The Monday Read

All I can manage right now edition

Japan seems to be getting deeper into its economic coronapocalype, creating a security divide between salaried employees at larger companies and freelancers/employees at smaller companies.

The government seems to be slowly coming around to the need to support people, but the gears are grinding slowly. Let’s hope they hurry up a bit and provide enough support to keep people going.

I’m also a little bit overwhelmed at the moment, so you may see fewer posts on the blog than you’re used to. Apologies and I’m hoping to get back to normal as soon as possible.

This week’s links

  1. Glad I didn’t sign up to buy one of these: Harumi Flag at half-mast
  2. Bill Gates on what to do next: What our leaders can do now
  3. Nice speech from an early retiree: SOME THOUGHTS ON THE REAL WORLD BY ONE WHO GLIMPSED IT AND FLED
  4. True story: I overheard some of the student interns at my wife’s school talking about kabu (stock) yields and got really excited. Turns out they were talking about kabu (turnip) yields in the Animal Crossing game…
  5. A short primer: HOW TO REDUCE YOUR TAXES WITH JAPANESE REAL ESTATE INVESTMENTS
  6. The last few weeks have changed how I think about the world: Wounds Heal, Scars Last
  7. A look behind the scenes: How coronavirus almost brought down the global financial system
  8. Personal Finance During a Crisis
  9. Always surprise me: WHAT IS THE AVERAGE STARTING SALARY IN JAPAN?
  10. This snarky non-financial article brought me five minutes of joy: The world’s most insane coffee drinker
  11. That graph of what the UK economy might look like! Weekend reading: Whoops, there goes the economy
  12. Howard Marks’ latest: Knowledge of the future
  13. Things are getting bad for freelancers/small companies: Media Mix, April 19, 2020

What do you think? Anything good in there? I loved number 10 and number 11 shocked me!

This week’s books

I found a new author/series I like! So of course I have read three of them and almost finished the fourth 🙂

The Naturalist, by Andrew Mayne. Serial killer/detective stories with a great protagonist, good writing, decent plots, and just enough setbacks to keep them interesting. They are also reasonably priced! Perfect social distancing reading material. Somewhat violent but not too bad. Lots of science.

11 Responses

  1. Yes, the UK article in particular really brings home the importance of not trashing the economy in a’bid to save lives.’ No, it’s not that money is more important than people’s lives, but that there are many ways people other than COVID-19 that can cause people to die prematurely. Imagine, for example, how much more difficult it is going to be to fund the NHS just through normal times when government tax receipts collapse.

    I also wish to relay the following. I did this myself, though I used someone else’s template. I am not am epidemiologist not even a doctor. I do know basic statistics though and have not seen this analysis reported in any newspaper or other media in Japan.

    From a data perspective, the situation looks not too bad. There are several things to watch out for regarding spreading disease:

    – the worst situation would be exponential growth in the number of new cases (exponential meaning not just an increase in new cases, but an increase in the rate of new cases. The rate is calculated by comparing the number of new cases to the number of existing cases.)

    – linear growth (meaning a fairly constant rate of increase) means that ‘new records’ in absolute numbers will be made most days – this is in fact what the media seems to be focusing on, as it makes dramatic headlines – but it is not particularly significant statistically.

    – The daily data is variable, so the significant point is the 3 or 4 day running average. This eliminates variations in reporting due to factors such as speed of communication of data, days of the week, differences in speed of confirmation of tests, etc.

    – If the trend in 3-day average looks good or bad, we can double check by looking back to the past and see from the later data how long it has taken for previous totals to double.

    In short……

    Worst case – growth is exponential – we will see a rising average over time in the 3-day average, we will see constant ‘new records’ in raw numbers, the time taken for case numbers to double will be decreasing, day-by-day.

    Bad case – growth is linear – we will see a constant 3-day average, we will see ‘new records in raw numbers almost daily, the time taken for case numbers to double will be constant.

    Encouraging case – growth is non-exponential and decreasing – we will see a drop over time in the 3-day average, we will only sometimes see ‘new records’ in raw numbers, the time taken for case numbers to double will be decreasing.

    It is clear from the data that at least since April 1st (and probably earlier) we are in the third case, the most encouraging.

    There are several reliable data sources, it doesn’t matter which one you use, as long as you are consistent. (Both sources exclude the Diamond Princess and airport quarantine data, as they are largely irrelevant to the trend within Japan.)

    For Japan-wide, I am using this:

    https://toyokeizai.net/sp/visual/tko/covid19/en.html

    For Tokyo, this:

    https://stopcovid19.metro.tokyo.lg.jp/en/cards/number-of-confirmed-cases/

    Japan-wide data:

    Daily increase/New total/%increase/3-day average of increase/no. of days to double

    4/19 10219 (+565) +5.9% (6.6%)
    4/18 9654 (+627) +6.9% (6.6%)
    4/17 9002 (+585) +6.9% (6.4%)
    4/16 8442 (+478) +6.1% (5.9%)
    4/15 7966 (+457) +6.1% (6.6%)
    　4/14 7509 (+386) +5.4% (8.5%)
    　4/13 7123 (+507) +7.7% (10.8%)
    4/12 6616 (+714) +12.1% (12.3%)
    4/11 5902 (+656) +12.5% (12.3%)
    4/10 5246 (+579) +12.4% (11.2%) 10 days (at least)
    4/9 4667 (+499) +12.0% (9.4%) 9 days
    4/8 4168 (+351) +9.2% (9.3%) 8 days
    4/7 3817 (+248) +7.0% (10.3%) 8 days
    4/6 3569 (+378) +11.8% (12.0%) 8 days
    4/5 3191 (+346) +12.2% (11.5%) 7 days
    4/4 2845 (+304) +12.0% (10.5%) 7 days
    4/3 2541 (+235) +10.2% (10.4%) 7 days
    4/2 2306 (+199) +9.4% (10.1%) 7 days
    4/1 2107 (+220) +11.7% (10.0%)

    Tokyo data:

    4/19 3083 (+108) +3.6% (5.9%)
    4/18 2975 (+181) +6.5% (6.7%)
    4/17 2794 (+201) +7.5% (6.3%)
    4/16 2596 (+150) +6.1% (6.3%)
    4/15 2445 (+127) +5.4% (5.7%)
    4/14 2379 (+161) +7.4% (6.8%)
    4/13 2159 (+91) +4.4% (8.2%)
    4/12 2068 (+166) +8.7% (10.9%)
    4/11 1902 (+197) +11.6% (12.5%)
    4/10 1715 (+189) +12.4% (12.7%) 10 days (at least, maybe even 11)
    4/9 1516 (+184) +13.5% (10.9%) 9 days
    4/8 1338 (+144) +12.1% (9.1%) 9 days
    4/7 1194 (+79) +7.1% (10.1%) 8 days
    4/6 1115 (+83) +8.0% (13.0%) 8 days
    4/5 1032 (+143) +16.1% (14.7%) 7 days
    4/4 889 (+116) +15.0% (14.8%) 6 days
    4/3 773 (+89) +13.0% (14.1%) 6 days
    4/2 684 (+97) +16.5% (15.6%) 6 days
    4/1 587 (+66) +12.7% 5 days
    3/31 521 (+78) +17.6% 5 days

    To conclude:

    Despite the dramatic newspaper headlines, the data so far suggests that in both Tokyo (especially) and Japan in general the curve is starting to bend down significantly. Indeed, is has been doing so since very early April (April 2nd/3rd.) This means that, if the trend continues, we can expect the total number of cases to peak at about the end of April, enabling a lifting of the state of emergency. In other words, Japan is on schedule.

    Of course, if there is a second wave or change in the trend, these conclusions become invalid, but the data has been quite consistent so far.

    I have not read the above approach in any newspapers, nor seen it on any media reports. But that is because, as they say, “good news is no news.”

    1. Olivier, these charts are awesome and I like that the Tokyo Keizai chart provides an important criteria; seriousness and death. The headlines are all about people being infected which lacks context, as you mentioned. But it sure gets eyeballs and the associated advertising revenue.

      6 serious cases a day across the country sounds like something that could be manageable. In fact, if it wasn’t I’d probably be asking questions to my health minister.

      I wonder if someone has overlaid the CORVID data with year over year flu data yet.
      Thanks for sharing these.

      1. Haven’t seen that done. I do know the the flu season was extremely mild comparatively this year and, more importantly, very short. One reason it was short was because of the early end of the school year (due to Covid) and schools are one of , if not the, main transmitter of the flu most years.

    2. And I mis-typed: in the encouraging case, the time taken for the total to double will be INCREASING.

    3. I thought the same about the number of infections/deaths data, especially as it seems Japan is finally increasing testing and removing the legal need to hospitalize everyone who tests positive, regardless of need.
      On the other hand hospitals seem to be getting closer to being overwhelmed, at which point the mortality rate balloons and people start dying of other treatable conditions because they can’t get medical care.
      It’s a tricky one.

      1. Yes, it was kind of weird the way they have hospitalized everyone who tested positive and are now saying the system is overloaded. I note the announcement today, though they they have secured 210,000(!) hotel rooms for light and asymptomatic cases….

  3. And just to be as combative as usual:

    “Japan seems to be getting deeper into its economic coronapocalype, creating a security divide between salaried employees at larger companies and freelancers/employees at smaller companies.”

    I would replace “creating” with “highlighting.” No, not even “exposing”, as if you weren’t aware that salaried employees at larger companies enjoy better job security than freelancers or employees at smaller companies then, as the Americans say, “duh!” And that is the way it should be, as being a salaried employee at a large company can suck in many ways vis-a-vis the freedom of freelancing etc in better economic times.

    This doesn’t detract of course from the utmost importance of small and medium-sized businesses to the economy in general, even in Japan which, despite a reputation to the contrary, is still vastly comprised of such.

  4. On 10 — yes, truly insane. An insane waste of money, not just the price of coffee shop coffee, also money spent driving and adding to global warming.

    Is my way insane? I got a ceramic Korean hand roaster through Amazon. It takes 70 grams of raw coffee beans or 80g of the sundried higher moisture Brazilian beans I got most recently. That’s enough for 1.5 litre of strong filter coffee. I buy green beans through Amazon at between ¥1500 and ¥2000 a kilogram. I worked out that ¥2000 is roughly the cost of the 340g bags of supermarket UCC beans. Roasting to a bit after the second pop takes about 15 mins, mostly near the low setting on an LPG konrou. And I’m running the extractor hood fan and small LED light. So far I’ve tried Brazilian Chocolat beans, Papua Elimbari Gold and Brazil Topazio Sundried.

    I use a Zassenhaus hand grinder adjustable from espresso to cowboy saucepan. That’s 10-15 mins of exercise. The grind is finer than a pre-ground supermarket filter but my speed of filtration is faster than most – – I use a large size Melita ceramic cone with fluting and three holes and put in more than half of the hot water in quickly in one go and then pour in the rest when there’s space. I drink it black, no sugar.

    Reading 10 filled me with feelings of virtue (at least in this area of my life). I haven’t worked out the total cost but I’m pretty sure it’s cheaper than even 711’s machine brew.

    1. I’m not as virtuous as you, but my coffee habit is Kirkland Espresso Roast beans from Costco/Amazon (just over 2,000 yen per kilo), hand grind with a Hario ceramic mill, brew with Aeropress to a ridiculous strength, drink 3-4 cups a day.

      It’s all portable too, so I can take it with me on trips, etc.