The Perniciousness of Infinite Loss

We're risk mitigators, not value maximizers (and it sucks that it's rational)

Mar 14, 2022

This is a companion to The Returns to Volatility. They’re a finite set (order doesn’t matter). Read both. Or neither. Doesn’t matter. We’re just glad you’re here!

There’s an asymmetry among people. Amongst? Hmmmmm.

This asymmetry [among(st)’d’ve all of the other ones] is that some people have lots of resources and skills while others have few. I’m not going to go into an exploration of the correlation vs causation/geographic & genetic lottery/systemic & structural inequality root causes of this imbalance, because many smarter people than I have written very well about it (but these linked sources are def worth reading).

Yet the imbalances exist and often these result in fundamentally different decisions in a person’s day to day life. It’s trivial to say that some people have a different perpetual option set compared to others, but I’ll point it out anyway.

If you can comfortably afford a $200 dinner, that’s one of your options for dinner tonight. If you can’t afford it, then it’s not an option. Pretty simple.

Except from my fully unscientific point of view, this extended option set doesn’t seem to map neatly to like…big life decisions.

Plainly. Lots of super resource and skill rich people I know (and many others I have researched) tend to make essentially equivalent decisions to the less resource rich people (at least on a risk adjusted basis).

Which is weird because they’re subject to the same hedonic treadmill that we all are, so they’re not fundamentally enjoying their lives any more like some utility monster.

Meaning - resource rich folks should be shooting for the moon because they’re a relative bunch so their “life scoreboard” has a higher baseline. In other words, it’s harder to win their personal game to a degree. It could just be that money or success matters less to that group, but haha ok yea can’t keep that one together we all know that’s definitely not it (plus also happiness probably keeps climbing with wealth).

I think it’s because…

Infinite Loss feels bad, man.

Infinite loss means game over.

It can be literally thought of as death in the cases of natural systems (with prey and predators and stuff). In human civilizational terms, it also can mean literal death (more on this later), but it can also mean some sort of ostracization from your core social group or family. It can mean getting canceled in the cultural sense or literally (if you are on a TV show or Twitter). It can mean failing at something critically important to you or not meeting someone’s expectations. It can also mean losing on a bet of selling a naked call (don’t laugh at that). It’s your personal worst case scenario.

For those of us with any possibility of “infinite loss”, volatility is the enemy. The more the spread of your outcomes expands the more likely you’ll end up in the infinite loss scenario (which is bad for a rational you). Only an appropriately probabilistic infinite upside could balance it out, but since most of us start life much closer to infinite loss than infinite gain this isn’t a common scenario.

Accordingly, animals that can’t ever meaningfully reduce their risk of infinite loss spend most of their time worrying about not dying. This is a fully rational (even if a little sad) strategy for these animals to play their life’s game. That skittish mouse probably should stay skittish considering it’s brief candle can go out with attacks from above, sideways, AND below (seriously - rough life right here).

What does “worrying about not dying” mean tangibly? It means the mouse is consistently making decisions that reduce the volatility of his outcomes.

New smell? Hide from it. Strange noise? Gtfo of there. Come across an intriguing thing in your path? Leave it the hell alone unless it’s the middle of the night and you’re pretty dang sure everything that can kill you is asleep or somewhere else.

The mouse forfeits the possibility of outsize returns for the approximate certainty of risk avoidance and the mouse makes this trade most of the time.

The mouse does this because he can’t sufficiently reduce his risk of infinite loss (even if he’s filled with determination).

But. Who cares? Well.

Infinite loss in an economic context can mean bankruptcy for a pure financial actor. Or it can also mean the return to “pre-success” for classes of “newly successful” like immigrants or first generation college grads, or it can even be not getting a “high prestige” job for those competing in various high status pipelines (even though it might seem ridiculous to feel like that is a loss scenario to outside observers). The concept of Infinite loss completely depends on your individual context and imo explains a lot more big decision making logic than we generally give it credit.

Let’s make it real.

How could you possibly tell your work friends (who might be your only friends after college) that you’re leaving your industry to do that weird new scam currency thing?

How could you possibly explain in a few sentences at yearly family Christmas party that you turned down your full time position with Google or Amazon to start a company? How could you do that 30 times in a row that night to receive the same fake smile, concerned look as people struggle to understand why you would willingly leave a guaranteed stable career?

How could you possibly turn down that dream job offer you’ve gone through prep school, then Ivy league checkbox 1, then banking/consulting/PE/VC/whatever, then Ivy league checkbox 2 for even just to have your dream job as an option? You’d be an idiot to let all that work go to “waste” right?

Of course, a much more objective type of infinite loss is known viscerally by anyone even remotely in the neighborhood of poverty or has a memory of poverty in their lived experience.

How could you possibly take a chance on a that startup job when you remember exactly what that single slice plain bologna sandwich tasted like growing up? Even if a single failure doesn’t put you back on that path, it might put your family there if you’re one of the many who support extended relatives as part of the responsibilities as one of the few “success stories” (in the economic sense) from your family.

This latter example feels like it puts the first few examples into perspective (and it absolutely does), but this isn’t the struggle olympics. Almost everyone resource rich and rational (sometimes mutually exclusive tbh) knows that they are in an objectively advantaged geopolitical/societal/etc etc position. The thoughtful ones won’t dispute it, but that doesn’t change the fact that their relative enjoyment of life is fundamentally dependent on their local factors and decisions they have to make daily. So in a relative sense, their feeling of infinite loss is just as real (even as some quietly try to reconcile their fortunate lot in life).

All of this dancing is just to try thread the needle that this moderately warm take is relevant to you at all stages of life (except maybe Elon who’s transcended these ideas I think).

Make it Useful

The infinite loss scenario variable dominates all other decision variables on a long timeline if you don’t find a way to deal with it.

This logic applies to anyone who can lose enough “status” to significantly alter their social standing or shift within their peer group (approximating infinite loss enough to matter). This applies to people literally close to death (from acute or longitudinal reasons). This also applies to people that have a foundational view of linking themselves to some tangible thing in the world (job) and get validation from any number of external sources (grades/money/title).

Expanding further, anyone with any meaningful chance of infinite loss in their life will avoid volatility to some, likely meaningful degree.

This isn’t inherently good or bad.

It just is.

Except some of us like living deliberately and by not realizing the spectre of Infinite Loss, we have been consequentially shaped by something we did not understand previously. Now that we understand it, we can control it. We can shape it right back.

We can also remove it from our calculus if we so choose.

So, assuming you choose to do so, how can you remove your Infinite Loss downside risk?

One (very hard) way to do it is adjust the internal definition of your worst case criteria of acceptable infinite loss. If you can somehow become okay with a more bearish bear case then you’re off to the races. Not easy or recommended but some people manage to do this. Philosophy probably helps.

The other way (still hard, but easier) to do it is to achieve some minimum viable version of the low volatility path. This gives you a baseline to fail back to in case the volatility gets too unlucky in your high volatility path. What this looks like is situationally dependent. It could be an internship or could be a few years at a “blue chip” firm. In tech, it often looks like going to Google or whatever and puttering around for a while. VCs sometimes whine about how this depletes the startup talent pool, but then they only invest in ex-FAANG founders (circle of life I guess). This is a perfectly rational choice depending on your personal definition of infinite loss. It derisks your follow on decisions (to a degree) and builds a stable platform from which to jump.

The danger with the latter technique though is when the derisking becomes the end rather than the means. Unfortunately, it’s comfy being comfy. That’s why people tend to stay comfortable once they get a taste of it. So have a plan (1 year at Google). Write it down (scribble scribble). Stick to it (quitspo).

Adjust only if circumstances seriously change. But don’t preoptimize! Just execute your plan.

Ok. Good. Your downside is capped. Now how do you maximize volatility?

Check this essay out for starters.

But ultimately it comes down to making high standard deviation decisions rather than higher short term NPV decisions (which often self select you out of the most asymmetric upside result matrices)

Remember that the goal isn’t to get the highest score (obliquely it is), it’s to create the opportunity to generate the widest range of scores and the way you’ve capped your Infinite Loss optimizes the probability distribution for the upside.

Wysr by Cameron Armstrong

Discussion about this post