rational smash and grab

when you lose trust in system solvency, there is no longer a marshmallow test to pass

a few of us were chatting about ideas like “$2000 stimmy checks” and “fiscal solvency” over on twitter the other day.

and some time worn themes were rehashed.

coyote (creator of coyote’s law) had a useful take and as i unabashedly love a good “marshmallow test” conversation, this one sucked me in.

his points here are valid in a deep way. redistribution kills growth and incentives to growth and you wind up locked in place, equal in negative-sum stasis and everything he says is certainly correct.

but then i got to thinking about why this was not quite sitting right with me and something ominous snapped into focus:

what i fear is that there is also a second operator in effect, a sort of tipping point for trust systems, one we may have crossed.

much of creating and perpetuating a high-trust, high-function system is more than just a marshmallow test demonstrating the low time preference needed to withhold immediate gratification and build useful things. this is a necessary but not sufficient condition, or perhaps, another way of looking at it is that for a low time preference society to work, there is another aspect of gratification withholding to consider:

keeping a high-trust society running essentially amounts to getting large numbers of people to all reliably not defect in a massively multiplayer iterative prisoner’s dilemma.

you keep your word, i keep mine. i don’t try to take your things, you don’t try to take mine. we respect rights and property. no one uses the state to plunder others.

this is the foundation of social contract and the golden rules that beget golden ages.

but as more and more people lose trust in the system, they begin to defect and seek to smash and grab. if you suspect that a system is likely to end or fail soon, this looks all the more attractive.

in essence, in order to see the benefit of delayed gratification or self-limitation to preserve a valued system, you have to have faith that the delay will bring benefit or that the system will persevere and survive.

lose that faith, and the whole equation changes and there is nothing irrational or short sighted about that fact.

when the mood switches to this:

all the calculus switches with it.

most people think they understand prisoner’s dilemmas, but few really do in the deep game theory sense.

it gets presented a bunch of ways, but the basic construct always looks like this:

2 players who have committed a crime are separated by police and asked to rat the other one out. if no one squeals, both get one year in prison. but if one squeals and the other does not, the squealer goes free and the one who stayed mum gets 3. if both squeal, both get 2 years.

like a social contract or a high-trust society, it’s inherently a system that runs entirely on trust. what makes it prospectively unstable is that each player can improve their situation from “cooperation” by being the only one betraying. but, if the other player suspects that you will betray them, then their only rational choice is to betray you too. to see why this is so, consider:

if you are player A and are confident that player B will betray you, then you’re really only choosing from 2 boxes: you can stay quiet and get 3 years or turn rat and get 2. even if you are wrong and player B cooperates with you and does not rat, you get 0. so you become a rational ratfink. there’s no benefit to cooperating with those you do not believe will cooperate back.

for you not to betray player A, you need high confidence that player A will also cooperate. the minute you lose it, choosing to cooperate is a losing choice for you.

thus, the upper right and lower left corners are not stable in an iterative game. the minute someone betrays, so must you.

so you can either have high trust (upper left) or low trust (lower right). those are the only stable equilibria.

and once you have lost faith in the other player’s intent to cooperate, it’s all bottom right, all the time.

the upper left quadrant becomes impossible and the very idea of it rapidly fades from your consciousness and from your rational planning.

this works just like “social contract” or “using the state to take people’s rights or property.”

we can either trust each other, pay the cost of being high-trust and not plundering one another, or you can reach for the benefits of betrayal and then trigger others to do the same which forces society to the low-trust equilibrium where everyone is worse off.

now consider the current situation:

so, for a 27 year old today who believes (almost certainly correctly) that there is no possible way that social security will still exist, much less be being paid by the time that they retire and who also knows that they are being taxed like it will be and that these taxes are not going to stop:

why not defect?

if you keep following the rules, it’s an obvious loser for you. you’re getting plundered to pay for the retirement of the richest generation in human history.

why not go “smash and grab” high time preference?

take your $2k today and run. who cares if the problem gets worse? it’s rearranging deckchairs on the titanic anyhow.

when passing a marshmallow test on government solvency amounts to failing a test of basic financial literacy, this invalidates the marshmallow test.

you’re not withholding immediate gratification for future gain, you’re just getting fleeced like your name was sean the sheep.

consider what this means for young voters:

so long as the clear fiscal insolvency of the federal government impends and you suspect that you are being made to pay in to a system from which you will never be able to take out:

why would you play?

why not jack up your time preference to “immediate” and take as much as you can from the existing system before it falls over? it’s going to do that anyway and obviously, someone will benefit from plundering it to death. so why them and not you?

see the dilemma in which we’re all becoming prisoners?

lack of faith in a solvent future means smash and grab in the present. you want to know why populism, debt write-offs, bonus checks, etc are suddenly so popular? it may not be that we’re losing rationality but rather because the young are finding it. they no longer believe the upper left quadrant is possible.

perhaps they are right.

if you see low-trust behavior and yourself getting plundered, it’s 100% rational and arguably even moral to say “to hell with this system. i see no benefit to cooperating with it. it’s stacked against me.”

in essence, they are saying “you will not keep your contract with me, why should i honor it with you?” if you want us to play in your system, you must convince me that this system serves my future and not just yours.

this makes a loss of faith in societal solvency into societal solvent.

presuming this is so, it would seem to imply that perceptions of bleak futures for the overall system make for a populist and kleptocratic present. we’re in the societal “red box” where the only choice is “how do i respond to the betrayal of others?” and there is no option to land at “everyone cooperates” anymore.

that’s how a society comes apart and cannot find its way back.

the classic game theory solution to this is called “tit for tat.” you betray me, then i betray you in the next round. then, in the round after that, i cooperate to show a willingness to move back to the upper left. and i see what you do.

if you cooperate as well, then we can rebuild trust.

but if you don’t, i stop offering to cooperate until you stop betraying me.

that’s rational play.

and so, in a system like the current entitlement and fiscal structure, this is a problem because any young person who is rudimentarily numerate can see the losses, the debt, the wall ahead. there is no marshmallow to come. so why cooperate?

this leads me to honestly wonder:

is it even possible to fix this populist/tribal smash and grab without fixing the federal deficit? is the true “giant sucking sound” our moral fiber being slurped from the society by the implacable and awful logic of the game theory effects of losing trust in systems and their capacity to generate future rewards?

is this what the final tytler stage feels like?

just losing prisoner’s dilemma after prisoner’s dilemma because trust has been lost while being asked to trust the system?

because i’m starting to think it does.

and that sounds like we’ve moved from “dependence” to “bondage.”

“a democracy cannot exist as a permanent form of government.

it can only exist until the voters discover that they can vote themselves largesse from the public treasury.

from that moment on, the majority always votes for the candidates promising the most benefits from the public treasury with the result that a democracy always collapses over loose fiscal policy, always followed by a dictatorship.”

and the only way out of this is the recovery of high-trust structures and institutions that are worthy of sacrificing today to sustain tomorrow.

from that faith comes courage and from that courage, once more, liberty.

and that’s the whole ballgame.