When the coronavirus pandemic gained momentum in the spring, Sweden chose a less restrictive containment strategy than most other countries did, including its Nordic neighbors. Some medical experts and economists, in and out of Sweden, have criticized its policy for being too lenient. Others, from Elon Musk to National Review columnists, have hailed it as a role model for allegedly keeping the economy open, staving off the consequences of a harsher lockdown. President Trump’s medical adviser Scott Atlas has advocated that the United States adopt the Swedish approach to the pandemic.

Sweden’s strategy indeed likely helped the economy — but this came at too high a cost, in terms of lives lost. Taking a similar approach in the United States would, in all probability, be even more costly, because unlike Sweden and other European countries, the United States does not have a centralized, publicly funded health-care system with universal coverage.

Observers outside of Sweden have often misunderstood its handling of the pandemic. It is true that the government never imposed a formal lockdown and that day-care centers and elementary schools remained open. But the Public Health Agency has strongly recommended social distancing, working from home and avoiding unnecessary travel, among other precautions; compliance was high. The agency instructed people above 70 years of age to avoid socializing with others, and visits to old-age homes were banned. The government prohibited public gatherings of more than 50 people. The Swedish summer holidays, when many people leave the cities and towns for their summer houses in the countryside, may have worked as an informal lockdown, slowing the spread of the virus.

Even for observers within Sweden, it has been difficult to ascertain the rationale behind its pandemic strategy. Throughout, the Swedish Public Health Agency remained reticent about the motives behind its policy approach. Officials did not explain how they weighed economic considerations and wider public health concerns, and whether they drew up policies with an objective of achieving herd immunity or based on a worry that rigorous restrictions would be unsustainable.

It might be most accurate to characterize the Swedish “strategy” as one that began with misjudgments of the risk of large-scale spread and the extent of asymptomatic contagion. The health agency did not try to dissuade families from going skiing in the Alps during the winter holidays in late February, though reports regarding an outbreak in northern Italy soon surfaced. A recommendation to avoid unnecessary travel to that region did not come until March 6. When travelers from the skiing resorts returned, they were not quarantined. In media interviews, representatives of the health agency even criticized employers who, on their own initiative, quarantined personnel returning from hot spots. The other Nordic countries imposed limits on public gatherings of five to 10 people in mid-March, but Sweden did not introduce its much higher limit until the end of the month. Even as infections surged in April and May, test-and-trace-programs were not boosted, on the grounds that they would not be effective in a situation with large virus spread. These programs began to be substantially expanded only in the late spring. Very recently, in mid-September, the Public Health Agency recommended, for the first time, that people should quarantine themselves if someone in their household is infected.

What has been the economic effect? Like other countries, Sweden has been hit hard economically. During the first six months of 2020, the gross domestic product fell by 8.5 percent. Unemployment is projected to rise to almost 10 percent in the beginning of next year.

The drop in GDP is considerably smaller than in southern European countries and the United Kingdom, and one to three percentage points smaller than in Denmark, Germany and the United States. The GDP fall, however, is larger than in Finland and Norway. It’s difficult to meaningfully evaluate the impact of different coronavirus strategies using this metric, because of the differences between the countries’ economic structures. For example, Sweden depends less on tourism, an industry hit particularly hard by the pandemic, than do Italy and Spain. Still, the lenient Swedish approach to the pandemic, involving fewer formal restrictions, likely did dampen the economic impact of the pandemic.

But the death toll here has been much higher than in our Nordic neighbors. As of Tuesday, the cumulative number of deaths from coronavirus infections per million people was 52 in Norway, 64 in Finland, and 118 in Denmark, according to Johns Hopkins University. In Sweden it was 581 — not that far below the United States with 673. Comparisons of infection rates are less reliable because of differences in the testing volume, but here, too, Sweden stands out, relative to its neighbors. Registered cases in Sweden are slightly above 106,300, compared with around 13,800 in Finland and 16,600 in Norway — each with about half the population of Sweden.

Have the economic gains from the Swedish strategy been worth the cost? The following are crude back-of-the-envelope calculations.

Assume that the differential Swedish approach dampens the GDP fall this year by 1 percentage point. This represents a gain of approximately $5.6 billion. Also suppose that the approach has caused 5,000 extra deaths — a reasonable guess from comparisons with other Nordic countries. How could one estimate that loss of life in economic terms? The value of a statistical life, used by the Swedish Transport Administration in its cost-benefit analyses of investment in traffic security, is approximately $4.6 million. Using this number, the economic cost of lost lives would be as high as $22.9 billion — clearly outweighing the benefits from the smaller GDP fall. One might argue that the value of lost lives should be set much lower, as the vast majority of deaths have been among elderly people, with fewer years left to enjoy life: 89 percent of the dead in Sweden have been above 70 years of age, and 67 percent above 80. One reaches the break-even point in my calculation if one lowers the value of an average life lost to $1.12 million. For the Swedish approach to be “profitable,” the average life lost must be valued lower than that.

A more complete analysis should consider other factors as well: future health consequences for covid-19 patients, the crowding out of other medical treatments during the pandemic, and the benefits of maintaining a more normal life and keeping more people attached to the labor market. But even a simple calculation, looking only at lives lost, suggests that the costs of the so-called Swedish “strategy” were too high. One thing Sweden may perhaps have done right was to keep day-care centers and elementary schools open. Closing them would have prevented many parents from working, at great cost to their economic output and their livelihoods, and to pupils’ opportunities to learn. There appears to be no evidence that these institutions played an important role in spreading the virus in Sweden: An early study found that day-care personnel and teachers were not at higher risk of infection compared with the labor force in general. There is clearly a need, however, for more research into the potential role of schools in viral spread, in Sweden and elsewhere.

The jury is still out on how well Sweden copes with the pandemic in the longer run: Case numbers, hospitalizations and deaths fell to low levels in August and early September but are now rising again. Based on the numbers so far, however, it appears that Sweden’s failure to adopt a more cautious approach in the early phase of the pandemic caused an unnecessarily large number of deaths, most of them among the elderly. In my view, one would have to attach an unreasonably low value to their lives to conclude that the economic gains outweighed the human losses.

Read more: