How did some countries manage to achieve a very low covid-19 prevalence?
(See updates below.)
Non-African countries with at least one million inhabitants but less than 1000 covid cases per one million inhabitants include, notably: Laos, Vietnam, Cambodia, Taiwan, Thailand, China (ex. Wuhan), New Zealand, South Korea, Cuba, Hong Kong, Australia (ex. Victoria) and Singapore (ex. migrant workers). Other major countries with a low covid prevalence include Japan, Finland and Norway.
The typical ‘reactive’ measures cannot explain this: it cannot be masks, as most of the worst affected countries have introduced mandatory masks, too; it also cannot be national lockdowns, as most of the worst affected countries have had lockdowns, too; and it cannot be PCR mass testing, as many of the worst affected countries have very high testing rates, too.
In fact, many low-covid countries never imposed a lockdown (including Taiwan, South Korea, Japan, Australia (excluding Victoria), Laos and Cambodia), and many low-covid countries never performed mass PCR testing (including Japan, Laos, Vietnam, Cambodia and Cuba).
Instead, the single most important factor has been early border controls – as of January or February 2020 – something all low-covid countries indeed did. There are two overlapping groups of countries with successful early border controls: islands and countries neighboring China.
Border controls are easiest for islands, which most low-covid countries indeed are, including Taiwan, New Zealand, Australia, Japan, Cuba, and also, essentially, South Korea, Hong Kong and Singapore. Many other, smaller islands also did well, including Iceland, Greenland, Sri Lanka, Madagaskar, Mauritius and Haiti, but not complex island states such as Indonesia and the Philippines, which ‘despite’ extended lockdowns couldn’t contain the coronavirus.
In addition, countries neighboring China – many of which already had experience with the 2003 SARS-1 epidemic – also introduced early border controls. This group includes Thailand and Vietnam, and in extension Laos and Cambodia, plus Hong Kong, Taiwan, South Korea and Japan, and also Singapore. Thailand, in particular, appears to have been very lucky, as some Chinese tourists could enter the country until March, although requiring a ‘health certificate’ since January.
Even with early border controls, however, a few infected people may already have entered the country. These people need to be identified and isolated very quickly. This can be done in a high-tech way (by rapid PCR testing, as in China, Taiwan and South Korea), or in a low-tech way (by batch isolation, as in Vietnam and Thailand; Vietnam isolated up to 200,000 people).
Moreover, most low-covid countries have been isolating potentially infected people not at home – where they may infect their family and neighbors –, but in dedicated isolation facilities. This is easiest for authoritarian countries (such as China, Vietnam and Thailand), but democratic low-covid countries like Australia and New Zealand have been doing this, too, using empty hotels.
Within Australia, only the state of Victoria and its capital city Melbourne failed to properly isolate infected people and, as a result of this, entered into a nightmarish months-long lockdown.
Are there (non-African) low-covid countries without early and strict border controls? No.
A few countries happened to avoid the spring wave, but got caught in the autumn wave. This group includes the Czech Republic (and most of Eastern Europe) as well as Uruguay in South America; some US states (notably in the Midwest) also belong to this group. Interestingly, in the US, the US President wanted to close borders early, but senior health experts were against it.
The WHO, in fact, even “urged countries not to close borders to foreigners from China”.
The hypothesis of a ‘pre-existing immunity’ due to similar coronaviruses, e.g. in Southeast Asia, doesn’t seem to hold: while Thailand, Laos and Vietnam have few covid cases, their direct neighbors Myanmar (Burma), Indonesia, Malaysia, Bangladesh and the Philippines have many cases.
In contrast to influenza epidemics, primary schools have not been a major driver of the coronavirus pandemic. Secondary and tertiary schools are a different and more complex matter, however.
In conclusion, if you don’t want trouble with the coronavirus, don’t let the coronavirus in. Just a few weeks of delay may make all the difference, as in the cases of France vs. Germany or Norway vs. Sweden. In contrast, once the coronavirus is already widespread in a country, most of the much-discussed measures have turned out to be largely ineffective, and often destructive.
In the case of the highly infectious coronavirus, even the idea of “protecting the high-risk groups” has turned out to be very difficult, if not impossible, in a high-prevalence environment. This is shown by the fact that in many Western countries, about 50% of deaths occurred in nursing homes.
In terms of economic impact, IMF data clearly shows that the harder and longer a lockdown, the stronger the economic contraction. However, most lockdowns occurred in response to an already high infection rate due to late border control. The economic impact moreover depends on the structure of an economy (e.g. the importance of the tourism and export industries).
The question of covid prevalence is different from the question of covid mortality. The latter depends mostly on demographics and possibly on genetic, immunological and lifestyle factors.
Many Western countries and Russia show high-prevalence and high-mortality. Many African countries and India show high-prevalence but low-mortality. Countries like Vietnam and Cambodia show low-prevalence and (likely) low-mortality. Countries like Japan and Taiwan show low-prevalence but (probably) high-mortality, although likely not as high as in Western countries.
Related: Do lockdowns work? Only if you lock the borders down, too (Quillette, February 2021)
Update: Obesity and the Pandemic: New Insights (June 2021)