Incremental COVID-19 News More Likely Positive Than Negative

Incremental news on COVID-19 is likely to point towards greater healthcare capacity and lower mortality rates than current consensus. On balance, this is bullish for risk assets, despite risks of second waves after the reopening of some economies.

Source: Johns Hopkins CSSE, Macro Hive

As the global COVID-19 curve flattens, policymakers are considering gradually lifting mitigation measures (Chart 1).  Such moves however could lead to a resurgence of infections, especially in countries where curve flattening has been achieved through strict measures such as New Zealand, Israel and Austria (Chart 2).  By contrast, a second wave of infections appears less likely in Australia, Greece and Germany, which have flattened their curve despite less stringent measures (due to data issues I am not going to discuss China).

Source:  Johns Hopkins CSSE, Oxford University Blavatnik School of government, Macro Hive

Despite a likely resurgence in infections in some countries, economic reopening could still gain momentum for four reasons.  First, the risks of COVID patients overwhelming healthcare systems have been receding.  Contrary to expectations, countries have been able to expand their healthcare capacity through measures such as: building additional hospitals; repurposing existing equipment; and ordering their local industries to switch production to needed supplies and machineries.  In addition doctors are rethinking the use of respirators.

Second, the relationship between the intensity of confinement and curve flattening varies across countries.  Chart 2 shows the change in the slope of the epidemic S curve against the average value of the intensity index in the 15 days to April 22^nd.  Countries such as Taiwan, Australia and Greece have managed to flatten their curve with only limited restrictions.  By contrast, in France, South Africa and India strict confinement measures have produced only limited curve flattening.  As more information on the epidemic and economic impact of lockdowns becomes available, countries where lockdowns have under-delivered could face pressure to ease them.

Source: COVID Tracking Project, Macro Hive

Third, the symptoms that COVID-19 patients experience seem to be lightening, global mortality is stabilizing and recoveries net of deaths are increasing as relative to the number of cases.  In the US, hospitalization rates are stabilizing (Chart 3).  This could reflect that the more vulnerable members of the population are first to experience acute symptoms and that, as the epidemic takes root, symptoms become less intense.  In addition, infection rates in countries are converging (Chart 4).

Source: John Hopkins CSSE, Macro Hive

Fourth, the actual mortality rate could be significantly lower than the current global rate of 7%.  There are two broad issues with assessing mortality rates.  First, even if countries had the resources to study each death to determine its cause, there is no straightforward way of doing so in patients where multiple health conditions are present.  In practice countries follow different standards, which makes cross country comparisons difficult.

Excess deaths relative to a historical average are probably more reliable indicators but are not very precise, either, because: base line changes year to year; factors other than COVID-19 can drive excess deaths; and excess deaths need to be assessed over a long enough period as the most vulnerable are likely to get hit first, which could see a spike in deaths followed by a decline.  Excess death data for Switzerland, Spain and Lombardy displays this pattern.

The second issue with mortality rates is the unreliability of the case count, which is biased both by a lack of testing and by sample selection bias.  In most countries, access to testing remains limited and policy makers have understandably opted to test primarily symptomatic individuals.  Consequently, if COVID transmission remains asymptomatic, testing could be missing a large share of the infected which would exaggerate mortality rates.

The most reliable mortality rate estimate we currently have is Iceland’s, not only because this country of 364,000 has tested 12% of its population, but also because testing has consisted of viral rather than antibodies tests and the former is more reliable than the latter.  Iceland’s mortality rate has been stable at 0.6%, which is compares with 0.3% for the 1957-58 influenza pandemic and 0.1 to 0.2 % for the regular flu.

Antibodies tests are just starting and likely to have reliability issues that could bias results, both positively and negatively.  Nevertheless early results points at higher infection rates than suggested by viral testing: early results in California, Germany, and the Netherlands suggest an underlying infection rate ranging from 3 to 15%, while a recent modelisation exercise in France suggests  an underlying infection rate of about 5%.  This compares with confirmed cases of 0.2% of the population in the US, Germany, France and the Netherlands.

Epidemic markers have recently become drivers of equity and FX market performance.  In countries where the epidemic has been tamed through stringent measures, this outperformance could, however, reverse with economic re-opening.  This could lead to markets discounting the efficiency of the policy response rather than the epidemic intensity alone.  In addition, incremental COVID-19 news seems more likely to be bullish than bearish risk.  They could point at higher healthcare capacity, weaker symptom intensity and lower mortality rates, as well as more broadly reduce the uncertainty surrounding COVID-19.