Context and Purpose

This document is the white paper of the Working Group, a collective of four pseudonymous professionals based in Tokyo. Patrick McKenzie served as the primary author. It was circulated discreetly during the week of March 25th, 2020, and is reproduced here unedited as a historical record. Given the rapid evolution of the covid-19 situation at the time, the contents should be read with that date firmly in mind. The URL hosting this copy is intended to remain stable for citation purposes; the Working Group does not manage any other versions of this document online.

Core Challenges Identified

The Working Group’s assessment centered on a set of structural and procedural problems that, in their view, hindered an effective domestic response to the emerging pandemic. Their observations were grounded in the operational realities of Tokyo in late March 2020.

Testing Capacity and Strategy

A primary concern was the nation’s limited polymerase chain reaction (PCR) testing capacity. The group argued that the official strategy of conserving tests by focusing on severe cases and known clusters was insufficient. They emphasized that without widespread testing, particularly of mild and asymptomatic cases, the true extent of community transmission could not be measured. This data gap, they contended, made evidence-based policy decisions regarding lockdowns or school closures nearly impossible.

Healthcare System Fragmentation

The group pointed to the atomization of medical care as a critical vulnerability. Specifically:

  • Many private clinics lacked the protocols or incentives to screen for covid-19, leading to potential transmission within waiting rooms.
  • Large hospitals, facing the risk of nosocomial infection, began refusing non-urgent consultations and turned away patients with fevers, pushing them toward an already strained public health system.
  • The absence of a centralized, digital health record system meant that patient outcomes and epidemiological data were siloed, impeding coordinated contact tracing and resource allocation.

The Working Group highlighted several structural impediments to a rapid response. The legal framework governing infectious disease control was perceived as rigid, granting the state significant power while simultaneously making decisive action procedurally difficult. They noted confusion over who held the authority to mandate business closures or school shutdowns, with responsibilities divided between the national government and local governors. This ambiguity, they argued, created a diffusion of responsibility that delayed action during the critical early weeks of the outbreak.

Proposed Tactical Responses

Despite identifying significant hurdles, the white paper put forward a series of actionable recommendations aimed at mitigating the impact while working within the existing system.

Increasing Testing Efficiency

Rather than only appealing for more PCR kits, the group proposed a more practical approach: the use of pooled sampling. By testing groups of, for example, five to ten swabs simultaneously, laboratories could drastically increase throughput with the same reagent supply. They advocated for this to be deployed in high-prevalence areas to gauge community spread, alongside reserving individual tests for diagnostic confirmation in hospitals.

Establishing a Field Hospital Network

To address the risk of overwhelming existing hospitals, the white paper recommended the immediate establishment of designated covid-19 treatment centers. These would ideally be repurposed facilities with independent ventilation systems to house mild-to-moderate cases. This would serve two purposes: freeing up beds in acute-care facilities for the critically ill, and breaking the cycle of infection that occurs when covid-positive patients are admitted to general wards.

Protecting the Medical Supply Chain

The report flagged the vulnerability of the supply chain for personal protective equipment (PPE) and basic supplies like surgical masks. They advised against relying on spot-market purchases of N95 masks and instead recommended that hospitals form a collective purchasing cooperative to aggregate demand and secure supply lines. This was framed as an emergency measure to ensure that frontline workers were not forced to ration protective gear.

Data-Driven Public Communication

The Working Group urged officials to move away from vague directives toward specific, actionable communication. They recommended the release of granular data on hospital capacity, case counts by municipality, and testing numbers. They stressed that a public that understood the rationale behind restrictions—such as avoiding the "3 Cs" (closed spaces, crowded places, and close-contact settings)—was more likely to comply with them during the extended state of emergency they predicted.

Japan’s Public Health Response Is Built on an Undercount

Japan’s widely reported success in containing covid-19 is not supported by the evidence. Official case counts are likely understated by a significant margin—possibly by a factor of five or more. The cluster-based containment strategy has not prevented community spread, and the country is not implementing meaningful social interventions. Without immediate policy shifts, Japan faces a national-scale public health crisis within weeks.

Update, March 26th: The government’s panel of experts has now stated that infections are likely “rampant,” concurring with the assessment below. No official projections consistent with this white paper have been published, except for one from Osaka.

Why the Official Numbers Don’t Add Up

As of March 24th, Japan reported 1,135 cases (excluding those aboard the Diamond Princess, charter flights, and attending officers), of which 859 were active, 54 critical, and 41 fatal. The government’s stated approach is to treat and contain diagnosed clusters while taking a wait-and-see attitude on broader measures. If the true case count is much higher than reported, the premise that clusters are being quickly identified and contained collapses.

Statistical evidence suggests containment failure has already happened. Several independent lines of reasoning point to a substantial undercount.

Asymptomatic Carriers Are Being Missed

Japan’s testing policy has excluded asymptomatic individuals except for those with deep, direct contact with a confirmed case (濃厚接触者), and initially only tested even those individuals if they had fever or respiratory symptoms. This policy is expected to miss asymptomatic infections, and the data supports that expectation.

Ministry of Health, Labor, and Welfare statistics for patients diagnosed via PCR on March 10–14 and March 23 show that asymptomatic patients consistently make up roughly 10% of all diagnoses (ranging from 10.24% to 11.29%). This rate is far below what should be expected for covid-19. On the Diamond Princess, where all passengers were tested repeatedly, the National Institute of Infectious Diseases reports that 48% of infected patients were asymptomatic at the time of sample collection.

If the true asymptomatic rate is higher than the observed rate, and we assume perfect identification of all symptomatic patients, then a large population of infectious but asymptomatic carriers exists outside official counts. They are not subject to restrictions, monitoring, or contact tracing—and they do not know they can spread the disease.

Estimating the true number of infections under this assumption, using March 22nd MHLW data:

907 / (1 - 0.48) ≈ 1,750 true infections

That is approximately 70% higher than the reported count for that day. Every missed symptomatic infection also implies a missed asymptomatic infection.

Other estimates support similar conclusions:

  • Iceland: Its chief epidemiologist, with near-universal testing capacity, reports that 50% of infections are asymptomatic. If the ratio held in Japan, the country would have ~80% more infections than publicly reported.
  • Japanese evacuees from Wuhan: A pre-print in the Journal of Infectious Diseases by a Japan-based team, using exhaustively tested returnees, arrives at an asymptomatic rate of 30.8%, implying ~30% more domestic cases than reported (again, assuming perfect symptomatic case identification).
  • Diamond Princess modeling: A Eurosurveillance pre-print estimates the true asymptomatic rate at 17.9% using more sophisticated modeling, still implying an undercount.

Deaths Imply Far More Infections Than Reported

A pre-print from the Centre for Mathematical Modelling of Infectious Diseases uses delay-adjusted case fatality ratios to estimate under-detection. Reasoning backward from observed deaths, the research estimates Japan’s detection rate at between 15% and 35%—meaning the country is undercounting cases by a factor of approximately 3X to 6X.

Other Countries Detect Japan-Origin Infections Japan Misses

Singapore traced four coronavirus cases to individuals who had recently been in Japan. Singapore’s airport, with roughly 64,000 passengers arriving from Japan over a two-week period, found infections at a rate of about 47 basis points—5X the reported 9 basis point infection rate in Japan. This suggests tourists are not stumbling into the exact same clusters as residents; rather, they are encountering unscreened community transmission. If tourists are affected only by community spread, the official Japanese count would be understated by 25X or more.

Containment Has Already Failed

Prefectural officials are starting to acknowledge the reality. In Osaka and Hyogo, the current epicenters, a document dated March 16th and shown on TV by Governor Yoshimura on March 21st admits that cluster containment is failing, citing new infections without a detected link to an existing cluster. It states: “It is believed that infections without surveilled chain to a cluster continue to increase and that therefore a rapid increase in infections has already begun.”

Aichi Prefecture (Nagoya) faces a distinct crisis: over half of its high-grade infectious disease beds (161 of them) were already in use as of March 11th. As of March 22nd, the prefecture reported 103 hospitalized patients, of whom 27 are not linked to any known cluster. Nagoya is likely in a state of uncontrolled outbreak, with medical care expected to suffer within days.

Japan’s overall approach—cluster identification, surge medical attention, and voluntary behavior changes—has not held. The voluntary measures have been modest and insufficient to prevent the situation from worsening.

Internal travel is largely unimpeded. The Tokaido Shinkansen carried roughly 500,000 passengers per day between Tokyo and Osaka in normal times; February usage declined by only 8%. An outbreak in one major metropolitan area will almost certainly metastasize nationwide without aggressive movement restrictions, especially on public transport.

Societal Preparation Remains Inadequate

Public cooperation and mask-wearing culture are not a strategy. Breakout infections are already being observed. Spread in Japan should be assumed to resemble that of peer nations taking minimal precautions.

Concrete actions so far: nationwide school closures on February 27th, two weeks before spring holiday, and discouragement of large events—but only after a live music event generated Osaka’s first cluster. Otherwise, it is largely business as usual.

  • Mask wearing in central Tokyo is below 30%, including at well-attended outdoor hanami parties.
  • Reporters have not observed social distancing or universal mask use even at press conferences about the epidemic.
  • Government guidance recommends masks mainly in enclosed, high-density spaces with vocalization, but has not closed bars or restaurants and appears to carve out mass transit and hanami.

April Will Bring a National Health Crisis

Osaka’s own forecast, included in the governor’s document, projects 3,374 infections (including 227 severe cases) before April 3rd, with a compounding rate exceeding 6X per week. New York, with aggressive mitigation, has been compounding at approximately 2X per week.

Two scenarios illustrate the near-term trajectory:

  1. Optimistic: Using doubling rates from peer nations under aggressive measures, with current official counts, Japan would see on the order of 3,000 cases, including more than 200 severe cases, in each of Nagoya, Osaka, and Tokyo by the end of April.
  2. Pessimistic: Using the Osaka working group’s growth estimates and without aggressive intervention, infections could be ten to one hundred times that number.

Both scenarios lead to an “overshoot”—the breakdown in care capacity that Japan has explicitly warned about—with sharp increases in deaths, including among patients with unrelated conditions as the medical system becomes overwhelmed.

Tokyo announced on March 23rd that it had 118 beds appropriate for high-level infectious disease patients, with plans for an additional 700 severe-care beds and 3,300 for moderate cases. The real constraint is not beds but skilled personnel. Japan has an ongoing shortage of such staff even in normal times and cannot surge them from unaffected regions—which will soon be affected themselves.

Applying even optimistic growth rates to an understated baseline yields a grim picture. A realistic projection must account for the true infection count:

We project a true count of over 500,000 infections, including more than 5,000 severe cases, with a breakdown in care in Nagoya, Osaka, and Tokyo before the end of April. Other national breakdowns are almost certain. April will not be the worst month.

White Paper Methodology and Conclusion

This analysis is based on published official statistics, pre-print research, and cross-national comparisons. The figures presented here are estimates from a range of reputable sources, not firm predictions. The core finding is consistent across all methodologies: Japan’s official case count is materially lower than the true number of infections, and the gap is large enough to invalidate the current public health strategy.

Japan must act immediately. The time for a wait-and-see posture has passed. Immediate, concerted, and aggressive policy interventions are required to prevent a public health catastrophe that will otherwise unfold within weeks. Tomorrow will be worse than today—and the costs of delay will be measured in lives.