> But they find that the second effect predominates: “Working from home results in an absolute reduction of 14.7 percentage points in the annualized probability of an alert (misconduct report) per trading employee.”

Well, then either WFH decreases risk of committing crimes... or they still happen just as much and it increases the chance they'll go undetected? A bit worrying that TFA doesn't discuss this, though I haven't read through to the the study itself.

It could just reduce work, and crime as a second order effect, as crime still takes work. But I would theorise that it could decrease crime because it increases the likelihood of getting caught. Discussing a criminal conspiracy is much easier to do in person rather than over Teams, or any other communication software that leaves a record.

I mean while we're speculating we might as well throw in the other possibilities - like working from home might make you happier, resent your job less and less likely to try to fuck over your employer. In reality we don't know, but I don't think we need to assume it's due to lower productivity or increased sneakiness.

Trader crimes are not so much about screwing the employer. It's often aligned with the incentives of the employer i.e. make as much money as possible. The crime is more in that they use unlawful methods (or take on too much risk) to earn money for their employer and thus themselves via bonuses (and status, promotions etc).

> Trader crimes are not so much about screwing the employer. It's often aligned with the incentives of the employer i.e. make as much money as possible.

This really isn't the case, at least for large banks.

Bank boards and CEOs would much rather have lower trading performance than get pulled up by regulators. It's very difficult for a bank CEO to get fired for being to cautious and regulator friendly. Mediocre financial results can always be excused in earrings calls with de-risking or social responsibility and few shareholders are interested in the political fallout of publicly disagreeing with that.

Or that instead of sitting at your computer for another 4 hours, once you're done with your work, you're done. Yes, you could warm the seat for another couple of hours once you've finished your actual work. You could do that. On the other hand, Tears of the Kingdom just came out.

Maybe it feels guiltier to commit crimes within one’s home, even perhaps around one’s children or family while working at home (or even just photos and other stimuli that evoke emotions and memories of family or even just self.)

> or they still happen just as much and it increases the chance they'll go undetected?

What you said doesn't relate to this from the OP. First up, the article should at least address this question before we can consider your supposition.

For that to explain the gap I think you would have to consider how many of those employees who get written up live with their families vs more detached living arrangements. With that consideration, it seems unlikely that change alone could make up 15% for the overall group.

Indeed, hence the word “maybe”

maybe they are also less productive working from home - they get less done overall, and less crimes.

That was my guess. A 14% reduction in efficiency or perhaps just hours worked sounds about right.

it's almost like WFH helps insulate employees from toxic cultures. Making otherwise reasonable people feel less pressure to be unreasonable.

Sure you can but there's less immediate pressure from management and peers to do stupid and or criminal things. You can't just leave a meeting room without getting fired. You can totally take ten minutes to think of a response and an excuse for the delay on slack. Also... People are less inclined to ask people to commit crimes via text then they are via voice in their place of business. You can imagine why.

The magnitude of intensity you get on Slack is 100x less that what you'd experience face to face.

You can, but that's a paper trail. It's harder to push people into doing something criminal over email and slack without being self incriminating

> they still happen just as much and it increases the chance they'll go undetected

I would assume it’s a decrease in criminality. Compliance-caught and self-reported violations eclipse tattle-told ones. (Questioning the stability of this ratio through WFH is valid.) Being at home reduces the risk of the last. But it drastically increases the risk of the first, given the lack of unmonitorable live conversations.

I suppose one could measure markets for signs of statistically-detectable fudging? This would only cover a small set of trading violations, however.

Or..the kind of trader that typically commits fraud/misconduct prefers working from their office station/trading desk?

"If you are around your naughty friends all day, their naughtiness will rub off on you; some number of people who wouldn’t independently do financial crimes will do those crimes due to peer pressure"

I believe that's the TLDR of the article and I fully agree. If you put a lot of competitive people into a highly competitive environment and some of them brag about gaining an edge by skirting the rules a bit, everyone else will also start pushing boundaries just to remain on the same rank on the competitive ladder.

It's a classic Red Queen's Race: https://en.wikipedia.org/wiki/Red_Queen%27s_race

Isn't it possible that financial crimes require discrete collaboration to avoid detection and a WFH situation prevents that from happening?

Or conversely that WFH makes financial crimes easier to commit and harder to detect so we're seeing a drop in detected incidents but the real numbers haven't changed?

Or that the there's another variable related to both that's causing it (like increased other vices that the article speculates)?

Or that this is just random chance and financial crimes were going to decrease during this time period no matter what?

I don't see any of those possibilities discussed in the underlying study. Though they do add several important disclaimers like all this data coming from a single firm in the UK, that they can't rule out trade volume as a driver (fewer trades means fewer crimes), and that WFH is still a relative novelty so behavioral patterns observed now might not be the same as those observed long term.

It's definitely an interesting finding but I wish they had addressed other causes more robustly instead of speculating that WFH was the cause per se.

The discrete collaboration thing seems likely to me. Evidence in insider trading cases not infrequently involves some way for Bob who knows about X to tell Jim, who trades something related to X before X is public. If Bob and Jim are in the same office building this can happen very casually, and it's extremely deniable whereas if Bob has to make a Teams call to Jim, now that's traceable and they might both be called to explain it.

Or perhaps the regulators and compliance officers meant to catch this stuff are less efficient out of the office?

Could be many things at this point. Way too early to call.

Plus, if negative behavior rubs off on you - wouldn't positive behavior? And if negative behavior rubs off on you better in the office vs. WFH, isn't that invalidating much of the argument most have for WFH?

Interesting crime is considered more efficient in-office but legit business isn't. And if the argument is culture I think it proves the point even more so.

I actually don't think WFH vs. in-office has much to do with this personally, but I find how the narratives change quite interesting.

It could not even be peer pressure, but just employees observing "hey, they can get away with this, if they can so can I"

Red Queen is an underrated analogy for so many things

While of course the archive trick is great, for anyone who really liked this article you can sign up for Matt Levine's newsletter and get these in your inbox many times per week.

https://twitter.com/matt_levine/status/994296126055608320?la...

Used to be on a bulge bracket trading desk. This happens a lot. People do stupid things for money. Especially French derivative traders.

I also bet there is less harassment (sexual or otherwise) at home.

There is a lot of behavior that can be hidden in in-person interactions that leaves a trail if you try to do it remotely.

Well I guess because they are not just doing crimes. They're doing organized crimes. Organized crimes required effective coordination, and WFH curbs exactly that.

Sounds like a questionable methodology to me.

In my experience traders generally aren't allowed to work from home in investment banks (even all throughout the pandemic) with which I am familiar, except for a handful. That handful being the head of a desk etc who in fact does little day to day trading and has a more managerial job.

Many people also WFH who are technically traders are in sales and doing very little day to day trading.

People, as usual, reading tea leaves.

I wonder if sales teams makes fewer sales at home ;)

I have walked down a street named Wall in NYC but I've never been to Wall St and I know nothing firsthand about the culture there. I have, however, participated in very petty childhood and teenage crimes with other troubled boys, was a buyer of controlled substances for decades, and have been a gigging musician for periods of time, so I've been around criminality.

One thing about every crime that involved more than one perpetrator that I know about is, there is a period of conspiring beforehand, which happens - must happen - away from the eyes of others. I encountered many weed dealers over the years who just used their cellphones as convenient appointment books/ledgers for the police, and I tended to not go back to them, cause they'll be out of business soon anyways.

If you put a bunch of people controlling large sums of money in a single floor of an office building, it will be very easy to gather in a bathroom stall and plan the execution of any sort of crime over a few rails. They're basically being forced into it, because the environment itself guarantees obscene and ubiquitous graft.

If they necessarily must communicate over digital means, they still can do it, there's plenty of ways to get encryption that works well enough to evade the pigs. But how many of these guys are actually smart enough to implement that? Most of them are going to assume that people smarter than them are keeping an eye. I connect to a remote desktop for work, and I take it for granted that they have my laptop's camera on when I'm connected. Because the light is on. Bet theirs is on too. It is probably more terrifying to them than Big Brother to Winston Smith. Any eye is terrifying to a criminal, and these fellows, they are still criminally-minded, they are just denied the crucial quiet, dark spaces in which to plan, so they are finding ways to stay happy with merely a million or two per year, with maybe a trip once or twice a year to plan a big score.

If the USA really wanted to fix this, they would put every one of these people in a room by themselves and never let them talk to each other unless they're working for the same client I suppose. From home or at the office doesn't matter, but they should have twenty microphones around them at any moment they are working, each one monitored by a committed Marxist who would just love to hand them over to The State for Processing.

> Any eye is terrifying to a criminal

I find this to be an interesting comment because there is supposedly an effect where simply an image of an eye being present tends to cause people to behave better.

https://en.wikipedia.org/wiki/Watching-eye_effect

So, I guess the test would be to retrofit all dark collaboration corners of trading firms with pictures of eyes and see what impact that has on the detection of crime.

Yah, let's try some new art in the bathrooms of Wall Street and see if that fixes anything.

edit: makes about as much sense as the idea that a rock star is gonna solve shit.