POLICY RESEARCH WORKING PAPER 2 69 Risk Shifting and Long-Term Risk shifting and incomplete contracting lie at the heart of Contracts the agency relationship inherent in the procurement Evidence from the Ras Gas Project and financing of large-scale projects such as power plants, oil and gas pipelines, Mansoor Dailami and liquefied natural gas Robert Hauswald facilities. An investigation of Ras Gas bonds provides empirical evidence of the risk-shifting consequences of contractual incompleteness. The World Bank World Bank Institute Governance, Regulation, and Finance Division November 2000 POLICY RESEARCH WORKING PAPER 2469 Summary findings Risk shifting and incomplete contracting lie at the heart Consistent with theoretical predictions, they find that of the agency relationship inherent in the procurement the risk factors affecting the sales and purchase and financing of large-scale projects such as power agreements drive perceptions of market risk for Ras Gas plants, oil and gas pipelines, and liquefied natural gas bonds. In particular, Ras Gas yield spreads reflect the (LNG) facilities. Resolving this agency problem is critical market's risk assessment of Kepco. Other priced risks are in structuring the nexus of long-term contracts- energy price and foreign currency exposure (which construction, operating, output sale, and financial influence Ras Gas credit spreads through their impact on contracts-commensurate with the projects' underlying Kepco), Korean economic variables, and spillovers from technological and market organization. turbulence in European and Latin American emerging By investigating the Ras Gas bonds-the largest and debt markets. most liquid global project bonds ever issued in an The authors' analysis shows that the design of each emerging market economy-Dailami and Hauswald contractual arrangement is not independent, because risk provide empirical evidence of the risk-shifting factors relevant to one contract determine the price and consequences of contractual incompleteness. risk premium of the other. Despite heavy capitalization They relate the credit spreads of Ras Gas bonds to the and partial guarantees by the parent companies of Ras bond spreads of the Korea Electric Power Company Gas, the off-take agreement essentially determines the (Kepco), the major customer, in the context of a 25-year riskiness of the bonds. Dailami and Hauswald interpret supply agreement, the oil price index used to price the this as evidence of the nexus-of-contracts view of the LNG, emerging debt market returns, and various firm in the presence of contractual incompleteness: systematic and unsystematic risk variables. Investors bear all residual risks and price their financial claims accordingly. This paper-a product of the Governance, Regulation, and Finance Division, World Bank Institute-is part of a larger effort in the institute to disseminate the lessons of experience and best practices in infrastructure finance and risk management. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact William Nedrow, room J3-283, telephone 202-473-1585, fax 202-334-8350, email address wnedrow@worldbank.org. Policy Research Working Papers are also posted on the Web at www.worldbank.org/research/workingpapers. The authors may be contacted at mdailami@worldbank.org or rhauswald@rhsmith.umd.edu. November 2000. (30 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas ahout development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center Risk Shifting and Long-Term Contracts: Evidence from the Ras Gas Project Mansoor Dailami World Bank Institute The World Bank Washington, DC 20433 and Robert Hauswald Kelley School of Business Indiana University Bloomington, IN 47405-1701 The authors would like to thank Peter Christoffersen, Tim Crack, Bim Hundal, Lutz Kilian, Gregory Randolph, John Strong, Haluk Unal and Volker Wieland for stimulating discussions, and Panos Kogkalidis for excellent research assistance. I Risk Shifting and Long-Term Contracts: Evidence from the Ras Gas Project 1. Introduction The presence of risky debt in a firm's capital structure is known to lead to ex post conflicts of interests between the firm's equity holders and bondholders.' Given eguity's convex pay-off profile, firms that are managed in the interest of equity holders have an incentive to shift risk to their bondholder. Excessive risk taking, if not controlled through contractual and pre- commitment devices, could result in suboptimal investment decisions (over- or under- investment). The associated investment distortions cause deadweight losses to the economy. The risk shifting incentives of risky debt and the associated suboptimal investment problem arise from the interplay of pay-off profiles and informational asymmetries between equity holders and bondholders, e.g., in circumstances where monitoring and contracting costs prevent bondholders from sufficiently observing investment choices made by inside managers. To address such agency problems, recent research has advanced a variety of incentive compatible solutions, including financial contracting (restrictive bond covenants, and convertibility clauses, Kahan and Yermack, 1998), project finance (Berkovitch and Kim, 1990), tax policy (John, John and Senbet 1991), and reform of bankruptcy procedures (Schwartz, 1997). Through or because of such devices, it is claimed, borrowers can credibly signal to lenders that they choose efficient levels of risk. This strand of literature, however, neglects a further source of risk shifting that has a direct incidence on debt finance: contractual incompleteness. Thus, in markets where transactions are characterized by a high degree of uncertainty and relationship specificity conditions, contracting parties often have to leave many contingencies and provisions out of ex ante agreements, creating incomplete contracts. Prime examples of such transactions are investments in large scale projects, i.e., power plants, oil and gas pipelines, and lquified natural gas (LNG) facilities. In this paper, we analyze risk shifting as a consequence of contractual incompleteness2 and relate its sources to the price of risky debt in the context of project finance. The project in question is the Ras Laffan Natural Liquified Gas Company (Ras Gas), set up to extract, process and sell liquefied natural gas from a field off the shore of Qatar. Ras Gas signed a 25 year supply 'See Jensen and Meckling (1976), Myers (1977), Smith and Warner (1979), John (1987), Green (1984). 2 The theoretical underpinnings of contractual incompleteness are well understood in the literature on contract the- ory, whether in its principal-agent or transaction cost setting. According to the transaction-cost approach, contract incompleteness is attributed to high transaction costs of writing, negotiating of contracts, and costs associated with monitoring of contractual performance. (See, for instance, Joskow (1987, 1988), Hart (1988), and Aghion and Bol- ton (1992). 4 contract to sell most of the output to the Korea Gas Corporation (Kogas) for re-sale to the Korea Electric Power Company (Kepco) for electricity generation3. The contractual incompleteness primarily stems from the long-term sales contract, the absence of a spot market for liquefied natural gas, and the economics of LNG fired power plants. At the same time, the incentive to shift risk to investors arises from potential opportunistic behavior by the output buyers and the specificity of the required investment in LNG infrastructure. If investors are rational and markets reasonably efficient from an informational point of view, any material change affecting the buyer of a project's output should immediately feed through to the price of debt and equity. From an analytic perspective, the Ras Gas joint venture offers the unique advantage of having two large global bond issues outstanding and only one major off-taker, Kogas, on behalf of Kepco. Hence, any change affecting the prospects of Kepco and, more largely, the electricity market in Korea should be reflected in the Ras Laffan bond yields. The Asian and Russian financial crises just provide two such external events that adversely affected investor expectations and, in case of the former, Kepco's perceived ability to honor the off-take agreement. Related events revolve around rating changes for Korea and Kepco that permit the definition of further events, both negative ones (downgrades) and, more recently, positive changes (upgrades). Finally, Kepco has both ADRs and global bonds outstanding whose prices and yields provide valuable information about the market's assessment of the company's prospects. Given the contractual arrangements in the off-take agreement and bond covenant that creates a legal and economic link between the output purchase and debt contracts we formulate several hypotheses regarding risk factors. Any risks that can not be explicitly addressed in the nexus of contracts comprised of the off-take agreement and debt contract (covenant) is ultimately borne by investors.4 Having bond yields of both the seller and buyer of the output allows us to investigate which risks are priced by bondholders and, hence, shifted to them. In particular, we expect that any deterioration in the economic prospects of Kepco should increase the likelihood of breach of contract. Ras Gas bondholders' exposure to Kepco comprises several components: exposure to the Korean economy and its demand for energy (electricity, gas), and corporate exposure to Kepco proper in the form of operational, financial and regulatory risks. Since Kepco yields reflect both corporate and Korean exposure we attempt to differentiate the two by including a set of Korean control variables. Finally, both Ras Gas and Kepco bonds belong to the emerging debt market segment of global bond markets so are subject to emerging market risk and contagion (the spill-over effects from the Asian and Russian financial crises). 3The parties later renegotiated the initial contract doubling its volume and dropping price guarantees for Ras Gas in return for the significant economies of scale realized. 4 Contrary to a large company, projects such Ras Gas have only one cash flow stream from which all debt obliga- tions are to be met and dividends paid to project sponsors. Construction, start-up and operational phase tend to be clearly separated fitting the typical time structure of financial contracting models. Consequently, one can easily iso- late the various risks and conflicts of interest involved and analyze their incidence on the financial and contractual arrangements. Furthermore, output tends to be sold to very few customers and often a single off-taker. Conse- quently, investor and creditor concerns necessitate explicit links between capital structures and off-take agreements creating a web of interwoven contracts. (For further discussion of project finance see Kesinger, and Martin (1998), and Finnerty (1996), Dailami and Leipziger (1998). 5 Using a sample of daily data from January 1997 to March 2000 we find that Ras Gas yield spreads exhibit a very high degree of persistence. By far the most important explanatory variable for both levels and changes in credit spreads is the Kepco credit spread: investors rationally anticipate on the incidence of the off-taker's financial and economic condition on the riskiness of their bond. However, we also find evidence for over-reaction: while Ras Gas spreads widen with contemporeanous Kepco spread movements, they narrow in lagged ones. The price of crude oil, which serves as a reference for LNG pricing, comes out largely insignificant: investors seem to disregard commodity price risk.. While rising energy prices increase revenue and debt service capacity it also raises the cost of electricity generation for Kepco leading to higher breach of contract risk. Further investigation shows that the direct energy price impact on Ras Gas is insignificant but that the indirect impact via Kepco's financial position is highly significant: revenue and off-taker quality effects off-set each other but the indirect effect via Kepco yields survives. In terms of Korean exposure variables we find evidence of Ras Gas exposure to the Korean currency both directly and indirectly through Kepco yields despite the fact that the off- take agreement is USD based. Also, Ras Gas bond holders acquire significant Korea country risk which, in light of the energy dependence of the Korean economy, makes sense. Finally, we find significant evidence for "guilt by association" effects related to financial contagion. As returns in European, Asian, Middle Eastern and Latin American emerging debt markets fall Ras Gas spreads are predicted to widen considerably. In particular, the impact of contemporaneous and past events in European emerging debt markets stands out. This responsiveness reflects the turmoil in global debt markets following the Russian financial crisis of summer and fall 1998 which heavily affected other market segments. Similarly, Ras Gas spreads exhibit a delayed reaction to Asian, Middle Eastern and Latin American emerging debt markets with only a few but highly significant lagged returns accounting for most of the explanatory power. The paper is organized as follows. In the next section we provide background information on the Ras Gas project and its web of inter-related contracts. Section 3 describes the project- specific sources of contractual incompleteness and risk factors. Section 4 contains a description of our variables and how they are related to the various risk factors. In Section 5, we summarize the results of our empirical analysis. Next, we carry out a decomposition of the risk factors' direct and indirect effects. Section 7 concludes. We relegate all tables to the Appendix. 2. Ras Gas: Project and Contract Design The Ras Laffan Natural Liquified Gas Company Limited (Ras Gas) is a joint venture between the Qatar General Petroleum Corporation (70%) and Mobil Corporation of the US (30%). Ras Gas has the right to develop lOm tons of liquified natural gas (LNG) annually from Qatar's North Field, the world's largest unassociated natural gas field with about 380bn cubic feet of confirmed recoverable reserves (about 9% of global gas reserves). To this end, Ras Gas has constructed a 5.2 MMTA (metric tons per annum) liquification facility in Ras Laffan consisting of two 6 identical LNG processing trains, offshore drilling platforms, storage facilities, pipelines and port loading facilities. Construction was completed in summer 1999 and cost approximately USD 3.4b including interest during the initial contract phase. As is customary in such projects most of the output was sold through long-term supply contracts even before construction started. The principal off-taker, Korea Gas Corporation (Kogas), is a state-owned company whose shareholders include the Republic of Korea (50%), Korea Electric Power Corporation (Kepco: 34.7%) and regional governments (15.3%). As such, Kogas shares its credit rating with the sovereign rating of Korea as does Kepco, which is currently being privatized. Most of the Ras Gas LNG bought by Kogas, who has a legal monopoly of gas sales and purchases in Korea, is currently re-sold to Kepco for electricity generation. Kepco, in turn, is about to double its existing LNG powered electricity generation in the next years which makes it Ras Gas' de facto off-taker with only the balance going toward residential and commercial gas supplies. Kogas-Kepco currently accounts for more than 75% of the projects expected revenue. State of Qatar Mobil Corp. 100% 100% Qatar General Gas, Inc. Republcof Korea Electric Regional Petroleum Corp. b M s c Power Corp. Governments 70% 30% 50% 34.7% 15.3% Ras Laffan Liquefied Natural Gas Co. Ltd. Korea Gas Corporation I | l~~~~~~~~~~~~~~~I Onshore EPC Platforms EPC Pipelines EPC Korea Electric Contract Contract Contract Power Corp. JGC Corp./ The |McDermott-EPTM| M.W.Kellogg East, Inc./ Chiyoda Saipem Sp.A. Company Corp. Exhibit 1. Ras Gas Project Participants Taken together the two sales and purchase agreements (SPAs) with Kogas account for a total of 4.8 MMTA of LNG. Beginning in the last quarter of 1999 Kogas will receive LNG shipments for 25 years on a take-or-pay basis. Under such an agreement the purchaser (Kogas on behalf of Kepco) is obligated to pay for the gas whether or not they take delivery. Hence, Kogas can make a cash payment in lieu of delivery which are credited against charges for future deliveries. The off-take agreement effectively indexes LNG prices to world crude oil prices and contains minimal off-take quantities. As part of the second off-take agreement, which doubled the quantity of output sold and thus yielded significant economies of scale to Ras Gas, the initial off-take agreement was significantly modified. In particular, the parties replace a price guarantee (floor price for LNG) with a minimum off-take guarantee (minimum quantity to be purchased 7 and its evolution over time). Exhibit 2 summarizes the final capital structure and construction budget. Ras Gas Two-Train Financing Uses of Sources of Funds (USD millions) Funds (USD millions) Capital cost 2, /5U Bonds 1 ,20U0 Finance & Bank/ ECA 1,400 Interest during Construction 650 Equity 800 I otal project 1 otal project Cost 3,400 Cost 3,400 Exhibit 2. Ras Gas Construction Budget and Capital Structure The preceding discussion reveals that a project such as Ras Gas represents a nexus of contracts contingent on each other. For instance, debt financing and take-off agreements are intimately linked since the latter serves as security for the former. The corresponding financial transactions reflect this reality. They attempt to find an optimal balance of the various parties' rights and obligations and serve to distribute risks to the entities best suited to bear them. The following figure summarizes the contractual arrangements and cash flow structure of the Ras Gas project making the nexus of contracts transparent in terms of the precise cash flows mechanics. In order to minimize moral hazard in payments Kogas will make payment for shipments directly to an off-shore trust account whose administrator then services public and private debt and remits the balance to Ras Gas for operational expenses and dividends. Other ~~~~~~~~Korea Gas Corp. Cnest Buyers numer Buyers ~xients 3 Payments Paymentr Royalties and Taxes' Offshore Account Debt Service' 5~~~~ ......... .......[3. r Dvi n , .14 Stae of Qatar | Mobil Corp. Bon Truste Facility Agent 100% 1100%ll LQGPC~~. ......... .......... -LMblQ ||Bodler|Banks | 70%30 Operating CoSts 2 ERas Laffan LNG Co. Ltd. Note: Superscript numbers indicate payment order of priority. Exibit 3. Ras Laffan Contract and Cash Flow Structure 8 3. Ras Gas, Kogas and Kepco: Contractual Incompleteness and Risk Shifting A large scale project such as Ras Gas typically requires a huge up-front investments with a high degree of asset and relationship specificity. By their very nature, the necessary physical assets cannot readily be removed and utilized elsewhere.5 As a result, there is a danger that the company and its financial backers suffer opportunistic behavior such as unilateral renegotiation of contracts or the redefinition of property rights. In the absence of a well functioning legal system that is willing to define and enforce property rights and contractual clauses the physical assets - always subject to hold-up problems - are of limited value as security to investors. Hence, while the debtholders are, in theory, secured the location of the assets in Qatar and the lack of credible legal institutions (enforcement) renders them inadequate for creditor protection. Instead, the off-take agreement with Kogas provides the only security. However, the SPA suffers from contractual incompleteness that arises from the interplay of the investment characteristics, Kogas-Kepco's use of the LNG, and certain peculiarities of the world market for LNG. At the heart of the problem lies the non-existence of a spot market for LNG due to the lack of transportation capacity.6 Consequently, the parties ofien build dedicated vessels for LNG transportation tied to a specific project7 and, in order to protect the very specific investments in physical infrastructure, sign long-term off-take agreements. Hence, most LNG is sold through long-term supply contracts, which, in turn, severely hinder the emergence of spot markets. In their absence, the most important hold-up risk consists of breach of contract or unilateral renegotation by Kogas, the off-taker. These risks are directly passed through to debtholders: they are locked into the project and, hence, vulnerable to opportunistic and strategic behavior from the customer. Investors risk being the victims of what has been called the obsolescent bargain should the Korean customers not honor the off-take agreement (Klein and Roger 1994). Kepco's use of LNG fired power stations for peak load generation exacerbates the contractual incompleteness in the off-take agreement. These plants are the marginal ones meant to smooth out demand cycles and most likely to be shut down in case of weakening electricity demand. While the Asian financial crisis does not seem to have significantly affected the overall LNG market, electricity demand was significantly affected. In Korea, electricity demand, which C Contracting problems arising from relationship specificity, sunk costs, and the associated "hold-up" problem, were first described in other areas of economics by Klein, Crawford and Alchian (1978), and Williamson (1979, 1983). Three types of solutions have been proposed in the literature that balance incentives for ex-ante efficient invest- ments and ex-post trade efficient: (i) writing contracts with proper legal remedies in case of breach of contract (Shavell, (1980, 1984); Rogerson, (1984)), (ii) agreeing on a rule for the re-negotiation of contracts (Aghion, De- watripont and Rey, 1994), and (iii) writing option contracts (Noldeke and Schmidt, 1995). 6 With the availability of LNG tankers not tied to specific projects the nascent LNG market for immediate delivery is expected to develop into a full-fledged spot market over the next decade. ' Part of the off-take agreement with Kogas-Kepco stipulated the construction of landing, storage and regasification facilities in Korea as well as 7 to 8 LNG tankers, two of which have already been built and two more to be com- pleted each year between 2000 and 2002. 9 had been growing by 10% annually during the previous decade, declined in 1998 by about 3.6% and Kepco, using LNG fired power plants as peak load reserves, correspondingly reduced its LNG purchases from Kogas by as much as 22%. The much decreased demand for LNG by Kepco explains why Kogas had to defer about 10% of its 1998 LNG cargoes. However, electricity demand has recently picked up (8.1% increase in 1999) and while demand growth is expected to fall short of initial forecasts Kepco still plans to add about 20,000 MW of generation capacity including LNG fired power stations over the next years. The resulting bilateral monopoly between producer and customer now leads to an interesting contractual dynamic that directly impacts the financial backers of the project. Investors, for instance, negotiating over the terms of the contract after an investment has already been made, may be taken advantage of since there is no other use for their investment. Anticipating such a scenario, contracting parties are hesitant to make relationship-specific investments without adequate contractual protection. In the Ras Gas project, signing and renegotiating the SPA with Kogas provided ex ante security, all other things being equal. However, the Asian financial crisis and ensuing recession in Korea forcefully brought out that the parties to such a project can not anticipate on all possible contingencies.8 At the same time, from the off-taker's perspective, commitment to a long-term contract such as a Sales and Purchase Agreement (SPA) poses the difficulty of not knowing at the time of contracting the future value of the output, i.e., the availability of re-contracting and alternative suppliers. She faces a loss in the real option value of its flexibility. Conversely, a project such as Ras Gas faces the danger that their dominant buyer walks away from the contract as alternative sources of supplies are more cheaply available elsewhere than through the SPA. Hence, the off- taker has always an implicit real option through breach of contract. This scenario came about in 1998 when Kogas started to defer LNG cargoes as Kepco diminshed its purchases in response to reduced electricity demand. While the renegotiated Ras Gas-Kogas SPA included deferral options, meant to pre-empt breach of contract, the risks from their exercise is directly passed on to investors and, especially bondholders, given the total lack of an LNG spot market. Since the assets are highly relationship specific through the construction of dedicated LNG terminals and storage facilities, the necessary project specific tanker fleet and the absence of a spot market, long-term supply contracts such as the 25 year SPA between Ras Gas and Kogas seem to offer a remedy. However, as we have argued the remedy is far from perfect since there always exists the threat of unilateral renegotiation or breach of contract. These incomplete contract risks are passed on to debtholders who presumably price them into their investment decision. Economic circumstances may change so that the absence of enforceable, complete contracts means that investors have to reassess their initial financing decisions. Here, Ras Gas affords the unique opportunity to analyze the price (yield) evolution of its outstanding global bonds to identify the sources of contractual incompleteness and how they shape project risk, a topic to which we now turn. 8 In the absence of sufficient contractual protections, the outcome is likely to be an inefficiently low investment, often referred to as the under-investment phenomenon (Hart and Moore, 1988). 10 4. Project Risk Factors and the Ras Gas Bonds The two Ras Gas bond issues proved to be in very high demand. Despite increasing the issue size, they sold out on the first offering day (December 16, 1996) and were twice over-subscribed. The long bond due 2014 had a total size of USD 800m and was priced at an issue yield of 8.2940% or 187.5 basis points above 15 year US Treasury bond yields (interpolated). Issued as both a global and 144A (foreign) bond it was targeted at institutional investors with strong international demand (20% international, 80% US based investors). The bond trades actively which permits us to use its spread over US Treasuries to gauge market perceptions of changes in Ras Gas' prospects and, hence, its riskiness. From an analytic perspective, Ras Gas offers the advantage of few, clearly identified risk factors whose effects make themselves felt through the nexus of contracts underlying the company. The deeper root of Ras Gas very particular risk profile is the sales arrangement with one dominant customer. Hence, we would expect that events affecting the economic prospects of Korea, Kogas and Kepco should be reflected in Ras Gas bond yield spreads over US Treasuries. In the sequel, we discuss several key risk variables that have a direct incidence on the project's economic and financial success and, therefore, should determine the riskiness of Ras Gas bonds in terms of yield spreads. The first variable behind the postulated chain of contractual risks is the oil price, which effectively determines the price for LNG through energy equivalent conversions. Electricity generation from LNG becomes non-viable at an oil price in excess of USD 23.00. As a direct consequence of the project's nature and contract structure the oil price is one of the principal state variables affecting Ras Gas. In the sequel, we will use the logarithm of the price of Brent (BRENT) - one of two commonly used crude oil reference standards for LNG pricing9 - to analyze the incidence of world energy prices on the riskiness of Ras Gas.' We also decompose the BRENT variable into three bands (below USD 14, between USD 14 and 23, above USD 23) based on the project's economics. S&P reckons that below USD 14 per barrel various guarantees by Mobil will have to be used to insure full debt service coverage while above USD 23 electricity generation from LNG becomes uneconomical. The contractual provisions of the off-take agreement permit us to separate demand from price risk because Kogas, by and large, has committed to buying a fixed amount of output annually." Hence, demand risk essentially translates into breach of contract risk. As Kogas does not have any publicly traded global bonds outstanding and is reselling most of its Ras Gas purchases to Kepco, we take the yield spread of the Kepco 7.75% global bond maturing in April 2013 (KORELES) over 10 year US Treasury yields to measure the economic and financial prospects of the LNG buyer as assessed by capital markets. The fortunes of Kepco and its bonds 9 Gas prices turned out to be non-significant which is not really surprising given that about 0.12 metric tons of LNG are priced as one barrel of crude oil. ' Diagnostic testing reveals that logarithms offer superior fit over levels for several of the independent variables. " Kogas can vary gas shipments by deferring about 5% per annum up to a total of 10% which must be paid for within 5 years whether Kogas accepts delivery or not. 11 are closely tied the Korean economy so that we expect the bond yield to reflect both idiosyncratic Kepco factors (operational, regulatory and financial risks) and Korea country risk. However, even if the Korean economy does not experience any difficulties and electricity demand is strong Kepco might suffer operational or unrelated financial problems that might force it to breach the off-take agreement with Kogas and, ultimately, Ras Gas. Ras Gas' fortunes also depend on Korea's macroeconomic environment. Independent of oil price fluctuations the economic situation of Korea might deteriorate to a point where LNG sourced according to the contract becomes too expensive or is simply not needed as electricity and gas demand slump, e.g., in a recession. This risk is all the more relevant that 40% of LNG imports are primarily used for marginal (peak load) generation and that Kepco already reduced its LNG purchases by 22% in response to weakened electricity demand after the Asian financial crisis. Hence, a severe recession might cast serious doubts over Kogas-Kepco's ability to honor the contractual commitments. In addition to KORELES, the spread of the Kepco global bond, we include a second measures to analyze this effect: the logarithm of the Korea Composite Stock Index (KOSPI). In a similar vein, we also use KEPCO, the logarithm of Kepco's stock price, as a measure for Ras Gas' direct exposure to the Korean economy. A further risk factor is the credit quality of the off-taker which, in Kogas' case, might reflect both systematic changes in the Korean macroeconomic environment, the industry structure (i.e., loss of gas monopoly, privatization) or purely idiosyncratic risks. However, Kogas and Kepco share their credit ratings with the Republic of Korea, whose rating over the sample period has varied from AA- to B+ back to BBB. How important the credit quality of the off-taker is to debtholders can be seen from the fact that the Ras Gas bond convenants restrict additional SPAs to buyers rated A or better."2 We, therefore, construct a rating index (KRR) that reflects not only the changes in S&P credit ratings but also their magnitude. Exposure to foreign currency risk can drastically change the economic viability of a project, as evident from the South-East Asian experience. In Ras Gas' case, this risk appears to be of relatively minor concern as all revenues and costs accrue in USD. However, by the very nature of the off-take agreements, the customer still poses a very subtle and indirect currency risk. Both Kogas and Kepco generate their revenue in local currency so that an adverse currency movement (devaluation or depreciation of the Korean Won against the USD) might imperil their ability to honor the SPA. The Asian financial crisis was a stark reminder of this fact: as the Korean Won depreciated against the USD the effective cost of LNG to Kogas and Kepco doubled in local currency terms. Hence, exchange rate risk when borne by the off-taker has a tendency to transform itself into a credit risk. To untangle the two effects, we include KRW, the logarithm of the KRW-USD exchange rate, as an explanatory variable. Finally, we need to control for the effects of financial contagion and other "guilt by association" characteristics of financial markets. Events and particularly financial crises in emerging economies have a tendency to affect other, seemingly unrelated countries through the propagation of financial shocks via global capital markets. In the data these effects show up as 12 In spring 1997 Kogas, Kepco and Korea were still rated AA- satisfying the bond covenants for the second SPA. 12 correlated shocks. In order to better understand how these propagation mechanisms affect the Ras Gas project and its bond yields we use the JP Morgan emerging market bond regional indices, i.e., Asia, Middle East, Europe and Latin America. We summarize the direct and indirect effects via Kepco credit spreads of the various variables on Ras Gas yield spreads and their first differences in the following table: Dependent Variable Ras Gas Yield Spread Changes in RG Spread Effect Direct Indirect Direct Indirect Variable Description Coefficient Coefficient Coefficient Coefficient RGS(- 1) Lagged Ras Gas spread persistent BRENT (Log) oil price indeterm. + indeterm. + BRENT<14 (Log) oil price below USD 14 insign. + indeterm. + BRENT: 14-23 (Log) oil price: USD 14 to 23 indeterm. + + + BRENT>23 (Log) oil price above USD 23 indeterm. + indeterm. + KORELES Kepco yield spread + + KORELES(-I) Lagged Kepco spread insign. persistent persistent KEPCO (Log) Kepco stock price insign. - indeterm. KRW (Log) Korean Won FX rate + + insign + KOSPI (Log) Korea Stock Price Index insign insign KR rating Korean country rating index + + insign. + ASIA Emerging debt returns Asia EUROPE Emerging debt returns Europe LAT Emerging debt returns Latin Am. MEA Emerging debt returns Middle East Exhibit 4. Explanatory Variables S. Empirical Analysis of Interacting Risks: Breach-of-Contract Factors In the sequel, we investigate how risk factors are transmitted through the nexus of contracts represented by the take-off agreement and bond covenant to bondholders. Our analysis relies on daily data covering the period from January 1997 to March 2000 that is drawn from various sources, mainly Bloomberg, IDC and Baseline. Whenever we found missing observations we cross checked the time series with other news sources and filled in the missing data or, if this was not possible, deleted the observation. Depending on the included variables we have 725 to 750 observations before taking lags. Our dependent variable is the spread of the 17 year Ras Gas bond'3 yield over the 10 year US Treasury yield. The explanatory variables are the above mentioned risk factors affecting the contractual relationships at the heart of the Ras Gas project. '3 The bond is highly liquid and widely traded contrary to the 10 year one and, therefore, constitutes a much better measure of investor and market assessment of the project's prospects. 13 Ras Gas Spreads Ras Gas Spread KEPCO Spread a O Prices 12 - 35 10 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~30 8 As the pecedig plo sugests he daa ishighl volaile. able1 in he Apendi20 2J I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 113/97 4/3/9r7 7/3/97 10/3/97 1/3/98 4/3/98 7/3/98 10/3/98 1/3/99 4-3-9-73-9-1-39 13 00 Exhibit 5. Ras Gas bond spreads, KEPCO bond spreads and Oil Prices As the preceding plot suggests the data is highly volatile. Table 1 in the Appendix contains summary statistics for the entire sample period that further highlight this point. The correlation matrix indicates that some of the variables are highly collinear, a point which we will address in our statistical specifications. The preceding diagram clearly indicates the two defining events during the sample period: the Asian financial crisis that hit Korea in December 1997 and the Russian financial crisis that shook emerging debt markets in August 1998. In the attempt to uncover the complex dynamics underlying the time series in Exhibit 5 and explain their interactions from a nexus of contracts view of the firm, we estimate variants of the following empirical relationship in the sequel: RGS, = #o + lax,RGS,, +ABRENT, + 12I,KORELES,, 0SI!SL Osls0!L +* 3KEPCO, +/4 KR W, + 65KOSPI, + /6KRR, + [Y,-MEA,, +Y2 ,ASIA,, +73 ,EUR,, +y4 ,LAT, ,]+* where RGS, is the spread of the Ras Gas bond over 10 year US Treasury yields, BRENT, the logarithm of the Brent blend oil price index, KORELES, the spread of the 2013 7.75% Kepco global bond over 10 year US Treasury yields, KEPCO, the logarithm of the Kepco stock price, 14 KR W, the logarithm of the Korean Won - US Dollar spot rate, KOSPI, the logarithm of the Korea Composite Stock Price Index, KRR, a credit rating index for Korea, Kepco and Kogas, and MEA,,ASIA,,EUR,,LAT, the continuously compounded daily returns of the JP Morgan regional total return indices in USD for emerging markets in the Middle East-Africa, Asia, Europe and Latin America, respectively. In terms of estimation strategy, we start with the two key variables depicted in Exhibit 6, the oil price and Kepco bond yield spreads, and successively add explanatory variables to the regression equation estimated by OLS. First, we focus on the supply contract specific variables of oil price and Kepco bond yield. Next, we will add contemporaneous and lagged emerging debt market returns to analyze systematic effects such as spill-overs and contagion. We then include variables related to Korean country risk before estimating models with all variable categories. It turns out that for any number of variables and their lags the residuals exhibit high serial correlation. Adding a lagged dependent variable generally fixes this problem as evidenced in Table 2 (Durbin and Watson d statistic). Given the high frequency of the data, it comes at no that daily credit spreads exhibit a very high degree of persistence: the coefficient on lagged spreads is close to unity. Tables 2 to 7 in the Appendix report the best fit estimation results after specification testing eliminated highly insignificant control variables. As we would expect, Ras Gas spreads vary positively with Kepco credit spreads: the second specification in Table 2 indicates that a 100 basis point increase in Kepco spread widens the Ras Gas spread by about 2 basis points. Hence, the perceived riskiness of the off-taker feeds through immediately to Ras Gas yield spreads which is consistent with the nexus of contract view of the firm. In light of the contractual incompleteness of the off-take agreement, credit migration by the dominant output buyer affects market risk perceptions of the firm in terms of yield spreads: off-take risk is shifted from Ras Gas owners to its bondholders as predicted by debt agency theories. Including the lagged Kepco spread reveals that bondholders overreact to news about the off-taker. Initially, Ras Gas spreads widen by 15.5 basis points for every 100 basis point increase in Kepco spreads. However, the next day they narrow by 13.6 basis points (coefficient on the lagged Kepco yield spread) all other things being equal. A comparison between the second and third regressions reported in Table 2 shows the 2 basis points spread widening is the net reaction over a two day period. Since further lags of the Kepco spread come out insignificant we interpret the change of sign between the initial and lagged reaction as a sign of over-reaction: Ras Gas bondholders over-react to news about the off-taker. The results in Table 2 show that this pattern is stable across all specifications. One possible explanation for the market's apparent overreaction to changes in the off-taker's credit spread might lie in Kepco's importance for debt service. Since most revenue comes from one single source, any deterioration in its prospects calls into question the financial viability of Ras Gas. Regarding oil prices we find that the coefficient is marginally significant at best. Independent of the set of control variables energy prices do not seem to significantly determine Ras Gas yield spread levels and, hence, the bond's riskiness as priced in global markets. It is 15 quite remarkable that markets seem to view the oil price as irrelevant because it determes the output price and, ultimately, the revenue that Ras Gas will generate from LNG sales. In the sequel we will further investigate this puzzle by decomposing the impact of BRENT into its direct revenue and indirect credit migration effect via Kepco and looking at different oil price bands. Turning to the impact of emerging markets on Ras Gas credit spreads we include contemporaneous and up to five lags of daily regional emerging debt market returns. Successively testing for their statistical significance and eliminating the most insignificant variables leads to the emerging market propagation pattern reported in specifications 4 and 5 of Table 2. As one would predict, Ras Gas spreads vary negatively with emerging debt market returns. The remaining significant lags hint at the time structure of the shock propagation mechanism behind this contagion effect. While Ras Gas spreads show a particularly strong contemporaneous reaction to European emerging debt markets (dominated by Eastern European countries and Russia in particular) the other debt markets' impact is delayed. European and Latin American debt returns are particularly persistent: 3 and 5 days later they have their most significant effect. Adding Korea specific variables (specifications 6 to 8 in Table 2) reveals that the Korean rating index KRR is insignificant but the Kepco price KEPCO significant at the 1% level. Concerning exchange rate exposure as measured by the exchange rate KRW we find that coefficient comes out negative. The negative association of Ras Gas spreads and the Korean Won - US Dollar exchange rate is puzzling: one would have expected that breach of contract and default risk increases as the Won depreciates, i.e., KRW increases. Instead, a depreciation of the Won seems to reduce Ras Gas credit spreads. The last regression reported in Table 2 puts together the three components of Ras Gas bond riskiness: variables representing contract risk, Korean exposure and emerging debt market spill-over effects. The net impact of Kepco yields is still 2 to 3 basis points, the significant Korea variables do not change in either identity or magnitude and the same is true for the emerging debt market contagion structure. Given the very high degree of persistence in Ras Gas yield spreads as evidenced by a coefficient of about 0.98 we next look at the hypothesis that the spread might follow a random walk by testing for unit roots. Dickey-Fuller and Augmented Dickey-Fuller tests show, that we can not reject the hypothesis that Ras Gas spreads follow a random walk (see Table 3). The same is true for Kepco spreads and Brent oil prices. Similarly, we can not decisively reject co- integration between Ras Gas spreads and Brent or Kepco spreads by Engle-Granger and Augmented Engle-Granger tests (see Table 3). However, error correction models of the Ras Gas credit spread do not perform well. Given that Kogas and Kepco buy 75% of Ras Gas output and but that Ras Gas is only one among many suppliers to Kogas - Kepco (and the marginal at that) the absence of a statistical equilibrium relationship comes as no surprise. Kepco influences Ras Gas in equilibrium but not the reverse. 16 Since we can not easily distinguish between persistence and unit root in the dependent variable we replicate the preceding analysis for changes in Ras Gas credit spreads, i.e., taking first differences of the dependent variable to address possible non-stationarities: ARGS, = /f0 + /A1BRENT, + E 2,-_KORELES,_J + /J3KEPCO, + 84KRW, + /5KOSPI, + ,/6AKRR, + X[Y1,,MEA,,I + y2,ASIA,, + y3,EUR,, +y3 LAT,-,]+ , 0SL The results in Table 4 and diagnostic testing reveal that first differences in the independent variables leads to inferior statistical performance so that we keep them in levels except for the rating variable KRR. We also experimented with using Kepco yields rather than the credit spreads but the results are virtually identical. Table 4 reports the results for best fit regression specifications. The results confirm our earlier findings from the level analysis that markets price output buyer related risk factors and that, therefore, the project shifts risk to bondholders. With spread changes as dependent variable, the oil price's logarithm becomes significant at the 10% or 5% level whereas the Won exchange rate becomes insignificant. The emerging debt markets contagion patterns are very similar and the over-reaction result also obtains: again, the coefficients on the contemporaneous and lagged Kepco spreads change in sign. 6. Direct and Indirect Effects of Risk Factors To the extent that both Ras Gas and Kepco fall into the same emerging markets and energy bond categories, we would expect that a common set of factors affect both their credit spreads. In order to distinguish the direct, contract related effects from the indirect, common factor induced ones we next specify a simultaneous equation model of the Ras Gas and Kepco yield spreads. We use the contractual arrangements to derive restrictions on the system's parameters. Since the off-take agreement effectively collateralizes the Ras Gas bonds, their credit spread critically depends on the prospects of Kepco, the ultimate buyer of its LNG output. The reverse does not hold. Kepco has a well diversified portfolio of fuel types and sources and, in any event, LNG is used for peak- load electricity generation. In addition, we conjecture that the Korean control variables primarily influence Ras Gas bond yields and credit spreads through Kepco's exposure to the Korean economy: the Ras Gas' Korean exposure acts through Kepco. We estimate the simultaneous equation system by Maximum Likelihood methods. We can then carry out diagnostic tests to determine which risk factors affect Ras Gas bonds directly and which ones operate indirectly through their incidence on Kepco's financial health. Under the restriction that Kepco spreads influence Ras Gas ones but not the reverse we arrive at the following specification:14 14 Ras Gas spreads in the Kepco equation come out insignificant as expected. 17 RGS, = fi, + Ya-,,RGS,,+ ,01, IBRENT, + Y. X2,-,,KORELES,, 0:IL Os1sL + 331 KEPCO, +/J4, KRW, + Pl 1KOSPI, + 6l KRR, + Zk[YI,IMEA,1 + Y2,,ASIA,, + y3,1EUR,, + Y4 IILAT_,J + 0IL KORELES, =f/02 + Al12BRENT, + E132.,2KORELES,, OIL + /32KEPCO, + ,42KR W, + /852KOSPI, + /862KRR, + E >l 2MEA,, + r2 ,2ASIA,_, + Y3,-12EUR,-, + r4,-,2LATI, ] + 62 0IL As the lagged Kepco spreads are insignificant in the Ras Gas spreads equation we drop them in the sequel. Similarly, all emerging debt market returns are insignificant in the KORELES equation as evidenced by the first simultaneous equation specification in Table 5. The results show that the oil price impacts the riskiness of Ras Gas through its effect on the financial position of Kepco rather than directly: while BRENT is insignificant in the RGS equation it is highly significant in the KORELES equation. We interpret this outcome, which holds for the full set of explanatory variables as well as sub-sets (see specification 3 and 4 in Table 5), as evidence that investors discount the direct revenue effect of oil prices on Ras Gas. Instead, they are more concerned about the impact of energy prices on Kepco's financial health. An increase in crude oil prices by USD 2.72 translates into a widening of Kepco spreads by about 18 basis points so that the indirect impact on Ras Gas credit spreads is about 0.66 basis points (systems 3 or 4 in Table 5). A comparison with equation 3 in Table 5, which also includes BRENT in the Ras Gas spread equation, reveals that the direct effect on Ras Gas spreads is 0.24 basis points. However, it is so highly insignificant that capital markets apparently disregard the oil price revenue effect in the pricing of Ras Gas default risk. In terms of the emerging markets shock propagation mechanism, the surviving significant emerging debt market returns confirm our earlier lag structure. Europe and the Russian financial crisis had the most lasting impact on investor perceptions. However, Kepco credit spreads seem to be unaffected by emerging debt markets. As we conjectured, the country risk variables primarily affect Kepco's credit spread through their impact on Kepco's economic and financial fortunes. In terms of direct effects on Ras Gas, only Kepco's share price KEPCO as an indicator for Korea's electricity demands together with the Won exchange rate come out significant. As economic prospects improve so does the risk profile of Kepco as reflected in stock prices. Hence, its yield spreads and, ultimately, Ras Gas credit spreads should decrease. The second significant country risk variable is again the Korean Won - USD exchange rate. Equation systems 3 or 4 reveal an interesting pattern: the RGS equation in each of the two specifications confirms the previously identified puzzle that a weakening Won on average directly decreases market risk perceptions as spreads decline. However, the indirect impact via Kepco yields in the KORELES equation clearly exhibits the conjectured currency exposure effect in terms of a positive KRW coefficient. As the Won depreciates Kepco yields rise as servicing Kepco's foreign debt and its oil, coal and gas purchases all denominated in USD become more 18 expensive. Consequently, Kepco's financial position deteriorates, increasing its riskiness and, hence, requiring higher yields to compensate bondholders for more risk: Korean Won exposure clearly is an indirect exposure. The direct, negative impact on the yield spread might now be explained in terms of improved economic outlook and the high correlation of the Won with the country rating. Alternatively, a weakening Won might lead to higher exports so that an economy as energy dependent as Korea's would require more gas and electricity which decreases the demand, i.e., direct breach of contract risk all other things being equal. In Table 6, we replicate the simultaneous equation analysis for RGS and KORELES in first differences. As in the previous analysis (see Table 4) levels of the explanatory variables perform better than their differences with the exception of the rating variable KRR. When the Kepco yield spread equation is specified in levels we obtain results that, by and large, mirror the results in Table 5. However, the insignificant coefficient on KORELES in the ARGS equation indicates that the model is misspecified. Repeating the estimations with AKORELES (specifications 3 and 4) reveals several new effects. First, changes in Kepco yield spreads respond to Asian and Latin American debt markets while levels do not (see Table 5). Second, energy prices are marginally less significant in the AKORELES than the ARGS equation so that direct and indirect impact are very comparable. Finally, the only significant Korean variables are KOSPI and the rating index, both for changes in RGS and KORELES. As Korea's economic prospects improve as measured by a higher stock price index, both Ras Gas and Kepco credit spreads change by less. In order to further investigate the insignificance of the oil price variable BRENT we next compose it into three distinct bands: oil prices under USD 14, between USD 14 and 23, and above USD 23. The choice of these bands corresponds to the contractual provisions and economics of electricity generation from LNG. Below approximately USD 14 per barrel, price support and debt service guarantees by Mobil to the bondholders will be triggered. Above USD 23, the electricity generation from LNG becomes uneconomical so that breach of contract risk escalates at such prices all the more that the LNG fired power plants are Kepco's marginal ones. As a result, we split the oil price's logarithm into three explanatory variables in all our estimations: A BRENT, = 6, lIBENT,:514)BRENT, + g2 1{14, + AIIBRENT + E /32, KORELES,_, + /3KEPCO, + J4KRW, + J5KOSPI, + /6KRR, OS'ISL OS/SE + Z ,-IMEA4l +y2,_,AS14_- +Y3, EUR;, +74,ILAT I-1]+6, where the dependent variable RGS is the spread of the Ras Gas bond over 10 year US Treasury yields, BRENT the logarithm of the Brent blend oil price index, KORELES the spread of the 2013 7.75% Kepco global bond over 10 year US Treasury yields, KEPCO the logarithm of the Kepco stock price, KRW the logarithm of the Korean Won - USD spot rate, KOSPI the logarithm of the Korea Composite Stock Price Index, KRR a rating index for Korea, and MEA, ASIA, EUR, LA T the continuously compounded daily returns of the JP Morgan regional total return indices in USD for emerging markets in the Middle East-Africa, Asia, Europe and Latin America. Specification 1 2 3 4 5 6 7 8 Dep. Variable RGS RGS RGS RGS RGS RGS RGS RGS Variable Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient P-Value P-Value P-Value P-Value P-Value P-Value P-Value P-Value Constant 1.9684 -.1300 -.0980 0.01273 -.1132 .8682 .6125 .4158 .0004 .0374 .1011 0.1515 .0488 .1637 .0115 .0778 RGS(- 1) .9755 .9819 0.99812 .9834 .9679 .9650 .9817 .0000 .0000 0.0000 0.0000 .0000 .0000 .0000 BRENT -.3427 .0424 .0309 .0368 -.0089 .0038 .0348 .0571 .0340 .1064 .0462 .7345 .8518 .0584 KORELES .5972 .02496 .1558 .1499 .1631 .1609 .1523 .0000 .0000 .0000 0.0000 .0000 .0000 .0000 KORELES(-I) -.1365 -.1312 -.1230 -.1213 -.1282 .0000 0.0000 .0000 .0000 .0000 KEPCO .0704 l ___________ __________ .00 16 _________0 KRW -.1934 -.1248 -.0762 l ____________ __________ .0219 .0006 .0207 KOSPI .0717 .0238 KRR .0050 .3450 EUR -2.13609 -2.0387 -2.0053 -1.9789 -1.9841 0.0007 .0005 .0005 .0006 .0006 LAT -2.27699 .0129 LAT(-1) -2.09425 -1.5725 -1.5412 -1.5594 -1.5663 0.0218 .0624 .0660 .0621 .0626 MEA(-2) -3.09962 -2.7622 -2.8062 -2.8438 -2.8012 0.0040 .0058 .0048 .0042 .0051 EUR(-3) -3.07837 -3.2586 -3.3701 -3.2883 -3.2260 0.0000 .0000 .0000 .0000 .0000 LAT(-5) -2.57983 -2.1537 -2.0545 -2.1006 -2.1405 0.0050 .0114 .0151 .0127 .0116 Obs. 724 724 724 719 719 719 719 719 Adj. R2 _.63995 .99558 0.99596 0.99569 .99631 .99636 .99638 0.99633 DW d Stat. .02340 1.90390 2.00086 1.89391 2.05283 2.05791 2.05229 2.06149 26 Table 3: Unit Roots and Cointegration Testing for unit roots and co-integration we appeal to Dickey-Fuller, Augmented Dickey-Fuller, and Augmented Engle-Granger tests of the form, respectively, where in the AEG test case the residuals are drawn from the corre- sponding co-integration equation: Ay, = 8, + It- )y,-, + 6,, A a, + (at - )_+ u, where the dependent variabley is either RGS, the spread of the Ras Gas bond over 10 year US Treasury yields, or KORELES, the spread of the 2013 7.75% Kepco global bond. The (one-sided) simulated asymptotic critical values for the r statistic for unit root tests under the null hypothesis are drawn from the Davidson and McKinnon (1993). F Dependent ARGS, ARGS, AKORELES, AKORELES, Residuals: Residuals: Variable RGS RGS KORELES BRENT Variable Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient r -Value r -Value r -Value r -Value r -Value r -Value Constant .0097 .0099 .0277 .0318 RGS(-1) -.0027 -.0028 -1.076 -1.126 DRGS(-1) .0846 KORELES(- 1) -.0072 -.0085 -1.700 -2.029 DKORELES(- 1) .2079 E(-1) -132 -.0041 -2.262 -1.212 _ DE(- 1) .0671 .0940 | Crit. value: I% -3.43 -3.43 -3.43 -3.43 -3.90 -390 Crit. value: 5% -2.86 -286 -2.86 -2.86 -3.34-3.3 Crit. value: 10% -2.57 -2.57 -2.57 -2.57 -3.04 -3.04 Crit. value: 97.5% 0.24 0.24 0.24 0.24 -0.30 -0.30 Obs. 724 723 724 723 723 723 Adj. R2 .00022 .00596 .00261 .04444 .00931 .00885 DWdStat. 1.83052 2.01473 1.58662 1.92602 1.98719 2.01310 27 Table 4: Changes in Ras Gas Spreads ARGS, = A0 + AiBRENT, + , ,2_,KORELES,, + /3,KEPCO, + ,64KRW, + ,36KOSPI, +f6AKRR, + E,',ME4, ,ASIA _,y3 ,EUR,P +3,LAT1,,]+, where the dependent variable ARGS is the first difference of the Ras Gas bond spreads over 10 year US Treasury yields, BRENT the Brent blend oil price index in logarithms, KORELES the spread of the 2013 7.75% Kepco global bond and DKORELES its first difference, KEPCO the Kepco stock price in logarithms, KRW the Korean Won - USD spot rate in logarithms, KOSPI the Korea Composite Stock Price Index in logarithms, KRR a rating index for Korea, and MEA, ASIA, EUR, LA T the continuously compounded daily returns of the JP Morgan regional total re- turn indices in USD for emerging markets in the Middle East-Africa, Asia, Europe and Latin America. Specification 1 2 3 4 5 6 7 8 Dep. Variable DRGS DRGS DRGS DRGS DRGS DRGS DRGS DRGS Variable Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient P-Value P-Value P-Value P-Value P-Value P-Value P-Value P-Value Constant -.1319 .0008 .0189 .3312 .1364 -.1448 .0048 .0626 .0283 .8438 .6575 .2437 .2534 .0121 .2475 .8187 BRENT .0365 -.0064 .0560 .0617 .0423 .0594 .6701 .0336 .0058 .0233 DBRENT -.0185 -.9211 .9073 .9524 KORELES .1597 .1550 .1429 .1530 .1455 .0000 .0000 .0000 .0000 .0000 KORELES(-I) -.1512 -.1468 -.1386 -.1443 -.1326 .0000 .0000 .0000 .0000 .0000 DKORELES .1565 .1566 .1499 .0000 .0000 .0000 KEPCO -.0149 -.0659 .6858 .0290 KRW -.0406 -.0506 .2438 .1261 KOSPI -.0282 -.0506 .0730 .5983 .0118 .0642 DKRR .0380 .0384 .0397 .0302 EUR -2.0585 -2.0501 -2.057 .0005 .0006 .0005 LAT(-I) -1.8165 -1.8228 -1.755 .0331 .0346 .0391 MEA(-2) -3.1005 -3.0727 -3.058 .0022 .0027 .0025 EUR(-3) -3.1628 -3.1450 -3.195 .0000 .0000 .0000 LAT(-5) -2.4008 -2.4026 -2.317 .0052 .0057 .0069 Obs. 724 724 724 724 724 719 719 719 Adj. R2 .12408 .10983 0.1100 .13066 .13514 .20389 .18877 .20901 DW dStat. 1.98276 1.94776 1.9487 2.00345 2.00340 2.03588 1.99468 2.05788 28 Table 5: Simultaneous Equations: Ras Gas Spread Levels RGS, = ,8, + la_,,RGS,, + ,8 ,BREN7T + 62,KORELES, + 2Jj, ME4 , + Y2,1ASI4. + y3-,EU R, + y4 ,11LAT ,]+ -, 0!0!L 05lL KORELES, = /A.2 +fl2BRENT + 42 /,2KORELES,, +)632KEPCO, +642KRW +A4 KOSPI, + /3.AKRR +e,2 where the dependent variable RGS is the spread of the Ras Gas bond over 10 year US Treasury yields, BRENT the Brent blend oil price index, KORELES the yield on the 2013 7.75% Kepco global bond, KEPCO the Kepco stock price in logarithms, KRW the Korean Won - USD spot rate in logarithms, KOSPI the Korea Composite Stock Price Index in logarithms, KRR a rating index for Korea, and MEA, ASIA, EUR, LA T the daily returns of the JP Morgan regional total return indices in USD for emerging markets in the Middle East-Africa, Asia, Europe and Latin Amer- ica. We use Maximum Likelihood estimation (asymptotically equivalent to Generalized Least Square). Specification 1 2 3 4 Dep. Variable RGS KORELES RGS KORELES RGS KORELES RGS KORELES Variable Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient P-Value P-Value P-Value P-Value P-Value P-Value P-Value P-Value Constant -.0962 -.6018 -.1037 .3452 .5383 .6506 .5397 .6536 .1072 .2639 .0832 .5649 .0288 .2924 .0193 .2894 RGS(- 1) .9785 .9782 .9653 .9642 .0000 .0000 .0000 .0000 BRENT .0335 .1276 .0359 .1883 .0024 .1816 .1796 .0809 .0110 .0613 .0009 .9069 .0014 .0013 KORELES .0206 .0210 .0370 .0373 .0000 .0000 .0000 .0000 KORELES(-1) .9586 .9446 .9469 .9469 .0000 .0000 .0000 .0000 KEPCO .2191 .0625 .2382 .0633 .2380 l== ______ .0047 .0047 .0023 .0012 .0023 KRW .2696 .2636 -.1090 .2195 -.1083 .2193 .0003 .0003 .0028 .0034 .0024 .0034 KOSPI -.2354 -.5073 -.5145 -.5137 .0000 .0000 .0000 .0000 DKRR .2431 .0717 .2898 .0695 .2899 .0000 .0000 .0000 .0001 .0000 EUR -2.1776 -1.3301 -2.0705 -1.9705 -2.0013 .0003 .3363 .0004 .0006 .0005 ASIA(- 1) -.9312 -1.0715 -.92728 .0754 .3682 .0605 MEA(-2) -2.8146 .1324 -2.8355 -2.9679 -3.0781 .0070 .9554 .0045 .0025 .0017 EUR(-3) -3.3677 -.0418 -3.4154 -3.4000 -3.4034 .0000 .9762 .0000 .0000 .0000 LAT(-5) -2.2099 -.0261 -2.1757 -2.1589 -2.1215 .0127 .9897 .0105 .0096 .0111 Obs. 719 719 719 719 719 719 719 719 Log-Likelihd 554.8793 554.8793 576.4339 576.4339 591.7214 591.7214 589.9580 589.9580 DW d Stat. 1.9353 1.5691 1.9411 1.7267 2.0017 1.7684 2.0025 1.7684 29 Table 6: Simultaneous Equations: Ras Gas Yield Spread Changes ARGS, = ,B.+ ja,,RGS1, + &BRENT + 21KORELES, + XIj -,ME4, + r2-1,ASI, + y3 ,,EUR, + r4,,LAT_,] + e,, 023 .0540 .0196 .0722 .0717 .0285 .2239 .0914 .1538 .0915 .5399 .0245 .0193 .3834 .0035 .0090 .0463 KORELES .1551 .1510 .1582 .1415 .0363 .0000 .0000 .0000 .0000 .0000 KORELES(-1) -.1360 -.1124 -.1487 -.1317 .9470 .0000 .0000 .0000 .0000 .0000 DKORELES .1387 .0000 KEPCO .0713 .0626 .2376 .0347 -.0141 .0015 .0050 .0025 .2459 .8307 KRW -.1201 -.1017 .2243 .0035 -.0027 .0015 .0074 .0033 .9019 .9661 KOSPI -.5077 -.1034 -.0696 .0000 .0066 .4081 DKRR .0382 .0413 .0716 .2897 .0382 .2910 .0289 .0195 .0000 .0000 .0374 .0000 EUR -1.9365 -2.0216 -1.9817 .0008 .0006 .0005 ASIA(-I) -.92218 -1.0567 -.92898 .0653 .0365 .0601 MEA(-2) -2.9791 -3.2279 -2.9539 .0027 .0013 .0026 EUR(-3) -3.3475 -3.2658 -3.4137 .0000 .0000 .0000 LAT(-5) -2.1336 -2.3750 -2.1167 .0115 .0055 .0113 Obs. 724 719 724 719 719 719 719 719 Adj. R2/Log-L .99596 .99638 .12835 .21364 592.4560 592.4560 535.7771 535.7771 DW d Stat. 2.00686 2.05491 1.99629 2.05796 2.0029 1.7697 2.0080 1.7923 Policy Research Working Paper Series Contact Title Author Date for paper WPS2455 The Effects on Growth of Commodity Jan Dehn September 2000 P. Varangis Price Uncertainty and Shocks 33852 WPS2456 Geography and Development J. Vernon Henderson September 2000 R. Yazigi Zmarak Shalizi 37176 Anthony J. Venables WPS2457 Urban and Regional Dynamics in Uwe Deichmann September 2000 R. Yazigi Poland Vernon Henderson 37176 WPS2458 Choosing Rural Road Investments Dominique van de Walle October 2000 H. Sladovich to Help Reduce Poverty 37698 WPS2459 Short-Lived Shocks with Long-Lived Michael Lokshin October 2000 P. Sader Impacts? Household Income Martin Ravallion 33902 Dynamics in a Transition Economy WPS2460 Labor Redundancy, Retraining, and Antonio Estache October 2000 G. Chenet-Smith Outplacement during Privatization: Jose Antonio Schmitt 36370 The Experience of Brazil's Federal de Azevedo Railway Evelyn Sydenstricker WPS2461 Vertical Price Control and Parallel Keith E. Maskus October 2000 L. Tabada Imports: Theory and Evidence Yongmin Chen 36896 WPS2462 Foreign Entry in Turkey's Banking Cevdet Denizer October 2000 I. Partola Sector, 1980-97 35759 WPS2463 Personal Pension Plans and Stock Max Alier October 2000 A. Yaptenco Market Volatility Dimitri Vittas 31823 WPS2464 The Decumulation (Payout) Phase of Estelle James October 2000 A. Yaptenco Defined Contribution Pillars: Policy Dimitri Vittas 31823 Issues in the Provision of Annuities and Other Benefits WPS2465 Reforming Tax Expenditure Programs Carlos B. Cavalcanti October 2000 A. Correa in Poland Zhicheng Li 38949 WPS2466 El Nino or El Peso? Crisis, Poverty, Gaurav Datt October 2000 T. Mailei And Income Distribution in the Hans Hoogeveen 87347 Philippines WPS2467 Does Financial Liberalization Relax Luc Laeven October 2000 R. Vo Financing Constraints on Firms? 33722 Policy Research Working Paper Series Contact Title Author Date for paper WPS2468 Pricing, Subsidies, and the Poor: Ian Walker November 2000 S. Delgado Demand for Improved Water Services Fidel Ordofiez 37840 in Central America Pedro Serrano Jonathan Halpern