Free Read: Is the Oil Market Operating on Borrowed Time?

There have been many opinions floating around as to why oil prices haven't spiked the way you'd expect after losing roughly 20% of global supply overnight. Estimates regarding the so-called offset mechanisms — bypass pipelines, reduced Chinese imports, increased US exports, SPR releases, and demand destruction — add up to a combined cushion of 17.5 mb/d. That's a large number, and if it holds, it explains the relative calm. In this piece I want to go through each mechanism and ask the question the table doesn't: how durable is it really?

Start with the bypass pipelines, because they've received the most attention. Saudi Arabia's Petroline — the 1,200km East-West artery running from Abqaiq to Yanbu on the Red Sea — is now pumping at its full capacity of 7 million barrels a day. The UAE's ADCOP pipeline from Habshan to Fujairah adds another 1.8 mb/d, and has been running at full capacity since the start of the war. Together these pipelines are doing meaningful work and deserve credit for that. But there are two constraints sitting quietly underneath the headline numbers. The first is physical: Yanbu's two export terminals have a combined loading capacity of roughly 4.5 million barrels a day — which means the pipeline is moving more oil than the port can actually load onto tankers. The second constraint is geopolitical, and it's the more serious one. All of this oil, once it reaches Yanbu or Fujairah, must still transit Bab el-Mandeb to reach Asian buyers. Oil exports through Bab el-Mandeb nearly doubled to 7.2 million barrels per day in April precisely because of this rerouting — and Iran knows it. Greater Houthi involvement at that strait is so worrisome because it has been the one outlet untouched by Iran's chokehold on Hormuz. This week, Yemen's Houthis declared a complete ban on Israeli shipping in the Red Sea. The bypass pipeline story, in other words, rests on a second chokepoint remaining open — one that is increasingly in play. 

The reduced Chinese imports figure — cited at 3.5 to 4 mb/d in the table — requires more careful framing. China hasn't cut imports because its demand fell. It cut imports because it spent all of 2025 stockpiling aggressively at cheap pre-war prices and is now drawing those barrels down instead of buying at current levels. China added an average of 1.1 million barrels a day to its strategic oil inventories in 2025, reaching nearly 1.4 billion barrels by December. That was a deliberate strategic bet, and Beijing got it right. But those inventories aren't infinite. According to various estimates, China is likely to return to buying large volumes of oil within weeks after selling down inventories but that is not until September. But so far the most reliable, long term offset mechanism is reduced Chinese oil appetite to buy from markets. This mechanism has an expiry date, and it's probably sometime around Q3. But before that we may not see any external pressure from China.

Increased US exports have been real — US crude exports jumped to 5.2 million barrels per day in April, a more than 30% increase over pre-war February levels. As highlighted in my previous research pieces (Monday Macro View) U.S. production is expected to stay strong. Pressure pumpers are engaged to their optimum levels further hinting towards elevated activity. Our FSC and FJC models continue to show stability. A very interesting paper by Atlantic Council highlighted that why exports shouldn't be banned and why they won't help prices at home. In this regard this offset mechanism may have yet another long term validity. 

Source: Robin Brooks, X

On the other hand, SPR releases are the most straightforward case of a finite buffer, and the numbers here are becoming difficult to ignore. The reserve is slipping to its most exhausted level since the Reagan era and will continue sinking as the Trump administration leans on it. It currently holds approximately 350 million barrels, with roughly 20% of total capacity required to remain operational — putting the functional floor around 70 million barrels. The API's CEO has now publicly flagged that reserves are entering a range of genuine concern. The SPR was designed as a bridge for short disruptions. The bridge was never meant to last this long. Even on a global level we cannot expect that emergency stockpiles will continue to make up for lost barrels. Refinery runs cannot go below 60%, pipelines need oil as well thoguhtout thier length to function. 

Finally, demand destruction. Most estimates are 2 to 4 mb/d. The IEA's own numbers tell a different story — a contraction of 420,000 barrels per day by end of 2026, and even with that the IEA still expects the oil market to end the year in deficit. What we are seeing is rationing, economic slowdown, and forced efficiency — not a structural shift away from oil. COVID came closest to real demand destruction at scale and recovered within months. Our estimates show that such a demand destruction may not be more than a million barrels even in extreme cases. Counting 2 to 4 mb/d here is, at best, assuming the worst case scenario.

Taken together, the 17.5 mb/d offset assumes that every mechanism fires simultaneously and holds indefinitely. The pipelines are capped and geopolitically exposed. The Chinese buffer reverses by Q3. US export gains reflect redirection more than new supply. The SPR is near its functional floor. And demand destruction is a fraction of what the table implies. The market is calm because these buffers are still in place. The more important question is what happens when they start to give — and that's what Part II will look at.