RIVM and NICE provide new figures up to April 18. My earlier discussion of 14 days ago used data from April 2. The Dutch death toll has risen from 1339 to 3601: with one unspecified age, so I will use 3600.

The excel sheet has been updated here. The London Imperial College infection fatality rates (IFR) are still not useful for Holland, and the Canary in the Mine principle for the 70-79 age group still gives results that seem reliable – see the former weblog entry for the explanation.

Comparison of the main tables for now and 14 days ago gives these results:

  • The 80+ group has become much more unreliable. Apparently GP doctors and families have a greater tendency not to send such patients to the hospital anymore, given the emotional and physical burden (no contact, low survival rate). This group has 12% more deaths than hospitalised cases, which means that testing is now done more often outside of hospitals. However, the 70-79 group still seems to opt for the hospital. (This also means that statisticians must check that their time series for the D / H ratio maintain the same definition. It is important to check the performance of hospitals.)
  • If the canary works okay, then the (cumulative) prevalence of infections has risen from 92391 (0.5%) to 236432 (1.4%). Officially reported infections were only 16% and are now only 13% of all infections.
  • The implied “registered” infection fatality rate (IFR) is about the same: was 1.4% and is 1.5%.
  • Deaths are still much underreported however. Deaths outside of hospitals are still hardly tested. The CBS mortality data indicate, as RIVM reported, that there might be another 2000 untested deaths. That is, the “surplus deaths” compared to other years are twice as high as the officially reported deaths attributed to Covid-19.
  • It is tempting to infer that the true IFR might also be 3% but this neglects the profile of infections, since people are not infected now randomly, but there are reports that the virus is spreading more rapidly in home-care for the elderly. The officially reported number of cases in the ICUs and the number of deaths is flattening, but if GP doctors and families have decided to avoid hospitalisation, then there might be no real flattening.
  • Thus, there still is no true indication that the reproductive number is below 1, though the national lockdown is of such quality that we may expect that it is, at least for the people outside of home-care for elderly. However, the economic cost of this lockdown is huge. There is something very bizarre about the official policy of locking down the country to save people but also allow that the virus spreads in home-care for the elderly who are precisely the most vulnerable.
  • While there were 0 deaths below age 50 at the beginning of April, there are now 23 such deaths. This still indicates that the virus works as its own vaccine for this group if we exclude comorbidity.

These new data only affect the left hand side (LHS) of the table. They do not change the fundamental insight on the right hand side (RHS) that the virus works as its own vaccine for the younger group without comorbidity. This means that the suggestion from the former weblog entry for an exit strategy with quarantine zones still stands. If we want to start with this on May 1 then only 11 days are left for planning, i.e. at all levels of government, agencies, companies and families.

Update April 21: The table with the exit scenario did not yet contain the implied IFR’s. These have been included now, see below. I now assume that the overflow of infections from the exposed population of 12 million to the vulnerable group of 5 million will be 4% of the latter, with 25% of development of disease, so that still 1% is in the danger zone. The vulnerable above have an IFR of 5.8% and below an IFR of 6.3%. The less vulnerable above have an IFR of 0.06% and below an IFR of 0.076%. The overall IFR’s do not compare, since the weights of the groups have been adjusted (above has population weights and below has deliberate status of infections).

All politics is local. Infections are local too. In the Covid-19 epidemic it makes sense to consider some sort of management of social behaviour, with attention to the local character of quarantine.

  1. Use information about disease status and risk.
  2. Insulate the vulnerable (elderly and diseased) by stricter rules.
  3. Allow a controlled gradual re-introduction of the virus as its own vaccine, so that the economy can recover.

Update 2020-04-07: (1) Sake de Vlas and Luc Coffeng had the same idea of gradually re-introducing the virus, see their preprint of April 1 2020. They use 10 regions of each 1 million people, with transport of ICU patients between them. I am thinking of hospitals and their own smaller service areas, and quarantined patches within those, so that transport is normal. There will be similar numbers of re-infected nationwide, and a same period of 16 months. De Vlas & Coffeng do not yet mention the death count of the strategy. (2) Paul Romer suggests daily testing of randomly 7% of the population, when those tests are available, and also considers daily randomly allocating 50% of the population to lockdown when there are no tests yet (and self-quarantine if you get symptoms). The latter are ways at containment / suppression of a homogeneous population while I would tend to look at isolation / immunity and ways to scale up ICU capacity for a heterogeneous population.
Update 2020-04-19: RIVM has reported more than 3,500 deaths now, as officially attributed to Covid-19, but also a rise of untested deaths overall. Official deaths below the age of 50 are 23, which corroborates below numerical setup. Table 3 is slightly adapted to better fit with table 2.

Colour coding

The idea of colour-coding for the status of infection is rather natural. Fruits and flowers already use the gimmick.

China now has this app: (1) green = travel relatively freely, (2) yellow = home isolation, (3) red = confirmed Covid-19 patient who must be in quarantine. (Guardian)  I had suggested this type of coding in 2004 for HPV that had no vaccine at that time yet – see this working paper – though this particular virus soon got a vaccine.

The last weeks I have been reconsidering the coding scheme. There are more groups to deal with. It is also better to use red for the quarantine borders between the different groups. My present suggestion is this overall scheme:

Covid-19 Colour Coded Groups

The reasoning behind the code is to use the natural combination of R, G and B as much as possible. This is the overall scheme.

Covid-19 RGB Colour Coding

The following is an example map. It contains two hospitals: one for the vulnerable and uninfected and one for the infected and infectables. The example has a geographic layout, but one would start out with a functional allocation, and then see how it would work out geographically. One would suppose that the homes of the elderly, or at least the entries and exits, are surrounded by red barriers, and guards checking who crosses the boundary. Basically one would try to identify larger regions that have the same code, so that people can move freely within their region. Indeed, villages that are unaffected would set up village border patrols again. Make visits by appointment: not only the doctor and hairdresser but also schools and cafes. Below calculation shows that a cohort of less-vulnerable (young and non-diseased) persons can be of size of 270,000 persons, such that if the virus goes rampant in that cohort then their allocated hospital would still have enough ICU beds to serve the severe cases amongst those less-vulnerables. It is an option to deliberately infect such cohorts, since for the less-vulnerables the virus has little effect and is basically a vaccine for itself (though it can hurt). For Holland, such a scenario would take 16 months. See the discussion below.

Covid-19 Colour Coded Map

Why would we consider such colour coding of the disease status ?

Findings by the Imperial College Covid-19 team

The Imperial College Covid-19 team estimates that R0 =3.87 from a study on European nations, with an Infection Fatality Rate (IFR) 0.66. For the UK they calculate an IFR 0.9. Their underlying estimates using data from China and the Chinese conventions at hospitalisation are not at their website but originally at the medical archive and now (shorter) in The Lancet. Their “impact paper” by Neil Ferguson et al. (March 16 2020) sums up the current message on Covid-19:

  1. Best is suppression of all infections down to the capacity of the health system, and there-after start with surveillance with tracing and isolating infections, for the whole period till we have a vaccine, which may take 12-18 months. This scenario fits with keeping R[t] < 1. Taiwan likely gives the best practice for containment / suppression of Covid-19. It will be quite a challenge for the health care system to set up such a surveillance system, e.g. requiring at least five investigators per suspected infection.
  2. Worse is the idea of mitigation, i.e. to try a combination with trying to get herd immunity. This plays with the idea that R[t] > 1 so that eventually a large majority of the population builds up immunity. This would come with a huge risk of overburdening the health care system. However, the Imperial College team has not yet considered an approach of structured quarantines, see below.

Perhaps not all arguments have been mentioned why Covid-19 will be with us for at least a year, and why it is best to plan for at least two years. The reasons are rather standard from an introductory course in infectious disease:

  • getting a vaccine takes time … and then give it to seven billion people on the planet
  • health care is under strain, will perform less well, thus allowing breaches and ever newer infections
  • the virus already shows mutations and is likely to continue to do so: every new mutation would require a swift reaction for new containment – but the health care system already is under strain.

We can expect waves of new infections, like with the flu, but then 10 times more infectious / deadlier than the flu, with the risk of shorter intervals because of faster mutations. The Northern hemisphere now benefits from the upcoming Summer, but in Autumn the reduced health because of the common cold and flu will combine with Covid-19, causing increased joint mortality. This upcoming Summer should rather not be wasted.

Hammer and dance

What the Imperial College proposes as “Containment” is called “Hammer & Dance” by Tomas Pueyo. Pueyo employs a slightly different terminology. He compares “Do nothing” with “Mitigation” (towards herd immunity) and “Containment / suppression“. He advises the latter for the USA, and presents his variant as a “hammer and dance” approach, which is still the Imperial College proposal:

  • first contain / suppress by lockdown to the level of health care capacity (likely not in a constant state of emergency),
  • and then follow the Taiwan and South Korea model of slow release but suppression by trace and quarantine of new infections.

In Pueyo’s graph, observe the distinction between the horizontal axis and the capacity of the health system just above it (the height of the green curve in the “ongoing” phase). Pueyo also points to the epidemic calculator by Gabriel Goh.

Containment and Hammer & Dance are not enough

Given the expectations above, the Imperial College “suppression” and Pueyo’s “Dance” are dubious. There is no way how we can currently reduce the number of infections down to the manageable level except by lockdown. Perhaps the USA and Europe with more resources can try to copy the Taiwanese example but can they really, and what about other regions ? Given that the world is not Taiwan, the world now has only the option of lockdown, continued hammering, and this has nasty effects:

  • In lockdown, there are the “collateral” deaths of persons who would normally receive care but who remain untreated. The Dutch Volkskrant newspaper reports that 40% of normal hospital care has been cancelled. (a) People with a disease are vulnerable to Covid-19 and may fear a greater risk of infection within the health system itself. (b) Health care resources are reallocated from normal care to Covid-19 related cases. The latter is rational, given that untreated infections are a risk for the whole population. Not treating infected cases (by hospitalisation or self-quarantine with supervision) creates such risk. We must compare the collateral deaths to the “avoided deaths by treating those with infections“. PM. See my earlier weblog about the value of life.
  • In lockdown, the economy is severely affected. Bankruptcies would strain the legal system. The government currently tries monetary, financial and tax arrangements, but real production would collapse, and more money chasing fewer goods will mean rising inflation and the need for price controls. Thus, we are getting a war time economy. For Holland, production already goes from expected positive growth of 1.7% to -1.2%, a loss of at least 2.9% of GDP, or some EUR 24 bn of EUR 800 bn – see the CPB-scenario’s of 2020-03-26. Each death has come along together with a loss of EUR 30 million. See also Richard Baldwin at VoxEU on preparing for the second wave.

The Imperial College team and Pueyo present the “mitigation / controlled infection” scenario as too risky with too many deaths. One tends to agree with them, except for above nasty effects that they don’t actually discuss.

If “mitigation” comes with a degree of control then there are some aspects that are worth considering. When Covid-19 is relatively harmless for a large low-risk group, it can work as its own vaccine. The objective of this present weblog is to show a way how to enhance control: by better identifying and handling of the various quarantine categories.

Dutch data about Covid-19: using the 70-79 group as the Canary in the Mine

The following uses data of April 2.

The Imperial College estimates give problems for the Dutch data. With 121 deaths in the Dutch 60-69 age group, the London age-specific IFR gives 5500 infected in the population while their “symptomatic cases per hospitalised” gives 7663 symptomatic cases in the population, which is too much since we are assuming that the flu season is over. Holland has 29 hospitalised children of age 0-9, and the London symp / hospital rate for this group gives 29000 symptomatic children in the Dutch population, which would create panic if true. Looking the issue over, I cannot find a match. It must be remarked that the Dutch “reported number of cases” is rather useless, because of the lack of tests, and their preferred application to medical personel rather than patients. Also the death count is understated since non-hospitalised deaths are not tested. See Table 1 below.

However, the Dutch 70-79 age group may be used as canary in the mine. The number 2951 of “reported cases” will be accurate for this group, since they do not belong to medical personel. These patients will have some symptoms (like “feeling really sick”) and not be tested for nought. The reported number of 2951 means only 0.19% of the whole age group. The Imperial College IFR for this group gives an estimate that 8137 would be infected, or a share of 0.005346 or 0.5%. We arrive at the problem that we are not in the steady state. Either these elderly “infected but non-patients-yet” have a stronger immune system or they are due to arrive at the hospital at a later moment. With lack of other information, we can still presume that this is the overall rate of infection in Holland. When we apply this rate to the whole population, then we get age-group specific rates of hospitalisation and IFR that show the same pattern as in China and the London research group. Especially relevant is the “hospitalised per infected ratio” (H/I). See Table 2 below.

Intermediate conclusions

The intermediate conclusions are:

  1. On April 5, Holland has about  92.391 infected persons, or 0.5% of the population. The reported number of cases by RIVM is 16% or 1/6 of the true number. The current IFR for Holland is 1.4% because of the high share of elderly people (with comorbidity). If you are younger than 60 then your IFR is 0.04% (4 basispoints) and for 60+ it is 5.5%.
  2. If 100% of the population would get the virus then there will be 20,945 deaths younger than 60 years and 197,896 deaths in the 60+ group. The normal deaths in 2019 were 151,737 persons. Life expectancy would roughly reduce by 1% – till there is the vaccine.

A scenario with a Dance with Managed Quarantine

Table 2 in the last column (at the bottom RHS) has the option of using the health care capacity for the coming two years:

  1. The group of 60+ is put under quarantine, so effective that at most 1% of them gets infected. This would cause the death of 1979 persons in that group. (Actually, it will be wise to also include the younger diseased in this vulnerable group, see below.)
  2. The group younger than 60 years is put under quarantine, but also: step by step exposed to the virus, in cohorts of size 271,360, using the virus as its own vaccine, so that eventually herd immunity at 75% of this group is attained (using R0 = 4 and 75% = 1 – 1/R0). This would cause the death of 15,709 persons in this group. (But there will be less deaths if we shift the younger diseased.)
  3. It would take 16 months to achieve this. By that time, there ought to be a vaccine, and the vulnerable people in the population can be vaccinated, while the less-vulnerable people already will have achieved herd immunity for their section.

More detailed calculations are in Table 3 and this excel workbook. I suppose that the population still will grow a bit. Blue letters and figures are parameters that can be adjusted. The other colour coding is taken from above. For this calculation, the vulnerable group consists of 60+ and the younger diseased, so that the protected group counts 5,000,000 people. Implications are:

  • A vulnerable person who gets infected anyway (slips through quarantine – the 1%) needs 3 weeks of ICU time, while a less-vulnerable person takes 1.5 weeks of ICU time. Given the required loads of service, 328 covid-beds serve the vulnerable and 1272 covid-beds serve the less-vulnerable. There are still 800 beds for non-covid ICU cases, allocated to the different quarantine areas. The parameters can be adjusted to different values, and then the scheme might take a different number of months.
  • In this scheme, the number of deaths is reduced from above 218,841 to 19,661 (a bit different from above rougher 17,688) over a 16 months period. The normal death count is about 150,000 per year, so this rises with 10% per year over a period of 16 months. The causes for the much lower death toll are: (a) the strict protection to the vulnerable group, (b) the shift of the group of younger vulnerables from the younger group to the vulnerable group.
  • While the less-vulnerable group would be the “economicially productive” group of society, they could restart their business in two manners: (1) first under the quarantine of “unsymptomatic and untested” (Cyan) group (with restricted number of contacts), then a pause for the phase of cohort infection and recovery (Blue, for some Magenta and some death (black, not shown)), then (2) secondly as recovered and likely immune and no longer a carrier (Green) (with restricted number of contacts even when no carrier, when it is not clear what the recent contacts have been).
  • For an evaluation, we need an estimate of how many “collateral lives” would be lost, if we do not restore some normality.

For people and goods crossing borders, testing is important (not only whether one carries the virus but also whether one once did). Such tests are now in short supply, and when they come in supply then the priority is for the health system. Overall, they would be important for the “dance” phase. However, with this scenario, they will also be important for the checking of the quarantine boundaries and the management of the deliberate infection of the cohorts.


Overall, it seems possible to start up the economy again in the beginning of May, without the risk of another wave of infections, provided that society finds a way to manage and control the states of quarantine.

NB. For an evaluation, we need an estimate of how many “collateral lives” will be lost, if we do not restore some normality.

PM. See the excel workbook for details and references to authors who inspired this kind of calculation. The CBS Statistics Netherlands StatLine tables do not provide five-year age groups of January 1 2020 yet, and beware that there are different subgroups of 95+.

Disclaimer. Limited earlier experience in research on infectious diseases

In 2002-2004 I collaborated at Erasmus Medical Center on the modeling of the Human Papilloma Virus (HPV) as the cause of cervical cancer. My background in modeling and also logistics was relevant because diseases may look like a Markov logistics process with stages and transition probabilities. There can be the same issues of test reliability, criteria of lives-saved or life-years-gained, and cost-effectiveness of screening and treatment. I also followed the discussion about the SARS epidemic of 2003. My period at Erasmus MC was too short to allow for publishing peer reviewed papers but let me mention two working papers.

  1. Working paper 2004: Modifying behaviour with a passport. At that time there was no HPV-vaccine yet. An option was to manage human behaviour. The status of infection can be recorded in the medical dossier: free (green) or carrier (red). While children can gets warts, an assumption might be that children start out uninfected by the harmful HPV variants (status green). When couples meet and want to get into a serious relationship – in the sense of sharing their germs – then they can show each other their status of infection in their medical dossier and discuss the implications. From this working paper, we may take the idea of recording the status of infection, and using colour coding for clear communication. For Covid-19, it is better to use “red” (alarm, or hungry in Chinese restaurants) for the barrier between zones and groups.
  2. Working paper 2003: On the value of life. This compares the lives-saved and life-years-gained measures, and develops a compromise: a “unit-square-root” measure, that regards each life as 100% and takes the square root of the relative gain. This is discussed in the former weblog entry.

The national lockdown causes the “collateral deaths” of persons who would normally receive care but who remain untreated. The Dutch Volkskrant newspaper reports that 40% of normal hospital care has been cancelled.

  • People with a disease are vulnerable to Covid-19 and may fear a greater risk of infection within the health system itself.
  • Health care resources are reallocated from normal care to Covid-19 related cases. The latter can be rational, given that untreated infections are a risk for the whole population.

This shift in the burden of disease and death might be acceptable to a large extent. We must compare these “collateral deaths” to the “avoided deaths by treating those with infections“. However, at issue is whether some elements in this shift of the burden are dubious. We can understand these aspects a bit better when we have a better grasp of what is called “the value of life”.

Below, I will discuss a particular theoretical case of “collateral death”:

Consider a person of 20 years of age, with a potential gain of 60 additional years of living, who might not get proper treatment because the Intensive Care Units (ICU) are occupied with Covid-19 patients, who are elderly and might perhaps gain 10 more years of living. PM. Women in Holland at age 80 still have a life expectancy of a bit more than 9 years on average (statline). (It is not known whether this changes or whether they will have a full recovery from Covid-19.)

There is my 2003 working paper: On the value of life. This compares the lives-saved and life-years-gained measures, and develops a compromise: a “unit-square-root” measure. This regards each life as 100% and takes the square root of the relative gain – see below. The paper was intended for macro-economic issues and not for triage at the micro-level, but let us now investigate how it would work at the micro-level. PM. Said paper still suffers a lot from being a working paper, and requires much editing for readability.

Let us first look briefly at the flu and then proceed with above theoretical case at the ICU.

Relation to the flu

RIVM provides the following data for two flu-periods in Holland. Rather than the case fatality rate (CFR) (deaths per infected) we have the symptomatic fatality rate (SFR) (deaths per symptomatic). A well-known statement is that the flu is “an old man’s friend” – but this is disputed by some. I did not find the percentage at ICU. The Dutch health system is accustomed to the flu and some categories of (elderly) patients with the flu and comorbidity are no longer sent to the ICU.

Table 1. Flu incidence in Holland 2017-2019

RIVM data

Winter 2017 – 2018

Winter 2018 – 2019

Length of period (weeks)



Reported symptomatic people






Surplus deaths



Symptomatic Fatality Rate (SFR)



Covid-19 is “at least ten times deadlier than the regular flu” (quote) – a recent estimate is 7 times – and likely more contagious and more prone to mutation. At issue is whether the properties of Covid-19 warrant a different treatment at the ICU as is happening in these weeks. The Dutch health system is not accustomed to Covid-19 yet, and very likely they now admit Covid-19 patients who would not be admitted in the future. Let us consider the admission criterion.

The ICU admission criterion

Dutch ICU currently have the “incremental probability of survival” admission rule (document). This is the Δp = p b between the probability of survival of the treatment at the ICU (p) and the probability of surviving not at the ICU (b = background risk). Observe:

  • Patients with Covid-19 tend to have a worse Δp at the ICU than similarly diseased patients without Covid-19. Normally, we would not see many people with Covid-19 at the ICU (with comorbidity). Apparently, the health care system is not used to Covid-19 yet. They try to save patients who in retrospect wouldn’t have a chance at admission. (An aspect is that it is reported that 80% of the patients requiring breathing machines are overweight – an aspect of the obesitas health crisis.)
  • A comparable situation exists with influenza. I did not find a report about the use of ICU for flu patients. Quite likely, the availability of beds made it possible in the past that also categories of patients with influenza were admitted, also with comorbidity. The difference with Covid-19 is the increase in the case fatality rate – but this points to fewer admissions.
  • However, people have little reason to go to the hospital when they will not be treated. Untreated infections are a risk for others and thus it is better to present some scope for treatment. Keeping patients at the ICU is a form of quarantine, though an expensive one.

My 2003 paper On the value of life assumed a p = 100% chance of success of treatment (with a known subsequent life expectancy). Let us now discuss the case with a different value of p.

PM. The term “lives saved” might be emotionally biased. It seems better to say “lives extended”. Eventually everyone dies, and the term “lives extended” indicates that the important information of “how much” is not mentioned.

A treatment criterion in general

The general setup is: (i) maximise performance given costs, or (ii) minimise costs given a level of performance. Public Health budgets tend to be given, so we choose the first approach.

Let there be two types of treated patients in numbers n1 and n2. We select these treated numbers from the pools waiting for treatment n*1 and n*2. Costs of treatment per patient are c1 and c2, for example because of different lengths of stay at the ICU. Available resources are C, and these can be time of medical personel or plain costs. The probabilities of success are p1 and p2, with performance outcomes s1 and s2. All variables are nonnegative.

We can impose an additional (moral) condition that one type of patient gets a weight λ. The weight λ means that a person in that group in the new objective function becomes λ times as important as a person in that same group in the old situation. The effect depends upon the success criterion, see below. Sometimes this condition is imposed by making different beds for different types of patients. We might also manipulate the pool sizes. When a pool of cheap patients with a high treatment score is large, so that they get all the treatments, we may restrict the pool size in order to allow treatments for the other type.

The above gives the linear programming model for the variables n1 and n2:

Maximise    p1.s1.n1 + λ p2.s2.n2

Subject to  n1 ≤ n*1    and    n2 ≤ n*2    and    c1.n1 + c2.n2 ≤ C

The following diagram presents the C-simplex and the feasible region restricted to the n* pool-variables. Due to linearity, the objective function will tend to select one of the end-points. The shown optimal point is {n°1, n*2}. When the objective function and the feasible region have the same slope, then –c1/c2 = – p1.s1 / (λ* p2.s2), and λ* = s1/s2 (p1/p2) / (c1/c2). In this case we would select patients at random, but still keep representative numbers for the types of patients.

Figure. Linear programming for triage

Let us use this model for the theoretical example case from above, namely of the patient of 20 years without Covid-19 and the patient of 80 years with Covid-19. Let type 1 be the youngsters and type 2 be the elderly.

Simplification by using costed-patients

Each type of patient comes with a standard cost, and the linear programming problem becomes a bit more tractable by using this constancy.  Above problem can be simplified by looking at “costed-patients” qi = ci.ni. The conditions on the pool size become qi = ci.nici.n*i = q*i. Then we get:

Maximise (p1.s1/c1) q1 + λ (p2.s2/c2) q2

Subject to q1 ≤ q*1 and q2 ≤ q*2 and q1 + q2 ≤ C

For this formulation it may be easier to calculate the randomisation value of λ* = (p1.s1/c1) / (p2.s2/c2).

Since the maximand can be scaled arbitrarily, we may divide by the coefficient of the second group, and get this expression, so that it is fully clear that taking λ = λ* gives a maximand that is parallel to the cost condition. The shadow price of the cost condition will be max[λ , λ *].

Maximise    λ* q1 + λ q2

Subject to q1 ≤ q*1 and q2 ≤ q*2 and q1 + q2 ≤ C

Cohort size

The “direct gain ratio” is s1 / s2. For lives-extended, the direct gain ratio is 1 / 1 = 1, while for life-years gained above example gives 60 / 10 = 6, meaning that treating one younger person successfully gives the same effect as treating 6 elderly persons successfully.

By comparison λ* is the “effective gain ratio”, in which the direct gain ratio is corrected for the relative risk p1/p2 and the relative cost of treatment. The effective gain ratio means that treating 1 younger person (with the given chance of success) has the same effect as treating λ* elderly patients (with their chance of success) (and excluding the manipulation with λ).

Assuming that all costs are depleted, we can write c1.n1 + c2.n2 = as n2 (c1.n1/n2 + c2) = C. Taking μ = n1 / n2 as the “selected ratio”, then we can take a single cohort as consisting of μ youngsters + 1 elderly, or μ  + 1 persons. The cost per cohort is k = c1 μ + c2 = C / n2, using that there are n2 cohorts.

While above manipulation of choosing λ at λ* obviously can be done, it still leaves the problem of randomisation. The manipulation of the maximand is not so effective in this kind of problem. It is more effective to manipulate the pool of youngsters n*1. Above we tended to assume that this pool was exogenously given, but when this pool is so large that they claim all treatments, and we want to put a limit to this, then we effectively reduce the pool size. A relevant variable to consider is the selected ratio μ or the composition of the cohorts. If μ is chosen, then n2 = C / k, and then n*1 = n1 = μ n2 = μ C / k = μ C / (c1 μ + c2).

The choices of λ and μ are different since they apply at different aspects of the problem, i.e. the maximand or the boundary condition. Yet we can conceive of some combination. The value of λ does not always mean a representative cohort size. Only when the slopes happen to be equal and there is randomisation, then, in some cases, we might first treat μ = λ* youngsters before treating an elderly patient again (which single person also weighs as λ*).

Case 1. Using mortality only

The success criterion of lives-extended gives s1 = s2 = 1. The objective function is parallel to the feasible region when λ = λ*[lives] = (p1/p2) / (c1/c2), or p1 / p2 = λ c1 / c2.

The “incremental probability of survival” is a subcase that only compares p1 and p2.  This neglects costs, or sets c1 = c2, e.g. neglecting duration of treatment, or in fact sets λ = c2 / c1, i.e. compensating for duration of treatment (so that there is randomisation when the two probabilities are equal). The rule is also silent about the pool sizes. (In practice ICU have more criteria, with some implied λ.)

We would not discriminate patients when λ = 1, i.e. without additional moral judgement and when only the lives-extended criterion and the parameter values determine who is to live and who not. (We do not decide who dies: nature does.) The choice of  λ = 1 still allows for the happenstance that the slopes would be equal by chance. For example, when the younger person costs 2 financial units (or weeks) and the older person costs 3, then the slopes are the same when the survival probabilities have the same ratio: p1 = 2/3 p2. For other values of the survival probabilities, however, one type of patient is preferred to the other type.

E.g. when p1 = 0.80% and p2 = 0.51%, the maximand is 0.80 n1 + 0.51 n2 or 0.40 q1 + 0.17 q2, and then clearly the first type will be preferred. Only an additional quantitative restriction n*1 might prevent the neglect of all patients of type 2.

For another value than λ = 1, the objective function is 0.40 q1 + λ 0.17 q2. We randomise if λ = λ* = 40 / 17 = 2.35. This means that one old patient in the new maximand is valued as 2.35 old patients in the original maximand. If we take λ* as the cohort parameter μ, then after randomly choosing 2.35 youngsters, we would randomly choose 1 person from the elderly (who has weight 2.35 too).

The lives-extended measure neglects the “how much” of the life-expectancies.

Case 2. Using life-years-gained

The success criterion using life-years-gained gives s1 = 60 and s2 = 10. We find λ = λ[lys] = 60 / 10 (p1/p2) / (c1/c2) = 6 λ[lives].

We may also find the maximand of 24 q1 + λ 1.7 q2, and the objective function is parallel to the feasible region for λ*[lys] = 24 / 1.7 = 14.1. The latter means that 1 old person in the new maximand counts as 14.1 old persons in the original maximand. Perhaps with a bit more freedom, we might say that 1 year of life for an elderly person counts as 14.1 years for a younger person. For the cohort size, we might treat μ = λ* = 14 youngsters and then treat an elderly person (who counts as 14 too).

PM. The 10 life-years actually gained for an elderly person are regarded in the new maximand as 14.1 * 10 = 141 life-years gained. This reflects 141 / 60 = 2.35 youngsters. But the elderly are not really such youngsters, and we must also account for the the effect of the other parameters.

The life-years-saved criterion is biased in age and sex: it gives advantage to the young and women. Both criteria of lives-extended and life-years-gained have their drawbacks. There is scope for a compromise.

Case 3. Unit-square-root measure

The unit-square-root (UnitSqrt) measure uses:

a = age

d = life expectancy with disease, without treatment (for the ICU: d = 0)

x = expected life-years-gained when treatment is successful

patient score = Sqrt[x] / Sqrt[a + d + x] = Sqrt[x / (a + d + x)]

Substituting all parameter values gives the maximand 34.6 q1 + λ 5.7 q2.

For above young person the score is Sqrt[60 / (20 + 60)] = 0.866. Above Covid-19 patient has Sqrt[10 / (80 + 10)] = 0.333. We get λ = λ[sqrt] = 0.866 / 0.333  (p1/p2) / (c1/c2) = 2.6 λ[lives]. We thus have a compromise value between the ratio of only survival 1 / 1 = 1 and the ratio of life-years of 60 / 10 = 6, namely 0.866 / 0.333 = 2.6. For the example case, the “middle of the road” character of the UnitSqrt measure also shows from the randomisation value of λ* = 6.1: this lies between the earlier other two values of 2.4 and 14.1.

Collecting results

We can tabulate our findings for the discussed example. Obviously, this table applies to this particular example only. There can be quite some discussion about what kind of success measure, and possible “correction” or “discrimination” by means of the n* and λ and μ. In practice, an ICU may have quite different reasoning as well, like on the availability of medical personel for particular treatments.

While it remains possible to select one of these objective functions and be neutral with λ = 1, and allow an extreme outcome (except for the happenstance of parallel lines), it seems more likely, as said, that the more relevant choice concerns the cohort size μ + 1. Above discussion gives considerations for a reasoned choice of μ as one of the “gain ratio’s” mentioned in the table. Above discussion suggests the compromise value, for this particular example case, of μ = 6, i.e. treat first 6 youngsters of said type and then treat an elderly person of said other type.

Table 2. Comparing success measures for triage, a = {20, 80} and x = {60, 10}

p = {0.80, 0.51}
c = {2, 3}
Lives extended
Life-years gained
Unit square root
Success measure s = {1, 1} s = {60, 10} = x s = Sqrt[x / (a + d + x)]
Direct gain ratio (only s1/s2) 1 / 1 = 1 60 / 10 = 6 0.866 / 0.333 = 2.6
Maximand 80 n1 + λ 51 n2

40 q1 + λ 17 q2

48 n1 + λ 5.1 n2

24 q1 + λ 1.7 q2

69.3 n1 + λ 16.9 n2

34.6 q1 + λ 5.7 q2

Weight formula
λ[lives] λ[lys] = 6 λ [lives] λ[sqrt] = 2.6 λ[lives].

Effective gain ratio (λ*)

λ* = 40 / 17 = 2.35 λ* = 24 / 1.7 = 14.1 λ* = 34.6 / 5.7 = 6.1


The 2003 paper originated from a macro-economic context of the allocation of the budget over different types of treatment. At the micro-level, a hospital is faced with a queue of patients with all their own characteristics of age a, life expectation d, effect of treatment x, now extended with probabilities and costs c. We could order patients on their costed scores p / c Sqrt[/ (a + d + x)], and apply a λ per category. Obviously this is only a suggestion from theory.


Some conclusions are:

  1. The 2003 paper did not look at a probability of survival p other than 100%, and now we have found an useful adaptation, namely above maximand with the LP properties.
  2. The current admission criterion for the ICU of “incremental probability of survival” would tend to favour youngsters because of their better conditions and survival probabilities. The current admission of many elderly Covid-19 patients must derive from other considerations, like inexperience with Covid-19 and the effect on quarantine and containment. The creation of additional ICU beds might be seen as additional only.
  3. The admission criterion of the “incremental probability of survival” looks at “lives saved”, or rather “lives extended”. This causes questions about cost comparisons (e.g. length of stay at the ICU), pool sizes (n*) and moral values (λ). Switching to “life-years gained” comes with a larger information load and partly provides answers to those questions, but also causes more questions, since this criterion is biased on age and sex. The UnitSqrt measure takes each life as 100% and would remove the latter bias. It would eliminate this aspect in the choice of λ (except for the choice of other functions than the square root), so that its choice would be more dependent upon the parameters on costs and survival probabilities.
  4. While the original paper derived from the macro context, and this present weblog entry explored the micro context, the latter also highlights that the present public discussion about Covid-19 deaths does not yet consider the aspects on the life-years. The latter will be required for an evaluation of the wider consequences, like on the “collateral deaths” and on how to “prepare for the second wave”. However, the life-years are a biased criterion and it would seem to be advisable to see more statistics that use the unit square root too.


(1) In 2002-2004 I collaborated at Erasmus Medical Center on the modeling of the Human Papilloma Virus (HPV) as the cause of cervical cancer. My background in modeling and also logistics was relevant because diseases may look like a Markov logistics process with stages and transition probabilities. There can be the same issues about test reliability, criteria of lives-extended and life-years-gained, and cost-effectiveness of screening and treatment. I also followed the discussion about the SARS epidemic of 2003. My period at Erasmus MC was too short to send in papers for peer review. (2) Now I am an elderly male and advantaged by the lives-extended and disadvantaged by the life-years criterion.


PM 1. We did not use “quality adjusted life-years” (qaly). The application with the UnitSqrt would be the same but the information load would be larger, and likely not available at an ICU.

PM 2. A bit more about the additive character in this weighted averaging: The current setup is that the patient with age a has a life expectancy of d without treatment, and supposedly also when the treatment would fail. If d = 0 then p is the acute probability of survival due to the treatment (relevant for an ICU). Some possible variants provide an indication that above linear weighted averaging is the relevant approach.

  • xeabt = p (xd) + (1 – pd = p x + d is the life expectation when admitted but before the treatment. This variable is less useful, see PM 3. Also Sqrt[p x / (a + d + p x)] for this state is not so useful.
  • scorexp = p Sqrt[x / (a + d + x)] + (1 – p) Sqrt[0 / (a + d + 0) is the expected score (relevant)

PM 3. Suppose that the outcome measures at failure are f1 and f2, and that s1 and s2 are measured incremental to such failure. Then we might consider maximising the total effect:

Maximise    {p1.(s1 + f1) + (1 – p1).f1}  n1 + {p2.(s2 + f2) + (1 – p2).f2}  n2

Which becomes:

Maximise    p1.s1.n1 + p2.s2.n2  + f1.n1 + f2.n2

With this maximand, there is a bias towards patients who would already have a good score when the treatment fails. This maximand is not the relevant one, because the failure outcome does not depend upon the treatment. Another maximand would include the background risk b, but this is also not affected by the treatment.

To the International Association for Official Statistics (IAOS),
Royal Statistical Society (RSS),
American Statistical Association (ASA),
Société Française de Statistique (SFdS) and
International Association for Research in Income and Wealth (IARIW)

Dear presidents Pullinger (IAOS), Ashby (RSS), Martinez (ASA), Marin (SFdS) and Reinsdorf (IARIW),

Your societies and associations have made public statements in support of Andreas Georgiou, former president of El.Stat Statistics Greece.

I have looked at the case and arrived at the conclusion that Georgiou was guilty as charged for the violation of duty, as was indeed confirmed by the Greek Supreme Court in 2018. It appears that Andreas Georgiou, Hallgrimur Snorrason (representative of Eurostat at El.Stat in 2010) and Walter Radermacher (Eurostat 2008-2016) have provided you with false testimonies about the Greek law of March 9 2010 that created El.Stat and the European Code of Practice of 2005 that was in force in 2010. They knew about the true legal situation and in fact worked to make changes in both the law (December 2010) and the Code (2011).

The documentation is in my new book “Forum Theory & A National Assembly of Science and Learning“. My discussion of the actual figures in national accounting of Greece 2009 is on pages 200-206, and there seem to have been made some arbitrary choices that are not fully clear to me. My discussion identifies many more points where important information is lacking. Overall, I advise that there will be a thorough investigation. I hope that you will indeed apply due diligence.

I already informed ISI and FENStatS via this letter, now online in slightly edited form to make it quotable for others. I regard the now online letter to ISI and FENStatS as an integral part of my letter to you now. I have the same requests for you as stated in that letter. Please inform your membership but please refer to my book instead of trying to rephrase the points in your own words because points might get lost in translation again. This letter to you is online now too.

Reading parts of my book, Richard Gill, emeritus professor in mathematical statistics in Leiden, and former chair of VVSOR (a founding member of FENStatS), informed me that he has revised his view, from an earlier signing of support to a (now disputed) statement by ISI. Klaus Kastner, retired banker who blogs on Greece, also has revised part of his view.

My disclaimer: I would like to see a thorough investigation in Dutch economics.

Sincerely yours,

Thomas Cool / Thomas Colignatus
Econometrician (Groningen 1982) and teacher of mathematics (Leiden 2008)
Scheveningen, Holland

PS. At IARIW, Peter van de Ven (formerly CBS now OECD) was IARIW president in 2010-2012, when the issue of El.Stat case arose. Van de Ven proposed to IARIW in 2016 to support Georgiou, apparently with deficient study of the underlying events. At CBS in 2009, Van de Ven removed the Tinbergen & Hueting figure of environmentally Sustainable National Income (eSNI) from the Dutch monitor on sustainability, using fallacies, see THAENAES Chapter 25 p289. I copy to Van de Ven at OECD and Kees Zeelenberg at CBS who is involved in IAOS – OECD.

To the President of [the International Statistical Institute (ISI)] and the secretary general of [the Federation of European National Statistical Societies (FENStatS)]

Dear professor Bailer (ISI) and dr Silva (FENStatS),

There is this new (open access) book by me:

Forum Theory & A National Assembly of Science and Learning

The idea is that scientists and scholars create an assembly of their own, to improve the quality and impact of science and learning. The book discusses various problems. Some main problems are about national accounting and official statistics. There is the Tinbergen & Hueting approach on environmentally Sustainable National Income (eSNI). There is the El.Stat / Georgiou case.

The book webpage is [here].

The PDF of the book is open access [here].

A short text (discussing 2 of 7 storylines) is [here (pdf)].

Though the Tinbergen & Hueting case is much more important, it so happens that ISI and 80 former chief statisticians around [IAOS and FENStatS] put out statements of support for Georgiou, and it may be that you are alerted on this problem in more accessible manner just now.

Allow me to invite you to look at my deconstruction of “Greek statistics” in Part 6, starting p153.

My diagnosis is that the statistical associations, including you, have been misinformed by Walter Radermacher (Eurostat and current president of FENStatS), Hallgrimur Snorrason (representative of Eurostat and associated with ISI) and Andreas Georgiou (El.Stat). They knew in 2010 that the Greek law was that Georgiou had to seek approval by his board. He is guilty as charged for the conviction of violation of duty. I copy to them but do not have the email address of Georgiou.

One might regard this as an old case, but Eurostat also arranged that the National Statistical Offices of the EU are now under a single head, while CBS Statistics Netherlands since 1892 was under supervision by a multiperson board. This change in governance has risks for the future. My book indicates that CBS likely did not protest strongly enough because the Dutch government managed to appoint a leadership that had no background in official statistics itself.

Since the Part on “Greek statistics” is lengthy, let me also point to p200-206 for some formulas en tables on the calculation of the 2009 debt and deficit.

[…] Richard Gill, former chair of [VVSOR] (a founding association of FENStatS), already informed me that he revised his opinion now (from signing an earlier ISI statement), see below under (1). (PM. I copy to Fred van Eeuwijk, current [chair] of [VVSOR].) Klaus Kastner, a retired banker who blogs on Greece, also has revised part of his view, see under (2). They may not have digested my book fully since it is just in print.

I already informed Walter Radermacher about my finding, since his background is in environmental statistics, and around 1990 he had friendly contacts with Roefie Hueting. Unfortunately, Radermacher did not comprehend the Tinbergen & Hueting approach, and has misrepresented it too. My criticism on the work by Radermacher is restricted to these two issues that I have looked at. I [expect him] to correct when errors are clarified to him. While I informed Radermacher earlier this week, he indicated that he had no time for this, and this is a complication.

Another complication is that Radermacher currently is the President of FENStatS. The Vice-President of FENStats is professor Maurizio Vichi, who also happens to be the thesis supervisor for the Radermacher thesis of 2019. I checked the thesis on the two issues of eSNI and the Georgiou affair. In my judgement, the thesis is biased and misrepresenting on these two issues. It should be retracted. Radermacher knew in 2010 that Georgiou was in violation of the Greek law, and as director of Eurostat assisted him in breaking the law of an EU Member State. He should have insisted that Georgiou would come clean with his board instead of helping him to bypass them. Subsequently, see my discussion of the actual debt and deficit calculations.

Thus, my request to Radermacher and Vichi and FENStatS is that Radermacher and Vichi step down from their positions at FENStatS, to that there is no conflict of interests between FENStatS and their positions on the thesis and Radermacher’s position in this case ueberhaupt. Hopefully Radermacher might reorganise his priorities and consider the criticism in my book.

After sending this email to you, I will relay it to some journalists, so please do not be surprised if they would contact you and know about this email. It also seems best that I include this email on my website, and check there for the upcoming link.

Earlier, see under (3), I already informed the DG CBS dr. Tjark Tjin-A-Tsoi about my book, and the finding that the erroneous ISI declaration had also been signed by some CBS researchers who had failed to check upon the Greek law, European Code of Practice 2005 [in force in 2010], and the particular data, and the false testimonies by Georgiou, Snorrason and Radermacher. I have been supplying the DG CBS with drafts of my book in the last month[s], and I offered CBS a final week to consider the argument, and provide for a more structured approach in informing the world of statistics about my book and finding. Alas, CBS did not indicate to me that they would take this opportunity. Thus, I am left with no other option than inform the world of statistics myself. The most efficient way to do so, at least for me, likely is to inform the media, though generally the media can be quite chaotic overall.

Let me also express that I am very annoyed concerning ISI. In 2012 I wrote to ISI with a request of an investigation, in which there was also attention for the views of the (other) board members. Instead, ISI apparently listened only to Snorrason at ISI. It strikes me as very unprofessional to allow someone to be judge in his or her own case. In BCC I will copy to some former El.Stat board members. My 2012 letter to ISI is [here].

Let me also express my [appreciation] to CBS for inviting Hueting and De Boer in 2019 to present their new book on national accounting and eSNI, which lecture was attended by some sixty statisticians at CBS. CBS also made a video (in Dutch) available, and since De Boer was late I myself had to step in for a moment. We must clearly distinguish the open and unencumbered discussion of issues on content at the “statistics work floor” and the issues of governance and management which gain more attention in my book. The link to the Hueting & De Boer lecture in 2019 is [here].

If you were to write on this, then my advice to you is to inform me with a draft text, to allow me to comment on possible errors. Please revise your (online) texts that contain erroneous statements, by heading them with clearly visible “Retracted, see link…” while keeping the remainder “as is” so that others can check the history of your errors. Please send your apologies to the Greek government, with copies to the press, for trying to infringe upon the separation of powers. Please inform your membership about this email and my book. Please do not try to restate my book in your own words, thereby taking attention away from my book, but simply correct and refer to the book for the details. Please respect my analysis that the Greek case is only an example, while the real issue is the creation of said assemblies, that also are intended for a better anchor for future governance of official statistics.

Sincerely yours,

Thomas Cool / Thomas Colignatus
Econometrician (Groningen 1982) and teacher of mathematics (Leiden 2008)
Scheveningen, Holland

(1) ================================

From: Richard Gill
To: Thomas Cool / Thomas Colignatus
Cc: […]
Subject: Re: N.a.v. uw ondertekening van de ISI verklaring uit 2013 over Andreas Georgiou
Date: Sun, 26 Jan 2020

Ik wil best zeggen dat mijn mening veranderd is. Ik heb het gezegd. Je mag me citeren.

(2) ================================


(3) ================================

Date: Sun, 09 Feb 2020
To: DG CBS dr. T. Tjin-A-Tsoi, Kees Zeelenberg, […]
From: Thomas Cool / Thomas Colignatus
Subject: Published “Forum Theory & A National Assembly of Science and Learning”

Dear dr Tjin-A-Tsoi and dr. Zeelenberg and […],

My book is now online for all:

Forum Theory & A National Assembly of Science and Learning

[pdf relocated]

The printshop will take some more days.

You might enjoy the national accounting on page 200-206. I just included this this weekend, while finishing the book, and it appears that finishing touches can be quite important.

If you would have comments, feel welcome to send them. I can include additions and corrections in a file with “Reading Notes”, and eventually there might be a 2nd edition.

Let me call your attention to some CBS employees mentioned on p292 who did not properly check the Georgiou case and who signed the erroneous ISI declaration. Allow me to request the DG to forward this email to them. […] is an appreciated contact of mine from the student days in Groningen, and I send him my regards.

I also wonder whether you would have some advice as to how to approach the world of official statistics on this book and its findings. It is easy for me to dispatch some emails to some of the associations and persons involved. It might be less effective and needlessly chaotic. If there would be scope that the CBS would take a week to digest the book and formulate a statement, and be willing to discuss the draft of the statemnet with me (with me only giving advice), then this would seem to be advisable. Please let me know in the coming days whether you would be willing to do so.

Sincerely yours,

Thomas Cool / Thomas Colignatus
Econometrician (Groningen 1982) and teacher of mathematics (Leiden 2008)
Scheveningen, Holland

Science and scholarship are much appreciated in our societies when we consider the lavish funding by tax payers and private donors. In contrast there is also a structurally weak position in the very functioning of our democracy. Our best research minds are allowed to discover the wonders at the frontier of knowledge but their findings might not be actually used. In examples like climate change, overpopulation, miseducation, inequality, political science of electoral systems, and the revolutions in computing, biology and medicine, and so on, and even in the issue of Greek statistics of the national debt and deficit of Greece around 2009, we see that science and learning are hardly listened to, and that policy makers and opinion leaders follow their own illusions, leading the world towards foreseeable disasters. Humanity as a whole acts mindless, drunk or crazy.

Trias Politica versus Tessares Politica

There is a system to the madness. Our democracy has the Trias Politica separation of powers between de Executive, Legislative, and Judiciary branches of government. What is lacking is the Epistemic branch. This means that the Trias Politica model allows too much room for manipulation of information. The alternative is a High Definition Democracy that has this Epistemic branch to protect information. This would give a Tessares Politica – from the ancient Greek “foursome”.

Epistemic branch

The Epistemic branch has two elements.

  • The first aspect concerns economic planning and the management of the State. The national budget is crucial for policy making. The Legislative branch or Parliament votes on the national budget to authorise the Executive to tax and spend funds. The economic planning agency resides currently under the Executive. However, planning is systematically abused. We better create an Economic Supreme Court at the same constitutional level, with the task to control the quality of information, and the power to veto the budget if it contains misleading information according to the court. This leaves all freedom for politicians to determine policy but they will lose the room they have now for manipulating information.
  • The second aspect concerns all fields of research. Parliament currently has the Senate and the House of Commons. In the French Revolution of 1789 the Chamber of the Clergy was abolished. In an alternative path of history, the clergy could have developed into scientists and scholars, and the revolution avoided, and then nowadays we would have a Chamber for Science and Learning.

If we want to improve the world and democracy then these are the two structural positions to consider.

Forum Theory

Forum theory is the approach by the Dutch cognitive psychologist, student-achievement tester, methodologist and philosopher of science Adriaan de Groot (1914-2006), also famous for his study of chess grandmasters. Forum theory provides us with the diagnosis that science and learning operate as a forum, i.e. a market subject to particular conditions. The approach investigates processes that enhance quality of science and learning, and formulates ways to protect and encourage those. Forum theory suggests that science and learning will be improved by a National Assembly. This contrasts with the current dominance by the National Academies of Science and Scholarship.

Such Academies tend to consists of mainly elderly researchers who have made their mark years ago and who are selected by co-optation. This leaves the mass of researchers out of the picture, currently at the front of research and making their mark. The common researchers on the work floor, e.g. in the laboratory or at the computer terminal, are locked up between a stone wall and a firy pit, and tend to be quite frustrated that the existing knowledge is neglected or misrepresented by politics, while the elevated colleagues at the Academy tend to focus on their next conference. Jonathan Swift (1667-1745) in his Gulliver’s Travels already made fun of the Royal Society.

This situation can be changed by the creation of a National Assembly of Science and Learning, with a Floor for the mass of researchers, while the National Academy forms the Senate of the Assembly. The Assembly improves governance of the forum, the forum itself, and research integrity. Researchers in science and learning can create their National Assembly actually quite simply. They can set up a foundation, give rules of operation, recruit members, organise elections and have a constitutional meeting. With sufficiently large membership the operating costs can be covered. The next step is to show results.

The difference between the current Academy and the sketched Assembly is that the latter has the full weight of the collected research body in a country, with the legitimacy of having been elected by them. The Assembly can do investigations and support conclusions, and speak for science and learning with an authority that now is lacking for the Academy. Over time Parliament could accept the National Assembly of Science and Learning as its third chamber, e.g. called The Study.

Greek national accounting and statistics in 2010

My book with above analysis and proposal also looks at the example of 2010, when Statistics Greece (El.Stat) director Andreas Georgiou sent figures on the 2009 Greek national debt and budget deficit to Eurostat without first seeking approval by his board. The Greek court system judged that he was in violation of duty, irrevocably by the Greek Supreme Court in 2018. Today there is an international uproar in national accounting and statistics that Georgiou’s conviction is a miscarriage of justice and an attack on the independence of Official Statistics.

However, these protesting statistical societies and associations have failed to check the Greek law of March 2010 that created El.Stat, and they also refer to the European Statistics Code of Practice of 2011 or 2017 while the code in force in 2010 was from 2005. Georgiou is guilty as charged.

By all looks of it, he also – advised by Eurostat – added the Simitis swaps of 2001 to the deficits, thus increasing the deficit of 2009 by about 2% points, instead of using the proper stock-flow adjustment. The deficit figure has the purpose to show the operating difference between income and expenditure, and is not the place to record hidden debts that fall from the closet. It is remarkable that these sobering points have not been recognised in all comments and protests. Also the Financial Times and the Wall Street Journal supported Georgiou while overlooking these facts.

People can bet their reputation on false information though. Georgiou and the leadership of Eurostat knew about the Greek law of March 2010 and the Code of Practice of 2005, because it was their job to know this at the time. In 2010 they even undertook a change of those very regulations. They however did not say so to the courts and their international colleagues.

Instead, board member and professor of econometrics Zoe Georganta, while lacking full information because of Georgiou’s obstruction, still pointed to crucial questions on content, and those were considered by the Greek courts but inadequately treated by those protesting statisticians and media.

Impact on the EU statistical system

Cause for worry is that the European Union has now restructured its statistical system so that each national statistical office has a single head with full authority, while we know that a single person is much more sensitive to political or commercial pressure or own illusions than a multiperson board.

Warranting the quality of scientific and learned information

This is one of more information scandals that my book deconstructs. The Greek case is peanuts compared to the discussion about climate change. Overall, the world is served not only by a better position of science and learning in our system of democracy, but also by a better internal functioning of the forum of science and learning itself. There is a clear need for a National Assembly of Science and Learning that can investigate such issues without interference by political interests and mistaken hierarchies that exist now.

Thomas Colignatus is the scientific name of Thomas Cool (1954), econometrician (Groningen 1982) and teacher of mathematics (Leiden 2008). See the book “Forum Theory & A National Assembly of Science and Learning” (2020) at https://mpra.ub.uni-muenchen.de/98568 and on his website http://thomascool.eu

Today, January 31 2020, at midnight Central European Time, Brexit will happen, even though it is unclear what the British voters think about it. Brexit is neither “by the will of the people” nor “against the will of the people” but merely “without the will of the people“.

A proto-democracy generates uncertainty

The UK is only a proto-democracy and no proper democracy, see this evaluation in the APS Newsletter Physics and Society, January 2020, p18-24, which looks at the USA but the argument for the UK is quite similar.

On Brexit, uncertainty abounds:

  • The Brexit Referendum Question of 2016 was a political manipulation and unacceptable for a decent statistical questionnaire, see here p14. The situation was “garbage in, garbage out”, with ample opportunity for populism.
  • The UK preferences were rather dispersed about the options for Leave or Remain, see here p6.
  • See my summary about Brexit’s deep roots in confusion on democracy and statistics p18.
  • The UK election of December 12 was for the House of Parliament and not about Brexit. Boris Johnson had all candidates for the Conservative Party pledge to support Brexit, which runs against the principle that members of the House must represent their district. These elections thus violate the very principle of the UK proto-democracy.

The UK proto-democracy has “district representation” with “first past the post“, which means that a party may get a majority in the House of Parliament without a majority in the electorate. In the UK 44% voted for the conservatives but they still got 56% of the seats. Thus 56% of the UK voters do not want a government by the Conservatives.

Thus we still do not know what voters think about Brexit too. While Brexit was much discussed, and caused voters to switch to the Conservative Party, it still was not the only issue on the table, and it still is unclear what voters think about Brexit on balance.

The UK has the curious phenomenon of the “Re-Leavers”. These voters chose Remain in 2016 but now switch to Leave merely because this was the majority outcome in the referendum, and they “want to respect the outcome”. However, this is not how democracy works. A vote is about what you think yourself and not about what the former outcome was. Obviously these Re-Leavers are free to exercise their democratic right to think whatever they want, but this kind of thinking destroys the possibility to determine what people actually want.

YouGov tracker

The YouGov tracker is the best summary information about the general sentiment on the issue, but it is a poll and no electoral statement. Let me quote the tracking at this moment, because it always changes:

Between party dynamics

Adam McDonnell and Chris Curtis of YouGov discuss a post-election survey of December 17 2019, and here are their underlying data (for us page 3). The dynamics between the UK parties are remarkably large. Their key graph for our purposes is the following. For example 27% (figure not printed) of the Conservatives voted Remain in 2016: 22% (shown) of those switch to the LibDem, likely because LibDem are Remain. However, 65% of the Remain Conservatives stick to their party, perhaps because they regard the issue less relevant than other issue of the Conservatives, or perhaps they are ReLeavers. Of the LibDem who voted Leave in 2016 still 46% voted LibDem though it had become a Remain party, perhaps because they thought that LibDem would not gain power anyway.

Labour and LibDem could have made a deal to oppose the Conservative candidates with only one candidate from Labour / LibDem, in proportion to the forecasted vote shares. In that case, the LibDem could have assured a referendum on Brexit. During the elections, Jeremy Corbyn was criticised that he did not take a stand on Brexit, but his party was clearly divided, and his offer of a referendum was a fair option. At most five years from now there will be new elections. These are the Conservative “battlegrounds“, where this party could lose a seat by small number of voters.

Beware of John Curtice

John Curtice’s diagnosis on Channel 4, November 27, was:

“This is pretty much a binary election. Hung parliament, then we’re almost undoubtedly heading towards an extension and a second referendum, and lord knows what the outcome of that will be. Or we get a majority and we go out on January 31 and Boris is charged with the task of negotiating an alternative outcome. Ironically at the end of the day we’ve kind of stumbled into this election, but as the way it’s turning out, it’s actually providing us with a fairly clear binary choice.”

The latter is clearly nonsense, already before the election outcome. Above dynamics of UK voters shows that voters did not see a binary vote on Brexit and clearly had various considerations other than Brexit.

John Curtice is a renowned professor who on Election nights predicts the district outcomes with amazing accuracy. The problem however is that Curtice doesn’t see or explain that the true problem for the UK lies in the lack of equal proportionality in the general election. Curtice is locked in his electoral worldview like a hamster in a running wheel. Whatever he thinks and says here is in service of the current disproportionate electoral system in the UK, and then still produces nonsense.

In sum, it is the current electoral system that created the mess on Brexit and its misleading referendum question in 2016. If the UK had had equal proportional representation (EPR) like in Holland to start with, then Nigel Farage could have gotten his 12,5% of the seats in the House, and then the political discussion would have had greater restraint on the truth of the matter.

Brexit is still a mess, and now the eggs are scrambled

The solution for the present mess lies not in a new referendum on Brexit, as Curtice accepts, but in equal proportional representation (EPR). Referendum questions are manipulative, and voters cannot negotiate in polling stations. With EPR, representatives in the House can deconstruct manipulation and can negotiate. The current UK system gives only district winners, and they may be locked to a party line and cannot represent the diversity of views within their districts. The latter was already a fairy tale in 1900 and even more in 2020. Again, see my evaluation in the APS Newsletter Physics and Society.

Let the UK reboot itself. A big problem for UK voters now is: if the UK would rejoin the EU then it would have to accept the euro.