5 ms·

I can tell you one thing. At least at smaller warehouses under-reporting of injuries is common. So whatever statistics are being used I'd be careful with. One

I can tell you one thing. At least at smaller warehouses under-reporting of injuries is common. So whatever statistics are being used I'd be careful with. One trick I heard about. If you can get the employee count up, you can get your reported injury rates down. Ie, two warehouses, both doing 1 injury per 10,000 packages shipped. If one has double the employee count it can make some of the reported injury rates go lower. If you ever see 5 guys watching one guy dig a hole - that employer is going to have great injury rate reporting even if the guy actually digging a hole gets injured just as often as anyone else digging a hole. Not sure if they figured out ways of mixing in management / white color jobs into the warehouse numbers. But I'm surprised no one discusses underreporting at smaller warehouses. And comparing to retail workers at walmart also seems a bit weird?

makomk 5y ago
The trouble is that there's a really obvious media bias - a report showing that Amazon warehouse injuries were 80% lower definitely wouldn't get this kind of credulous media treatment, if anyone bothered publishing it in the first place, and this has obvious effects on everyone's perceptions.
- WalterBright 5y ago
  We see that in another clickbait article today on "brutal" Amazon: https://news.ycombinator.com/item?id=27370589 https://news.ycombinator.com/item?id=27370589 I'm getting more and more skeptical of the news I read.
  - dariusj18 5y ago
    Good, remain skeptical, it's important.
jdlshore 5y ago
Related: Studies of aviation and construction have found that as incident rates decline, fatalities increase: "The underreporting that results from the implementation of such safety programs really means that you are shooting yourself in the foot. Your organization does not have a great safety culture because it has low numbers of incidents. In fact, the opposite is true. This has been shown in various industries already. A study of Finnish construction and manufacturing from 1977 to 1991, for example, showed a strong correlation between incident rate and fatalities, but reversed (r = –.82, p<0.001). In other words, the fewer incidents a construction site reported, the higher its fatality rate was (see Figure 7.1)." —The Field Guide to Understanding 'Human Error', chapter 7
- slownews45 5y ago
  I've got limited exposure to this, but we see this in some other fields. I just worked with a group that said they had no "defects". Actually, what you have is no reported defects and no clue what you are missing. The folks reporting defects in this space get a lot more hassle, but actually know what is going on in their ops.
- treesrule 5y ago
  is it worth getting the new edition of this book (its like 25$ more for me)
  - jdlshore 5y ago
    Don’t know, I read it on Safari a few weeks ago. I didn’t compare the two editions.
    - treesrule 5y ago
      Thanks
- andy_ppp 5y ago
  Of course, this is almost certainly true in medicine as well, if you can’t report mistakes you can never learn from them.
- GrantZvolsky 5y ago
  My anecdotal experience corroborates your quote: when I was 13-14 I occasionally did truck unloading/stage construction for an entertainment company. When another below-legal-age friend of mine got injured while unloading a truck, the managers begged him to make up some story in the hospital. I stopped taking stage construction gigs after another kid was killed when a ton of sound equipment crushed him. From what Amazon warehouse workers I know tell me, the safety measures are on a different level to anything I was used to when I did blue collar work. edit: I looked up one of the events we did[1] to make sure I got the age right and I was 15 when it took place. [1]: https://www.youtube.com/watch?v=bjHEMzyakrY https://www.youtube.com/watch?v=bjHEMzyakrY
PragmaticPulp 5y ago
> Not sure if they figured out ways of mixing in management / white color jobs into the warehouse numbers. I once worked as an engineer in a pseudo-office that was actually just part of the company’s warehouse that had been sectioned off and given some trendy decorations. The company was running out of warehouse space but rather than let us move into a real office, they kept us in the warehouse and rented more warehouse space next door. I could never understand why they did it, but your comment had me wondering now.
- gruez 5y ago
  or maybe warehouse space was cheaper than office space? or that the managers want easy/quick access to the warehouse to micromanage workers?
- slownews45 5y ago
  There can be other factors like tax credits for certain types of spaces / work / areas for adding employees etc that can drive some weird behavior. This can be really obvious with some govt contracting work which can have additional requirements on labor matters.
Anechoic 5y ago
And comparing to retail workers at walmart also seems a bit weird? The comparison is with Walmart warehouse workers, not retail workers. See the actual report posted by others.
gruez 5y ago
>If one has double the employee count it can make some of the reported injury rates go lower. [...] Not sure if they figured out ways of mixing in management / white color jobs into the warehouse numbers. Or just hire a bunch of part time people? Or do the numbers already adjust for FTEs?
sct202 5y ago
Except in rare edge cases, very few companies are going to burn money in excess labor to water down their injury rates like this. Workers comp insurance is not so expensive that it would make sense to double your labor force to do the same output.
stefan_ 5y ago
I don't understand. Workplace injuries are not a "shit happens" scenario - it is not good enough to consistently report high numbers and shrug your shoulders. The reporting is merely one half. You must make explicit efforts to avoid the proximal cause of these workplace injuries. When some worker at Tesla gets hit by a robot arm, nobody goes "duh watch out for the robot". There is an inquiry why it was possible for a worker to be in the work area of the robot in operation and get hit.
bryanrasmussen 5y ago
>If you ever see 5 guys watching one guy dig a hole ok first off this does not seem to be the way that any American employer would do things - I did live in the country for 20+ years and worked a lot of menial labor - never have I seen any employer have 5 guys standing around watching one guy dig a hole.
mandelbrotwurst 5y ago
Wouldn't increasing your labor costs by 400% outweigh whatever benefit you gain in reducing reported injury rates?
psds2 5y ago
If we are talking injury rate per holes dug or packages shipped then the number of people working on the task doesn't affect the injury rate.

makomk 5y ago

The trouble is that there's a really obvious media bias - a report showing that Amazon warehouse injuries were 80% lower definitely wouldn't get this kind of credulous media treatment, if anyone bothered publishing it in the first place, and this has obvious effects on everyone's perceptions.

WalterBright 5y ago
We see that in another clickbait article today on "brutal" Amazon: https://news.ycombinator.com/item?id=27370589 https://news.ycombinator.com/item?id=27370589 I'm getting more and more skeptical of the news I read.
- dariusj18 5y ago
  Good, remain skeptical, it's important.

jdlshore 5y ago

Related: Studies of aviation and construction have found that as incident rates decline, fatalities increase: "The underreporting that results from the implementation of such safety programs really means that you are shooting yourself in the foot. Your organization does not have a great safety culture because it has low numbers of incidents. In fact, the opposite is true. This has been shown in various industries already. A study of Finnish construction and manufacturing from 1977 to 1991, for example, showed a strong correlation between incident rate and fatalities, but reversed (r = –.82, p<0.001). In other words, the fewer incidents a construction site reported, the higher its fatality rate was (see Figure 7.1)." —The Field Guide to Understanding 'Human Error', chapter 7

slownews45 5y ago
I've got limited exposure to this, but we see this in some other fields. I just worked with a group that said they had no "defects". Actually, what you have is no reported defects and no clue what you are missing. The folks reporting defects in this space get a lot more hassle, but actually know what is going on in their ops.
treesrule 5y ago
is it worth getting the new edition of this book (its like 25$ more for me)
- jdlshore 5y ago
  Don’t know, I read it on Safari a few weeks ago. I didn’t compare the two editions.
  - treesrule 5y ago
    Thanks
andy_ppp 5y ago
Of course, this is almost certainly true in medicine as well, if you can’t report mistakes you can never learn from them.
GrantZvolsky 5y ago
My anecdotal experience corroborates your quote: when I was 13-14 I occasionally did truck unloading/stage construction for an entertainment company. When another below-legal-age friend of mine got injured while unloading a truck, the managers begged him to make up some story in the hospital. I stopped taking stage construction gigs after another kid was killed when a ton of sound equipment crushed him. From what Amazon warehouse workers I know tell me, the safety measures are on a different level to anything I was used to when I did blue collar work. edit: I looked up one of the events we did[1] to make sure I got the age right and I was 15 when it took place. [1]: https://www.youtube.com/watch?v=bjHEMzyakrY https://www.youtube.com/watch?v=bjHEMzyakrY

PragmaticPulp 5y ago

> Not sure if they figured out ways of mixing in management / white color jobs into the warehouse numbers. I once worked as an engineer in a pseudo-office that was actually just part of the company’s warehouse that had been sectioned off and given some trendy decorations. The company was running out of warehouse space but rather than let us move into a real office, they kept us in the warehouse and rented more warehouse space next door. I could never understand why they did it, but your comment had me wondering now.

gruez 5y ago
or maybe warehouse space was cheaper than office space? or that the managers want easy/quick access to the warehouse to micromanage workers?
slownews45 5y ago
There can be other factors like tax credits for certain types of spaces / work / areas for adding employees etc that can drive some weird behavior. This can be really obvious with some govt contracting work which can have additional requirements on labor matters.

Anechoic 5y ago

And comparing to retail workers at walmart also seems a bit weird? The comparison is with Walmart warehouse workers, not retail workers. See the actual report posted by others.

gruez 5y ago

>If one has double the employee count it can make some of the reported injury rates go lower. [...] Not sure if they figured out ways of mixing in management / white color jobs into the warehouse numbers. Or just hire a bunch of part time people? Or do the numbers already adjust for FTEs?

sct202 5y ago

Except in rare edge cases, very few companies are going to burn money in excess labor to water down their injury rates like this. Workers comp insurance is not so expensive that it would make sense to double your labor force to do the same output.

stefan_ 5y ago

I don't understand. Workplace injuries are not a "shit happens" scenario - it is not good enough to consistently report high numbers and shrug your shoulders. The reporting is merely one half. You must make explicit efforts to avoid the proximal cause of these workplace injuries. When some worker at Tesla gets hit by a robot arm, nobody goes "duh watch out for the robot". There is an inquiry why it was possible for a worker to be in the work area of the robot in operation and get hit.

bryanrasmussen 5y ago

>If you ever see 5 guys watching one guy dig a hole ok first off this does not seem to be the way that any American employer would do things - I did live in the country for 20+ years and worked a lot of menial labor - never have I seen any employer have 5 guys standing around watching one guy dig a hole.

mandelbrotwurst 5y ago

Wouldn't increasing your labor costs by 400% outweigh whatever benefit you gain in reducing reported injury rates?

psds2 5y ago

If we are talking injury rate per holes dug or packages shipped then the number of people working on the task doesn't affect the injury rate.