“Follow the science” sounds like a wise course of action in the Covid-19 pandemic, or for climate change and a host of other difficult issues.

“Follow the science” sounds like a wise course of action in the Covid-19 pandemic, or for climate change and a host of other difficult issues. Business leaders trying to anticipate government policy would do well to understand where science is taking politicians. But the simple mantra of follow the science misses the key lesson of economics: choices are about trade-offs.

We cannot have everything we want. That is a fundamental principle of economics presenting in Chapter One of most textbooks, but it’s also true far afield from economics. If we want to spend more time with family, we have to give up time doing other things. If we want longer life expectancy, we have to give up some risky behaviors. That fundamental economics principle makes following the science impossible.

If we want fewer deaths from Covid-19, what do we have to give up?. It’s not just jobs and economic growth we must give up. Reducing Covid-19 deaths may increase deaths from other causes. Following the science requires understanding all the other health impacts of policy actions.

Some cancer screening save lives, though the screening conflicts with social distancing practices. This is not a case of which science to believe, but a need to fully integrate all of the scientific evidence about a myriad of issues. Scientific knowledge may lack the precision to know if the Covid-19 risks of a mammogram outweigh the cancer detection benefits of the procedure. Following the science is complicated.

Psychology research from before the pandemic found that more isolated people have a higher risk of death relative to more socially connected people. That risk of death comes from many other diseases, which presumably interact with psychological health. How do we take the health benefits of social connection into account? Ideally we’d compute risks of Covid-19 transmission as well as risks associated with social isolation. And ideally we’d run the calculations for different groups. The elderly may have different risks than the young or middle-aged. Going to that level of granularity may require us to rely on just one or two scientific studies specific to the group, rather than the more numerous studies of the general population. Targeted research often has fewer subjects in the study, reducing our confidence in the results. But most of the science shows markedly different risks for different populations, so that cannot be ignored.

And how should quality of life be weighed against mortality risk? People’s preferences tend not to prioritize life above all else. People engage in dangerous activities such as rock climbing, hang gliding and scuba diving. People engage in risky occupations that pay higher wages, including fishing, logging and roofing. Few people eat a diet calculated for longevity, preferring some risk of health problems to the pleasures of sweets and red meat.

Given people’s preferences, as demonstrated by their actual choices, following the science should reflect the pleasure that people get from risky activities, even if that risk involves Covid-19.

The most common criticism of science-based policy is that scientists don’t fully understand the problem. There’s truth in that assessment, but we have to make decisions based on the best information we have available. To give this point its due, note that early assessments of Covid-19 have turned out to be wrong. For example, Science News reports that China’s first assessment showed limited ability of the virus to be transmitted from one person to another. Then scientists acknowledged person-to-person transmission, but believed that most of the spread was from people who were symptomatic. They later came to understand that the disease was being passed along by asymptomatic infections.

Early-stage scientific knowledge must be taken as provisional, but still it’s the best information we have at the time. It would be nice to postpone decisions until years of study have been completed, but that does not work when people are dying in the present. In fact, this problem illustrates an issue familiar to economists: knowledge is a scarce good, which is often developed over time and after substantial commitment of resources. So we must decide whether to act today based on knowledge that is incomplete and possibly inaccurate, or suffer along until we have invested in developing much greater knowledge.

Using current scientific knowledge makes sense, but it must be done with humility. Scientific knowledge evolves over time. Analysis of what the science says must reflect tradeoffs among different causes of mortality, as well as quality of life preferences as demonstrated by behavioral choices that people make.

Yes, follow the science. But no, that’s not simple.

Originally published at Forbes