Risk Management 101: The Precautionary Principle


Is it worse to send an innocent person to prison, or to let a guilty person escape punishment?

In liberal society, we believe that it is far worse to take away an individual’s liberty without just cause (authoritarian cultures see it differently). We express this belief by setting a higher bar for conviction than acquittal. For the court to send the accused to prison, the prosecution must prove “beyond all reasonable doubt” that the defendant is guilty – I guess that’s why they use the word ‘conviction’.


The Trade-Off

We know that this approach will occasionally let criminals off the hook, but that is a price we are willing to pay to maintain the liberal and pro-social presumption of innocence in society. Various philosophers have tried to put a number on the acceptable number of crimes that would go unpunished in order to protect the liberty of one innocent person, but this is a fool’s errand. What matters is that we know which side caution lies.

It is possible to guarantee that no innocent person ever goes to jail. It would be cheap and easy too: stop sending people to jail. Unfortunately, that policy also guarantees that no crime would ever be punished, which is not at all desirable. Similarly, we could guarantee that no crime ever went unpunished by sending all accused to jail. It is an open question as to which of these societies would be the least pleasant.

Instead, we accept this imperfect trade-off, choosing to resolve it by maintaining the presumption of innocence until it can no longer stand, because that is the better mistake to make.


The Scientific Threshold

A similar trade-off exists in science, where it is far more harmful to believe a relationship exists when it does not, than to miss a true relationship where the data was inconclusive. Imagine building a bridge with inadequate materials, or selling medical treatments based on faulty science. It only takes one mistake or false positive to destroy the lives of thousands of people. In fact, you don’t have to use your imagination. We have done it many times already, and Thalidomide is first to mind.

In both law and science, we recognise that we have a lot of decisions to make and that we are unlikely to reach the right conclusion every time. We will try our best, but in the face of inconclusive evidence, we need something to guide our decision-making. How do we handle the grey areas? Where do we put the benefit of the doubt?

Statisticians call this a trade-off between Type 1 and Type 2 errors. The medical scientists call them false positives and false negatives – we wrote about them previously. Normal people just call it erring on the side of caution, and that you are “better safe than sorry”.


A Decision-Making Framework

This concept leads us to a simple framework that you can use every day to make better decisions. In fact, many of us are already familiar with it.

If you are in university and you are preparing for final exams, how much should you study? That’s a difficult question to answer exactly, so let’s reframe the question. Which is the worst mistake to make: to study too much, or too little?

Given that these exams account for 100% of a 3-year degree result, I would say it’s better to study too much. You might lose a few hours of gaming to ‘excess’ study, but those hours cost less than repeating the whole year. Speaking as a risk manager, I’d rather have a margin of safety around my goals.

If you have an important appointment at 3pm, and it will take you about 30 minutes to get there, at what time should you leave? If it's important, then I'd imagine that being too late is much worse than being too early. Whatever time you're thinking of leaving, go 10 minutes earlier. Err on the side of caution.

If there’s a storm coming, should you batten down the hatches earlier or later? It’s possible that the storm might miss you, but I’d say batten down the hatches now. If the storm passes by, you haven’t lost much. If you leave it too late, you could lose everything.


There was a Storm Coming Alright...

That was the situation that Ireland’s policymakers found themselves in January 2020. There was a nasty contagious virus circling the globe. It was causing all sorts of trouble where it originated in China, and it had infected dozens of different countries, including most of Ireland’s closest neighbours.

Ireland’s leaders had a choice: batten down the hatches or wait to see if the storm will pass by.

If they had battened down the hatches too soon, some travellers would have been inconvenienced and the government might look a bit silly. Businesses would take a hit on that quarter’s revenue too, and the industry groups would be very sour about that.

On the other hand, if they went too late, they would allow a deadly, contagious virus – one for which they had neither vaccine nor cure – into the population. That would risk an outbreak that could kill tens of thousands, destabilise our society, and degrade our quality of life for years to come. By allowing the virus to spread around Ireland and then leave our shores to infect other populations, this policy would also sustain the outbreak and help to turn it into a pandemic.

It was a volatile and uncertain environment. There was no rule book, no memorised answers, no leadership from Brussels. What should Ireland do? On which side did the caution lie?

In my opinion, doing too little too late was the worst mistake we could have made in that situation. I think doing too much too soon would have been the better option, or better safe than sorry. I believe I share that opinion with millions of Irish people.

Sadly, Ireland’s policymakers didn’t see it that way. Fine Gael and NPHET did nothing in January and February to prevent the virus entering the population, and we have been living with the consequences of that decision ever since. Ireland’s media didn’t seem to have a problem with it. Nor did Ireland’s expert community, who were completely absent in January and February 2020 and only found their voice after the damage had been done.


First, Do No Harm

What I find most remarkable about what our policymakers didn’t do in January and February 2020 is that they had all the information they needed to make these decisions. I have presented the ideas in this piece from my perspective: that of a statistician, and a risk manager. But these aren’t niche or esoteric ideas. They have been understood and communicated for centuries, even millennia.

Perhaps the most famous formulation of erring on the side of caution or “better safe than sorry” is the Hippocratic Oath. It is inconceivable that the doctors and medics on NPHET – and the then-Taoiseach who had also trained as a doctor – weren’t familiar with either the Oath, the phrase “First, do no harm”, or the precautionary principle that threads this whole piece together.

The failure of Ireland's policymakers to apply the precautionary principle in January and February of 2020 allowed the virus into the population to circulate freely, and that one error explains the vast majority of the deaths and suffering that the Irish people have experienced ever since. It was a grievous error, and grievously have we paid for it.