COVID-19 has taken a devastating toll on Wisconsin, some of this is attributable to the disease itself — fear, serious illness or, even worse, the loss of a loved one. Some of it has been caused by our response to the disease — the loss of a job or the destruction of a family business.

While the novel coronavirus is dangerous and requires a serious response (it is not the equivalent of the flu), “sheltering in place” and waiting for the virus to go away is not sustainable. Businesses will fail and families will be unable to support themselves. Supply chains will break down and trillions of dollars in wealth will be destroyed. A second Great Depression could exceed the devastation wrought by the first with its own devastating impact on our mental and physical health. A “hard pause” to “flatten the curve” was not intended to — and cannot — be maintained indefinitely.

So Wisconsin’s economy must open and soon. However, wholesale reopening with no social distancing whatsoever could prove to be equally devastating. In addition, allowing businesses to reopen will do little good unless we can devise some way for Wisconsinites to assess the risk of renewed activities and take appropriate steps for their own safety.

The approach put forth by Wisconsin Manufacturers and Commerce (WMC) and state Senator Chris Kapenga (R-Delafield) represents a middle ground between Governor Tony Evers’ “Safer at Home” and President Donald Trump’s “Re-Open America.” The model provides for a smart and nimble response to local outbreaks as they might occur. It will also provide Wisconsin businesses with predictable and sensible guidelines to re-open — moving beyond the often arbitrary “essential” vs. “non-essential” distinctions. It provides a way for us to properly assess the risk associated with patronizing a particular business.

The ‘Back to Business’ Plan

The WMC ‘Back to Business’ proposal to reopen Wisconsin takes into account four factors at the county level: hospital capacity, infection rates, population density, and the extent to which a particular business involves close human interaction. The plan recognizes that Wisconsin is a big state and that the spread of the virus differs dramatically across the state. Put differently, while the virus travels, it is not present to the same degree everywhere.

The WMC plan takes this variation into account. It uses this seven-day rolling average for each county. Population density is based on census data for each county. Interactive concentration is a measure of how much person-to-person contact a business requires. Retail establishments, for example, will be in the highest risk category while businesses where more social distancing is possible, such as an accounting firm, will be in lower risk categories. Hospital capacity represents the extent to which local resources are currently being taxed by COVID-related cases. Because the overall goal of curve-flattening is to keep case loads within our capacity to treat them, this is a vitally important measure to include in any plan to re-open.

Each of these factors are equally weighted and are measured on a 3-point scale on a weekly basis. These factors are multiplied together to create a risk score. Each week, every business in the county will visit a website to determine their score for the week, and the amount of social distancing that will be required.

Factor 1: Accounting for Variation in Infection Rates

“One size fits all” solutions that fail to take into account the variance in local conditions generally aren’t the best fit. The same is true with COVID-19.

Risk of infection is based on the number of people you interact with who have been infected. One measure we could use is the number of confirmed cases. It is not perfect. The actual number of infections is certainly much higher due a lack of testing over time and because it appears that a high percentage of infected persons COVID are either asymptomatic or have only mild symptoms. The number of confirmed cases will presumably reflect most of the cases severe enough to have sought medical attention.

The WMC plan uses a different measure. It focuses on the percentage of persons who test positive for the virus. This can be a good measure because it tells us what’s happening recently (as opposed to a cumulative infection rate which tells us what has happened from the inception of the crisis.). More fundamentally, if testing is sufficiently widespread, it can capture the trend in cases where persons have mild symptoms or are asymptomatic, ensuring that any differing trend in that population will be reflected in our measure. With adequate testing, it may also be an earlier indicator of spread of the disease. So using this measure does require a certain level of testing and a consistency in the criteria for who is being tested (so that changes in the percentage of those testing positive is not caused by changes in who is being tested). But it does not require universal testing.

Just as with infection rates, this measure shows that the spread of the virus throughout the state is uneven. Particularly in rural areas, the risks have remained quite low. Using data from the Department of Health Services, we calculated the seven-day rolling average of the percentage of persons who have tested positive for COVID-19 by county.

Data reflects the week ending April 17, 2020.
Data reflects the week ending April 17, 2020.

The variation in infection rate across the state is substantial. If infection rates in Vilas County are less than half that in Brown County, it is simply not reasonable to require the same level of social distancing in both places.

The WMC proposal deserves credit for taking into account this variation in county infection rates and allowing for local variation in the levels of social distancing depending on local circumstances. The governor’s plan does not. We should not keep the entire state under lockdown because of problems in a few places. To be sure, vigilance against spread of the disease is appropriate everywhere. But “sheltering in place” and other more extreme measures may not be.

It should also be noted that it is not only conservatives who are endorsing more regionalized reopening plans. Pennsylvania under Democratic Governor Tom Wolf, for example, has announced that certain parts of Pennsylvania with low infection rates may begin a phased reopening plan soon.

While it is true that the virus can “travel,” this pattern has been consistent over the past six weeks. More importantly, by looking at the rate of positive tests, the WMC plan uses a “sentinel” method that can pick up changes in the presence of the disease rather than lock down places where there is no substantial risk.

Factor 2: Accounting for Population Density

Population density is among the most important factors in predicting the spread of epidemics. This is intuitive. Where people live closer together, social distancing is more difficult and there are more readily available “hosts” for the spread of the virus. We see this in our own state, where among the highest rates of infection and death have been around Milwaukee, and nationwide where New York has experienced the greatest devastation. The bottom line is that the more people who are in close contact, the more risk for contracting the virus.

While including population density as a static measure in the model is likely to cause pain for residents of Milwaukee County (including one of the authors) and other urban areas, this is based on sound science. While one can argue about the weight that should be placed on this factor, it should be part of the assessment. There are, in fact, studies that show this variable can be among the most important predictors of spread of the disease.

Factor 3: A Dynamic Model

The local circumstances vary on an almost daily basis with COVID-19. New “hot spots” have popped up all over the country, while other areas have peaked and faded in the level of urgency required.

By requiring that companies “check in” on a weekly basis with regard to their county’s status, this changing status can be accounted for. If a particular area has a spike in infections, or a straining of their hospital capacity, more social distancing will be implemented to stem the growth of the virus. At the same time, if circumstances in an area improve, more regular economic activity can begin again much more rapidly than is the case under our current blanket, statewide order.

In order for this plan to be effective, it is vital that we continue to grow our testing capacity. We must be confident that the rate of infection found for a particular area is actually representative of the true extent of infection, and we are not there yet.

But a few caveats are in order. It is not necessary that testing be universal as long as it is sufficiently broad. To date, we have focused testing on certain occupational groups and persons exhibiting certain symptoms. But we would also do well to conduct a certain number of tests of the general population.

One promising proposal would be to test all of those who present at a physician’s office or clinic for routine care. While this is not a perfectly representative population, it may be adequate to pick up changes in the spread of the disease. There may be other proposals for more randomizes testing as well.

If we cannot adequately ramp up testing by the date that we wish to begin using something like the WMC plan to reopen, policymakers might want to consider using the infection rate for risk rating until adequate testing comes on line.

To the extent that this testing challenge can be met, the approach that is being put forth by business leaders represents a viable approach to get Wisconsin working again without requiring the whole state to be closed for an indeterminate amount of time.

Rick Esenberg is President and General Counsel for the Wisconsin Institute for Law & Liberty.

Will Flanders is the research director for the Wisconsin Institute for Law & Liberty.