Oil report

RPT-COLUMN-Coronavirus and the impact on oil consumption: Kemp

(Repeats from Tuesday. John Kemp is a Reuters market analyst. The views expressed are his own)

LONDON, Feb 4 (Reuters) - Oil traders are struggling to estimate the severity and duration of the hit to global consumption from the outbreak of coronavirus in China, based on evidence from previous epidemics of coronaviruses and influenza.

Medical researchers recognise three global pandemics of influenza in the course of the 20th century, in 1918/19 (“Spanish influenza”), 1957/58 (“Asian influenza”) and 1968/69 (“Hong Kong influenza”).

Before the advent of modern record keeping, there are other probable candidates for global influenza pandemics, including 1847 and 1889 (“Report on the pandemic of influenza 1918-19,” British Ministry of Health, 1920).

More recently, there have been coronavirus outbreaks in 2003 (severe acute respiratory syndrome) and 2013 onwards (Middle East respiratory syndrome coronavirus).

And there have been several influenza epidemics and pseudo-pandemics that were quickly confined, including in 1947 (U.S. service personnel in Asia), 1976 (U.S. service personnel at Fort Dix) and 1977/78 (“Russian influenza”).

Most influenza and coronavirus outbreaks have followed virus mutations as they have passed back and forth between human and animal hosts, which is why China’s live animal markets have been such a high-risk factor.

By their nature, epidemics tend to spread rapidly through a susceptible population that has never encountered the virus before and has no immunity, but the rate of infections and fatalities then tends to peak and fade quickly.

Most uncontrolled outbreaks have lasted roughly three months, though in some instances there have been multiple waves, such as the three waves of influenza during the summer, autumn and winter of 1918/19.


The transmission and severity of an outbreak is largely determined by two parameters: the basic reproduction number (R0) and the case fatality rate (CFR).

The basic reproduction number is a measure of the expected number of others an individual host with the virus will infect without containment measures.

In practice, the actual transmission rate (R) may be lower than the R0. But as long as the actual value of R is greater than 1, the epidemic will accelerate, as the number of infected individuals grows exponentially.

Eventually, when enough individuals have contracted the disease, developed immunity, recovered, and are no longer susceptible, the transmission rate will decline, which is why epidemics eventually fade out.

R will eventually decline below 1 because there are simply not enough susceptible individuals left in the population for the virus to continue spreading, as many have already been exposed and died or recovered.

Containment measures aim to bring down the transmission rate more rapidly through quarantine, social distancing, cancelling public events, focusing on super-spreaders, such as healthcare workers, and intensified personal hygiene.


The case fatality rate measures the number of infected individuals who eventually die of the disease or complications, such as secondary bacterial infections.

One reason the pandemic of 1918 proved so deadly is that in an era before antibiotics many of those who contracted viral influenza went on to contract bacterial pneumonia.

In most instances, viruses tend to be most lethal to the young, the old, those with already weakened immune systems and those with underlying medical conditions.

Several million deaths around the world each year are directly or indirectly attributable to the ordinary seasonal outbreaks of influenza, most of them in higher-risk categories.

The pandemic of 1918 was unusual because a high proportion of the deaths were prime age individuals, for reasons that are still not fully understood.

Case fatality rates are difficult to estimate because the mildest cases may not be reported or recorded leading to an overestimate of mortality.

In general, there is a trade off between the reproduction rate and the case fatality rate, because the virus needs to keep its host alive long enough to transmit it to others.

The most lethal viruses tend to have relatively low transmission rates, because they kill the host too quickly, while the highly transmissible viruses tend to be less lethal, giving them more opportunity to infect other hosts.

But reproduction rates and case fatality rates are notoriously difficult to estimate accurately even after an outbreak has been controlled.


In any respiratory epidemic, the biggest impact on the economy and oil consumption comes from the containment measures, such as quarantine and social distancing, taken to bring the epidemic under control.

Some measures will be ordered by the government to deal with a public health emergency but others will be taken voluntarily by businesses and individuals concerned about limiting exposure to the disease.

Social distancing measures can have a large negative impact on both business and consumer spending as well as on manufacturing production, services provision and transportation networks.

Like the course of the epidemic, the economic impact tends to be acute rather than lasting, compressed into the space of a few weeks or months.

As the outbreak burns itself out, or is effectively controlled, the need for extreme quarantine and social distancing measures declines and activity returns to normal.

As the outbreak fades, concern about health risks is eventually displaced by commercial pressure to resume normal activities.

Businesses, employees, transport companies, schools are forced to resume near-normal activity in order to earn revenue, get paid, and pass examinations.


If a typical epidemic lasts less than 13 weeks, peaking around half-way through, the impact on economic activity is likely to follow a similar pattern.

The main exception is where the epidemic returns in multiple waves, which is unusual, but happened in 1918/19.

In London, the first wave of deaths peaked in week 28 of 1918 (with most deaths between week 25 and week 33). The second peaked in week 44 (with most deaths occurring between 42 and 51).

The third wave peaked in week 9 of 2019 (with most deaths starting in week six and ending by week 14), according to contemporary statistics compiled by Britain’s Ministry of Health.

The first wave was comparatively mild with only a few hundred deaths in the capital. The second was much more severe, killing 3,000-4,000 people per week at its peak. The third was somewhere in between.

In the current outbreak, the principal impact on China’s oil consumption is likely to be concentrated in the first quarter of the year, especially in January and February.

By March and April, the number of new cases and fatalities should start to decline, and more normal business activity and oil consumption should gradually be restored.

There is likely to be some lingering impact on economic activity and oil consumption as businesses and households absorb the loss of revenue and income from the first quarter, but it should fade.

But the biggest economic and oil consumption impact will most likely be felt in the first quarter and should start to fade from the start of the second and largely have disappeared by the end of the third.

The principal risk is fresh outbreaks in other centres in China or overseas, with a second or third epidemic wave later in the year, which would magnify the economic and oil consumption impact.


China’s oil consumption averaged around 14.5 million barrels per day (bpd) in 2019, according to the U.S. Energy Information Administration (“Short Term Energy Outlook,” EIA, Jan. 14).

For the sake of illustration, if coronavirus depresses China’s consumption by 10% on average in the first quarter, 2% in the second quarter and 1% in the third, the total impact will be the loss of around 450,000 bpd on average in 2020.

BP’s chief financial officer said on Friday that coronavirus could reduce oil consumption by about 300,000-500,000 barrels per day, roughly 0.5% of global oil demand, which is in line with this guesstimate.

China, together with India, has accounted for more than 50% of global incremental oil consumption in recent years, so any slowdown would have a major impact on the global production-consumption balance.

Oil prices have already fallen sharply to force U.S. shale producers and the OPEC+ group of major oil exporters to implement corresponding reductions in production this year. (Editing by Barbara Lewis)