Did Swedish COVID-19 policy perform well?

Especially in 2020 (but also in 2021-2022), in response to COVID-19 Sweden chose a very different set of measures from many other countries, including voluntary rather than mandatory social distancing. Nonetheless, Sweden did roll out a comprehensive vaccination program. Did the chosen policy perform well?

Search terms:

  • Swedish covid policy
  • Did Swedish COVID-19 policy perform well?

Swedish COVID-19 policy performed well

  •   Argument for As per all-cause deaths from Björkman et al. 2023, Sweden performed better than many European countries implementing a much more stringent lockdown.[1]
    •   Objection It is unclear the countries are comparable since they have different population network densities and topologies. Thus, one may naively compare the Swedish figure of 158 vs. Danish 97, but it can be that Denmark has much denser population network. Thus, the relatively good Swedish figure (worse than Denmark and Norway but not Finland) can possibly be explained not so much by the chosen policy but rather by the population network density and topology.
      •   Objection If most of old and vulnerable people live in large cities, e.g. in nursing homes in Stockholm, the relevant population density can well be comparable in that most of the risk of death exposure is located in densely populated areas. A source for the hypothesis is missing.
      •   Objection It also works the other way: what if the countries are not comparable as for the healthcare quality and density? Thus, what if Germany performed worse than Sweden despite its better and denser healthcare? This would need to be properly investigated, especially whether Germany really has better and denser healthcare.
  •   Argument for Sweden performed well as per the Norberg 2023 article[2].
    •   Objection The article was published by Cato Institute, a libertarian think tank. This makes it more likely to be biased in favor of less severe lockdown.
      •   Comment True enough. One has to look into the article, read it critically and check its sources.
  •   Argument against Sweden performed poorly based on confirmed COVID-19 deaths as per a Los Angeles Times article Hiltzik 2022, which references Brusselaers et al. 2022.[3]
    •   Objection All-cause deaths seem to be a much more reliable indicator than confirmed COVID-19 deaths, in part since different countries count confirmed COVID-19 deaths differently.
    •   Objection An article using the language of "devastating picture of Swedish policies and their effects" and "its record is disastrous" should probably not be taken seriously. Not conclusive, but suggestive.
      •   Objection One should analyze the article's data-based argument rather than attending too much to language.
        •   Comment True in principle, but such signs as this kind of language are a crude indicator that the article author is not serious enough and that ideally, a much better source making the same substantive argument should be found.
    •   Objection In the chart shown in the article, Sweden performed only marginally worse than Germany.
    •   Objection The article does not explain its choice of countries to be shown in the chart; no obvious selection criterion is apparent.
    •   Objection The article does not present any numerical indicator correlation analysis; all its correlation analysis is informal. Thus, the article could present a table of countries and for each country there would be an indicator of severity of lockdown and some indicator of mitigation success. Then, the article could ask not whether Swedish policy specifically performed well but rather whether there is a statistical correlation between severity of lockdown and death mitigation success.
    •   Objection It is perhaps not a good idea to take an article from someone identified as a business columnist seriously. Not conclusive but suggestive.
  •   Argument against Sweden performed poorly as per Brusselaers et al. 2022.[4]
    •   Objection The article abstract states that "During 2020, however, Sweden had ten times higher COVID-19 death rates compared with neighbouring Norway" and provides no figures from 2021 and 2022. Data from Björkman et al. 2023 reveal that ratio relations were very different in the years 2021–2022. As per Björkman et al. 2023, in 2020-2022 Sweden had an excess mortality of 158 deaths per 100,000 inhabitants whereas Norway had 129 deaths per 100,000 inhabitants, far from a 10x ratio. This renders Brusselaers et al. 2022 very inconclusive.
  •   Argument against Per Figure 1 in Norberg 2023 showing cumulative COVID-19 deaths per million for 2022-June 2023, Sweden's 2,322 performed much worse than Norway's 1,024 and worse than Netherlands' 1,309, Denmark's 1,485, Finland's 1,802 and Germany's 2,098. That is arguably not a good performance but rather an average one.
    •   Comment In so far as "good" stands in contrast to "excellent" or "very good", at least "moderately good" can reasonably apply, since Sweden is somewhere in the middle of the pack listed in Figure 1, the worst being Greece with its figure of 3,578.
    •   Objection Different countries count their COVID-19 deaths differently. That is why excess deaths determined from all-cause deaths were argued by multiple serious sources to be a much more reliable indicator than COVID-19 deaths.
    •   Objection Norway could have a lower population network density and a different topology; it could be that Norway's nursing homes did not present an entirely analogous risk of death exposure as Stockholm's accumulation of nursing homes. Admittedly, while this is somewhat plausible, it requires a careful analysis and verification.
      •   Objection This does not apply to Netherlands, which also performed so much better than Sweden.


  1. The Swedish COVID-19 approach: a scientific dialogue on mitigation policies - PMC by Anders Björkman, Magnus Gisslén, Martin Gullberg, and Johnny Ludvigssoncorresponding author, 2023
  2. Sweden during the Pandemic by Johan Norberg, 2023, Cato Institute
  3. Column: Did Sweden beat the pandemic by refusing to lock down? No, its record is disastrous by Michael Hiltzik, March 2022
  4. Evaluation of science advice during the COVID-19 pandemic in Sweden by Nele Brusselaers, David Steadson, Kelly Bjorklund, Sofia Breland, Jens Stilhoff Sörensen, Andrew Ewing, Sigurd Bergmann & Gunnar Steineck, 2022, nature.com

See also


Further reading



Older articles, which fail to account for the full multi-year picture: