Lessons From Foreign Investment Disputes

Lessons From Foreign Investment Disputes

February 01, 2003

By Peter F. Fitzgerald

During the 1990s, there was a massive inflow of foreign investment into infrastructure projects in developing countries.  Host country government contractual support for these projects and political risk insurance both played important roles in inducing companies to invest in power plants, toll roads, port development, telecommunications projects and other infrastructure projects.  With political and economic crises erupting in Asia in the late 1990s and Latin America more recently, these foreign investment projects have come under stress and the value of both host government support and political risk insurance has been tested.  What lessons have been learned?

Host Government Support

In the 1990s, host governments, with limited budgetary resources and borrowing capacity to develop crucially needed infrastructure, embarked on a process of privatization and invited foreign companies to build, own and operate infrastructure projects.  In order to attract the massive investment needed, the risks associated with investing in the politically-volatile emerging markets had to be mitigated in a credible way.  Host government support, backed with political risk insurance and guarantees often provided by development and export credit agencies of the home country governments, served as the basis for the necessary risk mitigation.

To illustrate both the type of support typically provided, as well as the types of problems that ensued when political and economic crises erupted, the case of the investments made by CalEnergy Company, Inc. (now MidAmerican Energy Holdings Company) in two geothermal power projects — the Dieng and Patuha projects — in Indonesia is instructive.  The CalEnergy case has the added advantage of having been so well publicized that the facts are neither confidential nor in dispute.

The structure of each of the Indonesian projects sponsored by CalEnergy resembled the structure of most infrastructure projects developed in emerging markets in the 1990s.  CalEnergy formed a special-purpose Indonesian company to build, own and operate its geothermal power project.  This project company entered into a joint operating contract with Pertamina (the state-owned oil company that controlled the geothermal resource) and an energy sales contract with Pertamina and P. T. (Persero) Perussahaan Listruik Negara, or “PLN,” a corporation wholly-owned and controlled by the government of Indonesia.  Under the energy sales contract, the project company agreed to develop a geothermal power project and PLN agreed to a take-or-pay obligation pursuant to which it committed to buy all of the project company’s available electricity for thirty years at a price denominated in US dollars.  In addition, the Indonesian government, pursuant to an undertaking issued to the project company and signed by the Ministry of Finance, agreed that the Indonesian government would “cause PLN ... to honor and perform” its obligations under the energy sales contract.

This contractual structure — that is, an offtake contract with a state-owned utility being the principal asset and source of all revenues of a special-purpose project company owned by a foreign investor, plus a guarantee of that contract by the host government — was the structure that marked most infrastructure projects during this period.  It allocated the “commercial” risks of the project — for example, the risks of building and operating the project — to the equity investors and their lenders from the private sector, but allocated most of the “political” risks to the host government.  Since the revenues under the energy sales contract were effectively guaranteed by the government of Indonesia, the energy sales contract provisions provided the contractual basis for allocating certain political risks to the host government.  The risk of devaluation of currency, for example, was allocated to PLN through the denomination of payments under the energy sales contract in US dollars, and further allocated to the Indonesian government under its performance guarantee.  Similarly, the risk that changes in Indonesian law might adversely affect the project was also allocated to PLN and the Indonesian government through provisions in the energy sales contract requiring PLN to pay an increased tariff for the power purchased if changes in Indonesian law increased the cost of production of the power.

The contractual allocation of political risk to the host government was not in itself sufficient, however, to induce the commercial banking community to lend the massive amounts of money needed for infrastructure projects in emerging markets.  Foreign investors and their bankers came to the investment process in the 1990s with the experience of the Latin American debt crisis of the 1980s still fresh in their minds.  Many bankers had held host government guarantees in connection with loans to Latin American parastatal companies, only to come to the painful realization when foreign exchange crises erupted in the 1980s that, although Walter Wriston may have been technically correct when he remarked that governments do not go bankrupt (because there is no bankruptcy process to manage the situation when governments cannot pay), governments certainly do run out of money and fail to meet their obligations.

Political Risk Insurance

With the international bankers still in the process of recuperating from Latin American debt crisis losses, it was clear to all that it would be difficult to raise the massive amounts of money needed for a large infrastructure project on the basis of host government commitments only; a creditworthy backstop to those commitments would be needed.  Political risk insurance and political risk guarantees from development agencies (such as MIGA and OPIC), the export credit agencies (such as US Export-Import Bank) and, by the late 1990s, from the private sector political risk insurers (such as Sovereign, Zurich, AIG, and Lloyds), played a crucial role in providing this backstop.

The political risk insurers first had to grapple with two problems in determining how to provide effective political risk insurance coverage to an infrastructure project with a contractual structure such as CalEnergy’s.  First, political risk insurers traditionally covered the risk of expropriation — that is, the risk that the host country government would take action in violation of international law that would substantially deprive the investor of the benefits of its investment.  Typically, this type of coverage would only be triggered if the host government took some kind of affirmative action to interfere with a project.  However, the fundamental risk in 1990s infrastructure projects was the risk that the government would fail to pay when required to do so.  Thus, the question arose as to whether expropriation policies covered this risk.

Also, policies required that the host government’s action must violate international law.  Under international customary law, it is not illegal for governments to breach contracts.  However, according to one authoritative source used in the United States (the “restatement” of US foreign relations law), a state is responsible for any injury caused by its breach of a contract with a foreign national if the foreign national is denied adequate channels to address his complaint, or is not compensated for any breach determined (by pursuing appropriate channels) to have occurred.

Therefore, the mere breach itself is not a violation of international law, but a violation of international law occurs if the host government frustrates the working of the dispute resolution process (typically offshore arbitration) in a manner constituting a denial of justice, or if the dispute resolution process yields an arbitral award or judgment in favor of the foreign investor and the host government refuses to pay it.  Given the lack of clarity inherent in the international law violation requirement of expropriation policies, foreign investors sought political risk insurance coverage tailored to the precise risk posed by the infrastructure project contractual structure.

In responding to the needs of investors for coverage, OPIC led the way by modifying its traditional expropriation coverage to include a form of “disputes coverage,” a variant of coverage that OPIC had long provided to US contractors with contracts to provide goods or services to host governments and that essentially underwrote the integrity of the dispute resolution process under these contracts.  The disputes coverage provided that if the host government failed to pay under its contract, the insured obtained an arbitral award in its favor against the host government, and the host government failed to pay within an agreed time period, then OPIC would pay under its political risk insurance contract.  By agreeing to backstop a host government’s refusal to pay an arbitral award, OPIC served as a reliable and creditworthy backstop to host government obligations.  Other development agencies and political risk insurers followed OPIC’s lead and this structure proved to be a financeable way forward for the development of infrastructure projects.

The Indonesian Experience of CalEnergy

In connection with its investments in geothermal power projects in Indonesia in 1994, CalEnergy obtained expropriation and disputes coverage from OPIC and Lloyds.  Pursuant to its coverage, CalEnergy would need to demonstrate that it obtained an arbitral award against both PLN and the government of Indonesia, the government failed to pay for 90 days, and the government’s failure to pay was a violation of international law.

In the fall of 1997, Indonesia began a rapid descent into political and economic crisis, triggered by the Asian economic crisis, marked by a free fall in the value of the Indonesian rupiah, and resulting in the collapse of the Suharto regime. (When the energy sales contract was signed in 1994, the US dollar had a value of approximately 2,450 rupiah.  Within months of the onset of the Asian economic crisis, the exchange rate plummeted to approximately 15,000 rupiah to the US dollar.) Having attracted a surge of foreign investment over the previous four years into 27 independent power projects being developed in response to Indonesia’s growing electricity needs, in the fall of 1997 Indonesia was faced with the difficult prospect of having to pay the independent power producers for electricity in US dollars, notwithstanding that it could not pass on the large devaluation costs to the end users of the electricity.  This is not to mention the fact that the economic crisis had led to a reduced demand for power in Indonesia.  With a view to decreasing its foreign exchange outflows in the midst of this currency crisis, a series of presidential decrees were issued postponing, suspending or reviewing most of the independent power projects, including CalEnergy’s projects.

By the summer of 1998, the situation had deteriorated to the point where the president of PLN was quoted in the press as saying that, if possible, all independent power producer contracts would be cancelled and challenging the independent power producers to sue PLN.  With the completion of one of the CalEnergy projects in the summer of 1998, PLN failed to pay for power as required by the energy sales contract and PLN representatives informed CalEnergy that the plant was essentially being shut down for “political reasons.” The Indonesian government also failed to pay under its performance guarantee.

With a clear breach of the energy sales contract by PLN and a failure of the government to honor the performance guarantee, CalEnergy commenced arbitration against PLN in August and obtained awards against PLN of more than $570 million in May 1999. (The energy sales contract provided for United Nations Commission on International Trade Law, or “UNCITRAL,” arbitration in Jakarta.) CalEnergy then filed expropriation claims with OPIC and Lloyds claiming that the government of Indonesia had violated international law and deprived CalEnergy of the only asset of value to it —the energy sales contract.

Also in May 1999, CalEnergy commenced UNCITRAL arbitration in Jakarta against the government of Indonesia for breach of the performance guarantee.  However, the following month, the Indonesian government sought injunctions to stop the arbitration from going forward.  In July 1999, the central district court in Jakarta issued the injunction, purporting to suspend the arbitration against the Indonesian government and fining all participants $1 million a day if they continued with the arbitration.

In response, the three-member arbitral panel moved the arbitration hearings offshore to The Netherlands.  The government, thwarted in its attempt to frustrate the arbitral process (it failed in an attempt to obtain an injunction from a Dutch court to enjoin the proceeding in The Netherlands), then resorted to coercing the arbitrator it appointed to the panel from participating in the proceedings.  The arbitral tribunal nevertheless entered an interim default award against the government of Indonesia on September 26, 1999 and a final award on October 16, which the government then refused to pay.

With OPIC and Lloyds now facing what were clear violations of international law by the Indonesian government, an insured that had the full value of its project taken without any compensation, as well as arbitral awards that had been obtained and not paid, CalEnergy’s claims were paid in full one month later in the amount of $290 million.  It remains the largest political risk insurance claim paid to date.  Thus, thanks to political risk insurance, CalEnergy was able to recoup its investment within one month of obtaining a final arbitral award against the Indonesian government and within approximately one and a half years from the initial payment default under the ESC.

The Experience of Other Investors

Although CalEnergy’s experience was arduous, it ended with a recouping of its investment, which otherwise would likely have been lost.  Other foreign investors in Indonesia (and elsewhere) have not had such positive results.

The facts surrounding the Kahara Bodas power project in Indonesia start out virtually identical to the CalEnergy situation.  Much like the CalEnergy projects, the Kahara Bodas Company was formed in 1994, principally by Caithness Energy and Florida Power & Light (each of which owned 40.5% of the company), to develop two geothermal power projects in Indonesia.  The contractual structure of the projects was similar to the CalEnergy projects.  Disputes under its operating and energy sales contracts with PLN and Pertamina were to be resolved by UNCITRAL arbitration in Switzerland.  The presidential decrees in the fall of 1997 unilaterally suspended the projects, and the project company, KBC, commenced arbitration against PLN and Pertamina in April 1998.

KBC obtained an arbitral award against Pertamina and PLN for $261 million in December 2000.  The defendants then sought to appeal the award in the Swiss courts without success.  KBC sought enforcement of the award against offshore assets of Pertamina in courts in the US, Hong Kong and Singapore.  A US federal court in Houston confirmed the award and entered judgment against Pertamina for $261 million on December 4, 2001.  Pertamina’s appeal of this decision to a US appeals court is still pending.  In February 2002, KBC filed garnishment and discovery requests with respect to Pertamina’s US assets in New York, as well as in Texas and Delaware. (Among other assets, KBC is seeking to attach the proceeds of liquefied natural gas sales made by Pertamina that are paid into trust accounts at Bank of America in New York.)

Pertamina has vigorously defended these actions and KBC has been engaged in costly and time-consuming international litigation in its attempt to get paid.  Although KBC was initially successful in freezing $320 million of Pertamina funds in the Bank of America accounts in New York, in April 2002 Pertamina obtained a New York federal court order holding that 95% of the funds held in the New York accounts belong to the government of Indonesia and not Pertamina. That decision was upheld on appeal, so it appears that only about $16 million of funds have effectively been frozen.

KBC also had some initial success with courts in Hong Kong and Singapore similarly freezing Pertamina assets.  But in August 2002, Pertamina obtained an order from the central district court in Jakarta annulling the arbitral award, and Pertamina has sought to use the annulment order to its advantage in the proceedings in the US, Hong Kong and Singapore.  It is arguing that those courts should refuse to enforce the award based on article V(1)(e) of the “New York Convention on the Recognition and Enforcement of Arbitral Awards.” The New York Convention was designed to ensure that parties resorting to offshore arbitration need not relitigate the merits of the dispute when seeking to enforce the award in countries that are parties to the convention.  It provides for few defenses to enforcement, but article V(1) (e) provides that the recognition and enforcement of an award “may be refused” if the award “has been set aside or suspended by a competent authority of the country in which, or under the law of which, the award was made.” As the agreements that provided for offshore arbitration were governed by Indonesian law, Pertamina seeks to persuade courts in these various jurisdictions that they should exercise their discretion and not enforce the award.

The KBC case continues to play out with complex litigation that has moved from Switzerland, through US courts in Texas, New York and Delaware, among others, and on to Hong Kong, Singapore and Jakarta.  Although the ultimate outcome remains uncertain, one thing is clear: Cal Energy recouped its investment in full by payment from its political risk insurers within one month of obtaining an arbitral award; KBC has had its lawyers chasing assets worldwide for over two years now since obtaining its arbitral award in December 2000, and it still seems pretty far from recovering anything.

KBC is hardly alone.  Investors in the other power projects in Indonesia have also run into difficult problems trying to recoup their investments.

The $2.9 billion Paiton power project has chosen to renegotiate its power purchase agreement rather than to pursue arbitration.  Given the size and complexity of the Paiton project, too much may have been at stake for Paiton to pursue the arbitration and litigation route aggressively. (Prior to settling its differences through negotiations, a few litigation skirmishes were fought.) Press reports indicate that PLN has to date renegotiated 20 of its agreements with independent power producers, suggesting that most of the independent power producers in Indonesia determined that the arbitration and litigation routes were not in their interest.

Investors in projects in Pakistan, India and elsewhere have experienced problems similar to KBC in being thwarted by host governments when they attempted to exercise their contractual right to offshore arbitration of disputes.  In the midst of a foreign exchange or other political and economic crisis, host governments have failed to honor the contractual obligations that are the linchpins of infrastructure projects and have attempted to frustrate the dispute resolution process by seeking to prevent the arbitral process from proceeding.

Lessons Learned

These experiences of CalEnergy, KBC and other investors in developing country infrastructure projects in the 1990’s have taught some valuable lessons about political risk coverage.

Lesson number one: Political risk insurance coverage is valuable.  The CalEnergy experience — especially in contrast to the experience of similarly situated projects like the Kahara Bodas or Paiton projects — demonstrates that political risk insurance may be of crucial importance in ensuring the integrity of contractual structures relying on host government support.  The problem with host government contractual support is that it is most needed during periods of political and economic dislocation, but during such periods it is often unrealistic to expect that the host government will be in a position to meet its obligation.  When countries experience political and economic crises, host government contractual obligations will be of uncertain value without the creditworthy backstop of a political risk insurer.

Lesson number two: Arbitrations against host governments are likely to proliferate.  Foreign investment projects proliferated in the 1990s with the demand of emerging markets for infrastructure.  Project agreements, concession agreements, host government guarantees and bilateral investment treaties that provide investors with protection against expropriation all generally provide for the arbitration of investment disputes.  With political risk insurance contracts structured to protect investors against the failure by host governments to pay arbitral awards, we are likely to see more arbitrations against host governments than in the past.

Without political risk coverage, the foreign investor faces a threshold decision as to whether to take a host government to arbitration over a failure to pay or to renegotiate the obligation to its detriment.  However, if the government cannot meet its current obligation, why is it more likely to be able to make the type of large buyout payment that would be awarded as damages by an arbitral tribunal? Moreover, the KBC experience suggests that even if the government can meet its obligation, the foreign investor may be facing years of complex and costly international litigation in order to compel the government to comply.  Without political risk insurance coverage, an investor will often feel that it has no real effective dispute resolution alternative to accepting a unilateral renegotiation of its contract with the host government on terms imposed on it by the host government.

As political risk insurance coverage makes arbitration of the dispute an effective alternative (because the insured is assured of a creditworthy obligor once the award is obtained), host governments are likely to see a rise in arbitrations against them for breaches of their agreements.  Since the insured investor’s interest in the project company and arbitral award is assigned to the insurer as a condition to claim payment, host governments will also probably find themselves in negotiations with the home governments — for example, the US government to the extent the coverage is issued by OPIC or US Ex-Im Bank — or the World Bank Group in the case of MIGA coverage — rather than the foreign investor.  The US ambassador played an important role in negotiations between OPIC and the government of Indonesia, for example, after OPIC paid the CalEnergy claim.

Lesson number three: Since governments can be relied on to attempt to frustrate enforcement of arbitral awards, it is important to draft the arbitration clause with the aim of making such frustration difficult.  Whether CalEnergy or Kahara Bodas in Indonesia, Enron in India, or the Hub power project in Pakistan, when foreign investors have sought to enforce their contractual rights against host governments by commencing arbitration, the host governments have aggressively sought to enjoin the arbitrations from proceeding.  As political risk contracts typically require the insured actually to obtain the arbitral award (in order that the insurer has salvage rights against the host government after payment of the claim), attention needs to be given to drafting the arbitration clause in a manner designed to overcome these likely attempts by governments to frustrate the process from continuing.  In addition to providing for offshore arbitration in a New York Convention jurisdiction hospitable to arbitration, the arbitration provisions should provide, for example, that the parties shall have no access to the local court system with respect to a dispute under the agreement until after an award is made and then only for enforcement of an arbitral award.

Lesson number four: Here are some tips for managing the claim process.  When pursuing claims against political risk insurers, a number of complex issues typically arise.  First and foremost, it is imperative that the insured keep its insurers informed of problems and current developments as soon as it becomes aware that political problems may give rise to a claim.  Most political risk insurance contracts require the insured to do so (as, among other things, the insurer may be in a position to use its influence to resolve the problem).  Providing full and current information will, at a minimum, preclude defenses by the insurer and, at best, may actually lead to the insurer assisting on resolving the problem before it leads to a claim. (MIGA, in particular, has shown an ability to assist the investor in this manner.)

One thorny issue that sometimes arises is that political risk insurance contracts typically require the insured throughout the claim process to negotiate in good faith with the host government and generally to pursue all remedies as though it were uninsured.  Although this provision is straightforward in its application in expropriation policies, it is less clear how this provision works in the context of disputes coverage.  If the host government indicates a willingness to settle an arbitration at a low amount and the insured refuses to settle (since it has paid high premiums for a policy standing behind the arbitral award now within its grasp), does the insurer have a defense to claim payment? Having bargained with the insurer for it to stand behind an arbitral award, it would not seem appropriate for an insurer to try to defend on the basis of an unwillingness of an insured to settle for less in this situation.  These clauses should probably be addressed when negotiating the policy.  In this context, it is not reasonable to expect the insured to act as though it were uninsured.

Another provision that needs to be looked at carefully when negotiating the policy is the assignment requirement in connection with claim payment.  Given that most infrastructure projects are project financed and lenders will want a lien over all project company assets (including any arbitral award), it is important to make sure that all the insurer will require to be assigned is the insured’s interest in the project company and arbitral award and not the award itself, which will likely be encumbered.  The industry has had enough problems with the insurers’ requirement that the insured’s equity interest be delivered free and clear of liens. (This often requires a complex carve-out from the lender’s pledge agreement.) It simply would not be possible or reasonable to expect lenders to relinquish their lien on an arbitral award in this context and care should be taken to ensure that the language in the policy does not appear to require this.

Lesson number five: Confidentiality agreements may pose a problem.  Political risk insurance contracts have tight confidentiality clauses.  One problem with these clauses is that if the insured feels the insurer is wrongfully denying its claim and unreasonably forcing it to a costly arbitration, it is deprived of its ability to complain to others in the market.  As the ability to complain about perceived unfair treatment often serves as a moderating influence on market participants, consideration should be given to negotiating these clauses in a manner where the confidentiality requirement with respect to a claim would no longer be required after the claim has been denied.