Decision Making under Uncertainty – Are there rules?
Dr. Anthony Fauci recently spoke to NPR about his first meeting with the new president of the United States, Joe Biden:
“[President] Biden…reiterated something he had already told us. He said, ‘Right now, everything we do is going to be based on science and truth, and if things go wrong and we make mistakes, we admit them, and we try and fix them.’ I said hallelujah, you know? (laughter).”
This reminded me of a much earlier statement by one of Biden’s predecessors. Franklin Roosevelt, addressing the 1932 graduating class of Oglethorpe University in Atlanta, in the depth of the Great Depression, told them:
“The country needs, and, unless I mistake its temper, the country demands, bold, persistent experimentation. It is common sense to take a method and try it: If it fails, admit it frankly and try another. But above all, try something.”
Lately Americans have been accustomed to being told by their candidates and their leaders that whatever policy they are advocating will definitely work, while those of their political opponents must fail. This approach is simplistic, doomed to failure, and unworthy of a free, intelligent, mature people.
Many recent policies of our government, hailed by supporters as perfect solutions at the time of their implementation, have gone on either to partial or complete failure. In fact, we would say that since every major policy initiative of any size – any “bold experiment” – must by its nature involve tradeoffs, every significant initiative must be compounded of failures and successes, or, more accurately, costs and benefits. Moreover, those costs and benefits are never shared equally by every citizen. Politics involves choice. As is often said, “Elections have consequences.”
Throw in the undeniable fact that every major initiative, even if executed more or less as intended, always seems to abide by the Law of Unintended Consequences, and one can see that major policies are subject to a huge amount of uncertainty.
In short, as FDR implied, we live in a world of uncertainty, and every bold policy entails risk, and risk implies the possibility of failure. But the best American presidential administrations have always been in the business of bold experimentation. Therefore, while there are no certainties in life, we can guarantee that some of the things this administration tries will not work as intended, and will have to be altered if they are to achieve positive results.
Hence, while every policy any administration adopts must be given adequate time to succeed, if, after that time, it does not succeed, a mature and competent administration should “admit it frankly and try another.” Too often lately, politics has been seen as a blood sport where any admission of error is seen as politically unacceptable, and any change of policy is branded as hypocritical “flip-flopping.”
As scenario consultants dealing with decision making under uncertainty, we would like to propose a different approach, whether you are a political leader, a business leader, or any other kind of decision-maker. Tell your constituents that you assume that not every initiative you take will work perfectly as designed, and that you plan to learn from your inevitable mistakes, and change course. Proudly build “hypocrisy” and “flip-flopping” consciously into policymaking. As ex-UK Prime Minister John Major said, paraphrasing Albert Einstein, “the politician who never made a mistake never made a decision.”
Some laws already contain “sunset provisions” designed to shut down the initiative after a certain time period. Perhaps every piece of legislation should instead contain a “review provision” laying out the criteria by which the law would either be shut down, continued, or perhaps even expanded.
In any case, people who are saluting the Biden shift to “Everything we do is going to be based on science” should recall that the scientific method is based on trial and ERROR. If you are doing big things, you will almost certainly make some big errors.
It is possible that much recent impatience with government may have resulted from “experts” pretending to a certainty that they could never possess. Government has been seen as overpromising and underdelivering, or else pretending that policies do not involve unevenly distributed costs and benefits. Even a policy that provides huge net benefits to society as a whole may impose huge direct costs on a very large number of citizens.
Experts tend to favor policies that provide net benefits to society, but especially those that provide very large benefits to people who look a lot like the experts. Large costs that may be imposed on people who do NOT look like the experts are likely to be minimized by those same experts. But those distant, unfamiliar people have a way of making their feelings known at the ballot box, to the consternation of the “experts,” who had assumed that their values, e.g. efficiency, environmentalism, the raising up of previously disadvantaged, or the like, either were or should be universally shared.
In the current harsh political climate, any perceived failure of a policy is going to be treated harshly by the opposition, even if its net benefits greatly outweigh the costs. Maybe the best way to begin to change that climate is by openly declaring that your administration is in the “bold experimentation” business, and that you expect to make mistakes and to change course at times. We’ve tried it the other way; there does not seem to be much to lose.
Further overpromising and failures of democratic government could put us in a position like one which John Maynard Keynes described after yet another failed international financial conference in 1933:
“It is now evident that there was no cat in the bag, no rabbits in the hat—no brains in the head…. The fiasco of the Conference merely increases the general cynicism and the lack of respect toward those in power. This growing lack of respect is, as recent examples elsewhere have shown, one of the most serious things which can befall a democracy.”
Or any organization operating under uncertain conditions.