Drugs, Suicide, and Crime: Empirical Estimates of the Human Toll of the Shutdown

September 7, 2020

As the coronavirus pandemic continues across the world, leaders and policymakers have scrambled to respond to the growing health crisis. In the United States, multiple state governors have issued statements urging their citizens to follow social distancing guidelines.

Other governors have taken more extreme measures, issuing orders to effectively lock down the entire state economy. The current goal of these responses has been to slow the spread of the virus in the hope of reducing the strain on the healthcare system. Discussion over the proper precautions is a necessity in such a time. 

There have been forecasted estimates of virus-related death totals for the US from as high as 10 million, to 2.2 million, to more conservative estimates of 5,000. The models used to estimate the potential death rates are not without criticism and repeated adjustment. Sampling bias may be a significant problem. These data errors are an important problem to resolve as policymakers use these models to inform their responses. 

The difference between social distancing and complete economic shutdowns is too dramatic not to be taken seriously. It is imperative that more testing be conducted to provide better access to data, as well as the health benefits that come with knowing who does and does not have the virus. However, as important as it is to get the cost of not shutting down right, it is also important that policymakers properly weigh the cost of the economic shutdowns themselves. 

Getting the cost right is not simply a matter of valuing “profits over people” as the social media memes may suggest. Rather, even in times of crisis, the ability to operate in a functioning economy is important for the people within it. The economy is the people, and the people are the economy. The ability to continue to function in a market system does matter to individuals within the system, particularly when the inability of business to remain open and continue to employ them is in question. 

We have already started to see some of these human effects as the unemployment has quickly rocketed beyond even the early initial projections. A rise in unemployment is correlated with a number of negative socio-economic effects. For some, these effects can be quite deadly, particularly when the changes are rapid as is currently the case.

The economic predictions for the shutdowns may be as varied as those for the virus itself. The Federal Reserve’s James Bullard has noted that unemployment may rise to as much as 30%. Treasury Secretary Steven Mnuchin has estimated a possible unemployment rate of 20%. Bullard’s number is higher than the unemployment seen in the US during the Great Depression (25%), and both estimates are significantly higher than the unemployment during the Great Recession (11%). Even if we take the more conservative estimate of 20% unemployment, that is a 16.5% rise in unemployment from its recent historic lows of 3.5% unemployment. 

Though it is difficult to estimate how long this downturn may linger, that is a severe shock to the economic system. It is possible that people return to work and economic activity returns in strength in short order after the shutdowns are lifted. Even then, the costs of shutting down will have been quite large. However, it is also possible that some businesses who had to pause activity for a month or more may not be able to return at all. The recession could be longer than some economists are projecting. If the economy does linger in its downturn, the human costs to the shutdown will inevitably begin to increase. 

A 2017 NBER paper finds a 3.6% increase in the opioid death rate per 100,000 people for a 1% rise in unemployment. There were 14.6 opioid deaths per 100,000 in the US in 2018. If we use the more conservative estimate of a 20% unemployment rate without a quick return to lower levels, then there would be an estimated 59.4% rise in deaths per 100,000, leading to an increase of 8.7 deaths for a total of 23.3 for opioids. 

With a current US population of 331 million, there are 3,310 groups of 100,000, meaning there is potential for an additional 28,797 deaths from opioids annually. Consider that for 2018, the CDC reports that there were 67,367 deaths from all-drug deaths, with 46,802 of those coming from opioid use. The 46,802 deaths were considered an opioid crisis. A possible 75,599 should not be dismissed quickly.

The negative effects will not be felt just through opioid use either. The numerical increase in deaths provided above is only for opioid users, but the all-drug death number will rise as well. In a 2018 study, Bruguera, et al, found that of the 180 drug users they surveyed about use during the Great Recession, 58.3% reported an increase in use while only 25.6% reported decreasing use, resulting in greater all-drug use for the period. Similarly, Mulia, et al, (2014) connects a rise in alcoholism to economic loss during the Great Recession. The CDC estimates that 2,200 people die in the US just from alcohol poisoning annually, not to mention the additional alcohol-related deaths that occur. In 2017 alone, there were also 22,246 deaths resulting from alcoholic liver disease.  As the jobless rate increases and the economic losses continue to mount, these numbers are likely to rise.

The deaths related to economic downturns go beyond those from chemical dependency, also. The mental toll is not inconsequential. For example, Blakely, et al, (2003) find that being unemployed may also increase the risk of suicide two to threefold. Milner, et al. (2014) similarly finds that unemployment is associated with a higher relative risk of suicide, with prior mental health issues being a key factor in that association. While a study by Kerr, et al, (2018) did not find that unemployment is directly linked to suicides, it did find a significant link between poverty, suicide, and alcoholism. 

When breaking the population into age groups, Lin and Chen (2018) do find that unemployment does have a direct impact on older portions of the population, the portion of the population many of the current shutdowns are most meant to protect. Whether it is the direct unemployment effect or the potential poverty produced from the economic shutdown that leads to greater suicides, an increase from the 48,344 suicides and 1,400,000 suicide attempts in the US in 2018 should give decision-makers pause during their response to this pandemic. 

Increased harm to oneself is not the only harm caused by economic downturns. There is also the threat of rising crime in general. Ajimotokin, et al, (2015) estimate that a one percent change in unemployment will increase the property crime rate by 71.1 per 100,000 people and the violent crime rate by 31.9 per 100,000 people. 

With our estimated 16.5% rise in unemployment, we could see a significant increase in both property and violent crimes. The violent crime also may add to the death toll in this period.  Kposowa and Johnson (2016) find that unemployed workers are over 50% more likely to become homicide victims than those who are employed. They also find people not in the labor force to be 1.3 times more likely to be victims than those who are employed. As workers become discouraged due to an inability to find jobs during a recession, their lives as well as their livelihoods are called into question.

The future during such a pandemic is largely uncertain, and misinformation is rampant in the current panic. Policymakers face tough decisions as they navigate the issues of data collection, virus transmission, and economic ramifications of doing too little or too much. It is vitally important, literally life and death, that the proper costs and benefits are weighed with the decision on how much and how long to shut down economic activity through the pandemic.

*AIER is a 501(c)(3) Nonprofit registered in the US under EIN:04-2121305