Europe’s even worse off than we thought

Here’s the Financial Times:

On Tuesday night, the health authority in the Grand Est region said two-thirds of its 620 old people’s homes had been affected by the coronavirus pandemic and 570 residents had died.

Those 570 people are not recorded in France’s official coronavirus death toll, which reached 4,032 on April 1, but so far counts only those who have died in hospital. Eastern France, which has had 1,112 hospital deaths, was the first region in the country to be badly hit by the pandemic.

Recent studies in Italy comparing recorded Covid-19 deaths with overall death rates in specific regions also suggest the country’s death toll is far higher than the official total of more than 13,000, already the world’s highest.

In an Italian retirement home in Mediglia, outside Milan, 52 of the 152 elderly staying there had died from Covid-19 infection by last week.

In the province of Bergamo, 2,060 deaths were attributed to the virus in March. However L’Eco di Bergamo, a local newspaper, found that a total of 5,400 deaths occurred in the province in March, up from just 900 in the same month in 2019.

Based on the Bergamo data, it seems possible that the actual Italian death total is already well over 20,000, not the 14,000 official total.  China’s not the only country where coronavirus deaths are somewhat undercounted.  Yet I continue to believe the death rate data is superior to caseload data, once adjusted for demographic factors such as age.  Caseloads have likely been grossly undercounted almost everywhere, with a few possible exceptions like Iceland.

I also continue to believe that the Chinese data showing a dramatic decline in the epidemic between February and March is broadly accurate, despite an undercount in level terms.

Many people have trouble understanding my view of China.  I believe the censoring of Wuhan doctors was the single worst mistake of the entire epidemic, and have said so repeatedly.  If China had been ruled by a Western government, those doctors would not have been censored and the epidemic would have been addressed more quickly (by a few weeks).  However the epidemic would also have been addressed much more incompetently by a Western-style government, and today the epidemic would be totally out of control in China, with far more deaths.  The optimal solution would have occurred if the Nationalists had won the Chinese civil war, as Taiwan has one of the few governments that has handled this crisis with any degree of competence.

And Tyler Cowen points out that the CCP is making hay out of the fact that we also censor our doctors when they issue warnings.  Can this country get any stupider?

PS.  People talk a lot about Italy, but Spain’s been hit just as hard, albeit a few days behind Italy. Roughly 24,000 of the 50,000 global deaths are in Italy and Spain, not even accounting for unreported cases.

PPS.  Alex Tabarrok points out that plenty of masks are available, but the FDA won’t allow them into the country.  Yeah, “globalization” is to blame.  And don’t bother donating your extra masks to your hospital; they won’t take them.  You can’t make this stuff up.

PPPS.  Ten days ago, I pointed to California (and should have mentioned Washington as well) as a place that was doing somewhat better due to getting a head start on social distancing.  Since then, their share of cases and especially deaths has dropped further.  California has nearly 12% of the US population but 4.4% of cases and 4.0% of deaths.  It’s fallen to number 6 in deaths, whereas Washington (once number one) has fallen to 5th in deaths and 10th in cases.  Both states continue to see deaths climb much more slowly than the national average.  Outside of the NYC area, Michigan and Louisiana have done extremely poorly.  Within my area (Orange County) the rich areas were hit hardest, although perhaps that’s now changing a bit.  I recall that AIDS started as a mostly white disease in America, then shifted to minorities.

Updated figures:



25 Responses to “Europe’s even worse off than we thought”

  1. Gravatar of Carl Carl
    2. April 2020 at 10:44

    The hospital wouldn’t take our stash of masks which has been sitting untouched by humans in a back closet for a couple of years. So we gave them to my wife’s niece who’s an emergency room nurse and she took them in and they used them.

    That’s good news about California and Washington. I’ve never had much time for our government here in the Bay Area, but on shelter-in-place they made the right call.

  2. Gravatar of Michael Rulle Michael Rulle
    2. April 2020 at 11:04

    I have come to think it is possible that Biden’s health will have worsened enough that it may be very hard to nominate him. Some people think that Cuomo should be the replacement, which I find absurd, apart from the coronavirus—-but, Newsom, even though a rookie governor was the Lt. Gov plus SF mayor. If I were Dems and wanted a Biden replacement, he is the obvious choice.

  3. Gravatar of James Alexander James Alexander
    2. April 2020 at 12:07

    Just bought us an oximeter. Various reasons. But researching what the readings mean for covid I cam across a few good examples of seriously good academic papers out of China on how they learnt quickly on the job. Lots more papers like this.

    Chinese consumer demand for exotic wild food is terrifying. The fact that the CCP can’t control it shows how unshackled are Chinese consumers. Too free to choose!

  4. Gravatar of Brian Donohue Brian Donohue
    2. April 2020 at 12:17

    Very good post.

    In Europe, there appear to be two hot radiation points: northern Italy, which also bled into Switzerland and Spain is one, and Benelux, which also bled into France, is the other.

    The Italian radiation is earlier and looks more lethal. It’s noteworthy that smaller Switzerland has done better than Italy or Spain, but they’ve still been hit hard.

    Luxembourg is a crazy place- 600,000 citizens but 200,000 foreigners work there, including lots of health care workers. Also hard hit, but they have done better than their larger Belgian and Dutch neighbors.

    Small countries seem to have an advantage in responding. Maybe there is a lesson for the EU project.

  5. Gravatar of msgkings msgkings
    2. April 2020 at 12:49

    @Michael Rulle:

    Aren’t you a Republican?

  6. Gravatar of dtoh dtoh
    2. April 2020 at 13:05

    Pursuant to my earlier post, mass media has now begun picking up the BCG story

  7. Gravatar of Ray Lopez Ray Lopez
    2. April 2020 at 13:12

    Another quality post by Sumner; this blog is becoming my favorite ‘go to’ site for accurate predictions about Covid-19. It’s as close to a crystal ball you can get.

    One minor point about China, which indeed has been on the vanguard of this disease, in fighting it, as Sumner says; it’s being reported now on Bloomberg that China is finding more asymptomless carriers of Covid-19, the more they test. If so, social distancing is going to be a years long, now a months long, endeavor. But rest assured Dr. Sumner’s keen mind is working on explaining it. I feel better just knowing that. And money (maybe) is NOT neutral! Stock market will tell us (wealth effect, notice if the Fed is intervening, it has stopped the downward rout since March 13).

  8. Gravatar of BlueSilverWave BlueSilverWave
    2. April 2020 at 15:01

    Re: Michigan, I wonder how much of that is related to how tightly the auto industry is tied to manufacturing China. DTW had daily flights to Wuhan before this all went down. Plus, as soon as the pandemic started becoming obviously out of hand in the media, all of the auto industry pulled back their people – especially suppliers. I wonder if the greater Detroit area became a massive hotbed of Coronavirus because of the sheer number of auto employees that came home?

  9. Gravatar of ssumner ssumner
    2. April 2020 at 15:52

    Brian, You said:

    “Small countries seem to have an advantage in responding. Maybe there is a lesson for the EU project.”

    Not so much in per capita terms. Some small places like Iceland and Luxembourg and San Marino and Andorra are high in per capita terms.

    dtoh, That would be great news.

    Ray, About 50% of cases are asymptomatic. But in that case, if you can get symptomatic cases down to a very low level, then you’ve also reduced asymptomatic cases:

    Blue, Could be that auto industry connection, as you say.

  10. Gravatar of ssumner ssumner
    2. April 2020 at 15:55

    James, What’s the oximeter good for? Maybe I should get one. (I have bad lungs.)

  11. Gravatar of Brian Donohue Brian Donohue
    2. April 2020 at 18:28

    Come on, San Marino and Andorra have about 100,000 people combined.

    Luxembourg and Switzerland have higher per capita CASES but lower per capita DEATHS than their larger neighbors. Almost as if they are casting a wider testing net or something.

  12. Gravatar of Student Student
    2. April 2020 at 22:05


    A pulse oximeter measures oxygen saturation in the blood. I bought one in case I catch it (my lungs are iffy as well). If your Sp02 registers between 92% and 90%, it’s time to get to a hospital. So I got it to monitor myself if I were to get covid19. You can basically use it to stay at home and know when it’s time to get to the hospital.

    US number check… GFs have been bouncing pretty weirdly the last few days… where the 7 day rolling average is lower than the 3 day (1.08 vs 1.14)… that suggests an uptick in the spread over the last few days.

    My model projects we hit GF=1.0 (the inflection point) on April 9th (as of early morning 4/3/2020). We should add ~32,500 new cases today (4/3) for a closing total of ~277,300 at the end of the day. The social distancing is working. We should have about half as many cases as we would have had if things were trending as they were 2 weeks ago. At this rate, we won’t hit a million cases until late April. That’s good.

  13. Gravatar of Michael Rulle Michael Rulle
    3. April 2020 at 07:05


    “if I were the Dems……….”

  14. Gravatar of Michael Rulle Michael Rulle
    3. April 2020 at 07:16

    Many have commented on the difficulty of determining the cause of death with people who are already ill when they contract Covid-19. As I understand it, the current protocol in most of the world of counting is to assign the cause of death to Covid-19 in instances when there are multiple diseases/illnesses also present. This is biased toward overstatement

    And while certain small cities in Northern Italy have atrocious excess death rates which are very hard to ignore—-it is still not representative of the population as a whole—they still can be confusing cause and effect—after all–we are LOOKING for the cities with this profile.

  15. Gravatar of msgkings msgkings
    3. April 2020 at 08:05

    @Michael Rulle:

    You didn’t say that, and you aren’t ‘the Dems’. You’re their opponent. So your advice should maybe be taken with a little salt?

    Or should the Reps take their electoral advice from Nancy Pelosi perhaps?

  16. Gravatar of msgkings msgkings
    3. April 2020 at 08:07


    Cases are not the best metric to track, because as testing ramps up we get more cases, which doesn’t mean the disease is spreading as much as we are testing more. The deaths number is the one to watch to see if we are containing this. Dr Fauci said this as well.

  17. Gravatar of msgkings msgkings
    3. April 2020 at 08:08

    @M Rulle: My apologies, you did say that. Still, is a Rep the right person for them to get advice from?

  18. Gravatar of Student Student
    3. April 2020 at 09:33


    I understand that point but I don’t see any real good reason to believe deaths are any better metric. They are a function of age distributions, obesity and other preexisting conditions, the extent to which the medical system is swamped, viral mutations…

    Plus deaths are following the same basic trend as cases (just lagged like 2-4 weeks). Right now, deaths in the US are doubling about every three days, just like cases were 3 weeks ago or so. As a result, I don’t see any empirical reason to believe the argument you are making… in spite of the fact it seems logical.

  19. Gravatar of msgkings msgkings
    3. April 2020 at 10:09


    It’s simply that deaths are what we are trying to prevent. Since most people who actually get the virus range from feeling nothing at all to being very sick (but surviving), we are clearly shutting the world down to prevent mass deaths.

    That’s how we will know we are winning, that’s why it’s a better metric. I agree with Dr. Fauci on this.

  20. Gravatar of ssumner ssumner
    3. April 2020 at 10:26

    Brian, OK, I didn’t know your definition of small countries. You mean sort of small, but not tiny.

    Student, Is the inflection point the day when the daily total of new cases peaks?

  21. Gravatar of James Alexander James Alexander
    3. April 2020 at 12:14

    Oximeters measure the oxygen level in your blood. Hospitals use them routinely they are the little clip that doctors put on your finger. They cost between $25 and $100 – depending on availability (which is a bit stretched right now, they all come from China, obv). Weak lungs can be improved by exercise and the improvement measured by the oximeter.

  22. Gravatar of Student Student
    3. April 2020 at 13:31



  23. Gravatar of Student Student
    3. April 2020 at 13:59


    Well put.

  24. Gravatar of ssumner ssumner
    4. April 2020 at 08:13

    Thanks James and Student.

  25. Gravatar of Jg Jg
    4. April 2020 at 19:43

    I think California locked down just two days before NY.

Leave a Reply