The original argument
In the Goldman Sachs Global Institute discussion, Jason Bordoff argues that the prolonged disruption of Hormuz demonstrates both the strait’s importance and the resilience of a system supported by earlier investment. The absence of a sustained $200-a-barrel oil price does not make the disruption immaterial. It suggests that substantial buffers have absorbed part of the shock.
Before the crisis, approximately one-fifth of global oil and LNG trade passed through the strait. In the source account, Brent rose from around $70 before the war to an average a little above $100, reaching $126 in April. Several mechanisms helped prevent a larger increase: relatively comfortable initial supply and inventories; Saudi and Emirati bypass pipelines carrying more than 5 million additional barrels a day; strategic stock releases and commercial inventory reductions supplying another 5–6 million; and a production and export response from the United States.
China also affected the balance. By slowing strategic stockbuilding, using inventories, restricting refined-product exports and containing domestic demand, it reduced crude imports by roughly 5 million barrels a day between February and June in the source account. Resilience was therefore not an abstract market property. It reflected infrastructure, stored resources, purchasing power and decisions about demand.
A changing geography of energy power
The United States is no longer situated as it was in earlier oil crises. It is a leading oil and gas producer, and its economy uses less oil per unit of output than in the past. China, meanwhile, is more than a large importer: inventories and the capacity to adjust demand can influence the marginal global balance.
The costs fall unevenly. Importers with weaker fiscal capacity are less able to compete for scarce cargoes. The source account records jet-fuel spot prices approaching $200 a barrel in some markets and nearly 200 emergency conservation, rationing or tax measures across dozens of developing economies. Energy security therefore cannot be reduced to a distinction between producers and importers. Production, inventories, fiscal resources, electrification, alternative infrastructure and market influence all shape exposure.
From just-in-time to just-in-case
Bordoff’s broader argument is that redundancy is regaining policy value. Alternative pipelines, ports, strategic stocks, domestic refining capacity and dependable supply chains provide options when an expected route fails. Saudi Arabia and the UAE plan further bypass capacity; Iraq and Kuwait are exploring alternatives. The United States faces the task of replenishing and modernizing strategic oil reserves that the source account places near a five-decade low. India, South Africa and Southeast Asian countries are also considering larger storage and refining buffers.
These options must be financed before they are needed. Their value is easy to underestimate when the cheapest route operates normally. The shift toward a “just-in-case” economy brings those insurance-like costs back into decisions about infrastructure.
Electrification and the Chinese buffer
Bordoff treats China’s electrification as an energy-security strategy as well as a climate strategy. The source account records electricity at more than 32 percent of final energy consumption, compared with approximately 22 percent in the United States and 21 percent in Europe. More than half of new vehicle sales in China are electric. Clean-energy investment reached about $625 billion in 2024, close to one-third of the world total, and the relevant industries contributed more than 10 percent of Chinese GDP in 2025.
Electrification helps explain why an oil-supply shock need not translate one-for-one into domestic transport disruption. Large oil reserves—estimated in the source account at about 1.4 billion barrels and only partly drawn down—may provide another buffer if the crisis persists. Bordoff’s geopolitical analogy is that Beijing could acquire a form of marginal market influence once associated with spare production capacity in Riyadh.
The picture is not a simple replacement of fossil fuels. China is expanding renewable energy and nuclear power while also using coal to support security. Its concentration of solar, wind and battery manufacturing, alongside mineral processing, means other countries must compare different dependencies rather than imagine that dependence can disappear.
The US position is equally mixed. The shale revolution and lower oil intensity reduce some macroeconomic vulnerability; domestic gas around $3 per MMBtu contrasts with prices near $20 in parts of Europe and Asia in the source account. Bordoff nevertheless warns that export restrictions would weaken production incentives and supplier credibility. Energy security, industrial policy and climate policy may support electrification while also sustaining fossil-fuel buffers. Electrification and decarbonization need not advance at the same pace.
The data center connection
For AI infrastructure, the relevant extension is from inexpensive energy to energy that is deliverable, dispatchable where required, and backed by credible alternatives. A low-priced renewable power-purchase agreement is not by itself a guarantee of continuous physical supply. The reliability requirements of computing workloads differ, but facilities that place a high value on continuity must examine what happens when their normal route fails.
A large campus might combine a grid connection, long-term procurement, storage, on-site generation and backup arrangements. Nuclear or gas supply and multiple network paths may form part of that design in some locations. This is a menu of possible options, not a requirement that every site build every technology. The economic question is which combination provides useful redundancy at an acceptable cost.
Who finances the option?
The relevant contractual questions extend beyond dollars per megawatt-hour. Who funds standby capacity and grid reinforcement? Which obligations apply during an interruption? Can backup resources participate in capacity or other grid-service arrangements, subject to their operating limits? Is apparent redundancy genuinely independent, or do several options rely on the same fuel supplier or transmission corridor?
A US–China comparison can examine how different institutions create and pay for these options. China’s electrification, domestic coal, network construction and state coordination offer one set of mechanisms; US gas resources, market institutions and private investment offer another. The proposition to test is that AI competitiveness depends partly on dependable, affordable and genuinely diverse energy routes, not only on average electricity prices.
About this reading
This commentary brings together a reading of the linked publication and an analysis of its relevance to data center infrastructure. Figures and event descriptions refer to the source’s publication context; the data center implications are identified as analysis rather than attributed findings.