With the rise of work-from-anywhere arrangements, corporations are finding new ways to monitor their employees during work hours. But in some industries, worker surveillance has already been the norm for sometime. Since 2018, truck drivers have been forced to maintain an Electronic Logging Device (ELD) in their cabs. While the devices were originally intended to simply monitor hours of operation, they’re increasingly tracking a range of things, such as driver eye movements. To talk more about driver surveillance, we spoke to Cornell professor Karen Levy -- the author of the book Data Driven — to understand more about this phenomenon. The transcript has been lightly edited for clarity.
Key insights from the pod:
Why electronic logging devices exist — 3:25
How ELDs became mandatory — 6:05
What ELDs can do — 7:20
Are ELDs even making the roads safer? — 13:25
How to cheat the ELD — 21:39
What data collection means for worker power — 25:34
The future of workplace surveillance — 33:14
---
Tracy Alloway: (00:10)
Hello and welcome to another episode of the Odd Lots podcast. I'm Tracy Alloway.
Joe Weisenthal: (00:14)
And I'm Joe Weisenthal.
Tracy: (00:16)
Joe, do you remember when we had Gord Magill on to talk about trucking?
Joe: (00:20)
Yeah, that's an iconic, I mean it was recent, but that's an Odd Lots classic already! And a bunch of people have reached out, that was a great episode.
Tracy: (00:29)
It was. And he said, I mean, he said a bunch of things that were very interesting, but he said something in particular where he was talking about an upcoming book by a Cornell University professor called “Data Driven.” And the idea was to explore the role of technology in the trucking industry.
Joe: (00:49)
Yeah, absolutely. And you know, I think this is a really important area to explore because of course, as we've discussed with Gord — but also other episodes that we've done on trucking — truckers are monitored increasingly. They use these ELDs — electronic logging devices — that track hours, things like that. And you know, obviously this affects truckers today, but you know, we may all be tracked by ELDs as workers eventually.
Tracy: (01:14)
Right. So there's two interesting things here. So one, you would expect that monitor monitoring technology, like the ELDs that they're using would be sort of anathema to a lot of the truckers. Like a lot of truckers go into it thinking they're going to be independent, they have the freedom of the road, and then it turns out that their eye movements and brainwaves are going to be tracked and monitored. So that's interesting. And then secondly, given the shift towards work from home for a wide variety of workers it does seem like surveillance technology in general could become more of a thing.
Joe: (01:51)
Well, and the other really important thing that I think maybe says something profound about how business works or capitalism works is this core problem Like, why does ELD exist in the first place? It attempts to solve the problem of truckers on the road for too long. That creates fatigue. That creates accidents. That’s the impulse behind these rules that curtail how long someone could be on the road. But rather than address why are they on the road so long? Why is there trucker fatigue? Why has fatigue been part of trucking since forever rather than figuring out different ways to do that? It’s like, let's just come up with a monitor system rather than addressing the core issue that creates the need for these time constraints and monitoring systems in the first place.
Tracy: (02:34)
Right. So the question is also whether or not technology is the right solution to the problem it's trying to solve. Okay. Well I am very happy to say we have the perfect guest, because we are gonna be speaking with the author of the book that was referenced by Gord. We are going to be speaking with Karen Levy. She is an associate professor at Cornell University and also the author of “ Data-Driven: Truckers Technology and the New Workplace Surveillance,” which just just came out. So Karen, thank you so much for coming on Odd Lots.
Karen Levy: (03:03)
Thanks so much for having me.
Tracy: (03:05)
This is a really interesting topic and I'm glad we have the chance to dive into it further. Maybe just to begin with, you know, Joe kind of alluded to this in the intro, but what exactly is the problem that this workplace surveillance technology that's been mandatory in the trucking industry, I think since 2017, what exactly is it trying to solve?
Karen: (03:25)
Yeah, I mean, the way you guys talked about it kind of in the lead up, I feel perfectly encapsulates it. So the ELD, the Electronic Logging Device as you mentioned, is federally mandated in the trucking industry. All long-haul truckers have to buy and install and use them. And ostensibly the goal is related to safety, right?
As you alluded to, we know that truckers are really tired. We know they're incredibly overworked and under slept. And obviously that’s high stakes, right? There are thousands of truck related accidents a year and fatalities from those accidents. It costs billions of dollars every year. Nobody wants that, right? Including truckers. Nobody wants the roads to be unsafe. I don't want to be next to a tired trucker on the road. We're all aligned on that. But what's interesting as you point out is the way that we have sort of chosen to address that in this case is via technology.
So the ELD is sort of the centerpiece of this effort to make truckers comply with the timekeeping regulations that they're subject to. I think this is something Gord talked about too on the episode. Truckers since the thirties have been bound to work no more than a certain number of hours every day and every week. And that's for safety reasons.
The problem, as y'all have discussed, you know, on Gord's show and on other shows is that the way truckers are paid does not align with those incentives, right? They're paid by the mile. Truckers have the saying, if the wheel ain't turning, you ain't earning, right? Like they're only paid for the time that they're actually moving down the highway, not to like sit around resting, right? Or not to do all the other things that are an important essential part of being a trucker but aren't compensated.
So the result is that truckers just want to stay on the road as much as they can because that's how they make a living. They're in pretty dire economic straits to begin with. It's not a mystery why they end up doing these things, right? And so the ELD is sort of seen as like an answer to this, right? A way to sort of police this behavior by making it ostensibly harder, although not impossible, for truckers to tamper with the log books that they used to keep. So they used to do this using paper and pencil. And that is a really easy system to kind of falsify or make it look like you're running legal when you're not. And so the ELD is sort of seen as an answer to that.
Joe: (05:29)
I always was a big fan of the song “ Convoy” and there's a line about tearing up your swindle sheets, which I didn't really realize the origin of that term until I read your book. And the swindle sheets was a nickname for the old physical pieces of paper that the truckers used to log their hours. Talk about, you know, they had all different ways, I guess, of sort of swindling. Lying on these cards. How long had that effort been in place to get truckers to give up the physical pieces of paper in favor of these ELDs?
Karen: (06:03)
So ELDs first start popping up in, you know, discussions around regulation in the 1980s and there's a long history of kind of how they eventually become mandatory. At first the government starts to do things like, ‘Well, you know, maybe we'll make these mandatory for people who have like really bad safety records, right?’ for some subset of the industry that we kind of think of as like habitual violators. Then there's a proposed mandate in the 2010s that goes nowhere, right? There are legal challenges, it's kind of a long and rocky road to get to the ELD mandate that we eventually end up with in 2017. But we do eventually get there.
And, you know, a lot of the kind of the research that's informed my book is trying to understand that transition, looking at the arguments that get made on either side of the ELD mandate, you know, from the 1980s and even before. There's some predecessor technologies too that, you know, I think set the stage for this up to, you know, post-mandate looking at what the effects have been on the industry.
Tracy: (06:55)
So we definitely want to dive into those various arguments. But before we do, can you talk a little bit about the technology that goes into ELDs. Because my understanding is they sort of range from pretty basic models to more sophisticated things that actually, you know, can measure your brain waves to see how awake you are or how much attention you're paying to the road. So talk to us about the variety of ELDs that we have right now.
Karen: (07:20)
Yeah, this is really crucial. So the ELD itself captures some important data, but as you say, sometimes maybe not that much, like a fairly minimal amount basically — where the truck is and how long it's been being driven, right? Those are basically the core requirements of the regulations.
However, it's almost as though the government told everybody like, you have to buy a phone that makes phone calls or something, right? You can. But it's very hard to find a phone that only makes phone calls, right? If you're going to buy a phone that makes phone calls, you're also going to buy a phone that takes photos and connects to the internet and all this other stuff, right? Because the tech tends to bundle these things together. And the same thing happens in ELDs. So ELDs are actually more commonly like a module in what's sometimes called a fleet management system, or FMS, people sometimes call 'em like a Qualcomm, because Qualcomm used to have a big market share in this area.
There are all kinds of names for them. But now what people think of as the ELD actually is kind of like, you know, it's almost like naming the part for the whole, like it sort of stands in for this much broader range of data collection technologies. And as you point out Tracy, those can include things, you know, it commonly includes stuff like two-way messaging or alerting a back office about your fuel use or how hard you're braking or how fast you're going or, you know, whether you're changing lanes without signaling — all kinds of like pretty fine grain driving behavior data, which is just like very different from what trucking work has looked like for a long time, right? Like, if you ask folks why they get into trucking, many of them say explicitly it's because they don't want that, right? They don't want someone looking over their shoulder kind of measuring how they do their work.
And you know, some of the stuff that's coming up on the horizon more recently includes the types of stuff you allude to, right? Things like driver-facing cameras have become much more common. Sometimes those are augmented by artificial intelligence that will do things like try to assess how fatigued a driver is. And they might do that by doing things like measuring whether the driver’s eyelids are fluttering, you know, when you start to fall asleep, your eyelids start to flutter, your head nods. They can measure those kinds of things and infer that the driver is tired. There are wearable devices that sometimes get sort of integrated into these systems or used kind of in a complementary way to these systems that do things like you said, that measure a driver's brain waves, measure a driver's heart rate, you know, collecting these other kind of biometric signals to alert the back office or to alert, you know, the driver's safety manager or whoever, [about] kind of what state they're in.
Joe: (09:44)
So technically the law just requires the bare minimum of like, okay, something tracking the truck, the hours on the road, etc., that needed an electronic version of the log book. But it's like this Trojan Trojan horse. Once you get that electronic device in the truck itself, then why not add all these other things? Where is the impulse coming from to add these other features onto the ELD? Who's benefiting from this and who's making the decision that we want to stick a camera in the face of the trucker and see their face at all time and start counting eyelids opening and closing.
Karen: (10:24)
I think that's exactly right. It's like a scaffold or a Christmas tree and once you have the Christmas tree, you can just hang more ornaments on the Christmas tree because you’ve got the Christmas tree now. So what has happened is that yes, the government has started to say, we need to collect a certain set of information digitally.
And then a lot of the impetus for more data collection comes not necessarily from the government per se, but from tracking companies, right? Because those workplace analytics, as you alluded to in the setup to the conversation, workplaces across all kinds of industries and all kinds of professions are increasingly interested in using software or sensors or other technologies that are cheaper and easier to come by to measure stuff about what their workers are doing. And so in some sense, trucking companies are doing what all kinds of employers are doing, which is measuring more information about what their workers are doing. And so in some sense, trucking companies are doing what all kinds of employers are doing, which is measuring more information about what their workers are doing. It actually doesn't end there though, right?
On top of that there are third parties that become very interested in this data too, right? So insurance companies become interested in it. Companies that sell parking space reservations to truckers, other stuff like that, they become interested in it. So the data becomes very valuable to many different parties. It's just not valuable to the trucker, right? The trucker is kind of isolated against all these parties that are interested in gathering this data about what he's doing.
Tracy: (11:47)
Can you talk a little bit more about the insurance angle? Because this is sort of a recent theme for us. This idea of insurers incentivizing different types of behavior, collecting more granular data in order to do that. Does the installation of ELDs, does that actually, you know, reduce insurance rates for instance for truck drivers or truck driving companies?
Karen: (12:11)
So in what I have seen, insurers I think at this point are mostly just interested in getting the data rather than really doing much with it beyond just like getting the data and integrating it into their modeling. So what I've seen is insurance companies offering what they sometimes call a ‘plugin discount.’
This is not so different from what Progressive does with commercial passenger vehicles, right? Where you get some discount on your premium for just installing the thing or just giving data access. I haven't yet seen that it's actually had a measurable impact on actual rates within the industry. But you can imagine this is the goal, right? Is to do something like that.
Joe: (12:53)
So I mean we have had this mandate in effect a few for a few years now. Is there evidence that it's making the road safer?
Karen: (13:02)
No, there's not. There's not evidence that they’re making the road safe. Thank you for that set up. No. So what we have seen is that in the few years after the ELD mandate took effect, truck crash fatalities went up. So they hit a 30-year high the year after the mandate. Crash rates in some segments of the industry, they've gone up. This is not based on my analysis, this is based on like quantitative analysis by some business school professors that I talk about in the book. But yeah, if the goal is safety, there is no demonstration that even on its own terms the ELD is succeeding at making the road safer.
And if you think about why that is there are a couple of reasons that may be happening. So one of them that's like really intuitive, is if you tell people like, ‘Hey, you know, you need to get from point A to point B in 11 hours’ — about, roughly. ‘Let me try to get home for Christmas in 11 hours,’ I will do that. But if it's 11 hours and five minutes or if it's 12 hours, it's not the end of the world. And if I need to stop and get a cup of coffee or if I need to pull over and see if my tire looks weird or something, I can do those things right? I'm not going to drive like a bat out of hell to do it because I know that that flexibility is sort of built into the system.
Now if I tell somebody you have exactly 11 hours and you have no more because I'm tracking this, I'm monitoring you, people will drive very differently, right? And that's very intuitive. So what we have seen is that rates of speeding and reckless driving have gone up because people feel this extra rigidity in the rules. And so — not shockingly — they compensate for this lost productivity by trying to make up for that in the way that they drive. So that's one clear mechanism.
Another thing that we don't have as clear data on this, but it clearly seems to be happening based on my conversations with drivers. You know, trucking is kind of an aging workforce anyway. The median trucker age is, I think in the late forties. And a lot of folks, if you talk to them, they are not interested in this, right? Veteran drivers who have been driving professionally for maybe decades, millions of miles without accidents, you tell them ‘Guess what we're putting in the truck now?’
They're not going to stick around for that. Trucking has this very high turnover rate anyway. People churn in and out of the industry all the time. And those are the drivers that are the safest. You want those drivers who are the most resistant in many ways to this kind of oversight. That's the guy you want to be next to on the road. You don't wanna be next to the 18-year-old who just got their CDL who maybe doesn't know any different, right? Like never grew up trucking any other way. It's hard to find a blue collar job that doesn't involve a lot of oversight and managerial surveillance. So those are the folks that will stomach this kind of thing, but they're not the safest drivers.
Tracy: (15:38)
Actually this reminds me, I wanted to ask you, you know, you studied this issue, I think when we spoke to Gord, he mentioned a a 10-year study or something like that. I'm not sure if that timeframe is entirely accurate, but talk to us about how you actually went about gathering data and anecdotes for your work.
Karen: (15:56)
Yeah, so it has been... Gord is right. It has taken me a very long time to process this study, but that's been good. I've spent like a quarter of my life thinking about trucking at this point, which, you know, life comes at you fast. But I started this study in 2011. I was a grad student, I was studying sociology and I was a lawyer before that. So I was really interested in kind of thinking about law and how does it function on the ground and how do people respond to it?
And especially I was really interested in kind of law and technology and what happens when we surveil people. What happens when we collect a bunch of information about what people are doing to assess whether they're breaking the rules. And I was looking around for like, where's a place where I can see this transition in action? I can really see that happening on the ground so I can get a good sense for like how that unfolds or how it changes the way people relate to one another.
And just by chance — it was completely a fluke — I heard a story on the radio about trucking and how truckers were upset because they were dealing with this, this was again 2011. So the ELD mandate was being discussed, although it wasn't in effect for several years after that, but about this mandate. And I thought, well, that's maybe it. And I didn't know any truckers, like I have no trucker truckers in my family, but I went that day to a truck stop just to see what is it like to talk to a truck driver? Like, is it easy? Will they talk to me? And I was instantaneously hooked. I found that truckers had such interesting stories, they were so forthcoming, they were incredibly generous, you know, in talking about the stuff that I admit I had never thought about.
Tracy: (17:28)
It sounds like the origins of Odd Lots’ interest in trucking as well. ..
Joe: (17:33)
We keep doing trucking episode, obviously. It’s a fascinating angle. You know, you mentioned the legal background. And this brings up something, a point that I thought was really interesting in your book, which is that society kind of depends on people being able to break the law a little bit. You can't actually run a society on everyone always strictly hewing to the law and being punished if they don't. Can you explain that a little bit more? Because I think that's key when you talk about how with an ELD you can't be 10 minutes late. You can't drive an extra 10 minutes. But what is this concept of like, we need a little bit of loose areas?
Karen: (18:11)
I think this is exactly right, right? It's very easy and intuitive to say like, ‘Well if we have a law, people should follow the law.’ And therefore, you know, if we assume that the law is legitimate or has been reached in a democratic way or whatever, people should just follow it. But that doesn't really hold up too much scrutiny, right?
And there's tons of places where we see that actually social life would kind of fall apart if people actually followed laws to the letter. You know, one of the examples I use in the book is just speeding, right? I think if you were told that you were, if you got a ticket for driving 66 in a 65, not as a trucker, but just as a regular car driver, people would be like, ‘what?’ That's not the law and, like, it is technically, but nobody really expects that that's the law.
Another thing I talk about in the book that kind of hits this home is in labor actions and, you know, collective action. A pretty common worker strategy is the ‘work to rule’ labor action. I don't know if this is something that's familiar to Odd Lots listeners, but in work to rule what you do, it's like a work slowdown and you achieve it by like suddenly you really pay attention to all the rules that are in the rule books right?
Tracy: (19:19)
Right, your contract spells out that you work this amount of hours every week and you work exactly that amount of hours.
Karen: (19:25)
And you just do it right? And the whole point is that by following the rules that is not actually what we necessarily want people to do. Because if you do it, that's the slowdown, right? The reason it's effective is because that's clearly not the expectation most of the time. And this happens just all over the place.
A few years ago in New York City, and in many cities, there are these big moves like these Vision Zero kind of projects where they're like, ‘we should have no jaywalking because jaywalking leads to traffic accidents.’ And then when police start really cracking down on jaywalking, there's some economic analysis where people say, you know, actually jaywalking is really efficient and we would lose like a good deal of economic efficiency if people actually follow these rules because we actually want people to bend the rules a lot of the time, right?
We've built a society in which following a bunch of rules to the letter can be really inefficient or ineffective or can cause other harms. And that's definitely the case in trucking, right? Clearly a lot of problems with the status quo in trucking, I'm not suggesting that everything is great in trucking and the labor conditions are wonderful and if we got rid of the ELDs everything would be fine. I don't view it that way, but it is definitely the case that we have all sort of come to rely on the rules being more like guidelines in this context, right? And that people fudge them in order to move stuff at the pace at which we kind of all demand it. So when we crack down on a rule without kind of considering that broader context, you end up with this weird situation where you haven't addressed the underlying reasons people are breaking the rule, you're just kind of policing them harder for it.
Tracy: (20:56)
You know, Joe mentioned the swindle books earlier, like the the analog log books that truck drivers used to keep. And I remember I used to fly planes when I was like nine years old.
Joe: (21:09)
I didn't know that you flew.
Tracy: (21:09)
Well because my dad's a pilot.
Joe: (21:11)
No, I knew that, but I didn't know you were a pilot.
Tracy: (21:12)
Well, I'm not, but I had a pilot's log book and I was supposed to put it like my hours in my tiny little Cessna, but I was nine years old and so I just just used to pretend to fill it out all the time.
Joe: (21:24)
I didn't know anything about this.
Tracy: (21:25)
But so my question is, moving from the analog logbooks to the ELDs, how infallible actually is that system? Can that information be tampered with or falsified?
Karen: (21:39)
Yeah, it is a great question. And I love that you yourself have falsified a log book. So the whole idea, the rhetoric around the ELD is this is a tamper-proof version of the thing that people have been doing for a long time. And certainly it is more difficult, right? Or it presents different kinds of challenges to falsify data in an ELD than it would using pencil and paper where like anybody could do it with five minutes of thought. But it's not impossible, right? As you point out. In one of the chapters of the book I go through all of these different strategies that truckers and other folks use to kind of still bend the rules sometimes, right? Because as I mentioned, you still sometimes need that or like the economy depends on that or they're expected to do it by their managers or whatever it is.
One of the things people will do is, and they did this with paper logs too, what's sometimes called a ghost log. So you sign in as Karen, right? If you're Karen and you do your 11 hours and then you sign in as Joe or as Tracy, right? Using some other [name]. And then you have 11 more hours, right? And sometimes, some companies, they often get like a demo account, like a John Doe kind of account when they sign up with an ELD vendor. And so they like have these kind of like slush [accounts], you know, drivers that they can use in some cases. Not everybody is doing this, but that's a strategy that was described to me. Other things folks will do, I was sitting with a safety or with a dispatcher at a trucking company and she was talking to this driver and the driver was like, ‘well, I'm like four miles from where I need to be, but I'm out of hours, what do I do?’ And she was like, ‘well, pull over and then roll to the place you have to get to. Like, roll to the drop off point at less than 15 miles an hour.’ And the reason was because the ELDs all have a threshold, which is usually set by the company and that threshold is like under 15 miles an hour it won't register as driving, right? So she knew that it wouldn't register as driving.
What's interesting in both of these examples actually, is that it's not necessarily the trucker that's pushing this, right? It's oftentimes the company that's telling the driver ‘Here's what you need to do,’ right? It’s kind of a ‘winky’ strategy because they kind of want it both ways in some sense, right? They want the control that ELDs afford them, but they also just want the stuff to move. So sometimes they kind of coerce or compel drivers to do these things.
Joe: (24:13)
So, you know, one of the things — and I think this is a thing that's going to matter for all workers, especially workers who work from home, we know that that’s going to increase the amount of surveillance on workers. Like, are you actually at your desk responding to customers, etc., whatever the job. But I think one thing is that all this like data collection is going to affect everyone, is sort of this idea of ‘who knows about how work is done? Where is the knowledge stored about how to do a good job?’
And so in the case of trucking, at one point the idea is like, okay, the truck drivers, they really know their area, they know the road, they know the warehouses that they go to and there's some sort knowledge base that exists primarily in the head of the truck driver that is difficult to replicate. And then I imagine, you know, okay, now with ELDs, maybe a lot of that knowledge isn't at the level of the driver, but maybe at whatever — the fleet owner or some other entity. But I also have to imagine that even for the owner of the trucking fleet, they may be losing to some third party company that's building a black box AI model that even the company can't access. And so the locus of all this knowledge and you know, the power that one has with that knowledge, gets sort of distributed.
Can you talk a little bit about the shifting distribution of power and knowledge within the industry as these manufacturers of ELD equipment, third party data providers, software companies that provide the databases, become like, ‘oh we have the database, have the information?’
Karen: (25:43)
Yeah, I think that's exactly right. The information moves right? In trucking, people used to talk about it like you're the captain of your ship. You get to make all the calls because you're there, nobody else is there and you know what the conditions are like and nobody else does because nobody else is there.
But as you point out now more of that data is distributed to the firm or to a third party. And what's important to recognize about that, I think, is not only does that change the game because somebody else has the data, but the data they have is not quite the same as the data you have, right? They have some kind of abstracted guess about what things are like. They know how many hours you registered on your ELD and they're going to kind of assess whether you have the right to be tired, right? Or whether you have the right to stop or they have some sense of what the weather is like because they can look at the Weather Channel, but that's not the same as being in the place and knowing what it's like.
So as things get abstracted, a lot of times those kind of more abstract data sources get used to challenge the accounts of workers on the ground because they're measurable, right? They're the kinds of things we can measure. And this happens again, not just in trucking, but all over the place. So I have a study from a different context where I like looked at retail workers and kind of how technology is affecting them and there too, retail workers in some settings do this practice that's called client telling where you like keep a book of business and you know the sizes of your favorite customers and stuff and you know when their birthdays are and that really is empowering.
Some stores, those workers, they work on commission and they have a really good relationship to certain customers. Increasingly they're not allowed to keep their own books of business. They have to input that data into some centralized system. And what that means is that when it comes time for them to argue for a raise or something like they don't have those bargaining chips because as you said, the data's not in their head. It's now owned by the company or it's owned by the software vendor or whatever. And so they're more substitutable, right? It reduces their bargaining power vis-a-vis the company. And I think you see this kind of thing happening just across the board really in a lot of different labor contexts.
Tracy: (27:51)
So just on this topic, and this is kind of a Devil's Advocate question, but our last trucking episode we were talking a lot about how truck drivers end up working more hours than they're actually paid for because they end up going to warehouses or depots or whatever and they have to wait hours and hours and hours to pick up or drop off a load.
If you have this kind of surveillance technology, could it not be used to measure, for instance, how much time you're idling and then maybe used to avoid inefficient warehouses or ameliorate some of those problems?
Karen: (28:28)
Yeah, definitely. I definitely think that can be. I don't think it redeems the whole thing, but it is a silver lining, right? And exactly those kinds of tools have been built using ELD data that kind of give you, it's almost like the thing where Google will tell you ‘is there a long wait at this restaurant?’ It'll do the same thing for truckers being like, you know, is there a long detention period on average at a particular shipper? And that potentially can be useful to you, right? To the extent you get to control whether you know which loads you're going to take to help you kind of control your business.
So I agree that that kind of transparency can be useful, right? And it's one example of where, you know, it's kind of a more labor friendly way of using this data to actually help truckers out. I don't think it redeems the whole because it's a nice thing to kind of put on top of all of this additional control. There would be other ways we could ensure that truckers don't lose a bunch of time to detention that would be much more direct. But I'm in agreement with you that that is like kind of a bright spot.
Joe: (29:24)
It also seemed like, look, we just had these two years of terrible supply chain disruptions. Why didn't all this data help us then? I mean maybe because this is a once in a hundred year crisis we just experienced, but you figure like, well if all this data was so great at improving supply chains, well it didn't exactly step up to the plate.
God, there's so many fascinating angles of this. Something else in your book that really struck me, you mention in many of the cases the truckers you really want on the road are the ones who might be most repelled by having this electronic monitoring device staring them in the face all day. Can you talk a little bit about these efforts that the ELD makers and companies have put forth to sort of like get people comfortable with them?
Karen: (30:12)
Yeah. So one thing that companies sometimes will do, and I've talked to ELD vendors that have talked about these strategies and to trucking companies themselves that will use them, is that like in rolling out an ELD system or a fleet management system, you know, a lot does depend on kind of the messaging around that or what kind of culture the company builds around that. And there's a bunch of obviously the tools can be used in all kinds of different ways and there's a bunch of variation in how companies roll them out.
One kind of interesting thing I saw happening is that, you know, because there's now all this data on, like there are scorecards that can be generated for each driver based on fuel efficiency and how many safety incidents they have and that kind of thing. That becomes really easy. And then that also facilitates like not only measuring the driver but comparing the driver against other workers, right? Or incentivizing the driver kind of gamifying it, right? Making it a game about like, ‘well, who can have the lowest fuel efficiency? That's a big one, right? Because that's a big cost driver for trucking firms.
And so companies will do all kinds of stuff to sort of incentivize drivers to sort of comply, you know. Like a really basic thing is just putting up a list of who's the most fuel efficient driver, which they of course now have that data, like putting that up, you know, in the central office or something. There's some pictures in the book of some of these, what these charts look like. They might attach a small financial incentive to it. So I talked to a company, they drove for restaurants and they're like ‘Oh, well, we'll give you a gift certificate at the restaurant that you drive to,’ which doesn't cost them very much, but I guess is nice for the driver.
My favorite one is where a company — they were interested in the drivers driving efficiently — and so what they decided to do was issue these like small bonuses to drivers’ wives in the wives’ names when their husbands met whatever the fuel efficiency goal was. And the thinking... I asked like, ‘well, why the wives?’ And once you get the wife onboard, it’s very clever.
Tracy: (32:08)
That is so insidious.
Joe: (32:10)
Yeah. ‘Honey, I'm so proud of your ELD -compliant fuel-efficient driving.’ That’s what people want to hear when they get home.
Karen: (32:21)
I know. That's what I say to my husband every night. You have to sort of admire the ingenuity, but it is an interesting strategy, right? Exactly like you said, to sort of bring the family dynamic into this kind of stratification of the work.
Tracy: (32:36)
Well Karen, one of the reasons we we wanted to have you on is, I mean we are fascinated by the trucking industry in general, but we mentioned this in the intro, as the world sort of shifts to work from home, or at least as there's a little bit more work from home than there was before the pandemic, it does feel like there is an opportunity for more of this kind of workplace surveillance technology to be deployed to a broader workforce. Can you talk to us a little bit about what your research tells us about the application of this tech to the rest of the labor market?
Karen: (33:13)
Yeah, I definitely agree that workplace surveillance has a very long history, right? And the motivations behind it are not new. So companies have always wanted more productivity, more efficiency, and watching your workers to kind of get at that is something that, you know, dates back to at least the industrial revolution, probably before that, right? This is not a new strategy in some ways, but it is changing in some important respects, right?
And one of those that I talk about in the book is, well first just that you can integrate these things into new workplaces. So truck cabs for a long time were pretty immune to the kinds of workplace oversight that you saw in warehouses or in factories or even in offices, right? Just because of the nature of the work. And in some ways they're like a lagging indicator, right? They're catching up with some other types of blue collar work.
But there are other things I think about, kind of what the trucking case tells us. And as you mentioned, post-pandemic, there's a lot more concern from managers about workers kind of not giving it their all or shirking, or not paying attention or doing those kinds of things.
And so these types of data, like data tracking, whether that's location tracking where your workers are or detecting what they're looking at on a screen when they're supposed to be working or what windows they have open. You see the stuff in school settings too, like the same kind of like proctor software that you sometimes see for taking remote exams, right? Functionally is not so different from some of this workplace technology tracking, like productivity monitoring.
More and more backend workplace systems like Microsoft 365 include these capabilities that will report back to a manager about how many minutes you spend in PowerPoint or whatever, stuff like that. So this kind of measuring work, this is just happening all over the place and I really don't see that going away anytime soon.
One of the tricks here, one of the problems is that when you measure stuff you inherently lose a lot of context. And oftentimes the things you can measure, like how many emails somebody sent or whatever is not an amazing indicator of whether they're doing like meaningful work. And so workers a lot of times, you know, still have to do the meaningful work, but they also are sort of burdened with making themselves legible to these systems that are sort of incentivizing or gamifying these things that may or may not be all that meaningful.
So I do think that those dynamics are just popping up across a lot of blue collar work, but also what we think of as kind of more white collar professionalized industries like medicine and law. There was this really great article in the New York Times a few months ago that maybe folks saw, by Jodi Kantor and Arya Sundaram, where they talk about this kind of productivity monitoring that's happening and one of the examples that they use is of a hospice chaplain who gets some certain number of points from her employer based on how many visits she makes to, like, dying people. I mean it's just super, super dystopian, but it's like, they talk about a range of different industries and it's clear that this is just sort of coming for everybody in its own, in its own way.
Tracy: (36:15)
Well, I look forward to our dystopian future where Joe and I are just sending each other two word emails constantly in order to look more productive.
Karen: (36:23)
Gaming the algorithm.
Joe: (36:25)
Prepping for a podcast.
Tracy: (36:26)
Yeah. All right. Karen, we're going to have to leave it there. Thank you so much for coming on Odd Lots and talking about your work. Absolutely fascinating.
Karen: (36:35)
Oh, it's my pleasure. Thanks for having me.
Tracy: (36:49)
Joe, that was interesting.
Joe: (36:51)
That was so interesting. There were so many threads. I feel like we could have talked to Karen for a long time, so many different threads about how business works. Where power is. The purpose of laws and rules. So many interesting things.
Tracy: (37:06)
Yeah, I do think that point about how we have rules in place, but actually flexibility to work outside of the rules is really important, both because it kind of lubricates just what we're doing — like living in New York would be really, really difficult if we actually never jaywalked for instance. But also here's an element of power to it. And you mentioned this, but if you have a manager in a workplace and they say like, ‘oh, you know, don't worry about this deadline. Or you're supposed to work from nine to five, but actually I don't mind if you come in slightly late.’ That's also a bargaining chip for workplace relationships.
Joe: (37:45)
Absolutely. And then you think, okay, all these businesses love the idea of having this software that allows us to track our workers and track our workers productivity. But then the software maker itself becomes a source of power because then they see the equivalent data for 50 other companies that they service and then they can aggregate that data and then they know things about how your employees benchmark against employees from other companies that you as a business owner don't have. And then they sell you things on top of that. So I think that as more jobs are sort of under this surveillance, it seems like it's going to be this sort of like fascinating power dynamic [of] who has the information about how stuff really works.
Tracy: (38:25)
Yeah. And also the thorny issue of measuring productivity and quality and this is something, not to get really navel gaze-y here, but in the media industry, this is something that's been going on for years where, you know, we can look at pure traffic numbers and say like, this article generated a lot of traffic, but you can compare it with something that maybe didn't get as much traffic and say, well this is objectively a better article. This one has more nuance and more detail, but maybe it wasn't as popular. That's always been a sort of issue in journalism.
Joe: (38:55)
But Tracy, you and I, we've always been good at traffic.
Tracy: (38:59)
We do both.
Joe: (39:00)
I've always liked those rules because usually I’m pretty good at them.
Tracy: (39:03)
We do quality and and quantity here at Odd Lots. But the gamification of a workplace algorithm is an interesting one.
Joe: (39:11)
Tons of stuff to talk about. That was fascinating.
Tracy: (39:13)
Shall we leave it there?
Joe: (39:21)
Let's leave it there.
You can follow Karen at @karen_ec_levy.