Electricity as Coercion: Is There a Risk of Strategic Denial of Service?
This guest post is co-authored by Joshua Busby, associate professor of public affairs at the Robert S. Strauss Center for International Security and Law at the LBJ School at the University of Texas at Austin, Sarang Shidore, a visiting scholar at the LBJ School at the University of Texas at Austin, and Morgan Bazilian, director of the Payne Institute and a professor of public policy at the Colorado School of Mines.
Increasing interconnection of electricity systems both within and between countries has much promise to help support clean energy power systems of the future. If the sun isn’t shining or wind isn’t blowing in one place, an electricity grid with high voltage transmission lines can move electricity to where it is needed. This shared infrastructure and increased trade can possibly serve as a basis for peace between neighbors in conflict, but it may also serve as a tool of coercion if the electricity can be cut off by one party.
Cross-border trade in electricity is currently dominated by Europe – 90% of the $5.6bn electricity trade market happens there, but in the future increased trade in electricity, particularly in Asia, is set to grow dramatically. The boldest proposal comes from the Chinese organization GEIDCO which has, with the backing of the State Grid Corporation of China (which reportedly has over 1 million employees), promoted regional and even global grid integration.
On the one hand, such grid integration could foster greater interdependence in conflict zones and facilitate more shared interests. But there is another concern, what we call a strategic denial of service. This would be a form of what Farrell and Newman refer to as “weaponized interdependence,” a situation where one country uses a shared relationship asymmetrically to extract political concessions from another party.
Emerging economies
China is providing ample financial support for electricity and energy initiatives through the Belt and Road Initiative (BRI) and the Asian Infrastructure Investment Bank (AIIB). As much as two thirds of BRI projects, worth some $50 billion, has been invested in the energy sector. Some observers have already raised concerns about what China’s overtures in this space might mean for its neighbors. Phillip Cornell, writing for the Atlantic Council, warned that despite the benefits of grid integration:
"Even if local grids are independently operated, deep interconnection means that supply and demand will increasingly be matched across the super-grid, making them more interdependent. It may be managed by 'international rules and operation code' as Liu [Zhenya, GEIDCO's chairman] insists, but those will be defined by a regional authority where China is bound to have major influence."
The scope for cross-border trade in electricity isn’t only Chinese-led. Even as India has been trying to integrate its domestic grid through what it calls green energy corridors, the country is also supplying electricity to some of its neighbors. India already exports some 660MW to Bangladesh, and Indian firms are building power plants which could meet as much as 25% of Bangladesh’s electricity needs. While India is currently supplying power to Nepal, Nepal has the potential capacity to supply hydroelectric power to India. Nepal and Bangladesh are also considering electricity trade through the intermediate Indian network. Hydro plants in the Mekong Delta from Laos already supply electricity to neighboring Thailand, making it its top source of foreign exchange. Other projects include CASA 1000, a proposed power line to link the Kyrgyz Republic and Tajikistan as well as an interconnection linking a hydro power station in Malaysia’s Sarawak to West Kalimantan that should diminish Indonesia’s dependence on imported oil.
Can electricity be used coercively?
Can a state use electricity as a coercive tool like the way Russia has used natural gas? Is this technically possible? What are the limits? In the case of gas, you have a product that can be physically stored for periods of time, whereas electricity is a much more ephemeral product that, absent viable storage at scale, is lost as waste heat if not transmitted to end users. Though a breakthrough in storage might take away some of the urgency of the threat of service denial, it wouldn’t remove it in the event of a prolonged outage.
Technically, the process of denial of electricity service is not all that difficult. Shutting down power service across a transmission system is just a matter of operations control (in the absence of good governance, power markets, contracts, etc. that are all in place to avoid disconnection of service). It can be done virtually instantaneously to disconnect power from any node on the system. Think of “rolling blackouts or brownouts” when different parts of a service area are shut down for various technical or economic reasons.
Curtailing electricity to another country is potentially costly for the coercer. Curtailing electricity transmission to a neighbor does mean foregoing payments for the electricity (provided the neighbor was paying their bills). But, it could be one that states will use to generate benefits such as higher rates of payment for electricity or, more broadly, to extract concessions on other matters of political importance. For some energy sources like hydro power, the water has other potential uses. This could enhance the attractiveness of using service denial as a coercive tool since the owner of the hydro could presumably monetize its water in another way.
A state might try to insulate its vulnerability to service denial through widespread conservation or building in extra reserve capacity, but in some settings and seasons, demand reduction might not be so easy to achieve. A country might be able to find alternative sources of electricity either from other neighbors or by powering up more expensive domestic sources of generation, though those arrangements could take time or prove much more costly than the existing cross-border arrangement.
The flipside of denial of service would be demand curtailment, which a state might pursue if it was attempting to punish a neighboring electricity supplier by reducing its revenues.
Has this been done?
During the Cold War, the Soviet Union was able to maintain dominance over Eastern and Central Europe by tying their energy supplies to the Soviet energy grid, reducing their scope for independent action. Though privatization in the early years of the break-up of the Soviet Union provided these countries with more independence, Gazprom made a conscious effort to acquire assets back under Russian control, sometimes under commercial conditions that could be construed as coercive, particularly in the natural gas space.
Baltic states and Poland remained tethered to the Russian electricity grid. It was not until 2018 that Estonia, Latvia, Lithuania, and Poland completed an agreement to decouple from Russia and transition to the EU grid by 2025 at the cost of $1.2 billion. Fears of potential Russian service denials helped them overcome remaining obstacles.
There have also been some examples in the post-Cold War era of denial of service and other forms of coercion related to integrated grids and interdependence.
In July 2018, Iran cut off a portion of power to Iraq over unpaid fees in the midst of a summer a heat wave. While this may have merely been to ensure Iraq paid its balance, other states have employed similar tactics for more expansive purposes.
In February 2019, the Trump Administration threatened secondary sanctions on Iraq to discourage its purchase of imported Iranian electricity and natural gas. Here, the service denial is not by the generator but by an influential third-party who has its own political axe to grind with the Iranians. In June 2019, the United States provided Iraq with a temporary four-month sanctions waiver to allow Iraq to get through the summer by importing products from Iran, lest the country experience a wave of unrest as it did in 2018 when Iran cut the power.
In May 2019, the United States and the Maduro government in Venezuela clashed over the rightful ruler of the country. Before and after the departure of Venezuela’s diplomats, left-wing U.S. protesters occupied the Venezuela embassy in Washington, D.C. to prevent supporters of the opposition Juan Guaidó from seizing the embassy. In the midst of the dispute, the power to the embassy was cut off by the electricity provider PEPCO, raising questions about political involvement by the Trump administration.
If we think of the embassy as sovereign territory of Venezuela, this would qualify as a case of cross-border service denial and speaks to the potential vulnerability of other such enclaves such as military bases that may depend upon electricity grids of host countries. In July 2016, Turkey temporarily cut power to the major US airbase in Incirlik in the wake of the coup attempt against President Erdoğan, underscoring these concerns about base vulnerability.
Cyber-security experts have also raised the prospect of denial of service by hackers who might be able to penetrate an electricity grid and take it off line. Given that communications, transport, and health care infrastructure all rely on electricity, the cascading effects of such an outage could have far-reaching consequences. If carried out by shadowy non-state actors, it might also be harder to attribute responsibility to a state actor. In December 2015, in the first known instance, Russian hackers were able to briefly take off line three Ukrainian distribution companies.
Which states might deploy this strategy?
We are more likely to see strategic denial of service where markets and contracts give private actors limited legal recourse in the event of supply disruptions. Moreover, strategic denial of service may be more common where there are large power asymmetries between neighbors, particularly but not limited to non-democracies. The more powerful state can use the size of its military apparatus or economy as additional leverage to extract concessions. In our view, we are less likely to see a poorer, smaller country like Nepal or Lesotho try to deny electricity to a more powerful neighbor, given the risks of reprisals. Similarly, in the event of demand curtailment, we might see powerful states use this tactic against neighbors that are highly reliant on electricity export revenue.
As Farrell and Newman argue, those that control central nodes are likely to possess asymmetric power at key chokepoints: “Specifically, states with political authority over the central nodes in the international networked structures through which money, goods, and information travel are uniquely positioned to impose costs on others.” In a bilateral sense, a small chokepoint would be the ability for one country to deny service to one downstream power importer, but if a single actor exercises control upstream over the entire transmission network, that would provide them with asymmetric power over a wider group of actors.
If no state controls a single node but several states together control electricity exports, that would require the kind of collective action that is less likely to occur in most regions of the world. In the event that the ambitious Desertec project moves forward, a study of the scope for North African countries to use renewable energy denial to Europe concluded that it was unlikely to succeed unless all five exporter countries curtailed service.
Authoritarian countries may use this tactic more than democracies. In authoritarian governments, private actors may have less arms-length relationships with the government and thus be susceptible to pressure to cut off service to foreign countries. However, powerful democratic countries may also use this tactic against adversaries and non-democracies. Outside the electricity space, we have even seen the United States try to use its control of SWIFT banking system to coerce other democracies to avoid trade with Iran. In the electricity arena, it is less clear when democracies might use denial of electricity as a tactic. That said, if the U.S. government did in fact have a role behind the scenes in cutting off power to the Venezuelan embassy, this would be an example.
As countries seek to balance their electricity needs to have the cheapest, greenest source of power when they need it, they may become both importers and exporters of energy. This may reduce the temptation for a state to unilaterally cut off electricity to its neighbor, lest the whole cooperative relationship fall apart. However, in a world of increasing concerns about sovereignty, we may see fewer of these interconnections to start with, absent confidence building measures and institutions.
Can this tactic be prevented?
To reduce the risks of coercive actions in cross border electricity trade, regional governance treaties and related multinational institutions should be created to oversee the implementation of agreements for the grid's operation. This could be akin to a neutral regional grid operator that has representation from all countries. ASEAN, for example, is helping develop the regulatory framework as the ASEAN Power Grid is built and knits countries together. OLADE -- the Organización de Energia Latinoamerica – is seeking to play a similar role in Latin America.
Ideally, markets, contracts, and legal forms of dispute resolution would also help ensure that politically motivated service denials do not happen, but market mechanisms on their own are unlikely to establish confidence in grid integration across borders. Regional institutions remain an important means of regulating the trade, along the lines of transborder water governance that Lucia De Stefano and collaborators write about in terms of institutions to apportion water, deal with shocks, and carry out dispute resolution. Chinese acquisition of electricity assets in Portugal, Greece, and Italy have led Europeans to raise questions about whether existing forums for transmission operators such as ENTSO-E ought to be elevated to a regulatory body.
Other regions are likely to be even less coordinated in terms of regulatory oversight. As Cornell points out, the vision for GEIDCO from the founder is one of decentralized, technical administration like the internet, without central control, but that actually betrays how the internet is subject to national level suppression as we have seen in countries like China with the Great Firewall. Other new Chinese-led institutions like the AIIB are subject to multilateral oversight, suggesting a governance model that might attenuate some of these concerns.
In the absence of institutions to guard against politically motivated service denials, countries will remain disconnected or even seek to decouple their systems from neighbors deemed too risky. In much of the electricity space where the potential is largely untapped, it would mean foregoing many of the benefits associated with integrated grids. Poorer, weaker countries needing power might have few options and accept the Faustian bargain that puts them at the mercies of more powerful neighbors. At a moment when our collective emissions of greenhouse gases already have tied us together in mutual vulnerability to climate change, it would be a shame if joint efforts to address the problem got caught up in the return of great power politics.