California Summer Lockdowns Have Been Ineffective

In response to the summer surge in infections in California, Governor Newsom re-instituted strict controls on July 1st closing most indoor businesses and then re-imposed a stay at home order on July 13th.  We take a look at how effective these lockdown efforts were in controlling the spread of the virus.

A simple examination of Rt (reproductive rate over time) of CA from shows that there was little variation in July.  Rt is directly proportional to the number of infectious contacts.  If July lockdowns were effective in any significant sense we would see a marked decrease in Rt.  We see very little movement.  Rt was already decaying for several weeks prior to the governor’s actions, and for the month of July Rt was fairly flat.  There is no correlation between these recent lockdown activities and a reduction in infectious contacts.  The lockdown efforts have had little effect other than to kill small businesses.

Infectious Contacts INCREASED in July Despite Government Efforts

Using our CDM analysis toolset, we take a deeper look to assess mitigation effectiveness by adjusting dynamic contact trends across age groups and to match the actual death rate curves.  We assume 50% asymptomatic rates for COVID-19 infections.  A base simulation of 2 million discrete agents was used and results are scaling to the population of 39 million CA residents.

We would expect an overall reduction in connectivity (potentially infectious contacts) in July if CA lockdown efforts were effective.  This assumption of was made in our previous analysis which understated death rates in August: California Roadmap for a Balanced Recovery.   To match actual date of death trends, our current analysis shows that connectivity must have increased in July, despite the attempt to return to the lockdown state of March/April.

The original lockdown in March was effective at slowing the spread with large compliance and little mask usage,  but since June, subsequent efforts, including emphasis on mask wearing, are having little effect on slowing the spread. In general this can be chalked up to mitigation fatigue and significant spread of the virus through means other than those under government control (private interactions, family events and socializing).

The government has limited control over people and is exerting disproportionate and punitive control over businesses and institutions with little effect, effectively flattening the economy without flattening the curve.  Worse, the focus on protecting the entire population rather than the elderly and the vulnerable is likely resulting in more deaths.

Analysis Summary

To match the current death trends in California, the contact rates were increased in July in the model as shown below. By the end of August, we estimate close to 24% of the population has been infected, and close to 33% of the 20-40 year old age group.  Significant progress has been made towards the HIT and the infection velocity is naturally slowing.

Dynamic Connectivity (Drives Infectious Contacts)
Fit of Actual Date of Death vs. Modeled Cum Deaths
New Infections and Infection Fatality Rate over Time
CA SIRD Results
Summary Outcome Data Through Month End August

Date of Death Extraction

Determination of actual date of death is critical to calibration of the analysis.   The graph below shows the derived date of death the reported death date data sets found at the COVID Tracking Project.  Our approach is far more accurate than a rolling average of reported deaths and is detailed here: Reported versus Actual Date of Death.   The most recent 2 weeks represent incomplete counts.


CA is Approaching Herd Immunity

Despite punitive government efforts, the virus will inevitably spread among younger and middle age adults until we eventually hit the herd immunity threshold (HIT) and then die out.  Our analysis in general shows that the HIT is established when approximately 25% to 30% of the population is infected and 40% to 50% of the young adult cohort (for more information see our article: Are We Closer to Herd Immunity than Most Experts Say?.

New infections peaked in early August, despite overall increasing connectivity.  Remember current connectivity is still much lower than the pre-COVID-19 baseline and as such the HIT is significantly lower.  Connectivity can be increasing and infections can still be in decline.

Overall, California is on a path to reach the HIT threshold this fall.  This trend is similar to what we have seen in Florida, Texas and Arizona where the disease is now in steep decline.  Infection distributions that show there is more mobility in younger cohorts, as would be expected, as risk for COVID death is minimal for young people and middle age adults.

5 thoughts on “California Summer Lockdowns Have Been Ineffective

  1. Have you analyzed Spanish Flu 1918 data? Are we on track to see 600,000 Americans die, like it happened during that period?

    How do you come up with this conclusion: “Worse, the focus on protecting the entire population rather than the elderly and the vulnerable is likely resulting in more deaths.” Likely? How? Why? You offer no evidence.

    What about this conclusion: “Overall, California is on a path to reach the HIT threshold this fall. This trend is similar to what we have seen in Florida, Texas and Arizona where the disease is now in steep decline.” They were in decline before, and then started leaking again when they relaxed their lockdowns. Why?

    And then this gem: “Infection distributions that show there is more mobility in younger cohorts, as would be expected, as risk for COVID death is minimal for young people and middle age adults.” As would be expected? Do you know more than virologists, infectious disease doctors, and other experts? They don’t know what makes COVID so infectious, so how do you know more than them?

    Why does US population make up only 4% of the world, but makes up 25% of the total number of cases? Does it have something to do with our slow approach to this virus? Perhaps we didn’t lockdown soon enough?

    So, to summarize: you’re suggesting that we do away with lockdown because the data shows were not going to have a repeat of the Spanish Flu Pandemic where we saw upwards of 600,000 people die?

    1. No we have not analyzed the Spanish flu. This was a different bug, likely different mutations, with much higher mortality rate (closer to 2.5% versus the standard rate of 0.1% for a typical flu). Stanford article here: CDC article describes the likelihood that we were looking at different viral genomes between the first and 2nd and 3rd waves. The mortality curve of the Spanish flu looked like a W. It heavily affected the children and young adults in addition to the elderly. COVID-19 predominately affects the elderly. Excellent CDC article here describing these details: Key differences: COVID-19 is far less lethal than the Spanish flue and only affects the elderly and those with chronic comorbidity conditions. As yet there is no evidence of any significant mutations (although that remains a possibility).

      Our analysis assesses the effectiveness of lockdown measures re-instituted in July and concludes that these mitigation measures are ineffective. So rather than ineffectively focus on protecting everyone, we suggest using that energy to protect the elderly and vulnerable populations. It is clear that there is mitigation fatigue, and if people in general have an attitude of malaise, this is likely to spread to folks who should really be careful.

      Relative to herd immunity, please review our post on herd immunity thresholds and how these are affected by heterogeneous population variations in connectivity and susceptibility: There is a strong consensus that the true HIT is much lower than 67%, and is in the range of 30%. See this MIT article: Population immunity is slowing down the pandemic in parts of the US California is approaching this number.

      Relative to the higher rates in the US, we are testing at high levels, so we would expect more cases. Relative to mortality the virus is very much a first world problem, affecting countries with older populations and excellent healthcare, where people with chronic conditions survive to old age. Much of the world population has a younger demographic with fewer co-morbidity conditions; With a lower percentage of the population surviving to old age with various chronic health conditions COVID-19 has fewer targets as it has little effect on the young and healthy in these countries.

      Additionally reporting of cases and deaths in other parts of the world is likely inconsistent. We know only 6% of US cases were from COVID-19 alone. So who knows how deaths are being counted around the world.

      We are suggesting lock-downs should be lifted and replaced with common sense guidelines advocating protection of the vulnerable and the elderly. Wear a mask indoors, social distance, work remotely, wash your hands, stay home if you are sick. Take extra precautions if you are at risk.
      Read our article here:

    1. The first graph from shows that the R(t) dropped below 1 and was level during the period where restrictions were imposed in July. Our model shows is calibrated to the death rate (fit to approximate actual date of death). To replicate this death rate, we had to INCREASE connectivity in the summer. Thus, we conclude that the lockdown measures were not effective at reducing potentially infectious contacts.

  2. Excellent analysis and rebuttal. On another level, there is the improper/misleading metric of “PCR test.” Essentially meaningless. Also, on 3.24.2020, Redfield/CDC violated federal law by unilaterally modifying the criteria for death certificates, and did so just for covid–last time modified 2003, and when modified, it is modified for all causes, yet 2020 modification was just for covid! . As of 8/28, of 160,000 reported covid deaths, if you applied the unmodified criteria, only 9000 of the 160,000 would be confirmed as Covid. This was without any notice, or any review. When seeking to modify death certificate criteria, CDC must notify public and permit comment for 60 day period, and OMB must certify. All federal agencies, including the Centers for Disease Control and Prevention (CDC), are lawfully required to comply with the Paperwork Reduction Act (PRA) and the Information Quality Act (IQA). Data being collected, analyzed, and published by any federal agency is required to meet the highest standards for accuracy, quality, objectivity, utility, and integrity as defined by the PRA, IQA, as well as additional guidelines issued by the Office of Management and Budget (OMB). This violation is just the tip of the iceberg…

Leave a Reply

Copyright 2020 Covid Planning Tools