State owned or not – China’s connected capitalism.

Ping An Life Insurance – A politically connected, private company.

China’s economy is often caricatured as a two horse race – politically connected SOEs against heavily handicapped private enterprises. The latest volume of the China Quarterly has an article by Yuhua Wang, Beyond Local Protectionism: China’s State–Business Relations in the Last Two Decades, takes us well beyond this concept to take a close look at thee political ties of both private and state-owned firms listed on China’s stock markets.

Using listed company data from 1992, 2002 and 2012, Wang counts the number of directors in each company who have served, or currently served in a range of government positions. Ninety per cent of listed companies have directors connected to the national government, local government, the National People’s Congress or political consultative committee (described as a national parliament),  a local people’s congress or consultative committee (described as a local parliament), a national Chinese Communist Party Congress, a local Communist Party congress or the People’s Liberation Army.

The 90 per cent figure is across all ownership types. Wang doesn’t specifically separate listed companies controlled by SOEs and those that are ‘private’, but does look at the effect ‘state-owned share’ of each company on the likelihood of specific political connections.  We see that a higher state-owned share correlates with more connections with national and local (executive) governments, and the PLA, and negatively with representation in a People’s Congress or Consultative Committee.

WangYuhua.png

Source: Figure 2 Wang Yuhua (2016),

He finds that companies connected with Parliaments are more likely to have a lower state share, and higher profits, consistent with the argument that rich businessmen are co-opted into these pseudo-representative forums as a way of co-opting them in support of the political regime.

 

Three paths to SOE innovation

Early this week I was able to attend a seminar on Chinese innovation by Bruce McKern – his new book China’s Next Strategic Advantage: From Imitation to Innovation. The review of McKern’s book in the Economist is also reviews Paper Tigers, Hidden Dragons: Firms and the Political Economy of China’s Technological Development by Douglas Fuller from Zhejiang University. The Economist describes Fuller’s book as is “scathing in his indictment of state capitalism”

By showering state-owned enterprises and “state-favoured” private firms with soft money and protecting them from market discipline and bankruptcy, the state removes any incentive for them to upgrade their technological capacity.

Of course, many of China’s most exiting ‘innovative’ firms for innovation, Alibaba, Xiaomi, Huawei are private. Hai’er is not a state-owned enterprise (it appears to be some sort of collective). But a paper on technology catch up by Yangao Xiao, Andrew Tylecote and Jiajia Liu suggests that the story about SOEs is not as simple as the Economist suggest (I haven’t yet read Fuller’s book, but from what I can see of it on Google Books he treats the SOE questions a lot more rigorously).

The authors choose three companies from Sichuan province for detailed case studies on their technological upgrades at “Yibin Grace Group Go Ltd” (Grace), “China National Erzhong Group Co ” – the No.2. Heavy Machine Factory (CNEGC) and “Changhong Electronics Group Corpration” (Changhong). All are state owned in different forms.

  • Grace is 59% owned by Yibin city, and 39% owned by Sichuan province, making it a ‘city SOE’. It makes viscose fibre.
  • CNEGC is a central SOE, with over 12,900 employees. It makes machinery for steel mills.
  • Changhong was originally owned by the military the make radar screens for aircraft, but started making TVs with a Matsushita production line in 1979 at the start of China’s ‘opening up’. It is a provincial SOE owned by Sichuan province.

None of these fit the simple stereotype about lazy SOEs.

By 1997 “Grace was on the edge of bankruptcy with 3000 employees and a production of 21,000.” The city government when on a talent search among all its senior employees to find an “exceptional manager” to turn the firm around. A 30 year old sales director, Feng Tao was given job on the understanding that “if Grace failed he would not be blamed; if he succeeded he would be generously rewarded”. As a local SOE, Grace didn’t have a lot of capital but it did have a lot of workers – being an inland province, it couldn’t sack them. Rather than upgrade its equipment to the latest technology from America or Europe, it innovated its own “2S” technology which allowed it simultaneously to spin multiple strands of yarn on its existing machines. This allowed “large increases in output at very low cost”. With the help of the Sichuan IPR Bureau, Grace attempted to protect its innovation using patents. It wanted to prevent other firms using its technology although it was ‘politically obliged’ to license it to some other SOEs. By 2009 it Grace had 245 R&D staff, and a 33% share of China’s viscose market.

In the 1990s, CNECG was helping build steel mills, but because key components had to be sourced from abroad it do ‘80% of the job but only received 20% of the profits’. An internal project team was set up to design a “hot coil box”, which suited Chinese conditions. It eventually developed a coil box that was 30-40% cheaper than imported coil-boxes, and captured half of the Chinese market.

Changhong was not one of the three ‘favoured’ firms which had been selected to develop Chinese through joint ventures with foreigners. Nevertheless, through the 1980s and 1990s Changhong was able to become a market leader through cut-throat price competition, and raised capital on the stockmarket. The Chairman of the company in 1993 persuaded Zhao Yang, then a top lecturer in physics at Tsinghua University, to come back to his home town and join Changhong. Zhao Yang became chief engineer in 1995 and Chief Executive Office in 2000. Zhao wanted to cut back on production and sell down inventory in order to generate cash for R&D. The provincial government was concerned about employment effects, and so rotated Zhao Yang into a role as a Deputy city Mayor instead.

Changhong had been a leader in old-style CRT televisions, which by the early 2000s were being superseded by flat-screen digital TVs instead. China had just entered the World Trade Organization, so protection for domestic TV manufacturers was being removed. Changhong seemed done for.

Luckily, Zhao returned as Chairman in October 2004, and Sichuan had started receiving central government funds to help ‘develop the West’ – some of which were put into Changhong whose R&D spending finally increased dramatically. From 2004-2006, Changhong lost an average of RMB 1 billion annually, but spent 6 per cent of its revenue on research and development. In 2008, Changhong bought a Korean OLED manufacturer and by 2011 had 11 per cent of the Chinese market.

These three SOEs clearly aren’t innovation paper tigers.

The Yibin city government did not have deep pockets to protect Grace from bankruptcy. So it needed to find its most entrepreneurial manager to save the company.In contrast to other central SOEs, whose heads are rotated between companies on 5 year contracts, CNEGC was run on more military lines ‘red experts’ who spend their entire careers within the firm. Changhong, a provincial SOE, did get some ‘soft money’ from the central government’s Western development push, but it used it precisely to upgrade technological capacity.

These three SOEs are probably not representative of the state-sector. They were selected because of their technological upgrading, and so it’s not surprising that innovation was found. But they do reveal diverse and (dare I say) innovative forms of corporate governance and management that do reflect diversity in SOE management today. There is simply not a ‘one size fits all’ model any more – different combinations of ownership rights between different players create different incentives for corporate managers and different performance outcomes. SOE products (viscose thread, components for steel mills, colour TVs) may not be at the ‘cool end’ of innovation in China (smart phones, modular buildings and internet-financial services) but it’s too glib for the Economist to say that state-owned firms have no incentive to innovate.

Stargazing China’s Central SOEs

Eight hundred and forty-five central state-owned enterprises report to the Central Government’s Finance Ministry. But only 106 of these are managed by the State Assets Supervision and Administration Commission (SASAC) of the State Council.

Who manages the rest?

I’ve rendered this colour-coded map to show the galaxy of central SOEs.

CentralSOEMapLabelled.png

Red dots are central government agencies. Cyan dots are SOEs under central government agencies. Blue dots are universities that sit under the Ministry of Education – their SOEs are in magenta.

I’ve added some grey dots for companies that are listed on Shanghai, Shenzhen or Hong Kong stock markets and owned directly by a central SOEs. Think of these as orbiting planets. I include listed subsidiaries of listed subsidiaries also in grey.

In green, I highlight the listed companies owned by Central Huijin, which is the main holding company for most of China’s state-owned financial sector.

There’s still some uncharted territory on this map – mainly SOE holdings through unlisted companies. I don’t scale the dots according to the size, so a university-owned SOE is the same size as State Grid or Sinopec.

Nevertheless, this picture gives an idea of the diversity and complexity of state-ownership structures in China – even at the central level.

My next mission, to explore and map the galaxies of provincial and local SOEs!

Local SOEs going global

Last week The Australian newspaper reported that “China’s powerful state-owned ­enterprises are tipped to become the major driving force in overseas real estate investment in Australia following major reforms ordered by the Chinese government to clean up the troubled sector and encourage these groups to ­invest abroad.” But while the state sector as a whole may be powerful, and there certainly are powerful SOEs, it isn’t true that all SOEs are powerful or even strategically important to China’s central government. The distinction between China’s 106 central SOEs and thousands of SOEs that are administered at provincial and county levels is crucially, but unfortunately not usually made.

 

Ming Hua Li, Lin Cui and Jiangyong Lu, in a 2014 paper in the Journal of International Business Studies, argue why this distinction matters.The authors make a theoretical case for why national and local SOEs are different, not just in China – but for India, Indonesia, Malaysia, South Africa and Vietnam.

MingHualietal

Ming Hua Li et al, Figure 1

Central SOEs, which are more likely to enjoy privileged monopoly positions domestically and internationally, are expected to act as ‘national champions’, which include meeting the economic and sometimes political policy goals of the nation. The authors describe them as following a “stronger, politically laden and non-commercial logic”.  By contrast, local SOEs enjoy significantly more managerial autonomy and follow a logic which is “increasingly profit-driven to serve local economic development objectives.”

The paper is theory driven, but does advance eight empirically-testable propositions about how central SOEs abroad might act compared to local SOEs.

  1. The greater the degree of administrative decentralization in a country, the higher the level of managerial autonomy of local SOEs relative to central SOEs
  2. The greater the degree of fiscal decentralization in a country, the lower the extent of institutional support received by local SOEs compared with central SOEs
  3. The greater the degree of industrial restructuring and consolidation in a country, the lower the level of monopoly power of local SOEs relative to central SOEs.
  4. The greater the degree of market liberalization in a country, the higher the level of market orientation of local SOEs relative to central SOEs.
  5. Relative to central SOEs, local SOEs are more likely to follow a gradual internationalization path when conducting outward FDI.
  6. Relative to central SOEs, local SOEs are more likely to engage in business diversification when conducting FDI.
  7. Relative to central SOEs, local SOEs are more likely to choose a joint ownership structure than a sole ownership structure when conducting FDI.
  8. Relative to central SOEs, local SOEs are more likely to pursue greenfield investment rather than M&A when conducting FDI.

China’s state-owned tobacco monopoly

Earlier this year, I published a paper asking “Where Have China’s State Monopolies gone?”.  Because of the way Chinese industrial survey data doesn’t aggregate enterprises with the same industry group, the paper argued that the conventional measure of industrial concentration, the Herfindahl-Hirschman Index (HHI), would under-estimate the prevalence of monopoly in China’s industrial economy. In the paper, I observed that “the 20th largest (industrial) subsector, the manufacture of cigarettes, has an HHI of 0.041, making it only the 148th most concentrated industry, and yet the entire tobacco industry is unashamedly monopolized by the State Tobacco Monopoly Administration”.

The entire story of China’s tobacco industry, and the role of the state in it from China’s late-Republican period to the present, is the subject of “State-Market Interactions in China’s Reform Era“, by Junmin Wang. This is a tremendous book that combines fieldwork, including 133 interviews, with a sophisticated framework concerning key questions in state ownership.

Tobacco is worthy of a closer look for a few reasons.  Firstly, while the Chinese tobacco industry is monopolised by state-owned enterprises (SOEs), these SOEs are not regulated through the State-owned Assets Supervision and Administration Commission (SASAC) that have overseen most industrial SOEs.  Since 2003, the “State Tobacco Monopoly Administration” at central and local levels, oversees the enforcement of the state monopoly, the Chinese National Tobacco Industrial Corporation (CNTIC) is in charge of all tobacco enterprises, including cigarette factories, and the China National Tobacco Trade Corproation (CNTTC) oversees domestic and international trade.

After assuming that all central SOEs involved in cigarette manufature are actually part of the same group, I showed that the HHI measure leaps from 0.041 (an unconcentrated market) to 0.842 (almost completely monopolised).

But this book tells a fascinating story of fierce local competition, despite the formal state monopoly.  As the author explains, the central state monopoly was formally established in 1984 with the goal of preventing local governments from allowing the tobacco industry to ‘grow out of the plan’. New township and village enterprises (and eventually private enterprise) were allowed to grow outside the state-planning mechanisms when it came to textiles and light manufactures, but to protect the state’s own tobacco revenues, the entire tobacco industry was put under the jurisdiction of a new central bureau. Tobacco would remain as an island of central planning and state-ownership, with quotas assigned by the central authority even as planning was eventually abandoned.

We might expect that old-school central state monopoly and a commitment to central planning would guarantee the ossification of the tobacco industry, despite the surrounding tumult of ‘reform and opening up’ in other sectors.

This isn’t what happened – Wang explains how local governments, left thirsty for fiscal revenue, effectively colluded with local bureau of the central monopoly administration to circumvent the strictures of the plan.  Local governments competed amongst each other for market pair, so that while the state part of the “state monopoly” was strictly enforced (there were no private competitors) the monopoly part was difficult.  Provinces informally traded quota assignments and sales rights amongst themselves, and one enterprising municipality (Yuxi, in Yunnan province) even combined the monopoly regulator, local cigarette manufacturer, and tobacco trading company into a “one-stop shop”, all under the common leadership of the entrepreneurial manager of the cigaratte factory.  The result was the Hongta Tobacco Group which became China’s “tobacco empire”.

In theory state ownership softens the budget constraint and weaken firm performance, in pratice Wang argues that in Yunnan “local governments surrended their direct control to the SOEs and played a supporting role in their economic activities”.  While in a formal sense, the company is centrally state-owned subject to a central planning mechanism, Hongta Group, supported by Yunnan province, took on a life of its own far removed from that of the comfortable state monopoly.

Wang also attempts to explain a contradiction between the decentralisation of industry in the 1980s and 1990s, with the center’s drive to create ‘national champions’ that could take on international competitors once China joined the World Trade Organization in 2001. The number of cigarette factories in China dropped from 202 in 1996 to 30 in 2008, after the central government encouraged mergers between provinces. Wang explains this not as an attempt to retreat from decentralisation within the Chinese domestic market (which gives a company like Hongta the chance to rise), but in order to have “internationally competitive, big Chinese corporations” in competition against the existing bohemoths.

Wang stresses the argument that local governments ‘have become “rational” economic agents and market participants‘ in the sense that they can no longer be thought of as merely implementing agencies of the central government. It shows what on paper is central government ownership and state monopoly doesn’t prevent the development of markets in some form.

 

I’ve done some work on China’s paper industry, and was already familiar with the Hongta brand because of Zhuhai Hongta Renheng Paper Company runs a large paper mill in Guangdong Province. According to the 2015 Almanac of China Paper Industry, the paper mill in Zhuhai started in 1991 as a joint venture between Yunnan Hongta Group and a Singapore company.  But today it Zhuhai Hongta today would be classified as a central SOE – since its original shareholdings have been augmented b a new cotnrolling investment from the Shenzhen-Listed Foshan Huaxin Packaging Ltd , itself controlled by the China Paper Corporation, a wholly state-owned subsidiary of central SASAC SOE – China Chengtong Holding Group Copmany. 

Bosses and bureaucrats

In a capitalist democracy economy, achieving the top executive of one of the world’s largest companies might be the ultimate goal of a corporate career. But in China, heads of central state-owned enterprises (SOEs) are part of a larger career structure integrated by the Communist Party. A 2012 article we considered last year looked at mechanism through which the Party controls appointments of top SOE executives.

20110401161536697538

The Central Organization Department and SASAC hold a joint forum on central SOE recruitment.

Another 2012 article, published in Chinese by Ruilong YANG, Yuan WANG, Huihua NIE (“The Political Promotion for Quasi-Government Officers: Evidence from Central State-Owned Enterprises in China”) provides an even closer look at the careers of top SOE executives, by looking at factors that led to their promotion. In the case of a successful SOE head, promotion may mean becoming a provincial governor or party secretary, a Government Minister, or even (eventually) a spot on the Politburo.

This is because SOEs are part of a personnel system that encompasses every important position in China. While the State-Owned Assets Supervision and Administration Commission (SASAC) acts on behalf of the central government as owner of central SOEs, the right to appoint management is shared with the Party. This makes SOE leaders ‘quasi-officials’, in the terms of this paper.

SASAC’s First Bureau for the Administration of Corporate Executives ‘shares’ the power to appoint the leadership of the most important 53 central SOEs with the Party’s Central Organization Department. This gives those leaders ‘Vice-Ministerial’ ranking, equivalent to a Deputy Minister of a department, or a Deputy Governor of a Province. Leaders of the remaining SOEs are formally appointed by SASAC’s Second Bureau for the Administration of Corporate Executives, subject to Organization department approval. These are bureau level appointments, giving their leaders the same rank the mayor of a third-tier city.

The organization department’s intimate involvement in this process places SOE leaders in the same career structure as other government officials. This is an area that is already widely studied. According to work cited in this paper, performance indicators for local government officials include morality (得), capacity (能), hard work (勤), results (绩), and integrity (廉). The results measure is the part in which GDP growth famously plays a role. But some research suggests that after controlling for a local official’s own personal political connections, the performance measure is no longer significant. This could be because politically connected officials are more likely to be assigned to faster-growing areas.

Assessing the performance of local officials is difficult – there are social as well as economic metrics that the Party values, in both cases local results are influenced by factors outside the control of leadership, and in many cases local statistical reporting may be either manipulated (for the purpose of improving promotion prospects) or downright unreliable.

Judging the performance of SOE leaders has fewer confounding factors. The formal method by which SASAC assesses performance of SOE executives (“Interim Measures for Assessment of the Operational Performance of Persons in Charge of Central Enterprises”) has been in place since 2004. (Key SOE executives are remunerated based on an annual letter grade from A to E). They now have limited political or social obligations,  which means straight financial measures for performance can be appropriate on the basis of their audited financial statements. Finally, the SOE head has more control over the enterprise than a local official has over his local area.

The authors linked annual financial data of central SOEs to biographical information on the SOE’s top two leaders gleaned from corporate and official websites. The top two leaders are usually the Party Secretary and the CEO. (In a company with a board, the Chairman of the Board tends to be Party Secretary). After excluding observations for which financial or leadership information is not publicly available (typically defence-related SOEs), the authors used a database of 87 companies and 189 leaders, giving 589 observations from 2008-2011.

A promotion is observed if an SOE leader moves to a higher ranked position, either amongst SOEs (for example, from a Bureau level SOE to a Vice-ministerial level SOE) or elsewhere in the system (for example, becoming a government Minister or Provincial Governor). A move from being second-in-charge to first-in-charge of an SOE is also a promotion (for example, if the CEO becomes Chairman of the Board). The move into a Party organ at the same level is also considered a promotion.

To judge SOE performance, the authors look at growth in operating revenue. (Operating revenue is the basis of the Fortune 500, and also Chinese measures of leading companies). This departs from the official SASAC assessment, which uses profit rate (net profits divided by average net assets), which the authors argue is unreliable (and also often negative), and economic value added (which was not published before 2010).

To measure the political connections of SOE leaders, they identify whether leaders are members or alternate members of the Central Party Committee or the Central Disciplinary Inspection Committee (8% of sample), whether leaders are members of the National People’s Congress (11% of sample), whether they have previous work experience in central government or party organs (25% of sample), and whether the SOE’s headquarters is located in Beijing (75% of sample). All these factors give SOE leaders better access to national leaders.

Controlling factors include the leader’s tenure at the firm (average 4 years), his age (average 54), whether he is over 60 (relevant for retirement, 13% are), whether he holds a PhD (21% do), whether the SOE is Vice-Ministerial or Bureau level, and the size of the firm as measured by the log of revenue in the previous year.

The authors find growth in revenue to be a significant determinant of promotion prospects (growing the value of state capital by comparison is not significant). In terms of political connections, being a member of one of the central committees is significant, but a role on the National People’s Congress, central-government work experience and being in Beijing is not. Leaders with doctoral degrees are more likely to be promoted, but their performance ranking (linked to their salary) does not relate to promotion.

The last point is particularly interesting. The authors interpret to argue that SOE leaders resemble government officials (concerned about building empires and future promotions) than professional managers (concerned about their own compensation from the firm)

 

Lenin Ltd.

The role of the Chinese Communist Party, while not prominent on the English language websites of SOEs is nevertheless not hidden.  In a previous post, we considered the provisions of China’s corporate law and the CCP’s own constitution that embed the Party within what is otherwise a modern market corporate form.

Jiangyu Wang, from the National University of Singapore, provides the best exposition I’ve read so far on the “twin governance structures” for legal and political control of SOEs.

 

The legal structures are easily recognizable to a western observer, with its shareholders, represented by a board of directors who appoint and oversee the management of the firm, embodied in a separate legal entity. In the case of China, the State Council representing the whole people, is the shareholder. Its State Owned Assets Supervision and Administration Commission (SASAC) acts on behalf of the state owner to participate in decision making and elect management personnel “in accordance with law”.

Party control exists outside this legal framework though. First, and most simply, Party members who are directors, managers or employees of the firm are obliged by virtue of their membership to follow Party decisions, maintain Party discipline, and provide feedback back to the Party. (The degree to which Party interests deviate from profit interests, or to which individual Party members are actually themselves monitored for compliance is an open question).

More important is the Party organization in the SOE, which gains significant authority by virtue of the fact that key members of the legal organs of the company overlap with key positions in the party group. The Party Secretary is typically also Chairman of the (legal ) Board. By virtue of party discipline, once the Party organization makes its decision, it is binding on the directors and management when it comes time to vote in the corporate decision making.

Finally, there is the Party involvement in careers of SOE managers. This doesn’t just include the appointment of top SOE leaders by the organization department  (about which Kjeld Erik Brodsgaard has written extensively ) but also, according to Wang, the appointment, management and supervision of all SOE officers at and above middle management by the SOE’s own Party cell.

What are the implications of this?  Wang argues that when push comes to shove, this means that the interests of the CCP will inevitably trump those of outside minority shareholders, should their arise a conflict between profit maximization and Party interests.

What is less clear is when such conflicts might occur. Wang gives examples of SOEs being expected to “volunteer” time and materials in the event of natural disasters as an example of a case in which SOEs go above and beyond what a publicly listed non-state firm might do (although even these would probably be willing to chip in for PR or corporate social responsibility motives). Moreover, this could be thought of as one way in which the state takes a non-cash dividend from the SOE which might make up for lower than average financial returns on capital for the state.

There are cases where foreign policy or security interests might tend toward commercially risky decisions. This might involve oil investment or other forms of resource exploitation, for example. But for the thousands of soes operating in ordinary competitive sectors of China’s economy, or simply seeking small profits abroad in ordinary commercial sectors, it is hard to see what the divergent Party or national interest there could be.

World GDP to 2050: Conditional convergence for the masses

HubbardSharmWorldGDPto2050

My friend Dhruv Sharma and I project gross domestic product (GDP) for 140 world economies from 2020 to 2050 based on United Nation’s demographic projections, the International Monetary Fund’s GDP statistics and estimates of potential labour productivity derived from the World Economic Forum’s Global Competitiveness Index (GCI) and a methodology published by the Australian Treasury. (Naturally, the views expressed in this paper represent the views of the authors and not those of the Australian Treasury.) We review the conceptual framework underpinning this model, and identify its core assumptions. Finally, we highlight potential applications for this model, including: considering the dispersion of global economic activity; assessing the potential scale of activity across different trading blocs; and quantifying the impact of domestic policy reform scenarios in individual economies. Rather than provide an exhaustive analysis of the results, we make the data and results freely available.

Paper from East Asia Bureau of Economic Research (EABER)

Data and Interactive Results: https://goo.gl/SJI5QA