Barclays' FX Research team explains why Ebola can no longer be ignored...
1. Is Ebola a significant market risk?
Before the current Ebola outbreak began in December, the disease was already notorious and fear inspiring despite having afflicted only 2,345 people in the 37 years since its discovery, nearly all of whom were in remote villages in Central Africa and 1,546 of whom died. Yet, as long as sporadic outbreaks remained contained and confined, it had little impact on the global economy or markets. The current outbreak in West Africa may be changing that.
The West African Ebola outbreak that began in Guinea is notably different from prior outbreaks in ways that significantly increase its chances of having a broader impact. In particular, Ebola’s likelihood of spreading to larger, more integrated economies has increased. It remains a tail risk, but has jumped in probability to one that can no longer be ignored.
There are several important differences between this Ebola outbreak and previous episodes.
First is its size: confirmed cases and mortalities – 4,087 and 2,071, respectively – already are greater than all prior outbreaks combined and growing at an uncontrolled pace. Including probable and suspected cases increases those numbers to 7,470 and 3,431, respectively (Figure 1), and both the US Centers for Disease Control (CDC) and the World Health Organization (WHO) believe these numbers are underestimated, perhaps by a factor of 2.5.
Second, the current outbreak is more urban than prior outbreaks, with Freetown, Sierra Leone and Monrovia, Liberia being among the hardest hit areas.
Third, and perhaps most importantly, the current outbreak is on the edge of a vast, densely populated area of 334 million people that is an ideal breeding ground for disease (Figure 3). As Robert Kaplan described it 20 years ago when its population was about half as large, “the Lome-Abidjan coastal corridor – indeed, the entire stretch of coast from Abidjan eastward to Lagos – is one burgeoning megalopolis that by any rational economic and geographical standard should constitute a single sovereignty, rather than the five (the Ivory Coast, Ghana, Togo, Benin, and Nigeria) into which it is currently divided.” Somewhat presciently, Kaplan noted this densely populated, impoverished, crime-ridden area was a perfect Petri dish for pestilence.
Figure 2 presents statistics on some of the socio-economic and health factors that make the region particularly susceptible to disease. The region’s young population – median age is 18.3 years – suffers the worst of poverty’s fellow travellers, including low literacy rates, poor access to clean water and to improved sanitation, and few health professionals and facilities. Illiteracy and residual warlordism from recent civil wars were key contributors to the rapid spread of Ebola in Sierra Leone and Liberia as health workers had difficulty educating a distrustful population on prevention, and in cases have been attacked on rumors they brought the disease. Making matters worse, Ebola has disproportionately stricken healthcare workers, reducing already low ratios to the general population.
A fourth difference is an active debate over how contagious the virus is now and what its potential to become more contagious is. Researchers have established that Ebola’s genetic structure is subject to a rapid pace of mutation and have documented numerous genetic changes to the virus, Zaire ebolavirus, one of five documented strains of Ebola, both from previous outbreaks and since the current outbreak began. The implications of these changes for the virus’s behaviour and morbidity are not yet known and will require significant further study. Health officials and researchers are not in agreement about the virus’s current and potential means of transmission. The WHO and CDC maintain that Ebola can only be transmitted via direct contact of an infected, symptomatic person’s bodily fluids with open lesions or with mucus membranes. Yet, the Public Health Agency of Canada and the Federation of American Scientists warn that aerosolized transmission, though unconfirmed in human-to-human transmission, has been documented in non-human primates and pigs.
Adding to the debate over the contagiousness of the strain of Ebola behind the current West African outbreak is its rapid expansion and higher-than-usual rate of infection of health professionals. The current outbreak is between 6.8 and 12.4 times larger – confirmed versus suspected cases – than the next largest outbreak, the first one in (then) Zaire and Sudan in 1976. As noted above, this may reflect a slow initial response combined with cultural differences between West Africa and Central Africa. But the higher rate of infection among health professionals is more difficult to explain. Through late August, the latest date for which complete statistics were available, 120 healthcare professionals had died from Ebola, or roughly 8% of confirmed, probable and suspected cases. By comparison, in the 1976 outbreak, when Ebola was an unknown new pathogen, 11 healthcare workers died from the disease, 2% of total deaths. Among the health professionals killed by the disease were five of 58 authors of the genomic study of the current outbreak, cited above, including Sheik Humarr Khan, an expert on hemorrhagic fevers and director of the Lassa fever program for the Sierra Leone Ministry of Health and Sanitation.
A more robust domestic and international response may still turn the tide of Ebola. The spread of the West African outbreak to both Senegal and to Nigeria was quickly contained and no new cases have manifest. In the last month, global attention has increased and with it resources devoted to combating the disease’s spread. Indeed, there already are signs that its pace of expansion has slowed.
Figure 4 plots the rate of growth of new Ebola infections – confirmed; confirmed and probable; and confirmed, probable and speculated – at a monthly rate between WHO situation reports (the longer time series in Figure 1 is for combined confirmed, probable and suspected cases). All three measures have fallen sharply to just under 50% per month since the surge of the late summer, when the average rate of change at a monthly rate rose to over 100%. However, this only takes the rate of expansion back to its July level, and the late summer surge may have reflected improved accounting. Furthermore, the geographic distribution of new cases continues to expand within Guinea, Liberia and Sierra Leone, and is expanding towards Côte d’Ivoire.
A strong external response may reinforce the trend of slowing expansion and ultimately contain and extinguish the West African outbreak. If it does not and Ebola’s eastward expansion continues at even the reduced current pace into the more populous countries of West Africa, the mathematics of geometric progression implies terrifyingly large numbers of cases in the not-distant future. The path of cumulative Ebola cases is projected, in log scale, in Figure 5 under two growth rate assumptions: the average rate of growth of confirmed cases since end June (66% per month) and since end August (54%). Both paths are more conservative than the CDC’s projection of 550,000 cases by January – even more so relative to the CDC’s “worst-case scenario” of 1.4 million – because the CDC includes probable and suspected cases in their starting point. But even more conservative projections that start with only confirmed cases and use the slower September growth rate suggest that there could be 1 million cases by the first week of August, 2015. If growth averaged its June-September pace, there could be 1 million cases by the second week of June 2015.
Case numbers that large raise the probability that Ebola could become endemic – that is permanently resident – in the West African population of 334 million. Beyond the human tragedy that implies for West Africa, the prospect of endemicity increases the probability of Ebola having a greater impact on the global economy and financial markets. Afflicted countries’ relatively small weight in global GDP, trade and financial markets so far has limited Ebola’s effects on global markets despite an already larger-than-SARS death toll.10 But a sustained outbreak in a larger or more globally integrated economy likely would have a significant impact on global markets.
That endemicity of Ebola to West Africa, or even a larger outbreak, greatly increases the chances of the disease breaking out to other more economically integrated economies is evident in the fact that cases have already been confirmed in Dallas, Texas and Madrid, Spain despite an outbreak of only 7,000 people in three small countries without direct air traffic to either the United States or Spain. If the projections in Figure 5 are fulfilled – much less the worse estimates of the CDC – the likelihood of Ebola spreading to other economies would rise sharply. The economic and market effects of sustained global outbreaks is another source of uncertainty, undertaken in the next question.
2. What lessons does SARS hold for Ebola?
The 2003 SARS outbreak, primarily in East Asia, is a good starting point for analysis of the potential effects on economies and on global financial markets of an expansion of Ebola outside of West Africa. But one of the lessons from the SARS experience is that it may not be an appropriate analogue for a possible Ebola breakout from Africa.
Barclays re-examined some of the lessons of the SARS episode for Asian economies in Revisiting Asia’s SARS experience, 20 August 2014. Among the lessons: 1) supply disruptions were minimal because the disease did not afflict enough people to curtail labor supply meaningfully; 2) the primary impact was a panic-driven decline in demand, especially for local services and for tourism, but there was some short-lived effect on trade and investment as decisions were delayed due to uncertainty; 3) once the disease was controlled and panic subsided, activity rebounded rapidly.
There are important differences between the SARS and West African Ebola outbreaks, however, both in current fact and in potential. SARS, an airborne respiratory virus, is more contagious than Ebola, but only about 1/8th as lethal. The greater lethality of Ebola, if its spread continues, may imply a supply effect for currently affected economies that was not present with SARS, especially in the dominant, labor-intensive agrarian sector. SARS afflicted several economies that were deeply integrated in the global economy, including Canada, China, Hong Kong, Singapore, Taiwan, and Vietnam; whereas Ebola is confined to much smaller, much less globally integrated economies. The SARS afflicted economies also benefitted from much better health statistics than do West African economies (see Figure 3). As noted in Question 1, this may explain Ebola’s rapid spread in the current outbreak.
The more interesting differences may be around potential. If Ebola continues to expand as explored in Question 1 and becomes endemic to West Africa, its then likely eruptions into the broader world also may be different from the SARS experience. First, while Ebola is not now as contagious as SARS, its greater mortality likely will inspire as much panic as SARS. But the difference with SARS would be that the persistence of a large human host reservoir would mean any panic may not fade as rapidly, particularly if it spawned sporadic mini outbreaks.
Second, the persistence of a large human host reservoir in West Africa would increase the probability that Ebola became epidemic or even endemic elsewhere in two ways: the continuous flare-ups of outbreaks would give the virus more chances to take hold elsewhere; and, as discussed in Question 1, a human host reservoir would give it greater opportunity to evolve into a more communicative, higher morbidity disease. SARS never became epidemic or endemic anywhere, and as such was not able to significantly affect the supply chain of the global economy. A sustained outbreak of a high mortality disease like Ebola in any large or important economy in the global supply chain would imply significantly larger impact than SARS caused.
3. What do you need to monitor and what could be Ebola’s effects?
As noted above, for Ebola to become a serious market risk, it needs to more acutely threaten demand or supply in larger or more economically integrated economies. Ebola’s threat to the global economy and markets would increase markedly if it either became endemic to West Africa or became more contagious and increased its rate of morbidity. But the response of larger, more economically integrated countries to any breakouts of the West African Ebola virus is likewise important.
What should investors monitor to assess the risks to markets from Ebola? In our view, three things:
1. A sustained pace of expansion within West Africa, or conversely containment and recession that eliminates the likelihood of endemicity;
2. Evidence of increased contagiousness among new cases, either due to new transmission vectors (like aerosol) or a longer period of asymptomatic contagion;
3. The success of countries outside of West Africa in containing the periodic cases that likely will escape the region, or already have in the case of Dallas, Texas.
Ebola’s expansion can be monitored in WHO situation reports on the West African Ebola outbreak: does the rate of infection continue to slow and remain restricted to Guinea, Liberia and Sierra Leone, or does it expand to new countries and sustain a rapid pace of expansion? Monitoring contagiousness is more difficult as it will be months before even preliminary analyses of the already documented changes in the Ebola genome are mapped to phenotypical outcomes (physical or chemical changes that determine the virus’s survival and means of replication). The sole case of Ebola in Dallas, Texas is simultaneously concerning and reassuring. There are already several well documented missteps in containment of the victims’ contacts, and yet the lack of further infections – so far – is reassuring that Ebola may be less dangerous in wealthier, healthier societies.
If Ebola becomes endemic and more contagious, its threat to the global economy and markets likely will be greater than SARS’ impact in 2003. Without endemicity in West Africa, mishandled outbreaks outside of Africa may have more temporary, SARS-like effects on local economies and markets. The greatest potential impact of Ebola is a combination of the two: endemicity and/or greater contagiousness combined with a mishandling of an outbreak outside of West Africa. That combination, for the reasons noted in Question 2, likely would have a significantly greater impact on the global economy, perhaps through both demand and supply channels, and as such on global financial markets.
Endemicity of Ebola in West Africa that spawned continuous periodic outbreaks elsewhere, or more destructively a sustained outbreak in a large or globally integrated economy, likely would lead to lower average growth with greater volatility, lowering risk free yields and increasing risk asset spreads. Those countries that best demonstrate their ability to avoid contagion, or best manage Ebola outbreaks when they occur, likely will see lower risk premia attached to their assets and higher equilibrium exchange rate values.
It must be emphasized again that endemicity of Ebola anywhere – including West Africa – remains a tail risk. But it is a tail risk that has left the fringes.