7 ms·

The Gender Earnings Gap: Evidence from Over a Million Rideshare Drivers [pdf]

43 points by malchow 9y ago

prepend 9y ago
This is interesting. What interventions can be made to reduce this pay gap? With the reasons in the paper, should this gap be corrected (eg, force maximum speed, pay higher for slow drivers but only gender specific, etc etc).
- fatjokes 9y ago
  > pay higher for slow drivers but only gender specific ... that's wage discrimination and is illegal.
  - mc32 9y ago
    Illegal or not it would certainly be gamed.
    - T2_t2 9y ago
      "I identify as a non-binary person - what do I get paid?"
      - mc32 9y ago
        probably trend toward the one you most closely resemble behaviorally.
      - erikpukinskis 9y ago
        What does your drivers license say?
- dragonwriter 9y ago
  > What interventions can be made to reduce this pay gap? Does the pay gap exist after adjusting for insurance costs and expected long-term driver-born per mile cost not covered by insurance, which Uber drivers bear? Since the cause of the higher pay seems directly linked to the reason for the gender-based difference in insurance rates, this might be something to consider.
- pixel_fcker 9y ago
  Should interventions be made to reduce the gap?
  - mlevental 9y ago
    what exactly are you implying? is it that it'll correct itself naturally (free market forces) or that it doesn't need to be corrected?
    - pixel_fcker 9y ago
      I'm implying that it doesn't need to be corrected.
      - mlevental 9y ago
        why do you think it doesn't need to be corrected? you don't see how an equitable distribution of resources between the sexes is much more productive for the whole? nm the morality of it.
        pixel_fcker 9y ago
        I believe strongly in an equal distribution of opportunity between the sexes. How would you propose fixing the gap that the paper describes? Simply paying women drivers 7% more? Would that be fair?
        mlevental 9y ago
        resources should be allocated to correct for this >The second factor is work experience. Even in the relatively simple production of a passenger’s ride, past experience is valuable for drivers. i.e. women should be given training to compensate for the a priori difference in experience. women should also be educated about this >hourly earnings on Uber vary predictably by location and time of week, and men tend to drive in more lucrative locations. the obvious question becomes why aren't women driving during more lucrative times. is it because those times are late nights during which it's potentially more dangerous for a woman driver. this is of course a systemic issue of a larger system. my point is that usually when someone claims "accidental/benign inequality" all one needs to do is expand the scope to discover the malignant root.
- Zanni 9y ago
  Since the pay gap is fair (in the sense that it's driven by personal preferences and experience), there are only two reasons to reduce it: 1) if you think it's going to cost you women drivers and 2) if you think that men are driving dangerously fast. Point 2 is something that Uber actually has enough data to assess and address, but the better motivation would be safety. Reducing the pay gap would be an ancillary benefit. As for point 1, I think the only option is education. Point out that pay increases with experience, and by how much. Maybe draw drivers' attention to more lucrative areas? Though I think that's contingent on experience as well. Beyond that, any attempt to address the pay gap seems unfair and, as fatjokes points out, possibly illegal.
- neveroffensive 9y ago
  > What interventions can be made to reduce this pay gap? This is a scary question to me. Is making arbitrary changes to align the world more with your definition of fairness really a good idea? Isn't this the precursor to authoritarian/totalitarian thinking? People chose to drive faster, people chose to prefer faster rides, neither of these things are "wrong" in any way. The market is simply rewarding those who provide service which is most desirable. I'm not saying it's fair, but it's not unfair in a way we should try and control IMHO. Any human attempts to correct the arbitrary injustice/unfairness in nature are only bound to shift/perpetuate it. Edit: My point really is, nature is evil, but by making choices ourselves which harm others to decrease natures evil, we become evil ourselves. It's the classic trolley/surgeon problem. I argue it's better not to flip the switch on the track and let nature take it's course because otherwise you become the arbiter, and a murderer. It seems more heartless with the trolley problem, but it's conforms to intuition with the surgeon scenario.
- tenpies 9y ago
  > What interventions can be made to reduce this pay gap? I don't see how you could do this without rewarding inefficiency and inexperience. More efficient workers with more experience should absolutely get paid more than new workers that don't do as good a job. Our entire society is structured on this concept.
  - danjoc 9y ago
    >I don't see how you could do this without rewarding inefficiency and inexperience. It's easy though. Just pay women more. Uber can take a smaller cut of your pay if you have two x chromosomes. Another approach would be to allow passengers to chose their driver's gender. More passengers will chose female over male drivers.
    - aisofteng 9y ago
      That's wage discrimination and is illegal.
      - danjoc 9y ago
        And paying women less is legal? https://www.eeoc.gov/laws/types/equalcompensation.cfm https://www.eeoc.gov/laws/types/equalcompensation.cfm "The Equal Pay Act requires that men and women in the same workplace be given equal pay for equal work." "If there is an inequality in wages between men and women, employers may not reduce the wages of either sex to equalize their pay." Seems pretty straightforward. Pay has to be equal. Can't lower wages. Therefore, pay women more. Easy as pie.
        Geeek 9y ago
        "for equal work" you missed the most important line. This paper is showing that they are not doing equal work. Not time wise nor trip wise.
        danjoc 9y ago
- omarforgotpwd 9y ago
  Better protections for female drivers that make them less likely stop driving for Uber. For example having an option to favor Women or only pickup women, or only pickup women and men with a high rating, etc so that they can feel more safe driving for Uber. If a female Uber driver has controls that allow her to stay out driving longer than she would normally be comfortable with, she can generate higher earnings and close some of that gap.
  - throwaway1748 9y ago
    Yes, but men are statistically more likely than women to be the victim of violence. Male drivers are taking on just as much, if not more of a safety risk driving for Uber, why should special accommodation be made for women?
    - omarforgotpwd 9y ago
      Because the men are not as bothered by the risk as women and continue driving for Uber whereas women drivers do not.
- tengbretson 9y ago
  >What interventions can be made to reduce this pay gap? It would actually be pretty simple to eliminate this pay gap. All we would need to do is make it illegal to base the amount you pay someone on things like how much you actually valued their service.
little_data 9y ago
"Furthermore, while Uber now allows riders to tip their drivers in-app, this did not become available until June 2017, which is outside the scope of our data. We do not believe that cash tips – which were possible before in-app tipping – had a material impact on driver earnings." The study did not have any actual tip data for drivers. Tipping was always an option in Uber before June 2017, it just became mandatory for the app to ask a user after each ride. I really think the study is missing some crucial data surrounding tips.
- beisner 9y ago
  Anecdotally, nobody I know ever tipped an Uber driver before tipping was introduced in the app, so there being a negligible tipping component would not surprise me at all.
- yyyyip 9y ago
  On the freakonomics podcast episode with the study authors, they said that on average women earn over 10 percent more than men from tips.
- jelgt2011 9y ago
  > Tipping was always an option in Uber before June 2017, it just became mandatory for the app to ask a user after each ride. That's not true, tipping was not possible in the app, at all.
millermp12 9y ago
"We find that the entire gender gap is caused by three factors: experience on the platform (learning-by-doing), preferences over where/when to work, and preferences for driving speed." So men make more because they are more likely to take initiative, be more flexible, and drive faster? What exactly is the problem?
pbiggar 9y ago
So the service can be unbiased, but the society/system it exists in will still be biased.
- neveroffensive 9y ago
  > but the society/system it exists in will still be biased Is biased even the right word here? Maybe I'm just nit-picking, but bias implies some level of intention IMHO. The "bias" here is really just the coincidental effects of nature. I'd say it's more like, "So the service can be unbiased, but trends will inevitably emerge from natural arbitrary properties of the system".
- malchow 9y ago
  My read of the study is that it more or less proves the opposite: the wage gap is the result of women having such powerfully different preferences that it shows up even when we know discrimination is impossible.
  - pbiggar 9y ago
    You are letting my point sail over you. How do women come to those "powerfully different preferences"? Do we think that all women got together and decided to only drive at less valuable times? Or is it more likely that there are systemic/societal reasons why women as a group pick less-valuable times to drive?
    - dang 9y ago
      > You are letting my point sail over you. Please edit such swipes out of your comments here. Besides crossing into incivility, they add no information and make divisive topics even more divisive. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
    - Geeek 9y ago
      Can it be systemic yet not societal? Like biological origins?
konschubert 9y ago
Here is the money quote: > We can explain the entire gap with three factors. First, through the logic of compensating differentials, hourly earnings on Uber vary predictably by location and time of week, and men tend to drive in more lucrative locations. The second factor is work experience. Even in the relatively simple production of a passenger’s ride, past experience is valuable for drivers. A driver with more than 2,500 lifetime trips completed earns 14% more per hour than a driver who has completed fewer than 100 trips in her time on the platform, in part because she learn where to drive, when to drive, and how to strategically cancel and accept trips. Male drivers accumulate more experience than women by driving more each week and being less likely to stop driving with Uber. Because of these returns to experience and because the typical male Uber driver has more experience than the typical female—putting them higher on the learning curve—men earn more money per hour. The residual gender earnings gap that persists after controlling for these two factors can be explained by a single variable: average driving speed.
sebastos 9y ago
The entire point of the paper is to show that you can observe a statistical anomaly like wage disparity whenever you compare two different populations, even when you essentially control for job effectiveness and isolate out any influence from malicious discrimination. The Uber case is being used to crystallize an example where there exists a bias in the results, and yet it is clearly not due to prejudice. It is illustrating the point that just because there is a pay-gap doesn't mean the source of the pay-gap is employers who don't properly compensate their female employees. Non-malicious systematic influences can easily cause such an effect. And yet you still have people in here wondering what we can do about this new problem, or how this just goes to show that it's really the SYSTEM that's biased, man. All I can do is shake my head at this point.
- mlevental 9y ago
  these two things are in such direct contradiction >systematic influences can easily cause such an effect >how this just goes to show that it's really the SYSTEM that's biased that i have no idea what you're really trying to say. > Non-malicious no one cares if it's malicious or benign. ask a black person whether they care about the intentions of those they're being racially profiled by (for example). it doesn't matter. what matters if systemic inequalities exist - and here is proof that they do.
  - T2_t2 9y ago
    This is either some next level trolling, or really odd. > The entire point of the paper is to show that you can observe a statistical anomaly like wage disparity whenever you compare two different populations, even when you essentially control for job effectiveness and isolate out any influence from malicious discrimination.
    - mlevental 9y ago
      >statistical anomaly >systematic influences >bias in the results >SYSTEM that's biased you really don't see that who i'm responding to isn't at all consistent in what they're trying to say? the first 3 aren't mutually consistent and the fourth makes no sense when contextualized against those prior inconsistencies.
tomlock 9y ago
>Overall, our results suggest that, even in the gender-blind, transactional, flexible environment of the gig economy, gender-based preferences (especially the value of time not spent at paid work and, for drivers, preferences for driving speed) can open gender earnings gaps. Not sure this is the right conclusion to draw. If Uber driving is less safe for female drivers, it's hardly a "gender-based preference" that's driving the gap. If women are spending more of their flexible hours doing housework or caring for children than men do, this may not be a "preference". While the paper does a great job identifying factors correlated with lower income, does it really indicate an individual's "preference", or does it indicate that the situation they find themselves in either enables or prevents uber participation?

43 points by malchow 9y ago

This is interesting. What interventions can be made to reduce this pay gap? With the reasons in the paper, should this gap be corrected (eg, force maximum speed, pay higher for slow drivers but only gender specific, etc etc).

fatjokes 9y ago
> pay higher for slow drivers but only gender specific ... that's wage discrimination and is illegal.
- mc32 9y ago
  Illegal or not it would certainly be gamed.
  - T2_t2 9y ago
    "I identify as a non-binary person - what do I get paid?"
    - mc32 9y ago
      probably trend toward the one you most closely resemble behaviorally.
    - erikpukinskis 9y ago
      What does your drivers license say?
dragonwriter 9y ago
> What interventions can be made to reduce this pay gap? Does the pay gap exist after adjusting for insurance costs and expected long-term driver-born per mile cost not covered by insurance, which Uber drivers bear? Since the cause of the higher pay seems directly linked to the reason for the gender-based difference in insurance rates, this might be something to consider.
pixel_fcker 9y ago
Should interventions be made to reduce the gap?
- mlevental 9y ago
  what exactly are you implying? is it that it'll correct itself naturally (free market forces) or that it doesn't need to be corrected?
  - pixel_fcker 9y ago
    I'm implying that it doesn't need to be corrected.
    - mlevental 9y ago
      why do you think it doesn't need to be corrected? you don't see how an equitable distribution of resources between the sexes is much more productive for the whole? nm the morality of it.
      - pixel_fcker 9y ago
        I believe strongly in an equal distribution of opportunity between the sexes. How would you propose fixing the gap that the paper describes? Simply paying women drivers 7% more? Would that be fair?
        mlevental 9y ago
        resources should be allocated to correct for this >The second factor is work experience. Even in the relatively simple production of a passenger’s ride, past experience is valuable for drivers. i.e. women should be given training to compensate for the a priori difference in experience. women should also be educated about this >hourly earnings on Uber vary predictably by location and time of week, and men tend to drive in more lucrative locations. the obvious question becomes why aren't women driving during more lucrative times. is it because those times are late nights during which it's potentially more dangerous for a woman driver. this is of course a systemic issue of a larger system. my point is that usually when someone claims "accidental/benign inequality" all one needs to do is expand the scope to discover the malignant root.
Zanni 9y ago
Since the pay gap is fair (in the sense that it's driven by personal preferences and experience), there are only two reasons to reduce it: 1) if you think it's going to cost you women drivers and 2) if you think that men are driving dangerously fast. Point 2 is something that Uber actually has enough data to assess and address, but the better motivation would be safety. Reducing the pay gap would be an ancillary benefit. As for point 1, I think the only option is education. Point out that pay increases with experience, and by how much. Maybe draw drivers' attention to more lucrative areas? Though I think that's contingent on experience as well. Beyond that, any attempt to address the pay gap seems unfair and, as fatjokes points out, possibly illegal.
neveroffensive 9y ago
> What interventions can be made to reduce this pay gap? This is a scary question to me. Is making arbitrary changes to align the world more with your definition of fairness really a good idea? Isn't this the precursor to authoritarian/totalitarian thinking? People chose to drive faster, people chose to prefer faster rides, neither of these things are "wrong" in any way. The market is simply rewarding those who provide service which is most desirable. I'm not saying it's fair, but it's not unfair in a way we should try and control IMHO. Any human attempts to correct the arbitrary injustice/unfairness in nature are only bound to shift/perpetuate it. Edit: My point really is, nature is evil, but by making choices ourselves which harm others to decrease natures evil, we become evil ourselves. It's the classic trolley/surgeon problem. I argue it's better not to flip the switch on the track and let nature take it's course because otherwise you become the arbiter, and a murderer. It seems more heartless with the trolley problem, but it's conforms to intuition with the surgeon scenario.
tenpies 9y ago
> What interventions can be made to reduce this pay gap? I don't see how you could do this without rewarding inefficiency and inexperience. More efficient workers with more experience should absolutely get paid more than new workers that don't do as good a job. Our entire society is structured on this concept.
- danjoc 9y ago
  >I don't see how you could do this without rewarding inefficiency and inexperience. It's easy though. Just pay women more. Uber can take a smaller cut of your pay if you have two x chromosomes. Another approach would be to allow passengers to chose their driver's gender. More passengers will chose female over male drivers.
  - aisofteng 9y ago
    That's wage discrimination and is illegal.
    - danjoc 9y ago
      And paying women less is legal? https://www.eeoc.gov/laws/types/equalcompensation.cfm https://www.eeoc.gov/laws/types/equalcompensation.cfm "The Equal Pay Act requires that men and women in the same workplace be given equal pay for equal work." "If there is an inequality in wages between men and women, employers may not reduce the wages of either sex to equalize their pay." Seems pretty straightforward. Pay has to be equal. Can't lower wages. Therefore, pay women more. Easy as pie.
      - Geeek 9y ago
        "for equal work" you missed the most important line. This paper is showing that they are not doing equal work. Not time wise nor trip wise.
        danjoc 9y ago
omarforgotpwd 9y ago
Better protections for female drivers that make them less likely stop driving for Uber. For example having an option to favor Women or only pickup women, or only pickup women and men with a high rating, etc so that they can feel more safe driving for Uber. If a female Uber driver has controls that allow her to stay out driving longer than she would normally be comfortable with, she can generate higher earnings and close some of that gap.
- throwaway1748 9y ago
  Yes, but men are statistically more likely than women to be the victim of violence. Male drivers are taking on just as much, if not more of a safety risk driving for Uber, why should special accommodation be made for women?
  - omarforgotpwd 9y ago
    Because the men are not as bothered by the risk as women and continue driving for Uber whereas women drivers do not.
tengbretson 9y ago
>What interventions can be made to reduce this pay gap? It would actually be pretty simple to eliminate this pay gap. All we would need to do is make it illegal to base the amount you pay someone on things like how much you actually valued their service.

little_data 9y ago

"Furthermore, while Uber now allows riders to tip their drivers in-app, this did not become available until June 2017, which is outside the scope of our data. We do not believe that cash tips – which were possible before in-app tipping – had a material impact on driver earnings." The study did not have any actual tip data for drivers. Tipping was always an option in Uber before June 2017, it just became mandatory for the app to ask a user after each ride. I really think the study is missing some crucial data surrounding tips.

beisner 9y ago
Anecdotally, nobody I know ever tipped an Uber driver before tipping was introduced in the app, so there being a negligible tipping component would not surprise me at all.
yyyyip 9y ago
On the freakonomics podcast episode with the study authors, they said that on average women earn over 10 percent more than men from tips.
jelgt2011 9y ago
> Tipping was always an option in Uber before June 2017, it just became mandatory for the app to ask a user after each ride. That's not true, tipping was not possible in the app, at all.

millermp12 9y ago

"We find that the entire gender gap is caused by three factors: experience on the platform (learning-by-doing), preferences over where/when to work, and preferences for driving speed." So men make more because they are more likely to take initiative, be more flexible, and drive faster? What exactly is the problem?

pbiggar 9y ago

So the service can be unbiased, but the society/system it exists in will still be biased.

neveroffensive 9y ago
> but the society/system it exists in will still be biased Is biased even the right word here? Maybe I'm just nit-picking, but bias implies some level of intention IMHO. The "bias" here is really just the coincidental effects of nature. I'd say it's more like, "So the service can be unbiased, but trends will inevitably emerge from natural arbitrary properties of the system".
malchow 9y ago
My read of the study is that it more or less proves the opposite: the wage gap is the result of women having such powerfully different preferences that it shows up even when we know discrimination is impossible.
- pbiggar 9y ago
  You are letting my point sail over you. How do women come to those "powerfully different preferences"? Do we think that all women got together and decided to only drive at less valuable times? Or is it more likely that there are systemic/societal reasons why women as a group pick less-valuable times to drive?
  - dang 9y ago
    > You are letting my point sail over you. Please edit such swipes out of your comments here. Besides crossing into incivility, they add no information and make divisive topics even more divisive. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
  - Geeek 9y ago
    Can it be systemic yet not societal? Like biological origins?

konschubert 9y ago

Here is the money quote: > We can explain the entire gap with three factors. First, through the logic of compensating differentials, hourly earnings on Uber vary predictably by location and time of week, and men tend to drive in more lucrative locations. The second factor is work experience. Even in the relatively simple production of a passenger’s ride, past experience is valuable for drivers. A driver with more than 2,500 lifetime trips completed earns 14% more per hour than a driver who has completed fewer than 100 trips in her time on the platform, in part because she learn where to drive, when to drive, and how to strategically cancel and accept trips. Male drivers accumulate more experience than women by driving more each week and being less likely to stop driving with Uber. Because of these returns to experience and because the typical male Uber driver has more experience than the typical female—putting them higher on the learning curve—men earn more money per hour. The residual gender earnings gap that persists after controlling for these two factors can be explained by a single variable: average driving speed.

sebastos 9y ago

The entire point of the paper is to show that you can observe a statistical anomaly like wage disparity whenever you compare two different populations, even when you essentially control for job effectiveness and isolate out any influence from malicious discrimination. The Uber case is being used to crystallize an example where there exists a bias in the results, and yet it is clearly not due to prejudice. It is illustrating the point that just because there is a pay-gap doesn't mean the source of the pay-gap is employers who don't properly compensate their female employees. Non-malicious systematic influences can easily cause such an effect. And yet you still have people in here wondering what we can do about this new problem, or how this just goes to show that it's really the SYSTEM that's biased, man. All I can do is shake my head at this point.

mlevental 9y ago
these two things are in such direct contradiction >systematic influences can easily cause such an effect >how this just goes to show that it's really the SYSTEM that's biased that i have no idea what you're really trying to say. > Non-malicious no one cares if it's malicious or benign. ask a black person whether they care about the intentions of those they're being racially profiled by (for example). it doesn't matter. what matters if systemic inequalities exist - and here is proof that they do.
- T2_t2 9y ago
  This is either some next level trolling, or really odd. > The entire point of the paper is to show that you can observe a statistical anomaly like wage disparity whenever you compare two different populations, even when you essentially control for job effectiveness and isolate out any influence from malicious discrimination.
  - mlevental 9y ago
    >statistical anomaly >systematic influences >bias in the results >SYSTEM that's biased you really don't see that who i'm responding to isn't at all consistent in what they're trying to say? the first 3 aren't mutually consistent and the fourth makes no sense when contextualized against those prior inconsistencies.

tomlock 9y ago

>Overall, our results suggest that, even in the gender-blind, transactional, flexible environment of the gig economy, gender-based preferences (especially the value of time not spent at paid work and, for drivers, preferences for driving speed) can open gender earnings gaps. Not sure this is the right conclusion to draw. If Uber driving is less safe for female drivers, it's hardly a "gender-based preference" that's driving the gap. If women are spending more of their flexible hours doing housework or caring for children than men do, this may not be a "preference". While the paper does a great job identifying factors correlated with lower income, does it really indicate an individual's "preference", or does it indicate that the situation they find themselves in either enables or prevents uber participation?