Resource Extraction Contracts Under Threat of Expropriation: Theory and Evidence

Abstract

We use fiscal data on 2,468 oil extraction agreements in 38 countries to study tax contracts between resourcerich countries and independent oil companies. We analyze why expropriations occur and what determines the degree of oil price exposure of host countries. With asymmetric information about a country's expropriation cost, even optimal contracts feature expropriations. Near linearity in the oil price of real-world hydrocarbon contracts also helps to explain expropriations. We show theoretically and verify empirically that oil price insurance provided by tax contracts is increasing in a country's cost of expropriation and decreasing in its production expertise. The timing of actual expropriations is consistent with our model. Disciplines Business | Economics | Public Affairs, Public Policy and Public AdministrationThis journal article is available at ScholarlyCommons: http://repository.upenn.edu/bepp_papers/140 RESOURCE EXTRACTION CONTRACTS UNDER THREAT OF EXPROPRIATION: THEORY AND EVIDENCE Johannes Stroebel and Arthur van Benthem*Abstract-We use fiscal data on 2,468 oil extraction agreements in 38 countries to study tax contracts between resource-rich countries and independent oil companies. We analyze why expropriations occur and what determines the degree of oil price exposure of host countries. With asymmetric information about a country's expropriation cost, even optimal contracts feature expropriations. Near linearity in the oil price of real-world hydrocarbon contracts also helps to explain expropriations. We show theoretically and verify empirically that oil price insurance provided by tax contracts is increasing in a country's cost of expropriation and decreasing in its production expertise. The timing of actual expropriations is consistent with our model.

Expropriation risk and FDI in developing countries: Does return of capital dominate return on capital?

“…For brevity, we present findings based on the ICRG data. 18 For example, Jensen's (2008) and Stroebel and Van Benthem's (2013) measure of expropriation risk, Constraints on the Executive from the Polity IV database, measures the extent of legislated constraints on the decision-making powers of the government executive; fewer constraints are interpreted as a lower cost of expropriation (and therefore higher risk). An advantage of this measure is that it is based entirely on objective categorical indicators (whereas ICRG and WGI ratings both combine subjective evaluations of survey and economic data).…”

“…38 Specifically, the government confiscates all foreign capital whenever its aggregate value exceeds the cost of expropriating. In the fully dynamic, repeated game environments of Cole and English (1991), Aguiar et al (2009), and Stroebel and Van Benthem (2013), the cost of expropriating is permanent exclusion from international capital markets (i.e., loss of reputation for honoring contracts). 39 Without loss of generality, we follow Eaton and Gersovitz (1984), Tomz and Wright (2010) and Hajzler (2014) and consider a static (two-period) game between foreign investors and a host-country government.…”

“…38 See, among others, Eaton and Gersovitz (1984), Cole and English (1991), Thomas and Worrall (1994), Albuquerque (2003), Aguiar et al (2009), Tomz and Wright (2010), Stroebel and Van Benthem (2013) and Hajzler (2014). 39 When cutting off all future investment cost is the maximum (trigger strategy) punishment that investors can impose, the cost to the host country is decreasing in the representative agent's (or government's) discount factor and increasing in the degree of risk aversion, since there is an insurance motive for accessing international capital markets when the host-country agent is risk averse.…”

“…39 When cutting off all future investment cost is the maximum (trigger strategy) punishment that investors can impose, the cost to the host country is decreasing in the representative agent's (or government's) discount factor and increasing in the degree of risk aversion, since there is an insurance motive for accessing international capital markets when the host-country agent is risk averse. However, even if the host-country agent is risk-neutral, a loss in expected future FDI inflows may be costly if FDI is associated with more efficient technology, as assumed in Stroebel and Van Benthem (2013) and Opp (2012). 40 In the fully dynamic environments considered by Cole and English (1991) and Stroebel and Van Benthem (2013), exogenous states are assumed to be independently and identically distributed (i.i.d.)…”

“…However, even if the host-country agent is risk-neutral, a loss in expected future FDI inflows may be costly if FDI is associated with more efficient technology, as assumed in Stroebel and Van Benthem (2013) and Opp (2012). 40 In the fully dynamic environments considered by Cole and English (1991) and Stroebel and Van Benthem (2013), exogenous states are assumed to be independently and identically distributed (i.i.d.) over time tation cost associated with loss in future foreign investment as well as any costs incurred by bringing foreign operations under state control, including legal costs associated with international arbitration claims against the host country, the use of less productive technologies, and potential diplomatic or economic sanctions imposed by investor country governments (such as relinquished aid flows or preferential trade access).…”