Lessons From Sweden: Freedom!

Sweden is brought up very often as an example of how to balance the impacts of the pandemic and the economy. Unfortunately, in most cases the data is used in the wrong way to draw black-and-white conclusions. As a guide to how other countries could reopen their economies Sweden indeed provides valuable lessons, but they are a lot more nuanced than most people believe. Here is the nuanced view.

What does the data say

What most people do is look at total cases or total deaths per country. In this perspective Sweden is worse than the average country, ranked 25th worst (cases) and 68th worst (deaths) among 185 countries today. Certainly better than some of the hardest hit countries in Europe even though those imposed long and strict lockdowns. To some people this seems a clear case that the Swedish approach works. However, if a country with 10 times as many people has 10 times as many cases, then they are really not doing 10 times worse. Rather they simply have 10 times more people...

Even in this perspective, Sweden has roughly 10 times as many deaths as neighboring countries Norway, Finland and Denmark. All with similar weather and demographics. This should raise skepticism even for the most ardent believers in the Swedish approach and spike a curiosity to learn more before promoting it.

The correct comparison is to look at cases and deaths PER CAPITA.

For more information on that see for example this post which compares countries and includes an updated Excel file to follow the logic and analyze at will.

In this correct perspective Sweden is the 8th worst country in the world today. And instead of being a factor of 3.75 worse than the average country, it is 90 times worse. This should certainly give us reason to pause...

Of course, looking at country totals, even per capita, does not tell the whole story. To find the truth we need to go more detailed. If we look per county, we see that by far the hardest hit county is Stockholm. Stockholm also happens to have the highest population density (364.9 people/km2), around triple the next highest density county (Sk?ne with 125.7 people/km2). So even correcting to a per capita measure the highest density area got hit harder. We see the same pattern in many other countries: New York City in US, Lombardy in Italy, Paris in France, and now Mumbai (most populated city in the world) is becoming a new epicenter in India. The reason is twofold: more people means a greater chance of someone introducing the virus, and more people crammed together means a lower distance between individuals to spread the virus faster.

This is brings us to:

The factors that impact the pandemic

The most important factors that determine how certain areas fare are, in (most probable) decreasing order of importance:

1. LUCK. Including timing. Some countries did not get travelers to bring the virus at all, or after the impact had already become clear in other countries.
2. Government diligence. How strict are travelers and new cases monitored and isolated.
3. Government preparedness. Were tests, protective equipment, ventilators available and was there a formal nationwide response plan ready to put into action.
4. Government decisiveness. How fast did countries react. How well do they align internally across government levels and with health institutes to provide clear and timely help and guidelines to their citizens and businesses.
5. Citizen cooperation. When governments did/do provide painful guidelines, how well did/do the citizens follow these guidelines. Or in countries where sensible guidelines were lacking how well did citizens respond autonomously.
6. Population density. Areas that are significantly more densely populated tend to have significantly higher infection rates. Especially major cities seem to follow a much different pattern than smaller cities, suburban areas, towns, villages and rural areas. Within all the latter, density seems to play a much smaller role, indicating there may be some critical threshold above which there is trouble.
7. Availability and quality of healthcare. Some countries that see large spikes in cases are able to keep number of deaths down because their healthcare prevents many deaths. Other countries and areas within countries were overwhelmed and death toll skyrocketed.
8. Climate. There is a hope that heat and humidity may help reduce infection rate in summer time or hotter regions. Contrary to popular believe this is neither proven nor made more likely based on existing research. There are too many other factors that dwarf anything we could measure in this regard. Only time will tell if this is true. And since many other coronaviruses do slow down in summer we can hope for the best, whilst we plan for the worst.

Unfortunately, for most countries the ship has sailed on 1,2, and 3, with the exception of ongoing diligence. Countries, not yet hit, can still work to prevent the economy-debilitating responses if they work hard to be ready. We have certainly discovered that economically and socially, prevention is much better than the cure.

How is Sweden different

If we look at mobility of the population of Sweden we can see that even though the government did not impose strict rules, the population did autonomously hunker down. Whilst not as much reduced as some other countries, mobility did reduce measurably. Compared to its neighboring countries that averaged a roughly 40% reduction in citizen movement, Sweden's movement reduced by 20%. This reflects a relatively high mutual trust of Swedish citizens in their government and vice versa, as well as trust in science.

In terms of population density, only two cities and surrounding areas are relatively dense: the capital Stockholm and Malm? in Sk?ne county. The rest of the country is relatively sparsely populated. Malm? did not get hit nearly as hard as Stockholm, which could be due to many of the above mentioned factors, but maybe it just did not hit the population threshold that makes infection go into overdrive. Let it be clear that is just a guess and only one data point. We should really look at this across hard hit countries around the world and look for a pattern before we can make life-or-death decisions on that.

One way Sweden is not different, surprisingly, is that its economy is suffering just as hard as the rest of us. Recent predictions show 2020 GDP shrinkage ranging from 6.9% to 9.7% for Sweden whilst its neighbors predict 6% to 6.5% and the rest of Europe sees GDP shrink by 6% to 9.1%. This dispels the myth that local lockdowns only hurt the local economy. If we relax, but our neighbors do not, we are not likely to see economic benefit. Naturally, the experience for local businesses and individuals is quite different from the economy as a whole.

What are the lessons

We cannot claim that Sweden's model is any better or worse than other models with regards to death toll or economy. Even if it were, it is much too soon to tell.

We cannot claim certainty about anything yet. We need to keep an open mind to all ideas and we need to be vigilant not to make life-or-death decisions on a hunch. Many hunches these days seem to be based on pseudo-science that anyone can easily disprove and then go viral on social media. The hunches are good to have, and most are worth further research. But hunch and pseudo-science based extreme decisions (full lockdown or full open society) are a recipe for disaster. Either for the death toll, for society or for the economy or all of the above. Short-term gains have a nasty habit of leading to long-term or permanent loss.

We can claim that local population density has a big impact. The exact shape of that needs further research. But we can be certain that much of what works in rural Montana will not work in New York City. What did not work in Stockholm may also not work in Phoenix. What worked in Malm? may work in Cincinnati. Any rules and guidelines should be made specifically for local conditions. Country-wide or state-wide orders should not be definitive.

We can claim that government dysfunction and citizen disobedience will be a key determining factor. In Sweden we see that the government created some minimal rules and then trusted its citizens to do what is right. The citizens in turn validated that trust and did what was right. Compare that to countries such as USA, Brazil, and UK, where the government did everything wrong they could possibly do wrong, and citizens logically have a strong distrust of those governments and then go out and taunt fate.

Important note: There are times when government dysfunction can go by for extended periods of time without direct measurable impact. And most of the time and on most subjects citizen disobedience is an expression of civil liberties that should be valued and cherished. This is not such a time, and the pandemic is not such a subject. This is a time where the impact of dysfunction and disobedience is painfully felt in both citizens' lives and livelihoods, businesses' survival and the economy.

If ever there was a time for governments to get their act together, for all parties to cooperate, this is it.

Many countries these days have polarized. This makes it especially hard to get over our differences. But putting world and country above party interest should be everybody's goal. Let's put our differences on ice for now. Until either this crisis is behind us or we can express it peacefully on an election day.

And we can claim that the most important benefit Sweden has over countries with strict lockdowns is freedom! So it did not hurt the pandemic as much as some people feared, and it did not benefit the economy at all as most people hoped. But it left the citizens with their freedom intact. Trust is the key prerequisite for this to be possible. Trust from the citizens that their government will not drop the ball. Trust from the government that their citizens will do what is right.

To take this out of the realm of possible to the realm of tangible. Let us look at the US. Because I live there and because it is arguably the most polarized country right now. If we can make it work here, we should be able to do so anywhere. There is no trust in the federal government, and in many places also no trust in the state government. But in most places there is trust in local government. We need to stop all federally and statewide enforced regulations, possibly with the exception of cordon sanitaires, IF AND ONLY IF, and when it is critical. Local government should be trusted by federal and state governments to do the right thing and to collaborate closely with neighboring local governments. The higher government's role is to provide guidelines and to enable! And it is crucially important that those guidelines are based on real science, not on pseudo-science and certainly not opinion, hunches, or hearsay. When states need to compete for lifesaving equipment or means are allocated based on party affiliation something is going terribly wrong. The higher government also has a role to monitor and correct local abuses of power or unwarranted deprivations of freedom, just like they have on many other issues. The role of the citizen is to follow the local rules and as much as possible abide to guidelines. The local government also has to enforce the rules. But actions that are legal in normal times should never lead to jail time. Reasonably high tickets for violating temporary restrictions as a financial deterrent should be part of the overall plan. This is hard and often painful, but we must trust that everyone is working towards the same goals: to limit deaths and to restart the economy.

That is the real lesson from Sweden:

There can be Freedom only if there is Trust

Now let's start by giving and earning trust. All around.