Valuation in International Arbitration: Measuring country risk in the Middle East

The following risks, which vary from country to country, can cause significant uncertainty about an investment’s returns:

• macroeconomic factors, such as inflation, currency stability, sovereign default and interest rates;
• political factors, such as the risk of expropriation, war, civil unrest and corruption, systems of governance and the presence or absence of investment treaties; and
• environmental factors, including the risk of natural disaster, the availability of power and fuel, and the quality of physical infrastructure.

The presence of such factors renders expected returns from a given investment less certain than those from an equivalent investment from which they are absent. In this article, we use the term country risk to refer to an investment’s exposure to uncertain outcomes as a result of its location in a particular nation state.

Country risk can be an important consideration in the assessment of damages in international arbitration, because valuation analysis depends on assessments of relevant risks (and the associated expected returns). Damages are often assessed by reference to:

• the value of an asset or business, on the basis that a wrongful act deprived the claimant of his or her economic rights in relation to the asset or business. In this case, damages might be assessed as the monetary value of the asset or economic right in question on the relevant date; or
• the value of profits a claimant has lost or will lose as a consequence of a wrongful act. This analysis may require a comparison between the value of the net cash flows that a claimant would have earned if no wrongful act had been committed and the value of the net cash flows the claimant is expected to earn given the wrongful act.

Irrespective of the approach, in assessing value the expert must account for the relevant risks attached to the asset in question, including relevant country risk.

The assessment of country risk can have a significant effect on the quantum of damages claimed or awarded. For example, in Gold Reserve Inc v Venezuela,¹ the amount claimed was reduced by US$130 million as a result of the tribunal’s findings in respect of country risk. This was almost 20 per cent of the award of US$713 million (before interest). This is one of several recent ICSID cases involving Venezuela in which experts and tribunals have reached a range of conclusions as to the quantification of country risk in their assessments of damages.² In these cases, a key legal issue has been the extent, if any, to which a state’s propensity to expropriate should be reflected in the valuation analysis.

Even in cases not involving expropriation, the presence and magnitude of country risk can significantly affect the assessment of quantum. It is not unusual for tribunals to be faced with two experts presenting different opinions and analyses. This is, in part, because a range of approaches are available to quantify country risk, as set out below. It may also reflect differing opinions on the extent to which the returns from the investment in question were, or would have been, affected by different sources of uncertainty.

In this article we explore some of the challenges and considerations involved in measuring country risk, in the context of a cash-flow-based damages assessment or valuation. We also briefly review the main methods used by practitioners to estimate country risk premiums. We conclude with illustrative results of such calculations, with reference to the major economies of the Middle East. This region is host to a range of markets with distinctive characteristics, and thus provides a useful context against which to understand the issues involved in estimating and applying country risk premiums.

Accounting for country risk in a cash-flow-based valuation

The valuation analyses which underpin assessments of damages often include a discounted cash flow (DCF) analysis. A DCF analysis measures the present value of a stream of net cash flows that a business or project will earn or would have earned. It involves two stages:

• preparing forecasts of the net cash flows that were or would have been expected to be generated by the project, business or asset in question. If valuing a business, the relevant cash flows are those that it would generate if it were financed wholly by equity; and
• pplying an appropriate discount rate to those forecasts, to calculate a monetary sum as at the valuation date. This discount rate should be the post-tax cost to the business of the weighted average return required by its providers of debt and equity capital.

Country risk factors may be incorporated at either or both stages of a DCF calculation. First, an expert may take into account known uncertainties in his or her cash flow forecasts, by (for example) making separate forecasts of cash flows given potential adverse or beneficial outcomes, then assigning probabilities to each of those scenarios to calculate the ‘average expected cash flows’ each year. To provide a simplified example, in the first year, there may be a 95 per cent chance of a business earning net cash flows of US$100 million, and a 5 per cent chance of it failing in the event of a possible legislative change (US$0). The average expected cash flows for that year are US$95 million.³

Second, an expert may adjust the discount rate used to calculate the present value of forecast future cash flows. The discount rate may be adjusted for country risk by including a country risk premium, which assumes that the investor in the relevant country requires a greater return, all else equal, to compensate for the higher level of uncertainty in that country compared to a mature, developed economy such as the United States. All else equal, including a country risk premium will increase the discount rate and decrease the present value of the future cash flows. For example, if a stream of expected net cash flows of US$95 million per year for 10 years is discounted at a rate of 10 per cent, its net present value is approximately US$580 million.⁴ If a discount rate of 15 per cent is applied instead, the value reduces to around US$480 million.⁵ Often, the results of a DCF analysis are sensitive to relatively small changes in the discount rate.

Identifying the most appropriate way to incorporate country risk in a DCF analysis is not always straightforward. First, the risks listed at the beginning of this article reflect the diverse factors that could give rise to country risk, from macroeconomic volatility to the threat of civil war. In a given situation, identifying the relevant risks or assessing their likelihood and potential magnitude can be challenging.

Second, as explained above, the expert must assess whether to account for risk factors by adjusting cash flow forecasts, or by adjusting the discount rate (or both). In principle, the cash flow forecasts to which discount rates are applied should reflect all possible outcomes weighted by their relative likelihood. However, giving effect to this will require estimates of the likelihood and possible financial impact of all sources of uncertainty. The discount rate should reflect only relevant risks. This would depend on whether the uncertainty in question could be avoided by investors being ‘diversified’, by which we mean a risk that is specific to a particular business or asset (and thus can be reduced or eliminated by combining several diverse investments in a portfolio). In principle, the rate of return on equity required by investors applied in DCF calculations is a measurement of ‘non-diversifiable risk’ – diversifiable risks are not relevant.⁶ 

Third, as we explain below, there are several approaches that can be used to estimate the country risk premium. Different approaches are based on different data, and measure different things. Some may be more appropriate in certain situations than others. Often, the available measures may not directly correspond to the risks identified as relevant to the valuation exercise.

Finally, there is debate among practitioners as to whether country risks themselves may be diversifiable. Some suggest that, because investors can access different international markets, those investors may consider most country risks to be diversifiable. If that is applicable to the investment in question, then there may be no need for a material country risk premium.⁷ Others hold the view that a critical mass of investors retain a home bias, and thus in practice the relevant investors may not be globally diversified.⁸ It is also suggested that not all country risks are diversifiable as a result of correlation between markets in different locations.⁹

Ultimately, developing an appropriate methodology requires analysis of the circumstances and nature of the business in question, as well as theoretical considerations. It is also important to be pragmatic, as data directly relatable to the identified risks may be hard or impossible to obtain. In the following section we explain how country risk premiums are often measured in practice. We then present premiums calculated for the major economies of the Middle East.

Measuring the country risk premium

The most common methods used by practitioners to estimate country risk premiums include:

• sovereign risk methods, where the premium is derived from a country’s credit rating, the yield on its sovereign bonds, or the spread charged on credit default swaps (CDS) for the relevant sovereign debt;
• equity market risk methods, which compare the variability of stock market returns in the country under consideration to that of a developed economy; and
• survey methods, which seek to collate and analyse country risk assessments used by investors, analysts and academics.

There are yet further methods available. Below, however, we consider the three methods listed above that, in our experience, are most widely used. We also identify some additional considerations relating to the calculation and application of country risk premiums.

Sovereign risk methods

One common method to estimate a country risk premium is to consider indicators of sovereign default risk. Experts may assess sovereign default risk as the difference between the yield on the local government’s bond (in US dollars or another convertible currency) and a comparable benchmark bond (often from the United States) or by reference to market-determined CDS spreads. If the sovereign does not issue foreign currency-denominated debt, the sovereign spread may be inferred to be the bond yield spread of other nations that do issue such debt and that have the same sovereign credit rating. Each of these methods measures the incremental returns required by investors in bonds issued by governments in emerging economies (as compared to US government bonds) and assumes that this reflects the incremental returns required by investors in equities in these countries. Part of the appeal of these methods is their simplicity and reliance on easily obtainable public data.

Sovereign risk methods may incorporate some adjustment to reflect the fact that government bonds are generally less risky than equity investments, which suggests that equities might face an increased country risk premium. In some cases, practitioners apply a uniform multiple of 1.5 to debt spreads to account for this issue. An alternative approach is to scale a premium based on CDS spreads by the difference in volatility between government bonds and equity markets in the country in question.¹⁰

Equity market methods

A further method is to compare the relative volatility of different stock markets. For example, US-based investors might observe that equities in an emerging market are more volatile than those on the New York Stock Exchange, and adjust their required return for emerging market investments accordingly.¹¹

In practice, this method can be problematic because volatility in stock markets outside of benchmark economies may not reflect country-specific risks. First, stock markets in emerging economies are often illiquid, which makes calculations of volatility less meaningful, as low observed volatility is not necessarily indicative of investors’ assessment of risk.

Second, such stock markets are also often dominated by a small number of large companies. Thus, volatility may be skewed based on the characteristics and circumstances of these companies, rather than the country where the companies are located. This is illustrated in Figure 1, which shows the liquidity of various Middle East stock markets and the largest company’s share of each exchange’s market capitalisation.

A further factor is that stock markets in emerging economies are often dominated by particular industries. For example, around three-quarters of the Bahraini stock exchange is comprised of financial services companies. As a result, volatility of the Bahraini exchange may reflect industry-specific rather than (or as well as) country-specific risk.

Due to these factors, our results below do not include country risk premiums measured on this basis.

Figure 1: Comparison of stock market liquidity and diversity across the Middle East

Survey methods

A third method to estimate country risk premiums is to ask investors about the returns they require in different locations. There are various surveys that ask companies, academics and analysts about the market risk premiums they use for investments in different countries. The difference between the premium required in a benchmark economy and the premium required in the host country of the investment in question provides an estimate of the country risk premium.

There are drawbacks associated with reliance on surveys, such as a lack of respondents for certain countries, as well as the possibility for respondents to interpret the same question differently. However, in our experience it is not uncommon for practitioners to take survey results into account. A benefit of the approach is that a practitioner’s selection of calculation methodologies or data inputs cannot influence the magnitude of the premium. Our analysis below includes country risk premiums based on the 2016 survey published by Professor Fernandez and colleagues of the IESE Business School titled, ‘Market Risk Premium used in 71 countries in 2016: a survey with 6,932 answers’.

Additional considerations

In determining how to apply the premium calculated for a particular country, a practitioner must consider additional factors. These factors show the need for practitioners to be sensitive to the context and purpose of the valuation so as to appropriately reflect the risks to which the investment in question is exposed.

By way of example, it may be relevant to consider whether the investment in question is exposed to country risk factors to a greater or lesser extent than other investments in the same country, perhaps because of its industry.

Additionally, experts should be aware that measures of country risk premiums may change over time. Periods of social unrest or political upheaval could lead to material changes in a short period. By way of illustration, a measure of the country risk premium of Egypt in December 2010, by reference to bond yields, was 2.7 per cent.¹³ At this time, President Mubarak had been in power for approximately 30 years. In January 2012, approximately one year later, following political turmoil and changes in the country’s leadership, the same measure of the country risk premium was 9.8 per cent.¹⁴ This illustrates the potential importance of the date of measurement in assessing a country risk premium.

Country risk in the Middle East

We show below illustrative calculations of country risk premiums for countries in the Middle East using some of the approaches reviewed above. First, we briefly highlight some characteristics of those countries that might influence investors’ assessment of country risk. Relevant characteristics include both political and economic factors. These are linked, as economies are shaped by governments’ resource allocation decisions, international relations and economic models.

The Middle East – A diverse political and economic landscape

Figure 2 summarises some chief macroeconomic and governance statistics for the emerging economies in the Middle East. The figure demonstrates clearly the comparative wealth, and low debt, of the major hydrocarbon exporting economies in the region.

Qatar is the world’s largest exporter of liquefied natural gas.¹⁵ Saudi Arabia, Iraq, UAE, Kuwait, Iran and Oman were all among the top 15 exporters of crude oil in 2015.¹⁶

Typically, for economies dependent on petroleum reserves, energy exports fund investment in infrastructure, social welfare and low-tax environments. There is also a high degree of reliance on expatriate labour. In the period following their dramatic fall in June 2014, low oil prices created substantial macro-economic imbalances, causing states to review public spending plans, deplete reserves and seek to raise debt.¹⁷

Several states have renewed efforts to diversify their economies and attract foreign investment.¹⁸ The UAE has enjoyed notable success in this regard.¹⁹ ‘Free zones’ such as the Dubai International Financial Centre have played a key role in attracting foreign investment to the United Arab Emirates, providing opportunities for 100 per cent foreign ownership of locally based businesses, and bespoke regulation.²⁰ This has not gone unnoticed in the region, and states such as Qatar, Saudi Arabia and Oman have announced plans to stimulate inward investment through increased use of such zones.²¹

The ‘Ease of Doing Business’ rankings published by World Bank Group, shown in Figure 2, rate the UAE as the 26th ‘easiest’ economy to operate in. Regionally, this is ahead of other countries such as Oman (66th), Qatar (83rd), and Saudi Arabia (94th). The ‘Corruptions Perceptions Index’ published by Transparency International appears correlated with the Ease of Doing Business rankings, albeit with some outliers.²² These statistics indicate that investors perceive the investment environment across these countries as different. However, such perceptions are not necessarily determinative of country risk. This is indicated in the premiums shown below.

Some other hydrocarbon-producing countries in the region, such as Libya, Yemen and Iraq have been severely affected by conflict in recent years. Such conflicts have caused the displacement of populations, affecting the economies of some other states, including Lebanon and Jordan in particular.²³ Along with Egypt, whose economy has also been beset by domestic conflict, Lebanon and Jordan have the highest levels of public debt as a proportion of GDP. This is shown in Figure 2.

The brief summary above provides a high-level primer of some relevant issues in assessing the investment environment in the Middle East. Below, we set out illustrative calculations of country risk premiums for some of the countries we have considered.

Figure 2: The Middle East – Macroeconomic and governance snapshot

Country risk premiums in the Middle East

Figure 3 summarises country risk premiums (CRPs), based on data as of January 2017, for six of the countries considered above: Qatar; Saudi Arabia; UAE; Bahrain; Egypt and Iraq. We present premiums calculated using sovereign risk methods (bond yield spreads, CDS spreads and sovereign credit ratings), in addition to those published in the IESE survey.

The results show the premiums applicable to the wealthier, hydrocarbon-exporting countries with stable governments (ie, Qatar, Saudi Arabia and UAE) are relatively low, around 1 per cent on most measures. For these countries, premiums are similar, indicating that distinctions between these countries as regards governance or macroeconomic conditions may not affect investors’ perceptions of risk significantly.

Political and economic security, on the other hand, appears to be a significantly more influential factor. This is shown in the materially higher premiums for Egypt and Iraq. Both of these countries have experienced conflict and instability in recent years. The range of results for these countries is also wider. In the case of Egypt, for example, country premium estimates range from 5.4 to 9.2 per cent.

Here, the choice of methodology could have a significant effect on the discount rate applied in a DCF calculation, and thus an expert’s quantification of damages. This demonstrates the importance of selecting an appropriate method for the project, business or asset in question.

Figure 3: Country risk premiums for selected countries in the Middle East, January 2017

Concluding remarks

It is clear from Figure 3 above that the different methods of estimating country risk can give rise to a range of estimates for the country risk premium. Which approach is most appropriate will depend on the circumstances of the country and the investment and may partly depend on practical matters, such as the availability and reliability of information. Ultimately, the preferred method should best reflect the risks to which the investment is actually exposed.

The characteristics of the company may be important. For example, if one considers two businesses operating in an economy where there is high inflation, currency devaluation and a history of uneven economic growth. The first is a construction company engaged on domestic projects, employing mainly local staff. The second is an oil refinery with an international customer base and a high proportion of expat workers. The expected cash flows of the construction company are likely to be directly exposed to uncertainties as a result of the country’s macroeconomic instability. The oil refinery, on the other hand, is likely to be less exposed to these risks – oil is traded globally and most of the company’s costs and revenues are in US dollars. It may be inappropriate to apply the same country risk premium to both businesses.

We explained that country risk factors may be accounted for in an expert’s work by adjusting cash flow forecasts or by adjusting the discount rate. A key difference between these approaches is that adjustments to cash flow forecasts require explicit assumptions in relation to the likelihood of various alternative outcomes, whereas the application of a higher discount rate cannot readily be reconciled to risk assessments in relation to individual risk factors. It is possible to unite the two approaches, as the effect of an adjustment to the discount rate can be explained with reference to the adjustments one would need to make to cash flow forecasts to achieve the same result. This may sometimes be a helpful analysis.

Experts can also seek to identify transactional evidence, which may provide a benchmark for the results of a valuation based on DCF (or may be the primary method of valuation). An expert might compare his or her DCF valuation with a value implied using the observed prices of comparable traded assets operating in the same country. If the two measures are consistent, or if the differences can be explained by the characteristics of the assets, then this could provide additional evidence that the expert’s analysis and conclusions are reasonable.

Finally, to ensure that country risk is treated appropriately in damages calculations, it is essential to be mindful of the limitations that exist in any analysis of country risk. In emerging markets in particular, there is likely to be missing data and issues that are uncertain. This is not necessarily a barrier to an appropriate assessment of losses. The most effective analysis will be based on a logical and consistent approach, and assumptions will – as far as possible – be demonstrably consistent with market conditions and the facts of the case.

Views expressed herein are those of the authors and not necessarily the views of FTI Consulting, Inc, its management, its subsidiaries, its affiliates, or its other professionals.

Notes

1. ICSID Case No. ARB(AF)/09/1: 848. 
2. ICSID Case No. ARB(AF)/09/1. ICSID Case No. ARB/10/19. ICSID Case No. ARB/07/27. ICSID Case No. ARB/10/5.
3. US$95 million = (95 per cent x US$100 million) + (5 per cent x US$0 million).
4. US$583.7 million = US$95 million x (1 – (1+10 per cent)-10) / 10 per cent.
5. US$476.8 million = US$95 million x (1 – (1+15 per cent)-10) / 15 per cent.
6. Commonly, this rate of return is estimated using a model known as the Capital Asset Pricing Model. It is calculated on the basis that the investors that are establishing the market prices of assets are diversified, and do not require additional expected return for risks that can be avoided by diversification. 
7. Valuation, Koller, Goedhart and Wessels, 2010: 728.
8. Measuring Company Exposure to Country Risk: Theory and Practice, Damodaran, September 2003: 4.
9. Measuring Company Exposure to Country Risk: Theory and Practice, Damodaran, September 2003: 4.
10. Measuring Company Exposure to Country Risk: Theory and Practice, Damodaran, September 2003: 11 to 12. Country Default Risk Spreads and Risk Premiums, Damodaran, January 2017, proposes an equity-market to debt-market multiplier of 1.23 based on an analysis of emerging market indices.
11. In financial theory, volatility and risk are considered closely related.
12. Egypt profile – Timeline, BBC News, 15 November 2016. Premium based on bond yield spread as at 31 December 2010, Bloomberg. Equity-market to debt-market multiplier applied of 1.5. 
13. Egypt profile – Timeline, BBC News, 15 November 2016. Premium based on bond yield spread as at 31 December 2010, Bloomberg. Equity-market to debt-market multiplier applied of 1.5.
14. Egypt profile – Timeline, BBC News, 15 November 2016. Premium based on bond yield spread as at 11 January 2012, Bloomberg. Equity-market to debt-market multiplier applied of 1.5. 
15. Qatar, US Energy Information Administration, update on 20 October 2015.
16.  Crude Oil Exports by Country, World Top Exports website, 30 January 2017.
17. Regional Economic Outlook: Middle East and Central Asia, IMF, October 2016: 15 and 19.
18. Regional Economic Outlook: Middle East and Central Asia, IMF, October 2016: 21.
19. Regional Economic Outlook: Middle East and Central Asia, IMF, October 2016: 21.
20. ‘The role of freezones’, Middle East Business Intelligence, 11 August 2013.
21. ‘The role of freezones’, Middle East Business Intelligence, 11 August 2013.
22. For example, Qatar’s ‘Ease of Doing Business’ ranking is 83 out of 190; its ‘Perception of Corruption’ ranking is 31 out of 176.
23. Regional Economic Outlook: Middle East and Central Asia, IMF, October 2016: 34.