Skip to main content

The COVID-crisis

Published by , Editor
Oilfield Technology,

In an exclusive article for the May/June 2020 issue of Oilfield Technology, Stephen B. Harrison, Nexant Energy & Chemicals Advisory, Germany, considers the COVID-crash’s potential implications for the Asia Pacific region.

On 20 April 2020, negative pricing for the May WTI futures contract sent a shockwave through the energy sector. Crude oil pricing structures have evolved over decades and have served the industry well during periods where supply has had ample time to adjust to long-term demand trends. They were never designed to work in a period of demand shock, such as that experienced worldwide in the first four months of 2020.

Figure 1. South China Sea drilling rig and support vessels.

For many Asian countries, which are major consumers and importers of crude oil, the COVID-crash might appear to be a once-in-a-lifetime buying opportunity to fill their strategic reserves and stock up before prices resume an uptrend. However, as any commodity trader will know, picking the low point in the market is never easy and over-committing to deliveries for which there is no storage or refining capacity is a high-risk game. Who would ever have expected that crude oil production and ownership could become a waste disposal problem?

What is the root cause of the COVID-crash?

Debates may rage for years about whether the market pricing structure was simply not set up for this ‘black swan’ event or whether geopolitical tensions also played a role in the COVID-crash. Could oil producers have cut crude production deeper and faster? Should they have reacted sooner to early warning indications from Chinese economic data that additional international coronavirus lockdowns would cause a more significant demand slump on a global scale? Incredulity, curiosity, a desire to analyse, the need to understand and a willingness to learn will all lead to the question: ‘Why did the oil price crash so dramatically?’

The simple answer is that supply and demand very quickly fell out of balance. There was an abrupt reduction in aviation, maritime, rail and road travel – causing a waterfall decline in refined products demand. Chemicals and plastics plants cut back their processing volumes, as a result of falling demand, as production lines in industries such as automotives and textiles shut down – pulling through fewer oil-derived petrochemical products. Electricity production slumped in line with the drop in industrial activity; fuel oil demand for power generation dropped accordingly. However, in Asia, the Pacific, the Americas, Africa, Europe and the Middle East, indeed all around the world, millions of barrels of crude more than the world could consume kept on being produced per day.

Will land-locked producers be hit hardest by the COVID-crash?

In any supply chain, storage between two processing steps is the buffer that absorbs short-term swings in demand. Static storage in the crude supply chain is fixed. However, redundant oil tankers can be used as temporary additional storage capacity. For offshore upstream operations, such as Malaysian rigs in the South China Sea or Australian production in the Bonaparte and Carnarvon Basins in the Indian Ocean, the use of tankers to store excess crude is conceivable.

Figure 2. Tazhong oilfield in Xinjiang, China.

This is not the case for Asia’s land-locked producers. Shuttering wells is expensive, but for onshore producers that outcome may come sooner than for offshore and nearshore producers. On the other hand, production cuts in other regions may redress the supply and demand balance and trigger an upward move in local demand and pricing just in time to save the day. Waiting for this to happen is a high-risk strategy based on hope however, and hope is not always the best decision-making tool. With this in mind, it is expected that land-locked producers with limited access to storage will be the first in line to shutter up their wells. Asia seems to be weathering the storm at present, but at the time of writing the US rig count has fallen 40% from 650 to 380.

To read the rest of this article, download Oilfield Technology's May/June issue for free:

Read the article online at:

You might also like


Embed article link: (copy the HTML code below):


This article has been tagged under the following:

Upstream news Upstream project news Oil & gas news