Quench a Thirst with Poison?—Local Government Financing Vehicles’ Past, Present and Future
Local government financing vehicles (LGFVs) are economic entities established by Chinese local governments and their departments and agencies through fiscal appropriation or injection of assets to perform the function of financing government-invested projects. LGFVs are the combined result of local governments’ exuberant investment demand, limited fiscal capacity and narrow financing channel. Their origin, development and operation model reveal an unignorable aspect of state capitalism in China—state capitalism at local level. The two terms of “de facto federalism” and “rent-state” help to understand this point. The “barbarian growth” of LGFVs substantially contributed to the high risk of local governmental debts. Efforts should be made to normalize and commercialize the operation of LGFVs, allow local governments to lawfully raise debts, and ultimately, reshape the role of local governments.
Key words: local government financing vehicles; local governmental debts; tax-sharing system; de facto federalism; land finance; Transformation of government functions
Mounting local governmental debts (herein “local debts”) and the associated fiscal and financial risks have become a sword of Damocles hanging over Chinese economy. According to a recent report by the National Audit Office (NAO), by the end of June 2013 the balance of local debts nationwide had reached the record high of 17.89 trillion yuan (2.93 trillion US dollars),  or about one third of China’s GDP. Compared to the number at the end of 2010 (10.72 trillion yuan, or $1.76 trillion),  this was a remarkable increase. And local government financing vehicles (LGFVs) were by far the largest borrower: Debts borrowed by LGFVs amounted to 6.97 trillion ($1.14 trillion), accounting for nearly 40 percent of the total local debts. 
LGFVs emerged in China in the 1990s when the so-called “tax-sharing system” was introduced and the central government began to take the lion share of tax. This left local governments with less money for such things as infrastructure construction and urban development.  Also, the intemperate infrastructure investment and spending by local governments in the early 1990s with money from bank lending and selling bonds overseas, and the resulting rampant inflation nationwide, prompted the central government to pass a severe budget law in 1994, forbidding local governments from running deficit or selling bonds. 
LGFVs apparently had their roots in the restrictions imposed on local governments. Local governments used such vehicles to obtain back loans as well as issue the so-called “Urban Construction Investment Bonds”, or Chengtouzhai, thus bypassing the prohibition of the Budget Law. The Asian Financial Crisis in 1998 came as an opportunity for LGFVs: it required spending to stimulate the economy and China was entering its golden period of urbanization, but local governments were strapped for cash.  But the real fuel was the Global Financial Crisis in 2008. Heavily stricken by the crisis, China announced a 4 trillion yuan (around 570 billion US dollars at the then exchange rate) stimulus package on November 5, 2008, equivalent to 12.5 percent of the 2008 GDP. The Central government, however, would fund only 1.18 trillion yuan of the stimulus, and the rest were to be “matched” by local governments. It was only logical that local governments again turned to LGFVs and the state-owned banking system. Indeed, total bank lending to such companies rose from 1.7 trillion yuan of outstanding loans at the beginning of 2008 to nearly 5 trillion yuan just two years later.  Furthermore, in the economic downturn after the crisis, many local governments initiated their own stimulus plans with government investment at the core. A recent report shows that since June 2012, the total amount involved in the stimulus plans of over ten provinces and municipalities has exceeded 20 trillion yuan, far more than the previous “Four Trillion Plan” of the central government.  Given this fact, it is no surprise that LGFVs have accumulated those mounting debts.
Though dressed up as business corporations, LGFVs have been more of a disguised arm of local governments. Their most—or even only—reliable asset is land. Often empowered by local governments as land banks, they use land as the main collateral to secure long-term loans from China Development Bank (CDB), originally a policy bank but has gone far beyond that over the years. With this initial funding, LGFVs are able to start operation and borrow further by seeking short-term loans from the state-owned commercial banks or selling bonds on the inter-bank bond market. Since the proceeds from selling or developing land are the main source to repay such debts, local governments behind LGFVs have all the reason and incentive to expropriate more land and keep the property price at a high level, so as to continue the cycle. That’s why despite the repeated announcements and increasingly severe measures of the national government since 2009 to lower or at least stabilize the property price, local governments have not been really interested in enforcing that policy and are always trying to find loopholes or seek exceptions. The accelerated urbanization process since the 1ate 1990s, and the expectation of its continuation and expansion in the foreseeable future, is the main driving force of this borrowing and investment spree.
This reveals an unignorable aspect of state capitalism in China—state capitalism at local level. Though constitutionally and legally China is a unitary state, in fact local governments (provincial and below) enjoy high degree of autonomy and freedom in policy enforcement, especially with regard to economic issues. This has been the case since the beginning of the reform and opening in the late 1970s. It is even argued that the actual powers of Chinese local governments dwarf those of the local governments under a federal system. Thus China is characterized as “de facto federalism”.  In this sense, there is not only a state at the central level vis-à-vis private sector, but also numerous “states” at various local levels—provinces, cities, counties and townships—vis-à-vis private sector at the corresponding levels. This partly explains why “(central) government orders don’t go beyond Zhongnanhai”. In fact, local governments contribute a lot to the “state advances, private retreats” （Guojin Mintui）wave since 2002.
In the rest of this article, Part II outlines LGFVs and the way they function, and provides some legal and policy explanations for their remarkable development. Part III looks into the major problems and risks associated with LGFVs. Part IV offers some suggestions for not only reforming LGFVs but also reshaping the role of local governments.
II. LGFVs and their development: what, how and why
A. Definition and features of LGFVs
The exact number of LGFVs remains an issue of discussion, partly due to the lack a clear definition of LGFV from the beginning, as can be seen below. The NAO report of 2013, however, stated that 7,170 LGFVs were audited.  This sheds some light on the multitude of LGFVs in China. Despite of their large quantities and various forms, for a long time there had been no official definition of LGFVs. Lacking clear guidance on what LGFVs should look like, how they should be organized, and what functions they can and can not perform, local governments in practice set up all kinds of investing and financing corporations, which, although shared the common label of LGFV, differed a lot in their size, functionality and corporate structure.
It was not until 2010, when LGFVs and their unregulated operations were increasingly recognized as posing substantial risks to the overall fiscal and financial system, and when the central government began to take steps to strengthen the regulation of LGFVs, that a somewhat summary definition was given. In a document promulgated by the State Council in June 2010,  the central government defined LGFVs as “economic entities with independent legal personality, which are established by local governments and their departments and agencies through fiscal appropriation or injection of assets such as land and equity, to perform the function of financing government-invested projects”. And according to a notice jointly issued by four heavyweight government departments to implement the LGFV Notice, such entities include all kinds of comprehensive investment corporations, such as construction & investment corporations, construction & development corporations, investment & development corporations, investment group corporations, state-owned assets management corporations and state-owned assets management centers, as well as industry-specific investment corporations, such as communications investment corporations. 
Based on this official definition as well as experience from the practice, it has been argued that four basic features can be attributed to LGFVs. First, they have strong government background, being established by local governments and their departments and agencies through fiscal appropriation or injection of assets such as land and equity. Second, their operation objective is to tackle the problem of funding insufficiency of local government-invested projects, and the money they borrowed mainly goes to infrastructure construction and public services projects of the cities. Third, they are independent economic entities with legal personality. Last, they more or less rely on the government credit to operate, since the money they borrowed might be directed to public services projects which do not generate cash flow or the cash flow generated is insufficient to repay the loan, and some kind of government guarantee is thus necessary. 
B. How LGFVs function
Due to the lack of lawful financing channels, as will be discussed below, local governments rely heavily on LGFVs to raise funds, and the capitalization of land plays a key role in the fundraising process. Typically it works like this: local governments inject as capital the land resources they own into LGFVs they established, and LGFVs use the allocated land as mortgage or the future revenue from the development of such land as pledge to apply for bank loans. Banks as a routine would ask the local governments to issue a commitment letter to the effect that they would serve as the “deep pocket” once LGFVs are unable to repay. According to a commentator, this model can be characterized as “land mortgage/pledge plus government commitment”. 
Figure 1: “Land mortgage/pledge plus government commitment” financing model
This operation model largely enables local governments to bypass the prohibition on borrowing money or being guarantors as imposed by relevant laws and regulations. On the one hand, formally LGFVs are independent legal persons, thus the obstacle that local governments themselves may not borrow from banks is overcome. On the other hand, since land is allocated by local governments to LGFVs as the latter’s self-raised capital, LGFVs can use the land per se as collateral to seek bank loans, and so the obstacle that local governments may not act as guarantors is also overcome.
CDB is behind the whole scene. In fact, this finance model is invented by CDB. Unique among Chinese banks, CDB is financed by bonds instead of deposits,  most with maturities of ten years or more that are bought by China’s commercial banks. That gives it the advantage in funding long-term infrastructure projects that might not generate a return on investment for many years. In this sense CDB is simply the best match for LGFVs. The new urban focus in China in the late 1990s unleashed a wave of state capital boosted by growing Chinese savings. Since CDB as a policy bank could not take people’s deposits directly, the cycle was something like this: the commercial banks used people’s savings to buy bonds sold by CDB on the nation’s bond market where the banks were the main investors. CDB would then help local governments set up LGFVs to borrow, and give them initial long-term loans. Thus dressed up and empowered, the LGFVs were free to go on a further borrowing spree seeking short-term loans from the commercial banks or selling bonds themselves on the bond market to banks and securities companies. 
The “Wuhu Model” serves as a classical case. In 1998, when Chen Yuan was appointed governor of CDB, the bank created a brand new lending model in the city of Wuhu in Anhui, home province of the then vice president Hu Jintao. First, CDB cooperated with the city government to transform the Wuhu Construction & Investment Co. Ltd. into a qualified financing platform. Then all the urban construction projects were bundled into a package, with the financing platform as the single debtor to borrow and repay, and a ten-year loan agreement of 1.08 billion yuan was signed between CDB and the platform. Finally, the city government established a “repayment reserve fund” to ensure the source of repayment. This was the so-called Wuhu Model. This CDB-LGFV lending model was copied and upgraded in Tianjin. In 2003, CDB signed a loan agreement of 50 billion yuan with the Tianjin government, with the loan proceeds being mainly used for the key infrastructure projects such as the construction of city expressways and subways.
One of the breakthroughs in this lending model is the association of the source of repayment of infrastructure loans with the income from land appreciation, i.e., the cost of urban infrastructure construction is covered by the income of land appreciation, so that the infrastructure construction funds can form a virtuous cycle; meanwhile the land income right serves as the pledge, and in case revenues from the infrastructure projects are insufficient to repay the loans, the income of land appreciation can ensure the safety of the loans.  This is totally understandable: as more infrastructures are built, land values can only go up, as would housing prices. Later on, the Wuhu model spread nationwide and became the standard model of financing urban construction. Especially after the Global Financial Crisis and the “Four Trillion Plan”, this finance model was eagerly followed by the large commercial banks, and LGFV loans became a significant part of their lending business.
C. Legal and policy factors underlying the development of LGFVs
In short, LGFVs are the combined result of local governments’ exuberant investment demand, limited fiscal capacity and narrow financing channel. With the acceleration of China’s industrialization and urbanization process, there has been a rapid increase in the demand for local construction investment. The introduction of the tax-sharing system in 1994 resulted in a mismatch between the fiscal power (Caiquan) and administrative power (Shiquan) of local governments, leaving them with decreased percentage of fiscal income but increased percentage of fiscal expenditure.  Legally unable to borrow directly, local governments established numerous investment and financing entities to undertake construction tasks, and use them to receive bank and trust money. The Central government acquiesced in or even supported this effort of local governments to tackle their financing problem. 
After the “Four Trillion Plan”, in order to solve the problem that local governments had insufficient funds to “match” the investment of the central government, the People’s Bank of China (PBC) and the China Banking Regulatory Commission (CBRC) jointly promulgated a guiding opinion in 2009,  stating that they “support qualified local governments to establish investment and financing platforms to issue financing instruments such as enterprise bonds and medium-term notes, so as to expand the financing channel of the matching funds for the central government invested projects”. This indeed ushered in an explosive development of LGFVs and greatly increased the potential risk.
To be clear, the development of LGFVs and more broadly, the accumulation of local debts, bore the hallmarks of the “transitional period” from the traditional planned economy to market economy. It resulted from the incomplete adjustment of the relation between government and market. More specifically, with the advance of the progressive reform featuring “decentralization of power and transfer of profits” from the central government, local governments gradually became administrative and economic units having their own development objectives. On the one hand, local governments now had their own independent economic interests, and the relevance of such interests to the local economic growth was greatly increased. On the other hand, the gradual decentralization of economic power provided local governments with unprecedented power and capacity to develop local economy and seek local interests. With the inertia from the planned economy period, however, most local governments were keen on “substituting” instead of “supplementing” the market, and invested extravagantly in various areas in competition with private enterprises, resulting in excessively broad government functions. As a result, the burden of fiscal expenditure was aggravated instead of alleviated. The expansion of fiscal expenditure pressured local governments to raise money through various channels, most remarkably the LGFVs, which in turn led to increasingly heavier burden of debts.
In China, the state has an unbeatable advantage over the private sector: it owns all the land. Moreover, land revenues or the so-called “land grant fees” are extrabudgetary revenues and don’t have to be included on the central government’s accounting of local budgets, meaning that there is little oversight on their use. This gives local governments much freedom in negotiate financing arrangements for projects with CDB and other banks. They simply need to devise a system to leverage the future value of land into large up-front loans, as indicated above by the Wuhu model. The excessive reliance of the fiscal capacity of local governments on land has given birth to the term “land finance” (Tudi Caizheng).
In fact, with the increasing significance of the land grant fees in the local fiscal income since the 1990s, China to some extent has had the feature of a rent-state. This, combined with the remarkably increased profitability of the state-owned enterprises and their contribution to the national fiscal income since the end of 1990s—which indicates that China still maintains some trace of an owner-state—means that even though China has begun to transform towards a tax-state, currently and in the near future, it remains a “hybrid” fiscal state: apart from tax, profit and rent are also significant sources of income of the state. This in turn means that the state has more independent resources and relies less on the society/private sector, as compared to a standard tax-state.  This also helps to explain the strong interest of local governments in infrastructure financing and investment, and more broadly, the rise of state capitalism in the past decade.
III. Major problems and risks associated with LGFVs
To be clear, the sustained operation of the LGFVs’ “land for loan” model depends on two preconditions, i.e., unlimited supply of land and the forever rise of the land price. But none of them can be continuously met in the real world. Once the land supply is in shortage or more probably, the land price drops, the debt-raising capacity of LGFVs will decrease sharply. And considering that a large portion of the LGFV debts are paid through the “borrow new to pay the old” cycle, the sharp decrease in their capacity to borrow new money is much likely to result in a break of the capital chain, thus triggering defaults.
A. The “barbarian growth” of LGFVs
Since Chinese law generally prohibits local governments from borrowing debts, it naturally does not provide for how to standardize and regulate the borrowing behavior of local governments. However, due to the reasons discussed above, it is in fact widespread practice for local governments to borrow money. Thus the combined result is that the motley borrowing practices of local governments are largely left unregulated. In this backdrop, LGFVs since the very beginning have been in a state of “barbarian growth”. They have two inborn defects, i.e., ambiguous relationship with local governments, and chaotic internal governance. On the one hand, LGFVs are in name independent corporate legal persons but in fact controlled by the governments, in name not-for-profit but in fact for-profit and often monopolistic, and in name borrowing on their own right but in fact relying on the guarantee of the governments. As pointed out by a commentator, “most LGFVs have become the de facto second line of fiscal income and expenditure for local governments at all levels. They have very close connections with the governments in personnel and financial matters. Both the bank loans and the urban construction investment bonds are guaranteed directly or indirectly by local governments. Their debts have constituted de facto hidden debts of local governments.”
On the other hand, LGFVs themselves often lack the internal control and outside supervision required by the Company Law. For example, corporate organs such as board of directors and board of supervisors are not essentially established, corporate governance structure is defective, regular information disclosure system is absent and cases of delayed disclosure, misrepresentation and omission are commonly seen.  This not only further blurs the line between LGFVs and local governments or government departments, but also adds to the risks inherent in LGFV debts.
B. Local governments’ liability for LGFV debts
In theory, since LGFVs are independent corporate legal persons, local governments as shareholder enjoy the protection of limited liability, and are only liable to the creditors to the extent of their capital contributions. In other words, local governments are not liable for those debts which can not be paid off with the LGFVs’ own assets. This seems to be consistent with the position of the central government. Article 4 of the LGFVs Notice provides: “Local governments bear limited liability for the LGFVs to the extent of their capital contributions, so as to internalize the risk of LGFV debts.” Article 4 of the Implementation Notice further prescribes: “For the debts newly incurred after the date of issuance of the LGFV Notice, local governments only bear limited liability to the extent of their capital contributions. If the debtors can not pay off all the debts, the creditors shall also bear the correspondent consequences.” Although the Implementation Notice does not specify what “the creditors shall also bear the correspondent consequences” means, but based on the context, a reasonable understanding is that the creditors shall themselves bear the risk that their lending can not be fully paid off in the event of the bankruptcy of LGFVs.
Judging from the actual practices, however, the answer does not seem to be so easy. First, as pointed out above, though in name LGFVs are independent legal persons, in fact they are interwoven with local governments in terms of assets, operation and personnel. Heads of the relevant government departments often serve as the legal representatives of LGFVs, and the whole process from establishment to finance to operation is taken care of by the governments. It is even pointed out that quite often LGFVs are merely a different name of the relevant government departments, and their operation and development depend on the arrangement of local governments.  Accordingly, “banks and even local governments tend to view LGFVs as a ‘tool’ of local governments’ investment and finance, instead of real economic entities”.  This being the case, it is only natural to ask whether the creditors can claim to pierce the “corporate veil” of LGFVs, and hold the relevant local governments—as the sole shareholder—to be personally liable for the former’s debts. I think this is at least possible.
Second, in the typical operation model of LGFVs mentioned above, i.e., the “land mortgage/pledge plus government commitment” model, local governments would issue a commitment letter to the creditors (banks) as a way of assurance. Article 8 of the Security Law explicitly prohibits state organs to act as guarantors, and Article 4 of the LGFV Notice also emphasizes that “local governments at all levels and their constituent departments…may not use fiscal income…or by any other means, directly or indirectly, to provide financing guarantees for the LGFVs”.  The point is, however, that the nullification of the guarantee contract for violating prohibitive provisions of law does not per se relieve the guarantor of his compensation liability.
More specifically, Article 5 (2) of the Security Law provides: “After the nullification of the security contract, faults, if any, of the debtor, guarantor or creditor, shall be affixed with due civil responsibilities.” In its judicial interpretation of the Security Law, the Supreme People’s Court further provides: “If the principal contract is valid and the security contract is void, the security provider and the debtor are jointly and severally liable for compensation for the financial loss of the creditor under the principal contract, provided that such creditor is not at fault. If the creditor and the security provider are at fault, the portion of civil liability borne by the security provider shall not exceed half of the portion that the debtor is unable to discharge.”  According to these provisions, local governments need to bear compensation liability to the lending banks at least to the extent of half of the portion of the debts that LGFVs are unable to pay. Not to mention that local government, based on non-legal considerations, might take the initiative to assume more liability. These all contribute to the risk of local debts.
IV. Some suggestions for reforming LGFVs and reshaping the role of local governments
A. Normalizing and commercializing the operation of LGFVs
The LGFV Notice requires that LGFVs be classified and cleared up. Only those LGFVs that are used to finance not-for-profit projects with stable business income and that mainly rely on their own income to repay debts, and those LGFVs that are used to finance for-profit projects, can be retained. Other LGFVs must cease to undertake financing tasks.  The retained LGFVs also need to be substantially reformed. According to the LGFV Notice, they should bolster their capital, improve their governance structure, commercialize their operation, and promote the diversification of investors and better the shareholder structure by introducing private investment. 
The key to implementing the above requirements, in my opinion, is to reform LGFVs into corporate legal persons with independent business operation, decision making and corporate management powers, pursuant to the principles of modern corporate governance, so that they are really separate and distinct from local governments and government departments. I would like to call this “de-governmentalization” of LGFVs. This certainly does not mean that local governments may not invest in LGFVs. On the contrary, they can invest in LGFVs and have all the rights and powers of shareholders. However, if they want to enjoy the protection of limited liability and avoid the “corporate veil” of LGFVs being pierced, they must stop at the role of shareholder and not go any further. In other words, the role of shareholder and that of management must be separated. Local governments enjoy the rights and exercise the powers of shareholders, but the operation and management powers should be independently exercised by LGFVs themselves.
And in order to reduce the feel of “tool”, LGFVs need to actively introduce private investment and other qualified investment, so as to diversify investors and improve the shareholder structure. Meanwhile, they should establish internal decision-making and supervisory organs that meet the requirements of relevant laws and regulations and can really function, enhance corporate governance and information disclosure, and make efforts to operate in an independent, transparent and market-based way.
The financing practice of LGFVs should also be normalized and regulated pursuant to the principle of market orientation. On the one hand, the requirements of the LGFV Notice and the Implementation Notice shall be fully carried out, and local governments shall be strictly prohibited from providing security for the LGFVs’ financing practice in violation of law. On the other hand, as recently stressed by the CBRC, banks should further improve their lending practice to LGFVs. For example, they should truly and objectively evaluate the credit risk of LGFVs according to the principle of market orientation, and base their lending and pricing decisions on the evaluation result. They should also control the total amount of loans to LGFVs, set strict conditions for new LGFV loans, and strengthen the examination and approval mechanism for such loans. 
B. Establishing and improving the local government bond issuance system
Article 28 of the Budget Law generally prohibits local governments from issuing bonds. This leads local governments to rely too much on bank loans, and meanwhile induces them to issue bonds in a disguised way through LGFVs. In the aftermath of the Global Financial Crisis, the State Council approved the Ministry of Finance (MOF) to issue 200 billion yuan of local government bonds (herein “local bonds”) on behalf of provincial governments and certain municipal governments. The repayment of the principal and interests must also be made through the agency of the MOF. From 2010 to 2013, the annual size of such issuance was 200 billion, 200 billion, 250 billion and 350 billion, respectively. Meanwhile, in 2011, the State Council further approved Shanghai Municipality, Zhejiang Province, Guangdong Province and Shenzhen Municipality to issue bonds by themselves on a pilot basis. Those governments may themselves organize the issuance of local bonds within the amount approved by the state council, but the repayment still needs to be made through the MOF. In 2013, the scope of the pilot was expanded to include Jiangsu Province and Shandong Province. 
In essence, both the “issuance and repayment through the MOF” model and the “issued by local governments and repaid through the MOF” model are using the credit of the central government as backing. Thus such bonds are not local bonds in a strict sense, and might be understood as some kind of “quasi-treasury bonds”. The insufficiency of those arrangements is obvious. First, the size of issuance is limited and can not fully satisfy the financing demands of local governments. Second, local governments are restricted in many ways during the issuance and/or repayment process, with low level of autonomy and flexibility. Third, such “local bonds” are essentially backed by the credit of the central government, which renders investors difficult to make differentiated, market-based investment decisions on the basis of their evaluation of the different credit and risk conditions of different local governments.
Experiences in managing local debts from other countries show that in a tax-sharing system, it’s necessary and reasonable for local governments to have moderate debt raising power. In other words, it’s the due right of local governments to raise debt under a normal tax-sharing system. In fact, in most market economy countries, it’s quite common for local governments to issue bonds. Therefore, rather than impose a general prohibition which to a large extent remains on paper, it’s better to allow local governments to issue bonds on their own right and put them under proper supervision and regulation.
The central government seems to be moving towards this direction. As a latest development, the State Council approved ten provinces and big cities to issue local bonds within the approved quota and repay the principal and interests all by themselves on a pilot basis in 2014.  Also, it is reported that the draft amendment to the Budget Law that is under the examination of the Standing Committee of the 12th National People’s Congress (NPC) proposes to lift the ban and allow local governments at the provincial level to issue bonds, with the size of issuance to be approved by NPC or its Standing Committee. 
Consistent with the move towards allowing local governments to issue bonds more freely, a credit rating system for local governments should be established and improved. The credit of local governments, as the issuer, should be comprehensively evaluated and rated by professional rating agencies according to market-based principles and standards, and the ratings should serve as a precondition and price basis for the issuance. Foreign experiences have already proved the significance of the credit rating system to the control of risk. Statistics show that from 1970 through 2011, among the rated local bonds only 71 bonds defaulted, while among the unrated local bonds 2,571 bonds defaulted.  In this sense, it’s good news to see that the MOF called for such credit rating system to be gradually established in China. 
C. Properly dividing the fiscal power and expenditure responsibility between the central and local governments
As mentioned above, there has been a mismatch between the fiscal power and administrative power of local governments, where national income mainly goes to the central government while expenditure responsibilities mostly remain with local governments. This partly caused the spending capacity of local governments to fall short of the need to provide public services, and was one of the major reasons for the continuing expansion of local debts. Therefore, in addition to establishing and improving the local bond issuance system, the division of the fiscal powers and expenditure responsibilities between the central and local governments needs to be duly adjusted. The first step should be to adjust the division of expenditure responsibilities. Efforts need to be made to define as soon as possible the administrative powers and expenditure/fund-raising responsibilities of the governments at all levels (central and local) in the area of basic public services, drawing experience from the general practice of market economy countries.  In this process, as pointed out by the Central Committee of the Communist Party of China, the administrative powers and expenditure responsibilities of the central government should be strengthened, and the central and local governments should then assume and share expenditure responsibilities according to the adjusted division of administrative powers.  Then the next step should be to make necessary amendment to the Budget Law and relevant tax laws and regulations, to provide local governments fiscal powers matching their administrative powers and expenditure responsibilities.
Furthermore, currently local governments in China to a large extent are still playing the role of “unlimited government” or “omnipotent government”, perform many functions that can and should be performed by the market. This is the real root of the booming LGFVs and mounting local debts. In order to contain the impulse of local governments to invest on borrowing, the government functions must be transformed to reduce direct involvement and intervention in the economy, and underline and highlight the social management and public service functions. The aim is to transform local governments from unlimited, omnipotent governments into limited, service-oriented governments. This is the ultimate solution in my opinion.
* Professor, Institute of Law, Chinese Academy of Social Sciences. I am grateful to Professor Benjamin Liebman and Professor Curtis Milhaupt, Columbia University Law School, for their precious suggestions and comments. All the errors and omissions remain my own.
 According to the definition of the National Audit Office (NAO), “local governmental debts” refer to the debts of all the four-level local governments (i.e., provinces, cities, counties and townships), including debts that local governments are directly liable to repay (herein “government debts”), debts guaranteed by local governments (herein “explicitly guaranteed debts”), and debts where local governments might be obliged to render certain assistance (herein “implicitly guaranteed debts”). The latter two are so-called “contingent liabilities” of local governments. See NAO, Audit Result on Nationwide Governmental Debts (QuanGuo Zhengfuxing Zhaiwu Shenji Jieguo), December 30, 2013, p. 1, available at http://www.audit.gov.cn/n1992130/n1992150/n1992500/3432077.html, last visited on May 18, 2014.
 Among the total amount, government debts, explicitly guaranteed debts and implicitly guaranteed debts were 10.89 trillion Yuan ($1.79 trillion), 2.66 trillion Yuan ($0.43 trillion) and 4.34 trillion Yuan ($0.71 trillion), respectively.
 See NAO, supra note 1, pp. 6-7. Other borrowers include government departments and agencies, public institutions, state-owned enterprises and public utilities. Ibid.
 Before 1980, under the "unified income and expenditure" fiscal system, almost all of local government revenue and profits were submitted to the central government, and were then redistributed to the provinces according to an expenditure plan. From 1980 to 1994, China used a divided budget system whereby central and local governments had their own independent financial interests. In 1994, China cancelled the contract system for taxation and began to implement the tax-sharing system. The main goal of this new policy was to strengthen the control of central government over sources of tax revenue, to raise the proportion of the central fiscal revenue in the national fiscal revenue and the proportion of fiscal revenue in GDP, and to resolve the conflict of interest between the central and local authorities. Therefore, taxes were divided into a central fixed tax, a local fixed tax, and a shared tax between the central and local authorities with a view to stabilizing the financial relationship between the two. Obviously, however, the objective of the tax-sharing system has not only been to improve the fiscal system but also to raise the proportion of the tax revenue that the central government would collect. See Miguel Elosua et al., Urbanisation in China: The Impact of the Tax-sharing System and the Definitions of New Strategies, UrbaChina Working Paper Series, No. 1 / November 2013, http://hal.archives-ouvertes.fr/docs/00/90/32/18/PDF/UrbaChina_working_paper_no1.pdf, last visited on May 18, 2014.
 Central government revenues jumped in the years after 1994 as a result, while local governments saw their share of revenues fall from 78 percent in 1993 to 45 percent in 2002. See Henry Sanderson & Michael Forsythe, China’s Superbank: Debt, Oil and Influence—How China Development Bank is Rewriting the Rules of Finance, John Wiley & Sons Singapore Pte. Ltd., 2013, p. 5.
 See Budget Law of the People’s Republic of China (passed on March 22, 1994 and effective as of January 1, 1995, herein “Budget Law”), Article 28, which provides that “local budgets at various levels shall be compiled according to the principles of keeping expenditures within the limits of revenues and maintaining a balance between revenues and expenditures, and shall not contain deficit. Except as otherwise prescribed by law or by the State Council, local governments may not issue local government bonds”.
 See Sanderson & Forsythe, supra note 6, p. 5.
 See Sanderson & Forsythe, supra note 6, p. 13.
 According to Lenin, state capitalism is a form of capitalism associated with state power and controlled and dominated by the state. Similarly, western scholars describe state capitalism as a form of capitalism dominated by the state and government, with economic growth as the core aim and by means of supporting specific enterprises or industries. See Hu Leming et al., State Capitalism and “China Model” (Guojia Ziben Zhuyi Yu “Zhongguo Moshi”), Economic Research Journal (Jingji Yanjiu), Issue 11, 2009, pp. 31-32.
 See Zheng Yongnian, De facto Federalism in China: Reforms and Dynamics of Central-local Relations (Zhongguo De Xingwei Lianbangzhi: Zhongyang-Difang Guanxi De Biange Yu Dongli), The Oriental Press, 2013, Preface to Chinese version. This is a translation from the original English version published in 2007 by World Scientific Publishing Co. Pte. Ltd.
 See NAO, supra note 1, p. 1.
 State Council, Notice of the State Council on Certain Issues Related to Strengthening the Regulation of Local Government Financing Vehicle Corporations (Guowuyuan Guanyu Jiaqiang Difang Rongzi Pingtai Gongsi Guanli Youguan Wenti De Tongzhi, herein “LGFV Notice”), issued on June 10, 2010.
 Ministry of Finance, National Development and Reform Commission, People’s Bank of China & China Banking Regulatory Commission, Notice on the Relevant Issues in Implementing the Notice of the State Council on Certain Issues Related to Strengthening the Regulation of Local Government Financing Vehicle Corporations (Guanyu Guanche Guowuyuan Guanyu Jiaqiang Difang Rongzi Pingtai Gongsi Guanli Youguan Wenti De Tongzhi Xiangguan Shixiang De Tongzhi, herein “Implementation Notice” ), Issued on July 30, 2010.
 Gu Gongyun & Hu Gairong, How to Regulate Local Government Financing Vehicles (Ruhe Guizhi Difang Rongzi Pingtai), Capital Shanghai (Shanghai Guozi), Issue 4, 2012, pp. 18-19.
 See Hongyuan & Guoping, Debt Risk of Local Government Financing Vehicles for the Perspective of “Land Finance”: Using the Financing Vehicles of City C in the Central Region as Example (“Tudi Caizheng” Shijiao Xiade Difang Zhengfu Rongzi Pingtai Zhaiwu Fengxian Yanjiu: Yi Zhongbu Diqu C Shi Rongzi Pingtai Weili), Journal of Xi’an University of Finance and Economics (Xi’an Caijing Xueyuan Xuebao), Issue 5, 2012, p. 74.
 See Article 28 of the Budget Law, supra note 7. See also Article 30 of the Law of the People’s Republic of China on the People’s Bank of China (revised on December 27, 2003 and effective as of February 1, 2004), stating that “the People’s Bank of China Shall not make loans to local governments or government departments at all levels”; Article 17 of the Lending General Provisions of the People’s Bank of China (promulgated on June 28, 1996 and effective as of August 1, 1996), stating that “the borrows shall be…enterprises or public institutions as legal persons, other business organizations, individual industrial and commercial households and…natural persons”; and Article 8 of the Security Law of People’s Republic of China (passed on June 30, 1995 and effective as of October 1, 1995, herein “Security Law”), stating that “state organs shall not be guarantors”.
 Established in 1994, CDB was originally a policy bank. Not allowed to take deposit, it could only raise funds through selling bonds to commercial banks. Pushed by its former governor Chen Yuan, CDB underwent a reform of commercialization, and was approved by the State Council in 2008 to transform into a commercial bank. It was authorized to step into the business of commercial banking, including taking public deposits except household deposits. See Niu Juanjuan, CDB Becomes the Fifth Largest State-owned Commercial Bank after the Transformation (Guokaihang Gaizhi Hou Cheng Diwujia Daxing Guoyou Shangye Yinhang), Financial News (Jinrong Shibao), December 23, 2008, Page 1. However, five years has gone and little seems to have been really done towards becoming a real commercial bank. There have even been signs of “returning to a policy bank”. For example, nowadays CDB still relies on bond market instead of taking public deposits. See Liu Zhongsheng et al., CDB Reform Marches Backwards (Guokaihang Gaizhi Kaidaoche), http://www.nbd.com.cn/articles/2013-09-25/775973.html, last visited on May 18, 2014.
 See Sanderson & Forsythe, supra note 6, pp. 5-6.
 The tax-sharing system remarkably changed the allocation system of national income as well as the percentage of central government and local governments in the total income. The allocation of expenditure responsibilities, however, has not been adjusted accordingly. The result is a situation where national income mainly goes to the central government while expenditure responsibilities mostly remain with local governments. Statistics show that since 1994, the percentage of local fiscal income in the total fiscal income sharply decreased from 78% in 1993 to around 45% for a long time, while the percentage of local fiscal expenditure in the total fiscal expenditure increased from 71.3% in 1993 to around 80% in 2009. See Luo Guimei, A Study on LGFVs and the Related Risks (Difang Zhengfu Tourongzi Pingtai Ji Xiangguan Fengxian Yanjiu), Master’s Thesis, Southwestern University of Finance and Economics, April 2011, pp. 17-18.
 See Li Xiaowu, Legal Thinking on the Borrowing and Financing System of Local Governments (Difang Zhengfu Juzhai Rongzi Zhidu De Falv Sikao), Jiangxi Social Sciences (Jiangxi Shehui Kexue), Issue 8, 2013, p. 160.
 See PBC & CBRC, Guiding Opinions on Further Adjusting the Credit Structure to Promote Steady and Rapid Growth of National Economy (Guanyu Jinyibu Jiaqiang Xindai Jiegou Tiaozheng Cujin Guomin Jingji Pingwen Jiaokuai Fazhan De Zhidao Yijian), March 18, 2009.
 See Ma Jun, State Governance and Wealth Management: Public Budget and State Building (Zhiguo Yu Licai: Gonggong Yusuan Yu Guojia Jianshe), SDX Joint Publishing Company, 2011, pp. 17-18.
 Huang Zhilong, Problems of and Suggestions on the LGFVs (Difang Zhengfu Rongzi Pingtai Zhaiwu Wenti Yu Duice Jianyi), China Price (Zhongguo Wujia), Issue 8, 2011, p. 30.
 See Yang Song & Zhang Yongliang, LGFVs’ Direction of Development (Difang Zhengfu Rongzi Pingtai De Fazhan Fangxiang), Law Science (Faxue), Issue 10, 2012, pp. 31-32.
 See Li Xiao’an & Zhou Xuzhong, Credit Risk of LGFVs and the Disorder of Local Government Credit: from the Perspective of Law (Tourongzi Pingtai Gongsi Xinyong Fengxian Yu Difang Zhengfu Xinyong Shixu: Yi Falv Wei Shijiao), Journal of Beijing Administrative College (Beijing Xingzheng Xueyuan Xuebao), Issue 2, 2012, p. 89.
 Gu & Hu, supra note 16, p. 18.
 According to Article 4 of the Implementation notice, “directly or indirectly provide financing guarantees for the LGFVs” includes without limitation the following: issue guarantee letter for LGFVs; promise to give liquidity support and provide provisional repayment funds in case LGFVs encounter difficulty in repayment; promise to assume part of the repayment obligation in case LGFVs can not repay the debts; promise to incorporate the repayment funds of the LGFVs into the government budgets.
 See Article 7 of the Judicial Interpretation of the Supreme People's Court on Some Issues Regarding the Application of the Security Law of the People's Republic of China (promulgated on December 8, 2000 and effective as of December 13, 2000).
 Specifically, those LGFVs that are used to finance not-for-profit projects and mainly rely on fiscal funds to repay debts shall not undertake financing tasks any further, and the relevant local governments shall, after make clear the repayment liability and put in place repayment measures, deal with those LGFVs properly. As to those LGFVs that in addition to financing the above mentioned not-for-profit projects, are also responsible for the construction and operation of such projects, they shall, after the repayment liability is made clear and repayment measure put in place, be divested of their financing business and not retain the function of financing platform. See Article 2 of the LGFV Notice.
 See CBRC, Guiding Opinions on Strengthening the Risk Regulation of the LGFV Loans in 2013 (Guanyu Jiaqiang 2013 Nian Difang Zhengfu Rongzi Pingtai Daikuan Fengxian Jianguan De Zhidao Yijian), April 9, 2013.
 See MOF, Measures for the Pilot Scheme of Self-Issuance of Bonds by Local Governments in 2013 (2013 Nian Difang Zhengfu Zixing Fazhai Shidian Banfa, herein “Self-Issuance Measures”), promulgated and effective as of June 25, 2013.
 Research Institute for Fiscal Science of the Ministry of Finance, China’s Local Governmental Debt Risk and the Countermeasures (Woguo Difang Zhengfu Zhaiwu Fengxian Yu Duice), Review of Economic Research (Jingji Yanjiu Cankao), Issue 14, 2010, p. 21.
 See MOF, Measures for the Pilot Scheme of Self-Issuance and Self-Repayment of Local Government Bonds in 2014 (2014 Nian Difang Zhengfu Zhaiquan Zifa Zihuan Shidian Banfa), promulgated and effective as of May 19, 2014.
 See Guan Qingfeng, The Ban on Local Bonds Is to Be Lifted, Precautions Shall be Taken against Provincial Government Lobbying Beijing for Higher Quota (Difangzhai Kaizha Yingfang Shengzhengfu Jinjing Paozhai), Beijing Youth Daily (Beijing Qingnian Bao), April 24, 2014, Page A07.
 See Liu Baoliang, A Thorough Survey on the Size of Local Debts: Regulators Push for Credit Rating (Difangzhai Da Modi: Jianguanceng Litui Xinyong Pingji), China Economic Herald (Zhongguo Jingji Daobao), October 24, 2013, Page B01.
 See Self-Issuance Measures, Article 19.
 See Research Institute for Fiscal Science of the Ministry of Finance, supra note 36, p. 24.
 See the Decision of the Central Committee of the Communist Party of China on Some Major Issues Concerning Comprehensively Deepening the Reform (adopted at the Third Plenary Session of the 18th Central Committee of the Communist Party of China on November 12, 2013), V. 19.