Pubblichiamo in lingua originale un studio sul processo decisionale durante l’emergenza da COVID-19, a firma di Dilectiss Liu, docente di Filosofia all’università di Ludwig-Maximilians-University di Monaco

Introduction

This article explains why different governments have implemented different strategies in response to the COVID-19 pandemic. The article is divided into seven parts. In part 1, the article introduces how decision making works. In 2, we take a look at why strategies for the same issues may differ between governments. In 3, we examine the common assumption that there is a tradeoff between public health and the economy. In 4, the article explains the difference between mitigation and containment. In 5, the article explains how shutdowns and lockdowns can curb a pandemic. In 6, the article explains the importance and difficulty of reference classes in statistical reasoning, especially with regard to the data on COVID-19. In 7, we compare and analyse the strategies of the following governments in response to the COVID-19 pandemic: China, Australia, Japan, and Germany.

1. Decision Making

Prima facie, good decision making seems straightforward: It is a function of beliefs and interests. If I believe that my cup contains coffee, and I want coffee, then I should grab the cup. However, in the real world, there are difficulties with both weighing our interests and settling on certain beliefs.

We often have multiple interests. Even for something as simple as deciding whether to take an umbrella with me, I want to stay dry and to not carry an umbrella. These wishes can come into conflict.

Two factors amplify this conflict. One, we often have no clear idea how our multiple interests compare and rank. I might not know if getting wet is worse than the burden of carrying an umbrella; and if I do think that it’s worse, I might not know how much worse it is. Two, we often aren’t certain of all the relevant facts in the world. I don’t know whether it will rain. I only know that there is a chance that it will rain.

In such a simple situation, we can still have a relatively clear ranking of decisions. If it isn’t likely to rain and I don’t mind getting wet, then I shouldn’t bring an umbrella. If it’s very likely to rain and I hate getting wet, then I should bring an umbrella.

However, in complex situations such as the COVID-19 outbreak, decision making can be difficult. Although there seems to be only one general action plan: to restrict social contact, there are multiple ways in which a restrictive measure can be devised. There are multiple factors at play, such as infection rate, case fatality rate (CFR), required cash flow of the economy, the psychological effect of keeping people home for weeks, the habits of the people living in a certain country, long-term effects etc. Moreover, many of these issues are unclear. We therefore cannot predict how a particular style of shutdown would affect a society, especially in the long run. Governments rely on experts to inform on particular issues, such as public health, psychology, economics etc. They then devise strategies that aims to weigh these expert advices.

2. Strategic Differences

Different societies use different strategies for making decisions, reflected in government policies and individual behaviours. These differences are due to different interests, different situations (and therefore beliefs), and most importantly, the difficulty of predicting the future.

For example, Germans are known to be risk averse. Their decision making strategies lean towards minimising risk. On an individual level, German households on average save a much larger percentage of their income than their Anglophone or European counterparts, and they are more weary of online services. On a national level, the German economy is supported by the middle class. In particular, industrial production such as civil engineering projects or car manufactures. This contrasts with America’s high-risk high-reward economy that relies more on investments and debt, and where the national GDP is supported primarily by the top 20% wealthiest investors. German society’s risk aversion is a cultural phenomenon of valuing security above convenience or maximising utility, but it is also a very real response to combat the unpredictability of the future.

The background situation of a country matters for decision making. For example, given that the USD currently inflates at a rate of 2.33%, and that the EUR currently inflates at a rate of 1.5% in Germany, and given that the American economy focuses on debt and investment, it might be unwise for American households to save their income on par with German households.

3. Public Health and the Economy

We are still in the middle of an ongoing crisis. The number of active cases worldwide is growing, and the economy is still in a state of uncertainty. Therefore we do not yet know the final outcome of different policies. However, we do have a good base for the current analysis.

This pandemic has tested each and every one of us, by appearing to pit public health against economic interests. We often think of a shutdown as sacrificing the economy for curbing the spread of the virus. Many see this as a worthy sacrifice. On the flip side, some folks (esp. from America and Britain) have argued against shutdowns. They see the death of around 3% of a country’s population as a worthy sacrifice against a potential economic depression.

However, the construal of such an opposition between public health and the economy is a mistake. Public health and the economy are intertwined and directly affect each other. If people begin to die or become ill en masse, then the market would see a corresponding decrease in value due to decreased labour and spending. A healthy population is necessary for a healthy economy. On the flip side, the economy is ultimately about people’s welfare. If a country’s economy is in poor condition, then more people are likely to die from poor health due to the unavailability of resources such as food or medication.

In the medium to long-term, we cannot divorce public health from the economy, and we simply cannot trade one off with the other. Governments aim at protecting both in the long run. The apparent ‘tradeoff’ is not between public health and the economy, it is between the damages that is done to our welfare by the spread of COVID-19 versus that of restrictive policies. On the one end, if governments do nothing, then the virus will spread at a rate far greater than our medical systems can cope. The result would be mass deaths. Much of these deaths would be preventable with proper medical care. On the other extreme, if governments implement total lockdowns – banning everyone from leaving their homes, then production would halt, including essential supplies such as water, food, medicine. Furthermore, many people would stop receiving income as businesses go bankrupt. People would quickly begin to starve and scramble for supplies. Therefore, governments must find a balance between the two extremes. We have seen that governments have implemented partial shutdowns or/and lockdowns to curb the spread of COVID-19, allowing essential goods and services to remain running, while providing economic stimulus and relief in the form of cash payments.

Almost all governments have been trying to curb the spread of COVID-19 via reducing social contact. However, the degree to which each government has implemented its restrictive measures differ. This is to be expected, since as per section 2, situations differ between countries, and these differences can be intricate and complex. Governments can choose to either mitigate or to contain the pandemic. To understand the reasons behind choosing one goal over another, we need to understand what each is meant to do, and which measures governments must implement to achieve their goals.

4. Mitigation and Containment

There is a difference between containment and mitigation. The goal of containment is to eradicate a virus from the population. The goal of mitigation is to slow down the spread of a virus, though allowing it to potentially infect the entire population. Whether an outbreak would be contained or mitigated depends on how a virus behaves and how societies respond.

A successful case of containment was with the outbreak of SARS-CoV-1 (commonly known as ‘SARS’) in 2004. Only approximately 8,000 confirmed cases had been reported worldwide, with approximately 10% fatalities. Two crucial factors about SARS contributed to the relative ease of containing it. First, the virus was only contagious after symptoms appeared in the host, which allowed the effective isolation of infected persons. Second, the virus had a high CFR of around 10%. Compared to a less fatal virus, the SARS-CoV-1 would be more likely to kill its host before it could spread.

A failed case of containment was with the HIV virus. The primary reason for our inability to contain the HIV is that HIV hides itself very well. It can incubate for more than 15 years. This means that we may fail to detect its presence in a person even as it spreads to new hosts via bodily fluids. Societal responses won’t help unless everyone is forced to have mandatory testing and certain forms of isolation thereafter. 

The success of containing the SARS-CoV-1 might make us think that perhaps we could also contain the SARS-CoV-2 (COVID-19), given their close resemblance. However, we have already learned of two crucial differences. First, COVID-19 is far more infectious, and studies have shown that an infected host could transmit the virus even before symptoms appear. This makes it virtually impossible to effectively isolate those who have the virus. Second, COVID-19 has a relatively low CFR, with a current official figure of 3.4% and a realistic figure of possibly much lower, given that not all those who have been infected have been tested. Therefore, for a government, it’s worth asking whether we should try to mitigate or to contain the spread of COVID-19.

Governments do not always try to contain a virus. If professionals in epidemiology believe that a virus cannot be effectively contained, then it might be preferable to try and mitigate the virus. This is because a measure for mitigation is less restrictive and disruptive than a measure for containment. For example, some governments such as that of Germany aim to mitigate the virus. Others such as Denmark aim to contain the virus. This difference in the final goal is crucial for determining which restrictive measures to implement.

5. Shutdowns and Lockdowns

A shutdown is the partial or total shutting down of a system. In this case, we are talking about the social-economic system. The goal of a shutdown is to reduce social contact, thereby curbing the spread of COVID-19. It works by restricting or closing the flow of goods and services, thereby reducing the means for people to come into physical contact. For example, closing our theatres is a highly effective means of stopping large groups from gathering in closed areas. We have seen also negative examples where a choir rehearsal has turned disastrous.

A shutdown by itself is not meant to contain an outbreak. It aims to mitigate an outbreak, so that the medical system can cope with those who have become ill from an infection. The degree of a shutdown can range from shutting down only theatres, museums, and places of mass gatherings to shutting down all goods and services, and even parks. Most places have started with a more relaxed shutdown of only theatres and museums, and over time, increased the severity of the shutdown. Almost all countries affected by COVID-19 have implemented some form of shutdown. Some stricter than others. However, many countries aim not to only mitigate, but also to contain the virus. This is where a lockdown comes in.

A lockdown is the isolation of one group from another. A group can be identified as a country, a city, or even an individual. On the one extreme, some of the most lenient lockdown measures have been implemented in Germany, where even the national borders are not completely closed for individual commuters who have a valid reason. On the other hand, Australia has implemented a strict lockdown of both external borders and individuals. People who leave their homes without an officially recognised reason would be now committing a criminal offence.

While a shutdown works passively by coercing people to not gather, a lockdown is an active measure that banns people from gathering. Ideally, if the maximal period of incubation is 27 days, then a lockdown of 27 days should stop new infections from happening across isolated groups, since the targeted groups would no longer be able to come into contact with each other. After 27 days, all those who had received the virus before the lockdown should show symptoms. In other words, all cases of COVID-19 would be known, and those who have been infected could then be isolated and the rest of society could then return to a normal working life.

However, in reality, no government can implement a complete lockdown that cuts all social contact. People need food and medicine, and possibly transportation in order to get food or medicine. Even if we rely completely on delivery services, the postal workers could also transmit the disease to previously uninfected people. This is not to mention how challenging it is for most people to stay home without suffering distress, which can lead to public unrest. An example of such a challenge can be seen in recent debates on whether people should be allowed in parks. There are arguments for, and arguments against.

That being said, even if social contact cannot be completely cut, it is still significantly cut with a lockdown. Therefore, a lockdown would at least slow down the infection on top of what a shutdown is doing. Note that this is not reflected immediately in newly confirmed cases, since there would be a delay between infection and positive testing, and this can depend on many factors such as the availability of test kits. Furthermore, people usually only take tests after symptoms develop, which means that this delay between infection and confirmation could be up to 27 days.

6. How to Interpret Data

A dataset by itself presents no useful information. If I tell you that there are 10,000 deaths, you would be puzzled and wonder what in the world I’m referring to. You might wish to ask me questions such as: In which period did these deaths occur? Where did the deaths occur? What caused these deaths? If I tell you that these deaths are the total number of people on earth who died in 2019, you might question whether I was missing a few ‘0’s. However, if I tell you that these were the deaths that occurred last month in Italy, due to some new disease, you would again doubt my claim, but in the opposite direction – you would be shocked at how high it is. The extra information I gave in this example – period, population, cause – provide the context in which our data could be interpreted. In statistics, these background information form the reference class with which we could compute the probability of some event from data.

When we estimate the probability of some event, we are usually thinking about conditional probability. When you read about the probability that it will rain tomorrow, you are reading about the conditional probability that it will rain given that the forecast is correct. This is represented as P(rain|correct forecast). Intuitively so, the conditional probability is computed upon the probability that the forecast is correct: P(correct forecast), which is known as the prior. We often neglect the prior because we tend to assume our prior to be 1, which means we take it to be true, however unjustified our assumptions may be. However, for complex issues such as considering the CFR of COVID-19: P(death|COVID-19), and how governments should respond, it is important to take priors seriously. For example, although the estimated P(death|COVID-19) on the 3rd of March is 3.4%, the actual CFR is dependent on a number of factors. If the population tends to be older, then we would actually want to know P(death|COVID) given that the infected person is, say, older than 45: P(death|COVID & >45 yrs old). In that case, we have a much higher number of 23-47% rather than 3.4%. Moreover, there is likely to be a general misestimation of CFR, since not every infected person has been positively tested. This can be due to the limitation of tests, to false negatives, to the existence of asymptomatic cases, to the number of young people who simply cannot be bothered. So the number 3.4% is actually the probability of death given that one has been positively tested with COVID-19: P(death|COVID & positively tested). It is not the probability of death given that one has been infected with COVID-19: P(death|COVID).

The CFR of 3.4% is based on the reference class of the positively tested population. This is an example of a faulty reference class, though arguably the best we could do with our limited data. We must understand that deciding on a reference class isn’t always straightforward, and even if we do have an ideal reference class on paper, it is not always easy to obtain the data. Consider the question: What would be the chance that you would come into contact with an infected person on the 8th of April if you were to be in New York City? Assume that there weren’t any restrictive measures taken and people go about their lives as usual. Also assume that people do not reside in high rise apartments and spreads around randomly. Even then it is very difficult to estimate the likelihood in question. At first, it may be tempting to simply divide the number of confirmed active cases by the total population. The confirmed active cases per million population of COVID-19 in NYC stood at around 9000 as of the 8th of April. Then the percentage of the population who had been confirmed to be infected was around 0.9%. However, this would be a fairly inaccurate number for estimating the chance for you to come into contact with an infected person.

As mentioned earlier, the number of confirmed cases does not reflect the number of actual cases. It depends not only on how much of the population have been infected, but also on how much of the population have been tested. The tests per million population stood at around 17,471 which entails that 1.7% of the population had been tested. One might want to say that since only 1 in every 57 people had been tested, we could simply multiply the number of confirmed cases by 57. However, this would only work if the tests had been conducted randomly among the population. This clearly isn’t the case – most of the tests had been conducted, rightly so, on those with symptoms. So what we need is a way to estimate how much of the untested population were infected. One reference class we could use is the number of people who had been confirmed to be dead from COVID-19. Some might object that not all deaths due to COVID-19 have been accounted for. Nonetheless, it’s still more accurate than using confirmed cases. So now we need a more accurate CFR.

There are various strategies we could use, such as using statistics from a country with a relatively low CFR but high test rate. However, we have a relatively more robust reference class for CFR in the special case of the Diamond Princess cruise. There the entire population had been tested, and it turned out that on the cruise, the CFR is 0.99%. Of course, the demographics is skewed towards the more elderly. On the other hand, medical care was available on command. We could provide a more detailed estimate of CFR by investigating the demographics on board, and compute P(death|positively tested & age group x). Then we can use that for the corresponding death rate per age group in NYC and estimate the number of infected population. Moreover, we can also improve our accuracy by considering the false positives and false negatives of tests. However, we are using an example to illustrate the difficulty of interpreting data, and this simplified calculation should be a warning against those who jump to conclusions due to unfounded claims.

Suppose that approximately 4,100 in NYC died from COVID-19 as of the 8th of April. Then the total cases, given our assumptions, should be 4,100 multiplied by 101, which is around 414 thousand, or around 5% of the NYC population. This rate is P(COVID|from NYC). This would be the probability of you coming into contact with an infected person given that you have come into contact with a person from NYC. If our question asked about your chances only given that you are in NYC, then we would also need to consider the population density of the city.

7. Strategies of Governments in China, Japan, Australia, Germany

Now let us compare the strategies of China, Australia, Japan, and Germany. China and Australia represent governments who implemented a relatively strict measure. However, they present very different background situations such as density, political system, economic system, culture, timing. Japan represents the other end of the spectrum, with very lax measures. Germany is somewhere in between. The aim here is to understand why different measures have been implemented and what kind of outcome we could expect.

7.1 China

The Chinese government implemented a very strict measure – a heavy lockdown along with a severe shutdown. They completely locked down the city of Wuhan (where COVID-19 was first reported) when there were 830 confirmed infections and 25 confirmed deaths, on the 23rd of January. The lockdown was almost total. It eventually blocked travel across the city borders. Sometimes local authorities went as far as physically blockading or destroying the roads that connected different regions. On an individual level, people were eventually banned by early February from going out for anything other than for necessity such as grocery shopping, and even grocery shopping was restricted. High-tech surveillance for COVID-19 has been implemented by tracking testing and social contact via their essential phone apps and face recognition cameras. This style of lockdown was eventually implemented across several Chinese cities as the virus spread. The number of newly reported cases began to decrease without salient rebound since the 13th of February, almost three weeks after the initial Wuhan lockdown. As a result, lockdowns and shutdowns, including restaurant shutdowns, have been gradually lifted city by city. Since the 8th of April, even the lockdown on Wuhan has been lifted. Although, mobile surveillance continues to track movement and possible COVID-19 infections. Social distancing measures continue to take place.

There are several important reasons why such a strict measure was implemented beginning at around 0.6 confirms cases and 0.017 deaths per million people. The goal was clearly to contain the virus as soon as possible.

First, a lockdown was implemented because it was speculated at the time that an infected person could transmit the virus without symptoms. This is now considered to be highly probable. As we have explained, a lockdown works by in effect weeding out those who are infected but show no symptoms.

Containing the virus as soon as possible is beneficial not only for public health, but also for the economy. In Chines culture, people believe in and follow a practical maxim based on the proverb: ‘It is better to suffer short-term pain so that one will not have to suffer long-term pain.’ Moreover, Chinese New Year is a major public holiday in China, so most non-essential businesses were already shut as per usual. As a result, the damage from a restrictive measure could be mitigated. Shutdowns were naturally in place, and only needed to be extended rather than initiated. Moreover, the Chinese economy has historically focused on self-sufficiency. It is well diversified with large market shares in every major sector. The inability to trade for several weeks would unlikely impact the Chinese economy as severely as it would for nations that rely more heavily on trade.

There are key reasons why the Chinese government was able to swiftly implement such a strict measure. First, the relatively centralised government was able to pass new laws more easily than their Western counterparts. Second, high-tech surveillance already existed in China. The infrastructure was readily available for projects such as monitoring the dynamic demographics of COVID-19. It therefore costed little resources to implement a lockdown with tracking. Third, a key difference between Chinese and Anglo-American culture is that Chinese people tend to be more group oriented rather than individualistic. They are by large ready to give up individual utility for group utility, including the freedom to go out. This contrast can be seen across the board, and is most evident in the constant public dissent in the UK or Australia when similarly strict measures have been implemented.

7.2 Australia

Since the 19th of March, Australia has shut its external borders and closed its airports, at 756 confirmed cases and 7 deaths, which is about 31 confirmed cases and 0.27 deaths per million population. Since the 30th of March, Australia has been implementing a measure similarly strict to the measures implemented in China. This was at 4460 confirmed cases and 19 deaths, which is around 181 confirmed cases and 0.76 deaths per million population. For individuals in NSW, leaving home without one of the 16 officially listed reasons could land one in a hefty fine of 11,000 Australian dollars (AUD) or 6 months jail time. Recently, the Australian government has introduced high-tech surveillance like China and S. Korea, though on a voluntary basis. The Australian government plans to enforce the lockdown measures till at least the end of June. According to official data, the number of active cases have been dropping since the 5th of April without rebound thus far. Currently, the Australian government is issuing relief payments for businesses to keep them afloat, and for individuals with low income. There are several reasons for the strict measures in Australia, many of which differs from the ones for China.

The goal here is also to contain rather than to merely mitigate the virus. Although Australia has an extremely low physiological density (population density by arable land), at around 50 people per square kilometres, Australia is entering winter, as the northern hemisphere approaches Summer. The symptoms of COVID-19 is similar to that of influenza, which thrives in winter. Therefore the flu season can make the situation more complicated if cases increase exponentially from March. Furthermore, based on our knowledge of other coronaviruses, which all thrive in colder weather, the new COVID-19 could behave similarly, though our evidence is inconclusive. Nonetheless, the government has a strong incentive to play safe.

Public compliance is low in Australia. This is reflected in the recent need to pass a law that fines those who cough deliberately at officials, then claiming to have contracted COVID-19.

Resources, including medical supplies such as masks have been depleted already due to the ongoing heavy bushfires they have sustained since June 2019, infamously known as the Black Summer.

The Australian economy is in a dire situation. For example, the AUD had began to crash rapidly in early March. The Australian economy, unlike the Chinese economy, is much more specialised, with a focus on service such as higher education and tourism. Within the industries, Australia relies heavily on imports of material and food and exports of energy and natural resources such as coal or cotton. This limitation is due to the unique and protected ecology and a relatively weak manufacturing industry. To make things worse, the Australian economy, like the American one, has debt flooded markets. Interests rates for loans are relatively low in Australia, especially with regards to real estate. For a similar property in a similar neighbourhood, mortgage is often on par with rent, and sometimes even cheaper if one decides on a long repayment period. This encourages investors and even wage workers to buy properties as soon as they have enough for an upfront fee. With the emergency 130 billion AUD stimulus, Australia is therefore facing a financial crisis similar to that of America, though less obviously so due to the relatively miniature scale. Therefore, Australia cannot afford the added uncertainty of a lax measure, especially as it has been inflicted with heavy bushfires and is about to enter the flu season.

7.3 Japan

The Japanese government has implemented one of the weakest measures – a partial shutdown (officially known as a ‘state of emergency’) beginning on the 7th of April, applying to Chiba, Kanagawa, Saitama, Osaka, Hyogo, and Fukuoka, 7 of Japan’s 47 prefectures. This means that Japan has implemented an official restrictive measure at 4,257 confirmed cases and 93 confirmed deaths, equivalent to around 33.5 confirmed cases and 0.73 deaths per million people. Not only are the implemented measures lax, they have also been introduced relatively late. What’s more puzzling is that Japan has two key factors that make it more susceptible to the COVID-19 pandemic. First, it has one of the highest physiological densities on earth, at 2,886 per square kilometre, more than double of the physiological density in China. Second, Japan is the second oldest nation on earth after Monaco, at a median age of 48.4. On the other hand, although the growth of new cases and deaths in Japan is still accelerating, with no signs of a turn around, their growth rates have been relatively low compared to most countries with a similar or even lower density. This apparent contradiction is worth exploring.

The main reason for the relatively late implementation of any restrictive measures by the Japanese government is due to their somewhat unique and unfortunate economic situation. Japan is the fourth largest economy in the world by nominal GDP in PPP. However, like Australia, and perhaps even more unfortunately so, Japan has a lack of arable land and natural resources. They rely on imports and exports, especially for energy and food. This is reflected in the shocking -1.8% GDP growth in their fourth economic quarter of 2019, as other countries began to slow down trade, especially China and America, Japan’s two largest trading partners. For reference with the other top 5 largest economies: the growth rate of GDP in the fourth quarter of 2019 for America was at 2.1%, China at 1.5%, Germany at 0%, and India at 1.1%. This has been further amplified by their economic decline throughout the 2019 economic calendar, with a 0% growth already in the third quarter. On top of all this, Japan is struggling to economically recover and adapt its noticeably rigid workplace rituals and rules amidst this crisis.

There is a very important reason why even when the Japanese government did implement measures to mitigate the outbreak, the measures are at most giving local governors the power to close down businesses or public institutions. It appears extremely lax in comparison to measures by governments in China, S. Korea, Australia, Russia, Denmark etc. The main reason is because the Japanese constitution mandates strict protections for civil liberties, in light of authoritarian repressions during and before WWII.

Other factors may have contributed to the apparently contradictory statistics we see – on the one hand, an old and densely populated country on no restrictive measures, and on the other hand, a relatively slow growth of deaths due to COVID-19. This in turn may have contributed to the lax measures from the Japanese authorities. One, the Japanese are known for public order and for following rules. In particular, the society is used to the rule of wearing masks and practicing social distancing during an illness. This contributed to the fact that even without an enforced shutdown or lockdown, the public would voluntarily impose strict measures upon themselves. Two, Japan is one of the most socially isolated societies. Two key reasons contribute to this: a highly competitive and strict work environment for young people, and a thriving virtual marketplace that keeps individuals entertained amidst their busy lifestyle. This is well documented and reflected in their thriving service industry catered for lonely individuals. This may sometimes take the form of celebrity hosts/hostesses, and sometimes take the form of lonely deathbeds for the elderly. In fact, the high median age is due to their low birthrate, which is a consequence of their isolated lifestyle. The Japanese people are so used to being alone that many of them would not mind staying home during this pandemic. This is in stark contrast with Europeans who tend to prefer going out and socialising. A clear and notorious example of this is again, the violation of lockdown orders in Britain.

7.4  Germany

The German government has been implementing a partial shutdown and lockdown beginning 22nd of March, currently due to end on the 19th of April. The shutdowns included venues of mass gatherings such as museums and theatres. However, restaurants and cafés were exempt so long they observed certain restrictions. There has been an introduction of partial border controls for neighbouring countries beginning on the 16th of March, with public transport cancelled for cross-border movement to and from Switzerland and France. For those within Germany, varying degrees of lockdowns have been implemented, with Bavaria and Saarland being the strictest, ordering people to stay home. In most of the country, public gatherings of more than two people are banned, unless they are from the same household. This measure stands between the strict measures represented by China, Australia, S. Korea and the lax measures represented by Japan and Sweden. On the 22nd of March, the shutdowns and lockdowns commenced at 24,873 total confirmed cases and 94 deaths. This is around 297 confirmed cases and 1.12 deaths per million population, making it effectively later than even Japan in implementing a restrictive measure. As of the 7th of April, the number of active cases in Germany began to decrease.

There are two key reasons that contributed to Germany’s relatively late shutdown. As of early-mid March, the EU officials did not see closing borders as beneficial for curbing the spread of COVID-19. Therefore, the German government had been reluctant to violate the Schengen agreement and reintroduce border controls. However, this quickly changed for two reasons. First, the numbers of newly confirmed cases and deaths were rapidly increasing in Germany, Switzerland, France, and Italy. Second, even Schengen countries that were less severely affected by COVID-19 jumped to close their borders, unilaterally and effectively nullifying the Schengen agreement. For example, on the 13th of March, Czech Republic announced a stringent border closure, effective beginning on the 16th of March that banns foreigners from entering and residents from leaving. This means that they had decided to close their borders at 141 confirmed cases and 0 deaths. That was at around 13 confirmed and 0 deaths per million population.

The second key reason is the rapid initial spread, which is due to a combination of other factors. Germany’s initial 16 confirmed cases in Bavaria had all recovered by the 25th of February, which rendered Germany effectively free of COVID-19. However, the second wave of the outbreak that spread from northern Italy was extremely quick and unmonitored since there were no border controls within the Schengen area. Furthermore, the outbreak intersected with Germany’s annual February festival “Carnival”, which has been celebrated most fervently in Cologne and the surrounding areas in North Rhine-Westphalia. This had most likely contributed to the numbers that by the 29th of February, there were 68 confirmed cases in North Rhine-Westphalia alone, accounting for more than half of Germany’s total confirmed cases. This was despite the fact that the second wave of the virus was first observed in the southern state of Baden-Württemberg, where it stood at 15 active cases on the 29th. Due to the infectivity of this virus and the nature of exponential growth, Germany had 1,545 confirmed cases and 2 deaths by the 10th of March, 5 813 confirmed cases and 13 deaths by the 15th of March, less than three weeks since the initial 2 confirmed cases of the second wave on the 25th of February.

As for the relatively moderate measures, there are very specific reasons. On the 11th of March, the German government had already decided that since vaccines cannot be expected to be in the market anytime soon, it is likely that most people will eventually contract the virus. Therefore, the strategy of the German government was never to contain the virus, but to mitigate it. The goal here is to slow down the spread sufficiently so that the German medical system can cope with the influx of patients. One of the key reasons why there were so many deaths in Italy was due to insufficient medical resources. At times, doctors had to decide between who to treat. So the key is to buy time. For Germany, this appears to be a sensible strategy.

Germany is able to afford a long-term strategy that tightens and relaxes measures over time as needed. The German economy is run quite differently from an American-styled one. In short, the American economy is run on investments and debt, thriving on low interest rates and fast economic growth. In contrast, Germany has a big pile of cash. German households on average save around 10% of their disposable income, twice than that of the European average. While this is arguably a suboptimal economic strategy in times of thriving economic growth, it provides security during a crisis. Americans cannot afford a long-term measure that harms their economic growth, since they do not have reserves to fund even basic needs, and a recession would devalue investments. This is evident by the various desperate measures from the American government as of late such as printing 2 trillion USD and prematurely and wrongly announcing the end of the oil price war between Russia and Saudi Arabia in order to stimulate market prices. In contrast, Germany has only passed particular tax relief measures for businesses.