4 ms·
> But the idea that they sold 11M vehicles with hardware cheating the emissions test makes his claim fairly absurd. So a CEO must know every detail about every
by msandford 11y ago
> But the idea that they sold 11M vehicles with hardware cheating the emissions test makes his claim fairly absurd.
So a CEO must know every detail about every component of every subsystem of every system in every trim level of every car of every brand that the company sells?
If that's really true, we're headed back to the early industrial revolution. Nothing can have more than a few dozen or perhaps hundred parts, and integration of say electronics and mechanical things will be rare, as it'll be the rare CEO who knows enough about both to meet your standard.
- wpietri 11y agoNo. But broadly, a CEO is responsible for everything that happens on their watch. And they are specifically responsible for designing their organizations in such a way that the organization behaves responsibly. Ignorance cannot be an excuse for bad management. If we allow that, then executives can set things up so that they always appear ignorant. E.g., imagine a CEO says, "I don't care how you dispose of that toxic waste, just make sure that it's cheap. And of course it has to be safe [wink, wink]. Just don't bother me with details [wink, wink]." Any employee will know the real message.
- nmrm2 11y agoRight, like the "GM nod". The principle that CEOs are responsible for what happens on their watch is major reason cited for their levels of compensation. You don't get it both ways.
- msandford 11y agoI completely agree! It doesn't matter if this particular CEO knew about this particular detail, because ultimately they're responsible for the culture, the work environment, etc. If they made a bad work environment at the top, and that trickled down to someone deciding to go the cheat route rather than the honest route, that's a problem. If the CEO did things right and made a good work environment but hired a bad exec, and they applied pressure to people to cheat, and this all got done, again, can be traced back to the CEO. If the CEO did the right thing and hired a seemingly great exec and that exec hired a bad manager, well, perhaps the CEO didn't do such an exceptional job. Ultimately all organizational failures like this are failures of management, and the CEO is the top manager so it's all on him (or her) when things go sideways. But just because it's all on them doesn't mean that we can all safely conclude that they definitely had firsthand personal knowledge of the problem and gave their approval, explicit, implicit or otherwise.
- wpietri 11y agoHoly moly. I had missed reports of the "GM nod" and their cultural pathologies. What an enormous mess. 95% of people's smarts taken up with performing rituals that shield them from blame. All real communication is sub-rosa, with actual words said unrelated to meaning. The detail that really kills me is that many employees thought that taking notes in meetings related to safety issues was forbidden. There was no official word of that, of course, but lots of people believed that following unofficial, gossip-routed information was the way to succeed in their careers. What lunacy. They should have been allowed to go bankrupt.
- emodendroket 11y agoHey, it worked for Steve Cohen.
- JustSomeNobody 11y ago> So a CEO must know every detail about every component of every subsystem of every system in every trim level of every car of every brand that the company sells? Of course, not. Stop trying to change the argument. He would know that VW cars are getting better scores than other cars. These are numbers that CEOs and other managerial positions live for. He would know why. Though he may not understand the technical details of why.
- msandford 11y ago> Of course, not. Stop trying to change the argument. I'm not trying to change the argument. I'm taking your statement literally, and showing that it's entirely unworkable in practice. I'm not saying he shouldn't be fired or have to step down, but to presume that he personally, definitely knew is quite the leap. I can think of several scenarios where he didn't know and is entirely innocent which are quite plausible. I also think that regardless of whether he personally had knowledge of exactly what was being done or not, he still created a work environment which rewarded the shortcut and the cheat over the right thing and needs to go. But that's an entirely different line of reasoning than "he must have known!" as a justification. > He would know that VW cars are getting better scores than other cars. These are numbers that CEOs and other managerial positions live for. He would know why. Though he may not understand the technical details of why. No, not necessarily. You're making an awful lot of assumptions and none of them are even superficially argued, much less backed up by verification. This isn't CO2 where you can draw a pretty clear line correlation between fuel efficiency and CO2 emissions. This is NO2 emissions which are highly variable and depend on a great many factors. So it's entirely possible that for the same CO2 emissions one car might output far more NO2 than another. https://www.quora.com/Why-does-diesel-exhaust-contain-more-nitrogen-dioxide-than-other-combustion-engines https://www.quora.com/Why-does-diesel-exhaust-contain-more-n... I'm very sympathetic to the notion of "if you get to claim credit for the good, then you're also going to be held responsible for the bad" argument, generally I agree with it. This is not an exceptional case. But to argue that because this is a reasonable course of action, that the person in charge of a 500k employee company must have known a particular detail of a particular subsystem? Not reasonable. Not defending the person, the company, the actions, or anything like that but I am attacking your reasoning.