The intresting thing is there's no specific source of "piety" in the religiously/ philosophically divided West. Even if someone's moral failings get exposed, currently, every devil has an advocate.
The problem is complexity creep really and there is little serious effort in "code maintenance". The law itself demonstrates such problems. The Justinian Code was one example of a famously successful refactoring and ironically its creator despite his talent was also very corrupt.
One ancient solution which probably wouldn't scale well with needed complexity was "reading of the laws" - a periodic reading aloud of all laws on the books and if any were forgotten and went unobjected to they were automatically repealed as essentially irrelevant.
Just hard limiting complexity isn't a very good solution given increased complexity of the world and scale it often ammounts to "if I can't understand it ban it". It is populistic and anti-intellectual and a bit of a meme - that what can be easily understood is better and what can't is worse. Except reality doesn't work that way - the main reason why we bother with extra complexity and difficult to understand fields is because it works better.
Needless complexity may be a good concept to define abusability but defining it is tricky just like pinning down Hollywood accounting - where do you formally draw the line between overmarketing and intentionally torpedoing profit margins without proof of intent?
True now, too; notably, by way of the legal doctrine of respondeat superior.
The modern breakthrough seems to be ability to act without any direct action. Indirect communications mean that a speaker doesn't need any connection to the person whose action they're compelling, and mass communications mean that the threshold of explicitness drops from "reasonable person understands" to "someone is crazy enough to take this literally". Math-washing and automated systems make it possible to take a decision you can't explain and apply it to an unknown set of people. The result is everything from Wells Fargo fraud to legal systems that bill suspects for their own wrongful searches.
It's not a new thing, and that's worth remembering, but we seem to have gotten vastly better at it in the last few decades.
These days, we've gotten much better at directionless provocations.
'Stochastic terrorism' was coined back in 2011 to describe mass media rhetoric which produces statistically-predictable violence from an unknown source. Where Henry spoke to his court and produced a murder by his knights, someone speaking online or on television can stir violence from someone they've never met, who will quite likely be the sort of person easily written off as "disturbed" or "not behaving rationally". The specific occurrence is accurately dismissed as unpredictable, without admitting that on average such actors will be exposed to the content.
Similarly, giving an employee a target creates an obvious expectation. If you direct someone to triple their sales in a month, they might tell you "I think that's impossible to do legally." If they then do it anyway, you lack plausible deniability even if you didn't instruct them to commit a crime. But today, you can punch a sales target into HR software and assign it, with no feedback channel, to 1,000 employees you'll never meet. Some will break the law, but when the whole thing goes Wells Fargo, you had no reason to know. If employees complain up the chain, the message can disappear or decay "between" layers, with no way to show that any one party knowingly concealed anything.
This seems to happen everywhere conspiracies and fraud used to operate. The "smoke-filled rooms" choosing political nominees get replaced with common-knowledge biases across donors and superdelegates. Price-matching means cartels get replaced with 'independent' but reciprocal high prices. Targeted bribery gets replaced with generalized expectations of favorable laws, post-retirement board seats, and so on. Honestly, the more surprising thing at this point is when people do make these things explicit and get caught.
> Honestly, the more surprising thing at this point is when people do make these things explicit and get caught.
I suppose the effort involved to insulate oneself from liability is eventually seen as an unnecessary burden when one is accustom to getting away with things. I imagine a thief on his first heist who has spent a great deal of time planning and plotting only to find that it was actually quite easy to pull off, the security wasn't nearly as strong as he expected. Maybe on the next heist he is less cautious, and on the next heist even less, until he gets to the point where he's casually strolling in wearing flip-flops and sunglasses and just takes what he wants. Of course this is when he gets caught and the news paper shows this ridiculous image of an offensively cavalier thief who is supposedly a great heister. Then the public is shocked how such a dope could be a great thief (where is his mask and great stealth), but they only know the lazy thief, not the thief when he was disciplined. The lazy or stupid thieves are the ones that get caught.
When I was younger I briefly worked for Wells Fargo as part of their mortgage pipeline. It was immediately clear to me that their quality control system was deliberately designed to incentivize employee fraud in favor of getting mortgages sold.
There may be a sign-up bonus (rarely) or a golden parachute (even more rare), but the whole idea of the bonus is to be tied to the company performance.
https://www.macrotrends.net/stocks/charts/GS/goldman-sachs/n...
I don't see why this is supposed to be a problem. (Barring "require", but I'm betting you're just exaggerating there.)
Any incentive is going to encourage the incentivized party to break the law. That's the point of the law, protecting some people from what other people want to do to them.
Right now, I'm incentivized to take all your money, because more money is better than less money. And there have been people who saw this fact and decided the problem was the concept of money. But those people were and are insane.
I don’t think so. All you need is a simple quota that is too high to be achieved naturally. A common example is credit card signups. The company requires some percentage of a worker’s customers to be signed up for the company card. That percentage is too high to be achieved by honest means. Result: workers lie about the card, get people to fill out the form by saying it’s just a bonus card, or just sign people up without their knowledge.
I’d bet the same sort of thing happened here. They had a quota where a certain percentage of customers who came in for a free diagnostic had to be sold services. The actual percentage of customers who needed work done was lower than that. Result: workers fake the need for service, and executives keep their hands squeaky clean.
As I recall, Dilbert nailed this one decades ago. "Welcome to sales training. We don't ask you to act illegally, but it's pretty much the only way to reach quota. That's it for training."
It works best at large, high-turnover companies; even if only 10% of the people you hire are willing to break the law, you filter out the honest people pretty quickly. And once fraud is common enough, it becomes culturally accepted, or inescapable via rules like "fire the bottom 20%". Sometimes it even ends up taught as a "best practice" to honest hires who don't know better. Hence credit card signups, hence fake Wells Fargo accounts, hence Comcast assigning reps to upsell on a percentage of all calls, even cancellations and complaints.
Heck, this is pretty much how game dev 'crunch time' works too; it's not scheduled, so there's no one to blame, but development targets get set so that it's a predictable outcome on every project.
Barring "require", but I'm betting
you're just exaggerating there.
Imagine a delivery company where drivers are given a schedule based on average traffic conditions; and to reward hustle and punish laziness, drivers are fired if they're late too often. It's official policy that drivers should not speed. Sounds reasonable enough, right?Now imagine a driver for that company who has just been delayed by worse-than-average traffic. Not their fault, a crash caused a tailback. They can speed to get back on schedule, otherwise they'll be fired.
Would you consider it an exaggeration to say the driver had been required to break the law?
UPS set delivery targets for drivers that were too high to consistently achieve. When drivers fell behind, they sped and caused accidents, so UPS installed speed monitoring. Drivers switched to leaving the rear door open between nearby stops, so UPS installed door sensors. Drivers started leaving seatbelts unbuckled, so those got sensors.
UPS essentially frames this as fighting Goodhart's Law with brute force, trying to track so many metrics that they can't be gamed. Except, of course, that the targets aren't realistic. What actually happens are results like drivers sitting on buckled seatbelts, moving heavy packages in injury-causing ways, or marking "signature required" packages as no-shows without knocking to shave off time. None of that is permitted, UPS fires people who are caught doing it, but the rest of the time when it isn't caught, it's all convenient cost-savings.
There's nothing wrong with holding individuals accountable for their response to incentives, but it's possible to do that while also holding people accountable for the incentives they create.
My understanding is that UPS has partially bypassed this by setting per-route targets which update independently, but within a multi-driver route it could still apply. FedEx dodged it by franchising so that slow workers outright went into debt to the company, but then got sued on employees-as-contractors grounds. Overall, though, I think it's a huge and underrated effect limiting Taylorism. Designing strictly around efficient workers isn't viable if you can't find a lot of them, and using multiple workflows is rarely cost-effective.
There's a story in Rivethead about a car-assembly stamping job that was assigned to two people splitting tasks, but could be done by one person working very smoothly. New guys would do it in pairs, but experienced workers could each do a half-shift solo and duck out to the bar for the other half. As far as I know, time-motion studies never put an end to that, because it was too scarce a skill to schedule and organize for. (Of course, the inability to squeeze out more productivity has consequences for automation, but that's another story.)