Sweden, Covid and lockdown – a look at the data

Over the last year, the debate about lockdown has been driven to extremes – everyone has, by now, made up their mind. Sweden has been used as an example of either a liberal heaven or Covid hell. To the outside world, Sweden is a country that defied lockdown, carried on regardless and ended up with what is (now) the highest case-rate in Europe. In reality, Sweden shows that you don’t need lockdown to significantly reduce mobility: it forced down two waves. It failed to protect care homes, leading to a scandal of thousands of avoidable deaths. But the question is whether, by avoiding lockdown, it managed Covid while minimising damage to the economy, society, healthcare and schools. I looked at this in my Daily Telegraph column a few weeks ago. Seeing as this is such a contentious (and misrepresented) issue, I thought I’d include some of the data and sources here for those interested.

Sweden’s story is more complicated than the UK narrative allows. We tend to see the problem only in terms of Covid cases, deaths, vaccinations etc. The BBC reads out these figures every day on the news – but we seldom hear about the aspects of the pandemic. The effect on society, schools (Swedens were kept open for the under-17s), unemployment, the public finances and the likely long-term implications of wider economic damage. The fall in Sweden’s urgent cancer referrals, for example, was far less than that of the UK – meaning fewer avoidable deaths.  Sweden also argues that lockdown brings isolation which can be fatal to the elderly.

Lockdown-related damage has (controversially) always been excluded from calculations used inside UK government.  Seen through only one dimension – the spread of Covid – Sweden looks like a flop. But add other factors (as Swedes have always done) and the picture changes. 

Let’s start with a table where Sweden looks bad: Covid infections. Since the pandemic started, its per-capita infections have been amongst the highest in Europe – and remain so today.

But when you switch to looking at Covid deaths, Sweden about average How can that be, given that it has tolerated such high Covid cases for so long? Tegnell says that at present, the link between high cases and deaths has been broken by vaccinations: Sweden’s targeted vaccination programme means ‘we should not have to experience these high death rates again’. So a new wave of cases need not mean a new wave of hospitalisations – ergo, no need to inflict more harm on society and the economy. Here’s how the picture looks for Covid deaths, to date.

In rejecting lockdown, Sweden deliberately tolerated higher Covid levels than locked-down countries in the hope of minimising other damage and protecting more lives in the round. Last year, Sweden’s economy (which relies heavily on exports) fell 3 per cent vs 10 per cent for Britain. This isn’t just about money: years of experience of recession shows a clear link between economic downturns and public health damage: the effects are longer term. Most health economists would agree that a lower economic hit now means a lower excess death hit later.

To the outside world, Sweden is a country that rebelled – and now rues that rebellion. But Swedes are still pretty happy. Why?

In spite of three UK lockdowns, Sweden has ended up with fewer Covid deaths. On a per capita basis, the UK and Sweden had almost identical Covid death levels by the end of the first wave. Both amongst the worst in Europe, but locking down didn’t give Britain much of an advantage – a point Tegnell makes now and again.  But locked-down Britain was hit harder in subsequent waves (below, UK lockdowns in grey).

The problem with ‘Sweden vs its neighbours’ Lockdown advocates often compare Sweden with its neighbours, but this mistakes the nature of Scandinavia. Sweden has lots of trees, but as a country it is more urbanised than even Britain, for example (ie: 88pc of Swedes live in cities, vs 84pc of Brits).  Right at the start of the pandemic, Bristol Uni’s Oliver Johnson came up with an index to explain this point: population density and how it relates to Covid vulnerability. The below graph sums it up. Norway and Finland have very low population density,  Sweden is in line with the European average (higher than Germany).  On top of that we have Sweden’s exposure to globalisation, high foreign-born population etc. Sweden falls down in comparison to Denmark, which has higher population density but suffered less excess death and comparable economic damage.

Collateral health damage The closure of society and discouraging people from seeking healthcare will inflict fatalities that will show up as ‘excess deaths’ for other reasons. Sweden avoided kind of school closure damage that the IFS thinks will cost British pupils £350 billion in lost future earnings. Swedish breast cancer referrals were down by about 10 per cent in the first ten months of last year. But the British figures are closer to 30pc decline. But, importantly, Sweden did take a hit. So it’s wrong to attribute all declining A&E use, cancer diagnosis and economic damage to lockdowns (as some lockdown sceptics do) because Sweden suffered the same (albeit to a smaller degree).

The question is how far lockdown – and messages like ‘protect the NHS’ – exacerbate the deadly effect of healthcare hesitancy. The indirect deaths (especially of cancer) would take years to show up, so a 2020 graph will not give the whole story. Adjust for age (most Covid deaths were amongst the over-80s) and you end up with the below Oxford Uni study.  Sweden’s figure was up 1.5 per cent last year, the England figure was 10.5 per cent. Is this the lockdown effect, or a coincidence? It’s impossible to tell – it’s just a piece of a still-incomplete picture.

The myth of Sweden being a liberal utopia. One of the biggest errors in the UK’s lockdown response was Imperial College’s controversial cliff-edge hypothesis: the idea that measures/advice/anything less than full lockdown makes v little difference. But full lockdowns? Why, that shoots Covid out of the sky. (After lockdown, Imperial’s Neil Ferguson predicted fewer than 7,000 net Covid-excess deaths.) This cliff-edge theory is still advocated fiercely in some circles, used to make the case for early and hard lockdowns.

Sweden matters to the world because it adopted another theory: that smaller measures (a ban on mass meetings, a bunch of homeworking advice) would be enough. That you have to look at this in the round, and lockdown would inflict too much collateral damage. Swedes also were big on the fundamental point that the state has no business criminalising everyday life.

So was Sweden right? The below graph uses Google mobility data to show that mobility to Swedish workplaces fell by a quarter and in transit stations by a third by 23 March – enough, it seems, to force its Covid into reverse. This also shows mobility in Britain (in red) falling even further – and long before lockdown was imposed. This matters. The idea that Brits needed to be locked down because they are not as obedient or sensible as Swedes is simply not borne out by the below data. The UK public is shown to take far stronger evasive action than the Swedes and and long before it was made mandatory.

Britain and Sweden’s synchronised Covid waves Comparisons between the UK and Sweden are often rejected as absurd due to latitude, landscape etc. But our Covid experience has been eerily similar in the first and second wave. Whatever led Britain to get hit by Covid so badly at first also applied in Sweden. But as the below graph shows, Sweden twice forced down its Covid deaths without lockdown. This raises the question: how far would a similar common-sense, guidance-based approach have worked in the UK?

To repeat: this is not to say that Britain got it wrong and Sweden got it right. The above is simply a snapshot, clues in an unsolved mystery about a virus that keeps surprising us. It’s premature to draw any firm conclusions at this stage. But the above might help explain why Swedes, even now, are not pushing for lockdown, even with cases so high as to have them placed in the UK’s ‘red’ category. They seek to limit the damage inflicted on people’s lives and health and pursue a policy that will lead to least harm overall. 

There’s plenty Sweden got wrong (chiefly, a failure to protect the elderly in care homes). Tegnell is keen to say that it will take months, perhaps years, for the picture to settle. But to understand Sweden, you need to look at more countries than its neighbours. You need to look at more than just Covid deaths. You need to compare Sweden to countries with similar levels of urbanisation, not just its neighbours.

All this explains why Sweden refuses to lock down and tolerates what is, at the time of writing, Europe’s highest Covid levels. Polls show almost three-quarters of Swedes saying that the health authority’s handling of the crisis was either ‘good’ or ‘very good When these other points are factored in, it becomes easier to understand why, after all the last year has brought, Swedes still think they will be proved right in the end. 

PS: I’m keeping this post updated :  if you’re interested in this topic please do check back from time to time as new data arrives all the time.