The sharp increase in the oil price between 2003 and 2008 brought back a phenomenon commonly observed in the 1960s and 1970s. Countries are expropriating assets of independent oil companies – directly with large unexpected windfall taxes. Countries with recent expropriations include Bolivia, Ecuador, Algeria, Russia, China and Venezuela. The subsequent collapse of the oil price highlighted how exposed many countries are to oil-price fluctuations, and many government budgets needed to be slashed in response to the price decline.
Resource-rich countries and oil companies negotiate contracts that specify future tax payments that the company will make to the host country in exchange for the right to produce hydrocarbons. In addition to operational expertise, these contracts could, in principle, provide a country with valuable insurance against oil-price risk. At low oil prices, the country receives positive tax payments from what may have otherwise been an unprofitable project. At high prices, the oil company will request a share of the profits in return. However, at these high prices governments will be most tempted to expropriate and capture all immediate benefits. This generates a tension between a contract’s ability to provide the host country with oil-price insurance, and its exposure to expropriation risk.
Previous research has discussed the drivers of expropriations.
- Guriev et al. (2008, 2010) have argued that “we should expect more expropriations in periods of higher oil prices."
- Another prediction is that expropriation is more likely when there are fewer checks on the government so that the government cannot commit to not expropriating.”
In recent research, we conjecture that the same factors that affect the likelihood of expropriations should also affect the structure of the contracts between host countries and oil companies (Stroebel and van Benthem 2012a). In particular, it should affect the degree to which oil-price risk is shifted from the host country to the oil company.
We analyse how the optimal contract varies, with parameters such as the country’s cost from expropriation (e.g. domestic and international legal challenges, loss of reputation, loss of foreign direct investment and loss of oil-production expertise). Our theory suggests that optimal contracts with countries that cannot credibly commit to honouring terms must involve payments that rise and fall with the resource price. The notion is that higher payments when the resource price is high reduce the incentives to expropriate, while lower payments in low-price times keeping the oil company interested in signing. In short, we should see the countries that suffer from a lack of commitment bearing more of the resource-price risk, than nations which can make solid commitments.
We then test these predictions with data for 2,466 contracts in 38 countries. For each contract we employ a hydrocarbon tax simulator (generously provided by the energy consulting firm WoodMackenzie) to identify the degree of contract insurance.
Before running regressions, we analyse the shape of the contracts. We show that – despite a myriad of different taxes – the tax payments can be quite accurately approximated by a linear function of the oil price. While contracts rarely condition tax payments explicitly on the price, they condition the payments on things – like revenue and production – that are either independent of, or rise linearly with, the price. As a result, total tax revenues increase linearly in the price. When non-linear taxes are present, they play a minor role.
Assuming that the expected payoff to the oil company is fixed, the degree of oil-price insurance provided through the contracts is captured by the slope of the government revenue-oil price function1. This gives the slope an easy interpretation. At a specific oil price, the slope indicates the additional revenue per barrel the government gets for a permanent $1 increase in the oil price. For a linear contract, the slope is constant and the size of the slope reflects the share of the oil price risk carried by the host country.
As argued above, a large number of factors contribute to the cost of expropriation faced by the host country and our estimation strategy considers some of the most important.
- First, it includes the cost of domestic legal challenges to expropriation (Engel and Fischer 2010).
In our empirical implementation, we measure this by the 'constraint on the executive index' from the Polity IV project, as well as the 'investment profile score' from the PRS Group.
- Second, it includes the loss of a country’s reputation following an expropriation.
This reputation loss has been shown to depress foreign direct investment (Eaton and Gersovitz 1984). We measure the cost of this to a country by the level of per-capita FDI inflow.
- Thirdly, it includes the efficiency loss following the loss of operational and technical knowledge of the oil company (Opp 2008).
We take the cumulative hydrocarbon extraction of a country up to the point of contract negotiation as a proxy for a country’s efficiency in producing hydrocarbons. This is to capture the idea that countries which have developed more hydrocarbon projects will have acquired more technical expertise.
Our regression results suggest that the structure of contracts is related to the dangers of expropriations as theory suggests (see paper for details). Specifically, countries that are better able to commit (as a result of larger costs associated with reneging on them) tend to obtain contracts that shift a larger degree of oil price risk to the oil company.
These effects are economically significant. For oil companies only, a one-standard-deviation increase in the constraint on the executive index (measuring a country’s institutional quality or cost of expropriation) is associated with a decline in the revenue-oil price slope of 0.05. In other words, for each $1 change in the oil price, the change in government revenues decreases by $0.05. Similarly, a one-standard-deviation increase in FDI per capita leads to a fall in the price slope of 0.03. A one-standard-deviation increase in cumulative production leads to an increase in slope of 0.02. This means that the ability to commit to contracts increases a volatility-averse country’s expected welfare. These results suggest that resource holding countries can benefit in terms of instituting more favourable taxation terms from providing more recourse to foreign investors.
In a follow-on project (Stroebel and van Benthem 2012b), we consider the extent to which Bilateral Investment Treaties matter. Such treaties usually include a clear description of what is considered an unlawful expropriation. Violations of contractual agreements with the oil company become a breach of the investment treaty with the country of incorporation. This facilitates the seizure of certain assets held abroad by the expropriating country.
Commitment through such treaties can be strengthened by including provisions that broaden the asset base subject to seizure following an expropriation. Hence expropriation costs increase, allowing for more insurance to be provided. Indeed, we find evidence that having a larger number of bilateral investment treaties signed by a country is associated with that country obtaining more oil-price insurance through their tax contracts. All else equal, countries with 30 more treaties signed (a one-standard-deviation increase), have an average risk-sharing slope in their contracts that is lower by 0.02. We also provide suggestive evidence that an investment treaty with the US is of particularly high importance. Countries which have signed such a treaty have a 4.1% lower price slope on average.
This suggests that bilateral investment treaties can have important welfare benefits for developing countries. In particular, they allow resource-rich countries to shift a larger proportion of the risk associated with variations in natural resource prices to oil companies operating in that country. Consequently, we recommend that developing countries consider expanding their participation in international investment treaties.
More subtly, beyond increasing the number of their bilateral investment treaties, countries should also benefit from contracts that explicitly waive sovereign immunity for assets held abroad. These waivers, which are equivalent to posting a bond, make it easier for foreign investors to execute any arbitration claim they may be granted. Such provisions would increase the bite of existing treaties and should further allow host countries to increase the level of price insurance provided by their tax frameworks.
Guriev, S, A Kolotilin and K Sonin (2008). "High oil prices and the return of 'resource nationalism'", VoxEU.org, 12 April
Guriev, S, A Kolotilin, and K Sonin (2010), "Determinants of Nationalization in the Oil Sector: A Theory and Evidence from Panel Data." Journal of Law, Economics, and Organization, 27(1).
Engel, E, and R Fischer (2010) "Optimal Resource Extraction Contracts Under Threat of Expropriation." forthcoming in W Hogan and F Sturzenegger (eds) The Natural Resources Trap, Cambridge, MA: MIT Press.
Stroebel, J and A van Benthem (2012a), “Resource Extraction Contracts Under Threat of Expropriation: Theory and Evidence.” Forthcoming in Review of Economics and Statistics.
Stroebel, J and A van Benthem (2012b), “Investment Treaties and Hydrocarbon Taxation in Developing Countries,” MIT Press, CESifo Conference Volume (forthcoming).
Eaton, J, and M Gersovitz (1984), "A Theory of Expropriation and Deviations from Perfect Capital Mobility." Economic Journal, 94(373): 16-40.
Opp, M. 2008. "Expropriation Risk and Technology." PhD diss. The University of Chicago.
1 To compute the contract slope γi(p) at oil price p for country i, we define the incremental revenue related to a permanent per barrel oil price increase of Δp as:
where TGR(p) is the total undiscounted government revenue over the life time of the project, assuming the oil price is constant at p. Furthermore, RR represents remaining reserves measured in barrels that will be produced.