of bottlenecks in customs clearance, China is good compared to India and a bit behind Asian neighbors such as Korea and Thailand. In large samples of manufacturing firms the average clearance time for imported inputs was 8 days in China, compared to 11 for India, and 7 for Korea and Thailand. In more recent surveys we have been asking firms their worst customs experience in the past six months, which 22

23 highlights whether firms operate in an uncertain environment. In the Indian sample the typical firms worst experience was a three-week delay, compared to only 9 days in Shanghai (figure 2.8). Figure 2.8 Days to clear imported inputs through customs Last time Longest India China Korea Thailand 0 India Shanghai c The reasonably good port and customs facilities help Chinese firms connect effectively to the world market. Telecommunications Telecommunications is an increasingly important component of a nation s infrastructure. There has been a steady increase in the number of telephones per capita in China, from less than one telephone line per hundred people in 1990 to nearly 14 by 2001, though the telephone penetration rate is still lower than many other developing countries. Among seven Asian countries, China has fewer telephone lines per capita than 23

24 South Korea and Malaysia, but more than India, Thailand, Indonesia, and the Philippines (figure 2.9). Figure 2.9 Telephones per hundred people Korea Malaysia China Brazil Thailand India Source: International Telecommunications Union The number of telephone mainlines is only a part of the telecommunications infrastructure. Much telecommunications growth in developing countries now comes from mobile telephone providers, and China is no exception. By 2001 China had more mobile telephone subscribers than any country in the world. Of course, China is also the most populous country in the world, and mobile penetration has exploded throughout the world, so the number of mobile phones per capita paints a slightly different picture. By 2001 China boasted just over 11 mobile subscribers per hundred people better than India or Indonesia, but trailing far behind Korea, Malaysia, and the Philippines (figure 2.10). 24

25 Figure 2.10 Mobile phone users per hundred people, S. Korea 40 Malaysia 20 China Philippines 0 India Indonesia Source: International Telecommunications Union Power Access to reliable power at a reasonable cost is a prime concern for most manufacturing firms. Figure 2.11 shows electricity generation capacity in China and selected countries. While generation has clearly been growing, generation capacity has only just managed to stay above population growth. Moreover, China lags several countries in this measure. Thailand, Brazil, and Malaysia all have more generation capacity per capita than does China, though the Philippines and two states in India for which there are data lag behind. China has a number of big projects in the pipeline that will boost generating capacity substantially. 25

26 Figure 2.11 Electrical generation capacity MW/thousand people) 1 Korea Malaysia Brazil Thailand China Philippines Punjab (India) Orissa (India) Source: US Energy Information Agency Generation capacity, of course, is only part of the story. Reliable power means electricity delivered with few interruptions. Low prices for electricity from the grid are not helpful to firms if electricity is frequently unavailable. Our surveys of firms contain some illuminating information. For example, 30 percent of firms in China have backup generators compared to 69 percent in India, where electricity is less reliable. Chinese firms report, on average, losing about 2 percent of their output as a result of power problems, compared to 6 percent in Pakistan. Regulatory burden, governance, and corruption A country s general governance structure and business-government interactions are an important component of the investment climate. Broadly speaking, these include the burden firms face in complying with regulations, the quality of the services provided by these regulations and the extent to which corruption is associated with the procurement of such services. Five composite indicators from Kaufman, Kraay, and 26

27 Zoido-Lobaton (1999; 2002) capture the dimensions of interest. 7 These categories include: Government effectiveness measures bureaucratic delays, competence of officials, the quality of public service delivery and the independence of the civil service from political pressures. This grouping of indicators covers the elements needed for the government to design and implement good policies. Regulatory burden includes the number of regulations within a market, the number of markets that are regulated, competition policy measures, and price controls. This captures more of the outcomes of the policies and provides a sense of how market friendly the business environment is. Rule of law captures the extent of crime, property rights, tax evasion and the legal system s effectiveness. It indicates the enforceability of contracts and the predictability of rules. Corruption measures include the frequency and size of irregular payments. 7 These governance indices are the compilations of up to 60 sub-indices from various sources. Some qualifications should be noted when using these subjective measures to compare competitive environments. They revolve around the issue of the yardstick used in judging a country s performance. First, the point of reference itself can vary across country. In one study, Korean entrepreneurs rankings of problems were consistently higher than those of Indonesia. Few would believe the business environment in Indonesia to be superior to that of Korea. Rather, a more plausible explanation is that the Korean entrepreneurs had relatively high expectations of services their government should provide and shortfalls were readily registered. In contrast, if there are few expectations that a service will be performed, its failure is not seen as particularly problematic. Controlling for differences in country mean responses offers at least a partial solution. Another shortcoming of these measures is that they register very little change over time, and do not necessarily reflect that substantial improvements have been made in the substantive area over time. Again, this may be due to changes in the subjective yardstick against which they are being measured. For example, as the judicial system is reformed, expectations will rise as to what role the courts should play. Failure to meet expectations could lead to low rankings, even as the system improves These subjective rankings are more useful for making comparisons within a single country than for attaching great significance to the international ranking. Thus, one can look among Indian entrepreneurs to see the relative rankings of problems, for example infrastructure versus corruption, rather than looking at perceptions of corruption between India and its neighbors. Unfortunately these indictors are rarely used in this fashion. 27

28 Political instability and violence measures the incidence of coups, assassinations, riots, armed conflicts and provides a measure of the likelihood of a violent overthrow of a governing party. The individual measures of governance are plotted on governance pentagons for China and selected countries (figure 2.12). Figure 2.12 Governance pentagons China Political stability India Political stability Corruption Government effectiveness Corruption Government effectiveness Rule of law Regulations Rule of law Regulations Thailand Political stability Brazil Political stability Corruption Government effectiveness Indonesia Political stability Corruption Government effectiveness Rule of Law Regulations Corruption Government effectiveness Rule of Law Regulations Rule of Law Regulations Source: Kaufman, Kraay and Zoido-Laboton, 1999 The outer web indicates the best performance, and the inner web the median measure for the 174 countries for which measures were available. Thus, the larger the country s web, the better its governance is measured to be. China scores very high on political stability, near the median for government effectiveness, regulations, and rule of law, but below the median for corruption. 28

29 Some additional measures support the information presented in the pentagons. There is both good, bad, and conflicting news on the regulatory burden facing firms in China. The World Business Environment Survey, for example, found that managers in China spend about nine percent of their time dealing with public officials, far less than in India, and about the same as in transition European countries, Latin America, and even OECD countries (figure 2.13). Figure 2.13 Management time spent dealing with public officials on regulations, administration Percent of management time China India Transitional Europe Source: World Business Environment Survey (WBES) 2000 The World Bank Group LAC OECD The Global Competitiveness Report, on the other hand, ranked China significantly worse than the median on the same measure, and far worse than India. In terms of corruption, the Global Competitiveness Report ranked China among the worst in terms of business costs of corruption, worse even than, for example, Thailand and India. Likewise, China was worse than the median country on the frequency of irregular payments by firms, though in this case Thailand and India scored even more poorly (figure 2.14). 29

30 Figure 2.14 Corruption measures Ranking out of 75 countries surveyed in the Global Competitiveness Report, 2001 Business costs of corruption Time senior management spends with government agencies/regulators India China Frequency of irregular payments Strongest Weakest Entry and Exit Free entry and exit of firms is important to a healthy economy. Barriers to entry and exit block new, more productive and innovative firms from emerging. A growing body of literature documents the difficulty entrepreneurs face in establishing firms in developing countries (e.g., Djankov, et al., 2002; Emery, et al., 2000; Friedman, et al., 2000). Djankov, et al. (2002) compiled data on entry regulations in 85 countries, and discovered enormous variation in the number of procedures required to start firms across countries, ranging from a low of two in Canada, to as many as 21 in the Dominican Republic (with Bolivia and Russia a close second at 20). Likewise, the time required to establish a firm ranged from two to 152 business days. These procedures can be extremely costly to the economy the cost of official procedures (that is, not including bribes) for setting up a new business was 266 percent of per capita income in Bolivia. They find that stricter regulation of entry is correlated with more corruption and a larger informal economy. 30

31 Figure 2.15 Number of procedures and days to complete them to start a new business Number of procedures 16 Number of procedures Days to complete Days Source: Djankov 2000 China Malaysia Thailand India S. Korea Philippines 85 country average 0 Using Djankov s measure of business entry, China required 12 procedures to start a firm, more than the sample average of 10, and far more time was required to complete them at 92 days, versus a sample average of 47 days. These estimates suggest that it takes more time to start a business in China than in, for example, India, Thailand, or Malaysia (figure 2.15). These data, however, are somewhat contradicted by survey responses in the Global Competitiveness Report (figure 2.16). This measure places China at worse than the median in terms of the number of permits to start a business which they estimate at six but also suggests that the median number of days to start a business is only 30. In this case China does better than the median, and much better than India, which is near the bottom. Another way to determine the difficulty of entry and exit is to examine outcomes that are associated with these factors. In particular, entry and exit are crucial to a competitive economy, and one of the most telling indicators of whether 31

32 markets are competitive in a country is the productivity dispersion of firms within an industry. In a competitive market, with reasonably free entry and exit, dispersion should be low as unproductive firms either improve or leave the market. Higher dispersion indicates that less efficient producers are not being forced to improve their productivity or exit the market. Firm-level studies in a number of countries bear this out. 8 Subsidies or strict regulations that impede entry or exit can ultimately bolster high cost producers. When such firms remain in the market, more productive firms may not have adequate incentives or the ability to increase productivity or to grow. However, as competition increases, firms face greater incentives to innovate and greater penalties for failure to do so. Figure 2.16 Entry and exit Ranking out of 75 countries surveyed in the Global Competitiveness Report, 2001 (6 permits) China Number of permits to start a firm (10 permits) India Median number of days to start a firm Administrative burden for startups Hiring and firing of workers (30 days) (90 days) Strongest Weakest Loss of protection and greater competition from foreign firms can drive inefficient domestic producers to better exploit scale economies, eliminate waste, reduce managerial slack, adopt better technologies, or shut down. As a result, productivity dispersion should shrink as productivity levels rise in the face of greater competition. 8 See Hallward-Driemeier (2001) for a longer discussion; Levinsohn (1993), Haddad and Harrison (1994). 32

33 Productivity dispersion a measure of inefficiency tends to be associated with barriers to competition, such as administrative barriers to start a business or trade barriers to competition for India, China, Philippines, Thailand, Malaysia and Korea (figure 2.17). Figure 2.17 Dispersion of value added by worker Ratio of 75th percentile to 25th 8 6 India China Philippines Thailand Malaysia Korea Garments Electronics In Chinese garments and electronics, the higher performers have value added per worker that is 5 times that of lower performers. In electronics, productivity dispersion is lower in four East Asian countries where the World Bank has conducted similar surveys, while in garments only India performs worse than China on this measure. In Malaysia and Thailand the ratios are just below 3, and in Korea not much more than 2. Thus, more competitive countries in the group (as proxied by weeks to start a business and average tariff rates) have lower levels of productivity dispersion than do the less competitive countries. 33

34 Human Resources, Skills, and Technology Endowment The availability of inputs is a crucial component of the investment climate. In terms of human resources, this does not mean simply an abundant labor force which China certainly possesses but also skill levels and technological know-how. In this section we explore how educated the Chinese labor force is compared to other countries at the most basic level and in terms of higher education. We also explore measures of technological know-how and endowments in the country. Close to 16 percent of the Chinese population is illiterate. This is considerably better than India, and only slightly worse than Malaysia and Indonesia, but is much worse than the Philippines, Thailand, or South Korea (figure 2.18). Figure 2.18 Illiteracy and school enrollment, China India Indonesia Korea, Rep. Malaysia Philippines* Thailand Illiteracy rate Secondary enrollment Tertiary enrollment** *Philippines data is for Source: UNESCO, through SIMA ** No data for India and Indonesia Similarly, nearly 63 percent of the relevant population in China is enrolled in secondary school, better than the 50 percent in India, but below the other countries in the sample. 34

35 Even worse, China s tertiary enrollment rate is only 7.5 percent, which is extremely low compared to other Asian countries. Nonetheless, these relatively low figures still represent improvements over time. China s educational infrastructure, especially with respect to science and technology, has developed mostly since the 1978 reforms. Since then China has rapidly built higher educational institutions and sent tens of thousands of students to study abroad. Postgraduate enrollment in Chinese institutions has increased from nearly nothing prior to 1978 to more than 300,000 by the year Some recent research has begun to account for this vast increase in human capital, finding that it has contributed significantly to growth and welfare (Wang and Yao, 2001). An ongoing concern in China, however, is the large number of students who leave to study abroad and then do not return. In response to this concern in 1992 the government began to send delegations abroad to entice Chinese degree holders to return (Guo, 2000). While it is true that increasing numbers of students leave to study abroad, it is also true that an increasing number of students are returning to China once they have their degree (see figure 2.19). 35

36 Figure 2.19 Chinese students abroad 40,000 Number students studying abroad 30,000 20,000 10,000 Number students returned Source: Chinese Statistical Yearbook Data compiled by the U.S. National Science Foundation (NSF) sheds additional light on this phenomenon. The NSF collects information on, among other things, students from outside the US who received science and engineering PhDs from US institutions. The data includes numbers of students and those with firm plans to stay, where firm plans means the students have job offers in the United States. 36

37 Figure 2.20 Number of science and engineering PhDs received in the US, by citizenship 3,000 2,000 China India 1,000 Taiwan, China Korea Brazil Source: US National Science Foundation Figure 2.20 shows the number of students from China and other countries that received science and engineering PhDs in the United States in the 1990s. The figure shows a steadily increasing trend for China (except in 1997), which has far better representation than any other country, while the number of students from other countries remained fairly flat. The share of students from China that plan to stay in the US is exceeded only by students from India (figure 2.21). Moreover, the share of students with plans to stay remained relatively stable until 1995, when it began to increase. Whether this trend continued with the contraction of the US high-tech economy is unclear. 37

38 Figure 2.21 Share of science and engineering PhDs with firm plans to stay in U.S. 60 India China 40 Taiwan, China 20 Korea Brazil Some see the increasing numbers of students abroad and large share of students staying in the US as evidence of brain drain. On the other hand, ever increasing numbers of students are returning to China after studying abroad, suggesting more of a brain circulation. There is likely some truth to both sides. New research, however, is beginning to document gains to developing countries that emerge over a long period of time when large numbers of students at first remain abroad but then gradually return. Saxenian (1999; 2000) found that large numbers of people from India and Taiwan province came to the US as students and remained to work in Silicon Valley, learning US high-tech business and management models. Eventually, many of them began to return home, bringing those skills and education with them. Those re-patriats are largely responsible for new high-tech firms and success of new technology areas. Moreover, there is increasing evidence of a similar phenomenon beginning to take place in China. 38