Webinar Supplement: Coronavirus in Scandinavia
Stats, links, and numbers for 2020–05–06 Webinar with the Bastiat Society, ‘International Comparisons of COVID-19 Responses: Open vs Closed and Lessons to be Learned.’
In this piece I’m attaching some of the numbers, quotes, links, and further information that I draw on for today’s lecture with the Bastiat Society’s D.C. chapter. Some of these are updated versions of data that I have presented in several articles at the American Institute for Economic Research:
March 31st: What Has Sweden Done Right On Coronavirus
April 18th: The Scandinavian Experiment: Open vs Close
May 1st: Individual Freedom Works for Disease Mitigation Says WHO
All data are from publicly available and easily accessible sources. Don’t hesitate to contact me over email or twitter and I’ll be sure to help you out.
For my part of the lecture, I aim to get a few major messages across — particularly debunking some myths about Sweden’s corona-fighting as superficial or incomplete stories fill the media: it is not true that Sweden has turned into a laissez-faire paradise as libertarians around the world seems to believe; neither has the Scandinavian country done nothing, meerily sacrificing their elderly to the god of Mammon.
Contrary to what most media outlets seem to suggest, Sweden is not pursuing a “herd immunity” strategy — and it’s questionable that anybody even could at this stage as we don’t quite know if having had the virus confers immunity on the infected person.
What I’ve stressed in writing and public appearances so far is that Sweden is opting for a slightly different trade-off between protecting public health and protecting economic life — not intentionally, I believe, but nevertheless an outcome of their approach.
Note that I said “slightly” — not revolutionary different: gigantic support packages for businesses and individuals rival the American ones in size. Monetary policy is very similar: liquidity for banks and asset purchases by the Riksbank. We’re talking loans and grants for companies specially targeting airlines and firms promising not to shed their workforce; changes in unemployment benefits and favorable rules surrounding sick leave.
The same arsenal every other country applies.
Outcomes?
While there are tons of things that differ between countries and their experience of corona, we only get to see one thing: the sum outcome. From that we have to tease out what variable interacted with what shock to produce this outcome. To name a few: countries have had different initial shocks:
- In contrast to Norway and Denmark, Sweden had thousands of citizens returning from vacations in the Alps, rending moot their neighbours’ strategies of closing borders.
- Allegedly the strain of the virus can differ in which case death rates between identical countries reacting identically would differ.
- Underlying differences that always apply when we compare countries: health, culture, institutions, traditions
- And then of course the country’s response.
What we tend to do is associate a country’s outcome — which depends on interaction of all these factors — with that country’s response. Which is why the Swedish response stands out — and why the comparatively high Sweden death rates per capita become a point on contention.
Two more caveats:
At this point we have really poor data — even over those dying. There are inconsistencies in reporting; countries count covid-deaths differently; there are delays in when they arrive and how global data source report them. The damage to the economy is even harder to assess: we don’t get (reliable) GDP numbers for a few more quarters — and preliminary Q1s that are coming in both miss most of the effect and are likely to be revised later on.
Secondly, the relevant trade-offs are not clear. The only thing that matters is not number of people dying from covid: it might not even be the sum total of people dying, as other dimensions matter too: social, liberty, financial, economical. In one piece I wrote:
Health matters more than short-term income, but whoever tells you extremities are fooling themselves. “Muh GDP” is as wrong as “you can’t put a price on human life.” Yes, we can — and yes, we should. Anything else is madness.
Just gimme the God Damn Numbers
Having said all of that, I constructed some vague comparisons between the countries for which I had most insight. We do know — roughly — how many have died so far, and do we know — again roughly — how many people have lost their jobs. So let’s compare them as see if those numbers make sense.
Scientifically, they don’t. What we ideally wanna get at is the counterfactual: was the another trade-off available such that politicians could have saved some more lives by locking down earlier and/or locking down later? Is the financial and economic damage — most of which we haven’t seen yet — worth it?
What I think this table tells me is that it compares the trade-off we actually made. While we’d want to know if there were others available, that’s not really something we can easily gauge at this stage.
As people being treated in intensive care seems to me as a sizeable damage, I reported rates and job losses for deaths only and for deaths+total ICU treated.
Sweden (Health, job losses, news), Norway (job losses, health numbers), Denmark (Health, Job losses), U.S (Job losses from Labor dept, health from CDC and JHU). All numbers updated as of May 4th.
It is clear that Sweden’s per-capita death rate exceeds its Scandinavian neighbours by a lot (but interestingly enough not as high as most other hard-hit European countries: UK, Spain, Italy etc). It’s a bit of a stretch to use the entire country’s population, as — much like the U.S. — it’s a big-city story: Stockholm and its metropolitan area accounts for about two-thirds of all deaths, a bit like how New York City accounts for 30% of American deaths.
Bret Stephens writes aptly in New York Times that what applies for New York might not apply and/or be reasonable for America at large.
Some other relevant information is that the S-curve (not exponential! is clearly visible in the Swedish numbers — and, as far as I can judge the British regional numbers as well:
Some really interesting economic data came out mid-April when credit card companies compared weekly household consumption across different sectors with the same week previous year. The results were quite impressive:
- Obviously, spending on air traffic dropped by about 100%.
- Spending on hotels, restaurants, and clothing dropped to levels at about 50% (not zero!) of what they usually are during Spring weeks. Obviously e-commerce has boomed, but that’s from very low bases (1–2% of total spending).
- Spending at supermarkets and pharmacies boomed during a few weeks and have exceeded 2019 spending by roughly 10–15% per week. Again, not very surprising.
Here’s the really cool stuff:
- Spending at building suppliers fell during mid-March to levels at about 80% of their predicted level, after which they delivered a V-shaped return to levels only about a tenth below were it was last year.
- Household electronics boomed to about 150% of their 2019 levels — and has stayed that for weeks.
Although household spending is only a small part of the economy (and can’t be sustained unless there is associated production) it shows some pretty startling things: Swedes working from home set up work offices like crazy, buying needed equipment; they started those home-construction projects they have pushed aside — and they stocked up on food and medicine and the rest of things we’d expect people to buy during these times.
So what?
Well, what’s so interesting about the Swedish approach so far is that it seems much more sustainable than many other countries: while losses are huge and small businesses are suffering, Sweden is not “holding its breath”/”hibernating” like many other economies in the world are — and so can sustain these measures for a lot longer than many other places.
While the devastation and deaths and despair suffered so far are huge, we won’t quite know how bad and whether it was worth it until the dust settles.
