Tuesday, June 9, 2020

COST OF PREVENTING A DEATH FROM COVID-19 IN CANADA




What has been the cost of the economic lockdown and other mitigating policies imposed by Canadian governments to reduce the spread of the Covid-19 virus? What has been the dollar value of benefits created by these policies? Have the costs been more or less than the benefits?

A recent study has answered these questions for the United States. It found that the benefits much exceeded the costs. The problem with this study is that calculated benefits by using the Value of Statistical Life, which is estimated to be $10 million. This amount is much greater than other value of life figures used widely: $125 thousand, the quality adjusted life and $2.7 million, life-time earnings of the average Canadian (in 2017). The use of these lower values would show the US costs to be higher than the benefits of the country’s mitigation policies.

To avoid the need to choose a specific value of life figure in a cost/benefit study for Canada, this study instead estimates the cost of saving a life resulting from that country’s mitigation policies. It does so by dividing the cost of these policies by the number of lives saved. The resulting figure allows readers to decide whether the cost of preventing deaths was reasonable and whether the government should consider alternative mitigation strategies to fight the possible second wave of the pandemic or new outbreaks caused by mutations of the current virus.

Economic Costs

Economic costs are the result of government policies requiring Canadians to self-isolate, keep social distance from others and the lockdown of productive activities such as construction, the provision of hospitality, travel, and education. According to the International Monetary Fund this reduction will lower Canadian GDP in 2020 by 6.5 percent below that of 2019. Since GDP in 2019 GDP was $1.7 trillion, the value of reduced output in 2020 is estimated to be $113 billion (6.5% of $1.7).

It is important to note that government payments to individuals and business to prevent financial suffering and bankruptcies do not lead to lost output because these payments add to the public debt, which in the future will require interest payments that are made by taking money from some Canadians (the taxpayers) and transfer it to other Canadians (who hold the debt obligations).

Economic Benefits – Number of deaths prevented

Government Covid-19 fighting policies reduce the number of deaths that would have occurred in the absence of these policies. This information is found in computer models produced by infectious disease specialists, which predict the number of deaths likely to occur with and without mitigating government policies, the difference representing the number of deaths prevented.

Such a number was produced in the study by the Imperial College of London (ICL) and is used here. Only data for North America rather than Canada are available, which show 8 deaths per thousand without and 4 per thousand with mitigating policies (p.7). These data imply that with Canada’s population in 2020 at 37.7 million, there would be 301,600 deaths in the in the absence of and 150,800 with mitigation.

Cost per Life Saved

Model 1
Using the IMF estimates of $113 billion in cost and the ICL estimate of 150,800 lives saved, the cost per life saved is $749 thousand.

Model 2
The estimate of the cost used above draws on the IMF predicted loss of output in the year 2020. It is likely that the reduction in output will last beyond the end of that year. Assuming that the final cost it will be $226 billion, twice the 2020 amount, and that the number of lives saved remains the same, the cost per life saved is $1.5 million.

Model 3
At the end of May 2020, according to Health Canada, the number of deaths attributed to Covid infections was 6,583, a much smaller number than that predicted by the ICL model. To estimate the number of deaths prevented, this study assumes that the mitigation policies have reduced the numbers by 50 percent, the rate found in the ICL model. Under these assumptions, the number of deaths without mitigation would have been 13,166 and the reduced and the actual deaths 6,583 each. Given the economic cost of $113 billion as predicted by the IMF, the cost per death prevented is $17.2 million (113 divided by 6,583).

Model 4
It is safe to predict that the costs and number of deaths prevented will grow beyond the numbers used in the preceding estimates. If the final cost of the policies will be $226 billion, double the amount predicted by the IMF in 2020 and the number of deaths prevented is 9,875, which is 50 percent higher than the 6,583 recorded at the end of May, the cost per death prevented is $22.9. million.

Model 5
The pandemic-fighting policies have increased unemployment, which is known to raise the number of suicides. A recent study estimated this number to be 2,100, which reduces the number of deaths prevented to 7,775 from the 9,875 assumed in Model 4. With costs at $226 billion, the cost per death prevented is $29 million.

Policy Implications

In my judgement, the results of the first two models are the least realistic since they are driven by the ICL estimate of deaths prevented, which is much in excess of those found in Canada. Most realistic seem to be the result of Model 5, which assumes costs that are justified by economists’ growing concerns about the length of time required before  the economy is fully recovered and uses an assumption about the ultimate number of deaths that is consistent with the recently observed decline in numbers. The evidence on the relationship between unemployment and suicides is strong and the numbers found in the cited study seem reasonable.

According to Model 5, the estimated cost per life saved is $29 million, which is near three times the Value of Statistical Life figure of $10 million used in the US study cited above, which concludes that US pandemic fighting policies have resulted in benefits exceeding costs. Whether one considers that the same conclusion is warranted for Canada’s policies depends on one’s assessment of the value of life. But regardless of one’s views on this matter, the high costs raise questions whether other mitigating strategies exist, which are less costly, could have been used and should be considered for use in fights against future pandemics.

Costs of Alternative Mitigation Policies

A less costly method for fighting the Covid-19 pandemic uses policies that rely on three premises.

First, deaths caused by the Covid-19 virus are concentrated among the elderly. At the end of May, 6,538 Canadians with the infection had died. Of these, 4,694 or 72 percent were over the age 80; 1,640 or 25 percent were between the ages of 60 to 79. Therefore, 97 percent of Canadians who died from the virus were older than 60.

Second, the symptoms experienced by the vast majority of infected individuals below the age of 60 are mild, do not require hospitalization or lead to death. Some of the young require hospitalization or die in numbers remarkably similar to those afflicted every year by the seasonal influenza. These problems of the young should not be taken lightly but have not in the past been considered serious enough to warrant costly mitigating policies other than vaccination.

Third, individuals who have recovered from the infection are immune and will not become ill again for some significant length of time.

Assuming that these premises are correct or need only modest modification, the following policies promise to be much less costly than those used presently.

·     *  No distancing and lockdowns are mandated.
·      * A massive effort is made to shield the elderly from infection until a vaccine is developed.
·      * The small number of infected young who have severe, life-threatening symptoms receive all the     healthcare they require.

Such a policy regime would avoid the large costs caused by the currently used mitigation policies because the cost of protecting the elderly is certain to be far below the $113 billion of the current policies. The cost of treating the young needing hospitalization would also be relatively small.

The costs would also be lowered by the fact that the proportion of the young who have recovered from infections will relatively quickly be large enough to create herd immunity and allow a gradual reduction in the severity and cost of protecting the elderly and of treating young requiring hospitalization.

Sweden offers evidence on the working of this model in a real-world setting. The fact is that Sweden’s loss of output was much less than that of Canada’s. In the first quarter of 2020 Sweden’s economy grew 0.1 percent while that of Canada fell 2.1 percent. Other factors may have contributed to this difference, such as the decline in Canada’s energy industry, but are unlikely to be significant enough to negate the validity of this comparison.

However, the comparison of death rates much favors Canada. At the end of May the deaths per 100,000 were 41.9 in Sweden and 18.9 in Canada. This difference between these death rates is puzzling because it is almost certain that the age-pattern of deaths is the same in the two countries. If that is correct, over 90 percent of all deaths in Sweden are among the elderly and residents of care homes. It follows that Sweden’s higher death rate is caused by its failure to protect this age cohort as well as Canada has.

Why this happened raises important questions for future research. Are there differences in the two countries in the proportion of the elderly in the population, in the design and staffing of care homes, in the amount of resources devoted to insulating the elderly and health-care providers, and others?
In spite of the present lack of reliable answers to these questions, the analysis presented here suggests that policy makers in Canada should seriously consider the Swedish model when they have to deal with the expected second wave of infections in Canada and possible future problems created by aggressive mutations of the existing virus.

Would it be better to focus on the reduction of deaths rather than infections?