Economic losses and counterfactuals

Main article: Economic Systems

Main article: Systems thinking

$5.9 billion is a lot of money but probably no one actually “felt” the loss. This is usually the case with financial fraud. If someone robs our bank, we do not personally feel the effect of that. For one thing, the FDIC insures our accounts so the loss is only felt collectively. But another problem, obviously, is that much of what counts as a loss is only measurable in the future. In other words, the loss is the difference between what we would have received in the future (if the bad thing hadn’t happened) and what we actually received. Since we never lived in the future where the loss hadn’t happened, we don’t see it as a “loss”. Imagining the future where the loss hadn’t happened is a counterfactual scenario.

Counterfactuals are not just the bane of collective financial losses like those above but also of economic policy. If the government chooses policy A over policy B, we only get to live in the future where Policy A was selected. There is no “control group” future that allows us to make a real comparison. An understanding of policy B is the counterfactual scenario and it is essential to be able to do that, especially in cases where substantial losses occur.

The best way to run complex counterfactual scenarios is to model them with one of the simulation methodologies we discussed in the past. Our community should have the ability to rate policies not just on their actual effect but also against a carefully constructed counterfactual that serves as a control group. In fact, since policy effects are frequently confounded by other variables, the only way to know the effect of a policy is to compare its real effect against a counterfactual simulation where everything was the same except that the policy in question was not enacted. All of this can be approximated through simulation.

Counterfactual economic policy is difficult but losses due to crime and corruption shouldn’t be. It is estimated that white-collar crime, almost all of which is financial in nature, costs the US economy $1 trillion per year, whereas street crime costs only $15 billion per year. The $1 trillion is 4% of the entire economy. That is, if we run the counterfactual where financial crime didn’t occur, we would all be 4% richer per year. That’s a lot, especially compounded over time.

A community performing simulations on financial crime might trade off the cost of regulation or privacy intrusiveness vs. not losing money to some potential fraudulent scheme. It wouldn’t be easy to estimate, but after a few years of experience reasonable probabilities and correlations should become evident. One of the great strengths of our communal ratings-based system is the ability to experiment and optimize.

Another thing a community can do is make sure that counterfactuals are presented to members as the correct basis for evaluating any policy choice, or the effects of failure. If we hadn’t had the 2008 financial crisis, or had responded more effectively to Covid, what would our current status be today? This is the correct way to think about policy but we usually don’t do it. We normally just compare ourselves with our past. Simulations should be able to easily break down how each individual would have probably fared in the counterfactual scenario. Having this information would keep people more interested in policy and ensure the watchfulness over government that is frequently missing in our current political system. And, needless to say, if we could devise a system where financial losses are “felt” by real people, it might wake us up to the necessity of building integrity into economic distribution schemes in the first place.