Better Decisions in Anxious Times

COVID-19
Internal
The Pilot
Uncategorized
Pilot

By Jin Kang, Lydie Neill, and Steve Weber

The Problem

Making good decisions is hard.  Doing so day after day, in an increasingly fast-paced environment with too much information, too many stakeholders, too much constant scrutiny from inside and outside, and too much volatility is hard

But good decision-making is at the root of good strategy, communications, and stakeholder engagement. You can’t execute on decisions you haven’t made, and no one benefits from excellent execution on a bad decision. Recovering from unforced decision errors is just getting harder as the stakeholder environment becomes more fraught.

We’re writing this with a deep sense of empathy for our own team and for our clients. No one is immune, and we see the challenges of quality decision-making in our own daily lives as individuals and team members just as much as we see it in our work with clients. 

There are tons of useful advice and insights waiting to be put to use, from decision theorists, organization theorists, behavioral economists, and others. Think about the library of books on leadership or popular behavioral science, from Kahnemann to Tetlock to the classic study of Groupthink and many more. But we sense something new, different, and even more challenging in today’s decision environment which goes beyond what the current science and advice tries to grapple with. 

What’s new is a profound level of decision anxiety, and the challenges associated with what we call ‘decision-making under anxiety’. We’ll explain what that means, why it’s another level of hard, and what we’re learning from our practice that can be done to manage and improve it. But first, some context.

Defining uncertainty vs. risk in decision-making

The most important distinction between decision theory and practice is the difference between risk and uncertainty. It’s simple in principle but often tricky to navigate in practice.

Decision-making under riskwhen you have a decent model of the situation you are trying to impact or change, but you can’t put precise numbers on at least some important parameters in the model.  

  • Imagine a hotel in the Hudson River Valley that has a robust model of room demand for the month of September based on many years of data. One of the parameters in that model is probably ‘rainfall’.  You know the average rainfall over the last 50 years, and you have a good model of how rainfall impacts demand for rooms, but you don’t know how much rain will fall this September. So, you are making decisions under ‘risk’.
  • Risk is distinctive in that it can be priced and hedged. That history of rainfall data can be used to put a price on an insurance contract that ‘shares the risk’ of this particular September being really dry (upside risk) or really wet (downside risk). In practice, not all risks are hedge-able or hedged, but in principle they can be. Figuring out how to price risk is basically the history of financial market innovation in a nutshell.
  • Much of behavioral economics is about helping people make better decisions under risk. For example, prospect theory explains how people tend to be overly risk-averse in the domain of gains and overly risk-acceptant in the domain of losses. Put simply, when people are winning, they tend to buy too much insurance or overpay for it. When they are losing, they tend to do the opposite. Once you recognize that, you can develop decision processes that compensate.

Now consider decision-making under uncertaintywhen you don’t have a reliable model of the situation to start from, the situation you are trying to impact is truly unprecedented, and there’s no history or even strong theory to guide you. Or, when you do have a model of the world, but it turns out to be completely wrong (what Wall Street, confusingly, called ‘model risk’ in 2008-2009).

  • The beginning of the pandemic is a good recent example. No one had a decent model of how modern economies would function under stay-at-home orders and quarantines on a nearly global scale. That’s why no one had really hedged or bought insurance against such a possibility. No insurance company would have had any idea how to price such a contract. 
  • Uncertainty, as opposed to risk, cannot be priced or hedged. The Hudson River Valley hotel wouldn’t have a model of the impact of earthquakes on room demand in New York State. Demand might respond like it has in California, but it could easily be completely different since New York consumers have no past experience of earthquakes to draw on either. Should that fleetingly rare event come to pass, decisions under uncertainty would be unavoidable.

The uncertainty-risk distinction has real implications for decision-making

Risk can be hedged, though at a price. That’s why ‘sensitivity analysis’ is so valuable for decision-making. A classic sensitivity analysis:  The Hudson River Valley hotel asks itself: what happens to our room demand if rainfall is 20% above or below normal, and should we do anything about that? It might then offer a partial rebate offer for customers if it rains a lot, that keeps demand higher and maxes out revenue. It might swap part of its expected revenue with a sister hotel in Phoenix whose demand is impacted by hot temperatures (assuming hot temperatures in Phoenix are uncorrelated with wet months in upstate NY). It might buy rain insurance from a third party that has internalized those kinds of swaps.

Uncertainty typically can’t be hedged in the same way: no one would know how to price the hedge. To make better decisions under uncertainty, one approach is to ‘wind-tunnel’ decisions against multiple models of the situation you might be operating in, keeping in mind that you don’t know which one is ‘right’ or what may actually happen. This is exactly the situation in which Breakwater Strategy deploys ‘scenario thinking.’ Scenarios are multiple, diverse models of the world, and are not merely +/- variations on a single model. A robust decision is one that can win no matter which of these models turns out to be closest to truth and/or help you diagnose which of the scenario models is in fact closest to truth. But it’s a much more customized analysis that takes greater effort and as a result is typically reserved for high stakes decisions.

The Rising Challenge: Decision-Making Under Anxiety

In our recent look at the landscape of American political economy in 2022-2024, we highlighted anxiety as a newly important condition that shapes decision-making because of the multiple cumulating pressures converging at once in the post-pandemic economic and social-political maelstrom we’re navigating. We said in that context:

It’s long been known that capitalist economies are propelled by the sentiments of greed and fear. While that will remain true, in these scenarios we see a new dynamic in which anxiety – which is different from fear insofar as it is rooted in uncertainty about what may happen, rather than the dread of what is happening – becomes a factor impacting everything from employee productivity to financial markets. There may not be an antidote for this feeling, but organizations need to call it out and recognize its emotional and cognitive impact on decision makers and on the quality and coherence of decision processes.

We’ve observed over the last year or so that decision makers seem to experience anxiety differently than they do risk and uncertainty because anxiety is first and foremost an emotional state, more than it is a cognitive constraint. It’s about how you feel, and that occurs prior to how you process information – the primary challenge for risk and uncertainty.

This matters, we think, because most organizations are not as good at dealing with emotional (‘hot’) decision challenges as they are in dealing with cognitive (‘cold’) ones. The same is true for many individuals.

Consider some of the differences that most of us have experienced personally at one time or another: 

  • Anxiety is as much about the body as it is about the mind. Decision makers operating under anxiety have an aura of diffuse emotional activation, tension, and strain. It’s physically taxing and exhausting: anxiety causes heart rate and respiration to escalate, increased cortisol levels in the bloodstream, and disruptions in sleep patterns (people are literally kept up at night worrying about X or Y). We haven’t measured and proven it, but we suspect that most people, when operating under anxiety, will be meaningfully limited in their ability to make decisions in a set time frame.
  • Anxiety can be contagious within decision-making groups. That’s not a new observation: Who hasn’t sat in a meeting with anxious colleagues, and felt the anxious energy spread no matter how hard you try to hold it at a distance? Contagious anxiety can become self-reinforcing among teams and individuals, which means decision-making gets harder the more you do it, rather than easier with practice. The opposite of what we’d hope for and expect. 
  • Anxious decision makers tend to over-interpret noisy data flows. Anxiety seems to predispose people to pay too much attention to short-term signals and lose sight of trends and tendencies. It’s like reacting to today’s weather when you really want to make sense of the climate. This can take the form of over-indexing on a single bad news story or a single comment, even a single tweet thread, any of which in a broader context might be just an outlier and unimportant. We’ve also seen a tendency to careen from one such small signal to another way too quickly in a kind of reactive ping-pong mode. If you’ve been in a decision setting like that, you know what we mean when we say this can be not only dysfunctional for strategy but exhausting and dispiriting as well. 
  • Anxious decision makers seem to be overly subject to FOMO and mirror imaging. The way we’ve seen this manifest is in decision makers who seem to spend far too much time and energy watching what their competitors and reference groups are doing, with the underlying rationale that ‘they must know something we don’t’ about this crazy world we’re operating in, and so we should just copy what they do as a ‘least bad’ choice. Disguising this as a ‘best practice’ doesn’t make it any more functional. There’s certainly a place and time for assessing best practices but defaulting to a fast-follow of competitors to ease anxiety is rarely going to be a good idea.
  • Anxious decision makers tend to polarize people between flip-flopping and rigidity. Some anxious decision makers seem to flip-flop as the ‘weather’ changes and really struggle to stay with a course of action for long. Others, conversely, seem to become overly rigid about decisions they’ve made and refuse to listen to new incoming evidence which might suggest a change is warranted. We understand that re-opening a ‘closed’ decision can increase ambient anxiety and thus there’s a natural desire to avoid it — but senior decision makers are responsible for carrying exactly that burden. We don’t know which decision makers tend to fall into one or the other of these categories (flip-flop vs. rigid) yet, but we do know one thing for sure: a decision-making group that has a mix of both is going to quickly become scattered and dysfunctional.

To be clear: these observations are only that, not scientific hypotheses (at least not yet) that we’ve tested in any manner. But in naming them, we think that many decision makers and groups will nod their heads and recognize just a little bit more clearly that these tendencies exist in their own experience and often get in the way.  So, that’s a starting place, even in the absence of fully tested findings. 

What can be done about it?

Anxiety starts and is held deeply in the old reptilian part of the brain, which means that we shouldn’t overestimate the ability of the modern cerebral cortex to use rational thought and command the rest of the brain to simply turn the anxiety off. That doesn’t work in corporate decision-making any more than it does for individual people. But, just as with individual experience, the simple act of noticing — when the cortex can observe that the deep state of anxiety is present, put a name to it, and acknowledge that it is in fact happening right now, rather than just being caught up in it — can make a difference. 

So, in those situations, the first thing to try is to take a pause, watch, listen, and note the experience.

  • Ask yourself: do I feel anxious with this decision in front of me?  How do I know I’m anxious – for example, is it my stomach tightening up, my hands clenched, or something like that? Is this group feeling a mounting anxiety as we discuss the decision?  What signals am I seeing that lead me to think so — perhaps people are talking faster and faster, interrupting each other in ways they usually don’t, bouncing their legs up and down, cutting off valuable lines of inquiry as if they’re just trying to relieve the pressure, and so on. In online settings, we sometimes see this situation arise in the form of the awkward protracted silence on a Zoom call, where everyone is looking to someone else to provide some semblance of certainty to the table — when it just isn’t possible. 
  • It may sound trivial, but we’ve seen firsthand the value of a leader at exactly this kind of moment calling anxiety out for what it is in simple language and asking everyone on the team to take a step back and just be aware that anxiety is affecting their individual and group reasoning. Not a panacea by any means, but noticing and naming is almost always a surprisingly effective first step.

A second thing to try is a ‘climate time-out’.

  • Have the group take a half step back and ask of our own discussion and decision process that we are in: how much time and effort are we expending in talk about today’s weather, and how much on the climate? Different circumstances call for different mixes of course — and there are certainly crisis moments when the weather means everything, and the climate just has to get pushed off till later. The point is to make sure the decision-making group knows where it is at the moment, and assesses that it is where it wants and needs to be. You might be surprised at how often that is not the case, and how simply calling the question can work as a partial re-set.  

A third thing to try: do a thought experiment that blows up the best practice fallback. 

  • Pose the alternative hypothesis: what if our competitors and our reference groups know even less than we do, and are actually making really bad decisions right now?  What could we do to truly stand out in that context? What does our organization really stand for that is distinctive and what could we do to place that at the forefront of this difficult decision? These defining moments when collective indecision is plaguing an entire industry present valuable opportunities for organizations that are not afraid to challenge the status quo and present an alternative narrative or path forward that explains it. 

Modest Improvements Can Make a Difference

At the end of the day, organizations are susceptible to many of the same emotional biases that hinder individuals in quality decision-making. It’s not surprising that we’ve seen these challenges show up vividly in 2022, because the landscape we’re all operating on has given us plenty of reasons to be anxious. We’ve tried here to offer a few observations about what that means and what impacts it can have on decision-making, along with some modest experiments that we think can tilt the table back in favor of better decisions. 

There’s no silver bullet here, but there’s also never been a silver bullet for the cognitive biases that are the subject of so much decision hygiene research and advice — just human-scale interventions that try to make us function a little bit better that we otherwise would. When it comes to improving decision processes, crawl before you walk before you run is a reasonable objective. As with any skill, practice will lead to gradual but meaningful improvement.