China's World Trade Compliance

Thank you, members of the Commission, for the honor of addressing such a distinguished and thoughtful panel. Industrial subsidies in China derive from governmental dominance of the economy and from various factors including the central, provincial and municipal governments' strategic goals, patronage, and corruption. The subsidies include direct and indirect components that affect both the top and bottom lines of industrial operations. My statement stems from research that I have conducted over the last 7 years on business in China, some of which has been published in my book, The Chinese Tao of Business: the Logic of Successful Business Strategy (John Wiley & Sons). Answers to specific concerns follow.

In what form and in what industries do subsidies exist in China?

State subsidies flow into State-Owned Enterprises (SOEs) although some well-connected private firms also benefit from indirect subsidies such as Special Market Information. In 2002, the state controlled half the industrial output and SOEs still account for 35% of urban employment. Almost all of China's heavy industry and much of its technology lies in governmental hands. The government controls about a third of China's economy through SOEs in key sectors such as defense and utilities. The State Owned Assets Supervision and Administration Committee (SASAC) directly manages the top 190 or so SOEs, the biggest of which have international stock-market listings.

Subsidies exist in all industries that the Chinese state and provincial governments considered economically or militarily strategic, including Resource Extraction, Steel, Computing, Software, R & D, Environmental Services and Conservation, and Autos.

The subsidies exist in various forms, including:

1. Free to Low-cost Loans. The government exercises a vice-like grip on banks, stock markets and bond issuance and these translate to the ability to make grandiose loans. The most extreme statistics in the financial sector deal with loans outstanding. In three years from 2002 to 2004, loans increased by 58 per cent, or $785 billion. In 2003, new lending equaled almost one quarter of gross domestic product (GDP). A credit binge fueled this latest boom. Half of all bank loans go to SOEs. Most of these loans will never be repaid. Huawei for example, has a $10 billion credit line from China Development Bank.
2. Asset Injections: The SOEs' parent companies, usually municipal governments or ministries, provide their prot'g's with opportunities to acquire state-run businesses, such as toll bridges, at highly preferential terms.
3. No Break-even: Poor bookkeeping practices, and lax bottom-line considerations, grant SOEs freedom from the need to make profits, or to break even.
4. Subsidized Purchases: SOEs can purchase their components and raw materials below cost and directly from each other, affecting the competitiveness of certain sectors in the global economy. This tradition propelled the Chinese motorcycle industry's ability to buy control of virtually all Indian motorcycle companies short of Bajaj and turn them into assemblers of Chinese components.
5. International Bargaining Power: Beijing has used its enormous buying power to intercede for its SOEs with foreign suppliers and to reduce acquisition costs for raw materials. A recent example includes the Chinese government's aborted attempt to bully down the cost of iron ore for the Chinese steel industry below internationally negotiated price levels. The Chinese government has also secured contracts and exploration rights abroad for its SOEs.
6. Labor Controls: The government exercises various methods to control employees including the dang'an or employment dossier; and to reduce labor costs through injection of part-time and migrant workers and the use of prison labor.
7. Tax Breaks: Many SOEs avoid taxation or reduce it through tax breaks (although this can backfire if a company's management loses favor).
8. Energy and Land Subsidies: The state subsidizes gasoline and electricity. Currently, Beijing tightly controls the price of both gasoline and electricity at well below their true economic levels. The state also offers free land and utilities to SOEs and companies in key strategic sectors.
9. Sectoral Credit Allocation: The Chinese economy speeds up or slows down on a sector-by-sector basis on credit allocations by Beijing. Some sectors such as automotive, steel, ethylene and metals' smelting have come off the boil. Others sectors such as coal, railways and utilities are still getting huge infusions of policy-mandated credit. Very high levels of bureaucratic interference characterize credit allocations and industrial-project approvals in China and the state banking system does not allow the market to price capital.
10. Stock Listings: SOEs and Collectives form over 93% of the listing of approximately 1300 companies on China's Shanghai and Shenzhen Stock Exchanges. Provincial governments pressure government regulators to discriminate against private companies and give the precious slots to their ailing state dinosaurs. Indeed, private companies without state connections cannot obtain a listing on any Chinese stock exchange.
11. Cheap Technology: China runs a deficit on its technology trade with the rest of the world and foreign multinational companies control 80% of technological imports and exports in China. The Chinese have made little progress in either basic research or advanced design in vital industries. Despite this institutional flaw, SOEs such as Huawei owe much of their success to lax laws governing the theft of intellectual property.
12. Control over Distribution Channels: Provincial and municipal governments control distribution channels to allocate and to manage market share, to protect favored industries from competition and to shape investment patterns. Regulations on distribution incorporate considerable ambiguities leading to both legitimate differences in interpretation and considerable legal efforts to find loopholes. Central and provincial governments routinely use this ambiguity to confer privileges on favored companies or industries, and to withhold normal rights from companies or industries as a form of protectionism. Administrative guidance from various and competing sources can override the basic laws or regulations either explicitly or unofficially. Provincial or municipal governments may interfere with the national limits on distribution by their generosity (to lure investment or to meet local goals) or restrictions (to protect local interests). Guanxi with local army officials assumes particular importance for distribution. Some estimates suggest that the Peoples' Liberation Army (PLA) controls distribution of goods for up to about 80% of the Chinese population. Its control over manufacturing facilities also makes the PLA China's largest and most diversified manufacturer of industrial and consumer goods.
13.  Special Market Information: Relevant information for strategic decisions comes at a premium price in China and often includes what we in the U.S.A. would consider Insider Information. In China, the central government deliberately controls and disseminates information that it considers of strategic importance. When restrictions on distribution insulate foreign or Chinese companies from their customers, they also cannot undertake direct market research and have to rely on less-sophisticated surrogates. For example, General Motors' (GM) interns in Beijing have scoured the capital's streets to find out who is buying their cars after the intermediaries get them, so that GM can build guanxi with the buyers.

What complications, if any, in evaluating such subsidies are caused by the lack of transparency in China's government and economic systems?

Lack of transparency affects ability to monitor all forms of subsidy except perhaps Stock Listings. Opacity serves as a tax which:

1. Reduces ability to determine the true efficiency and productivity of China's labor and results in potentially sub-optimal foreign direct investment (FDI) decisions until after commitments are made. Consequently, our research has shown that FDI enjoys higher ROIs and ROEs across entire industrial sectors in India against China, including Capital Goods; Food Beverage and Tobacco; Materials; Pharmaceuticals and Biotech; and, Software and Services.
2. Reduces the ability of U.S. domestic producers to prove dumping, especially as so many of those affected are Small and Medium-sized Enterprises with limited resources.
3. Magnifies the weakness of China's statistical system which depends too much on reporting and too little on sampling; the statistical system shows a systematic bias to over-report growth at the bottom of the economic cycle and under-report it at the top, ie, to flatten out a much more volatile economic cycle. Recently, some foreign companies have started constructing their own physical-activity indices of everything from freight-barge traffic to power consumption and air miles flown to find true economic indicators, but the enormous expense constrains companies from doing this well.
4. Reduces the credibility of the SOEs' books. For example, in 2003, China's top 500 SOEs reported revenues of 4.07 trillion Yuan up 25% from the previous year; and profits of 334 billion yuan, up 33% from the previous year. Only 87 of the 500 reported making losses. Unfortunately, outright fraud aside, most SOEs' managers do not know their real profits and tell their supervisors what they want to hear.

Unreliability in macroeconomic data also seriously compounds the problem of estimating the effects of subsidies.

A. For example, in February 2002, the Chinese government said that China's GDP had grown by 7.3% in 2001, making it the world's fastest-growing economy. However, growth rates reported by individual provinces told another story. Only one, Yunnan, said its product had grown slower than the national rate. Taken together, the provincial figures produced a national growth rate nearly two points higher than the official rate! The National Statistics Bureau (NSB) conducts sample surveys and uses these to estimate the country's GDP and growth rate. The results have invariably disagreed with provincial figures. In 1995, the GDP growth rate suggested by provincial data averaged three percentage points higher than the figure of 10.5% produced by sample surveys. Opinions vary as regards the accuracy of the central government's estimates. However, in China, few scholars publicly attempt any detailed justification of alternative figures because of political sensitivity.

B. China's NSB also lacks the capacity to collect data outside normal information channels and lower-level officials interfere with its surveys. The numbers generated by provincial governments remain an important criterion in evaluating local officials' performance, creating an incentive for statistical falsification. The pressure to exaggerate statistics grew in the late 1990s as Chinese officials sought to pump up the economy to stave off the Asian economic slump's effects. Beijing declared that the country had to grow at least 7% a year to create jobs and to forestall social unrest. Not surprisingly, reported growth rates have not dipped below that level since.

C. Officials may also routinely underreport other sensitive data such as debt numbers, unemployment or even FDI to avoid tax payments and governmental scrutiny. The central government's methods at ascertaining the validity of data, a process it calls yasuo shuifen or 'squeezing the water', involves sample surveys, price-index adjustments and plenty of guesswork.

D. Technical difficulties, such as staff reductions among statistical analysts, have enhanced errors in data. No comprehensive measures exist for the size of the fast-growing private-business and service sectors or even for what constitutes FDI.

E. The Chinese government strictly controls economic and industrial data and even classifies some as state secrets. Routinely, Beijing has overvalued SOEs' stocks of unsold goods, and underestimated inflation. Other provinces underreport growth and activity: for example, Zhegiang province in Eastern China may have underreported growth to conceal the rapid development of private companies in its economy. Additionally, affluent provinces, such as Guangdong in Southern China, may have underreported growth to avoid paying more taxes to the central government. However, without more systematic data, economists cannot definitively state if these factors pushed up growth or even occurred.

F. Governmental officials downplay unemployment figures to mask the suffering that economic reforms and restructuring have caused. The official unemployment rate of 3.6% in 2001 excluded xiagang workers (laborers receiving small, monthly stipends from former companies and not counted as unemployed) that economists estimate to number about 10 million. The official rate also excluded farmers who left their fields to work in cities, a floating population of around 150 million unemployed migrants. Using international standards, China's unemployment rate in 2001 approximated 7.6 percent in rural areas and more than 8.5% in the cities, well above Beijing's red-flagged figure to indicate inevitable social turmoil.

G. Most disturbingly, the central government's debt numbers look highly erroneous. The Central Bank's governor, Dai Xianlong, confessed to Parliament in April 2002 that national domestic debt appeared much higher than the official numbers (16% of GDP) suggested. Dai said the figure appeared closer to 60% of GDP if one considered unfunded state-pensions' liabilities, local governments' debts, and major banks' nonperforming loans (NPLs). Dai's unusual candor may mask more bad news. Independent economists have discovered that Dai's statistics drew on China's yearbook GDP growth statistics. Debt more realistically appears closer to 100 or 125 percent of GDP. The Bank of China reported two different figures for its NPLs in 1999, one using Chinese accounting standards, another Western; the latter looms 2.6 times greater than the former. Moody's has openly called the books of China's 'Big Four' banks, 'meaningless.'

What subsidies will be utilized for the 11^th Five-Year Program Period?

I anticipate that all the subsidies that I identified will continue. The 11^th 5-year plan specifically identifies certain strategically important industries that will receive state subsidies. These include:

1. Integrated circuits and software including technology for 90-nanometer and smaller integrated circuits;
2. New-generation networks including digital TV networks and mobile communication;
3. Advanced computing including technology for petaflop computer systems;
4. Biomedicine including commercial production of vaccines;
5. Civil airplane including general purpose planes and helicopters;
6. Satellite applications including meteorological, oceanographic, navigation positioning and telecommunication satellites; and
7. New materials including high-performance materials in information biological and aerospace industries.

Researchers may have more difficulties monitoring the rate of subsidization as China's 11^th Five-Year Plan has only two numeric targets: per capita GDP in 2010 must be double the 2000 figure and 'each work unit must cut its use of energy by 20% of current levels by 2010.' The plan fails to mention raising the price of electricity and gasoline, and unlike the previous 10 years, sets no economic growth targets.

In Part Two, Prof. Haley covers how profitable and available those subsidies are and how profitable companies are that serve the China market.

(The entire U.S.-China Commission hearing record for April 4, 2006 is available at www.uscc.gov/hearings/2006hearings/transcripts/april_4/06_04_04_trans.pdf.)

Thank you, members of the Commission, for the honor of addressing such a distinguished and thoughtful panel. Industrial subsidies in China derive from governmental dominance of the economy and from various factors including the central, provincial and municipal governments' strategic goals, patronage, and corruption. The subsidies include direct and indirect components that affect both the top and bottom lines of industrial operations. My statement stems from research that I have conducted over the last 7 years on business in China, some of which has been published in my book, The Chinese Tao of Business: the Logic of Successful Business Strategy (John Wiley & Sons). Answers to specific concerns follow.

In what form and in what industries do subsidies exist in China?

State subsidies flow into State-Owned Enterprises (SOEs) although some well-connected private firms also benefit from indirect subsidies such as Special Market Information. In 2002, the state controlled half the industrial output and SOEs still account for 35% of urban employment. Almost all of China's heavy industry and much of its technology lies in governmental hands. The government controls about a third of China's economy through SOEs in key sectors such as defense and utilities. The State Owned Assets Supervision and Administration Committee (SASAC) directly manages the top 190 or so SOEs, the biggest of which have international stock-market listings.

Subsidies exist in all industries that the Chinese state and provincial governments considered economically or militarily strategic, including Resource Extraction, Steel, Computing, Software, R & D, Environmental Services and Conservation, and Autos.

The subsidies exist in various forms, including:

1. Free to Low-cost Loans. The government exercises a vice-like grip on banks, stock markets and bond issuance and these translate to the ability to make grandiose loans. The most extreme statistics in the financial sector deal with loans outstanding. In three years from 2002 to 2004, loans increased by 58 per cent, or $785 billion. In 2003, new lending equaled almost one quarter of gross domestic product (GDP). A credit binge fueled this latest boom. Half of all bank loans go to SOEs. Most of these loans will never be repaid. Huawei for example, has a $10 billion credit line from China Development Bank.
2. Asset Injections: The SOEs' parent companies, usually municipal governments or ministries, provide their prot'g's with opportunities to acquire state-run businesses, such as toll bridges, at highly preferential terms.
3. No Break-even: Poor bookkeeping practices, and lax bottom-line considerations, grant SOEs freedom from the need to make profits, or to break even.
4. Subsidized Purchases: SOEs can purchase their components and raw materials below cost and directly from each other, affecting the competitiveness of certain sectors in the global economy. This tradition propelled the Chinese motorcycle industry's ability to buy control of virtually all Indian motorcycle companies short of Bajaj and turn them into assemblers of Chinese components.
5. International Bargaining Power: Beijing has used its enormous buying power to intercede for its SOEs with foreign suppliers and to reduce acquisition costs for raw materials. A recent example includes the Chinese government's aborted attempt to bully down the cost of iron ore for the Chinese steel industry below internationally negotiated price levels. The Chinese government has also secured contracts and exploration rights abroad for its SOEs.
6. Labor Controls: The government exercises various methods to control employees including the dang'an or employment dossier; and to reduce labor costs through injection of part-time and migrant workers and the use of prison labor.
7. Tax Breaks: Many SOEs avoid taxation or reduce it through tax breaks (although this can backfire if a company's management loses favor).
8. Energy and Land Subsidies: The state subsidizes gasoline and electricity. Currently, Beijing tightly controls the price of both gasoline and electricity at well below their true economic levels. The state also offers free land and utilities to SOEs and companies in key strategic sectors.
9. Sectoral Credit Allocation: The Chinese economy speeds up or slows down on a sector-by-sector basis on credit allocations by Beijing. Some sectors such as automotive, steel, ethylene and metals' smelting have come off the boil. Others sectors such as coal, railways and utilities are still getting huge infusions of policy-mandated credit. Very high levels of bureaucratic interference characterize credit allocations and industrial-project approvals in China and the state banking system does not allow the market to price capital.
10. Stock Listings: SOEs and Collectives form over 93% of the listing of approximately 1300 companies on China's Shanghai and Shenzhen Stock Exchanges. Provincial governments pressure government regulators to discriminate against private companies and give the precious slots to their ailing state dinosaurs. Indeed, private companies without state connections cannot obtain a listing on any Chinese stock exchange.
11. Cheap Technology: China runs a deficit on its technology trade with the rest of the world and foreign multinational companies control 80% of technological imports and exports in China. The Chinese have made little progress in either basic research or advanced design in vital industries. Despite this institutional flaw, SOEs such as Huawei owe much of their success to lax laws governing the theft of intellectual property.
12. Control over Distribution Channels: Provincial and municipal governments control distribution channels to allocate and to manage market share, to protect favored industries from competition and to shape investment patterns. Regulations on distribution incorporate considerable ambiguities leading to both legitimate differences in interpretation and considerable legal efforts to find loopholes. Central and provincial governments routinely use this ambiguity to confer privileges on favored companies or industries, and to withhold normal rights from companies or industries as a form of protectionism. Administrative guidance from various and competing sources can override the basic laws or regulations either explicitly or unofficially. Provincial or municipal governments may interfere with the national limits on distribution by their generosity (to lure investment or to meet local goals) or restrictions (to protect local interests). Guanxi with local army officials assumes particular importance for distribution. Some estimates suggest that the Peoples' Liberation Army (PLA) controls distribution of goods for up to about 80% of the Chinese population. Its control over manufacturing facilities also makes the PLA China's largest and most diversified manufacturer of industrial and consumer goods.
13.  Special Market Information: Relevant information for strategic decisions comes at a premium price in China and often includes what we in the U.S.A. would consider Insider Information. In China, the central government deliberately controls and disseminates information that it considers of strategic importance. When restrictions on distribution insulate foreign or Chinese companies from their customers, they also cannot undertake direct market research and have to rely on less-sophisticated surrogates. For example, General Motors' (GM) interns in Beijing have scoured the capital's streets to find out who is buying their cars after the intermediaries get them, so that GM can build guanxi with the buyers.

What complications, if any, in evaluating such subsidies are caused by the lack of transparency in China's government and economic systems?

Lack of transparency affects ability to monitor all forms of subsidy except perhaps Stock Listings. Opacity serves as a tax which:

1. Reduces ability to determine the true efficiency and productivity of China's labor and results in potentially sub-optimal foreign direct investment (FDI) decisions until after commitments are made. Consequently, our research has shown that FDI enjoys higher ROIs and ROEs across entire industrial sectors in India against China, including Capital Goods; Food Beverage and Tobacco; Materials; Pharmaceuticals and Biotech; and, Software and Services.
2. Reduces the ability of U.S. domestic producers to prove dumping, especially as so many of those affected are Small and Medium-sized Enterprises with limited resources.
3. Magnifies the weakness of China's statistical system which depends too much on reporting and too little on sampling; the statistical system shows a systematic bias to over-report growth at the bottom of the economic cycle and under-report it at the top, ie, to flatten out a much more volatile economic cycle. Recently, some foreign companies have started constructing their own physical-activity indices of everything from freight-barge traffic to power consumption and air miles flown to find true economic indicators, but the enormous expense constrains companies from doing this well.
4. Reduces the credibility of the SOEs' books. For example, in 2003, China's top 500 SOEs reported revenues of 4.07 trillion Yuan up 25% from the previous year; and profits of 334 billion yuan, up 33% from the previous year. Only 87 of the 500 reported making losses. Unfortunately, outright fraud aside, most SOEs' managers do not know their real profits and tell their supervisors what they want to hear.

Unreliability in macroeconomic data also seriously compounds the problem of estimating the effects of subsidies.

A. For example, in February 2002, the Chinese government said that China's GDP had grown by 7.3% in 2001, making it the world's fastest-growing economy. However, growth rates reported by individual provinces told another story. Only one, Yunnan, said its product had grown slower than the national rate. Taken together, the provincial figures produced a national growth rate nearly two points higher than the official rate! The National Statistics Bureau (NSB) conducts sample surveys and uses these to estimate the country's GDP and growth rate. The results have invariably disagreed with provincial figures. In 1995, the GDP growth rate suggested by provincial data averaged three percentage points higher than the figure of 10.5% produced by sample surveys. Opinions vary as regards the accuracy of the central government's estimates. However, in China, few scholars publicly attempt any detailed justification of alternative figures because of political sensitivity.

B. China's NSB also lacks the capacity to collect data outside normal information channels and lower-level officials interfere with its surveys. The numbers generated by provincial governments remain an important criterion in evaluating local officials' performance, creating an incentive for statistical falsification. The pressure to exaggerate statistics grew in the late 1990s as Chinese officials sought to pump up the economy to stave off the Asian economic slump's effects. Beijing declared that the country had to grow at least 7% a year to create jobs and to forestall social unrest. Not surprisingly, reported growth rates have not dipped below that level since.

C. Officials may also routinely underreport other sensitive data such as debt numbers, unemployment or even FDI to avoid tax payments and governmental scrutiny. The central government's methods at ascertaining the validity of data, a process it calls yasuo shuifen or 'squeezing the water', involves sample surveys, price-index adjustments and plenty of guesswork.

D. Technical difficulties, such as staff reductions among statistical analysts, have enhanced errors in data. No comprehensive measures exist for the size of the fast-growing private-business and service sectors or even for what constitutes FDI.

E. The Chinese government strictly controls economic and industrial data and even classifies some as state secrets. Routinely, Beijing has overvalued SOEs' stocks of unsold goods, and underestimated inflation. Other provinces underreport growth and activity: for example, Zhegiang province in Eastern China may have underreported growth to conceal the rapid development of private companies in its economy. Additionally, affluent provinces, such as Guangdong in Southern China, may have underreported growth to avoid paying more taxes to the central government. However, without more systematic data, economists cannot definitively state if these factors pushed up growth or even occurred.

F. Governmental officials downplay unemployment figures to mask the suffering that economic reforms and restructuring have caused. The official unemployment rate of 3.6% in 2001 excluded xiagang workers (laborers receiving small, monthly stipends from former companies and not counted as unemployed) that economists estimate to number about 10 million. The official rate also excluded farmers who left their fields to work in cities, a floating population of around 150 million unemployed migrants. Using international standards, China's unemployment rate in 2001 approximated 7.6 percent in rural areas and more than 8.5% in the cities, well above Beijing's red-flagged figure to indicate inevitable social turmoil.

G. Most disturbingly, the central government's debt numbers look highly erroneous. The Central Bank's governor, Dai Xianlong, confessed to Parliament in April 2002 that national domestic debt appeared much higher than the official numbers (16% of GDP) suggested. Dai said the figure appeared closer to 60% of GDP if one considered unfunded state-pensions' liabilities, local governments' debts, and major banks' nonperforming loans (NPLs). Dai's unusual candor may mask more bad news. Independent economists have discovered that Dai's statistics drew on China's yearbook GDP growth statistics. Debt more realistically appears closer to 100 or 125 percent of GDP. The Bank of China reported two different figures for its NPLs in 1999, one using Chinese accounting standards, another Western; the latter looms 2.6 times greater than the former. Moody's has openly called the books of China's 'Big Four' banks, 'meaningless.'

What subsidies will be utilized for the 11^th Five-Year Program Period?

I anticipate that all the subsidies that I identified will continue. The 11^th 5-year plan specifically identifies certain strategically important industries that will receive state subsidies. These include:

1. Integrated circuits and software including technology for 90-nanometer and smaller integrated circuits;
2. New-generation networks including digital TV networks and mobile communication;
3. Advanced computing including technology for petaflop computer systems;
4. Biomedicine including commercial production of vaccines;
5. Civil airplane including general purpose planes and helicopters;
6. Satellite applications including meteorological, oceanographic, navigation positioning and telecommunication satellites; and
7. New materials including high-performance materials in information biological and aerospace industries.

Researchers may have more difficulties monitoring the rate of subsidization as China's 11^th Five-Year Plan has only two numeric targets: per capita GDP in 2010 must be double the 2000 figure and 'each work unit must cut its use of energy by 20% of current levels by 2010.' The plan fails to mention raising the price of electricity and gasoline, and unlike the previous 10 years, sets no economic growth targets.

In Part Two, Prof. Haley covers how profitable and available those subsidies are and how profitable companies are that serve the China market.

(The entire U.S.-China Commission hearing record for April 4, 2006 is available at www.uscc.gov/hearings/2006hearings/transcripts/april_4/06_04_04_trans.pdf.)