I am not defending Shipt and there is no doubt gig workers are in a very vulnerable position. However, the data analysis results as presented in the article do not support the article's main point. "40% are getting paid at least 10% less" is not unnatural to expect whenever pay is redistributed, especially since some 30+% are getting at least 10% more. Imagine a _hypothetical_ situation where Shipt is 100% on point and driving a fairer version of the algorithm patch removing a way for workers to "optimize" for short, well paid trips, resulting in pay cuts to those who had learnt how to do it, while not changing/increasing pay for everyone else. We would see the same kind of result: some portion of workers would get paid 10% less, some 10% more. This does show that workers are paid differently for the same work they have been doing, but does not prove the change is unfair.
I had similar thoughts. But let's not overlook the information asymmetry, which contributed to the dissatisfaction. I don't want to live in a world which is controlled unilaterally, and intransparently by a group of people who assume they have a full picture of the situation and assume they understand moral completely, and also don't think it necessary to explain how they think so highly of themselves.
It's an interesting question, as we have a spectrum from 'little to no transparency' through to 'full transparency' (which is pretty rare), and in the middle sits the usual approach of 'communications-team-led messaged quasi-transparency'. Difficult to know (without more info) where Shipt would have appeared on this spectrum, but given the issue, they're probably somewhere towards the 'insufficient transparency' end.
What's silly in this case is that (as others have pointed out) the new algorithm seems to have been reasonably equitable, with a genuine redistribution of payments, rather than just a cut overall. Shipt could have avoided this whole situation with a straightforward explanation of the changes, together with a few examples of the cases/jobs in which people would earn more or less.
I think the issue is that there was full transparency on pay (a fixed base rate plus a fixed percentage) and then it was changed without warning.
I work for a salary, which is fully transparent in the sense that I know what my next paycheck will be to the penny. (It’s not transparent in how it’s set, but it is week-to-week.) If my employer started paying me based on effort, and didn’t tell me what constituted effort, not only would I be pissed off but that would be completely illegal.
I’m not suggesting that this change is or should be illegal. But if it happened to me I’d find it extremely unfair.
If Shipt is actually trying to incentivize better performance, it seems the best way is to be completely transparent about the rewards algorithm. "Short high-value trips are now somewhat de-rated, and trips requiring more effort now have improved rewards, specifically ..." or whatever.
This "communications team" approach did everyone a disservice if Shipt mgt were really trying to improve results.
OTOH, if the actual goal was to screw workers harder, they accomplished that, as here ate arguments on HN about how this could be good for the workers, thus successfully obfuscating the goal of screw-the-workers.
Although as far as that graph reflects the the study’s results, the new distribution looks almost perfectly balanced: even more people experienced a “10%+” bump than a 10%+ reduction post-update.
By what mechanism do you suggest the worker-screwing is happening here?
even if total compensation was decreasing, the results for the company can be improved by being able to provide their services at a lower price point by cutting costs.
cutting costs is not "screwing workers". cutting costs is key to acting in a competitive market.
> I don't want to live in a world which is controlled unilaterally, and intransparently by a group of people who assume they have a full picture of the situation and assume they understand moral completely
I have thought about this topic for a while at the time that I worked with Law data (e.g. Family Law and Military Law), and I just came to the conclusion that several societal institutions and it's agents are inherently intransparent, even in situations where some "illusionist transparency" (there's transparency, but the magician deviates your attention to another side) is given (e.g. judiciary system, under-the-table political agreements, etc.).
That's one of the reasons I would like to have a more algorithmic society with human in the loop calling the final shots and placing the rationale on top. An algorithm will have human and institutional biases but in some sort, you can explain part of it and fine-tune it; a human making the final call on top of a given option would need to explain and rationally explain its decision. At best a human actor will use logic and make the right call, at worst it will transparently expose the biases of the individual.
<< That's one of the reasons I would like to have a more algorithmic society with human in the loop calling the final shots and placing the rationale on top. An algorithm will have human and institutional biases but in some sort, you can explain part of it and fine-tune it; a human making the final call on top of a given option would need to explain and rationally explain its decision. At best a human actor will use logic and make the right call, at worst it will transparently expose the biases of the individual.
I will admit that it is an interesting idea. I am not sure it would work well as a lot of the power ( and pressure to adjust as needed ) suddenly would move to the fine-tuning portion of the process to ensure human at the top can approve 'right' decisions. I am going to get my coffee now.
> a human making the final call on top of a given option would need to explain and rationally explain its decision.
To who? What you describe does not seem much different than the representation governments most of us here are accustomed to, other than the algorithm eases some day-to-day work required of the constituents. Already nobody cares, and no doubt would care even less if they could let an algorithm let them be even less involved.
I think that's a very fair point, but wouldn't that be true even if Shipt hadn't made any changes?
It feels to me like the problem wasn't the change. For all we know, the change was a net good thing. The bad thing was the context in which the change occurred.
>I don't want to live in a world which is controlled unilaterally, and intransparently by a group of people who assume they have a full picture of the situation and assume they understand moral completely, and also don't think it necessary to explain how they think so highly of themselves.
We already do; uncertainty is fundamental at all levels.
That statement is wrong either way. Looking at the graph, ~22% got a >10% cut while ~36% got a >10% raise. Overall ~43% got a cut while ~57% got a raise. And if there’s any doubt, later on they dropped the “at least 10% less” qualifier for the 40% figure in text:
> But we felt that it was important to shine a light on those 40 percent of workers who had gotten an unannounced pay cut through a black box transition.
Can’t believe they (accidentally? intentionally?) screwed up the very first concrete figure given in the article. Guess what, discussion is now based on the wrong figure.
Also, the y-axis of the plot is labeled “number of workers” when it should be “percentage of participating workers”, unless they had exactly 100 participants (they say they had 200+). Lousy presentation.
I’m all for transparency, obviously.
Edit: In addition, since participation is entirely voluntary, common sense tells me that the data they gather should skew more negative, since people negatively affected are much more likely to participate.
it seems clear to me from first principles and from experience [0] that the result from voluntary compensation sharing skews low compared to the actual population. it's the people who want to complain about low wages who participate in such schemes.
that's not to say sharing wages is bad. this can still provide upward pressure on said wages. but the people who participate in the beginning are likely from the lower end of the distribution.
[0] websites like levels.fyi seem to consistently skew low when estimating higher percentiles
To clarify, I was basing the "some 30%" on the wage change distribution histogram that comes somewhat further down in the article from the statement you quote.
> Mostly gray area because gig corps don't like actually treating the workers as contractors, they just like the lowered costs.
I don't follow this? Is this predicated on the fact that gig corps can choose not to work with contractors that don't meet their criteria? If so, how is this different from only using lumber yards that consistently meet your expectations?
The implication of the parent poster seems to be that there are legal requirements regarding contractors and legal requirements regarding employees but gig corps would prefer to treat their workers as one or the other class depending on which is to their benefit - which would be against the law because of the aforementioned concept "legal requirements".
Exactly this - there are differences in what you can require from someone on employment contract and external contracting company (whether that company is single person or not) and effectively making one category into another without actually reclassifying (like employing as employee) is considered fraud in most places.
There are possible frauds from both employer and employee side, but I will list some common "landmines" of miss-classification, though beware that they are picked across different jurisdictions and I do not remember which apply where. All examples are possible items that can be decided to be part of misclassification, usually from contractor to effective employee:
- requiring specific dress code is non-enforceable on contractors in many places
- contractor is not required to provide specific person to fulfill the job, only a person of appropriate qualifications (it's valid for there to be a check on those qualifications)
- in UK case, contractor might be asked to prove that they have a substitute to work in their place!
- [Poland, possibly other] having only one client is not illegal, but can be grounds for investigations and if it's your only client where you work for equivalent of full-time job, it will be evidence for tax fraud
- You can not enforce working hours on contractors in most jurisdiction, only specific deliverables (taking part of work meetings is deliverable, requiring availability in general of specific person at specific times can be grounds for reclassification)
- above is often linked with "gig economy" - rules regarding "contractors" needing to pick up available jobs etc. are often considered illegal skirting of employment law.
As sibling comment mentioned, more is available from your local (too) friendly search engine. And employment lawyers and HR specialists.
In this case would it matter if they were contractors or not?
There is a way of calculating the pay for a job. It is predictable. Publish the algorithm. Want to change it? Great. Update the documentation and then publish that. The workers should be able to calculate exactly what they are owed. They can decide to leave or stay.
Only in America are people deflecting by bringing up the employment status of people when the issue is a lack of transparency designed to allow wage theft.
You’re arguing that it’s sometimes legitimate to hide payment information. So I can hire someone for a job, and only I know what the job is worth. The worker just has to try it out and see if their valuation of the job is the same as my valuation?
The fact that some, maybe most, are making more money is irrelevant. The organisation can change the value of a particular job at any time and simply say “the algorithm made me do it”.
I don't think the reasons you are listing are valid at all.
1. If they can game it, they should. If that's a problem, then Shipt should fix their algorithm.
2. It's fine to data wrangle your way to an automated model for e.g. your company's growth projection or for predicting where you can best expand to find more customers. It's not okay to use it to unilaterally change pay agreements with your workers or "contractors".
3. The right of an individual to know what they're getting paid for their work outweighs any company's nebulous claim of its algorithm being a trade secret or something vague like that. Can you imagine businnesses actually operating like this? "You can't know what we'll pay you for this job, its ~~seeeecret~~. Just trust us."
It's amazing that anyone would even try to say that.
From now on I'm going to pay my rent according to a secret algorithm of my own devising and my landlord will just have to hope I'm generous this month. I can't have him gaming the lease agreement by only providing the things it says!
Not one of those reasons constitutes a valid excuse. Holy cow.
"Surely in a hotly competitive market" there is no need to hypothesise about what "surely" would happen according to some wishful thinking. We know from countless past observation exactly what happens in any market of low skill low investment labor, wages go straight to the level of the most desperate willing to be slaves and make nothing at all, because there always are enough of those for everyone else to take advantage of, and having no money they have no power to demand better. They can't afford lawyers and ad campaigns and lobbyists and politicians, and they can't afford to strike, and that same wonderful magic market means there is no one better to work for, all employers are essentially the same. Unions and strikes do exist, some places where they are allowed to, and it's always a big news story and a miracle when they actually accomplish even 10% of what they needed to once in a blue moon.
Maybe not cut in stone, but if the data are derived from the workers’ payout receipts, it would seem likely that this was the amount that was paid out by Shipt. Or do you mean that third parties might be skimming something off after the company pays out?
I mean, it wasn't "a clear case of wage theft" as these weren't wage workers; and in any case it's not like they weren't told ahead of time how much the order would pay.
A clear case of adverse selection in the old pricing model.
From a game-theoretic perspective in a gig marketplace you don’t want jobs that are strictly better, else sophisticated market participants (workers) will select the best ones leaving chaff - and a worse experience - for the less sophisticated participants.
What you are looking for is preference optionality, eg one Uber driver might prefer not to do very long trips, another might prefer it, and you ideally get paid fairly for either.
In this case as others have noted, it doesn’t actually sound like an unfair change. Perhaps communications could have been better though.
We are entering an era where corporations have perfect data. They can charge each customer exactly the maximum amount possible, and pay each worker the exact minimum amount possible
Its a marketplace with strict rate fixing. This is ironic since the libertarians who start these companies are hysterical reactionaries any time the government tries to do the same thing to make prices more fair for the public.
Central planning would be if a committee decided how many rides there would be each year, regardless of demand.
No, you ideally want all the jobs to be good. With the old model it was clear what the floor for getting something delivered was. Target could have instead adjusted order minimums or shrunk delivery zones instead.
Preference optionality is widely stated to be one of the features that gig workers like about the arrangement.
The options you suggest are also valid ways of homogenizing the jobs to reduce variance.
> Gamifying peoples livelihood is the problem
To be clear Game Theory applies to all economic interactions. Mechanism Design is the branch of Game Theory pertaining to market design to achieve desired outcomes, such as “avoid adverse selection in my gig work marketplace”.
Gamification is a specific application of video game design to economic interactions, it’s unrelated to what I’m discussing. (Examples of Gamification would be gaining experience points and levels for delivery, daily checking rewards, achievement badges, etc. - the general goal in Gamification is setting up a dopamine loop to encourage repeat use of the app. Hopefully it’s clear this is not what I was talking about.)
applied Game Theory in real human history a.k.a. wealth-building, has shown that the biggest empires with the most wealth and the best armies are built with slavery. So slavery did win, again and again and again. I don't think most modern people have any idea how deep and wide the history of slavery is ..
Game-theory is fun when you get good at it for designing markets and products, but let us not lose sight of the crucial discussion.. human beings with real lives are not equal to economic parts.
I used to fly into Pittsburg and take an Uber to my grandparents house 40 miles away. For years I could always get an Uber but then after the pandemic my trip was rejected so frequently that I had to start renting a car. So basically no one wanted the gig and Uber never told me my ride was unreasonable. Something changed in the algorithm to benefit drivers, Uber, or both, at the expense of the customer, and that info wasn’t made clear to the users.
I’m curious: would you have felt better if Uber had just rejected your ride request upfront (“Sorry, we can’t offer rides that far” or something)?
Or is this attitude of “hey, it’s a long shot, but let’s give it a try and see if anybody takes the job” closer to the attitude you’d like to see? If the latter, how would you communicate that to the users?
I had an occasion a long time ago where I needed to request an Uber for a ride similar to what you’re describing. At that time, apparently the driver didn’t find out the route until they’d committed to the ride. The guy swiped to say he’d picked me up, and more or less broke down in tears when he found out where I was needing to go. He lived 40 miles in the other direction, was going off shift, and would be driving the whole 90 miles home without any prayer of a passenger to cover the time or cost. In that case I ended up giving the guy a generous amount of cash to cover the imposition, but I couldn’t bring myself to use Uber for that route in the future.
Until recently, when I had to use Uber for that route again. This time it seemed like they’d gotten much better at accommodating drivers’ preference optionality: the guy who picked me up drove over 110mph all the way to the airport. Apparently when you drive like that, especially in an EV, the more miles the better…
He explained that he could dial into the app that he preferred longer trips and trips between areas that happened to be connected by this lawless highway.
Driving an EV at 110mph is going to sap the range horrendously. The power needed to overcome air resistance/drag is proportional to the cube of the speed.
Do the worker's even see how much they'll make up front? If not, how is this fair, or even legal? I'm doubtful anyone here would go and work for McDonald's with the agreement being that they pay you what they think you're worth, after the job's done. We all see the asymmetry at play, and how it'd be abused at a moments notice.
We use Shipt regularly and it’s a bit different than the other delivery apps. I now have a collection of favorite shoppers. While jobs still go to the pool, these people have a first shot at my order. I’ve learned the general availability of my favorites and tend to place orders when I think there’s a high chance they’ll be available to shop the order.
While it’s still not me selecting an individual contractor, it’s not the randomness of other apps.
So... they've turned it from "we provide a good quality service at a fair price" to "over time you will have a chance to pick favorites, but not consistently".
That's enshittification for your use of the service, and enshittification for the workers, too.
If it were as good or better than before, nobody would want to pay attention to when to schedule deliveries based on the deliverator. Since this person specifically wants some deliverators over others, that means that the quality of the service is less good. They need to spend time considering scheduling, where they did not before.
I don't worry about which USPS mail carrier delivers my mail -- I know it will be consistent and good enough. I happen to know who my usual carrier is, because I work from home and she likes to say hi to cats if they are in the front window. I also know the face of the usual UPS driver and the usual FedEx driver; they aren't here 6 days a week, but often enough that I recognize them.
In none of those cases do I expect a quality change based on the driver. I expect competence, and I get it so often that the exceptions really stand out.
From the Shipt workers' perspective, they now need to worry about customers discriminating among them rather than just getting the job done.
This “Shipt,” though, involves an opportunity for some degree of relationship to make a difference, right? Your mail carrier must deliver your package, the package is the package, it’s either delivered or not. Maybe there’s a small margin around the edge where one carrier is nice to the cats and the other isn’t.
These Shipt people, though, have to interpret your preferences and essentially act as your agent as they decide what to pick from the store shelves on your behalf. Sometimes they make decisions that you probably would have made, sometimes less so; sometimes they’re confident that you understand each other, sometimes they’re nervous and want to hassle you about each of 10 different little decision points. When you find somebody I work well with, isn’t it a positive that you get to try to keep that relationship for future transactions? Isn’t this the same dynamic underpinning virtually every in-person service, from your hair cutting human to the tradies who do work on your house to the dry cleaner?
For that matter, doesn’t it create a perverse incentive if worker doesn’t believe that trying to understand my preferences will ever pay off? That it’s a one-off game rather than an iterated series of games, and effort to excel and bring human judgment to bear is wasted because there’s no way to reward it?
Doesn’t the enshittification tend to require as a prerequisite that a platform is successful at alienating service providers from service recipients (and from each other) like that?
This is the question. All the ethical concerns are almost superfluous if the provider knows how much they're gonna earn, at a minimum, before they accept the gig. It's either worth it to them or it isn't.
If anything shoud be a regulation, this feels like the one to add: platform opportunities must estimate and prominently display the estimated time to complete the task and the minimum payout after platform fees.
> Remember, consumers are using platform apps like Uber because they don't trust the drivers on the other side.
I don’t understand where these complex theories about ride sharing apps come from.
People use Uber because it’s easy and it’s an app. Taxis did not have a universal app at the time.
If you talk to young Uber users, chances are they wouldn’t actually know how to call a traditional taxi if you asked. It’s either Uber or Lyft because those are the apps they’ve heard about.
Also, it’s common for drivers to work for both Uber and Lyft at different points, maybe the same time. There’s no real element of trust difference between the two options.
If there was a driver you trusted then you could text them or their taxi dispatcher. The problem is they won't come, will scam you, won't take you to a poor neighborhood, etc.
Uber solves one part of the equation. As I have recently learned, ride sharing apps just allow you to get in touch with the customer. In some countries, apparently, the drive will 'work with rider' outside app control. It is a weird cat and mouse game.
Having never used Shipt, I also find that part unclear:
> Target… offered same-day delivery from local stores. Those deliveries were made by Shipt workers, who shopped for the items and drove them to customers’ doorsteps. Business was booming… and yet workers found that their paychecks had become… unpredictable. They were doing the same work they’d always done, yet their paychecks were often less than they expected.
Edit:
> On Facebook and Reddit, workers compared notes. Previously, they’d known what to expect from their pay because Shipt had a formula: It gave workers a base pay of $5 per delivery plus 7.5 percent of the total amount of the customer’s order through the app. That formula allowed workers to look at order amounts and choose jobs that were worth their time. But Shipt had changed the payment rules without alerting workers. When the company finally issued a press release about the change, it revealed only that the new pay algorithm paid workers based on “effort,” which included factors like the order amount, the estimated amount of time required for shopping, and the mileage driven.
If the whole point of the algo change is to correct an unfairness by which a strict fee+cart value approach doesn’t reliably reflect the amount of work somebody’s being asked to do, isn’t this exactly the outcome we expect? That the people who were putting more work in now get more money, while the people who were benefiting from sniffing out the “easy” jobs now make something more in line with everybody else’s compensation?
It does seem unsporting on the company’s part to play coy about the details. I wonder what the imperative was there: to avoid squabbling with workers about what “effort” means? To reduce the chances of legal scrutiny in one of the thousands of jurisdictions they operate in? To preserve the flexibility to quietly turn the dial in their own favor in the future?
I’m reminded of how Uber caught flak over surge pricing, and ultimately dealt with that by making pricing completely opaque. Now they still might say “prices are a little higher because of the weather” if they decide to, but normally you don’t even expect to know whether your price for a given ride is based on their estimate of your desperation, their having sized you up as price-insensitive, driver supply, or what…
Apart from the workers and the company there is another important actor here - the clients. I think the point of the algo change was to better serve clients with smaller orders.
There is an inherent flaw in those algos - they can be played by bots that scan for the best orders, while workers without bots and customers with smaller orders are left hanging. Better to just pay by the hour - you agree to deliver any order thrown at you during your shift.
> The system used optical character recognition—the same technology that lets you search for a word in a PDF file
That's not correct, at least for "digitally-born PDFs" that were made on a computer and haven't been scanned. In that case, the PDF can be parsed directly, without OCR, to get text. That's what a tool like PyPDF2 does, for example.
>Those deliveries were made by Shipt workers, who shopped for the items and drove them to customers’ doorsteps.
I've seen Shipt's operations internally, and they don't go shopping for stuff at stores and then deliver them, unless that's a different part of the business.
“There’s no technical reason why these algorithms need to be black boxes; the real reason is to maintain the power structure.”
I’m kind of amazed that the article has the courage to say this out loud. The New York Times or any mainstream publication would never have been so honest.
If anything they would have said some weasel words like “some ex-associates of shipt have complained that the app’s compensation system is unfair.” Rather than just blurt out the truth, which is that it’s unfair by design because the owners of the app want to maintain a certain power relationship. It’s the kind of thing that everyone knows but is not allowed to say in printed form.
> I’m kind of amazed that the article has the courage to say this out loud. The New York Times or any mainstream publication would never have been so honest.
It's the IEEE, (we) Engineers are known for having an aversion to bullshit, and just for straight-up having no filter. I wish more of the world worked that way.
Wow, gig or not, workers should be paid transparently.
Which should be obvious but this is kind of the problem with enshittification where once a business feels they have a bit of a moat (like with a two sided marketplace) they will erode the service to take every advantage unless stopped by regulation. No one likes regulation because it's effectively crufty technical debt and our political system is far too slow, corrupt, or incompetent to effectively refactor it so the best we can do is either nothing and endure the enshittification or layer on more cruft, usually far after the fact multiple years and court fights later.
> They asked for a meeting with Shipt executives, but they never got a direct response from the company. Its statements to the media were maddeningly vague, saying only that the new payment algorithm compensated workers based on the effort required for a job, and implying that workers had the upper hand because they could "choose whether or not they want to accept an order."
> Did the protests and news coverage have an effect on worker conditions? We don’t know, and that’s disheartening.
> choose whether or not they want to accept an order."
If the app shows clearly what needs to be done (shop, order list, miles driven), and the pay the worker will earn, and asks if they want to accept, then IMO that's fine.
The business can set those offers however they like, even using a random number generator if they want, and IMO it's morally fine.
They can set offers however they like and you are free to not accept.
If the algorithm detects that you are likely to accept for little money and short you with lower offers compared to other users, is it still morally fine?
This runs into both the ideals and the limitations of the Free Market.
Ideally, there's incentive for people to collectively reach the most efficient solution through aggregated laziness and greed.
In practice, people only have so much bandwidth and shortcuts will be taken, options will be overlooked, and people will exploit or be exploited due to the blinders either put on willingly or forced on them--on top of our natural capacity for observing reality no matter how much information is provided.
Look to two-tailed tests when a flipside doesn't make sense.
Consider what risk might exist if you fear overpaying so much that you make a lowball offer yet someone feels compelled to accept. The product or service might be "done" but in a way that screws you over in the long run as well.
I think reasonable thing here would also allow contractors send counter offer. Maybe 10x 100x or 1000x. Then it would be up to side ordering to accept or reject one of those.
And if the company makes it a policy to never accept any counteroffer (which is legal and fair), you're back to the same system, without that feature existing.
Platforms sometimes care—by which I mean, achieve a market position that means sellers can’t afford not to use them, then leverage that power to force lots and lots of weaker people and entities to do what they want, possibly causing higher lowest-prices in the overall market in the process, so, also hurting buyers.
What if there's discrimination built in to the system? Maybe a business is willing to pay white people more, or women less. They can do that while still following your framework. Is that moral?
When they offer to pay you X for the job, but then pay you < X.
Or if they get you to pay them money upfront (ie. for uniforms) on the basis of 'workers earn $Y per day', but then change the rules so some workers don't earn Y per day and don't offer a refund of the upfront payment to unhappy workers.
Looks like shipt/target successfully converted gig work back from a percentage of revenue (percentage of cart value) to a task based rate. Workers lose when they can't capture value proportional to the revenue generation they support, only in proportion to their hours of labor.
If workers are low-skilled, easily replaceable and practically fungible then realistically speaking why would their employer pay them based on value-added?
> Workers lose when they can't capture value proportional to the revenue generation they support, only in proportion to their hours of labor.
Time-based contracts are pretty normal. I imagine most people on the planet are on them. There are exceptions - e.g. sales commissions - but to say that workers lose on the thing that most people do requires at least some elaboration.
>60 percent of workers were making about the same or slightly more under the new scheme. But we felt that it was important to shine a light on those 40 percent of workers
Absolutely pathetic investigative journalism on display. This is a hit piece thinly veiled under the guise of being pro worker that fails to support the main point of algorithmic management of gig workers is worse for everyone but the corporation employing it.
If anything, they proved that shipt's algo did exactly what it was designed and reported to do, make payments more fair.
> If anything, they proved that shipt's algo did exactly what it was designed and reported to do, make payments more fair.
They didn't prove that. It's entirely possible the algo was skimming something off the top. It's entirely possible the algo was disproportionately rewarding some people to the detriment of others. A lot of it depends on what one thinks is "fair"... and without transparency, we can't even judge whether it is or not... which itself could be argued is unfair.
What's silly in this case is that (as others have pointed out) the new algorithm seems to have been reasonably equitable, with a genuine redistribution of payments, rather than just a cut overall. Shipt could have avoided this whole situation with a straightforward explanation of the changes, together with a few examples of the cases/jobs in which people would earn more or less.
I work for a salary, which is fully transparent in the sense that I know what my next paycheck will be to the penny. (It’s not transparent in how it’s set, but it is week-to-week.) If my employer started paying me based on effort, and didn’t tell me what constituted effort, not only would I be pissed off but that would be completely illegal.
I’m not suggesting that this change is or should be illegal. But if it happened to me I’d find it extremely unfair.
If Shipt is actually trying to incentivize better performance, it seems the best way is to be completely transparent about the rewards algorithm. "Short high-value trips are now somewhat de-rated, and trips requiring more effort now have improved rewards, specifically ..." or whatever.
This "communications team" approach did everyone a disservice if Shipt mgt were really trying to improve results.
OTOH, if the actual goal was to screw workers harder, they accomplished that, as here ate arguments on HN about how this could be good for the workers, thus successfully obfuscating the goal of screw-the-workers.
By what mechanism do you suggest the worker-screwing is happening here?
cutting costs is not "screwing workers". cutting costs is key to acting in a competitive market.
I have thought about this topic for a while at the time that I worked with Law data (e.g. Family Law and Military Law), and I just came to the conclusion that several societal institutions and it's agents are inherently intransparent, even in situations where some "illusionist transparency" (there's transparency, but the magician deviates your attention to another side) is given (e.g. judiciary system, under-the-table political agreements, etc.).
That's one of the reasons I would like to have a more algorithmic society with human in the loop calling the final shots and placing the rationale on top. An algorithm will have human and institutional biases but in some sort, you can explain part of it and fine-tune it; a human making the final call on top of a given option would need to explain and rationally explain its decision. At best a human actor will use logic and make the right call, at worst it will transparently expose the biases of the individual.
Let’s re-audit the algorithm regularly; say, perhaps, a central committee revisits and revises the plan every 5 years?
I will admit that it is an interesting idea. I am not sure it would work well as a lot of the power ( and pressure to adjust as needed ) suddenly would move to the fine-tuning portion of the process to ensure human at the top can approve 'right' decisions. I am going to get my coffee now.
To who? What you describe does not seem much different than the representation governments most of us here are accustomed to, other than the algorithm eases some day-to-day work required of the constituents. Already nobody cares, and no doubt would care even less if they could let an algorithm let them be even less involved.
It feels to me like the problem wasn't the change. For all we know, the change was a net good thing. The bad thing was the context in which the change occurred.
We already do; uncertainty is fundamental at all levels.
That statement is wrong either way. Looking at the graph, ~22% got a >10% cut while ~36% got a >10% raise. Overall ~43% got a cut while ~57% got a raise. And if there’s any doubt, later on they dropped the “at least 10% less” qualifier for the 40% figure in text:
> But we felt that it was important to shine a light on those 40 percent of workers who had gotten an unannounced pay cut through a black box transition.
Can’t believe they (accidentally? intentionally?) screwed up the very first concrete figure given in the article. Guess what, discussion is now based on the wrong figure.
Also, the y-axis of the plot is labeled “number of workers” when it should be “percentage of participating workers”, unless they had exactly 100 participants (they say they had 200+). Lousy presentation.
I’m all for transparency, obviously.
Edit: In addition, since participation is entirely voluntary, common sense tells me that the data they gather should skew more negative, since people negatively affected are much more likely to participate.
that's not to say sharing wages is bad. this can still provide upward pressure on said wages. but the people who participate in the beginning are likely from the lower end of the distribution.
[0] websites like levels.fyi seem to consistently skew low when estimating higher percentiles
> It wasn’t a clear case of wage theft, because 60 percent of workers were making about the same or slightly more under the new scheme.
Your statement that "some 30+% are getting at least 10% more" assumes that there is no wage theft - which is not cut in stone.
Sometimes that goes afoul when it's ruled that they put requirements that are only valid for employees on contractors.
I don't follow this? Is this predicated on the fact that gig corps can choose not to work with contractors that don't meet their criteria? If so, how is this different from only using lumber yards that consistently meet your expectations?
The implication of the parent poster seems to be that there are legal requirements regarding contractors and legal requirements regarding employees but gig corps would prefer to treat their workers as one or the other class depending on which is to their benefit - which would be against the law because of the aforementioned concept "legal requirements".
- requiring specific dress code is non-enforceable on contractors in many places
- contractor is not required to provide specific person to fulfill the job, only a person of appropriate qualifications (it's valid for there to be a check on those qualifications)
- in UK case, contractor might be asked to prove that they have a substitute to work in their place!
- [Poland, possibly other] having only one client is not illegal, but can be grounds for investigations and if it's your only client where you work for equivalent of full-time job, it will be evidence for tax fraud
- You can not enforce working hours on contractors in most jurisdiction, only specific deliverables (taking part of work meetings is deliverable, requiring availability in general of specific person at specific times can be grounds for reclassification)
- above is often linked with "gig economy" - rules regarding "contractors" needing to pick up available jobs etc. are often considered illegal skirting of employment law.
As sibling comment mentioned, more is available from your local (too) friendly search engine. And employment lawyers and HR specialists.
There is a way of calculating the pay for a job. It is predictable. Publish the algorithm. Want to change it? Great. Update the documentation and then publish that. The workers should be able to calculate exactly what they are owed. They can decide to leave or stay.
Only in America are people deflecting by bringing up the employment status of people when the issue is a lack of transparency designed to allow wage theft.
You’re arguing that it’s sometimes legitimate to hide payment information. So I can hire someone for a job, and only I know what the job is worth. The worker just has to try it out and see if their valuation of the job is the same as my valuation?
The fact that some, maybe most, are making more money is irrelevant. The organisation can change the value of a particular job at any time and simply say “the algorithm made me do it”.
1. If they can game it, they should. If that's a problem, then Shipt should fix their algorithm.
2. It's fine to data wrangle your way to an automated model for e.g. your company's growth projection or for predicting where you can best expand to find more customers. It's not okay to use it to unilaterally change pay agreements with your workers or "contractors".
3. The right of an individual to know what they're getting paid for their work outweighs any company's nebulous claim of its algorithm being a trade secret or something vague like that. Can you imagine businnesses actually operating like this? "You can't know what we'll pay you for this job, its ~~seeeecret~~. Just trust us."
From now on I'm going to pay my rent according to a secret algorithm of my own devising and my landlord will just have to hope I'm generous this month. I can't have him gaming the lease agreement by only providing the things it says!
"Surely in a hotly competitive market" there is no need to hypothesise about what "surely" would happen according to some wishful thinking. We know from countless past observation exactly what happens in any market of low skill low investment labor, wages go straight to the level of the most desperate willing to be slaves and make nothing at all, because there always are enough of those for everyone else to take advantage of, and having no money they have no power to demand better. They can't afford lawyers and ad campaigns and lobbyists and politicians, and they can't afford to strike, and that same wonderful magic market means there is no one better to work for, all employers are essentially the same. Unions and strikes do exist, some places where they are allowed to, and it's always a big news story and a miracle when they actually accomplish even 10% of what they needed to once in a blue moon.
From a game-theoretic perspective in a gig marketplace you don’t want jobs that are strictly better, else sophisticated market participants (workers) will select the best ones leaving chaff - and a worse experience - for the less sophisticated participants.
What you are looking for is preference optionality, eg one Uber driver might prefer not to do very long trips, another might prefer it, and you ideally get paid fairly for either.
In this case as others have noted, it doesn’t actually sound like an unfair change. Perhaps communications could have been better though.
Central planning would be if a committee decided how many rides there would be each year, regardless of demand.
Gamifying peoples livelihood is the problem.
That is… exactly the point I made, when I said:
> you don’t want jobs that are strictly better
Preference optionality is widely stated to be one of the features that gig workers like about the arrangement.
The options you suggest are also valid ways of homogenizing the jobs to reduce variance.
> Gamifying peoples livelihood is the problem
To be clear Game Theory applies to all economic interactions. Mechanism Design is the branch of Game Theory pertaining to market design to achieve desired outcomes, such as “avoid adverse selection in my gig work marketplace”.
Gamification is a specific application of video game design to economic interactions, it’s unrelated to what I’m discussing. (Examples of Gamification would be gaining experience points and levels for delivery, daily checking rewards, achievement badges, etc. - the general goal in Gamification is setting up a dopamine loop to encourage repeat use of the app. Hopefully it’s clear this is not what I was talking about.)
Game-theory is fun when you get good at it for designing markets and products, but let us not lose sight of the crucial discussion.. human beings with real lives are not equal to economic parts.
Or is this attitude of “hey, it’s a long shot, but let’s give it a try and see if anybody takes the job” closer to the attitude you’d like to see? If the latter, how would you communicate that to the users?
I had an occasion a long time ago where I needed to request an Uber for a ride similar to what you’re describing. At that time, apparently the driver didn’t find out the route until they’d committed to the ride. The guy swiped to say he’d picked me up, and more or less broke down in tears when he found out where I was needing to go. He lived 40 miles in the other direction, was going off shift, and would be driving the whole 90 miles home without any prayer of a passenger to cover the time or cost. In that case I ended up giving the guy a generous amount of cash to cover the imposition, but I couldn’t bring myself to use Uber for that route in the future.
Until recently, when I had to use Uber for that route again. This time it seemed like they’d gotten much better at accommodating drivers’ preference optionality: the guy who picked me up drove over 110mph all the way to the airport. Apparently when you drive like that, especially in an EV, the more miles the better…
He explained that he could dial into the app that he preferred longer trips and trips between areas that happened to be connected by this lawless highway.
We use Shipt regularly and it’s a bit different than the other delivery apps. I now have a collection of favorite shoppers. While jobs still go to the pool, these people have a first shot at my order. I’ve learned the general availability of my favorites and tend to place orders when I think there’s a high chance they’ll be available to shop the order.
While it’s still not me selecting an individual contractor, it’s not the randomness of other apps.
That's enshittification for your use of the service, and enshittification for the workers, too.
I don't worry about which USPS mail carrier delivers my mail -- I know it will be consistent and good enough. I happen to know who my usual carrier is, because I work from home and she likes to say hi to cats if they are in the front window. I also know the face of the usual UPS driver and the usual FedEx driver; they aren't here 6 days a week, but often enough that I recognize them.
In none of those cases do I expect a quality change based on the driver. I expect competence, and I get it so often that the exceptions really stand out.
From the Shipt workers' perspective, they now need to worry about customers discriminating among them rather than just getting the job done.
These Shipt people, though, have to interpret your preferences and essentially act as your agent as they decide what to pick from the store shelves on your behalf. Sometimes they make decisions that you probably would have made, sometimes less so; sometimes they’re confident that you understand each other, sometimes they’re nervous and want to hassle you about each of 10 different little decision points. When you find somebody I work well with, isn’t it a positive that you get to try to keep that relationship for future transactions? Isn’t this the same dynamic underpinning virtually every in-person service, from your hair cutting human to the tradies who do work on your house to the dry cleaner?
For that matter, doesn’t it create a perverse incentive if worker doesn’t believe that trying to understand my preferences will ever pay off? That it’s a one-off game rather than an iterated series of games, and effort to excel and bring human judgment to bear is wasted because there’s no way to reward it?
Doesn’t the enshittification tend to require as a prerequisite that a platform is successful at alienating service providers from service recipients (and from each other) like that?
If anything shoud be a regulation, this feels like the one to add: platform opportunities must estimate and prominently display the estimated time to complete the task and the minimum payout after platform fees.
Remember, consumers are using platform apps like Uber because they don't trust the drivers on the other side.
Oh, I thought it was because "push button to summon car" is, like, super convenient.
I don’t understand where these complex theories about ride sharing apps come from.
People use Uber because it’s easy and it’s an app. Taxis did not have a universal app at the time.
If you talk to young Uber users, chances are they wouldn’t actually know how to call a traditional taxi if you asked. It’s either Uber or Lyft because those are the apps they’ve heard about.
Also, it’s common for drivers to work for both Uber and Lyft at different points, maybe the same time. There’s no real element of trust difference between the two options.
> Target… offered same-day delivery from local stores. Those deliveries were made by Shipt workers, who shopped for the items and drove them to customers’ doorsteps. Business was booming… and yet workers found that their paychecks had become… unpredictable. They were doing the same work they’d always done, yet their paychecks were often less than they expected.
Edit:
> On Facebook and Reddit, workers compared notes. Previously, they’d known what to expect from their pay because Shipt had a formula: It gave workers a base pay of $5 per delivery plus 7.5 percent of the total amount of the customer’s order through the app. That formula allowed workers to look at order amounts and choose jobs that were worth their time. But Shipt had changed the payment rules without alerting workers. When the company finally issued a press release about the change, it revealed only that the new pay algorithm paid workers based on “effort,” which included factors like the order amount, the estimated amount of time required for shopping, and the mileage driven.
Even convoluted ones, like commissions for sales, or shared tips, are covered by law. I do know this, as I know a number of salespeople and servers.
I suspect that the government needs to know what to tax, and obfuscated pay, means obfuscated taxes.
You're taxed on what you're paid. The government doesn't do a parallel calculation and tax you on that.
My experience is that governments are quite interested in where the money goes.
That is money that leaves the employer’s account and goes into the employee’s account (or government’s account for tax withholding).
I don’t see how this can be obfuscated.
Not my area of expertise. That's why I pay an accountant.
It does seem unsporting on the company’s part to play coy about the details. I wonder what the imperative was there: to avoid squabbling with workers about what “effort” means? To reduce the chances of legal scrutiny in one of the thousands of jurisdictions they operate in? To preserve the flexibility to quietly turn the dial in their own favor in the future?
I’m reminded of how Uber caught flak over surge pricing, and ultimately dealt with that by making pricing completely opaque. Now they still might say “prices are a little higher because of the weather” if they decide to, but normally you don’t even expect to know whether your price for a given ride is based on their estimate of your desperation, their having sized you up as price-insensitive, driver supply, or what…
That's not correct, at least for "digitally-born PDFs" that were made on a computer and haven't been scanned. In that case, the PDF can be parsed directly, without OCR, to get text. That's what a tool like PyPDF2 does, for example.
I've seen Shipt's operations internally, and they don't go shopping for stuff at stores and then deliver them, unless that's a different part of the business.
I’m kind of amazed that the article has the courage to say this out loud. The New York Times or any mainstream publication would never have been so honest.
If anything they would have said some weasel words like “some ex-associates of shipt have complained that the app’s compensation system is unfair.” Rather than just blurt out the truth, which is that it’s unfair by design because the owners of the app want to maintain a certain power relationship. It’s the kind of thing that everyone knows but is not allowed to say in printed form.
It's the IEEE, (we) Engineers are known for having an aversion to bullshit, and just for straight-up having no filter. I wish more of the world worked that way.
Which should be obvious but this is kind of the problem with enshittification where once a business feels they have a bit of a moat (like with a two sided marketplace) they will erode the service to take every advantage unless stopped by regulation. No one likes regulation because it's effectively crufty technical debt and our political system is far too slow, corrupt, or incompetent to effectively refactor it so the best we can do is either nothing and endure the enshittification or layer on more cruft, usually far after the fact multiple years and court fights later.
> They asked for a meeting with Shipt executives, but they never got a direct response from the company. Its statements to the media were maddeningly vague, saying only that the new payment algorithm compensated workers based on the effort required for a job, and implying that workers had the upper hand because they could "choose whether or not they want to accept an order."
> Did the protests and news coverage have an effect on worker conditions? We don’t know, and that’s disheartening.
If the app shows clearly what needs to be done (shop, order list, miles driven), and the pay the worker will earn, and asks if they want to accept, then IMO that's fine.
The business can set those offers however they like, even using a random number generator if they want, and IMO it's morally fine.
If the algorithm detects that you are likely to accept for little money and short you with lower offers compared to other users, is it still morally fine?
Ideally, there's incentive for people to collectively reach the most efficient solution through aggregated laziness and greed.
In practice, people only have so much bandwidth and shortcuts will be taken, options will be overlooked, and people will exploit or be exploited due to the blinders either put on willingly or forced on them--on top of our natural capacity for observing reality no matter how much information is provided.
Consider what risk might exist if you fear overpaying so much that you make a lowball offer yet someone feels compelled to accept. The product or service might be "done" but in a way that screws you over in the long run as well.
Trust is earned, and it flows both ways.
Platforms sometimes care—by which I mean, achieve a market position that means sellers can’t afford not to use them, then leverage that power to force lots and lots of weaker people and entities to do what they want, possibly causing higher lowest-prices in the overall market in the process, so, also hurting buyers.
At least with laws, there are clear adjudicators on the issues at hand.
When it becomes immoral from your perspective?
Or if they get you to pay them money upfront (ie. for uniforms) on the basis of 'workers earn $Y per day', but then change the rules so some workers don't earn Y per day and don't offer a refund of the upfront payment to unhappy workers.
Time-based contracts are pretty normal. I imagine most people on the planet are on them. There are exceptions - e.g. sales commissions - but to say that workers lose on the thing that most people do requires at least some elaboration.
Absolutely pathetic investigative journalism on display. This is a hit piece thinly veiled under the guise of being pro worker that fails to support the main point of algorithmic management of gig workers is worse for everyone but the corporation employing it.
If anything, they proved that shipt's algo did exactly what it was designed and reported to do, make payments more fair.
They didn't prove that. It's entirely possible the algo was skimming something off the top. It's entirely possible the algo was disproportionately rewarding some people to the detriment of others. A lot of it depends on what one thinks is "fair"... and without transparency, we can't even judge whether it is or not... which itself could be argued is unfair.