- “China Is the World’s Last ‘Zero Covid’ Holdout: The Government Has Staked Its Political Legitimacy On Controlling The Virus” – a New York Times headline
The recent outbreak in the Chinese city of Xian has spurred some of the toughest Covid countermeasures yet. Authorities have shut down the entire city of 13 million people, shuttering businesses, schools and factories, imposing virtual house-arrest on all residents. Leaving home is a violation. Not even food shopping is allowed. Instead, bags of vegetables – “leeks, cabbages, broccoli” – will be delivered to the locked-down inhabitants. Hundreds of quarantine camps are being prepared to hold people who may have been in proximity to an infected individual. The city’s economy is at a complete standstill.
Covid containment policies obviously involve a trade-off between economic outcomes and public health outcomes, and of course these are mutually interdependent. Strict measures (lockdowns, travel bans, etc.) raise the economic cost, but (hopefully) save lives.
So, is China’s hardline approach working? The question has both an economic answer and a medical one.
The Economic Impact
The economic picture in China is murky (as usual). However, even aside from the Xian crisis, a slowdown seems to be taking place across the country. (More on that in Part 3 of this series.)
The official Chinese GDP growth figure for Q3 2021 was 4.9% – the lowest quarterly GDP figure ever (except for the first two quarters of 2020, when the pandemic shock first hit). Western estimates are more pessimistic.
- “Monthly data showed China’s economy slowing sharply.… China’s tough virus curbs mean economic activity could be close to flat…” – Bloomberg
The declining Chinese stock market reflects this. It is the only major financial market to have declined this year.
In the real economy, China’s retail consumption “remains stubbornly weak” and other signs of distress have multiplied:
- “[There is] a slump in retail sales and property construction. Chinese consumption growth has rarely looked weaker… Real spending on consumer goods rose only 0.5% year over year in November… the weakest since at least 2011….The economic damage is becoming apparent.” – WSJ
China has chosen apparently decided to accept significant economic sacrifices to control the virus.
The Public Health Impact
China claims (officially) an extraordinarily low incidence of Covid mortality. However, there are significant gaps in the Chinese data (as described in the previous column). Essentially, Beijing put a halt to the reporting almost two years ago. It means that most of the inferential comparison studies have to leave China out.
The absence of reliable internal Covid statistics forces us to look at comparative benchmarks based on better data from other countries.
The Official Figures (From Reuters)
A comparison of the officially reported Covid Mortality rates for various countries in Asia makes it clear that China is an extreme statistical outlier.
[The source is Reuters; the methodology is described here; the data is shown in deaths per million population. These are not estimates. They are official figures published y the countries in question.]
Developed Asian Economies
Japan (GDP/Capita ~ $41K) and Korea ($32K GDP/Capita) are rich countries, fully industrialized, and important trading partners for China. Singapore ($60K GDP/Capita) is a very rich, very tightly controlled city-state, well known for its strict social policies and vigilant administration. All three countries are seen as having implemented successful Covid containment programs, with mortality rates 10-20 times lower than in Europe or North America.
Yet China claims to have a Covid mortality rate that is 30-50 times lower even than these three comparables.
Developing Asian Economies
China’s cities match up with those in its rich neighbors. But there are more than half a billion people in rural China, with incomes about 1/3rd of the average income of the urban Chinese population. Living conditions and health care deficits are common. Many rural Chinese are still unvaccinated. Socially and economically, rural China looks more like some of its developing Asian neighbors.
Yet, China claims a Covid mortality rate up to 300 times lower than these other Asian nations.
Japan and Singapore are islands, lacking any land borders. Korea has just one militarily closed border. These countries can more easily control traffic from abroad than China, which is a continental power, with many neighbors that share land borders with it, borders that are in some cases quite porous, open to legitimate and illegitimate human traffic. There are hundreds of thousands of refugees from North Korea in China today. Burmese refugees number in the tens of thousands. Smuggling and human trafficking from Vietnam is endemic, where border walls are being constructed.
How does China compare with its neighbors?
China in fact shares land borders with 14 other countries, a total of over 13,000 miles – the longest aggregate land borders of any country in the world. (Yes, even longer than Russia’s borders.) All of these countries are evidently hot-beds of Covid infection and death – hundreds of times higher than China. How does China keep all that bubbling contagion at bay?
Korea: The Best Comparable
The best close comparable for China, in terms of climate, culture, economic development, patterns of commerce, and Covid protocols, is arguably Korea.
South Korea is regarded generally as the “Emerging COVID-19 success story.” It has countered Covid aggressively and successfully.
- “South Korea’s response to COVID-19 has been impressive. Building on its experience handling Middle East respiratory syndrome (MERS), South Korea was able to flatten the epidemic curve quickly… It achieved this success by developing clear guidelines for the public, conducting comprehensive testing and contact tracing, and supporting people in quarantine to make compliance easier. The country successfully managed outbreaks in March and August and gradually gained control of a larger, more dispersed outbreak in December 2020. Overall, South Korea has shown success across three phases of the epidemic preparedness and response framework: detection, containment, and treatment.”
- “[The Korean response] was managed through restrictions on international travel, school closures, targeted suspensions of public gatherings, and closures of public entertainment venues. The primary focus has been a system of testing, contact tracing, and quarantine supported by mobile technology and data analytics.” – Brookings Institute Study
South Korea does not have the exposure to the many porous land borders with other countries that China has. That has certainly helped with the isolation measures, and no doubt kept mortality rates down. Experientially, the Korean Covid regime was similar to the Chinese, except for the city-wide lockdowns. “Korea required all travelers to quarantine in a hotel for 14 days. They had rigorous testing. They had rigorous contact-tracing. They mask up.” [A comment from a Korean colleague.]
Korea, unlike China, has excellent Covid-related data. There are no unexplained gaps, and the registration of cases seems to be very thorough. According to The Economist’s study cited below, Korea’s reported death toll conforms almost exactly to the predicted mortality rate.
Given all that, we would expect a similar public health outcome. Korea’s mortality rate is among the lowest in the world, at 108 deaths per 1,000,000 population.
Nevertheless, China’s reported death rate is said to be 34 times lower than Korea’s.
New Zealand: the Edge Case
Finally, consider the most extreme comparison available.
The island nation of New Zealand is the most geographically isolated country in the world. It has a very low-population density (just 11% of the population density of China). Both factors should be favorable with respect to controlling the spread of Covid. New Zealand is rich, with 4 times the per capita GDP of China. It is peaceful and well-governed. It has strong medical institutions, and universal government-run and publicly-funded healthcare. The country responded quickly to Covid and instituted a rigorous travel ban. The authorities articulated an aggressive “zero covid” type policy early on, which has been described as “an emblematic champion of proper prevention and response to the pandemic.”
- “Informed by strong, science-based advocacy, national leaders decisively switched from a mitigation strategy to an elimination strategy. The government implemented a stringent countrywide lockdown.” – New England Journal of Medicine
The country was able to contain the initial wave more effectively than almost any other nation. The first Covid case in NZ was diagnosed on February 26, 2020, and the last case was diagnosed in early May. (Recently, the Delta and Omicron surges have produced a handful of new cases.)
As a result of these natural advantages and smart policies, New Zealand achieved an extraordinarily low mortality rate of just 10.3 deaths per million – the lowest of any developed country by far. (Australia, with similarly stringent policies had a rate 8 times higher; most European countries were at least 100 times higher.)
Still – Still! – China claims an overall Covid mortality rate three times lower than New Zealand. And if we exclude the single 90-day period of Wuhan/Hubei’s Covid cases from January 1 to March 31 2020 – the only sample of quasi-complete mortality data from China in two years, which accounts for 97% of all officially reported Chinese Covid deaths – then China’s official Covid mortality rate over the past 20 months would be… wait for it…
- 73 times lower than New Zealand’s.
In any data analysis project, when a discrepancy in the data becomes “too large,” one must entertain the possibility that there is either a measurement problem, or a bias in the reporting, or outright falsification.
The Economist’s Study
To address the missing data problem, Western analysts have recently turned to alternative methods for estimating the prevalence and impact of Covid. The Economist has developed a new model for assessing Covid mortality. (The methodology followed by The Economist is fully published here.)
The results underscore the anomalous character of the Chinese data.
The model is intended to correct for systematic gaps and discrepancies in reporting of Covid deaths. These discrepancies can arise from bad record-keeping and administrative inefficiencies, from weaknesses in national governance (e.g., the effects of war, civil unrest), from diagnostic failures (Covid deaths that are mislabeled as due to other causes), or from incomplete reporting protocols (e.g., China does not include asymptomatic Covid cases in its published figures).
Discrepancies may also be the result of deliberate suppression, alteration or falsification of the mortality data for political purposes.
The Economist’s model attempts to estimate “excess mortality.” This refers to the deviation from the past averages and trends in death rates in a given country.
Death rates are normally very steady. For example, the U.S. death rate from 1950 to 2000 varies only about 1% from its average in any given year. So if the death rate rises much above that trend line, there is said to be “excess mortality.” By analyzing this relationship, and using a machine learning model to bring other sources of data to bear, the scope and some aspects of the source of the discrepancy between the reported Covid deaths and the total number of deaths can be estimated, with the goal of illuminating unreported Covid-related deaths.
As far as I can tell, The Economist’s model does not have the ability to discriminate the motive that gives rise to a discrepancy. They don’t differentiate between underreporting due to incompetence vs political manipulation. The model focuses on quantifying the discrepancy relative to the official figures.
Globally, the study concludes that Covid deaths are underreported by a factor of 2 to 4. That is, if the model is correct, there have been between 6 and 18 million more Covid deaths in the entire world than what has been officially reported.
Looking at individual countries, the model further confirms the peculiarity of Chinese Covid mortality figures.
(Only 17000% off? Ponder that, dear reader. Who knew percentages could even go that high?)
Or, to reverse the ratio – China has apparently reported less than 1% of the actual Covid deaths in the country. And keep in mind that almost all deaths of those it has reported – some 97% of them – come from that single 90-day period, for a single city (Wuhan, as noted in the previous column).
Can these results be questioned? Of course. Like any estimation procedure, there is inherent uncertainty. The model is built on various assumptions which can each be critically examined. The model’s output is given as a range, bounded by statistical likelihood parameters – e.g., the 95% confidence interval.
This means that this model cannot be expected to reveal fine differences. The distinction between 200% and 300% under-reporting is probably not bankable, so to speak.
But China’s undercount stands out. No other major country is even close to this discrepancy. The only comparably large figures come from countries that are currently involved in war or large-scale civil unrest, such as South Sudan, Chad, the Congo, Burkina Faso, where it is understandable that public health records may not be in good order. Even some of the most unsettled countries in the world currently, such as Libya (400%), Iraq (900%), Iran (100%), and Afghanistan (900%), Venezuela (1100%) – while they have large discrepancies and are clearly undercounting their Covid deaths – they are no where near China’s ratio of 17,000%.
- China’s reported Covid statistics are the outlier in the global picture; Beijing’s official numbers are 10’s or 100’s of times lower than would be expected from simple comparisons even with countries that have the aggressive and stringent countermeasures (e.g., Singapore, Japan, Korea, Australia, New Zealand)
- After Q1 2020, China stopped providing most mortality data; since that time, China’s reported mortality rate is more than 100,000 times lower than the major Western countries
- China’s willingness to provide data related to Covid in any way has been minimal; the pattern of denial, obstruction, and data destruction is striking
- China has a long history of public health scandals, including faulty vaccines (multiple), adulterated baby formula, and coverups and mismanagement related to the initial outbreaks of SARS (2003), Bird flu (2004), Bird flu again (2013), Swine Flu (2019), and Covid (2020).
Bottom line: China’s numbers cannot be trusted. Which ought not to be surprising, given how little we can trust China with respect to almost anything involving Covid.
In light of all this, the Economist’s study needs to be taken seriously. It is not a definitive conclusion; it is an estimate. It is a serious and well-founded attempt to make up for the appalling gaps in Chinese accounts. The numbers, with whatever caveats, are so large as to clearly indicate that China has been misreporting Covid mortality, at least since April 2020.
As to motives? The NY Times headline puts it succinctly: The Government Has Staked Its Political Legitimacy On Controlling The Virus. It is a battle Beijing cannot afford to lose.