The Law of China’s Local Government Debt: Local Government Financing Vehicles and Their Bonds

The Law of China’s Local Government Debt: Local Government Financing Vehicles and Their Bonds Abstract Local government financing vehicles (LGFVs)—companies capitalized and owned by local government and established for the purpose of raising funds for municipal infrastructure construction—emerged in China in the 1980s as a response to the severe constraints on indebtedness by local governments themselves. The mushrooming of their number and indebtedness has sparked fears about their ability to repay the debt and the consequences of a default. In addition to taking on bank debt, a number of LGFVs have also issued bonds. While observers have questioned the value of collateral typically offered as security for the bonds, we know of no extensive analysis to date of the legal quality of the collateral: what exactly are the bondholders being promised, and what is the status of those promises in the Chinese legal system? This Article is an attempt to answer that question, using data from two hand-collected samples of LGFV bond prospectuses from different regions in China in two different time periods. We find that current collateralization practices vary a great deal across bond issues and have changed over time, and we discuss the legal and other problems attendant upon each type of backup. Remarkably, we find that, unlike our initial sample of bond issues, recent bond issues virtually all state explicitly in the prospectus that they carry no security. Thus, the popular image of local governments wildly overpromising with guarantees they are not legally empowered to give seems, at least as far as recent bond issues are concerned, to be wholly wrong. This in turn calls into question the figures commonly provided for local government debt, since they often include LGFV debt that local government is neither legally nor morally obligated to pay. To be sure, they may wish to pay creditors voluntarily, but it is misleading to label as “debt” soft obligations of this nature. Creditors who have tried to force local governments to make good on their guarantees have uniformly failed, at most receiving half of what they sought. The argument that local governments have some politically enforceable obligation to pay on their guarantees does not seem supported by the evidence. Introduction Local government financing vehicles (LGFVs)—that is, companies capitalized and owned by local government1 and established for the purpose of raising funds for municipal infrastructure construction—emerged in China in the 1980s as a response to the severe constraints on indebtedness by local governments themselves.2 In the last several years, their number and indebtedness have mushroomed, sparking fears about their ability to repay the debt and the consequences of a default. In addition to taking on bank debt, a number of LGFVs have also issued bonds, something generally prohibited to local governments. While observers have justifiably questioned the economic value of land and other collateral typically offered as security for the bonds,3 we know of no extensive analysis to date of the legal quality of the collateral: what exactly are the bondholders being promised, and what is the status of those promises in the Chinese legal system? This Article is an attempt to answer that question. The primary data for this study comes from two hand-collected samples of LGFV bond prospectuses from different regions in China in two different time periods.4 Group I consists of nineteen issues from 2009 to 2011, and Group II consists of twenty issues from 2014. The prospectuses in Group I were selected with a view to understanding the picture in various parts of the country, but they were not selected on the basis of the repayment safeguards offered. The Group II prospectuses were selected entirely randomly from a large group of 2014 bond issues. We begin with a brief overview of the Chinese LGFV financing landscape. We then discuss in detail the main categories of backup5 for the bondholders. We find that current collateralization practices vary a great deal across bond issues and have changed over time. The three main ways—typically alternative and not additive—in which bonds are backed up are land use rights, local government guarantees, and guarantees from third-party companies. Some bond issues are explicitly not backed up by anything other than the issuer’s own assets. We discuss the legal and other problems attendant upon each type of backup. Remarkably, we find that, unlike our initial sample of bond issues, recent bond issues virtually all state explicitly in the prospectus that they carry no security. Thus, the popular image of local governments wildly overpromising with guarantees they are not legally empowered to give seems, at least as far as recent bond issues are concerned, to be wholly wrong. This in turn calls into question the figures commonly provided for local government debt, since they often include LGFV debt which, as we shall show, local government is neither legally nor morally obligated to pay. It may turn out that handwringing over LGFV debt is no longer necessary. In 2015, a new Budget Law came into effect that promised to regularize the issue of bonds by provincial-level governments, and in the spring of 2015 the Chinese government announced a plan whereby LGFV debt would be replaced by bonds issued by provincial-level governments.6 These policies have the potential to fundamentally change the regime described here; if the swap is completed, for example, many of the issues raised here would essentially be finessed away. Nevertheless, even if the swap is successful, we believe that the legal issues involved in LGFV debt are very likely to reappear at another time in another context, and therefore are still worth exploring and clarifying. In particular, it is important to clarify the legal picture in order to assess the government’s response now and its likely response in similar situations in the future. I. Municipal Government Financing and LGFVs A. Municipal Finance Generally LGFVs arose as a response to an irresistible force and an immovable object. The irresistible force was the heavy responsibility of local governments for the provision of services, a responsibility often unfunded by the central government. The immovable object was the inability of local governments to raise money through taxation or borrowing: local governments are not allowed to run budget deficits and may not generally issue bonds.7 The result has been the toleration—and during the financial crisis, the encouragement—of backdoor practices that permit subnational governments (SNGs) to meet their spending responsibilities. 1. Responsibility for Provision of Services Municipal governments in China are responsible for a vast array of government expenditures, including all vital services and infrastructure. Although China is formally a unitary state, in 2008,8 for example, central government expenditures were just 21% of national budgetary expenditures.9 By contrast, even federal states such as Canada and the United States exhibit much higher percentages: 43% (in 2002) and 51% (in 1999) respectively.10 At the same time, the intergovernmental financial system is weak. SNGs suffer large vertical fiscal gaps—that is, a mismatch between assigned tax revenues and assigned expenditures. For example, in 2008, SNGs were responsible for 79% of government expenditures but were directly allocated only 43% of tax revenues.11 The remaining expenditures must be met from intergovernmental transfers and nontax sources of revenue, and transfers are inadequate.12 2. Restrictions on Revenue Raising In spite of their considerable responsibilities, municipal governments and SNGs generally suffer considerable restrictions on their revenue-raising abilities. a. Restrictions on Taxing Power SNGs cannot meet financial needs simply by imposing new taxes at will. Nevertheless, the often-heard assertion that local governments in China have no taxing power needs to be qualified. It is true that both the 1992 Law on the Administration of Tax Collection13 and a 1993 decision of the State Council14—the decision that was the core of the 1994 tax system reform—purport to grant exclusive authority over taxes to the central government. But the rules contained in these documents have not been strictly adhered to by the central authorities themselves, and local governments have consistently maintained the power in practice to reduce effective tax rates through measures such as rebates.15 In the words of a leading scholar of the Chinese tax system, these and other facts cast doubt on whether the centralization of tax legislative power is indeed a fundamental normative principle for Chinese law and policy, and whether it is not instead, as a more skeptical view would hold, a mere slogan wielded by the central government as a non-neutral party in political contests. . . . [F]iscal federalism in taxation in China is still very much in an unsettled state, and does not represent a coherent model.16 As a legal matter, therefore, it is not completely clear that SNGs, at least in the form of People’s Congresses17 at the level of the province and certain cities, cannot impose (in substance or in effect) new taxes. As a political matter, on the other hand, it is clear that the central government almost always disfavors local imposition of new taxes, and in practice SNGs have not been imposing new taxes as a response even to severe financial pressures. Moreover, the center’s share of tax revenue, after subtracting revenue collected by the center and then returned to provinces, has been steadily increasing since the 1994 tax reform: from 21.2% in 1994 to 46.1% in 2014.18 b. Restrictions on Ability to Borrow SNGs are severely restricted in their ability to borrow from sources other than superior levels of government, who may not be willing to lend. It is not always clear, however, what the various prohibitions actually mean within the Chinese legal system. They might mean that the prohibited debt is legally void and unenforceable, like a contract entered into by a minor. The prohibited party is simply incapable of entering into a binding debt agreement. But a prohibition could instead mean that the legal system recognizes the validity of the debt, but that the governmental body or particular officials responsible for incurring it are subject to some kind of administrative or other sanction.19 If the legal system did not recognize the validity of the debt, the punished misbehavior could not occur. i. Bank Debt Governmental units in China have limited legal authority to borrow from banks. It is commonly said that SNGs are forbidden to borrow from banks,20 and there is some legal basis for this claim. The 1996 Lending Principles issued by the People’s Bank of China (PBOC), at that time the bank regulator, forbid lending to entities other than natural persons with legal capacity or enterprise legal persons, other economic organizations, or individual industrial-commercial households, in each case properly registered with the State Administration of Industry and Commerce.21 This list of permissible borrowers does not include governmental units, and so has been interpreted by some as banning loans to them.22 Note, however, that the Lending Principles can be viewed as a kind of internal directive within the banking system. Thus, it does not necessarily make loans to governmental units invalid, except perhaps via an argument that the borrower knew or should have known that the lender had no authority to lend. While the PBOC Lending Principles are at least arguably a lender-side prohibition, the Budget Law (in both the original 1994 version and its new 2015 version) has what can also be viewed as a borrower-side prohibition. Local governments are required to balance their budgets; they may not have deficits, and they are specifically forbidden to issue bonds absent authorization by statute or the State Council.23 Since the preceding article in the Budget Law specifically provided for the conditions under which the central government may take loans, it is reasonable to infer that, at least as far as the Budget Law is concerned, local governments may not do so under any conditions. On the other hand, articles 55 and 73 of the 1994 Budget Law clearly contemplated that borrowing could be a legitimate part of a SNG budget—although interestingly their corresponding articles in the 2015 Budget Law (articles 70 and 92) do not. Moreover, official documents and reports appear to confirm that SNGs do borrow directly from banks.24 An important and widely cited 2011 National Audit Office (NAO) report on local government debt explicitly distinguished LGFV debt from debt borrowed and owed directly by local governments, and found the latter to be substantial25 (but also unlawful).26 Whether bank debt is viewed as lawfully or unlawfully incurred, however, commentators appear to be unanimous that the obligation is legally valid. It is not, for example, similar to gambling debts or an obligation to pay usurious interest rates, both of which are unenforceable in some jurisdictions. The constant stream of government documents sternly warning local governments not to incur debt illegally is evidence of this.27 If the obligation were invalid, there would be no need to be concerned about local governments incurring it. At least in theory, the government could effectively eliminate all unlawful local government debt incurrence by simply making it known that creditors’ rights against local governments per se28 would not be enforced and their money would be lost, and following up with appropriate court action (or more likely, inaction). Unlawful lending would dry up immediately. Why the government has not adopted this policy option is not clear. There may be practical considerations of which we are not aware that vitiate the theoretical outcome. On the other hand, a recent Supreme People’s Court case suggests that the government may have decided to signal that it will pursue precisely this policy.29 ii. Bonds Local governments were specifically forbidden under the 1994 Budget Law to issue bonds except as permitted by statute or the State Council.30 As bond issues are highly visible, this prohibition was generally effective. Nevertheless, one local government—that of Qingyang District in Chengdu, Sichuan—managed astonishingly to issue bonds for 375 million yuan over ten years, from 1992 to 2002, before the violation was discovered by the Ministry of Finance and shut down.31 Only in 2011 did the State Council first provide the explicit permission required by law for four local governments (Shanghai, Zhejiang, Guangdong, and Shenzhen) to issue bonds directly,32 and in November of that year, the Shanghai municipal government issued bonds in the amount of 7.1 billion yuan.33 More recently, the provincial governments of Shandong and Jiangsu were granted permission to issue bonds.34 In general, therefore, fundraising through bonds had to be done through LGFVs—making such bonds in form corporate bonds—or not at all. There appears to have been some debate over how the revised Budget Law should treat local government bond issues, and successive drafts treated the matter differently. In the end, however, the final version of the amendments passed by the National People’s Congress Standing Committee provides that provincial-level governments may, with State Council approval, issue bonds.35 Analysts have, without contradiction from the central government, universally interpreted the revisions as liberalizing the regime for local government bond issues. This slightly misstates the content of the revisions. By adding a considerable amount of text to set forth the process for issuing local government debt, the revisions certainly hint that more such debt is contemplated. But although the central government may plan on approving the issuance of more local government debt, the approval process is actually tightened. Under the old Budget Law, local governments at many levels could issue bonds for many purposes, and State Council approval was adequate. Under the revised law, bond issues may be by provincial-level governments only and may not be for commercial projects. Moreover, the revisions add the requirement that an overall quota for local government debt must be approved by the National People’s Congress or its Standing Committee.36 The revisions could most accurately be understood as signaling that the government intends to approve quotas for local government debt as a routine matter and not as an exceptional measure, but that the debt issuers and the approval process will be tightly controlled from the top. B. LGFVs as a Response 1. History of LGFVs In order to avoid the constraints on direct borrowing, local governments began as early as the 1980s to turn to special-purpose vehicles (despite the absence at the time of a law on corporate organization) to raise funds for public investment. These entities were often created for a single purpose—construction of toll roads or utilities, for example, because such projects had a revenue stream that could be used to service the debt—and could take loans from international financial institutions.37 A breakthrough came in 1992 with the formation of the Shanghai Municipal Construction Investment and Development General Company. This was a general-purpose entity whose mission was the construction of facilities such as water supplies, roads, and sewers. This took the financing available for urban infrastructure in Shanghai from a few billion yuan per year to 17 billion in 1993 and 38 billion in 1994.38 This model soon spread to other areas, and most cities had established investment corporations by 2000.39 The separation from public finance became less stringent, and some local governments offered guarantees to lender banks.40 The investment companies also sold bonds; by the end of 2008, there were thirty-one LGFV bond issues totaling 46 billion yuan.41 The next stage—and the real leap forward—in LGFV development was occasioned by the financial crisis. The Chinese government announced a 4 trillion yuan stimulus, much of it to be funded by local government borrowing through LGFVs.42 In March 2009, the People’s Bank of China and the China Banking Regulatory Commission issued a joint notice encouraging the use of LGFV financing,43 and borrowing took off. Over 3,200 SNGs in China have the authority to establish LGFVs: 31 provincial-level units, 333 prefectural-level units, and 2,859 county-level units. They had established 10,321 LGFVs by the end of 2009, an increase of 7,000 over 2008.44 In 2009 alone, there were 115 LGFV bond issues in the total amount of 167.7 billion yuan. By 2009, LGFVs had already accumulated 5 trillion yuan in bank loans, and in that year alone took on another 3 trillion.45 Outstanding loans to LGFVs represented 240% of local government income, and 90% of the loans came from banks.46 According to the National Audit Office, which appears to calculate LGFV debt more conservatively, by the end of 2010, LGFVs carried a debt load of over 4.97 trillion yuan;47 by mid-2013, this had increased to 6.97 trillion yuan, almost 40% of the total 17.9 trillion yuan debt load of local governments.48 As state agencies became aware of the ballooning debt, they realized at the same time that very little was known about LGFVs. Since the middle of 2009, therefore, the government has engaged in a massive effort to find out more about them, and the State Audit Office launched a major investigation in the spring of 2011 covering all thirty-one provincial-level jurisdictions and involving 41,300 auditors. Furthermore, in 2010, the State Council issued a document intended to rein in their activities.49 This apparently had some effect: according to People’s Bank of China, the growth in annual borrowing by LGFVs slowed to under 20% in 2010, from over 50% in 2009.50 One source asserts that LGFVs incurred only 300 billion yuan in new debt in 2011.51 Nevertheless, a National Audit Office study of thirty-six local governments found that their indebtedness had increased by almost 13% over the two years of 2011 and 2012.52 As of the end of 2014, officially counted local government debt was put at 15.4 trillion yuan by China’s Minister of Finance, with 1.06 trillion yuan as approved government bond issues and the rest bank loans and borrowing through LGFVs.53 It was expected to reach 16 trillion by the end of 2015.54 Despite any slowing in the rate of increasing indebtedness, however, it seems clear that the total amount is unlikely to be repayable.55 In early 2012, central and local government officials began negotiating with banks over how to deal with the estimated 1.8 trillion yuan in loans to LGFVs scheduled to mature in 2012.56 According to contemporary news reports, “[b]anks are being asked to rejigger repayment schedules and let borrowers settle maturing debt with fresh loans under a risk-mitigation plan drafted by the central government.”57 A 2013 report by Nomura stated that half of LGFV debt due on 2012 would have been at risk of default were it not for local government support.58 Moreover, in late 2012, several Chinese government departments issued yet another joint notice intended to rein in LGFV borrowing, indicating that the problem had yet to be brought under control.59 News reports indicate that about three quarters of the 4 trillion yuan that local governments were to repay to banks by the end of 2012—this figure likely includes, but is not limited to, LGFV debt—was in fact rolled over.60 More recently, on December 31, 2013, the National Development and Reform Commission officially authorized local governments to roll over LGFV debt through yet more LGFV bond issues.61 The last two years have seen yet more government ambivalence about LGFVs. In October 2014, the State Council issued its famous Document 43,62 which was intended to put the brakes on LGFV financing.63 It essentially said that local governments could no longer borrow money through LGFVs. In the same month, the National Development and Reform Commission, which has jurisdiction over corporate bonds generally, issued a document raising the bar for corporate bond issues.64 In May 2015, however, perhaps concerned by an economic slowdown, three government agencies jointly issued Document 40,65 which was widely viewed as reversing many of the policies embodied in Document 43. In the words of one report, Under [Document 43], [LGFVs] were barred from borrowing additional funds starting this year, as the government sought to close what it dubbed “the back door” for localities to borrow. Instead, all borrowing would be done by the local governments themselves and be appropriately disclosed and reflected in their budget plans. The purpose was to rein in runaway local-debt growth and make local borrowings more transparent. According to the latest directive, local financing firms can continue to get loans from banks to fund ongoing projects. If the local firms have trouble repaying their bank debts, the rule says, their loan contracts should be “renegotiated and extended.”66 2. Structure of LGFVs LGFVs are usually organized under China’s Company Law as limited liability companies. Usually, either the local government itself or the local branch of the State Asset Supervision and Administration Commission (SASAC) is the sole shareholder.67 In some cases, the LGFV is a joint venture among two or three entities such as state-owned enterprises (SOEs) affiliated with the local government and other government or quasi-governmental agencies.68 Most LGFVs focus on local infrastructure construction projects as well as other public welfare projects such as education, environmental protection, and medical care.69 Some engage in side business activities of various kinds as well.70 In 2010, the State Council issued a notice that for various purposes classified LGFVs into three classes.71 The first class of LGFVs are those used to finance nonprofit public welfare projects which will not generate income, and therefore will pay off their debts mainly by cash transfers from government funds. The second class of LGFVs are those used to finance public welfare projects that return a steady profit, and therefore are expected to pay their debts with income from those projects. The third class of LGFVs are those used to finance nonpublic welfare projects.72 Many government agencies, including the National Development and Reform Commission and the Ministry of Finance, have issued various directives or notices that interpret and implement the Notice.73 3. LGFV Bonds As discussed above, while local governments may not issue bonds, companies may, and LGFVs are companies. As of 2013, approximately eight or nine hundred LGFVs out of a total of about 10,800 had publicly issued bonds.74 LGFV bonds are usually underwritten by a group of underwriters led by a lead underwriter and issued to Chinese institutional investors. Some bond issuances, however, are clearly aimed at non-Chinese investors. For example, the offering prospectus in an issue by a Beijing LGFV contains the following language: Pursuant to this offering, the Securities may not be offered or sold in the United States, except that the Securities may be offered or sold to persons who are “qualified institutional buyers,” as defined in Rule 144A under the Securities Act in reliance on the exemption from the registration requirements of the Securities Act provided by Rule 144A.75 The underwriters in that bond issue included BofA Merrill Lynch, Morgan Stanley, and Credit Suisse, among others, and the governing law was stipulated to be New York law.76 C. Confusion in the System Although the rules and principles of government and LGFV borrowing seem fairly clear, there is evidence of a great deal of confusion and variation in practice. This could be due to the sources’ being careless in their terminology, but could also be a sign that actual practice diverges widely from what the law prescribes. 1. Do LGFV Bonds Have an Implicit Government Guarantee? It is frequently asserted that LGFV bonds carry an implicit government guarantee even where such a guarantee is not explicit. One recent paper even asserts that this guarantee comes from the central government, not just the local government that owns the issuing LGFV.77 We question this assertion. Since an “implicit” guarantee is by definition unstated, no affirmative evidence can be adduced for it. The support for the assertion frequently comes from another assertion that bond buyers believe that the bonds are implicitly guaranteed. We call this the Tinkerbell theory of rights: if you believe in something hard enough, you can bring it into existence. But the Chinese legal system does not adopt this theory. More to the point, empirical evidence contradicts the claim that bond buyers believe an implicit guarantee exists. Most obviously, if LGFV bond buyers truly believed that all LGFV bonds carried a central government guarantee, we would expect to see identical credit ratings on all LGFV bonds. But we do not.78 We would also expect to see identical interest rates on bonds with identical maturities and liquidity characteristics. Again, we do not. A group of scholars working with a dataset of LGFV bond issues from 1997 to 2014 found that LGFV bond interest rates reflected both economic risk and political risk.79 When they were swapped for provincial government obligations under a program introduced in the spring of 2015, Moody’s rated the bonds Triple-A, regardless of which province issued them, on the grounds that they were (in the view of Moody’s) implicitly backed by the central government,80 a judgment Moody’s had not (to the best of our knowledge) made about LGFV bonds. The financial system itself recognized the riskiness of LGFV bonds: the clearing agency for exchange-traded debt announced that LGFV bonds rated lower than AAA could not be used as collateral for short-term loans because they were too risky.81 Finally, there is no support in law for the notion of an implicit guarantee. As discussed below,82 government guarantees are explicitly labeled invalid in the Guarantee Law of the People’s Republic of China (“Guarantee Law”), the only law that could make them valid. Moreover, in case the Guarantee Law was not clear enough, the central government has issued pronouncements both disclaiming its own responsibility for local debt83 and reminding the public that local government guarantees are invalid.84 Above and beyond these theoretical considerations, we now have a bit of empirical evidence as well. Alarming reports have recently surfaced of corporate debtors, both ordinary corporations and LGFVs, defaulting in unprecedented ways. In March 2014, solar equipment maker Chaori defaulted on a bond, marking the first corporate bond default in reform-era China.85 More worrisome from the standpoint of LGFV creditors is the reported first default of an LGFV on a bank loan, which came to light in mid-2014.86 Clearly this is one LGFV whose debt was not backed by the government. In light of the above, it is hard to see what evidence could support the claim that implicit government guarantees either exist or are even widely believed by investors to exist. Both documentary evidence and empirical observation suggest the contrary. 2. What Do National Audit Office Categories Mean? A further source of confusion is the National Audit Office’s way of categorizing LGFV and local government debt. The NAO has carried out two major investigations of LGFV and local government debt, the first in 2011 and the second in 2013. These investigations were carried out pursuant to instructions from the State Council first set forth in a 2011 document.87 That document charged the NAO with classifying local government debt into three types.88 Table 1. Local Government Debt as of June 30, 2013 (in Billions of Yuan).a Type of entity incurring debt  Debt for which local government bears repayment responsibility (Type I debtb)  Contingent government debt  Debt for which the local government is liable as a guarantor (Type II debtc)  Debt for which government may undertake a certain responsibility for assistance (Type III debtd)  Financing vehicle companies  4,075.55  883.25  2,011.64  Government departments and organs  3,091.34  968.42  0  Institutions with subsidized expenses  1,776.19  103.17  515.71  State-owned financing or holding companies  1,156.25  575.41  1,403.93  Institutions responsible for their own revenues and expenditures  346.29  37.79  218.46  Other units  316.26  83.14  0  Public service institutions  124.03  14.39  189.64  Total  10,885.92  2,665.58  4,339.37  Type of entity incurring debt  Debt for which local government bears repayment responsibility (Type I debtb)  Contingent government debt  Debt for which the local government is liable as a guarantor (Type II debtc)  Debt for which government may undertake a certain responsibility for assistance (Type III debtd)  Financing vehicle companies  4,075.55  883.25  2,011.64  Government departments and organs  3,091.34  968.42  0  Institutions with subsidized expenses  1,776.19  103.17  515.71  State-owned financing or holding companies  1,156.25  575.41  1,403.93  Institutions responsible for their own revenues and expenditures  346.29  37.79  218.46  Other units  316.26  83.14  0  Public service institutions  124.03  14.39  189.64  Total  10,885.92  2,665.58  4,339.37  a This Table is a direct reproduction of the data in NAO 2013, supra note 48, at 6–7 tbl.3. Totals may not equal sum of amounts in columns due to rounding. b See the definition in the text accompanying note 89. c See the definition in the text accompanying note 90. d See the definition in the text accompanying note 91. View Large Type I debt is described as “debt for which local government bears repayment responsibility,” and is defined more precisely as follows: Borrowing undertaken by local government (including government departments and agencies), institutions financially supported [by local government], institutions with a public purpose, government financing vehicles, and other relevant units, that is confirmed to be repayable from government budgetary finances, where the government bears direct responsibility for repayment. The first kind is local government bonds, national bonds converted to loans, foreign debt converted to loans, loans for overall development of agriculture, and other governmental debts converted to loans that have been confirmed to be repayable from government budgetary finances. The second kind is debt incurred by government financing vehicles, government departments and agencies, institutions financially supported [by local government], institutions with a public purpose, and other units through borrowing, owing, repurchase obligations, or other means that has been confirmed to be repayable from government budgetary finances (not including revenues from road tolls, tuition, and the like). The third type is policy-related amounts payable by local government food enterprises and supply and marketing enterprises.89 Type II debt is described as “debt for which the local government is liable as a guarantor,” and is defined more precisely as follows: Debt for which a local government (including government departments and agencies) has provided a direct or indirect guarantee such that when the debtor is unable to repay the debt, the government bears joint liability for repayment. The first type is borrowing undertaken by government financing vehicles, institutions financially supported [by local government], institutions with a public purpose, and other units that has been confirmed to be repayable from the operational revenues (including tuition fees), business revenues (including road tolls), or other non-[government] budgetary funds of the debtor unit, with respect to which the local government (including government departments and agencies) has provided a direct or indirect guarantee. The second type is borrowing undertaken by a local government (including government departments and agencies) that is to be repaid from non-budgetary funds; such debt is to be deemed debt guaranteed by government.90 Type III debt is described as “other relevant debt,” and is defined more precisely as follows: Borrowing undertaken for projects with a public purpose by government financing vehicles, institutions financially supported [by local government], and institutions with a public purpose that is to be repaid from non-budgetary funds, where the local government (including government departments and agencies) has not provided a guarantee (not including debt owed to other units or individuals). The government does not as a legal matter bear any liability for this type of debt, but when the debtor faces a debt crisis, it may be necessary for the government to shoulder responsibility for providing relief.91 The problem with these categories is that, as we have noted above, legal borrowing by governments and government departments is difficult—the PBOC Lending Principles seem to forbid it—and government guarantees are unquestionably invalid as a legal matter. What type of debt, then, is classed as Type I and Type II debt, and how is it different from Type III debt? The standard answer regarding Type I and Type II debt is that the NAO does not use strict legal categories for classifying debt, and includes in Types I and II obligations for which the local government is not liable in a strict legal sense, but can nevertheless be meaningfully said to be liable for a variety of reasons such as informal promises, budgetary commitments from the local government, or political pressures. The problem with this answer is that debt for which local government is not liable in a strict legal sense seems to be precisely the kind of debt that the State Council wants categorized as Type III debt. Moreover, it makes any figures for such debt useless for comparative purposes, since statistics for local government debt in other countries may or may not include amounts for which local government is not legally liable.92 One answer might lie in the distinction between repayment from budgetary funds and repayment from non-budgetary funds. In some circumstances, local People’s Congresses may provide that repayment of debt shall come from budgetary funds, and such debt is specifically excluded from Type III. Any budgetary commitment made by a People’s Congress may, of course, be later unmade by the same or a different People’s Congress, so such a provision is not legally binding. Nevertheless, perhaps it is considered sufficient to deem the debt Type I debt. The NAO reports offer further paradoxes in their breakdowns of Type I and Type II debt. Take Type I debt—debt for which the local government is said to be directly responsible—in the 2013 NAO report.93 Fully 28% (3.1 trillion yuan) of it is classified as LGFV debt.94 This makes no sense. If local government is directly (i.e., primarily) responsible for payment, then it, not an LGFV, is the debtor. If the LGFV is the primary debtor, then the debt should be classified as Type II or Type III debt. Nevertheless, the State Council’s instructions contemplate just such a subcategory, but it is difficult to know what exactly it meant. Moreover, slightly under 6% (614.6 billion yuan) of local government Type I debt is said to come from bond issues.95 In its 2011 report, the NAO put the figure at 551 billion yuan. As we have seen, local governments at that time did not have authority to issue bonds absent specific authorization from the State Council. And the numbers cannot be attributed to bonds issued by other entities that the local government is deemed to have guaranteed, whether lawfully or unlawfully, since such an obligation should clearly by Type II or Type III debt. The listed debt might be attributable to legally issued bonds, but one wonders. The first publicized instances of State Council permission for local governments to issue bonds were in late 2011.96 Such bonds could not have been included in the 2011 NAO Report, however, since the investigation on which it was based concluded in May 2011.97 As for Type II debt—debt for which local government is said to be contingently responsible through a guarantee—33% (883 billion yuan) is classified as LGFV debt.98 But as we have seen, local government cannot legally be even indirectly responsible for LGFV debt. A small amount (49 billion yuan) is even attributed to local government guarantees of local government bonds. It is difficult to know what this means: even if the bonds were somehow lawfully issued, debtors do not guarantee their own obligations. In any case, as noted below, in our review of bond prospectuses we did not find a single one in which a government purported to offer a legally binding guarantee. In short, despite the detail in the State Council’s instructions to the NAO, we cannot feel confident that we understand the criteria the NAO is using to classify local government debt. We can probably rule out certain criteria, however, including the criterion of legal obligations. Thus, it is important to recognize that when government officials and commentators speak of local government debt obligations in China, they are not speaking of legal obligations. Consequently, comparisons with legal obligations of local governments in other times and places are dangerous. II. What Stands Behind LGFV Bonds? Safeguards for Repayment A. Introduction LGFV bonds are risky. Although individual bond issues vary in their risk, it seems very unlikely that they can as a whole be repaid. As of the end of 2010, the total direct and contingent (via guarantees) obligations of SNGs were put by the NAO at 70%of their annual revenue sources.99 And at the same time, seventy-eight city-level and ninety-nine county-level governments had direct obligations of more than 100% of their annual revenues.100 In 2010, 358 LGFVs borrowed new debt to cover their old debt, and 148 LGFVs had overdue debt, totaling approximately 8 billion yuan.101 LGFV-funded projects typically do not generate large revenue streams; a survey conducted by the People’s Bank of China revealed that the average profitability of the assets owned by LGFVs is less than 1.3%.102 In some cases this is normal and intended: a highly efficient and worthwhile public works project may have positive externalities that cannot be charged for, and thus may not generate a cash return adequate to repay the investment. In other cases, the project itself may be misconceived: local leaders have strong incentives to show local GDP growth, banks (especially since the 2008 stimulus program began) have strong incentives to lend, and the result can be white elephant projects that contribute to GDP in the same way that digging a hole and filling it up again does: toll roads without cars, residential blocks without inhabitants, and airports without airplanes.103 In short, only a very small fraction of LGFV debt can be repaid in full by cash generated by their investment projects.104 These repayment concerns make necessary a closer look at what stands behind LGFV debt and, for the purposes of this Article, LGFV bonds in particular. In some cases, the bond issues offer what purports to be a formal security interest in assets or an income stream. Other issues purport to be secured by a guarantee. While the economic value of the assets is an important issue worth considering, the focus of this Article is on the legal status of the various mechanisms offered as a safeguard for repayment. Since whether such safeguards amount to a security interest in the legal sense is precisely one of the important issues to be determined, and because not all safeguards indeed amount to security interests but are not necessarily worthless for that reason, we shall generally use the term “repayment safeguard” or “backup” instead of “security interest.” This usage is in fact reflected in the offer prospectuses themselves, which almost always use the term “repayment safeguard measures” (偿债保证措施 or 偿还保障措施) to head a section setting forth the various reasons—economic, legal, formal and informal, binding and nonbinding—why investors should feel reassured about repayment. B. Safeguards As noted in the Introduction, for this Article we reviewed LGFV bond offer prospectuses from different regions in China in two different time periods and examined the proffered safeguards. Some bond issues explicitly stated that there was no security for the bonds, although they did offer other reasons for buyers to have confidence that they would be paid. Other issues offered security that would be recognized under the Guarantee Law: mortgaged land use rights, a pledge of future receivables, and third-party guarantees. Falling somewhere between nothing and something were assurances of the local government’s “strong support” (大力支持), including at times a description of the local government plans to allocate a certain amount of funds periodically to the LGFV in order to help it repay the bondholders. Several bond issues stated that funds for scheduled payments of principal and interest would be placed at some point into a special bank account that could only be used for such payments. As we shall discuss, the legal robustness of this structure is questionable. Absent a formal security interest in the account granted to bondholders, it is hard to see how other creditors could be prevented from getting their hands on such funds. Interestingly, nineteen of the twenty Group II prospectuses explicitly stated that there was no security for repayment, a marked change from the Group I prospectuses. In the following subsection, we provide a detailed examination of the various safeguard methods. We discuss why they do or do not qualify as a valid security interest, in what respects they are inadequate even if they do qualify, and how their deficiencies might be cured. 1. No Security A number of bond issues explicitly state that there is no security for repayment.105 In most cases, however, the offer prospectuses attempt to provide reasons for bondholders to have confidence that they will be repaid. In some cases, the assurance amounts to nothing more than a promise that the business will be run efficiently and show a profit. In other cases of both secured and unsecured debt, the offer recites that the LGFV has the “strong support” (大力支持) of the local government.106 This support can take the form of special business opportunities and even commitments to make good through appropriations any shortfall in LGFV repayments.107 Some offer prospectuses also note the existence of special bank accounts into which the local government will periodically make payments, and state that the funds from those accounts can be used only to pay back the debt. None of these safeguards amounts to a security interest in the sense of an asset of reasonably reliable value to which bondholders have top-priority access. First, the “special accounts” described in so many prospectuses do not seem to have any legal status under China’s Guarantee Law and thus do not offer any special protection for creditors. At any given moment, funds intended for the account are owned either by the local government or by the LGFV. (It is not clear who has formal ownership over the account itself, but it must be one or the other.) If the funds belong to the local government, they are off limits to LGFV creditors. Once transferred to the LGFV, they are available to all creditors, not just bondholders. The only way bondholders can protect these funds from other creditors is to have a proper security interest in them, but the disclosures say nothing about such a security interest having been established in the accounts. Without such a security interest, the prior claim of the bondholders to funds in the account is, at most, a contractual one between the LGFV and the bondholders—the LGFV could not, for example, use the funds for some other purpose—but that contractual claim cannot trump the interests of other creditors who did not agree to it. Second, the various assurances of support that come from local governments and People’s Congresses do not bind either them or their successors. If a People’s Congress passes a resolution promising to support a certain project, a successor can pass another resolution undoing the first. All that such resolutions can do as a legal matter is forestall an argument that a local government was somehow acting ultra vires in establishing a LGFV and having it issue bonds; they do not operate to bind the local government (or the local People’s Congress) to repay the bonds.108 Many of these assurances fall under the general rubric of informal guarantees by local governments of LGFV debt. Such guarantees must remain informal, however, because formal guarantees of indebtedness by local governments are prohibited, invalid, and unenforceable. The Guarantee Law provides “government agencies cannot act as guarantors, except in the case of sub-lending loans from foreign governments or international economic organizations as approved by the State Council of the People’s Republic of China.”109 Obviously, government guarantees for LGFV bonds do not fall within this limited exception. Indeed, there is a solid precedent on precisely this point: the 1999 bankruptcy of the Guangdong International Trust and Investment Corporation. In that case, all the “loan registrations, state and provincial guarantees, and ‘comfort letters’ with which foreign loans are trussed up count[ed] for little.”110 Investors have no realistic prospects of winning a case based on a government guarantee in case of default,111 and, tellingly, auditors are told to treat them as invalid.112 Various government agencies have periodically restated this rule just to make sure everyone gets the point. In June 2010, the State Council issued a notice emphasizing that local governments had to follow the rules of the Guarantee Law and must not issue guarantees.113 The following month, the Ministry of Finance, together with several other government departments, issued a notice further clarifying the State Council’s notice.114 This notice repeated that all guarantees were unlawful, although in a backhanded way seemed to acknowledge the binding nature of previous guarantees by stating that as of the date of the notice, local governments would no longer be liable for the debts of LGFVs (beyond, of course, the loss of their investment). In September 2010, the Ministry of Finance issued a “Notice on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels.”115 The notice states, inter alia, that local governments must “strictly implement the provisions of [the Guarantee Law], and may not issue guarantees, promises, comfort letters, or other forms of direct or disguised guarantee . . . they may not promise to bear liability for repayment of debt undertaken for the financing of other enterprises or institutions . . . .”116 And in December 2012, the Ministry of Finance together with several other departments issued yet another notice117 repeating the exact language of the 2010 notice. Litigation since the enactment of the Guarantee Law tells the same story, although with an interesting twist on the question of remedies. Up until the end of 2014, creditors who sued governmental entities on their guarantees seem to have received a consistent result in Chinese courts: the guarantee was held invalid and unenforceable, but the defendant guarantor was held liable for up to half (never more) of the unpaid debt on a theory that it shared fault with the creditor for the invalidity of the guarantee contract, and relevant law provided for liability in such cases.118 In May 2014, the Guangdong Province Higher-Level People’s Court held a press conference specifically to announce that in recent years they had processed a large number of lawsuits seeking to enforce guarantees against local governments and that such guarantees were invalid. It then presented what it called a typical case, in which the local government was held liable for one third of the unpaid debt on the theory of fault discussed above.119 One problem with this way of handling cases of government guarantees is that it seems to go directly against the intent behind the prohibition on government guarantees: to prevent local governments from becoming liable for the debts of third parties. Making local governments liable for half of the unpaid debt on a theory of fault does an end run around the prohibition, and means that for all intents and purposes local governments in fact do have the capacity to guarantee up to half the debt of third parties—even more if a court finds that the creditor was without fault. Although the courts’ unwillingness to let local governments simply disavow their promises without any consequences is understandable, surely this is not the result the drafters of the Guarantee Law intended or expected by their prohibition. A recent Supreme People’s Court (SPC) case from late 2014, however, has changed the picture in favor of local governments. In Bank of China v. Liaoning Provincial People’s Government,120 the creditor sued the Liaoning provincial government to enforce a guarantee the defendant had purportedly made, and the SPC found the government liable for no part of the debt. The case bears a very strong factual resemblance to a 2002 Guangdong province case in which the Foshan city government was found partially liable on a guarantee.121 In that case, the Foshan government had issued a document apparently called a “promise letter” (chengnuo han). The letter promised that if the debtor was late in paying the debt, the government would “take responsibility for resolving the situation” and “not cause [the creditor] to suffer economic loss” (bushi . . . zai jingji shang mengshou sunshi). The word “guarantee” was apparently never used. Despite the defendant’s argument that this promise letter created only a responsibility to try to get the debtor to repay and was not a formal guarantee, the court held that it clearly indicated an intention to issue a guarantee. Of course, the court also found that the purported guarantee was invalid, but finding that this language constituted an attempt to make a guarantee meant that the court could find the Foshan government liable for half the unpaid debt on a theory of fault. In the Bank of China case, however, the SPC looked at virtually identical language and found the defendant entirely free of liability. As in the Foshan case, the provincial government had issued a document entitled “Promise Letter” (承诺函), stating inter alia that the provincial government promised that if the borrower could not repay any amount owing under the terms of the loan, the government would “assist in resolving the debt owed by the borrower” to the bank and “would not let the bank suffer any economic loss” (不让贵行在经济上猛兽任何损失), language which is if anything slightly stronger than that of the Foshan case. Nevertheless, the SPC found that the promise did not constitute an attempt to make a formal guarantee, and constituted instead merely a kind of comfort letter that did not purport to be legally binding. Unlike in previous cases, the defendant provincial government was found not liable at all for any of the debt it had (at least arguably) purported to guarantee. Given that the Bank of China case is an SPC case and thus presumably influential, it suggests the following set of rules for future cases. First, if the word “guarantee” is not used, governments will not be held liable for any amount of debt that they purport to guarantee. Second, if the word “guarantee” is used, governments may be found to have attempted to make a guarantee, and if so, will be held liable for some portion of the debt depending on the court’s judgment as to the relative fault of the creditor and the guarantor for the invalidity of the guarantee. Courts are likely to find the guarantor at fault and liable for half, but not more, of the unpaid amount of the debt. In sum, guarantees by governmental entities are prohibited and invalid; courts will not recognize them and will not even attempt to enforce them. Still, some puzzles remain. First, if the guarantees are simply invalid, it is hard to understand why the government needs to keep telling local governments not to issue them.122 An invalid guarantee is not a problem for the purported guarantor; it’s a problem for the creditor. The government should be issuing warnings to creditors not to have faith in such guarantees. Second, both the State Council and the NAO seem to believe that unlawful government guarantees are nevertheless valid, since in the 2011 and 2013 NAO reports on local government debt, a category exists (i.e., Type II) for hard obligations represented by local government guarantees. Thus, we face essentially the same puzzle here that we examined in the context of direct debt: if the central government does not want local government guarantees to exist, why not take the route of consistently refusing to recognize their validity to any degree? Creditors would rapidly learn to ignore local government promises (or at least to consider them ephemeral and nonbinding). Legal systems that do not wish to see minors bound by contracts do not in general accomplish this purpose by threatening to punish minors who enter into contracts. Instead, they simply refuse to enforce such contracts, with the result that minors who wish to enter into contracts will find no willing counterparty. Even if local governments did purport to guarantee debt and the guarantee was in some formal sense legally valid, enforcement would be difficult. First, there are a host of practical problems to suing a government body in China, especially in a local court whose personnel appointments and finances are in the hands of the defendant.123 Enterprises and individuals are considered to be inferior, not equal, to the government. Therefore, the stakes in litigation against a government body as guarantor will typically go far beyond the issues in a particular case. In such a legal and social climate, many investors are unlikely to feel comfortable about bringing suit against a government guarantor. Second, even if a creditor did manage to bring and win a lawsuit, enforcing the judgment would remain another hurdle.124 Although the literature is replete with warnings about local governments giving formal (and therefore improper) guarantees, and the above-cited regulations seem to confirm that the phenomenon is widespread, to our surprise we did not find any such guarantees in the offer prospectuses we reviewed. In other words, no local governments purported to guarantee LGFV bond issues in any legally enforceable way. Can investors take comfort in the informal promises of support? To an extent, that depends on how much importance a local government places on its reputation for paying its debts. A problem with reputational sanctions lies in the fact that government officials at any given time have a high incentive to borrow in order to construct projects that will raise local GDP, one of the important indices on which their performance is graded.125 By the time the debt becomes due, that official will long ago have left office (possibly having been promoted) according to the pattern of five-year terms for local officeholders. The successor official may well be unwilling to honor a nonbinding pledge made by and for the benefit of his predecessor.126 Despite the dubious security offered by government assurances, the number of bond issues safeguarded by such assurances is the same as the combined number of issues safeguarded by the other three methods discussed in this Article. 2. Land Use Rights as Security Several bond issues put up land use rights (LURs) belonging to the issuer as collateral.127 The legal basis for this form of security is article 34 of the Guarantee Law of China, which provides that land use rights that can be legally transferred by the debtor can be used as collateral.128 “Land use rights” refers to what are in effect long-term (a maximum term of forty, fifty, or seventy years, depending on the use) transferable leaseholds in urban land. Although what amounts to permanent fee simple ownership, and hence the reversion,129 must by law remain with the state, the law allows such land use rights to be granted for a one-time fee. The LURs then enter the stream of commerce and can be sold, mortgaged, leased, and otherwise transferred. Land subject to this regime is called “granted land.” Urban land from which LURs have not yet been carved out and sold is called “allocated land”: someone is using it, but the user received it through a process of bureaucratic allocation, not via a market transaction. Allocated LURs are not transferable and cannot be pledged as security—or more precisely, allocated LURs that are pledged as security will have to be turned into granted LURs, with the accompanying payment, before they can be taken in satisfaction of the debt by the secured party.130 In other words, the secured party will end up paying twice: once when it makes the loan, and once when it takes the security. In all the prospectuses we examined where LURs were offered as security, the LURs were in all but one case said to be obtained by grant and are hence unproblematic in that sense.131 Issuers typically also disclosed that they had already obtained the land use right certificate from the relevant agency and that the land was free from other encumbrances. Offering prospectuses frequently included a detailed schedule of the issuer’s land use rights used as security, noting the method by which the LURs were obtained and the date. How good is this security? Although this study does not in the main address the economic value of the security, there are some aspects of the valuation process that are cause for concern and can lead to excessive or questionable valuations.132 Although valuations must be made by credentialed assessors, countless assessment agencies exist in China, with a wide range of competency and levels of professionalism. LGFVs can freely choose whoever will give them the highest assessment. Not all offer prospectuses disclose how the assessor was selected or on what basis the valuation was made—for example, whether it was made with reference to market value in an arm’s length transaction. Although the value of urban land in China has risen greatly since the mid-1990s,133 there is no way of knowing where it will go in the future.134 In fact, the market entered a major slump in early 2014.135 However, LGFV issuers still have a tendency to overvalue the land use rights they have.136 Ideally, problems of overvaluation should be solved in the bond rating process. China has a credit rating industry,137 and bond issues are rated. There is reason to believe, however, that Chinese bond rating agencies may be afflicted by the same problems that afflicted U.S. bond rating agencies in the run-up to the financial crisis. For example, one prominent firm, Dagong, downgraded its rating of U.S. government debt to a single A with a negative outlook in an apparent fit of pique after being denied status as a Nationally Recognized Statistical Rating Organization by the Securities and Exchange Commission (SEC) in 2010.138 At the same time, it gave a higher rating to the bonds issued by the Loudi LGFV even though the LGFV valued the real estate collateralizing the debt at prices similar to those of Winnetka, Illinois, where local household income is about 100 times that of Loudi.139 The significance of valuation problems is compounded by a common feature in the collateralization agreement: a one-way ratchet. In a number of issues, the issuer has the obligation to add more LURs as collateral if the value of current collateral falls below some multiple of the outstanding debt obligation, and the right to demand the release of some collateral if the value of the collateral is above some multiple of the outstanding debt.140 This structure could operate as a one-way downward ratchet on the value of collateral on the basis of two plausible assumptions: first, that land values, even if accurately assessed, can fluctuate, and second, that the issuer simply may not possess sufficient unencumbered LURs to provide more collateral when the obligation to do so arises. As a result, if land values go up, creditors will lose some collateral, and it may be impossible to recover that collateral when land values go back down again. It is sometimes thought that despite valuation problems, LGFV bonds secured by land use rights at least have something backing them up.141 There may, however, be legal infirmities in the collateral itself, in addition to problems of economic value. First, there may be other encumbrances on the LURs used as collateral, and the claims of bondholders may be inferior.142 The offering prospectuses typically recite that the LURs are unencumbered, but bondholders have few if any remedies if this assertion turns out not to be true. The LGFV could be sued for false disclosure, but the disclosure problem will typically be discovered only at the point when the LGFV is already insolvent. Even if individual corporate officers can be successfully sued, their personal wealth will be inadequate to make up actual losses. Second, the LUR may not have been properly granted.143 In some cases, LURs purport to be granted by the governing entity of a development zone, called a Development Zone Commission. But such commissions have no legal power to grant LURs; that belongs to the local municipal government.144 In other cases, the land may still be in the intermediate stage prior to being formally granted, when existing residents are being cleared off and various payments must be made.145 It should be noted that these are generic problems with the use of LURs as collateral; we did not find specific evidence of these problems in any of the offering prospectuses we reviewed. A more serious problem with the grant lies in the possibility that no grant fee was ever paid. Consider the mathematics, taking the Loudi Prospectus as an example. The bond issue proposed to raise 1.2 billion yuan. The prospectus, dated January 2011, states that the value of the LURs securing the bonds is approximately 2.1 billion yuan. Most of the LURs were granted just one year earlier, in 2010. Even if the LGFV got the LURs at a 50%discount to the market value of just a year later—a generous assumption—it should have paid just over one billion yuan for them. This leads to an obvious question: if it had enough cash to pay for LURs, why did it need to raise cash through a bond issue? The obvious answer is that it did not have the cash and probably paid for the LURs with an IOU. But Chinese law requires full payment for LURs before LURs can be transferred, and a transfer is involved when a creditor realizes on collateral.146 Even where the local government is the sole shareholder of the LGFV, whether it can inject LURs as a capital contribution (taking payment in stock of the LGFV) is not clear. In short, there are strong reasons to doubt that the LURs securing LGFV bonds have been paid for in full. Without full payment, they cannot be transferred and are useless to creditors. And when the LGFV is insolvent—precisely the time when creditors need the collateral—it is of course in no position to pay for the LURs. Similarly, article 39 of the Urban Real Estate Administration Law requires that LUR grantees take certain steps to develop the land before it can be transferred. One rule is that the development be undertaken in accordance with the terms of the contract. Since the local government, the sole shareholder of the LGFV, is on the other side of the contract, one could reasonably expect the terms not to be onerous. But there is another rule not so easily avoided: the requirement that 25%of planned investment in housing have been made before the LUR can be transferred. As many of the LURs are stated to be for housing purposes, this requirement cannot be avoided. Failure to make planned investment can even make the original grant contract void.147 In an extreme case, validly granted LURs can be voided entirely when the fee is not paid according to the terms of the contract.148 A final consideration is that even if the legal and economic status of the security is sound, realizing on it may prove difficult if it requires displacing existing residents.149 They may not be willing to go peacefully. Given that defaults are likely to occur in a time of general economic downturn and therefore with a higher potential for social unrest, it is very unlikely that government authorities would place a higher priority on protecting the legal rights of distant bondholders than on preventing demonstrations and other manifestations of social unrest. There can be little doubt that many of these infirmities exist. Several were noted with disapproval in a document jointly issued in late 2012 by several central government regulators.150 Current disclosure practices vary. In the Shaoyang and Loudi offer prospectuses, the disclosure is quite comprehensive. It includes the exact location, the size, and expiration dates of the LURs, whether there are encumbrances, and the nature of those encumbrances.151 Other issuers disclose only basic data about the location and value of the LURs, without addressing more critical concerns, such as the way the LUR was obtained and information about encumbrances.152 In any case, it is impossible to tell from the prospectuses what agreements have been made about land development and whether the LGFVs are in a position to undertake necessary development of the LURs being offered as collateral. The point here is simply that what appears to be and may in fact be a valid, top-priority LUR in an accurately assessed parcel of real estate can still bear hidden dangers that call for constant monitoring. 3. Future Receivables as Security Several bond issues purport to be secured by receivables due to the LGFV issuer under Build-Transfer (BT) Agreements between the issuer and certain local governmental agencies (usually the agencies in charge of finance, transportation, or construction) in which the issuer agrees to build a project and the agency agrees to purchase the project after it has been built.153 Interestingly, this structure seems to be legally robust, even though in substance it mimics a forbidden structure. As discussed above, if a local government wishes to finance a construction project, it may not do so by issuing bonds directly. Nor may it guarantee bonds issued by its wholly owned LGFV.154 Obviously, an LGFV may issue bonds on its own, but the trick is to have the local government give a legally binding commitment of its own to bondholders in the face of the above prohibitions. The pledge of receivables accomplishes this goal neatly. First, the issue of bonds by the LGFV is legally unproblematic, precisely because the LGFV is not a government entity. Second, the contractual commitment by the local government to purchase the project upon completion under the BT contract is also unproblematic.155 Governments and government agencies can commit themselves contractually to suppliers under ordinary principles of government procurement. Third, the pledge of the receivable, which gives bondholders priority over other creditors of the LGFV, is permitted under China’s Guarantee Law. As a result, bondholders get exactly what they would get if the local government issued the bonds directly or provided a guarantee for LGFV bonds (assuming either was legal): a legally valid government promise to pay money sufficient to repay the bonds and to which they have top-priority access. Needless to say, whether the local government will actually have the funds to make good on its promise to pay is an important issue, but the focus of this Article is on the legal status of the various safeguards for repayment available to bondholders. 4. Third-Party Business Entity as Guarantor Another commonly used safeguard in LGFV bonds is a guarantee from a third party business entity. The third party company can be either an investment holding company or a company specializing in the guarantee business. For example, in the Baoji LGFV bond, China National Investment and Guaranty Co., Ltd. acted as the guarantor for the bonds, providing a “full, unconditional, and irrevocable joint responsibility guarantee.”156 In the Leshan LGFV bond, the two classes of bonds were guaranteed respectively by Chongqing Three Gorges Guarantee Group Co., Ltd. and Beijing Capital Investment and Guarantee Co., Ltd. Respectively, and the guarantors provided a “full, unconditional, and irrevocable joint responsibility guarantee.”157 In a bond issuance by Mega Advance Investments Limited (“Mega”), a Beijing LGFV, the guarantor company promised to “irrevocably and unconditionally guarantee the due and punctual payment of the principal of and interest on, and all other amounts payable under, the Notes when and as the same shall become due and payable.”158 Furthermore, the guarantor agreed that its obligations under the guarantees would be “as if it were principal obligor and not merely surety.”159 Under Chinese law, a guarantee provided by a third party that can be independently held liable for the debt is a form of security under the Guarantee Law.160 Specifically here, the bonds are secured by the guaranteeing entity’s operating assets that can be legally transferred or disposed of in order to pay off the debt of the LGFV under the guarantee agreement signed between LGFV and the guarantee company. Accordingly, the disclosure in the prospectus involves material information about the guarantor company instead of the LGFV issuer, with the guarantor company’s income-generating operations and assets constituting the bulk of the disclosure, as in the case of the Mega bond issuance. A common type of third-party guarantor is a specialized guarantee company; as of the end of 2011, China had 8,402 of them, with assets purporting to total 931 billion yuan.161 Guarantees from such companies can, however, be very risky. Guarantee companies were originally created with the backing of local governments to help small and medium enterprises get bank loans.162 Banks normally prefer to lend to enterprises that have hard assets such as land or else have implicit state backing. Guarantee companies charge borrowers a fee and then provide a binding guarantee to the lending bank. In effect, then, the guarantee company offers default insurance, with the premium paid by the borrower. This means that the lender has to consider a number of things about the insurer. How many policies does it have outstanding, and how do its potential obligations match up against its assets? Even if that information is fully disclosed—and often it is not—the lender still needs to know how many more policies it will issue after the loan is made. Most importantly, one can have confidence in one’s insurer only if one is confident that the insurer’s risks on its various policies are not positively correlated with each other.163 In other words, if it is insuring one party against a default that will happen if the price of oil goes up, it should be insuring another party against a default in the same amount that will happen if the price of oil goes down. You do not want a flood insurer that does business only in New Orleans. Whether Chinese guarantee companies have in fact spread their risks properly is far from clear, and in any case not evident from the disclosure provided in bond prospectuses. Thus, it is very possible that they are vulnerable to broad economic shocks that hit all or most borrowers at once, such as a downturn in the property market or a significant slowing of growth. The risks of guarantee companies were vividly brought home in the 2012 collapse of the Zhongdan Investment Credit Guarantee Company.164 In January 2012, banks reacted to rumors of a liquidity crunch at Zhongdan by calling loans. The debtor companies in turn sought help from Zhongdan, but the volume of loans overwhelmed the company. Zhongdan had guaranteed 3.3 billion yuan in loans, but had only 5 million yuan immediately liquid and accessible when the calls for funds came. In the case of Zhongdan, there appears to have been malfeasance afoot. But even without malfeasance, the lack of real substance to its guarantees is likely duplicated at many other guarantee companies. An important aspect of due diligence for bondholders is to examine carefully the assets and liabilities of the guarantor companies. The assets of guarantor companies may be misstated or subject to multiple existing encumbrances.165 When the bulk of a guarantor company’s assets are its subsidiaries, it is critical to know about the indebtedness of those subsidiaries, ranging from convertible bonds to bank loans.166 Even if the guarantor company had no encumbrances or indebtedness, its asset structure could be problematic because of a high ratio of non-liquid assets to liquid assets such as cash. For example, in the Mega LGFV bond issue, the guarantee company’s portfolio assets include natural gas distribution and transmission assets, brewery operations, sewage and water treatment facilities and concessions, and expressways and toll roads. For the guarantee company in the Baoji LGFV bond issuance, China National Investment & Guarantee Co., Ltd., the ratio of non-liquid assets to liquid assets is approximately four to five.167 Its nonliquid assets include “sellable financial assets” and fixed assets.168 It is by definition difficult to cash out nonliquid assets when the debt obligation is due, and that difficulty undermines the value of the guarantee as security. III. The Debt Swap Program In the spring of 2015, the government announced a program under which local governments could issue bonds. Under the plan, known as a “debt swap,” the Ministry of Finance would, in separate tranches of quotas, authorize local governments at the level of province and separately planned city169 to issue bonds, the revenues from which would be used to pay off local government debt. The purpose of the plan is to allow local governments to save on interest costs by replacing short-term, high-cost, high-risk local government debt (including LGFV debt that is classed as local government debt) with longer-term, lower-risk and therefore lower-cost government bond debt.170 The first tranche of the swap was one trillion yuan and was to cover obligations (probably only Type I obligations) of local governments identified in the 2013 NAO report that were due in 2015.171 According to the Ministry of Finance, the one trillion yuan quota covered 53.8 %of the 1.86 trillion yuan in debts that came due in 2015, and a proportional allocation system meant that each province received a quota representing 53.8%of its direct indebtedness.172 The second tranche, also of one trillion yuan, was announced in June 2015.173 That additional quota was also intended to cover debts coming due in 2015. Since each tranche amounts to over 50% of the debt it is to cover, together they would seem to cover more than the debt in question. In any case, the quota for 2015 was ultimately raised to 3.2 billion yuan, of which 3.18 trillion yuan in swapped bonds was used by mid-December.174 Estimates of the amount needed to absorb remaining LGFV debt range from 11 to 25 trillion yuan.175 Although the Ministry of Finance announced that revenues from the swapped bonds could not be used to pay off debts that should be paid from the self-generated revenues of enterprises and institutions, there is wiggle room here: as discussed above, since the distinctions used by the NAO to determine precisely which debts are this kind of debt are not strict legal ones, it is not going to be self-evident which local debts “should” be payable solely from self-generated revenues of enterprises and institutions. The swap program is part of a larger central government effort to clean up LGFV debt and reduce or eliminate its use in the future. It is not, however, going to make the issues surrounding LGFV debt discussed in this Article disappear. First, although we cannot know exactly the criteria used by the NAO is classifying debt associated with local governments, it seems beyond doubt that a considerable amount of LGFV debt is Type II or Type III debt, or possibly not even counted as local government debt at all, whereas the swap program seems to cover only Type I debt.176 Second, since the very category of “LGFV” is not a legal or precisely defined one, the only way to effectively ban LGFV debt is to ban all debt in corporate entities owned by local governments. But the government has not gone this far. If it is ever more convenient for local governments to borrow through establishing wholly owned companies instead of directly in their own name, we will see LGFVs or their functional equivalent. Conclusion The contribution of LGFVs to China’s economy will probably always remain controversial. On the one hand, they were a key element in the government’s post-financial-crisis stimulus plan, which relied heavily on bank lending and relatively little on new government spending.177 They allowed local governments to get capital for infrastructure construction, and enabled China’s GDP to continue to grow. On the other hand, the only reason local governments needed LGFVs to raise capital was that the central government prohibited them from borrowing themselves—a policy that could have been changed whenever necessary, and that is now being relaxed (although only down to the level of the province and separately planned cities).178 It is not clear whether the infrastructure financed through LGFVs has in fact made China better off, since there is reason to believe that at least some of it is unlikely to produce an economic return over its cost. GDP grows whenever there is more economic activity than before, but not all economic activity—for example, the classic case of digging a hole and filling it up again—represents a good investment and leads to growth. It can retard growth if it just produces bad debt that must be ultimately borne by households, thus in turn leading to lower aggregate demand. Because it seems almost mathematically certain that at least some nontrivial number of LGFV bonds will not be eligible for the swap program and will default, it is necessary to think about the quality of the security they offer. It is difficult to generalize about bond repayment safeguards because of the variety of practices we have canvassed. As we have discussed, a government guarantee is not a valid security in Chinese law, either in theory or in practice, no matter what the bond prospectuses might say. On the other hand, the land use right obtained by grant, as well as the guarantee provided by a guarantee company, constitutes a valid security interest, although its economic value requires separate analysis. This Article’s analysis of the legal structure of LGFV bonds and local government debt more generally also has important lessons for understanding the Chinese legal system as a whole. A key part of the story of local government indebtedness is the inadequacy of legal categories to understand what is going on. Legal terms of art such as “debt,” “liability,” and “guarantee” are widely used, and yet it becomes quickly apparent that they are not being used in a legal sense. This means that data classified using these terms cannot be readily compared with data classified using these terms in other countries. In particular, it is clear that accounts of the level of China’s local government debt that include Type II and Type III debt simply cannot be accepted at face value. This is not debt that local governments have any actual obligation to pay. To be sure, they may wish to pay creditors voluntarily and may from time to time do so, but soft obligations of this nature have never been considered debt in China or in other countries, and it is misleading to label them as such. Creditors who have tried to force local governments to make good on their guarantees have uniformly failed. At most they have received half of what they sought, under a legal theory that a recent Supreme People’s Court case has made more difficult to invoke.179 The argument that local governments have some politically (even if not legally) enforceable obligation to pay on their guarantees does not seem supported by the evidence. In addition, we see here what can be observed in other areas of Chinese law as well: the tendency of some players and institutions in the system to recognize as legally valid acts that at the same time the system deplores and even prohibits.180 The system could attempt to discourage unlawful borrowing and guarantee-issuing by local governments by simply making such borrowing and guarantees legally invalid and unenforceable—in effect, invisible to the legal system. And in a sense it tries to do that, but not unambiguously. Its approach is instead to prohibit the acts and make them punishable, implicitly conceding that the act could be effective. For the state to prohibit and punish murder makes sense, because the murderer does not need the state’s help to make the murder effective. But to prohibit and punish the making of guarantees is different, because without the state’s cooperation the guarantee could not be effective and could do no damage in the first place. On the one hand, the courts are invalidating local government guarantees. But at the same time, instead of warning creditors not to accept such guarantees, the central government is warning local governments not to make them, and the State Council and the NAO categorize them as valid and enforceable when they count local government debt obligations. The broader lesson here is that acts that are prohibited under Chinese law cannot be assumed to be invalid even if they require state cooperation to be valid; they may or may not be, and a detailed analysis of the specific act in question, and related rules and practices, is required. There is no question that LGFV defaults will pose a challenge for China’s political system. The pressure for bailouts will be intense, but the cost of providing them, in terms of both actual money and added moral hazard, will be huge. Yet defaults will also pose a serious challenge to China’s legal system. If defaulting LGFVs are not bailed out, then the cost of default must land somewhere other than on the entity—probably the central government, perhaps the local government—that would have funded the bailout. There are legal rules about where it ought to land. Will they be effective? A great natural experiment is in the offing. Appendix Bond Prospectuses Group I Baoji Prospectus: 2011 Nian Baoji Shi Touzi (Jituan) Youxian Gongsi Qiye Zhaiquan Muji Shuomingshu (2011年宝鸡市投资(集团)有限公司企业债券募集说明书), March 2011, available athttp://bit.ly/ztl7j3 Datong Prospectus: 2011 Nian Datong Shi Jingji Jianshe Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年大同市经济建设投资有限责任公司公司债券募集说明书), June 2011, available athttp://bit.ly/zG2OYs Hebei Bohai Prospectus: 2011 Nian Hebei Bohai Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年河北渤海投资有限公司公司债券募集说明书), May 2011, available athttp://bit.ly/zk0jQ2 Hefei Prospectus: 2009 Nian Hefei Shi Jianshe Touzi Konggu (Jituan) Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年合肥市建设投资控股(集团)有限公司公司债券募集说明书), July 2009, available athttp://bit.ly/1Mhc4p1 Jiangyin Prospectus: 2009 Nian Jiangyin Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年江阴城市建设投资有限公司公司债券募集说明书), September 2009, available athttp://bit.ly/1JSEmWJ Jilin Prospectus: 2011 Nian Jilin Shi Chengshi Jianshe Konggu Jituan Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiwu Muji Shuomingshu (2011年吉林市城市建设控股集团有限公司市政项目建设债券募集说明书), March 2011, available athttp://bit.ly/yAetfO Leshan Prospectus: 2011 Nian Leshan Shi Guoyou Zichan Jingying Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年乐山市国有资产经营有限公司公司债券募集说明书), March 2011, available athttp://bit.ly/zixZYR Lishui Prospectus: 2011 Nian Lishui Shi Chengshi Jianshe Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年丽水市城市建设投资有限责任公司公司债券募集说明书), October 2011, available athttp://bit.ly/xzxbaZ Loudi Prospectus: 2011 Nian Loudi Shi Chengshi Jianshe Touzi Jituan Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年娄底市城市建设投资集团有限公司市政项目建设债券募集说明书), January 2011, available athttp://bit.ly/wNBdER Mega Prospectus: Offering Memorandum, Mega Advance Investments Limited, Unconditionally and Irrevocably guaranteed by Beijing Enterprises Holdings Limited, March 2011 (on file with first author) Nanjing Prospectus: 2009 Nian Nanjing Shi Chengshi Jianshe Touzi Konggu (Jituan) Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年南京市城市建设投资控股(集团)有限责任公司公司债券募集说明书), available athttp://bit.ly/1KYo9xL Panjin Prospectus: 2011 Nian Panjin Jianshe Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Faxing Gonggao (2011年盘锦建设投资有限责任公司公司债券发行公告), February 2011, available athttp://bit.ly/zyA1P6 Shaoyang Prospectus: 2011 Nian Shaoyang Shi Chengshi Jianshe Touzi Jingying Jituan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2010年邵阳市城市建设投资经营集团有限公司公司债券募集说明书), August 2010, available athttp://bit.ly/y33PrX Tianjin Binhai Prospectus: 2011 Nian Tianjin Binhai Xin Qqu Jianshe Touzi Jituan Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年天津滨海新区建设投资集团有限公司市政项目建设债券募集说明书), November 2011, available athttp://bit.ly/1WecbX1 Xianyang Prospectus: 2009 Nian Xianyang Shi Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年咸阳市城市建设投资有限公司公司债券募集说明书), September 2009, available athttp://bit.ly/1KcQb5H Yichang Prospectus: 2011 Nian Yichang Shi Chengshi Jianshe Touzi Kaifa Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年宜昌市城市建设投资开发有限公司市政项目建设债券募集说明书), November 2011, available athttp://bit.ly/AlIUxA Yichun Prospectus: 2009 Nian Yichun Shi Chengshi Jianshe Touzi Kaifa Youxian Zeren Gongsi Qiye Zhaiquan Muji Shuomingshu (2009年伊春市城市建设投资开发有限责任公司企业债券募集说明书), September 2009, available athttp://bit.ly/1f4JkT9 Yongzhou Prospectus: 2011 Nian Yongzhou Shi Chengshi Jianshe Touzi Kaifa Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年永州市城市建设投资开发公司市政项目建设债券募集说明书), December 2011, available athttp://bit.ly/womppr Zhenjiang Prospectus: 2010 Nian Zhenjiang Shi Jiaotong Touzi Jianshe Fazhan Gongsi Qiye Zhaiquan Muji Shuomingshu (2010年镇江市交通投资建设发展公司企业债募集说明书), December 2010, available athttp://bit.ly/zc42Eo Zhuzhou Prospectus: 2009 Nian Zhuzhou Shi Chengshi Jianshe Touzi Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年株洲市城市建设投资经营有限公司公司债券募集说明书), available athttp://bit.ly/1L3di8B Group II Baoshan Prospectus: 2014 Nian Baoshan Shi Guoyou Zichan Jingying Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年保山市国有资产经营有限责任公司公司债券募集说明书), May 2014, available athttp://bit.ly/1HD2ZjO Beichen Tianjin Prospectus: 2014 Nian Tianjin Shi Beichen Qu Jianshe Kaifa Gongsi Qiye Zhaiquan Muji Shuomingshu (2014年天津市北辰区建设开发公司企业债券募集说明书), April 2014, available athttp://bit.ly/1L1wdh7 Changzhou Prospectus: 2014 Nian Changzhou Gonggong Zhufang Jianshe Touzi Kaifa Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年常州公共住房建设投资发展有限公司公司债券募集说明书), available athttp://bit.ly/1gZ1IyQ Chengfa Prospectus: 2014 Nian Chengfa Touzi Jituan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年城发投资集团有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1KfcAj1 Dongtai Prospectus: 2014 Nian Di’erqi Dongtai Shi Chengshi Jianshe Touzi Fazhan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年第二期东台市城市建设投资发展有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1HD4Jd6 Ezhou Prospectus: 2014 Nian Diyiqi Ezhou Shi Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年第一期鄂州市城市建设投资有限公司公司债券募集说明书), May 2014, available athttp://bit.ly/1J82jun Hami Prospectus: 2014 Nian Xinjiang Weiwu’er Zizhiqu Hami Diqu Guoyou Zichan Touzi Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014 年新疆维吾尔自治区哈密地区国有资产投资经营有限公司公司债券募集说明书), July 2014, available at http://bit.ly/1PhOeJC Kunming Prospectus: 2014 Nian Kunming Dianchi Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年昆明滇池投资有限责任公司公司债券募集说明书), June 2014, available athttp://bit.ly/1IVbasI Kunshan Prospectus: 2014 Nian Kunshan Jiaotong Fazhan Konggu Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年昆山交通发展控股有限公司公司债券募集说明书), May 2014, available at http://bit.ly/1gv6qDw Laizhou Prospectus: 2014 Nian Laizhou Shi Guoyou Zichan Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年莱州市国有资产经营有限公司公司债券募集说明书), July 2014, available athttp://bit.ly/1L1zwFf Lin’an Prospectus: 2014 Nian Lin’an Shi Chengjian Fazhan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年临安市城建发展有限公司公司债券募集说明书), August 2014, available athttp://bit.ly/1P2AHVx Lu’an Prospectus: 2014 Nian Di’erqi Lu’an Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2013年第二期六安城市建设投资有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1PhPLiM Mudanjiang Prospectus: 2014 Nian Mudanjiang Shi Guoyou Zichan Touzi Konggu Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年牡丹江市国有资产投资控股有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1J0OnjT Ningxiang Prospectus: 2014 Nian Ningxiang Jingji Jishu Kaifaqu Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年宁乡经济技术开发区建设投资有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1DJlRmo Pizhou Prospectus: 2014 Nian Pizhou Shi Runcheng Zichan Jingying Jituan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年邳州市润城资产经营集团有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1DEMg56 Qidong Prospectus: 2014 Nian Di’erqi Qidong Shi Chengshi Touzi Jingying Zhongxin Qiye Zhaiquan Muji Shuomingshu (2014年第二期启东市城市投资经营中心企业债券募集说明书), April 2014, available athttp://bit.ly/1WeoW3S Shenmu Prospectus: 2014 Nian Shenmu Xian Guoyou Zichan Yunying Gongsi Qiye Zhaiquan Muji Shuomingshu (2014年神木县国有资产运营公司企业债券募集说明书), June 2014, available athttp://bit.ly/1PhRmp0 Shenyang Prospectus: 2014 Nian Shenyang Nongye Gaoxinqu Guoyou Zichan Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年沈阳农业高新区国有资产经营有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1DENjlw Taiyuan Prospectus: 2014 Nian Taiyuan Jingji Jishu Kaifa Zonggongsi Qiye Zhaiquan Muji Shuomingshu (2014年太原经济技术开发总公司企业债券募集说明书), April 2014, available athttp://bit.ly/1L1CzNG Xinjiang Prospectus: 2014 Nian Xinjiang Shengchan Jianshe Bingtuan Nong Shi’er Shi Guoyou Zichan Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年新疆生产建设兵团农十二师国有资产经营有限公司公司债券募集说明书), May 2014, available athttp://bit.ly/1DEOgu0 The views expressed in this Article are solely those of the authors and do not reflect the views or position of Memjet Technology, Ltd. The authors wish to thank Michael Forsythe and Henry Sanderson for the Bloomberg article that inspired this Article and for helpful advice on finding the kind of prospectuses we were looking for. We also wish to thank Andrew Collier, Nicholas Howson, Curtis Milhaupt, and Barry Naughton for their thoughtful comments. All English translations in this Article are by Donald Clarke unless otherwise indicated. Footnotes 1 In this Article, “local government” refers to a government one or more administrative levels below the central government—for example, a city, county, or provincial government. The term is used interchangeably with “subnational government” or “SNG.” 2 On local government debt in China generally, see Xun Wu, Policy Brief, China’s Growing Local Government Debt Levels, Mass. Inst. of Tech.(MIT) Ctr. for Fin. & Pol’y (Jan. 2016), http://bit.ly/xunwu. 3 See, e.g., China Cities Value Land at Winnetka Prices with Bonds Seen Toxic, Bloomberg News (July 13, 2011), http://bloom.bg/q0AV7P. 4 The prospectuses are listed in the Appendix. For ease of reference, the prospectus for any particular issue shall be referred to in this Article using the abbreviation provided in the Appendix. 5 We use this term deliberately in order to avoid the more narrow and technical connotations of “collateral.” Whether the proper steps to achieve collateralization have actually been taken is sometimes not clear. 6 See Du Tao (杜涛), Caizheng Bu Pifu 3 Wan Yi Edu Zhihuan Cunliang Zhaiwu 1 Wan Yi Yi Xiapi Gesheng (财政部批复3万亿额度置换存量债务 1万亿已下批各省) [Ministry of Finance Approves 3 Trillion Quota in Bonds-for-Debt Swap; 1 Trillion Already Allocated to Provinces], Jingji Guancha Wang (经济观察网) [Econ. Observer Network] (Mar. 6, 2015), http://www.eeo.com.cn/2015/0306/273066.shtml; Defusing a Bomb, Economist (Mar. 11, 2015), http://econ.st/1HIAuDX. For a more detailed discussion, see infra Part III. 7 See infra Part I.A.2.b.ii. 8 We will frequently provide statistics for the late 2000s because this is when LGFV activity took off. 9 See Wei Cui, Fiscal Federalism in Chinese Taxation, 3 World Tax J. 455, 456 (2011). The figure for 2009 is 20%. See Christine P. Wong, Paying for Urbanization in China: Challenges of Municipal Finance in the 21st Century, inFinancing Metropolitan Governments in Developing Countries 273, 281 (Roy W. Bahl et al. eds., 2012). 10 See Cui, supra note 9, at 456. 11 See id. at 457. Similar numbers are cited in the more recent Houze Song, Why China Can Avoid a Local Fiscal Crisis, Paulson Inst. (May 4, 2015), http://bit.ly/1Q24q28. Looked at another way, local government revenue is 20% of GDP, while local government expenditures are 28% of GDP. See China’s Regional Financial Crisis, The Banker, Feb. 1, 2016. 12 See Wong, supra note 9, at 281–90; Cui, supra note 9, at 456–57. 13 Zhonghua Renmin Gongheguo Shuishou Zhengshou Guanli Fa (中华人民共和国税收征收管理法) [Law of the People’s Republic of China Concerning the Administration of Tax Collection] (promulgated by the Standing Comm. Nat’l People’s Cong., Sept. 4, 1992, effective Jan. 1, 1993), https://perma.cc/FV2J-5PTS. 14 Guowuyuan Guanyu Shixing Fenshuizhi Caizheng Guanli Tizhi de Jueding (国务院关于实行分税制财政管理体制的决定) [Decision of the State Council Regarding the Implementation of the Tax Sharing Fiscal Administration System] (promulgated by the St. Council, Dec. 15, 1993, effective Dec. 15, 1993), https://perma.cc/8WZD-XLBD. 15 See generally Cui, supra note 9, at 463–65. 16 Id. at 457. 17 People’s Congresses are legislative bodies that exist at various levels of government. They appoint the leadership of, and are formally higher in rank than, the so-called “People’s Governments” at the same administrative level, which can be roughly analogized to an executive branch of government. 18 See Cui, supra note 9, at 469 & n.85; Lou Jiwei (楼继伟), Guowuyuan Guanyu 2014 Nian Zhongyang Juesuan de Baogao (国务院关于2014年中央决算的报告) [Report of the State Council on the Implementation of Budget and Final Accounts of Central and Local Governments], Zhonghua Renmin Gongheguo Caizheng Bu (中华人民共和国财政部) [Ministry of Fin. of the People’s Republic of China] (June 28, 2015), https://perma.cc/YJ5C-5BS6. 19 A legal system does this kind of thing when it enforces against a principal a contract entered into by an agent that exceeds the scope of the agent’s authority. The contract is valid as between the principal and the third party, but the principal can sue the agent for any losses. 20 See, e.g., Ye Peirong (叶裴荣), Woguo Difang Zhengfu Rongzi Tixi Yanjiu (我国地方政府融资体系研究) [A Study of The Local Government Financing System in China], 1 Xinan Jinrong (西南金融) [Sw. Fin.] 39 (2012) (citing Daikuan Tongze (贷款通则) [Lending Principles] (promulgated by the People’s Bank of China [PBOC], June 28, 1996, effective June 28, 1996), http://www.law-lib.com/law/law_view.asp?id=15263); China Said To Warn Banks as Local Officials Seek New Loans, Bloomberg News (July 13, 2011), http://bloom.bg/1VppWAI; Michael F. Martin, Cong. Research Serv., R42380, China’s Banking System: Issues for Congress (2012), http://www.fas.org/sgp/crs/row/R42380.pdf. 21 See Lending Principles, June 28, 1996, art. 17. 22 See, e.g., Ye, supra note 20, at 39. The author is described as an employee at the Chengdu branch of the China Development Bank, a wholly state-owned “policy bank,” and presumably speaks with some knowledge of actual practice within the banking industry. The same reasoning is cited by Liang Weihong (梁卫洪), Difang Ren Da Huandai Chengnuo de Falü Wenti Yanjiu (地方人大还贷承诺的法律问题研究) [Research on the Legal Issues in Local People Congress’s Promise to Repay Loans] (Apr. 20, 2005) (unpublished LL.M. thesis, Jinan (暨南) University) (on file with first author), as the reason for setting up LGFVs as loan recipients. 23 This was true under article 28 of the old Budget Law and remains true under article 35 of the recently revised Budget Law of 2015. See Yusuan Fa (预算法) [Budget Law] (promulgated by the Nat’l People’s Cong., Mar. 22, 1994, effective Jan. 1, 1995), https://perma.cc/VR3D-UD3F; Yusuan Fa (预算法) [Budget Law] (promulgated by the Nat’l People’s Cong., Mar. 22, 1994, amended Aug. 31, 2014, effective Jan. 1, 2015), https://perma.cc/6QT9-GEH5. 24 In reading the relevant literature, it is critical to distinguish references to direct SNG borrowing and SNG borrowing through LGFVs. The latter is often carelessly referred to as SNG debt. 25 SeeZhonghua Renmin Gongheguo Shenji Shu (中华人民共和国审计署) [National Audit Office of the People’s Republic of China], Quanguo Difang Zhengfuxing Zhaiwu Shenji Jieguo (全国地方政府性债务审计结果) [Results of a Nationwide Audit of Government Debt] tbl.3 (2011), https://perma.cc/9HFP-RZ96 [hereinafter NAO 2011]. 26 See id. sec. 3.1. 27 See, e.g., Caizheng Bu Guanyu Guifan Difang Geji Zhengfu Bumen Juzhai he Danbao Chengnuo Xingwei de Tongzhi (财政部关于规范地方各级政府部门举债和担保承诺行为的通知) [Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels] (promulgated by the Ministry of Fin., Sept. 2, 2010, effective Sept. 2, 2010), https://perma.cc/85RB-PMWM. This document cannot be interpreted as a warning to creditors that their rights will not be upheld, since it is labeled “not for public circulation” and was distributed only to potential debtors (local governments) and not to potential creditors (the banking system). 28 Note that we are not speaking here of creditors’ rights against financing vehicles established by local governments, which are the focus of this Article. 29 See our discussion of the Bank of China case infra note 120 and accompanying text. 30 See Budget Law, Mar. 2, 1994, art. 28 (“The 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. The local governments shall not issue local government bonds, except as otherwise prescribed by laws or the State Council.”). 31 See Caizheng Bu Guanyu Zhuanqing Chuli Chengdu Shi Qingyang Qu Zhengfu Weigui Fafang Difang Zhengfu Zhaiquan Wenti de Han (财政部关于转请处理成都市青羊区政府违规发放地方政府债券问题的函) [Letter of the Ministry of Finance Regarding the Problem of Handling the Illegal Issuance of Local Government Bonds by the Government of Qingyang District, Chengdu, on Its Illegal Issuance of Municipal Bonds] (promulgated by the Ministry of Fin., May 23, 2002, effective May 23, 2002), https://hk.lexiscn.com/law/law-chinese-1-33364.html; Li Jiuling (李九领), Woguo Difang Zhengfu Faxing Zhaiquan de Nengli he Suzhi Fenxi (我国地方政府发行债券的能力和素质分析) [An Analysis of the Capability and Quality of Issuance of Bonds by China’s Local Governments], Zhonghua Kuaiji Wangxiao (中华会计网校) [China Online Acct. Sch.] (Apr. 6, 2007), http://bit.ly/1X3cioL; Shi Hongxiu (时红秀), Difang Zhengfu Zhaiwu Wenti Zai Taolun (地方政府债务出路问题再讨论) [Another Discussion of the Problem of Local Government Debt], 3 Yinhang Jia (银行家) [Banker] 10, 11 (2010). 32 See Guanyu Yinfa “2011 Nian Difang Zhengfu Zixing Fazhai Shidian Banfa” de Tongzhi (关于印发“2011年地方政府自行发债试点办法”的通知) [Notice of the Ministry of Finance on Issuing the “Measures for the 2011 Pilot Program of Bond Issuance” by Local Governments] (promulgated by the Ministry of Fin., Oct. 17, 2011, effective Oct. 17, 2011), https://perma.cc/6S52-HC8C. The same four governments were given permission for another bond issue in 2012. See Guanyu Yinfa “2012 Nian Difang Zhengfu Zixing Fazhai Shidian Banfa” de Tongzhi (关于印发“2012年地方政府自行发债试点办法”的通知) [Notice of the Ministry of Finance on Issuing the “Measures for the 2012 Pilot Program of Bond Issuance” by Local Governments] (promulgated by the Ministry of Fin., May 8, 2012), https://perma.cc/9Z4K-KDSR. 33 See generally Shanghai Shoufa 71 Yi Difang Zhengfu Zhaiquan (上海首发71亿地方政府债券) [Shanghai Issues First Government Bonds Worth 7.1 Billion], Xin Jing Bao (新京报) [Beijing News] B05 (Nov. 16, 2011), http://bit.ly/zsMLRx; see also Guowuyuan Pizhun Zhejiang Deng Si Di Shidian Zixing Fazhai (国务院批准浙江等4地试点自行发债) [State Council Approves Four Areas Including Zhejiang for Trial Bond Issuance], Hexun Xinwen (和讯新闻) [Hexun News] (Oct. 20, 2011), http://news.hexun.com/2011/sdzw/ (providing more comprehensive discussions about the trial local government bond issuance in China). 34 See Yu Jian (于舰), Difang Zixing Fazhai Shidian Kuorong, Su Lu Liang Sheng Bei Naru (地方自行发债试点扩容 苏鲁两省被纳入) [Pilot Program of Bond Issuance by Local Governments Expands and Include the Provinces of Jiangsu and Shandong], Diyi Caijing Ribao (第一财经日报) [First Fin. News] (July 5, 2013), http://www.yicai.com/news/2013/07/2834475.html. 35 See Quanguo Renmin Daibiao Dahui Changwu Weiyuanhui Guanyu Xiugai “Zhonghua Renmin Gongheguo Yusuan Fa” de Jueding (全国人民代表大会常务委员会关于修改“中华人民共和国预算法”的决定) [Decision of the Standing Committee of the National People’s Congress of the People’s Republic of China on the Revision of the “Budget Law of the People’s Republic of China”] (promulgated by the Standing Comm. Nat’l People’s Cong., Aug. 31, 2014, effective Aug. 31, 2014) item 30, https://perma.cc/D2YP-LXU5 (amending article 28 of the 1994 Budget Law). 36 See generally Donald Clarke, Budget Law Revisions Tighten, Not Loosen, Central Government’s Control Over Local Government Bond Issues, Chinese Law Prof Blog (Oct. 3, 2014), http://bit.ly/chineselawblog-bonds. 37 See Wong, supra note 9, at 300ff.; Tu Dejun (涂德君), Woguo Chengtouzhai Shichang Fazhan de Sange Jieduan (我国城投债市场发展的三个阶段) [The Three Phases of Development of the Chinese Market for Local Government Bonds], 2 Zhongguo Zhaiquan (中国债券) [Chinese Bonds] 14, 15 (2010). 38 The faith-based quality of this kind of financing is exemplified by the fact that China at the time still had no law on corporate organization that would have governed such a company. 39 See Wong, supra note 9, at 300. 40 As will be discussed below, these guarantees have no legal validity. 41 See Tu, supra note 37, at 15. 42 See generally Victor Shih, Local Government Debt: Big Rock-Candy Mountain, 14 China Econ. Q. 26, 26–32 (2010). 43 Zhongguo Renmin Yinhang, Zhongguo Yinhangye Jiandu Guanli Weiyuanhui Guanyu Jinyibu Jiaqiang Xindai Jiegou Tiaozheng Cujin Guomin Jingji Pingwen Jiaokuai Fazhan de Zhidao Yijian (中国人民银行、中国银行业监督管理委员会关于进一步加强信贷结构调整促进国民经济平稳较快发展的指导意见) [Guiding Opinions of the People’s Bank of China and China Banking Regulatory Commission on Further Adjusting the Credit Structure to Promote the Rapid yet Steady Development of the National Economy] (promulgated by the People’s Bank of China, China Banking Regulatory Comm’n, Mar. 18, 2009, effective Mar. 18, 2009), https://perma.cc/7AB3-WKNL. 44 See Ye, supra note 20, at 38. 45 See Wong, supra note 9, at 301 (citing Hua Qiang (华强), Gai Gei Difang Rongzi Pingtai Daishang Jinguzhou (该给地方融资平台戴上紧箍咒) [A Restraint Collar Should Be Put on Local Government Financing Vehicles], Fenghuang Wang (凤凰网) [Phoenix Network] (June 6, 2010), http://bit.ly/1RNJz3Y). Another source states that borrowing in 2009 was 4 trillion yuan, leading to total indebtedness of 7.16 trillion yuan as of the end of the year. See Ye, supra note 20, at 38. The difference in the numbers may be accounted for by different ways of categorizing local government borrowing; the lower figure may exclude certain kinds of borrowing not directly connected to LGFVs. See also China’s Official Local Debt Could Exceed CNY 14 Trl, Some Hidden Debt Excluded, Caijing (Oct. 28, 2013), http://english.caijing.com.cn/2013-10-28/113485388.html. 46 See Shi, supra note 31, at 10. 47 See NAO 2011, supra note 25, tbl.3. 48 SeeZhonghua Renmin Gongheguo Shenji Shu (中华人民共和国审计署) [National Audit Office of the People’s Republic of China], Quanguo Difang Zhengfuxing Zhaiwu Shenji Jieguo (全国地方政府性债务审计结果) [Results of Nationwide Audit of Government Debt] 6–7 tbl.3 (2013), https://perma.cc/53NW-MCLQ [hereinafter NAO 2013]. The amount for local government debt includes amounts for which local governments are not legally obligated; we discuss this issue in detail in Part I.C.2 below. 49 Guowuyuan Guanyu Jiaqiang Difang Zhengfu Rongzi Pingtai Gongsi Guanli Youguan Wenti de Tongzhi (国务院关于加强地方政府融资平台公司管理有关问题的通知) [Notice of the State Council on Issues Concerning Strengthening the Management of Local Government Financing Vehicle Companies] (promulgated by the St. Council, June 10, 2010, effective June 10, 2010), https://perma.cc/8ECH-3X4M [hereinafter State Council 2010]. 50 See Zhan Xiangyang (詹向阳), Bianzheng Kandai Difang Zhengfu Rongzi Pingtai Fazhan (辩证看待地方政府融资平台发展) [Dialectically Viewing the Development of Local Government Financing Vehicles], 7 Zhongguo Jinrong (中国金融) [China Fin.] 38, 38–40 (2010). 51 See Sufang Xie (谢素芳), Zhengfu Juzhai: Guifan he Jiandu Shuangguan Qixia (政府举债: 规范和监督双管齐下) [Government Debt: To Regulate and to Supervise Simultaneously], 6 Zhongguo Ren Da (中国人大) [Chinese People’s Congress] 35, 35 (2012). 52 See NAO 2013, supra note 48, 6–7 tbl.3. 53 See China to Refinance 11 Trln Local Govt. Debt, Shenzhen Daily, Dec. 30, 2015. 54 Id. 55 See Shih, supra note 42, at 27. 56 See Kan Huo, Local Governments Hope to Dodge Debt Deadlines, Caixin (Feb. 24, 2012), http://english.caixin.com/2012-02-24/100360506.html. 57 Id. 58 See Bettina Wassener, Report Warns of Chinese Municipal Debt Risks, N.Y. Times (Sept. 26, 2013), http://nyti.ms/1unppTY. 59 Guanyu Zhizhi Difang Zhengfu Weifa Weigui Rongzi Xingwei de Tongzhi (关于制止地方政府违法违规融资行为的通知) [Notice on Putting a Stop to Local Governments’ Illegal Financing Activities] (promulgated by the Ministry of Fin., Nat’l Dev. & Reform Comm’n, People’s Bank of China, China Banking Regulatory Comm’n, Dec. 24, 2012, effective Dec. 24, 2012), https://perma.cc/3A2N-9REC [hereinafter MOF et al. 2012]. 60 See Simon Rabinovitch, China Averts Local Government Defaults, Fin. Times (Jan. 29, 2013), http://on.ft.com/1xaPzZP. 61 See Simon Rabinovitch, China Gives Local Governments Go-Ahead to Roll Over Debt, Fin. Times (Jan. 2, 2014), http://on.ft.com/ZGMINt. 62 Guowuyuan Guanyu Jiaqiang Difang Zhengfuxing Zhaiwu Guanli de Yijian (国务院关于加强地方政府性债务管理的意见) [Opinions on Strengthening the Regulation of Governmental Debts] (promulgated by the St. Council, Oct. 2, 2014, effective Oct. 2, 2014), https://perma.cc/TD7Q-8XLT. 63 See Xinhua News Agency, China Focus: China Targets Local Gov’t Debt Risks with New Rules, Renmin Wang (人民网) [People’s Network] (Oct. 3, 2014), http://en.people.cn/n/2014/1003/c90882-8790763.html. 64 See Huo, supra note 56. 65 Guowuyuan Bangongting Zhuanfa Caizheng Bu Renmin Yinhang Yinjianhui Guanyu Tuoshan Jiejue Difang Zhengfu Rongzi Pingtai Gongsi Zaijian Xiangmu Houxu Rongzi Wenti Yijian de Tongzhi (国务院办公厅转发财政部人民银行银监会关于妥善解决地方政府融资平台公司在建项目后续融资问题意见的通知) [Notice of the General Office of the State Council Transmitting Opinion of the Ministry of Finance, the People’s Bank of China, and the China Banking Regulatory Commission on Appropriately Handling the Follow-up Financing Issues of In-Process Projects of Local Government Financing Vehicle Companies] (promulgated by the Gen. Off. of the St. Council, May 15, 2015, effective May 15, 2015), https://perma.cc/8SBU-TYPM. 66 Lingling Wei, China Backtracks on Local Government Debt, Wall St. J. (May 15, 2015), http://on.wsj.com/1KSd5ku. 67 See, e.g., the information about the issuer provided in the Leshan, Yongzhou, Tianjin, Zhuzhou, and Nanjing prospectuses, listed in the Appendix. 68 See, e.g., the information about the issuer provided in the Hebei Bohai and Zhenjiang prospectuses, listed in the Appendix. 69 See Xu Yuan (徐瑗), Hu Zuliu: Caozuo Bu Touming Shi Difang Rongzi Zhenzheng Fengxian (胡祖六: 操作不透明是地方融资真正风险) [Hu Zuliu: Lack of Transparency in Operations Is the Real Risk of Local Financing], Xinlang Xinwen (新浪新闻) [Sina News] (July 5, 2011), http://bit.ly/zpeOu5 (interview with Hu Zuliu, Goldman Sachs (Asia) LLC’s Executive Director and Asia Pacific chief economist). 70 See Difang Zhengfu Rongzi Pingtai Gongsi Luanxiang (地方政府融资平台公司乱象) [The Chaos of Local Government Financing Vehicles], Dongfang Zaobao (东方早报) [Oriental Morning Post] A33 (June 28, 2011), http://bit.ly/1JF90xS. 71 See State Council 2010, supra note 49. 72 See id. 73 See, e.g., Guanyu Jinyibu Guifan Difang Zhengfu Tourongzi Pingtai Gongsi Faxing Zhaiquan Xingwei Youguan Wenti De Tongzhi (关于进一步规范地方政府投融资平台公司发行债券行为有关问题的通知) [Notice on Relevant Issues in Further Standardizing the Acts of Local Government Investment and Financing Vehicle Companies in Issuing Bonds] (issued by Office of the Nat’l Dev. & Reform Comm’n, Nov. 20, 2010, effective Nov. 20, 2010), https://perma.cc/KM6Z-4GUU; see also Guanyu Guanche Guowuyuan Guanyu Jiaqiang Difang Zhengfu Rongzi Pingtai Gongsi Guanli Youguan Wenti De Tongzhi Xiangguan Shixiang De Tongzhi (关于贯彻国务院关于加强地方政府融资平台公司管理有关问题的通知相关事项的通知) [Notice on Matters Related to the Implementation of the State Council’s Notice on Issues Regarding the Strengthening of Control over Local Government Financing Vehicles] (jointly issued by Ministry of Fin., Nat’l Dev. & Reform Comm’n, People’s Bank of China, and China Banking Regulatory Comm’n, July 30, 2010, effective July 30, 2010), https://perma.cc/BJQ7-V8UL [hereinafter MOF et al. 2010]. 74 Standard Chartered, On the Ground: Global Research: China—Local Debt: USD 3.6–4.0tn and Counting 2, Oct. 2, 2013 (on file with first author). 75 SeeMega Prospectus, listed in the Appendix, at iii. 76 See id., at cover p., 107. 77 See Andrew Ang, Jennie Bai & Hao Zhou, The Great Wall of Debt: The Cross Section of Chinese Local Government Credit Spreads (Georgetown McDonough School of Business, Research Paper No. 2603022, 2015; Columbia Business School, Research Paper No. 15–57, 2015), http://ssrn.com/abstract=2603022. 78 See BNP Paribas, Equities Research: China Banks: 10 Questions on the LGFV Debt Swap Programme 1 (June 25, 2015), https://perma.cc/6PRB-R6RE. 79 See Ang, Bai & Zhou, supra note 77. 80 See Mudi: Difang Zhai Pingji Weishenme Dou Shi AAA? (穆迪: 地方债评级为什么都是AAA?) [Moody’s: Why Is All the Local Debt Rated AAA?], Diyi Caijing (第一财经) [First Fin. & Econ.] (June 12, 2015), http://bit.ly/1TZd05g (citing Moody’s Vice General Manager Zhu Shuning). 81 See China’s Second LGFV Bond Sale in Doubt as State Support Pulled, Bloomberg News (Dec. 16, 2014), http://bloom.bg/1EtPTLF. 82 See infra Part II.B.1 and notes 109–22 and accompanying text. 83 See Guowuyuan Guanyu Jiaqiang Difang Zhengfuxing Zhaiwu Guanli de Yijian (国务院关于加强地方政府性债务管理的意见) [Opinions on Strengthening the Regulation of Governmental Debts] (promulgated by the St. Council, Oct. 2, 2014, effective Oct. 2, 2014), https://perma.cc/TD7Q-8XLT. 84 State Council 2010, supra note 49; MOF et al. 2010, supra note 73. 85 See Lingling Wei, Shanghai Chaori in Default on Bond Interest Payments, Wall St. J. (Mar. 6, 2014), http://on.wsj.com/1pBfS6s. The article does not add the qualification “reform-era China”, but we feel it is wise, since there may well have been defaults in the early years of the People’s Republic of China on bonds issued before 1949. 86 See Pete Sweeney, Chinese Bank Sues Local Govt Property Company in Rare Public Spat, Reuters (July 2, 2014), http://reut.rs/1unnleW. Of course, it could always be argued that the defaulting company was not an LGFV, but a different kind of company that was nevertheless wholly owned by the local government. But this comes perilously close to arguing that LGFVs are essentially by definition any government-owned company that we see after the fact getting its debt paid off by local government. This is not helpful to creditors or analysts ex ante. 87 Guowuyuan Bangongting Guanyu Zuohao Difang Zhengfuxing Zhaiwu Shenji Gongzuo de Tongzhi (国务院办公厅关于做好地方政府性债务审计工作的通知) [Notice of the General Office of the State Council on Properly Handling the Audit of the Governmental Debts] (promulgated by the Gen. Off. of the St. Council, Feb. 13, 2011, effective Feb. 13, 2011), http://laws.66law.cn/law-129676.aspx. 88 For an example of the classifications in use, see supraTable 1. 89 Notice of the General Office of the State Council on Properly Handling the Audit of the Governmental Debts, Feb. 13, 2011. The term used here, gongyong shiye danwei (公用事业单位), which we have translated as “institutions with a public purpose,” refers to institutions such as schools, hospitals, railroads, and public utilities. Here and elsewhere, “to be repayable” is intended to convey an obligation, and is not to be understood as a statement of fact about the borrower’s financial capability. 90 Id. 91 Id. (emphasis added). 92 For a canvassing of the various ways government “debt” can be defined, including the counting of unfunded pension liabilities or revocable legislative promises, see Jennifer Weiner, Assessing the Affordability of State Debt, Fed. Reserve Bank of Boston (Dec. 2013),http://bit.ly/weiner-statedebt (New England Public Policy Center Research Report 13-2). 93 NAO 2013, supra note 48, at 6. 94 See id. 95 See id. at 7. 96 See supra note 32 and accompanying text. 97 See NAO 2011, supra note 25, introductory sec. 98 See NAO 2013, supra note 48, at 6. 99 See NAO 2011, supra note 25, sec. 1.5. As discussed in the preceding subsection, and in Part II.B.1, what the NAO counts as direct and contingent debt liabilities are not necessarily such, even in the Chinese system. 100 See id., sec. 3.3. 101 See The Chaos of Local Government Financing Vehicles, supra note 70. 102 See Fan Xuyang (范盱阳) & Liu Ximing (刘西明), Shixi Difang Zhengfu Rongzi Pingtai De Guifan Yu Chuangxin (试析地方政府融资平台的规范与创新) [A Tentative Analysis of the Regulation and Innovation of Local Government Financing Vehicles], 12 Xin Jinrong (新金融) [New Fin.] 42, 42 (2010), http://bit.ly/1ErnCEX. 103 See Jason Subler & Lu Jianxin, Analysis: Local Government Debt Alarm May Delay China Muni Bonds, Reuters (July 20, 2011), http://reut.rs/1ADZ9J5. 104 See China Regulators, Officials Aim to Ease Local Finance Vehicle Loan Concern, Bloomberg News (July 27, 2010), http://bloom.bg/9InLcH (according to one source, only 27% of loans to LGFVs can be repaid solely by the profits generated by the projects funded by LGFV financing); see also Chinese Banks See Risks in 23% of $1.1 Trillion Loans, Bloomberg News (July 26, 2010), http://bloom.bg/bIaQ6H. Another survey revealed an even more troubling figure. According to a questionnaire survey that was conducted by the Institute of Finance & Banking, Chinese Academy of Social Sciences in 110 cities in 2010, only 10% of LGFVs relied on the income from financed projects as the primary source to repay their debt, whereas about 90% of LGFVs relied on other sources, such as land development arrangements and fiscal transfers, for repayment. See Lei Jiaxiao & Peng Bo, Difang Zhengfu Rongzi Pingtai Cunzai Tizhi Jizhi Fengxian (地方政府融资平台存在体制机制风险) [Institutional and Mechanism Risks Exist in Local Government Financing Vehicles], Keji Ribao (科技日报) [Sci. & Tech. Daily] (Jan. 30, 2011), http://bit.ly/1LLTerM. 105 See, e.g., Datong Prospectus, Jilin Prospectus, Panjin Prospectus, Tianjin Binhai Prospectus, Yichang Prospectus, and Zhenjiang Prospectus, listed in the Appendix . 106 SeeBaoji Prospectus, Jilin Prospectus, Mega Prospectus, Yichang Prospectus, Yichun Prospectus, and Zhenjiang Prospectus, listed in the Appendix. 107 See, e.g., Baoji Prospectus, listed in the Appendix. 108 According to one study, research in Guangdong province revealed that typically the general office of the county or municipal People’s Congress Standing Committee would issue a letter “acknowledging” the loan deal between the bank and the local government. See Liang, supra note 22, at 17–21. The letter would recite that the loan was in conformity with borrowing plans set out in the local government’s most recent Work Report approved by the local People’s Congress, and was “acknowledged” (queren) by the Standing Committee. According to the study, banks supply the form of such letters and insist that they be signed before making the loan. See id. at 20 (showing form of letter). Importantly, it is addressed to the local government, not to the lending bank. The study also reproduces the form of a local People’s Congress Standing Committee resolution supporting a loan to a LGFV. See id. at 20–21. The resolution recites the Standing Committee’s approval of the loan and states that any part of the loan not repaid by the LGFV will be repaid from the local government’s budget. 109 Danbao Fa (担保法) [Guarantee Law] (promulgated by the Standing Comm. Nat’l People’s Cong., June 30, 1995, effective Oct. 1, 1995), art. 8, https://perma.cc/GU58-JJ6Q. This prohibition has now been written into the Budget Law as well. See Quanguo Renmin Daibiao Dahui Changwu Weiyuanhui Guanyu Xiugai “Zhonghua Renmin Gongheguo Yusuan Fa” de Jueding (全国人民代表大会常务委员会关于修改”中华人民共和国预算法”的决定) [Decision of the Standing Committee of the National People’s Congress of the People’s Republic of China on the Revision of the “Budget Law of the People’s Republic of China”] (promulgated by the Standing Comm. Nat’l People’s Cong., Aug. 31, 2014, effective Aug. 31, 2014) item 30, https://perma.cc/7EJ9-6UBU. 110 GITIC’s Empty Offers, Economist (Jan. 14, 1999), http://www.economist.com/node/182583. 111 See Wu Daqing, Guanyu Guifan Difang Zhengfu Tourongzi Pingtai De Jidian Jianyi (关于规范地方政府投融资平台的几点建议) [Suggestions on Regulating Local Government Investment and Financing Vehicles], Jinrong Shibao (金融时报) [Fin. News] (Feb. 8, 2010), http://bit.ly/1JqyZdm. See also the discussion of recent court cases infra at notes 119–20 and accompanying text. 112 See Xiaomin Song (宋晓敏) & Yuan Tian (田原), Zhuce Kuaijishi Zai Difang Zhengfu Rongzi Pingtai Gongsi Shenji Zhong Ying Guanzhu de Wenti (注册会计师在地方政府融资平台公司审计中应关注的问题) [The Issues to Which Registered Auditors Should Pay Close Attention When Auditing Local Government Financing Vehicle Companies], 7 Caiwu Yu Kuaiji (财务与会计) [Fin. & Acct.] 54, 54–55 (2011). 113 See State Council 2010, supra note 49. 114 MOF et al. 2010, supra note 73. 115 Caizheng Bu Guanyu Guifan Difang Geji Zhengfu Bumen Juzhai he Danbao Chengnuo Xingwei de Tongzhi (财政部关于规范地方各级政府部门举债和担保承诺行为的通知) [Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels] (promulgated by the Ministry of Fin., Sept. 2, 2010, effective Sept. 2, 2010), https://perma.cc/85RB-PMWM. Oddly enough, the Notice itself is marked “Not to Be Made Public” (bu yu gongkai). It is mentioned in a 2011 press release from the Ministry of Finance: see Caizheng Bu Fuzeren Jiu Jiaqiang Difang Zhengfuxing Zhaiwu Guanli Dawen (财政部负责人就加强地方政府性债务管理答问) [Question Answering by the Leading Official Regarding Strengthening the Regulation of Governmental Debts], Zhonghua Renmin Gongheguo Caizheng Bu (中华人民共和国财政部) [Ministry of Finance of the People’s Republic of China] (Aug. 11, 2011), https://perma.cc/8C9N-Q96W. 116 See Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels, Sept. 2, 2010. 117 See MOF et al. 2012, supra note 59. 118 The theory of liability for fault in the invalidity of a guarantee can be found in general terms in article 58 of the Contract Law (liability for invalid contracts) and much more specifically in article 5 of the Guarantee Law, which states that when a guarantee contract is found to be invalid, the debtor, creditor, and guarantor shall each bear civil liability in proportion to their fault. This provision was then further elucidated by a Supreme People’s Court interpretation in 2000. This interpretation repeated that government guarantees were invalid and that the damage to the creditor from invalidity should be handled in accordance with article 5 of the Guarantee Law. It added, however, that where the creditor and the guarantor were both at fault, the guarantor’s liability was not to exceed half of the unpaid debt. See Zuigao Renmin Fayuan Guanyu Shiyong “Zhonghua Renmin Gongheguo Danbao Fa” Ruogan Wenti De Jieshi (最高人民法院关于适用“中华人民共和国担保法”若干问题的解释) [Judicial Interpretation of the Supreme People’s Court on Some Issues Regarding the Application of the Guarantee Law of the People’s Republic of China] (promulgated by Sup. People’s Ct., Dec. 8, 2000, effective Dec. 13, 2000), arts. 3,7, https://perma.cc/2AXK-AQ4V. In the cases we have reviewed, guarantors are routinely found at fault for making guarantees they should not have made, and creditors are equally routinely found at fault at the same time for carelessly accepting a guarantee from an entity they should have known did not have the capacity to make it. The most sophisticated exposition of this theory can be found in a pragmatic and well-reasoned judgment of the Guangdong Province Higher-Level People’s Court in the case of Foshan Shi Chancheng Qu Renmin Zhengfu Yu Jiaotong Yinhang Xianggang Fenhang Jiekuan Danbao Hetong Jiufen Shangsu’an (佛山市禅城区人民政府与交通银行香港分行借款担保合同纠纷上诉案) [Appeal Case of Foshan City Chancheng District People’s Government v. Communications Bank of China, Hong Kong Branch] (Guangdong Higher People’s Ct. Jan. 22, 2003), https://perma.cc/5SB6-J9ZA [hereinafter Foshan Case]. The case and the principles behind it are discussed by Professor Yang Lixin in Yang Lixin (杨立新), Chengnuo Han, Zuigao’e Baozheng, Wuxiao Baozheng Peichang Zeren (承诺函, 最高额保证, 无效保证赔偿责任) [Liability for Commitment Letters, Maximum Guarantees, and Invalid Guarantees], 110 Falü Zixun Wang (法律咨询网) [Legal Advice] (July 11, 2010), https://perma.cc/L93N-FP6B. There are other cases with the same reasoning and result: see, e.g., Zhen Zhengfu Danbao Qiye Jiekuan, Fayuan Pan Danbao Wuxiao, Zhen Zhengfu You Guocuo (镇政府担保企业借款 法院判担保无效 镇政府有过错) [County Government Guarantees Loan to Enterprise, Court Rules that Guarantee Is Invalid but that County Government Is at Fault], Chengdu Shangbao (成都商报) [Chengdu Bus. News] (July 9, 2010), https://perma.cc/7HWK-FU9S (reporting that the township government was found liable for half of unpaid debt on the theory of fault for invalidity of guarantee); Jian Zhengbo (简正波), Guojia Jiguan Danbao Wuxiao, Peichang Zeren Buneng Mianchu (国家机关担保无效 赔偿责任不能免除) [State Organ Cannot Be Exempt from Liability Even if Its Guarantee Was Invalid], Jiujiang Fayuan Wang (九江法院网) [Jiujiang Ct. Network] (May 16, 2012), https://perma.cc/TZ9V-B9AW (reporting a case from Jiangxi province where subcounty local government purported to guarantee a debt and was found liable for half the unpaid amount on the same theory); Chen Guiping (陈桂平), Guquan Zhuanrang Jiufen: Zhengfu Liandai Danbao Xiaoli Ruhe (股权转让纠纷—政府连带担保效力如何) [Equity Transfer Dispute: What Is the Validity of Government’s Joint and Several Guarantee], Zhongguo Lüshi Wang (中国律师网) [China Lawyer Network] (Dec. 30, 2013), https://perma.cc/B79C-8WAQ (discussing a 2005 Heilongjiang case involving the government of Tongjiang City with similar circumstances, similar liability, and similar theory). 119 See Zhou Weishe (周巍摄), Guanzhou Zhongyuan: Difang Zhengfu Danbao Beisu Tonglei Anjian Jinnian Yongxian (广州中院: 地方政府担保被诉同类案件近年涌现) [Guangzhou Intermediate People’s Court: Many Cases in Recent Years In Which Local Governments Are Sued on Their Guarantees], Yangchen Wanbao (羊城晚报) [Yangcheng Evening News] (May 22, 2014), https://perma.cc/52XU-V7QD. 120 Zhongguo Yinhang (Xianggang) Youxian Gongsi yu Liaoning Sheng Renmin Zhengfu, Hulu Dao Xinchang Baozheng Hetong Jiufen Ershen (中国银行(香港)有限公司与辽宁省人民政府、葫芦岛锌厂保证合同纠纷二审) [Second-Instance Adjudication Between Bank of China (Hong Kong), Ltd., Liaoning People’s Gov't, and Hulu Island Zinc Plant (Bank of China)] (Sup. People’s Ct. Dec. 22, 2014), https://perma.cc/8RLZ-TPSB. 121 Foshan Case, supra note 118. 122 See, e.g., Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels, Sept. 2, 2010. 123 Control over personnel and finances of courts in China rests with political power at the same administrative level at the court. Formally, this means the local People’s Congress and People’s Government respectively. In practice, it means the local Party authorities. 124 See Yao Liang (姚亮), Difang Zhengfu Rongzi Pingtai Gongsi Daikuan De Danbao Wenti Yanjiu (地方政府融资平台公司贷款的担保问题研究) [Study of the Problem of Guarantees for Loans to Local Government Financing Vehicle Companies], Yao Lüshi de Weibo (姚律师的微博) [Attorney Yao’s Blog] (Aug. 9, 2010), https://perma.cc/9HD6-9JSX. 125 There is some dispute in the literature over the degree to which the widespread story that promotion depends on delivering GDP is actually true. Certainly it is widely believed in the Chinese media. See, e.g., Tan Yingzi, Evaluation for Officials Hinges on Public Input, China Daily (Aug. 13, 2009), http://bit.ly/1sZtwZi (discussing new system for assessing public officials that will look at more than local economic output). See generally Victor Shih et al., Getting Ahead in the Communist Party: Explaining the Advancement of Central Committee Members in China, 106 Am. Pol. Sci. Rev. 166, 166–87 (2012), for a review of the academic literature; the authors conclude from their own research that there is in fact no relation. 126 This observation is quite common in the Chinese media. See, e.g., Zuo Xiaolei (左晓蕾), Renqi Nei Yusuan Ying Yueshu Ying Yu Dingfang Zhai Tongbu (任期内预算硬约束应与地方债同步) [Hard Budget Constraint During Term of Office Should Synchronize with Municipal Debt], Shanghai Zhengquan Bao (上海证券报) [Shanghai Sec. News] (Oct. 24, 2011), https://perma.cc/AQ4V-SRJ4. 127 See Loudi Prospectus, Hebei Bohai Prospectus, Lishui Prospectus, Shaoyang Prospectus, Xianyang Prospectus, and Yongzhou Prospectus, listed in the Appendix. See also Kelvin Soh & Aileen Wang, Special Report: China’s Debt Pileup Raises Risk of Hard Landing, Reuters (Oct. 10, 2011), http://reut.rs/mZERlu (noting that land use rights are often used as collateral for loans to LGFVs). 128 Danbao Fa (担保法) [Guarantee Law] (promulgated by the Standing Comm. Nat’l People’s Cong., June 30, 1995, effective Oct. 1, 1995), art. 34, https://perma.cc/GU58-JJ6Q. 129 A reversion is the technical term for the interest remaining in the hands of one who grants to another an interest in land with a shorter term than his/her own term. For example, if X has a fee simple (an interest in perpetuity) and grants a twenty-year lease to Y in year Z, then X’s remaining interest (a future interest that ripens to a fee simple in year Z+20) is called the reversion. 130 See Zhonghua Renmin Gongheguo Chengshi Fangdichan Guanli Fa (中华人民共和国城市房地产管理法) [Urban Real Estate Administration Law of the People’s Republic of China] (promulgated by the Standing Comm. Nat’l People’s Cong., July 5, 1994, effective Jan. 1, 1995, amended Aug. 30, 2007 and Aug. 27, 2009), art. 51, https://perma.cc/YT4A-U4WU. 131 In the Hami Prospectus, the land use rights purportedly securing the debt were disclosed to be so-called “allocated” (huabo) land and hence essentially worthless as security for the reasons explained above. It is also worth noting that at the end of 2012, the Ministry of Finance and other departments issued a joint notice requiring, among other things, that any allocated land transferred to LGFVs had to be paid for and converted to granted land. See MOF et al. 2012, supra note 59. The Hami Prospectus states that the parcels in question were acquired in May 2013, and the bonds were not issued until July 2014. Thus, the land use rights held by the LGFV in question appear to be in violation of Ministry of Finance rules. 132 See generally China Cities Value Land at Winnetka Prices with Bonds Seen Toxic, supra note 3. In this article, the reporter interviewed some governmental officials in charge of the LGFV financing. In the northeast city of Cangzhou, the person in charge of a large-scale bond sale for a port expansion project could not even locate on a map the land his company had supposedly put up as collateral. 133 See George E. Peterson, Land Leasing and Land Sale as an Infrastructure-Financing Option 6 (World Bank Policy Research, Working Paper No. 4043, 2006), https://perma.cc/9YVB-8JZZ. 134 Bullish analysts often point to various factors such as increasing urbanization as reasons why land prices will continue to rise. What they do not explain, however, is why such widely known factors are not already incorporated into the current price of land. 135 See Jamil Anderlini, Chinese Cities See Property Prices Fall, Fin. Times (Jan. 18, 2015), http://on.ft.com/1Mb2AMc. 136 See Chen Bingcai (陈炳才), Tian Qing (田青) & Li Feng (李峰), Difang Zhengfu Rongzi Pingtai Fengxian Fangfan Duice (地方政府融资平台风险防范对策) [Preventive Measures Against the Risks of Local Government Financing Platform Companies], 1 Zhongguo Jinrong (中国金融) [China Fin.] (2010), http://bit.ly/z0Bbe2. 137 See Naoki Murakami, The Capital Market and Credit Rating in China (Ctr. for China and Asian Stud., Nihon Univ. College of Econ., Sept. 2009), https://perma.cc/4WGN-YUYQ. 138 See Brian Milner, Shot Down by SEC, Chinese Rater Punishes U.S. Debt, Globe & Mail (Nov. 9, 2010), http://bit.ly/QuRV4o. 139 See China Cities Value Land at Winnetka Prices with Bonds Seen Toxic, supra note 3. Another credit rating agency, China International Capital Corp., gave the bonds its third-lowest rating, a non-investment grade. See id. 140 See, e.g., Hebei Bohai Prospectus, Xianyang Prospectus, and Yongzhou Prospectus, listed in the Appendix. 141 See, e.g., Tracy Alloway, China’s Uncollateralised, Cash Flow-Less, Local Government Loans,Fin. Times (UK) (June 7, 2011), http://on.ft.com/1NHzs0W (pointing out that while the valuation of LURs as collateral has attracted attention, the larger problem remains “those dodgy uncollateralized loans”). 142 See Wu, supra note 111. There is a lively metaphor for such a phenomenon in China: “to marry one daughter to more than one person.” Unlike bigamy, there is nothing per se unlawful about creating multiple encumbrances on the same collateral, and the law provides a method of deciding priority. The unlawfulness occurs when the debtor falsely claims that no previous encumbrances exist. 143 See Jiang Shulin (江抒琳), Xing Shihe (邢世和) & Wang Yuyin (王玉印), Tudi Diya Yu Jinrong Fengxian (土地抵押与金融风险) [Land Mortgage and Finance Risks], inKexue Heli Yong Di, Ren Di Hexie Xiangchu 2008 Nian Xueshu Nianhui Lunwen Ji (科学合理用地, 人地和谐相处2008年学术年会论文集) [Proceedings of 2008 Annual Academic Meeting on the Scientific and Reasonable Use of Land, with Man and Land Harmoniously Coexisting] 168 (Lin Yaohua (林耀华) ed., 2008). 144 See id. at 165. 145 See generally Taisu Zhang, The Pragmatic Court: Reinterpreting the Supreme People’s Court of China, 25 Colum. J. Asian L. 1 (2012). 146 See Zhonghua Renmin Gongheguo Chengshi Fangdichan Guanli Fa (中华人民共和国城市房地产管理法) [Urban Real Estate Administration Law] (promulgated by the Standing Comm. Nat’l People’s Cong., July 5, 1994, effective Jan. 1, 1995, amended Aug. 30, 2007), art. 39, https://perma.cc/YT4A-U4WU. 147 See Jiang, Xing & Wang, supra note 143, at 165. 148 See Urban Real Estate Administration Law, Jan. 1, 1995, art. 16. 149 This point is made by Jiang, Xing & Wang, supra note 143, at 166. 150 See MOF et al. 2012, supra note 59. 151 SeeShaoyang Prospectus, listed in the Appendix, at 62–66; see alsoLoudi Prospectus, listed in the Appendix, at 70–73. 152 SeeHebei Bohai Prospectus, listed in the Appendix. 153 SeeYichun Prospectus, Hefei Prospectus, Yichang Prospectus, Zhuzhou Prospectus, and Jiangyin Prospectus, listed in the Appendix. The Datong Prospectus notes the existence of the receivable as an assurance to bondholders, but does not state that a security interest has been created in their favor. 154 See supra Part II.B.1 and notes 109–26 and accompanying text. 155 We confirmed this point in a discussion in August 2014 with a group of Chinese lawyers specializing in local government debt. 156 Baoji Prospectus, listed in the Appendix, at 10. 157 Leshan Prospectus, listed in the Appendix, at 3. 158 Mega Prospectus, listed in the Appendix, at 95. 159 Id. 160 Danbao Fa (担保法) [Guarantee Law] (promulgated by the Standing Comm. Nat’l People’s Cong., June 30, 1995, effective Oct. 1, 1995), art. 7, https://perma.cc/GU58-JJ6Q. 161 See Patrick Chovanec, Falling Dominoes, Patrick Chovanec Blog (June 27, 2012), https://perma.cc/R3GA-EDCA. 162 I am indebted to Patrick Chovanec, No Guarantee, Patrick Chovanec Blog (May 19, 2012), https://perma.cc/DT8B-5A9A, for the analysis in this and the following two paragraphs. 163 The concept of correlated risk is explained more fully at Correlated Risks, World Fin., http://www.worldfinance.com/home/risk-encyclopaedia/correlated-risks (last visited Oct. 3, 2014). 164 See generally Yang Na & Ma Yuan, Fool’s Gold Behind Beijing Loan Guarantees, Caixin (Mar. 20, 2012), http://english.caixin.com/2012-03-20/100370485.html. 165 See Fan & Liu, supra note 102. 166 See, e.g., Mega Prospectus, listed in the Appendix, at 92–93 (“Description of Other Material Indebtedness”). 167 SeeBaoji Prospectus, listed in the Appendix, at 73. 168 See id.; see also Qiye Zhunze Di 22 Hao: Jinrong Gongju Queren He Jiliang (企业会计准则第22号—金融工具确认和计量) [Accounting Standards for Enterprises No. 22—Recognition and Measurement of Financial Instruments] (promulgated by Ministry of Fin., Feb. 15, 2006, effective Feb. 15, 2006), art. 18, https://perma.cc/42QT-KZJS (defining “sellable financial assets”). 169 The Chinese term is jihua danlie shi: a city separately listed under the state plan. The idea is simply that of a city sufficiently large and important to be treated as a province for economic planning purposes. Although the scope of economic planning has declined drastically in post-Mao China, the category of “separately planned city” remains when it is desired for whatever reason to include large cities within the scope of activities generally reserved for provinces. 170 See Du, supra note 6, Chen Jia, Expand Local Debt-Swap Plan to Further Reduce Burden, Says Official, China Daily, Dec. 22, 2015. For a detailed analysis, see Barry Naughton, Local Debt Restructuring: A Case of Ongoing Authoritarian Reform, 47 China Leadership Monitor (Summer 2015), http://hvr.co/1QBWqUu. 171 Caizheng Bu Youguan Fuzeren Jiu Faxing Difang Zhengfu Zhaiquan Zhihuan Cunliang Zhaiwu Youguan Wenti Da Jizhe Wen (财政部有关负责人就发行地方政府债券置换存量债务有关问题答记者问) [The Leading Director of the Ministry of Finance Answers Reporters’ Questions Concerning Issuing Government Bonds to Replace the Remaining Debt], Zhonghua Renmin Gongheguo Caizheng Bu (中华人民共和国财政部) [Ministry of Fin.] (Mar. 12, 2015), https://perma.cc/WRQ8-H3GX. 172 See id. 173 See Caizheng Bu Youguan Fuzeren Jiu Xiada Di 2 Pi Zhihuan Zhaiquan Edu Youguan Wenti Da Jizhe Wen (财政部有关负责人就下达第2批置换债券额度有关问题答记者问) [Responsible Person of the Ministry of Finance Answers Reporters’ Questions Concerning Allocation of Second Quota in Bonds-for-Debt Swap], Zhonghua Renmin Gongheguo Caizheng (中华人民共和国财政部) [Ministry of Fin.] (June 10, 2015), https://perma.cc/PRQ4-4FM5. 174 See Chen, supra note 170. 175 See China To Refinance 11 Trln Local Govt. Debt, supra note 53 (11.14 trillion estimate); China’s Regional Financial Crisis, supra note 11 (25 trillion estimate). 176 The definitions of these categories of debt are discussed in Part II.A. 177 See Supporting China’s Infrastructure Stimulus Under the INFRA Platform, INFRA Update (World Bank) (June 2010), http://bit.ly/1Libfx6 (“[O]nly a small part of the stimulus was reflected in the central government’s budget. Additional financing—largely bank lending—contributed almost two-thirds of the stimulus.”); Yu Yongding, China’s Response to the Global Financial Crisis, East Asia Forum (Jan. 25, 2010), http://bit.ly/1J7j7js (“The central government is financing one-quarter of the 4 trillion yuan package, in the form of direct grants and interest rate subsidies. Bank credit is the second most important source of finance for the stimulus package.”). 178 Although article 35 of the 2015 Budget Law reserves bond-issuing powers at the subcentral level to provincial-level governments, see supra note 36 and accompanying text, the Ministry of Finance in 2015 interpreted this to include separately planned cities. See, e.g., Difang Zhengfu Yiban Zhaiquan Faxing Guanli Zanxing Banfa (地方政府一般债券发行管理暂行办法) [Interim Measures for the Administration of the Issuance of Ordinary Bonds by Local Governments] (promulgated by the Ministry of Fin., Mar. 12, 2015, effective Mar. 12, 2015), https://perma.cc/H87R-8VXT. (The concept of separately planned cities is explained supra in note 169.) More recently, it expanded its interpretation still further to include free trade zones. See Guanyu Zuohao 2016 Nian Difang Zhengfu Zhaiquan Faxing Gongzuo de Tongzhi (关于做好2016年地方政府债券发行工作的通知) [Notice on Doing Well the Work of Issuance of Local Government Bonds in 2016] (promulgated by the Ministry of Fin., Jan. 25, 2016), sec. 6, para. 2, https://perma.cc/T7Z9-MG3P. The legal implications of this apparent disregard for a clear statutory rule are beyond the scope of this Article. 179 See supra note 120 and accompanying text for a discussion of the Bank of China case. 180 One can often see this phenomenon in real estate transactions. Consider the following report: In the Hongshan district of Wuhan City, 927 mu of collective mountain and forest land was sold illegally four times in succession, producing a profit for the work units and people involved of over 40 million RMB. A budget audit of the State Food and Drug Supervision and Management Bureau revealed that in December 2003, the organization services department of that bureau signed, without any authority, a contract with two companies to sell for 85 million RMB 216 mu of land in the Chanping District, Beijing. The companies were to use the land for running a business and development. That land had been provided to the Bureau without cost by the State. This land sale was not reported to the departments concerned for approval. This was done without any of the required legal procedures in violation of the Land Management Law and the regulations issued by the State Council to bring order into the real estate market. After the audit, the Bureau took the matter seriously and is now studying how to rectify the matter. Li Jinhua (李金华) [Director, National Audit Office], Guanyu 2003 niandu zhongyang yusuan zhixing he qita caizheng shouzhi de shenji gongzuo baogao (关于2003年度中央预算执行和其他财政收支的审计工作报告) [Audit Report on the Implementation and Financial Receipts and Expenditures of the 2003 Central Budget] (June 24, 2004), https://perma.cc/K9LD-BUXB (translation courtesy of David Cowhig). One way of rectifying the matter would simply be to refuse to recognize the sale, and to support the reoccupation of the land by whoever had the right to it before the sale. Buyers in illegal transactions would quickly learn their lesson and insist that legal procedures be followed before parting with their money. But that apparently is not a step the authorities were willing to take. © The Author(s) [2017]. Published by Oxford University Press on behalf of the American Society of Comparative Law. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Comparative Law Oxford University Press

The Law of China’s Local Government Debt: Local Government Financing Vehicles and Their Bonds

Loading next page...
 
/lp/ou_press/the-law-of-china-s-local-government-debt-local-government-financing-mVXYv8GRKl
Publisher
American Society of Comparative Law
Copyright
© The Author(s) [2017]. Published by Oxford University Press on behalf of the American Society of Comparative Law. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0002-919X
eISSN
2326-9197
D.O.I.
10.1093/ajcl/avx036
Publisher site
See Article on Publisher Site

Abstract

Abstract Local government financing vehicles (LGFVs)—companies capitalized and owned by local government and established for the purpose of raising funds for municipal infrastructure construction—emerged in China in the 1980s as a response to the severe constraints on indebtedness by local governments themselves. The mushrooming of their number and indebtedness has sparked fears about their ability to repay the debt and the consequences of a default. In addition to taking on bank debt, a number of LGFVs have also issued bonds. While observers have questioned the value of collateral typically offered as security for the bonds, we know of no extensive analysis to date of the legal quality of the collateral: what exactly are the bondholders being promised, and what is the status of those promises in the Chinese legal system? This Article is an attempt to answer that question, using data from two hand-collected samples of LGFV bond prospectuses from different regions in China in two different time periods. We find that current collateralization practices vary a great deal across bond issues and have changed over time, and we discuss the legal and other problems attendant upon each type of backup. Remarkably, we find that, unlike our initial sample of bond issues, recent bond issues virtually all state explicitly in the prospectus that they carry no security. Thus, the popular image of local governments wildly overpromising with guarantees they are not legally empowered to give seems, at least as far as recent bond issues are concerned, to be wholly wrong. This in turn calls into question the figures commonly provided for local government debt, since they often include LGFV debt that local government is neither legally nor morally obligated to pay. To be sure, they may wish to pay creditors voluntarily, but it is misleading to label as “debt” soft obligations of this nature. Creditors who have tried to force local governments to make good on their guarantees have uniformly failed, at most receiving half of what they sought. The argument that local governments have some politically enforceable obligation to pay on their guarantees does not seem supported by the evidence. Introduction Local government financing vehicles (LGFVs)—that is, companies capitalized and owned by local government1 and established for the purpose of raising funds for municipal infrastructure construction—emerged in China in the 1980s as a response to the severe constraints on indebtedness by local governments themselves.2 In the last several years, their number and indebtedness have mushroomed, sparking fears about their ability to repay the debt and the consequences of a default. In addition to taking on bank debt, a number of LGFVs have also issued bonds, something generally prohibited to local governments. While observers have justifiably questioned the economic value of land and other collateral typically offered as security for the bonds,3 we know of no extensive analysis to date of the legal quality of the collateral: what exactly are the bondholders being promised, and what is the status of those promises in the Chinese legal system? This Article is an attempt to answer that question. The primary data for this study comes from two hand-collected samples of LGFV bond prospectuses from different regions in China in two different time periods.4 Group I consists of nineteen issues from 2009 to 2011, and Group II consists of twenty issues from 2014. The prospectuses in Group I were selected with a view to understanding the picture in various parts of the country, but they were not selected on the basis of the repayment safeguards offered. The Group II prospectuses were selected entirely randomly from a large group of 2014 bond issues. We begin with a brief overview of the Chinese LGFV financing landscape. We then discuss in detail the main categories of backup5 for the bondholders. We find that current collateralization practices vary a great deal across bond issues and have changed over time. The three main ways—typically alternative and not additive—in which bonds are backed up are land use rights, local government guarantees, and guarantees from third-party companies. Some bond issues are explicitly not backed up by anything other than the issuer’s own assets. We discuss the legal and other problems attendant upon each type of backup. Remarkably, we find that, unlike our initial sample of bond issues, recent bond issues virtually all state explicitly in the prospectus that they carry no security. Thus, the popular image of local governments wildly overpromising with guarantees they are not legally empowered to give seems, at least as far as recent bond issues are concerned, to be wholly wrong. This in turn calls into question the figures commonly provided for local government debt, since they often include LGFV debt which, as we shall show, local government is neither legally nor morally obligated to pay. It may turn out that handwringing over LGFV debt is no longer necessary. In 2015, a new Budget Law came into effect that promised to regularize the issue of bonds by provincial-level governments, and in the spring of 2015 the Chinese government announced a plan whereby LGFV debt would be replaced by bonds issued by provincial-level governments.6 These policies have the potential to fundamentally change the regime described here; if the swap is completed, for example, many of the issues raised here would essentially be finessed away. Nevertheless, even if the swap is successful, we believe that the legal issues involved in LGFV debt are very likely to reappear at another time in another context, and therefore are still worth exploring and clarifying. In particular, it is important to clarify the legal picture in order to assess the government’s response now and its likely response in similar situations in the future. I. Municipal Government Financing and LGFVs A. Municipal Finance Generally LGFVs arose as a response to an irresistible force and an immovable object. The irresistible force was the heavy responsibility of local governments for the provision of services, a responsibility often unfunded by the central government. The immovable object was the inability of local governments to raise money through taxation or borrowing: local governments are not allowed to run budget deficits and may not generally issue bonds.7 The result has been the toleration—and during the financial crisis, the encouragement—of backdoor practices that permit subnational governments (SNGs) to meet their spending responsibilities. 1. Responsibility for Provision of Services Municipal governments in China are responsible for a vast array of government expenditures, including all vital services and infrastructure. Although China is formally a unitary state, in 2008,8 for example, central government expenditures were just 21% of national budgetary expenditures.9 By contrast, even federal states such as Canada and the United States exhibit much higher percentages: 43% (in 2002) and 51% (in 1999) respectively.10 At the same time, the intergovernmental financial system is weak. SNGs suffer large vertical fiscal gaps—that is, a mismatch between assigned tax revenues and assigned expenditures. For example, in 2008, SNGs were responsible for 79% of government expenditures but were directly allocated only 43% of tax revenues.11 The remaining expenditures must be met from intergovernmental transfers and nontax sources of revenue, and transfers are inadequate.12 2. Restrictions on Revenue Raising In spite of their considerable responsibilities, municipal governments and SNGs generally suffer considerable restrictions on their revenue-raising abilities. a. Restrictions on Taxing Power SNGs cannot meet financial needs simply by imposing new taxes at will. Nevertheless, the often-heard assertion that local governments in China have no taxing power needs to be qualified. It is true that both the 1992 Law on the Administration of Tax Collection13 and a 1993 decision of the State Council14—the decision that was the core of the 1994 tax system reform—purport to grant exclusive authority over taxes to the central government. But the rules contained in these documents have not been strictly adhered to by the central authorities themselves, and local governments have consistently maintained the power in practice to reduce effective tax rates through measures such as rebates.15 In the words of a leading scholar of the Chinese tax system, these and other facts cast doubt on whether the centralization of tax legislative power is indeed a fundamental normative principle for Chinese law and policy, and whether it is not instead, as a more skeptical view would hold, a mere slogan wielded by the central government as a non-neutral party in political contests. . . . [F]iscal federalism in taxation in China is still very much in an unsettled state, and does not represent a coherent model.16 As a legal matter, therefore, it is not completely clear that SNGs, at least in the form of People’s Congresses17 at the level of the province and certain cities, cannot impose (in substance or in effect) new taxes. As a political matter, on the other hand, it is clear that the central government almost always disfavors local imposition of new taxes, and in practice SNGs have not been imposing new taxes as a response even to severe financial pressures. Moreover, the center’s share of tax revenue, after subtracting revenue collected by the center and then returned to provinces, has been steadily increasing since the 1994 tax reform: from 21.2% in 1994 to 46.1% in 2014.18 b. Restrictions on Ability to Borrow SNGs are severely restricted in their ability to borrow from sources other than superior levels of government, who may not be willing to lend. It is not always clear, however, what the various prohibitions actually mean within the Chinese legal system. They might mean that the prohibited debt is legally void and unenforceable, like a contract entered into by a minor. The prohibited party is simply incapable of entering into a binding debt agreement. But a prohibition could instead mean that the legal system recognizes the validity of the debt, but that the governmental body or particular officials responsible for incurring it are subject to some kind of administrative or other sanction.19 If the legal system did not recognize the validity of the debt, the punished misbehavior could not occur. i. Bank Debt Governmental units in China have limited legal authority to borrow from banks. It is commonly said that SNGs are forbidden to borrow from banks,20 and there is some legal basis for this claim. The 1996 Lending Principles issued by the People’s Bank of China (PBOC), at that time the bank regulator, forbid lending to entities other than natural persons with legal capacity or enterprise legal persons, other economic organizations, or individual industrial-commercial households, in each case properly registered with the State Administration of Industry and Commerce.21 This list of permissible borrowers does not include governmental units, and so has been interpreted by some as banning loans to them.22 Note, however, that the Lending Principles can be viewed as a kind of internal directive within the banking system. Thus, it does not necessarily make loans to governmental units invalid, except perhaps via an argument that the borrower knew or should have known that the lender had no authority to lend. While the PBOC Lending Principles are at least arguably a lender-side prohibition, the Budget Law (in both the original 1994 version and its new 2015 version) has what can also be viewed as a borrower-side prohibition. Local governments are required to balance their budgets; they may not have deficits, and they are specifically forbidden to issue bonds absent authorization by statute or the State Council.23 Since the preceding article in the Budget Law specifically provided for the conditions under which the central government may take loans, it is reasonable to infer that, at least as far as the Budget Law is concerned, local governments may not do so under any conditions. On the other hand, articles 55 and 73 of the 1994 Budget Law clearly contemplated that borrowing could be a legitimate part of a SNG budget—although interestingly their corresponding articles in the 2015 Budget Law (articles 70 and 92) do not. Moreover, official documents and reports appear to confirm that SNGs do borrow directly from banks.24 An important and widely cited 2011 National Audit Office (NAO) report on local government debt explicitly distinguished LGFV debt from debt borrowed and owed directly by local governments, and found the latter to be substantial25 (but also unlawful).26 Whether bank debt is viewed as lawfully or unlawfully incurred, however, commentators appear to be unanimous that the obligation is legally valid. It is not, for example, similar to gambling debts or an obligation to pay usurious interest rates, both of which are unenforceable in some jurisdictions. The constant stream of government documents sternly warning local governments not to incur debt illegally is evidence of this.27 If the obligation were invalid, there would be no need to be concerned about local governments incurring it. At least in theory, the government could effectively eliminate all unlawful local government debt incurrence by simply making it known that creditors’ rights against local governments per se28 would not be enforced and their money would be lost, and following up with appropriate court action (or more likely, inaction). Unlawful lending would dry up immediately. Why the government has not adopted this policy option is not clear. There may be practical considerations of which we are not aware that vitiate the theoretical outcome. On the other hand, a recent Supreme People’s Court case suggests that the government may have decided to signal that it will pursue precisely this policy.29 ii. Bonds Local governments were specifically forbidden under the 1994 Budget Law to issue bonds except as permitted by statute or the State Council.30 As bond issues are highly visible, this prohibition was generally effective. Nevertheless, one local government—that of Qingyang District in Chengdu, Sichuan—managed astonishingly to issue bonds for 375 million yuan over ten years, from 1992 to 2002, before the violation was discovered by the Ministry of Finance and shut down.31 Only in 2011 did the State Council first provide the explicit permission required by law for four local governments (Shanghai, Zhejiang, Guangdong, and Shenzhen) to issue bonds directly,32 and in November of that year, the Shanghai municipal government issued bonds in the amount of 7.1 billion yuan.33 More recently, the provincial governments of Shandong and Jiangsu were granted permission to issue bonds.34 In general, therefore, fundraising through bonds had to be done through LGFVs—making such bonds in form corporate bonds—or not at all. There appears to have been some debate over how the revised Budget Law should treat local government bond issues, and successive drafts treated the matter differently. In the end, however, the final version of the amendments passed by the National People’s Congress Standing Committee provides that provincial-level governments may, with State Council approval, issue bonds.35 Analysts have, without contradiction from the central government, universally interpreted the revisions as liberalizing the regime for local government bond issues. This slightly misstates the content of the revisions. By adding a considerable amount of text to set forth the process for issuing local government debt, the revisions certainly hint that more such debt is contemplated. But although the central government may plan on approving the issuance of more local government debt, the approval process is actually tightened. Under the old Budget Law, local governments at many levels could issue bonds for many purposes, and State Council approval was adequate. Under the revised law, bond issues may be by provincial-level governments only and may not be for commercial projects. Moreover, the revisions add the requirement that an overall quota for local government debt must be approved by the National People’s Congress or its Standing Committee.36 The revisions could most accurately be understood as signaling that the government intends to approve quotas for local government debt as a routine matter and not as an exceptional measure, but that the debt issuers and the approval process will be tightly controlled from the top. B. LGFVs as a Response 1. History of LGFVs In order to avoid the constraints on direct borrowing, local governments began as early as the 1980s to turn to special-purpose vehicles (despite the absence at the time of a law on corporate organization) to raise funds for public investment. These entities were often created for a single purpose—construction of toll roads or utilities, for example, because such projects had a revenue stream that could be used to service the debt—and could take loans from international financial institutions.37 A breakthrough came in 1992 with the formation of the Shanghai Municipal Construction Investment and Development General Company. This was a general-purpose entity whose mission was the construction of facilities such as water supplies, roads, and sewers. This took the financing available for urban infrastructure in Shanghai from a few billion yuan per year to 17 billion in 1993 and 38 billion in 1994.38 This model soon spread to other areas, and most cities had established investment corporations by 2000.39 The separation from public finance became less stringent, and some local governments offered guarantees to lender banks.40 The investment companies also sold bonds; by the end of 2008, there were thirty-one LGFV bond issues totaling 46 billion yuan.41 The next stage—and the real leap forward—in LGFV development was occasioned by the financial crisis. The Chinese government announced a 4 trillion yuan stimulus, much of it to be funded by local government borrowing through LGFVs.42 In March 2009, the People’s Bank of China and the China Banking Regulatory Commission issued a joint notice encouraging the use of LGFV financing,43 and borrowing took off. Over 3,200 SNGs in China have the authority to establish LGFVs: 31 provincial-level units, 333 prefectural-level units, and 2,859 county-level units. They had established 10,321 LGFVs by the end of 2009, an increase of 7,000 over 2008.44 In 2009 alone, there were 115 LGFV bond issues in the total amount of 167.7 billion yuan. By 2009, LGFVs had already accumulated 5 trillion yuan in bank loans, and in that year alone took on another 3 trillion.45 Outstanding loans to LGFVs represented 240% of local government income, and 90% of the loans came from banks.46 According to the National Audit Office, which appears to calculate LGFV debt more conservatively, by the end of 2010, LGFVs carried a debt load of over 4.97 trillion yuan;47 by mid-2013, this had increased to 6.97 trillion yuan, almost 40% of the total 17.9 trillion yuan debt load of local governments.48 As state agencies became aware of the ballooning debt, they realized at the same time that very little was known about LGFVs. Since the middle of 2009, therefore, the government has engaged in a massive effort to find out more about them, and the State Audit Office launched a major investigation in the spring of 2011 covering all thirty-one provincial-level jurisdictions and involving 41,300 auditors. Furthermore, in 2010, the State Council issued a document intended to rein in their activities.49 This apparently had some effect: according to People’s Bank of China, the growth in annual borrowing by LGFVs slowed to under 20% in 2010, from over 50% in 2009.50 One source asserts that LGFVs incurred only 300 billion yuan in new debt in 2011.51 Nevertheless, a National Audit Office study of thirty-six local governments found that their indebtedness had increased by almost 13% over the two years of 2011 and 2012.52 As of the end of 2014, officially counted local government debt was put at 15.4 trillion yuan by China’s Minister of Finance, with 1.06 trillion yuan as approved government bond issues and the rest bank loans and borrowing through LGFVs.53 It was expected to reach 16 trillion by the end of 2015.54 Despite any slowing in the rate of increasing indebtedness, however, it seems clear that the total amount is unlikely to be repayable.55 In early 2012, central and local government officials began negotiating with banks over how to deal with the estimated 1.8 trillion yuan in loans to LGFVs scheduled to mature in 2012.56 According to contemporary news reports, “[b]anks are being asked to rejigger repayment schedules and let borrowers settle maturing debt with fresh loans under a risk-mitigation plan drafted by the central government.”57 A 2013 report by Nomura stated that half of LGFV debt due on 2012 would have been at risk of default were it not for local government support.58 Moreover, in late 2012, several Chinese government departments issued yet another joint notice intended to rein in LGFV borrowing, indicating that the problem had yet to be brought under control.59 News reports indicate that about three quarters of the 4 trillion yuan that local governments were to repay to banks by the end of 2012—this figure likely includes, but is not limited to, LGFV debt—was in fact rolled over.60 More recently, on December 31, 2013, the National Development and Reform Commission officially authorized local governments to roll over LGFV debt through yet more LGFV bond issues.61 The last two years have seen yet more government ambivalence about LGFVs. In October 2014, the State Council issued its famous Document 43,62 which was intended to put the brakes on LGFV financing.63 It essentially said that local governments could no longer borrow money through LGFVs. In the same month, the National Development and Reform Commission, which has jurisdiction over corporate bonds generally, issued a document raising the bar for corporate bond issues.64 In May 2015, however, perhaps concerned by an economic slowdown, three government agencies jointly issued Document 40,65 which was widely viewed as reversing many of the policies embodied in Document 43. In the words of one report, Under [Document 43], [LGFVs] were barred from borrowing additional funds starting this year, as the government sought to close what it dubbed “the back door” for localities to borrow. Instead, all borrowing would be done by the local governments themselves and be appropriately disclosed and reflected in their budget plans. The purpose was to rein in runaway local-debt growth and make local borrowings more transparent. According to the latest directive, local financing firms can continue to get loans from banks to fund ongoing projects. If the local firms have trouble repaying their bank debts, the rule says, their loan contracts should be “renegotiated and extended.”66 2. Structure of LGFVs LGFVs are usually organized under China’s Company Law as limited liability companies. Usually, either the local government itself or the local branch of the State Asset Supervision and Administration Commission (SASAC) is the sole shareholder.67 In some cases, the LGFV is a joint venture among two or three entities such as state-owned enterprises (SOEs) affiliated with the local government and other government or quasi-governmental agencies.68 Most LGFVs focus on local infrastructure construction projects as well as other public welfare projects such as education, environmental protection, and medical care.69 Some engage in side business activities of various kinds as well.70 In 2010, the State Council issued a notice that for various purposes classified LGFVs into three classes.71 The first class of LGFVs are those used to finance nonprofit public welfare projects which will not generate income, and therefore will pay off their debts mainly by cash transfers from government funds. The second class of LGFVs are those used to finance public welfare projects that return a steady profit, and therefore are expected to pay their debts with income from those projects. The third class of LGFVs are those used to finance nonpublic welfare projects.72 Many government agencies, including the National Development and Reform Commission and the Ministry of Finance, have issued various directives or notices that interpret and implement the Notice.73 3. LGFV Bonds As discussed above, while local governments may not issue bonds, companies may, and LGFVs are companies. As of 2013, approximately eight or nine hundred LGFVs out of a total of about 10,800 had publicly issued bonds.74 LGFV bonds are usually underwritten by a group of underwriters led by a lead underwriter and issued to Chinese institutional investors. Some bond issuances, however, are clearly aimed at non-Chinese investors. For example, the offering prospectus in an issue by a Beijing LGFV contains the following language: Pursuant to this offering, the Securities may not be offered or sold in the United States, except that the Securities may be offered or sold to persons who are “qualified institutional buyers,” as defined in Rule 144A under the Securities Act in reliance on the exemption from the registration requirements of the Securities Act provided by Rule 144A.75 The underwriters in that bond issue included BofA Merrill Lynch, Morgan Stanley, and Credit Suisse, among others, and the governing law was stipulated to be New York law.76 C. Confusion in the System Although the rules and principles of government and LGFV borrowing seem fairly clear, there is evidence of a great deal of confusion and variation in practice. This could be due to the sources’ being careless in their terminology, but could also be a sign that actual practice diverges widely from what the law prescribes. 1. Do LGFV Bonds Have an Implicit Government Guarantee? It is frequently asserted that LGFV bonds carry an implicit government guarantee even where such a guarantee is not explicit. One recent paper even asserts that this guarantee comes from the central government, not just the local government that owns the issuing LGFV.77 We question this assertion. Since an “implicit” guarantee is by definition unstated, no affirmative evidence can be adduced for it. The support for the assertion frequently comes from another assertion that bond buyers believe that the bonds are implicitly guaranteed. We call this the Tinkerbell theory of rights: if you believe in something hard enough, you can bring it into existence. But the Chinese legal system does not adopt this theory. More to the point, empirical evidence contradicts the claim that bond buyers believe an implicit guarantee exists. Most obviously, if LGFV bond buyers truly believed that all LGFV bonds carried a central government guarantee, we would expect to see identical credit ratings on all LGFV bonds. But we do not.78 We would also expect to see identical interest rates on bonds with identical maturities and liquidity characteristics. Again, we do not. A group of scholars working with a dataset of LGFV bond issues from 1997 to 2014 found that LGFV bond interest rates reflected both economic risk and political risk.79 When they were swapped for provincial government obligations under a program introduced in the spring of 2015, Moody’s rated the bonds Triple-A, regardless of which province issued them, on the grounds that they were (in the view of Moody’s) implicitly backed by the central government,80 a judgment Moody’s had not (to the best of our knowledge) made about LGFV bonds. The financial system itself recognized the riskiness of LGFV bonds: the clearing agency for exchange-traded debt announced that LGFV bonds rated lower than AAA could not be used as collateral for short-term loans because they were too risky.81 Finally, there is no support in law for the notion of an implicit guarantee. As discussed below,82 government guarantees are explicitly labeled invalid in the Guarantee Law of the People’s Republic of China (“Guarantee Law”), the only law that could make them valid. Moreover, in case the Guarantee Law was not clear enough, the central government has issued pronouncements both disclaiming its own responsibility for local debt83 and reminding the public that local government guarantees are invalid.84 Above and beyond these theoretical considerations, we now have a bit of empirical evidence as well. Alarming reports have recently surfaced of corporate debtors, both ordinary corporations and LGFVs, defaulting in unprecedented ways. In March 2014, solar equipment maker Chaori defaulted on a bond, marking the first corporate bond default in reform-era China.85 More worrisome from the standpoint of LGFV creditors is the reported first default of an LGFV on a bank loan, which came to light in mid-2014.86 Clearly this is one LGFV whose debt was not backed by the government. In light of the above, it is hard to see what evidence could support the claim that implicit government guarantees either exist or are even widely believed by investors to exist. Both documentary evidence and empirical observation suggest the contrary. 2. What Do National Audit Office Categories Mean? A further source of confusion is the National Audit Office’s way of categorizing LGFV and local government debt. The NAO has carried out two major investigations of LGFV and local government debt, the first in 2011 and the second in 2013. These investigations were carried out pursuant to instructions from the State Council first set forth in a 2011 document.87 That document charged the NAO with classifying local government debt into three types.88 Table 1. Local Government Debt as of June 30, 2013 (in Billions of Yuan).a Type of entity incurring debt  Debt for which local government bears repayment responsibility (Type I debtb)  Contingent government debt  Debt for which the local government is liable as a guarantor (Type II debtc)  Debt for which government may undertake a certain responsibility for assistance (Type III debtd)  Financing vehicle companies  4,075.55  883.25  2,011.64  Government departments and organs  3,091.34  968.42  0  Institutions with subsidized expenses  1,776.19  103.17  515.71  State-owned financing or holding companies  1,156.25  575.41  1,403.93  Institutions responsible for their own revenues and expenditures  346.29  37.79  218.46  Other units  316.26  83.14  0  Public service institutions  124.03  14.39  189.64  Total  10,885.92  2,665.58  4,339.37  Type of entity incurring debt  Debt for which local government bears repayment responsibility (Type I debtb)  Contingent government debt  Debt for which the local government is liable as a guarantor (Type II debtc)  Debt for which government may undertake a certain responsibility for assistance (Type III debtd)  Financing vehicle companies  4,075.55  883.25  2,011.64  Government departments and organs  3,091.34  968.42  0  Institutions with subsidized expenses  1,776.19  103.17  515.71  State-owned financing or holding companies  1,156.25  575.41  1,403.93  Institutions responsible for their own revenues and expenditures  346.29  37.79  218.46  Other units  316.26  83.14  0  Public service institutions  124.03  14.39  189.64  Total  10,885.92  2,665.58  4,339.37  a This Table is a direct reproduction of the data in NAO 2013, supra note 48, at 6–7 tbl.3. Totals may not equal sum of amounts in columns due to rounding. b See the definition in the text accompanying note 89. c See the definition in the text accompanying note 90. d See the definition in the text accompanying note 91. View Large Type I debt is described as “debt for which local government bears repayment responsibility,” and is defined more precisely as follows: Borrowing undertaken by local government (including government departments and agencies), institutions financially supported [by local government], institutions with a public purpose, government financing vehicles, and other relevant units, that is confirmed to be repayable from government budgetary finances, where the government bears direct responsibility for repayment. The first kind is local government bonds, national bonds converted to loans, foreign debt converted to loans, loans for overall development of agriculture, and other governmental debts converted to loans that have been confirmed to be repayable from government budgetary finances. The second kind is debt incurred by government financing vehicles, government departments and agencies, institutions financially supported [by local government], institutions with a public purpose, and other units through borrowing, owing, repurchase obligations, or other means that has been confirmed to be repayable from government budgetary finances (not including revenues from road tolls, tuition, and the like). The third type is policy-related amounts payable by local government food enterprises and supply and marketing enterprises.89 Type II debt is described as “debt for which the local government is liable as a guarantor,” and is defined more precisely as follows: Debt for which a local government (including government departments and agencies) has provided a direct or indirect guarantee such that when the debtor is unable to repay the debt, the government bears joint liability for repayment. The first type is borrowing undertaken by government financing vehicles, institutions financially supported [by local government], institutions with a public purpose, and other units that has been confirmed to be repayable from the operational revenues (including tuition fees), business revenues (including road tolls), or other non-[government] budgetary funds of the debtor unit, with respect to which the local government (including government departments and agencies) has provided a direct or indirect guarantee. The second type is borrowing undertaken by a local government (including government departments and agencies) that is to be repaid from non-budgetary funds; such debt is to be deemed debt guaranteed by government.90 Type III debt is described as “other relevant debt,” and is defined more precisely as follows: Borrowing undertaken for projects with a public purpose by government financing vehicles, institutions financially supported [by local government], and institutions with a public purpose that is to be repaid from non-budgetary funds, where the local government (including government departments and agencies) has not provided a guarantee (not including debt owed to other units or individuals). The government does not as a legal matter bear any liability for this type of debt, but when the debtor faces a debt crisis, it may be necessary for the government to shoulder responsibility for providing relief.91 The problem with these categories is that, as we have noted above, legal borrowing by governments and government departments is difficult—the PBOC Lending Principles seem to forbid it—and government guarantees are unquestionably invalid as a legal matter. What type of debt, then, is classed as Type I and Type II debt, and how is it different from Type III debt? The standard answer regarding Type I and Type II debt is that the NAO does not use strict legal categories for classifying debt, and includes in Types I and II obligations for which the local government is not liable in a strict legal sense, but can nevertheless be meaningfully said to be liable for a variety of reasons such as informal promises, budgetary commitments from the local government, or political pressures. The problem with this answer is that debt for which local government is not liable in a strict legal sense seems to be precisely the kind of debt that the State Council wants categorized as Type III debt. Moreover, it makes any figures for such debt useless for comparative purposes, since statistics for local government debt in other countries may or may not include amounts for which local government is not legally liable.92 One answer might lie in the distinction between repayment from budgetary funds and repayment from non-budgetary funds. In some circumstances, local People’s Congresses may provide that repayment of debt shall come from budgetary funds, and such debt is specifically excluded from Type III. Any budgetary commitment made by a People’s Congress may, of course, be later unmade by the same or a different People’s Congress, so such a provision is not legally binding. Nevertheless, perhaps it is considered sufficient to deem the debt Type I debt. The NAO reports offer further paradoxes in their breakdowns of Type I and Type II debt. Take Type I debt—debt for which the local government is said to be directly responsible—in the 2013 NAO report.93 Fully 28% (3.1 trillion yuan) of it is classified as LGFV debt.94 This makes no sense. If local government is directly (i.e., primarily) responsible for payment, then it, not an LGFV, is the debtor. If the LGFV is the primary debtor, then the debt should be classified as Type II or Type III debt. Nevertheless, the State Council’s instructions contemplate just such a subcategory, but it is difficult to know what exactly it meant. Moreover, slightly under 6% (614.6 billion yuan) of local government Type I debt is said to come from bond issues.95 In its 2011 report, the NAO put the figure at 551 billion yuan. As we have seen, local governments at that time did not have authority to issue bonds absent specific authorization from the State Council. And the numbers cannot be attributed to bonds issued by other entities that the local government is deemed to have guaranteed, whether lawfully or unlawfully, since such an obligation should clearly by Type II or Type III debt. The listed debt might be attributable to legally issued bonds, but one wonders. The first publicized instances of State Council permission for local governments to issue bonds were in late 2011.96 Such bonds could not have been included in the 2011 NAO Report, however, since the investigation on which it was based concluded in May 2011.97 As for Type II debt—debt for which local government is said to be contingently responsible through a guarantee—33% (883 billion yuan) is classified as LGFV debt.98 But as we have seen, local government cannot legally be even indirectly responsible for LGFV debt. A small amount (49 billion yuan) is even attributed to local government guarantees of local government bonds. It is difficult to know what this means: even if the bonds were somehow lawfully issued, debtors do not guarantee their own obligations. In any case, as noted below, in our review of bond prospectuses we did not find a single one in which a government purported to offer a legally binding guarantee. In short, despite the detail in the State Council’s instructions to the NAO, we cannot feel confident that we understand the criteria the NAO is using to classify local government debt. We can probably rule out certain criteria, however, including the criterion of legal obligations. Thus, it is important to recognize that when government officials and commentators speak of local government debt obligations in China, they are not speaking of legal obligations. Consequently, comparisons with legal obligations of local governments in other times and places are dangerous. II. What Stands Behind LGFV Bonds? Safeguards for Repayment A. Introduction LGFV bonds are risky. Although individual bond issues vary in their risk, it seems very unlikely that they can as a whole be repaid. As of the end of 2010, the total direct and contingent (via guarantees) obligations of SNGs were put by the NAO at 70%of their annual revenue sources.99 And at the same time, seventy-eight city-level and ninety-nine county-level governments had direct obligations of more than 100% of their annual revenues.100 In 2010, 358 LGFVs borrowed new debt to cover their old debt, and 148 LGFVs had overdue debt, totaling approximately 8 billion yuan.101 LGFV-funded projects typically do not generate large revenue streams; a survey conducted by the People’s Bank of China revealed that the average profitability of the assets owned by LGFVs is less than 1.3%.102 In some cases this is normal and intended: a highly efficient and worthwhile public works project may have positive externalities that cannot be charged for, and thus may not generate a cash return adequate to repay the investment. In other cases, the project itself may be misconceived: local leaders have strong incentives to show local GDP growth, banks (especially since the 2008 stimulus program began) have strong incentives to lend, and the result can be white elephant projects that contribute to GDP in the same way that digging a hole and filling it up again does: toll roads without cars, residential blocks without inhabitants, and airports without airplanes.103 In short, only a very small fraction of LGFV debt can be repaid in full by cash generated by their investment projects.104 These repayment concerns make necessary a closer look at what stands behind LGFV debt and, for the purposes of this Article, LGFV bonds in particular. In some cases, the bond issues offer what purports to be a formal security interest in assets or an income stream. Other issues purport to be secured by a guarantee. While the economic value of the assets is an important issue worth considering, the focus of this Article is on the legal status of the various mechanisms offered as a safeguard for repayment. Since whether such safeguards amount to a security interest in the legal sense is precisely one of the important issues to be determined, and because not all safeguards indeed amount to security interests but are not necessarily worthless for that reason, we shall generally use the term “repayment safeguard” or “backup” instead of “security interest.” This usage is in fact reflected in the offer prospectuses themselves, which almost always use the term “repayment safeguard measures” (偿债保证措施 or 偿还保障措施) to head a section setting forth the various reasons—economic, legal, formal and informal, binding and nonbinding—why investors should feel reassured about repayment. B. Safeguards As noted in the Introduction, for this Article we reviewed LGFV bond offer prospectuses from different regions in China in two different time periods and examined the proffered safeguards. Some bond issues explicitly stated that there was no security for the bonds, although they did offer other reasons for buyers to have confidence that they would be paid. Other issues offered security that would be recognized under the Guarantee Law: mortgaged land use rights, a pledge of future receivables, and third-party guarantees. Falling somewhere between nothing and something were assurances of the local government’s “strong support” (大力支持), including at times a description of the local government plans to allocate a certain amount of funds periodically to the LGFV in order to help it repay the bondholders. Several bond issues stated that funds for scheduled payments of principal and interest would be placed at some point into a special bank account that could only be used for such payments. As we shall discuss, the legal robustness of this structure is questionable. Absent a formal security interest in the account granted to bondholders, it is hard to see how other creditors could be prevented from getting their hands on such funds. Interestingly, nineteen of the twenty Group II prospectuses explicitly stated that there was no security for repayment, a marked change from the Group I prospectuses. In the following subsection, we provide a detailed examination of the various safeguard methods. We discuss why they do or do not qualify as a valid security interest, in what respects they are inadequate even if they do qualify, and how their deficiencies might be cured. 1. No Security A number of bond issues explicitly state that there is no security for repayment.105 In most cases, however, the offer prospectuses attempt to provide reasons for bondholders to have confidence that they will be repaid. In some cases, the assurance amounts to nothing more than a promise that the business will be run efficiently and show a profit. In other cases of both secured and unsecured debt, the offer recites that the LGFV has the “strong support” (大力支持) of the local government.106 This support can take the form of special business opportunities and even commitments to make good through appropriations any shortfall in LGFV repayments.107 Some offer prospectuses also note the existence of special bank accounts into which the local government will periodically make payments, and state that the funds from those accounts can be used only to pay back the debt. None of these safeguards amounts to a security interest in the sense of an asset of reasonably reliable value to which bondholders have top-priority access. First, the “special accounts” described in so many prospectuses do not seem to have any legal status under China’s Guarantee Law and thus do not offer any special protection for creditors. At any given moment, funds intended for the account are owned either by the local government or by the LGFV. (It is not clear who has formal ownership over the account itself, but it must be one or the other.) If the funds belong to the local government, they are off limits to LGFV creditors. Once transferred to the LGFV, they are available to all creditors, not just bondholders. The only way bondholders can protect these funds from other creditors is to have a proper security interest in them, but the disclosures say nothing about such a security interest having been established in the accounts. Without such a security interest, the prior claim of the bondholders to funds in the account is, at most, a contractual one between the LGFV and the bondholders—the LGFV could not, for example, use the funds for some other purpose—but that contractual claim cannot trump the interests of other creditors who did not agree to it. Second, the various assurances of support that come from local governments and People’s Congresses do not bind either them or their successors. If a People’s Congress passes a resolution promising to support a certain project, a successor can pass another resolution undoing the first. All that such resolutions can do as a legal matter is forestall an argument that a local government was somehow acting ultra vires in establishing a LGFV and having it issue bonds; they do not operate to bind the local government (or the local People’s Congress) to repay the bonds.108 Many of these assurances fall under the general rubric of informal guarantees by local governments of LGFV debt. Such guarantees must remain informal, however, because formal guarantees of indebtedness by local governments are prohibited, invalid, and unenforceable. The Guarantee Law provides “government agencies cannot act as guarantors, except in the case of sub-lending loans from foreign governments or international economic organizations as approved by the State Council of the People’s Republic of China.”109 Obviously, government guarantees for LGFV bonds do not fall within this limited exception. Indeed, there is a solid precedent on precisely this point: the 1999 bankruptcy of the Guangdong International Trust and Investment Corporation. In that case, all the “loan registrations, state and provincial guarantees, and ‘comfort letters’ with which foreign loans are trussed up count[ed] for little.”110 Investors have no realistic prospects of winning a case based on a government guarantee in case of default,111 and, tellingly, auditors are told to treat them as invalid.112 Various government agencies have periodically restated this rule just to make sure everyone gets the point. In June 2010, the State Council issued a notice emphasizing that local governments had to follow the rules of the Guarantee Law and must not issue guarantees.113 The following month, the Ministry of Finance, together with several other government departments, issued a notice further clarifying the State Council’s notice.114 This notice repeated that all guarantees were unlawful, although in a backhanded way seemed to acknowledge the binding nature of previous guarantees by stating that as of the date of the notice, local governments would no longer be liable for the debts of LGFVs (beyond, of course, the loss of their investment). In September 2010, the Ministry of Finance issued a “Notice on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels.”115 The notice states, inter alia, that local governments must “strictly implement the provisions of [the Guarantee Law], and may not issue guarantees, promises, comfort letters, or other forms of direct or disguised guarantee . . . they may not promise to bear liability for repayment of debt undertaken for the financing of other enterprises or institutions . . . .”116 And in December 2012, the Ministry of Finance together with several other departments issued yet another notice117 repeating the exact language of the 2010 notice. Litigation since the enactment of the Guarantee Law tells the same story, although with an interesting twist on the question of remedies. Up until the end of 2014, creditors who sued governmental entities on their guarantees seem to have received a consistent result in Chinese courts: the guarantee was held invalid and unenforceable, but the defendant guarantor was held liable for up to half (never more) of the unpaid debt on a theory that it shared fault with the creditor for the invalidity of the guarantee contract, and relevant law provided for liability in such cases.118 In May 2014, the Guangdong Province Higher-Level People’s Court held a press conference specifically to announce that in recent years they had processed a large number of lawsuits seeking to enforce guarantees against local governments and that such guarantees were invalid. It then presented what it called a typical case, in which the local government was held liable for one third of the unpaid debt on the theory of fault discussed above.119 One problem with this way of handling cases of government guarantees is that it seems to go directly against the intent behind the prohibition on government guarantees: to prevent local governments from becoming liable for the debts of third parties. Making local governments liable for half of the unpaid debt on a theory of fault does an end run around the prohibition, and means that for all intents and purposes local governments in fact do have the capacity to guarantee up to half the debt of third parties—even more if a court finds that the creditor was without fault. Although the courts’ unwillingness to let local governments simply disavow their promises without any consequences is understandable, surely this is not the result the drafters of the Guarantee Law intended or expected by their prohibition. A recent Supreme People’s Court (SPC) case from late 2014, however, has changed the picture in favor of local governments. In Bank of China v. Liaoning Provincial People’s Government,120 the creditor sued the Liaoning provincial government to enforce a guarantee the defendant had purportedly made, and the SPC found the government liable for no part of the debt. The case bears a very strong factual resemblance to a 2002 Guangdong province case in which the Foshan city government was found partially liable on a guarantee.121 In that case, the Foshan government had issued a document apparently called a “promise letter” (chengnuo han). The letter promised that if the debtor was late in paying the debt, the government would “take responsibility for resolving the situation” and “not cause [the creditor] to suffer economic loss” (bushi . . . zai jingji shang mengshou sunshi). The word “guarantee” was apparently never used. Despite the defendant’s argument that this promise letter created only a responsibility to try to get the debtor to repay and was not a formal guarantee, the court held that it clearly indicated an intention to issue a guarantee. Of course, the court also found that the purported guarantee was invalid, but finding that this language constituted an attempt to make a guarantee meant that the court could find the Foshan government liable for half the unpaid debt on a theory of fault. In the Bank of China case, however, the SPC looked at virtually identical language and found the defendant entirely free of liability. As in the Foshan case, the provincial government had issued a document entitled “Promise Letter” (承诺函), stating inter alia that the provincial government promised that if the borrower could not repay any amount owing under the terms of the loan, the government would “assist in resolving the debt owed by the borrower” to the bank and “would not let the bank suffer any economic loss” (不让贵行在经济上猛兽任何损失), language which is if anything slightly stronger than that of the Foshan case. Nevertheless, the SPC found that the promise did not constitute an attempt to make a formal guarantee, and constituted instead merely a kind of comfort letter that did not purport to be legally binding. Unlike in previous cases, the defendant provincial government was found not liable at all for any of the debt it had (at least arguably) purported to guarantee. Given that the Bank of China case is an SPC case and thus presumably influential, it suggests the following set of rules for future cases. First, if the word “guarantee” is not used, governments will not be held liable for any amount of debt that they purport to guarantee. Second, if the word “guarantee” is used, governments may be found to have attempted to make a guarantee, and if so, will be held liable for some portion of the debt depending on the court’s judgment as to the relative fault of the creditor and the guarantor for the invalidity of the guarantee. Courts are likely to find the guarantor at fault and liable for half, but not more, of the unpaid amount of the debt. In sum, guarantees by governmental entities are prohibited and invalid; courts will not recognize them and will not even attempt to enforce them. Still, some puzzles remain. First, if the guarantees are simply invalid, it is hard to understand why the government needs to keep telling local governments not to issue them.122 An invalid guarantee is not a problem for the purported guarantor; it’s a problem for the creditor. The government should be issuing warnings to creditors not to have faith in such guarantees. Second, both the State Council and the NAO seem to believe that unlawful government guarantees are nevertheless valid, since in the 2011 and 2013 NAO reports on local government debt, a category exists (i.e., Type II) for hard obligations represented by local government guarantees. Thus, we face essentially the same puzzle here that we examined in the context of direct debt: if the central government does not want local government guarantees to exist, why not take the route of consistently refusing to recognize their validity to any degree? Creditors would rapidly learn to ignore local government promises (or at least to consider them ephemeral and nonbinding). Legal systems that do not wish to see minors bound by contracts do not in general accomplish this purpose by threatening to punish minors who enter into contracts. Instead, they simply refuse to enforce such contracts, with the result that minors who wish to enter into contracts will find no willing counterparty. Even if local governments did purport to guarantee debt and the guarantee was in some formal sense legally valid, enforcement would be difficult. First, there are a host of practical problems to suing a government body in China, especially in a local court whose personnel appointments and finances are in the hands of the defendant.123 Enterprises and individuals are considered to be inferior, not equal, to the government. Therefore, the stakes in litigation against a government body as guarantor will typically go far beyond the issues in a particular case. In such a legal and social climate, many investors are unlikely to feel comfortable about bringing suit against a government guarantor. Second, even if a creditor did manage to bring and win a lawsuit, enforcing the judgment would remain another hurdle.124 Although the literature is replete with warnings about local governments giving formal (and therefore improper) guarantees, and the above-cited regulations seem to confirm that the phenomenon is widespread, to our surprise we did not find any such guarantees in the offer prospectuses we reviewed. In other words, no local governments purported to guarantee LGFV bond issues in any legally enforceable way. Can investors take comfort in the informal promises of support? To an extent, that depends on how much importance a local government places on its reputation for paying its debts. A problem with reputational sanctions lies in the fact that government officials at any given time have a high incentive to borrow in order to construct projects that will raise local GDP, one of the important indices on which their performance is graded.125 By the time the debt becomes due, that official will long ago have left office (possibly having been promoted) according to the pattern of five-year terms for local officeholders. The successor official may well be unwilling to honor a nonbinding pledge made by and for the benefit of his predecessor.126 Despite the dubious security offered by government assurances, the number of bond issues safeguarded by such assurances is the same as the combined number of issues safeguarded by the other three methods discussed in this Article. 2. Land Use Rights as Security Several bond issues put up land use rights (LURs) belonging to the issuer as collateral.127 The legal basis for this form of security is article 34 of the Guarantee Law of China, which provides that land use rights that can be legally transferred by the debtor can be used as collateral.128 “Land use rights” refers to what are in effect long-term (a maximum term of forty, fifty, or seventy years, depending on the use) transferable leaseholds in urban land. Although what amounts to permanent fee simple ownership, and hence the reversion,129 must by law remain with the state, the law allows such land use rights to be granted for a one-time fee. The LURs then enter the stream of commerce and can be sold, mortgaged, leased, and otherwise transferred. Land subject to this regime is called “granted land.” Urban land from which LURs have not yet been carved out and sold is called “allocated land”: someone is using it, but the user received it through a process of bureaucratic allocation, not via a market transaction. Allocated LURs are not transferable and cannot be pledged as security—or more precisely, allocated LURs that are pledged as security will have to be turned into granted LURs, with the accompanying payment, before they can be taken in satisfaction of the debt by the secured party.130 In other words, the secured party will end up paying twice: once when it makes the loan, and once when it takes the security. In all the prospectuses we examined where LURs were offered as security, the LURs were in all but one case said to be obtained by grant and are hence unproblematic in that sense.131 Issuers typically also disclosed that they had already obtained the land use right certificate from the relevant agency and that the land was free from other encumbrances. Offering prospectuses frequently included a detailed schedule of the issuer’s land use rights used as security, noting the method by which the LURs were obtained and the date. How good is this security? Although this study does not in the main address the economic value of the security, there are some aspects of the valuation process that are cause for concern and can lead to excessive or questionable valuations.132 Although valuations must be made by credentialed assessors, countless assessment agencies exist in China, with a wide range of competency and levels of professionalism. LGFVs can freely choose whoever will give them the highest assessment. Not all offer prospectuses disclose how the assessor was selected or on what basis the valuation was made—for example, whether it was made with reference to market value in an arm’s length transaction. Although the value of urban land in China has risen greatly since the mid-1990s,133 there is no way of knowing where it will go in the future.134 In fact, the market entered a major slump in early 2014.135 However, LGFV issuers still have a tendency to overvalue the land use rights they have.136 Ideally, problems of overvaluation should be solved in the bond rating process. China has a credit rating industry,137 and bond issues are rated. There is reason to believe, however, that Chinese bond rating agencies may be afflicted by the same problems that afflicted U.S. bond rating agencies in the run-up to the financial crisis. For example, one prominent firm, Dagong, downgraded its rating of U.S. government debt to a single A with a negative outlook in an apparent fit of pique after being denied status as a Nationally Recognized Statistical Rating Organization by the Securities and Exchange Commission (SEC) in 2010.138 At the same time, it gave a higher rating to the bonds issued by the Loudi LGFV even though the LGFV valued the real estate collateralizing the debt at prices similar to those of Winnetka, Illinois, where local household income is about 100 times that of Loudi.139 The significance of valuation problems is compounded by a common feature in the collateralization agreement: a one-way ratchet. In a number of issues, the issuer has the obligation to add more LURs as collateral if the value of current collateral falls below some multiple of the outstanding debt obligation, and the right to demand the release of some collateral if the value of the collateral is above some multiple of the outstanding debt.140 This structure could operate as a one-way downward ratchet on the value of collateral on the basis of two plausible assumptions: first, that land values, even if accurately assessed, can fluctuate, and second, that the issuer simply may not possess sufficient unencumbered LURs to provide more collateral when the obligation to do so arises. As a result, if land values go up, creditors will lose some collateral, and it may be impossible to recover that collateral when land values go back down again. It is sometimes thought that despite valuation problems, LGFV bonds secured by land use rights at least have something backing them up.141 There may, however, be legal infirmities in the collateral itself, in addition to problems of economic value. First, there may be other encumbrances on the LURs used as collateral, and the claims of bondholders may be inferior.142 The offering prospectuses typically recite that the LURs are unencumbered, but bondholders have few if any remedies if this assertion turns out not to be true. The LGFV could be sued for false disclosure, but the disclosure problem will typically be discovered only at the point when the LGFV is already insolvent. Even if individual corporate officers can be successfully sued, their personal wealth will be inadequate to make up actual losses. Second, the LUR may not have been properly granted.143 In some cases, LURs purport to be granted by the governing entity of a development zone, called a Development Zone Commission. But such commissions have no legal power to grant LURs; that belongs to the local municipal government.144 In other cases, the land may still be in the intermediate stage prior to being formally granted, when existing residents are being cleared off and various payments must be made.145 It should be noted that these are generic problems with the use of LURs as collateral; we did not find specific evidence of these problems in any of the offering prospectuses we reviewed. A more serious problem with the grant lies in the possibility that no grant fee was ever paid. Consider the mathematics, taking the Loudi Prospectus as an example. The bond issue proposed to raise 1.2 billion yuan. The prospectus, dated January 2011, states that the value of the LURs securing the bonds is approximately 2.1 billion yuan. Most of the LURs were granted just one year earlier, in 2010. Even if the LGFV got the LURs at a 50%discount to the market value of just a year later—a generous assumption—it should have paid just over one billion yuan for them. This leads to an obvious question: if it had enough cash to pay for LURs, why did it need to raise cash through a bond issue? The obvious answer is that it did not have the cash and probably paid for the LURs with an IOU. But Chinese law requires full payment for LURs before LURs can be transferred, and a transfer is involved when a creditor realizes on collateral.146 Even where the local government is the sole shareholder of the LGFV, whether it can inject LURs as a capital contribution (taking payment in stock of the LGFV) is not clear. In short, there are strong reasons to doubt that the LURs securing LGFV bonds have been paid for in full. Without full payment, they cannot be transferred and are useless to creditors. And when the LGFV is insolvent—precisely the time when creditors need the collateral—it is of course in no position to pay for the LURs. Similarly, article 39 of the Urban Real Estate Administration Law requires that LUR grantees take certain steps to develop the land before it can be transferred. One rule is that the development be undertaken in accordance with the terms of the contract. Since the local government, the sole shareholder of the LGFV, is on the other side of the contract, one could reasonably expect the terms not to be onerous. But there is another rule not so easily avoided: the requirement that 25%of planned investment in housing have been made before the LUR can be transferred. As many of the LURs are stated to be for housing purposes, this requirement cannot be avoided. Failure to make planned investment can even make the original grant contract void.147 In an extreme case, validly granted LURs can be voided entirely when the fee is not paid according to the terms of the contract.148 A final consideration is that even if the legal and economic status of the security is sound, realizing on it may prove difficult if it requires displacing existing residents.149 They may not be willing to go peacefully. Given that defaults are likely to occur in a time of general economic downturn and therefore with a higher potential for social unrest, it is very unlikely that government authorities would place a higher priority on protecting the legal rights of distant bondholders than on preventing demonstrations and other manifestations of social unrest. There can be little doubt that many of these infirmities exist. Several were noted with disapproval in a document jointly issued in late 2012 by several central government regulators.150 Current disclosure practices vary. In the Shaoyang and Loudi offer prospectuses, the disclosure is quite comprehensive. It includes the exact location, the size, and expiration dates of the LURs, whether there are encumbrances, and the nature of those encumbrances.151 Other issuers disclose only basic data about the location and value of the LURs, without addressing more critical concerns, such as the way the LUR was obtained and information about encumbrances.152 In any case, it is impossible to tell from the prospectuses what agreements have been made about land development and whether the LGFVs are in a position to undertake necessary development of the LURs being offered as collateral. The point here is simply that what appears to be and may in fact be a valid, top-priority LUR in an accurately assessed parcel of real estate can still bear hidden dangers that call for constant monitoring. 3. Future Receivables as Security Several bond issues purport to be secured by receivables due to the LGFV issuer under Build-Transfer (BT) Agreements between the issuer and certain local governmental agencies (usually the agencies in charge of finance, transportation, or construction) in which the issuer agrees to build a project and the agency agrees to purchase the project after it has been built.153 Interestingly, this structure seems to be legally robust, even though in substance it mimics a forbidden structure. As discussed above, if a local government wishes to finance a construction project, it may not do so by issuing bonds directly. Nor may it guarantee bonds issued by its wholly owned LGFV.154 Obviously, an LGFV may issue bonds on its own, but the trick is to have the local government give a legally binding commitment of its own to bondholders in the face of the above prohibitions. The pledge of receivables accomplishes this goal neatly. First, the issue of bonds by the LGFV is legally unproblematic, precisely because the LGFV is not a government entity. Second, the contractual commitment by the local government to purchase the project upon completion under the BT contract is also unproblematic.155 Governments and government agencies can commit themselves contractually to suppliers under ordinary principles of government procurement. Third, the pledge of the receivable, which gives bondholders priority over other creditors of the LGFV, is permitted under China’s Guarantee Law. As a result, bondholders get exactly what they would get if the local government issued the bonds directly or provided a guarantee for LGFV bonds (assuming either was legal): a legally valid government promise to pay money sufficient to repay the bonds and to which they have top-priority access. Needless to say, whether the local government will actually have the funds to make good on its promise to pay is an important issue, but the focus of this Article is on the legal status of the various safeguards for repayment available to bondholders. 4. Third-Party Business Entity as Guarantor Another commonly used safeguard in LGFV bonds is a guarantee from a third party business entity. The third party company can be either an investment holding company or a company specializing in the guarantee business. For example, in the Baoji LGFV bond, China National Investment and Guaranty Co., Ltd. acted as the guarantor for the bonds, providing a “full, unconditional, and irrevocable joint responsibility guarantee.”156 In the Leshan LGFV bond, the two classes of bonds were guaranteed respectively by Chongqing Three Gorges Guarantee Group Co., Ltd. and Beijing Capital Investment and Guarantee Co., Ltd. Respectively, and the guarantors provided a “full, unconditional, and irrevocable joint responsibility guarantee.”157 In a bond issuance by Mega Advance Investments Limited (“Mega”), a Beijing LGFV, the guarantor company promised to “irrevocably and unconditionally guarantee the due and punctual payment of the principal of and interest on, and all other amounts payable under, the Notes when and as the same shall become due and payable.”158 Furthermore, the guarantor agreed that its obligations under the guarantees would be “as if it were principal obligor and not merely surety.”159 Under Chinese law, a guarantee provided by a third party that can be independently held liable for the debt is a form of security under the Guarantee Law.160 Specifically here, the bonds are secured by the guaranteeing entity’s operating assets that can be legally transferred or disposed of in order to pay off the debt of the LGFV under the guarantee agreement signed between LGFV and the guarantee company. Accordingly, the disclosure in the prospectus involves material information about the guarantor company instead of the LGFV issuer, with the guarantor company’s income-generating operations and assets constituting the bulk of the disclosure, as in the case of the Mega bond issuance. A common type of third-party guarantor is a specialized guarantee company; as of the end of 2011, China had 8,402 of them, with assets purporting to total 931 billion yuan.161 Guarantees from such companies can, however, be very risky. Guarantee companies were originally created with the backing of local governments to help small and medium enterprises get bank loans.162 Banks normally prefer to lend to enterprises that have hard assets such as land or else have implicit state backing. Guarantee companies charge borrowers a fee and then provide a binding guarantee to the lending bank. In effect, then, the guarantee company offers default insurance, with the premium paid by the borrower. This means that the lender has to consider a number of things about the insurer. How many policies does it have outstanding, and how do its potential obligations match up against its assets? Even if that information is fully disclosed—and often it is not—the lender still needs to know how many more policies it will issue after the loan is made. Most importantly, one can have confidence in one’s insurer only if one is confident that the insurer’s risks on its various policies are not positively correlated with each other.163 In other words, if it is insuring one party against a default that will happen if the price of oil goes up, it should be insuring another party against a default in the same amount that will happen if the price of oil goes down. You do not want a flood insurer that does business only in New Orleans. Whether Chinese guarantee companies have in fact spread their risks properly is far from clear, and in any case not evident from the disclosure provided in bond prospectuses. Thus, it is very possible that they are vulnerable to broad economic shocks that hit all or most borrowers at once, such as a downturn in the property market or a significant slowing of growth. The risks of guarantee companies were vividly brought home in the 2012 collapse of the Zhongdan Investment Credit Guarantee Company.164 In January 2012, banks reacted to rumors of a liquidity crunch at Zhongdan by calling loans. The debtor companies in turn sought help from Zhongdan, but the volume of loans overwhelmed the company. Zhongdan had guaranteed 3.3 billion yuan in loans, but had only 5 million yuan immediately liquid and accessible when the calls for funds came. In the case of Zhongdan, there appears to have been malfeasance afoot. But even without malfeasance, the lack of real substance to its guarantees is likely duplicated at many other guarantee companies. An important aspect of due diligence for bondholders is to examine carefully the assets and liabilities of the guarantor companies. The assets of guarantor companies may be misstated or subject to multiple existing encumbrances.165 When the bulk of a guarantor company’s assets are its subsidiaries, it is critical to know about the indebtedness of those subsidiaries, ranging from convertible bonds to bank loans.166 Even if the guarantor company had no encumbrances or indebtedness, its asset structure could be problematic because of a high ratio of non-liquid assets to liquid assets such as cash. For example, in the Mega LGFV bond issue, the guarantee company’s portfolio assets include natural gas distribution and transmission assets, brewery operations, sewage and water treatment facilities and concessions, and expressways and toll roads. For the guarantee company in the Baoji LGFV bond issuance, China National Investment & Guarantee Co., Ltd., the ratio of non-liquid assets to liquid assets is approximately four to five.167 Its nonliquid assets include “sellable financial assets” and fixed assets.168 It is by definition difficult to cash out nonliquid assets when the debt obligation is due, and that difficulty undermines the value of the guarantee as security. III. The Debt Swap Program In the spring of 2015, the government announced a program under which local governments could issue bonds. Under the plan, known as a “debt swap,” the Ministry of Finance would, in separate tranches of quotas, authorize local governments at the level of province and separately planned city169 to issue bonds, the revenues from which would be used to pay off local government debt. The purpose of the plan is to allow local governments to save on interest costs by replacing short-term, high-cost, high-risk local government debt (including LGFV debt that is classed as local government debt) with longer-term, lower-risk and therefore lower-cost government bond debt.170 The first tranche of the swap was one trillion yuan and was to cover obligations (probably only Type I obligations) of local governments identified in the 2013 NAO report that were due in 2015.171 According to the Ministry of Finance, the one trillion yuan quota covered 53.8 %of the 1.86 trillion yuan in debts that came due in 2015, and a proportional allocation system meant that each province received a quota representing 53.8%of its direct indebtedness.172 The second tranche, also of one trillion yuan, was announced in June 2015.173 That additional quota was also intended to cover debts coming due in 2015. Since each tranche amounts to over 50% of the debt it is to cover, together they would seem to cover more than the debt in question. In any case, the quota for 2015 was ultimately raised to 3.2 billion yuan, of which 3.18 trillion yuan in swapped bonds was used by mid-December.174 Estimates of the amount needed to absorb remaining LGFV debt range from 11 to 25 trillion yuan.175 Although the Ministry of Finance announced that revenues from the swapped bonds could not be used to pay off debts that should be paid from the self-generated revenues of enterprises and institutions, there is wiggle room here: as discussed above, since the distinctions used by the NAO to determine precisely which debts are this kind of debt are not strict legal ones, it is not going to be self-evident which local debts “should” be payable solely from self-generated revenues of enterprises and institutions. The swap program is part of a larger central government effort to clean up LGFV debt and reduce or eliminate its use in the future. It is not, however, going to make the issues surrounding LGFV debt discussed in this Article disappear. First, although we cannot know exactly the criteria used by the NAO is classifying debt associated with local governments, it seems beyond doubt that a considerable amount of LGFV debt is Type II or Type III debt, or possibly not even counted as local government debt at all, whereas the swap program seems to cover only Type I debt.176 Second, since the very category of “LGFV” is not a legal or precisely defined one, the only way to effectively ban LGFV debt is to ban all debt in corporate entities owned by local governments. But the government has not gone this far. If it is ever more convenient for local governments to borrow through establishing wholly owned companies instead of directly in their own name, we will see LGFVs or their functional equivalent. Conclusion The contribution of LGFVs to China’s economy will probably always remain controversial. On the one hand, they were a key element in the government’s post-financial-crisis stimulus plan, which relied heavily on bank lending and relatively little on new government spending.177 They allowed local governments to get capital for infrastructure construction, and enabled China’s GDP to continue to grow. On the other hand, the only reason local governments needed LGFVs to raise capital was that the central government prohibited them from borrowing themselves—a policy that could have been changed whenever necessary, and that is now being relaxed (although only down to the level of the province and separately planned cities).178 It is not clear whether the infrastructure financed through LGFVs has in fact made China better off, since there is reason to believe that at least some of it is unlikely to produce an economic return over its cost. GDP grows whenever there is more economic activity than before, but not all economic activity—for example, the classic case of digging a hole and filling it up again—represents a good investment and leads to growth. It can retard growth if it just produces bad debt that must be ultimately borne by households, thus in turn leading to lower aggregate demand. Because it seems almost mathematically certain that at least some nontrivial number of LGFV bonds will not be eligible for the swap program and will default, it is necessary to think about the quality of the security they offer. It is difficult to generalize about bond repayment safeguards because of the variety of practices we have canvassed. As we have discussed, a government guarantee is not a valid security in Chinese law, either in theory or in practice, no matter what the bond prospectuses might say. On the other hand, the land use right obtained by grant, as well as the guarantee provided by a guarantee company, constitutes a valid security interest, although its economic value requires separate analysis. This Article’s analysis of the legal structure of LGFV bonds and local government debt more generally also has important lessons for understanding the Chinese legal system as a whole. A key part of the story of local government indebtedness is the inadequacy of legal categories to understand what is going on. Legal terms of art such as “debt,” “liability,” and “guarantee” are widely used, and yet it becomes quickly apparent that they are not being used in a legal sense. This means that data classified using these terms cannot be readily compared with data classified using these terms in other countries. In particular, it is clear that accounts of the level of China’s local government debt that include Type II and Type III debt simply cannot be accepted at face value. This is not debt that local governments have any actual obligation to pay. To be sure, they may wish to pay creditors voluntarily and may from time to time do so, but soft obligations of this nature have never been considered debt in China or in other countries, and it is misleading to label them as such. Creditors who have tried to force local governments to make good on their guarantees have uniformly failed. At most they have received half of what they sought, under a legal theory that a recent Supreme People’s Court case has made more difficult to invoke.179 The argument that local governments have some politically (even if not legally) enforceable obligation to pay on their guarantees does not seem supported by the evidence. In addition, we see here what can be observed in other areas of Chinese law as well: the tendency of some players and institutions in the system to recognize as legally valid acts that at the same time the system deplores and even prohibits.180 The system could attempt to discourage unlawful borrowing and guarantee-issuing by local governments by simply making such borrowing and guarantees legally invalid and unenforceable—in effect, invisible to the legal system. And in a sense it tries to do that, but not unambiguously. Its approach is instead to prohibit the acts and make them punishable, implicitly conceding that the act could be effective. For the state to prohibit and punish murder makes sense, because the murderer does not need the state’s help to make the murder effective. But to prohibit and punish the making of guarantees is different, because without the state’s cooperation the guarantee could not be effective and could do no damage in the first place. On the one hand, the courts are invalidating local government guarantees. But at the same time, instead of warning creditors not to accept such guarantees, the central government is warning local governments not to make them, and the State Council and the NAO categorize them as valid and enforceable when they count local government debt obligations. The broader lesson here is that acts that are prohibited under Chinese law cannot be assumed to be invalid even if they require state cooperation to be valid; they may or may not be, and a detailed analysis of the specific act in question, and related rules and practices, is required. There is no question that LGFV defaults will pose a challenge for China’s political system. The pressure for bailouts will be intense, but the cost of providing them, in terms of both actual money and added moral hazard, will be huge. Yet defaults will also pose a serious challenge to China’s legal system. If defaulting LGFVs are not bailed out, then the cost of default must land somewhere other than on the entity—probably the central government, perhaps the local government—that would have funded the bailout. There are legal rules about where it ought to land. Will they be effective? A great natural experiment is in the offing. Appendix Bond Prospectuses Group I Baoji Prospectus: 2011 Nian Baoji Shi Touzi (Jituan) Youxian Gongsi Qiye Zhaiquan Muji Shuomingshu (2011年宝鸡市投资(集团)有限公司企业债券募集说明书), March 2011, available athttp://bit.ly/ztl7j3 Datong Prospectus: 2011 Nian Datong Shi Jingji Jianshe Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年大同市经济建设投资有限责任公司公司债券募集说明书), June 2011, available athttp://bit.ly/zG2OYs Hebei Bohai Prospectus: 2011 Nian Hebei Bohai Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年河北渤海投资有限公司公司债券募集说明书), May 2011, available athttp://bit.ly/zk0jQ2 Hefei Prospectus: 2009 Nian Hefei Shi Jianshe Touzi Konggu (Jituan) Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年合肥市建设投资控股(集团)有限公司公司债券募集说明书), July 2009, available athttp://bit.ly/1Mhc4p1 Jiangyin Prospectus: 2009 Nian Jiangyin Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年江阴城市建设投资有限公司公司债券募集说明书), September 2009, available athttp://bit.ly/1JSEmWJ Jilin Prospectus: 2011 Nian Jilin Shi Chengshi Jianshe Konggu Jituan Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiwu Muji Shuomingshu (2011年吉林市城市建设控股集团有限公司市政项目建设债券募集说明书), March 2011, available athttp://bit.ly/yAetfO Leshan Prospectus: 2011 Nian Leshan Shi Guoyou Zichan Jingying Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年乐山市国有资产经营有限公司公司债券募集说明书), March 2011, available athttp://bit.ly/zixZYR Lishui Prospectus: 2011 Nian Lishui Shi Chengshi Jianshe Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2011年丽水市城市建设投资有限责任公司公司债券募集说明书), October 2011, available athttp://bit.ly/xzxbaZ Loudi Prospectus: 2011 Nian Loudi Shi Chengshi Jianshe Touzi Jituan Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年娄底市城市建设投资集团有限公司市政项目建设债券募集说明书), January 2011, available athttp://bit.ly/wNBdER Mega Prospectus: Offering Memorandum, Mega Advance Investments Limited, Unconditionally and Irrevocably guaranteed by Beijing Enterprises Holdings Limited, March 2011 (on file with first author) Nanjing Prospectus: 2009 Nian Nanjing Shi Chengshi Jianshe Touzi Konggu (Jituan) Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年南京市城市建设投资控股(集团)有限责任公司公司债券募集说明书), available athttp://bit.ly/1KYo9xL Panjin Prospectus: 2011 Nian Panjin Jianshe Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Faxing Gonggao (2011年盘锦建设投资有限责任公司公司债券发行公告), February 2011, available athttp://bit.ly/zyA1P6 Shaoyang Prospectus: 2011 Nian Shaoyang Shi Chengshi Jianshe Touzi Jingying Jituan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2010年邵阳市城市建设投资经营集团有限公司公司债券募集说明书), August 2010, available athttp://bit.ly/y33PrX Tianjin Binhai Prospectus: 2011 Nian Tianjin Binhai Xin Qqu Jianshe Touzi Jituan Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年天津滨海新区建设投资集团有限公司市政项目建设债券募集说明书), November 2011, available athttp://bit.ly/1WecbX1 Xianyang Prospectus: 2009 Nian Xianyang Shi Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年咸阳市城市建设投资有限公司公司债券募集说明书), September 2009, available athttp://bit.ly/1KcQb5H Yichang Prospectus: 2011 Nian Yichang Shi Chengshi Jianshe Touzi Kaifa Youxian Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年宜昌市城市建设投资开发有限公司市政项目建设债券募集说明书), November 2011, available athttp://bit.ly/AlIUxA Yichun Prospectus: 2009 Nian Yichun Shi Chengshi Jianshe Touzi Kaifa Youxian Zeren Gongsi Qiye Zhaiquan Muji Shuomingshu (2009年伊春市城市建设投资开发有限责任公司企业债券募集说明书), September 2009, available athttp://bit.ly/1f4JkT9 Yongzhou Prospectus: 2011 Nian Yongzhou Shi Chengshi Jianshe Touzi Kaifa Gongsi Shizheng Xiangmu Jianshe Zhaiquan Muji Shuomingshu (2011年永州市城市建设投资开发公司市政项目建设债券募集说明书), December 2011, available athttp://bit.ly/womppr Zhenjiang Prospectus: 2010 Nian Zhenjiang Shi Jiaotong Touzi Jianshe Fazhan Gongsi Qiye Zhaiquan Muji Shuomingshu (2010年镇江市交通投资建设发展公司企业债募集说明书), December 2010, available athttp://bit.ly/zc42Eo Zhuzhou Prospectus: 2009 Nian Zhuzhou Shi Chengshi Jianshe Touzi Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2009年株洲市城市建设投资经营有限公司公司债券募集说明书), available athttp://bit.ly/1L3di8B Group II Baoshan Prospectus: 2014 Nian Baoshan Shi Guoyou Zichan Jingying Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年保山市国有资产经营有限责任公司公司债券募集说明书), May 2014, available athttp://bit.ly/1HD2ZjO Beichen Tianjin Prospectus: 2014 Nian Tianjin Shi Beichen Qu Jianshe Kaifa Gongsi Qiye Zhaiquan Muji Shuomingshu (2014年天津市北辰区建设开发公司企业债券募集说明书), April 2014, available athttp://bit.ly/1L1wdh7 Changzhou Prospectus: 2014 Nian Changzhou Gonggong Zhufang Jianshe Touzi Kaifa Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年常州公共住房建设投资发展有限公司公司债券募集说明书), available athttp://bit.ly/1gZ1IyQ Chengfa Prospectus: 2014 Nian Chengfa Touzi Jituan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年城发投资集团有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1KfcAj1 Dongtai Prospectus: 2014 Nian Di’erqi Dongtai Shi Chengshi Jianshe Touzi Fazhan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年第二期东台市城市建设投资发展有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1HD4Jd6 Ezhou Prospectus: 2014 Nian Diyiqi Ezhou Shi Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年第一期鄂州市城市建设投资有限公司公司债券募集说明书), May 2014, available athttp://bit.ly/1J82jun Hami Prospectus: 2014 Nian Xinjiang Weiwu’er Zizhiqu Hami Diqu Guoyou Zichan Touzi Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014 年新疆维吾尔自治区哈密地区国有资产投资经营有限公司公司债券募集说明书), July 2014, available at http://bit.ly/1PhOeJC Kunming Prospectus: 2014 Nian Kunming Dianchi Touzi Youxian Zeren Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年昆明滇池投资有限责任公司公司债券募集说明书), June 2014, available athttp://bit.ly/1IVbasI Kunshan Prospectus: 2014 Nian Kunshan Jiaotong Fazhan Konggu Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年昆山交通发展控股有限公司公司债券募集说明书), May 2014, available at http://bit.ly/1gv6qDw Laizhou Prospectus: 2014 Nian Laizhou Shi Guoyou Zichan Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年莱州市国有资产经营有限公司公司债券募集说明书), July 2014, available athttp://bit.ly/1L1zwFf Lin’an Prospectus: 2014 Nian Lin’an Shi Chengjian Fazhan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年临安市城建发展有限公司公司债券募集说明书), August 2014, available athttp://bit.ly/1P2AHVx Lu’an Prospectus: 2014 Nian Di’erqi Lu’an Chengshi Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2013年第二期六安城市建设投资有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1PhPLiM Mudanjiang Prospectus: 2014 Nian Mudanjiang Shi Guoyou Zichan Touzi Konggu Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年牡丹江市国有资产投资控股有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1J0OnjT Ningxiang Prospectus: 2014 Nian Ningxiang Jingji Jishu Kaifaqu Jianshe Touzi Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年宁乡经济技术开发区建设投资有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1DJlRmo Pizhou Prospectus: 2014 Nian Pizhou Shi Runcheng Zichan Jingying Jituan Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年邳州市润城资产经营集团有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1DEMg56 Qidong Prospectus: 2014 Nian Di’erqi Qidong Shi Chengshi Touzi Jingying Zhongxin Qiye Zhaiquan Muji Shuomingshu (2014年第二期启东市城市投资经营中心企业债券募集说明书), April 2014, available athttp://bit.ly/1WeoW3S Shenmu Prospectus: 2014 Nian Shenmu Xian Guoyou Zichan Yunying Gongsi Qiye Zhaiquan Muji Shuomingshu (2014年神木县国有资产运营公司企业债券募集说明书), June 2014, available athttp://bit.ly/1PhRmp0 Shenyang Prospectus: 2014 Nian Shenyang Nongye Gaoxinqu Guoyou Zichan Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年沈阳农业高新区国有资产经营有限公司公司债券募集说明书), April 2014, available athttp://bit.ly/1DENjlw Taiyuan Prospectus: 2014 Nian Taiyuan Jingji Jishu Kaifa Zonggongsi Qiye Zhaiquan Muji Shuomingshu (2014年太原经济技术开发总公司企业债券募集说明书), April 2014, available athttp://bit.ly/1L1CzNG Xinjiang Prospectus: 2014 Nian Xinjiang Shengchan Jianshe Bingtuan Nong Shi’er Shi Guoyou Zichan Jingying Youxian Gongsi Gongsi Zhaiquan Muji Shuomingshu (2014年新疆生产建设兵团农十二师国有资产经营有限公司公司债券募集说明书), May 2014, available athttp://bit.ly/1DEOgu0 The views expressed in this Article are solely those of the authors and do not reflect the views or position of Memjet Technology, Ltd. The authors wish to thank Michael Forsythe and Henry Sanderson for the Bloomberg article that inspired this Article and for helpful advice on finding the kind of prospectuses we were looking for. We also wish to thank Andrew Collier, Nicholas Howson, Curtis Milhaupt, and Barry Naughton for their thoughtful comments. All English translations in this Article are by Donald Clarke unless otherwise indicated. Footnotes 1 In this Article, “local government” refers to a government one or more administrative levels below the central government—for example, a city, county, or provincial government. The term is used interchangeably with “subnational government” or “SNG.” 2 On local government debt in China generally, see Xun Wu, Policy Brief, China’s Growing Local Government Debt Levels, Mass. Inst. of Tech.(MIT) Ctr. for Fin. & Pol’y (Jan. 2016), http://bit.ly/xunwu. 3 See, e.g., China Cities Value Land at Winnetka Prices with Bonds Seen Toxic, Bloomberg News (July 13, 2011), http://bloom.bg/q0AV7P. 4 The prospectuses are listed in the Appendix. For ease of reference, the prospectus for any particular issue shall be referred to in this Article using the abbreviation provided in the Appendix. 5 We use this term deliberately in order to avoid the more narrow and technical connotations of “collateral.” Whether the proper steps to achieve collateralization have actually been taken is sometimes not clear. 6 See Du Tao (杜涛), Caizheng Bu Pifu 3 Wan Yi Edu Zhihuan Cunliang Zhaiwu 1 Wan Yi Yi Xiapi Gesheng (财政部批复3万亿额度置换存量债务 1万亿已下批各省) [Ministry of Finance Approves 3 Trillion Quota in Bonds-for-Debt Swap; 1 Trillion Already Allocated to Provinces], Jingji Guancha Wang (经济观察网) [Econ. Observer Network] (Mar. 6, 2015), http://www.eeo.com.cn/2015/0306/273066.shtml; Defusing a Bomb, Economist (Mar. 11, 2015), http://econ.st/1HIAuDX. For a more detailed discussion, see infra Part III. 7 See infra Part I.A.2.b.ii. 8 We will frequently provide statistics for the late 2000s because this is when LGFV activity took off. 9 See Wei Cui, Fiscal Federalism in Chinese Taxation, 3 World Tax J. 455, 456 (2011). The figure for 2009 is 20%. See Christine P. Wong, Paying for Urbanization in China: Challenges of Municipal Finance in the 21st Century, inFinancing Metropolitan Governments in Developing Countries 273, 281 (Roy W. Bahl et al. eds., 2012). 10 See Cui, supra note 9, at 456. 11 See id. at 457. Similar numbers are cited in the more recent Houze Song, Why China Can Avoid a Local Fiscal Crisis, Paulson Inst. (May 4, 2015), http://bit.ly/1Q24q28. Looked at another way, local government revenue is 20% of GDP, while local government expenditures are 28% of GDP. See China’s Regional Financial Crisis, The Banker, Feb. 1, 2016. 12 See Wong, supra note 9, at 281–90; Cui, supra note 9, at 456–57. 13 Zhonghua Renmin Gongheguo Shuishou Zhengshou Guanli Fa (中华人民共和国税收征收管理法) [Law of the People’s Republic of China Concerning the Administration of Tax Collection] (promulgated by the Standing Comm. Nat’l People’s Cong., Sept. 4, 1992, effective Jan. 1, 1993), https://perma.cc/FV2J-5PTS. 14 Guowuyuan Guanyu Shixing Fenshuizhi Caizheng Guanli Tizhi de Jueding (国务院关于实行分税制财政管理体制的决定) [Decision of the State Council Regarding the Implementation of the Tax Sharing Fiscal Administration System] (promulgated by the St. Council, Dec. 15, 1993, effective Dec. 15, 1993), https://perma.cc/8WZD-XLBD. 15 See generally Cui, supra note 9, at 463–65. 16 Id. at 457. 17 People’s Congresses are legislative bodies that exist at various levels of government. They appoint the leadership of, and are formally higher in rank than, the so-called “People’s Governments” at the same administrative level, which can be roughly analogized to an executive branch of government. 18 See Cui, supra note 9, at 469 & n.85; Lou Jiwei (楼继伟), Guowuyuan Guanyu 2014 Nian Zhongyang Juesuan de Baogao (国务院关于2014年中央决算的报告) [Report of the State Council on the Implementation of Budget and Final Accounts of Central and Local Governments], Zhonghua Renmin Gongheguo Caizheng Bu (中华人民共和国财政部) [Ministry of Fin. of the People’s Republic of China] (June 28, 2015), https://perma.cc/YJ5C-5BS6. 19 A legal system does this kind of thing when it enforces against a principal a contract entered into by an agent that exceeds the scope of the agent’s authority. The contract is valid as between the principal and the third party, but the principal can sue the agent for any losses. 20 See, e.g., Ye Peirong (叶裴荣), Woguo Difang Zhengfu Rongzi Tixi Yanjiu (我国地方政府融资体系研究) [A Study of The Local Government Financing System in China], 1 Xinan Jinrong (西南金融) [Sw. Fin.] 39 (2012) (citing Daikuan Tongze (贷款通则) [Lending Principles] (promulgated by the People’s Bank of China [PBOC], June 28, 1996, effective June 28, 1996), http://www.law-lib.com/law/law_view.asp?id=15263); China Said To Warn Banks as Local Officials Seek New Loans, Bloomberg News (July 13, 2011), http://bloom.bg/1VppWAI; Michael F. Martin, Cong. Research Serv., R42380, China’s Banking System: Issues for Congress (2012), http://www.fas.org/sgp/crs/row/R42380.pdf. 21 See Lending Principles, June 28, 1996, art. 17. 22 See, e.g., Ye, supra note 20, at 39. The author is described as an employee at the Chengdu branch of the China Development Bank, a wholly state-owned “policy bank,” and presumably speaks with some knowledge of actual practice within the banking industry. The same reasoning is cited by Liang Weihong (梁卫洪), Difang Ren Da Huandai Chengnuo de Falü Wenti Yanjiu (地方人大还贷承诺的法律问题研究) [Research on the Legal Issues in Local People Congress’s Promise to Repay Loans] (Apr. 20, 2005) (unpublished LL.M. thesis, Jinan (暨南) University) (on file with first author), as the reason for setting up LGFVs as loan recipients. 23 This was true under article 28 of the old Budget Law and remains true under article 35 of the recently revised Budget Law of 2015. See Yusuan Fa (预算法) [Budget Law] (promulgated by the Nat’l People’s Cong., Mar. 22, 1994, effective Jan. 1, 1995), https://perma.cc/VR3D-UD3F; Yusuan Fa (预算法) [Budget Law] (promulgated by the Nat’l People’s Cong., Mar. 22, 1994, amended Aug. 31, 2014, effective Jan. 1, 2015), https://perma.cc/6QT9-GEH5. 24 In reading the relevant literature, it is critical to distinguish references to direct SNG borrowing and SNG borrowing through LGFVs. The latter is often carelessly referred to as SNG debt. 25 SeeZhonghua Renmin Gongheguo Shenji Shu (中华人民共和国审计署) [National Audit Office of the People’s Republic of China], Quanguo Difang Zhengfuxing Zhaiwu Shenji Jieguo (全国地方政府性债务审计结果) [Results of a Nationwide Audit of Government Debt] tbl.3 (2011), https://perma.cc/9HFP-RZ96 [hereinafter NAO 2011]. 26 See id. sec. 3.1. 27 See, e.g., Caizheng Bu Guanyu Guifan Difang Geji Zhengfu Bumen Juzhai he Danbao Chengnuo Xingwei de Tongzhi (财政部关于规范地方各级政府部门举债和担保承诺行为的通知) [Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels] (promulgated by the Ministry of Fin., Sept. 2, 2010, effective Sept. 2, 2010), https://perma.cc/85RB-PMWM. This document cannot be interpreted as a warning to creditors that their rights will not be upheld, since it is labeled “not for public circulation” and was distributed only to potential debtors (local governments) and not to potential creditors (the banking system). 28 Note that we are not speaking here of creditors’ rights against financing vehicles established by local governments, which are the focus of this Article. 29 See our discussion of the Bank of China case infra note 120 and accompanying text. 30 See Budget Law, Mar. 2, 1994, art. 28 (“The 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. The local governments shall not issue local government bonds, except as otherwise prescribed by laws or the State Council.”). 31 See Caizheng Bu Guanyu Zhuanqing Chuli Chengdu Shi Qingyang Qu Zhengfu Weigui Fafang Difang Zhengfu Zhaiquan Wenti de Han (财政部关于转请处理成都市青羊区政府违规发放地方政府债券问题的函) [Letter of the Ministry of Finance Regarding the Problem of Handling the Illegal Issuance of Local Government Bonds by the Government of Qingyang District, Chengdu, on Its Illegal Issuance of Municipal Bonds] (promulgated by the Ministry of Fin., May 23, 2002, effective May 23, 2002), https://hk.lexiscn.com/law/law-chinese-1-33364.html; Li Jiuling (李九领), Woguo Difang Zhengfu Faxing Zhaiquan de Nengli he Suzhi Fenxi (我国地方政府发行债券的能力和素质分析) [An Analysis of the Capability and Quality of Issuance of Bonds by China’s Local Governments], Zhonghua Kuaiji Wangxiao (中华会计网校) [China Online Acct. Sch.] (Apr. 6, 2007), http://bit.ly/1X3cioL; Shi Hongxiu (时红秀), Difang Zhengfu Zhaiwu Wenti Zai Taolun (地方政府债务出路问题再讨论) [Another Discussion of the Problem of Local Government Debt], 3 Yinhang Jia (银行家) [Banker] 10, 11 (2010). 32 See Guanyu Yinfa “2011 Nian Difang Zhengfu Zixing Fazhai Shidian Banfa” de Tongzhi (关于印发“2011年地方政府自行发债试点办法”的通知) [Notice of the Ministry of Finance on Issuing the “Measures for the 2011 Pilot Program of Bond Issuance” by Local Governments] (promulgated by the Ministry of Fin., Oct. 17, 2011, effective Oct. 17, 2011), https://perma.cc/6S52-HC8C. The same four governments were given permission for another bond issue in 2012. See Guanyu Yinfa “2012 Nian Difang Zhengfu Zixing Fazhai Shidian Banfa” de Tongzhi (关于印发“2012年地方政府自行发债试点办法”的通知) [Notice of the Ministry of Finance on Issuing the “Measures for the 2012 Pilot Program of Bond Issuance” by Local Governments] (promulgated by the Ministry of Fin., May 8, 2012), https://perma.cc/9Z4K-KDSR. 33 See generally Shanghai Shoufa 71 Yi Difang Zhengfu Zhaiquan (上海首发71亿地方政府债券) [Shanghai Issues First Government Bonds Worth 7.1 Billion], Xin Jing Bao (新京报) [Beijing News] B05 (Nov. 16, 2011), http://bit.ly/zsMLRx; see also Guowuyuan Pizhun Zhejiang Deng Si Di Shidian Zixing Fazhai (国务院批准浙江等4地试点自行发债) [State Council Approves Four Areas Including Zhejiang for Trial Bond Issuance], Hexun Xinwen (和讯新闻) [Hexun News] (Oct. 20, 2011), http://news.hexun.com/2011/sdzw/ (providing more comprehensive discussions about the trial local government bond issuance in China). 34 See Yu Jian (于舰), Difang Zixing Fazhai Shidian Kuorong, Su Lu Liang Sheng Bei Naru (地方自行发债试点扩容 苏鲁两省被纳入) [Pilot Program of Bond Issuance by Local Governments Expands and Include the Provinces of Jiangsu and Shandong], Diyi Caijing Ribao (第一财经日报) [First Fin. News] (July 5, 2013), http://www.yicai.com/news/2013/07/2834475.html. 35 See Quanguo Renmin Daibiao Dahui Changwu Weiyuanhui Guanyu Xiugai “Zhonghua Renmin Gongheguo Yusuan Fa” de Jueding (全国人民代表大会常务委员会关于修改“中华人民共和国预算法”的决定) [Decision of the Standing Committee of the National People’s Congress of the People’s Republic of China on the Revision of the “Budget Law of the People’s Republic of China”] (promulgated by the Standing Comm. Nat’l People’s Cong., Aug. 31, 2014, effective Aug. 31, 2014) item 30, https://perma.cc/D2YP-LXU5 (amending article 28 of the 1994 Budget Law). 36 See generally Donald Clarke, Budget Law Revisions Tighten, Not Loosen, Central Government’s Control Over Local Government Bond Issues, Chinese Law Prof Blog (Oct. 3, 2014), http://bit.ly/chineselawblog-bonds. 37 See Wong, supra note 9, at 300ff.; Tu Dejun (涂德君), Woguo Chengtouzhai Shichang Fazhan de Sange Jieduan (我国城投债市场发展的三个阶段) [The Three Phases of Development of the Chinese Market for Local Government Bonds], 2 Zhongguo Zhaiquan (中国债券) [Chinese Bonds] 14, 15 (2010). 38 The faith-based quality of this kind of financing is exemplified by the fact that China at the time still had no law on corporate organization that would have governed such a company. 39 See Wong, supra note 9, at 300. 40 As will be discussed below, these guarantees have no legal validity. 41 See Tu, supra note 37, at 15. 42 See generally Victor Shih, Local Government Debt: Big Rock-Candy Mountain, 14 China Econ. Q. 26, 26–32 (2010). 43 Zhongguo Renmin Yinhang, Zhongguo Yinhangye Jiandu Guanli Weiyuanhui Guanyu Jinyibu Jiaqiang Xindai Jiegou Tiaozheng Cujin Guomin Jingji Pingwen Jiaokuai Fazhan de Zhidao Yijian (中国人民银行、中国银行业监督管理委员会关于进一步加强信贷结构调整促进国民经济平稳较快发展的指导意见) [Guiding Opinions of the People’s Bank of China and China Banking Regulatory Commission on Further Adjusting the Credit Structure to Promote the Rapid yet Steady Development of the National Economy] (promulgated by the People’s Bank of China, China Banking Regulatory Comm’n, Mar. 18, 2009, effective Mar. 18, 2009), https://perma.cc/7AB3-WKNL. 44 See Ye, supra note 20, at 38. 45 See Wong, supra note 9, at 301 (citing Hua Qiang (华强), Gai Gei Difang Rongzi Pingtai Daishang Jinguzhou (该给地方融资平台戴上紧箍咒) [A Restraint Collar Should Be Put on Local Government Financing Vehicles], Fenghuang Wang (凤凰网) [Phoenix Network] (June 6, 2010), http://bit.ly/1RNJz3Y). Another source states that borrowing in 2009 was 4 trillion yuan, leading to total indebtedness of 7.16 trillion yuan as of the end of the year. See Ye, supra note 20, at 38. The difference in the numbers may be accounted for by different ways of categorizing local government borrowing; the lower figure may exclude certain kinds of borrowing not directly connected to LGFVs. See also China’s Official Local Debt Could Exceed CNY 14 Trl, Some Hidden Debt Excluded, Caijing (Oct. 28, 2013), http://english.caijing.com.cn/2013-10-28/113485388.html. 46 See Shi, supra note 31, at 10. 47 See NAO 2011, supra note 25, tbl.3. 48 SeeZhonghua Renmin Gongheguo Shenji Shu (中华人民共和国审计署) [National Audit Office of the People’s Republic of China], Quanguo Difang Zhengfuxing Zhaiwu Shenji Jieguo (全国地方政府性债务审计结果) [Results of Nationwide Audit of Government Debt] 6–7 tbl.3 (2013), https://perma.cc/53NW-MCLQ [hereinafter NAO 2013]. The amount for local government debt includes amounts for which local governments are not legally obligated; we discuss this issue in detail in Part I.C.2 below. 49 Guowuyuan Guanyu Jiaqiang Difang Zhengfu Rongzi Pingtai Gongsi Guanli Youguan Wenti de Tongzhi (国务院关于加强地方政府融资平台公司管理有关问题的通知) [Notice of the State Council on Issues Concerning Strengthening the Management of Local Government Financing Vehicle Companies] (promulgated by the St. Council, June 10, 2010, effective June 10, 2010), https://perma.cc/8ECH-3X4M [hereinafter State Council 2010]. 50 See Zhan Xiangyang (詹向阳), Bianzheng Kandai Difang Zhengfu Rongzi Pingtai Fazhan (辩证看待地方政府融资平台发展) [Dialectically Viewing the Development of Local Government Financing Vehicles], 7 Zhongguo Jinrong (中国金融) [China Fin.] 38, 38–40 (2010). 51 See Sufang Xie (谢素芳), Zhengfu Juzhai: Guifan he Jiandu Shuangguan Qixia (政府举债: 规范和监督双管齐下) [Government Debt: To Regulate and to Supervise Simultaneously], 6 Zhongguo Ren Da (中国人大) [Chinese People’s Congress] 35, 35 (2012). 52 See NAO 2013, supra note 48, 6–7 tbl.3. 53 See China to Refinance 11 Trln Local Govt. Debt, Shenzhen Daily, Dec. 30, 2015. 54 Id. 55 See Shih, supra note 42, at 27. 56 See Kan Huo, Local Governments Hope to Dodge Debt Deadlines, Caixin (Feb. 24, 2012), http://english.caixin.com/2012-02-24/100360506.html. 57 Id. 58 See Bettina Wassener, Report Warns of Chinese Municipal Debt Risks, N.Y. Times (Sept. 26, 2013), http://nyti.ms/1unppTY. 59 Guanyu Zhizhi Difang Zhengfu Weifa Weigui Rongzi Xingwei de Tongzhi (关于制止地方政府违法违规融资行为的通知) [Notice on Putting a Stop to Local Governments’ Illegal Financing Activities] (promulgated by the Ministry of Fin., Nat’l Dev. & Reform Comm’n, People’s Bank of China, China Banking Regulatory Comm’n, Dec. 24, 2012, effective Dec. 24, 2012), https://perma.cc/3A2N-9REC [hereinafter MOF et al. 2012]. 60 See Simon Rabinovitch, China Averts Local Government Defaults, Fin. Times (Jan. 29, 2013), http://on.ft.com/1xaPzZP. 61 See Simon Rabinovitch, China Gives Local Governments Go-Ahead to Roll Over Debt, Fin. Times (Jan. 2, 2014), http://on.ft.com/ZGMINt. 62 Guowuyuan Guanyu Jiaqiang Difang Zhengfuxing Zhaiwu Guanli de Yijian (国务院关于加强地方政府性债务管理的意见) [Opinions on Strengthening the Regulation of Governmental Debts] (promulgated by the St. Council, Oct. 2, 2014, effective Oct. 2, 2014), https://perma.cc/TD7Q-8XLT. 63 See Xinhua News Agency, China Focus: China Targets Local Gov’t Debt Risks with New Rules, Renmin Wang (人民网) [People’s Network] (Oct. 3, 2014), http://en.people.cn/n/2014/1003/c90882-8790763.html. 64 See Huo, supra note 56. 65 Guowuyuan Bangongting Zhuanfa Caizheng Bu Renmin Yinhang Yinjianhui Guanyu Tuoshan Jiejue Difang Zhengfu Rongzi Pingtai Gongsi Zaijian Xiangmu Houxu Rongzi Wenti Yijian de Tongzhi (国务院办公厅转发财政部人民银行银监会关于妥善解决地方政府融资平台公司在建项目后续融资问题意见的通知) [Notice of the General Office of the State Council Transmitting Opinion of the Ministry of Finance, the People’s Bank of China, and the China Banking Regulatory Commission on Appropriately Handling the Follow-up Financing Issues of In-Process Projects of Local Government Financing Vehicle Companies] (promulgated by the Gen. Off. of the St. Council, May 15, 2015, effective May 15, 2015), https://perma.cc/8SBU-TYPM. 66 Lingling Wei, China Backtracks on Local Government Debt, Wall St. J. (May 15, 2015), http://on.wsj.com/1KSd5ku. 67 See, e.g., the information about the issuer provided in the Leshan, Yongzhou, Tianjin, Zhuzhou, and Nanjing prospectuses, listed in the Appendix. 68 See, e.g., the information about the issuer provided in the Hebei Bohai and Zhenjiang prospectuses, listed in the Appendix. 69 See Xu Yuan (徐瑗), Hu Zuliu: Caozuo Bu Touming Shi Difang Rongzi Zhenzheng Fengxian (胡祖六: 操作不透明是地方融资真正风险) [Hu Zuliu: Lack of Transparency in Operations Is the Real Risk of Local Financing], Xinlang Xinwen (新浪新闻) [Sina News] (July 5, 2011), http://bit.ly/zpeOu5 (interview with Hu Zuliu, Goldman Sachs (Asia) LLC’s Executive Director and Asia Pacific chief economist). 70 See Difang Zhengfu Rongzi Pingtai Gongsi Luanxiang (地方政府融资平台公司乱象) [The Chaos of Local Government Financing Vehicles], Dongfang Zaobao (东方早报) [Oriental Morning Post] A33 (June 28, 2011), http://bit.ly/1JF90xS. 71 See State Council 2010, supra note 49. 72 See id. 73 See, e.g., Guanyu Jinyibu Guifan Difang Zhengfu Tourongzi Pingtai Gongsi Faxing Zhaiquan Xingwei Youguan Wenti De Tongzhi (关于进一步规范地方政府投融资平台公司发行债券行为有关问题的通知) [Notice on Relevant Issues in Further Standardizing the Acts of Local Government Investment and Financing Vehicle Companies in Issuing Bonds] (issued by Office of the Nat’l Dev. & Reform Comm’n, Nov. 20, 2010, effective Nov. 20, 2010), https://perma.cc/KM6Z-4GUU; see also Guanyu Guanche Guowuyuan Guanyu Jiaqiang Difang Zhengfu Rongzi Pingtai Gongsi Guanli Youguan Wenti De Tongzhi Xiangguan Shixiang De Tongzhi (关于贯彻国务院关于加强地方政府融资平台公司管理有关问题的通知相关事项的通知) [Notice on Matters Related to the Implementation of the State Council’s Notice on Issues Regarding the Strengthening of Control over Local Government Financing Vehicles] (jointly issued by Ministry of Fin., Nat’l Dev. & Reform Comm’n, People’s Bank of China, and China Banking Regulatory Comm’n, July 30, 2010, effective July 30, 2010), https://perma.cc/BJQ7-V8UL [hereinafter MOF et al. 2010]. 74 Standard Chartered, On the Ground: Global Research: China—Local Debt: USD 3.6–4.0tn and Counting 2, Oct. 2, 2013 (on file with first author). 75 SeeMega Prospectus, listed in the Appendix, at iii. 76 See id., at cover p., 107. 77 See Andrew Ang, Jennie Bai & Hao Zhou, The Great Wall of Debt: The Cross Section of Chinese Local Government Credit Spreads (Georgetown McDonough School of Business, Research Paper No. 2603022, 2015; Columbia Business School, Research Paper No. 15–57, 2015), http://ssrn.com/abstract=2603022. 78 See BNP Paribas, Equities Research: China Banks: 10 Questions on the LGFV Debt Swap Programme 1 (June 25, 2015), https://perma.cc/6PRB-R6RE. 79 See Ang, Bai & Zhou, supra note 77. 80 See Mudi: Difang Zhai Pingji Weishenme Dou Shi AAA? (穆迪: 地方债评级为什么都是AAA?) [Moody’s: Why Is All the Local Debt Rated AAA?], Diyi Caijing (第一财经) [First Fin. & Econ.] (June 12, 2015), http://bit.ly/1TZd05g (citing Moody’s Vice General Manager Zhu Shuning). 81 See China’s Second LGFV Bond Sale in Doubt as State Support Pulled, Bloomberg News (Dec. 16, 2014), http://bloom.bg/1EtPTLF. 82 See infra Part II.B.1 and notes 109–22 and accompanying text. 83 See Guowuyuan Guanyu Jiaqiang Difang Zhengfuxing Zhaiwu Guanli de Yijian (国务院关于加强地方政府性债务管理的意见) [Opinions on Strengthening the Regulation of Governmental Debts] (promulgated by the St. Council, Oct. 2, 2014, effective Oct. 2, 2014), https://perma.cc/TD7Q-8XLT. 84 State Council 2010, supra note 49; MOF et al. 2010, supra note 73. 85 See Lingling Wei, Shanghai Chaori in Default on Bond Interest Payments, Wall St. J. (Mar. 6, 2014), http://on.wsj.com/1pBfS6s. The article does not add the qualification “reform-era China”, but we feel it is wise, since there may well have been defaults in the early years of the People’s Republic of China on bonds issued before 1949. 86 See Pete Sweeney, Chinese Bank Sues Local Govt Property Company in Rare Public Spat, Reuters (July 2, 2014), http://reut.rs/1unnleW. Of course, it could always be argued that the defaulting company was not an LGFV, but a different kind of company that was nevertheless wholly owned by the local government. But this comes perilously close to arguing that LGFVs are essentially by definition any government-owned company that we see after the fact getting its debt paid off by local government. This is not helpful to creditors or analysts ex ante. 87 Guowuyuan Bangongting Guanyu Zuohao Difang Zhengfuxing Zhaiwu Shenji Gongzuo de Tongzhi (国务院办公厅关于做好地方政府性债务审计工作的通知) [Notice of the General Office of the State Council on Properly Handling the Audit of the Governmental Debts] (promulgated by the Gen. Off. of the St. Council, Feb. 13, 2011, effective Feb. 13, 2011), http://laws.66law.cn/law-129676.aspx. 88 For an example of the classifications in use, see supraTable 1. 89 Notice of the General Office of the State Council on Properly Handling the Audit of the Governmental Debts, Feb. 13, 2011. The term used here, gongyong shiye danwei (公用事业单位), which we have translated as “institutions with a public purpose,” refers to institutions such as schools, hospitals, railroads, and public utilities. Here and elsewhere, “to be repayable” is intended to convey an obligation, and is not to be understood as a statement of fact about the borrower’s financial capability. 90 Id. 91 Id. (emphasis added). 92 For a canvassing of the various ways government “debt” can be defined, including the counting of unfunded pension liabilities or revocable legislative promises, see Jennifer Weiner, Assessing the Affordability of State Debt, Fed. Reserve Bank of Boston (Dec. 2013),http://bit.ly/weiner-statedebt (New England Public Policy Center Research Report 13-2). 93 NAO 2013, supra note 48, at 6. 94 See id. 95 See id. at 7. 96 See supra note 32 and accompanying text. 97 See NAO 2011, supra note 25, introductory sec. 98 See NAO 2013, supra note 48, at 6. 99 See NAO 2011, supra note 25, sec. 1.5. As discussed in the preceding subsection, and in Part II.B.1, what the NAO counts as direct and contingent debt liabilities are not necessarily such, even in the Chinese system. 100 See id., sec. 3.3. 101 See The Chaos of Local Government Financing Vehicles, supra note 70. 102 See Fan Xuyang (范盱阳) & Liu Ximing (刘西明), Shixi Difang Zhengfu Rongzi Pingtai De Guifan Yu Chuangxin (试析地方政府融资平台的规范与创新) [A Tentative Analysis of the Regulation and Innovation of Local Government Financing Vehicles], 12 Xin Jinrong (新金融) [New Fin.] 42, 42 (2010), http://bit.ly/1ErnCEX. 103 See Jason Subler & Lu Jianxin, Analysis: Local Government Debt Alarm May Delay China Muni Bonds, Reuters (July 20, 2011), http://reut.rs/1ADZ9J5. 104 See China Regulators, Officials Aim to Ease Local Finance Vehicle Loan Concern, Bloomberg News (July 27, 2010), http://bloom.bg/9InLcH (according to one source, only 27% of loans to LGFVs can be repaid solely by the profits generated by the projects funded by LGFV financing); see also Chinese Banks See Risks in 23% of $1.1 Trillion Loans, Bloomberg News (July 26, 2010), http://bloom.bg/bIaQ6H. Another survey revealed an even more troubling figure. According to a questionnaire survey that was conducted by the Institute of Finance & Banking, Chinese Academy of Social Sciences in 110 cities in 2010, only 10% of LGFVs relied on the income from financed projects as the primary source to repay their debt, whereas about 90% of LGFVs relied on other sources, such as land development arrangements and fiscal transfers, for repayment. See Lei Jiaxiao & Peng Bo, Difang Zhengfu Rongzi Pingtai Cunzai Tizhi Jizhi Fengxian (地方政府融资平台存在体制机制风险) [Institutional and Mechanism Risks Exist in Local Government Financing Vehicles], Keji Ribao (科技日报) [Sci. & Tech. Daily] (Jan. 30, 2011), http://bit.ly/1LLTerM. 105 See, e.g., Datong Prospectus, Jilin Prospectus, Panjin Prospectus, Tianjin Binhai Prospectus, Yichang Prospectus, and Zhenjiang Prospectus, listed in the Appendix . 106 SeeBaoji Prospectus, Jilin Prospectus, Mega Prospectus, Yichang Prospectus, Yichun Prospectus, and Zhenjiang Prospectus, listed in the Appendix. 107 See, e.g., Baoji Prospectus, listed in the Appendix. 108 According to one study, research in Guangdong province revealed that typically the general office of the county or municipal People’s Congress Standing Committee would issue a letter “acknowledging” the loan deal between the bank and the local government. See Liang, supra note 22, at 17–21. The letter would recite that the loan was in conformity with borrowing plans set out in the local government’s most recent Work Report approved by the local People’s Congress, and was “acknowledged” (queren) by the Standing Committee. According to the study, banks supply the form of such letters and insist that they be signed before making the loan. See id. at 20 (showing form of letter). Importantly, it is addressed to the local government, not to the lending bank. The study also reproduces the form of a local People’s Congress Standing Committee resolution supporting a loan to a LGFV. See id. at 20–21. The resolution recites the Standing Committee’s approval of the loan and states that any part of the loan not repaid by the LGFV will be repaid from the local government’s budget. 109 Danbao Fa (担保法) [Guarantee Law] (promulgated by the Standing Comm. Nat’l People’s Cong., June 30, 1995, effective Oct. 1, 1995), art. 8, https://perma.cc/GU58-JJ6Q. This prohibition has now been written into the Budget Law as well. See Quanguo Renmin Daibiao Dahui Changwu Weiyuanhui Guanyu Xiugai “Zhonghua Renmin Gongheguo Yusuan Fa” de Jueding (全国人民代表大会常务委员会关于修改”中华人民共和国预算法”的决定) [Decision of the Standing Committee of the National People’s Congress of the People’s Republic of China on the Revision of the “Budget Law of the People’s Republic of China”] (promulgated by the Standing Comm. Nat’l People’s Cong., Aug. 31, 2014, effective Aug. 31, 2014) item 30, https://perma.cc/7EJ9-6UBU. 110 GITIC’s Empty Offers, Economist (Jan. 14, 1999), http://www.economist.com/node/182583. 111 See Wu Daqing, Guanyu Guifan Difang Zhengfu Tourongzi Pingtai De Jidian Jianyi (关于规范地方政府投融资平台的几点建议) [Suggestions on Regulating Local Government Investment and Financing Vehicles], Jinrong Shibao (金融时报) [Fin. News] (Feb. 8, 2010), http://bit.ly/1JqyZdm. See also the discussion of recent court cases infra at notes 119–20 and accompanying text. 112 See Xiaomin Song (宋晓敏) & Yuan Tian (田原), Zhuce Kuaijishi Zai Difang Zhengfu Rongzi Pingtai Gongsi Shenji Zhong Ying Guanzhu de Wenti (注册会计师在地方政府融资平台公司审计中应关注的问题) [The Issues to Which Registered Auditors Should Pay Close Attention When Auditing Local Government Financing Vehicle Companies], 7 Caiwu Yu Kuaiji (财务与会计) [Fin. & Acct.] 54, 54–55 (2011). 113 See State Council 2010, supra note 49. 114 MOF et al. 2010, supra note 73. 115 Caizheng Bu Guanyu Guifan Difang Geji Zhengfu Bumen Juzhai he Danbao Chengnuo Xingwei de Tongzhi (财政部关于规范地方各级政府部门举债和担保承诺行为的通知) [Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels] (promulgated by the Ministry of Fin., Sept. 2, 2010, effective Sept. 2, 2010), https://perma.cc/85RB-PMWM. Oddly enough, the Notice itself is marked “Not to Be Made Public” (bu yu gongkai). It is mentioned in a 2011 press release from the Ministry of Finance: see Caizheng Bu Fuzeren Jiu Jiaqiang Difang Zhengfuxing Zhaiwu Guanli Dawen (财政部负责人就加强地方政府性债务管理答问) [Question Answering by the Leading Official Regarding Strengthening the Regulation of Governmental Debts], Zhonghua Renmin Gongheguo Caizheng Bu (中华人民共和国财政部) [Ministry of Finance of the People’s Republic of China] (Aug. 11, 2011), https://perma.cc/8C9N-Q96W. 116 See Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels, Sept. 2, 2010. 117 See MOF et al. 2012, supra note 59. 118 The theory of liability for fault in the invalidity of a guarantee can be found in general terms in article 58 of the Contract Law (liability for invalid contracts) and much more specifically in article 5 of the Guarantee Law, which states that when a guarantee contract is found to be invalid, the debtor, creditor, and guarantor shall each bear civil liability in proportion to their fault. This provision was then further elucidated by a Supreme People’s Court interpretation in 2000. This interpretation repeated that government guarantees were invalid and that the damage to the creditor from invalidity should be handled in accordance with article 5 of the Guarantee Law. It added, however, that where the creditor and the guarantor were both at fault, the guarantor’s liability was not to exceed half of the unpaid debt. See Zuigao Renmin Fayuan Guanyu Shiyong “Zhonghua Renmin Gongheguo Danbao Fa” Ruogan Wenti De Jieshi (最高人民法院关于适用“中华人民共和国担保法”若干问题的解释) [Judicial Interpretation of the Supreme People’s Court on Some Issues Regarding the Application of the Guarantee Law of the People’s Republic of China] (promulgated by Sup. People’s Ct., Dec. 8, 2000, effective Dec. 13, 2000), arts. 3,7, https://perma.cc/2AXK-AQ4V. In the cases we have reviewed, guarantors are routinely found at fault for making guarantees they should not have made, and creditors are equally routinely found at fault at the same time for carelessly accepting a guarantee from an entity they should have known did not have the capacity to make it. The most sophisticated exposition of this theory can be found in a pragmatic and well-reasoned judgment of the Guangdong Province Higher-Level People’s Court in the case of Foshan Shi Chancheng Qu Renmin Zhengfu Yu Jiaotong Yinhang Xianggang Fenhang Jiekuan Danbao Hetong Jiufen Shangsu’an (佛山市禅城区人民政府与交通银行香港分行借款担保合同纠纷上诉案) [Appeal Case of Foshan City Chancheng District People’s Government v. Communications Bank of China, Hong Kong Branch] (Guangdong Higher People’s Ct. Jan. 22, 2003), https://perma.cc/5SB6-J9ZA [hereinafter Foshan Case]. The case and the principles behind it are discussed by Professor Yang Lixin in Yang Lixin (杨立新), Chengnuo Han, Zuigao’e Baozheng, Wuxiao Baozheng Peichang Zeren (承诺函, 最高额保证, 无效保证赔偿责任) [Liability for Commitment Letters, Maximum Guarantees, and Invalid Guarantees], 110 Falü Zixun Wang (法律咨询网) [Legal Advice] (July 11, 2010), https://perma.cc/L93N-FP6B. There are other cases with the same reasoning and result: see, e.g., Zhen Zhengfu Danbao Qiye Jiekuan, Fayuan Pan Danbao Wuxiao, Zhen Zhengfu You Guocuo (镇政府担保企业借款 法院判担保无效 镇政府有过错) [County Government Guarantees Loan to Enterprise, Court Rules that Guarantee Is Invalid but that County Government Is at Fault], Chengdu Shangbao (成都商报) [Chengdu Bus. News] (July 9, 2010), https://perma.cc/7HWK-FU9S (reporting that the township government was found liable for half of unpaid debt on the theory of fault for invalidity of guarantee); Jian Zhengbo (简正波), Guojia Jiguan Danbao Wuxiao, Peichang Zeren Buneng Mianchu (国家机关担保无效 赔偿责任不能免除) [State Organ Cannot Be Exempt from Liability Even if Its Guarantee Was Invalid], Jiujiang Fayuan Wang (九江法院网) [Jiujiang Ct. Network] (May 16, 2012), https://perma.cc/TZ9V-B9AW (reporting a case from Jiangxi province where subcounty local government purported to guarantee a debt and was found liable for half the unpaid amount on the same theory); Chen Guiping (陈桂平), Guquan Zhuanrang Jiufen: Zhengfu Liandai Danbao Xiaoli Ruhe (股权转让纠纷—政府连带担保效力如何) [Equity Transfer Dispute: What Is the Validity of Government’s Joint and Several Guarantee], Zhongguo Lüshi Wang (中国律师网) [China Lawyer Network] (Dec. 30, 2013), https://perma.cc/B79C-8WAQ (discussing a 2005 Heilongjiang case involving the government of Tongjiang City with similar circumstances, similar liability, and similar theory). 119 See Zhou Weishe (周巍摄), Guanzhou Zhongyuan: Difang Zhengfu Danbao Beisu Tonglei Anjian Jinnian Yongxian (广州中院: 地方政府担保被诉同类案件近年涌现) [Guangzhou Intermediate People’s Court: Many Cases in Recent Years In Which Local Governments Are Sued on Their Guarantees], Yangchen Wanbao (羊城晚报) [Yangcheng Evening News] (May 22, 2014), https://perma.cc/52XU-V7QD. 120 Zhongguo Yinhang (Xianggang) Youxian Gongsi yu Liaoning Sheng Renmin Zhengfu, Hulu Dao Xinchang Baozheng Hetong Jiufen Ershen (中国银行(香港)有限公司与辽宁省人民政府、葫芦岛锌厂保证合同纠纷二审) [Second-Instance Adjudication Between Bank of China (Hong Kong), Ltd., Liaoning People’s Gov't, and Hulu Island Zinc Plant (Bank of China)] (Sup. People’s Ct. Dec. 22, 2014), https://perma.cc/8RLZ-TPSB. 121 Foshan Case, supra note 118. 122 See, e.g., Notice of the Ministry of Finance on Regulating the Borrowing and Guarantees and Promises of Local Governments at All Levels, Sept. 2, 2010. 123 Control over personnel and finances of courts in China rests with political power at the same administrative level at the court. Formally, this means the local People’s Congress and People’s Government respectively. In practice, it means the local Party authorities. 124 See Yao Liang (姚亮), Difang Zhengfu Rongzi Pingtai Gongsi Daikuan De Danbao Wenti Yanjiu (地方政府融资平台公司贷款的担保问题研究) [Study of the Problem of Guarantees for Loans to Local Government Financing Vehicle Companies], Yao Lüshi de Weibo (姚律师的微博) [Attorney Yao’s Blog] (Aug. 9, 2010), https://perma.cc/9HD6-9JSX. 125 There is some dispute in the literature over the degree to which the widespread story that promotion depends on delivering GDP is actually true. Certainly it is widely believed in the Chinese media. See, e.g., Tan Yingzi, Evaluation for Officials Hinges on Public Input, China Daily (Aug. 13, 2009), http://bit.ly/1sZtwZi (discussing new system for assessing public officials that will look at more than local economic output). See generally Victor Shih et al., Getting Ahead in the Communist Party: Explaining the Advancement of Central Committee Members in China, 106 Am. Pol. Sci. Rev. 166, 166–87 (2012), for a review of the academic literature; the authors conclude from their own research that there is in fact no relation. 126 This observation is quite common in the Chinese media. See, e.g., Zuo Xiaolei (左晓蕾), Renqi Nei Yusuan Ying Yueshu Ying Yu Dingfang Zhai Tongbu (任期内预算硬约束应与地方债同步) [Hard Budget Constraint During Term of Office Should Synchronize with Municipal Debt], Shanghai Zhengquan Bao (上海证券报) [Shanghai Sec. News] (Oct. 24, 2011), https://perma.cc/AQ4V-SRJ4. 127 See Loudi Prospectus, Hebei Bohai Prospectus, Lishui Prospectus, Shaoyang Prospectus, Xianyang Prospectus, and Yongzhou Prospectus, listed in the Appendix. See also Kelvin Soh & Aileen Wang, Special Report: China’s Debt Pileup Raises Risk of Hard Landing, Reuters (Oct. 10, 2011), http://reut.rs/mZERlu (noting that land use rights are often used as collateral for loans to LGFVs). 128 Danbao Fa (担保法) [Guarantee Law] (promulgated by the Standing Comm. Nat’l People’s Cong., June 30, 1995, effective Oct. 1, 1995), art. 34, https://perma.cc/GU58-JJ6Q. 129 A reversion is the technical term for the interest remaining in the hands of one who grants to another an interest in land with a shorter term than his/her own term. For example, if X has a fee simple (an interest in perpetuity) and grants a twenty-year lease to Y in year Z, then X’s remaining interest (a future interest that ripens to a fee simple in year Z+20) is called the reversion. 130 See Zhonghua Renmin Gongheguo Chengshi Fangdichan Guanli Fa (中华人民共和国城市房地产管理法) [Urban Real Estate Administration Law of the People’s Republic of China] (promulgated by the Standing Comm. Nat’l People’s Cong., July 5, 1994, effective Jan. 1, 1995, amended Aug. 30, 2007 and Aug. 27, 2009), art. 51, https://perma.cc/YT4A-U4WU. 131 In the Hami Prospectus, the land use rights purportedly securing the debt were disclosed to be so-called “allocated” (huabo) land and hence essentially worthless as security for the reasons explained above. It is also worth noting that at the end of 2012, the Ministry of Finance and other departments issued a joint notice requiring, among other things, that any allocated land transferred to LGFVs had to be paid for and converted to granted land. See MOF et al. 2012, supra note 59. The Hami Prospectus states that the parcels in question were acquired in May 2013, and the bonds were not issued until July 2014. Thus, the land use rights held by the LGFV in question appear to be in violation of Ministry of Finance rules. 132 See generally China Cities Value Land at Winnetka Prices with Bonds Seen Toxic, supra note 3. In this article, the reporter interviewed some governmental officials in charge of the LGFV financing. In the northeast city of Cangzhou, the person in charge of a large-scale bond sale for a port expansion project could not even locate on a map the land his company had supposedly put up as collateral. 133 See George E. Peterson, Land Leasing and Land Sale as an Infrastructure-Financing Option 6 (World Bank Policy Research, Working Paper No. 4043, 2006), https://perma.cc/9YVB-8JZZ. 134 Bullish analysts often point to various factors such as increasing urbanization as reasons why land prices will continue to rise. What they do not explain, however, is why such widely known factors are not already incorporated into the current price of land. 135 See Jamil Anderlini, Chinese Cities See Property Prices Fall, Fin. Times (Jan. 18, 2015), http://on.ft.com/1Mb2AMc. 136 See Chen Bingcai (陈炳才), Tian Qing (田青) & Li Feng (李峰), Difang Zhengfu Rongzi Pingtai Fengxian Fangfan Duice (地方政府融资平台风险防范对策) [Preventive Measures Against the Risks of Local Government Financing Platform Companies], 1 Zhongguo Jinrong (中国金融) [China Fin.] (2010), http://bit.ly/z0Bbe2. 137 See Naoki Murakami, The Capital Market and Credit Rating in China (Ctr. for China and Asian Stud., Nihon Univ. College of Econ., Sept. 2009), https://perma.cc/4WGN-YUYQ. 138 See Brian Milner, Shot Down by SEC, Chinese Rater Punishes U.S. Debt, Globe & Mail (Nov. 9, 2010), http://bit.ly/QuRV4o. 139 See China Cities Value Land at Winnetka Prices with Bonds Seen Toxic, supra note 3. Another credit rating agency, China International Capital Corp., gave the bonds its third-lowest rating, a non-investment grade. See id. 140 See, e.g., Hebei Bohai Prospectus, Xianyang Prospectus, and Yongzhou Prospectus, listed in the Appendix. 141 See, e.g., Tracy Alloway, China’s Uncollateralised, Cash Flow-Less, Local Government Loans,Fin. Times (UK) (June 7, 2011), http://on.ft.com/1NHzs0W (pointing out that while the valuation of LURs as collateral has attracted attention, the larger problem remains “those dodgy uncollateralized loans”). 142 See Wu, supra note 111. There is a lively metaphor for such a phenomenon in China: “to marry one daughter to more than one person.” Unlike bigamy, there is nothing per se unlawful about creating multiple encumbrances on the same collateral, and the law provides a method of deciding priority. The unlawfulness occurs when the debtor falsely claims that no previous encumbrances exist. 143 See Jiang Shulin (江抒琳), Xing Shihe (邢世和) & Wang Yuyin (王玉印), Tudi Diya Yu Jinrong Fengxian (土地抵押与金融风险) [Land Mortgage and Finance Risks], inKexue Heli Yong Di, Ren Di Hexie Xiangchu 2008 Nian Xueshu Nianhui Lunwen Ji (科学合理用地, 人地和谐相处2008年学术年会论文集) [Proceedings of 2008 Annual Academic Meeting on the Scientific and Reasonable Use of Land, with Man and Land Harmoniously Coexisting] 168 (Lin Yaohua (林耀华) ed., 2008). 144 See id. at 165. 145 See generally Taisu Zhang, The Pragmatic Court: Reinterpreting the Supreme People’s Court of China, 25 Colum. J. Asian L. 1 (2012). 146 See Zhonghua Renmin Gongheguo Chengshi Fangdichan Guanli Fa (中华人民共和国城市房地产管理法) [Urban Real Estate Administration Law] (promulgated by the Standing Comm. Nat’l People’s Cong., July 5, 1994, effective Jan. 1, 1995, amended Aug. 30, 2007), art. 39, https://perma.cc/YT4A-U4WU. 147 See Jiang, Xing & Wang, supra note 143, at 165. 148 See Urban Real Estate Administration Law, Jan. 1, 1995, art. 16. 149 This point is made by Jiang, Xing & Wang, supra note 143, at 166. 150 See MOF et al. 2012, supra note 59. 151 SeeShaoyang Prospectus, listed in the Appendix, at 62–66; see alsoLoudi Prospectus, listed in the Appendix, at 70–73. 152 SeeHebei Bohai Prospectus, listed in the Appendix. 153 SeeYichun Prospectus, Hefei Prospectus, Yichang Prospectus, Zhuzhou Prospectus, and Jiangyin Prospectus, listed in the Appendix. The Datong Prospectus notes the existence of the receivable as an assurance to bondholders, but does not state that a security interest has been created in their favor. 154 See supra Part II.B.1 and notes 109–26 and accompanying text. 155 We confirmed this point in a discussion in August 2014 with a group of Chinese lawyers specializing in local government debt. 156 Baoji Prospectus, listed in the Appendix, at 10. 157 Leshan Prospectus, listed in the Appendix, at 3. 158 Mega Prospectus, listed in the Appendix, at 95. 159 Id. 160 Danbao Fa (担保法) [Guarantee Law] (promulgated by the Standing Comm. Nat’l People’s Cong., June 30, 1995, effective Oct. 1, 1995), art. 7, https://perma.cc/GU58-JJ6Q. 161 See Patrick Chovanec, Falling Dominoes, Patrick Chovanec Blog (June 27, 2012), https://perma.cc/R3GA-EDCA. 162 I am indebted to Patrick Chovanec, No Guarantee, Patrick Chovanec Blog (May 19, 2012), https://perma.cc/DT8B-5A9A, for the analysis in this and the following two paragraphs. 163 The concept of correlated risk is explained more fully at Correlated Risks, World Fin., http://www.worldfinance.com/home/risk-encyclopaedia/correlated-risks (last visited Oct. 3, 2014). 164 See generally Yang Na & Ma Yuan, Fool’s Gold Behind Beijing Loan Guarantees, Caixin (Mar. 20, 2012), http://english.caixin.com/2012-03-20/100370485.html. 165 See Fan & Liu, supra note 102. 166 See, e.g., Mega Prospectus, listed in the Appendix, at 92–93 (“Description of Other Material Indebtedness”). 167 SeeBaoji Prospectus, listed in the Appendix, at 73. 168 See id.; see also Qiye Zhunze Di 22 Hao: Jinrong Gongju Queren He Jiliang (企业会计准则第22号—金融工具确认和计量) [Accounting Standards for Enterprises No. 22—Recognition and Measurement of Financial Instruments] (promulgated by Ministry of Fin., Feb. 15, 2006, effective Feb. 15, 2006), art. 18, https://perma.cc/42QT-KZJS (defining “sellable financial assets”). 169 The Chinese term is jihua danlie shi: a city separately listed under the state plan. The idea is simply that of a city sufficiently large and important to be treated as a province for economic planning purposes. Although the scope of economic planning has declined drastically in post-Mao China, the category of “separately planned city” remains when it is desired for whatever reason to include large cities within the scope of activities generally reserved for provinces. 170 See Du, supra note 6, Chen Jia, Expand Local Debt-Swap Plan to Further Reduce Burden, Says Official, China Daily, Dec. 22, 2015. For a detailed analysis, see Barry Naughton, Local Debt Restructuring: A Case of Ongoing Authoritarian Reform, 47 China Leadership Monitor (Summer 2015), http://hvr.co/1QBWqUu. 171 Caizheng Bu Youguan Fuzeren Jiu Faxing Difang Zhengfu Zhaiquan Zhihuan Cunliang Zhaiwu Youguan Wenti Da Jizhe Wen (财政部有关负责人就发行地方政府债券置换存量债务有关问题答记者问) [The Leading Director of the Ministry of Finance Answers Reporters’ Questions Concerning Issuing Government Bonds to Replace the Remaining Debt], Zhonghua Renmin Gongheguo Caizheng Bu (中华人民共和国财政部) [Ministry of Fin.] (Mar. 12, 2015), https://perma.cc/WRQ8-H3GX. 172 See id. 173 See Caizheng Bu Youguan Fuzeren Jiu Xiada Di 2 Pi Zhihuan Zhaiquan Edu Youguan Wenti Da Jizhe Wen (财政部有关负责人就下达第2批置换债券额度有关问题答记者问) [Responsible Person of the Ministry of Finance Answers Reporters’ Questions Concerning Allocation of Second Quota in Bonds-for-Debt Swap], Zhonghua Renmin Gongheguo Caizheng (中华人民共和国财政部) [Ministry of Fin.] (June 10, 2015), https://perma.cc/PRQ4-4FM5. 174 See Chen, supra note 170. 175 See China To Refinance 11 Trln Local Govt. Debt, supra note 53 (11.14 trillion estimate); China’s Regional Financial Crisis, supra note 11 (25 trillion estimate). 176 The definitions of these categories of debt are discussed in Part II.A. 177 See Supporting China’s Infrastructure Stimulus Under the INFRA Platform, INFRA Update (World Bank) (June 2010), http://bit.ly/1Libfx6 (“[O]nly a small part of the stimulus was reflected in the central government’s budget. Additional financing—largely bank lending—contributed almost two-thirds of the stimulus.”); Yu Yongding, China’s Response to the Global Financial Crisis, East Asia Forum (Jan. 25, 2010), http://bit.ly/1J7j7js (“The central government is financing one-quarter of the 4 trillion yuan package, in the form of direct grants and interest rate subsidies. Bank credit is the second most important source of finance for the stimulus package.”). 178 Although article 35 of the 2015 Budget Law reserves bond-issuing powers at the subcentral level to provincial-level governments, see supra note 36 and accompanying text, the Ministry of Finance in 2015 interpreted this to include separately planned cities. See, e.g., Difang Zhengfu Yiban Zhaiquan Faxing Guanli Zanxing Banfa (地方政府一般债券发行管理暂行办法) [Interim Measures for the Administration of the Issuance of Ordinary Bonds by Local Governments] (promulgated by the Ministry of Fin., Mar. 12, 2015, effective Mar. 12, 2015), https://perma.cc/H87R-8VXT. (The concept of separately planned cities is explained supra in note 169.) More recently, it expanded its interpretation still further to include free trade zones. See Guanyu Zuohao 2016 Nian Difang Zhengfu Zhaiquan Faxing Gongzuo de Tongzhi (关于做好2016年地方政府债券发行工作的通知) [Notice on Doing Well the Work of Issuance of Local Government Bonds in 2016] (promulgated by the Ministry of Fin., Jan. 25, 2016), sec. 6, para. 2, https://perma.cc/T7Z9-MG3P. The legal implications of this apparent disregard for a clear statutory rule are beyond the scope of this Article. 179 See supra note 120 and accompanying text for a discussion of the Bank of China case. 180 One can often see this phenomenon in real estate transactions. Consider the following report: In the Hongshan district of Wuhan City, 927 mu of collective mountain and forest land was sold illegally four times in succession, producing a profit for the work units and people involved of over 40 million RMB. A budget audit of the State Food and Drug Supervision and Management Bureau revealed that in December 2003, the organization services department of that bureau signed, without any authority, a contract with two companies to sell for 85 million RMB 216 mu of land in the Chanping District, Beijing. The companies were to use the land for running a business and development. That land had been provided to the Bureau without cost by the State. This land sale was not reported to the departments concerned for approval. This was done without any of the required legal procedures in violation of the Land Management Law and the regulations issued by the State Council to bring order into the real estate market. After the audit, the Bureau took the matter seriously and is now studying how to rectify the matter. Li Jinhua (李金华) [Director, National Audit Office], Guanyu 2003 niandu zhongyang yusuan zhixing he qita caizheng shouzhi de shenji gongzuo baogao (关于2003年度中央预算执行和其他财政收支的审计工作报告) [Audit Report on the Implementation and Financial Receipts and Expenditures of the 2003 Central Budget] (June 24, 2004), https://perma.cc/K9LD-BUXB (translation courtesy of David Cowhig). One way of rectifying the matter would simply be to refuse to recognize the sale, and to support the reoccupation of the land by whoever had the right to it before the sale. Buyers in illegal transactions would quickly learn their lesson and insist that legal procedures be followed before parting with their money. But that apparently is not a step the authorities were willing to take. © The Author(s) [2017]. Published by Oxford University Press on behalf of the American Society of Comparative Law. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

American Journal of Comparative LawOxford University Press

Published: Dec 1, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off