What Would a Utilitarian do with Volkswagen?


Vikram Bath

Vikram Bath is the pseudonym of a former business school professor living in the United States with his wife, daughter, and dog. (Dog pictured.) His current interests include amateur philosophy of science, business, and economics. Tweet at him at @vikrambath1.

Related Post Roulette

133 Responses

  1. Avatar Morat20 says:

    Actually, I would advocate jail time. Many of those who made this call are not US citizens, but even simply barring them from the US due to potential for arrest would undoubtedly cut into their business and personal lives.

    The potential monetary losses (stock loses, regulatory fines, legal losses due to a variety of potential lawsuits, RICO cases, etc) are pretty high, but I think one thing we’re having problem with is that while corporations might be fined, the individuals who made the call rarely pay a price. And if they do, that price generally does not really cut into their income or wealth.

    They feel no sting, no real loss. The corporation might go out of business, to be bought by someone else, but those who flouted the law generally land happily on their feet.

    I suspect that VW might have acted quite differently if those who made the call thought they might end up in jail if caught, rather than their company end up fined or sued.Report

    • Avatar Chris says:

      Punishing both the company and the individuals involved seems sufficient, though how do we determine who made the decisions? Is Volkswagon likely to be forthcoming about such things?

      Not that I don’t think the legal process could eventually tease it out, but by the time that happens, Volkswagon will undoubtedly have moved on, and it will sure look like they got away with it, even if some execs who’ve moved on face some minor legal difficulties that they can afford to drag out for extended periods of time.Report

      • Avatar Saul Degraw says:

        VW is owned and seems to be controlled by one family according to an article on Slate. Go after that family.Report

      • Avatar nevermoor says:

        If DOJ wants hard enough, DOJ gets access.Report

      • Avatar Morat20 says:

        The same way you do with ANY crime. You investigate. You dispose, subpoena, and otherwise dig so far into their corporate laundry that you know every player involved — or at worst, every player who tried to hide, wipe, delete, or shred the data.

        You use RICO and you flip the peons to go after the big fish. Bob from engineering was just doing his job — promise him immunity to go after his managers. You wrap everyone involved up with this.

        No criminal is EVER forthcoming about his or her crimes. I’m pretty sure we have, you know, people to investigate white collar crime. I just think we need to shake the deep belief that the appropriate penalty for corporate crime is corporate fines, rather than holding the corporate decision makers responsible for malfeasance.Report

        • Avatar Barry says:

          I’m seconding Morat – prosecutors have no problem getting really, really hard when the top defendants are not corporate laywers.

          I would bet that enough of the mid-level guys involved in this kept some evidence hidden just in case – ‘persuade’ them to turn it over, and use it.

          And note that even an unsuccessful prosecution is good, because it’ll put some fear into some of these people. If you knew that your fellow Board of Directors member Joe spent a year not sleeping, in serious fear of going to prison for what he did as CEO of his company, you might have second thoughts yourself.Report

      • Avatar Troublesome Frog says:

        Punishing both the company and the individuals involved seems sufficient, though how do we determine who made the decisions? Is Volkswagon likely to be forthcoming about such things?

        This is automotive industry embedded code, so it’s very likely that programmers can’t take a coffee break without filling out reams of paperwork and getting written approval from a half dozen people. Unless there’s some exception for the emissions control code, there’s no way for anybody to check in source code changes without record of what the change was, who committed it, who approved it, and what requirement or defect that approval traces back to.

        I think Volkswagen would have a very hard time hiding those types of records from determined prosecutors. Things will only get interesting once the engineering manager(s) who approved the changes in writing try to implicate their superiors for verbally approving them.Report

        • Avatar Barry says:

          “This is automotive industry embedded code, so it’s very likely that programmers can’t take a coffee break without filling out reams of paperwork and getting written approval from a half dozen people. Unless there’s some exception for the emissions control code, there’s no way for anybody to check in source code changes without record of what the change was, who committed it, who approved it, and what requirement or defect that approval traces back to.”

          Also, remember that in effect VW was claiming to have solved a major problem with an easy software fix. This is the sort of thing that upper management wants rolled out across all engines.Report

    • Avatar Snarky McSnarksnark says:

      Seconded, in all respects.Report

  2. Avatar Burt Likko says:

    There must surely be a third option: fine the hell out of them and subject future VW products to special scrutiny to conform to the same emissions standards as every other manufacturer for several years.Report

    • Avatar Tod Kelly says:

      The problem with this route, at least historically in the US over the past several decades, is that the fines and mandates are handed down, and then there is a legal appeal, and a few years later when no one cares any more the fine is essentially wiped clean (or at least settled for a not very punitive amount) and the mandates are largely waived.Report

      • Avatar Burt Likko says:

        If this is the case, it isn’t because there aren’t already-heavy fines and authority for heightened scrutiny in the law: it’s because the legal system gives way from imposing the maximum penalties to something the guilty party can tolerate.

        So, I blame the lawyers.Report

      • Avatar Patrick says:


        One workaround is to levy the fine, but also I think we should make it so that anyone who bought a TDI can drive it to a certified VW dealer, walk in with a copy of the title and the keys, and get back the full purchase price of their car.

        The people directly affected get all of their money back, VW gets tens of thousands in real damages, the cars are off the road.Report

        • Avatar Burt Likko says:

          What to do with the vehicles themselves, then? VW can export them and sell them in a market with emissions standards less demanding than those of the US?Report

          • Avatar oscargordon says:

            Cash for clunkers! What was the fate of every car?Report

            • Avatar Burt Likko says:

              Yeah, but the clunkers were still street-legal because the emissions standards are time-staggered (“grandfathered,” if you will). As @stillwater points out below (above? I can’t tell from where I’m writing), any individual car might, or might not, be street legal, and right now, no one knows with respect to any particular car or what percentage might be expected.

              Also, I forget: are we calling “cash for clunkers” a success or a failure?Report

              • Avatar Oscar Gordon says:

                Oh I’m in the failure category. I wasn’t suggesting junking all the cars was the right idea, only showing that we’ve already set that as an option.Report

          • Avatar DavidTC says:

            Why would we let VW export them?Report

            • Avatar Patrick says:

              They can export them after they fix them.

              If the destination countries undo the fix to get better gas mileage, well… they could do that anyway. There is only so much we can do when we’re talking international goods, here.Report

        • Avatar DavidTC says:

          That is exactly what I was thinking: Make VW buy back every single car.Report

  3. Avatar Saul Degraw says:

    VW Is certainly going to face the wrath of the trial lawyers and the governments of several counties. This is a world wide problem, no? I’ve read that this can be one of the most expensive recalls in any corporation’s history. It makes me wonder if VW could go bankrupt just like many asbestos manufacturers went bankrupt over their liability issues in the 1980s.Report

  4. Avatar Saul Degraw says:

    On a constitutional level, I wonder if revoking the right of VW to do business in the USA would raise Bill of Attainder issues.Report

    • Avatar LeeEsq says:

      Its an interesting question. The commerce clause gives Congress the right to regulate foreign commerce. This power has been usually interpreted broadly. I can’t recall any instance of congress forbidding one particular foreign company from doing business in the United States. If they have plants and other property in the United States, it could easily be considered a regulatory taking. There are probably procedural due process issues involved.Report

    • Avatar DavidTC says:

      No, but we could give them a fine that is literally impossible to pay.

      I know some people think the max fine is $37,500 per car, for a total of $18 billion, and that, in theory, is affordable. But that’s just the fine for using a *defeat device*…people seem to have failed to notice there’s also the fact that they, you know, sold a lot of cars that did not meet emission guidelines, which is *also* subject to fines, and they also put a lot of NOX in the air, again, subject to fines, and they lied to consumers, again, subject to fines, and lied to the EPA, again, fines…etc, etc…

      If the US government wish to, it basically could impose a fine that is so large that VW doesn’t have any option but to walk away from all their property in the US and not attempt to import any cars due to the fact we’d immediately seize them to cover their fine.

      But wait, you ask, couldn’t they sell cars to a third party that would then sell them here? Well, no, not if we seize any test cars before they can run them through the EPA tests, or seize their testing application fees and put them towards the judgement instead of the application.Report

      • Avatar scott the mediocre says:

        I think you will find that VW of America (however the corporation is structured) does not actually own much property in the US, particularly since their US market manufacturing left the US (for Mexico, mostly) long ago (as opposed to for example Toyota, Honda, etc.). Remember that most of the inventory and capital assets are owned by the dealers, who are very much victims in this, unless you think they somehow knew about the cheats, although they did at least hypothetically make some profits from selling the cheatin’ diesels.Report

        • Avatar Saul Degraw says:

          They can be plaintiffs as well!Report

          • Avatar Stillwater says:

            “Sue everybody!”


          • Avatar scott the mediocre says:

            Saul, I’m quite sure they will be. One of the things I almost sort of learned from 2009 and the Administration’s efforts to help GM in particular close dealerships is that there is a truly bewildering variety of state laws relating to auto dealerships – it’s a highly regulated business, not just between dealer and consumer but between dealer and manufacturer/importer too (most of the specific cases I recall had to do with GM dealerships; I dunno how different things are when (assuming) the regulated relationship is between dealer and VW of America.Report

        • Avatar DavidTC says:

          I think you will find that VW of America (however the corporation is structured) does not actually own much property in the US, particularly since their US market manufacturing left the US (for Mexico, mostly) long ago (as opposed to for example Toyota, Honda, etc.).

          The hypothetical point here isn’t actually to recover the money.

          It was asked if we could stop VW from selling cars here, and the point of what I said is to do that: We give them a fine they can’t pay, a fine that is bigger than their entire American market for the next two decades. And thus, with that unpaid fine hanging over their head, every car they *ship* here, to a dealer, can be seized before it can *reach* the dealer.

          The dealers, of course, would be free to sell their existing cars, except for the cheating ones.

          And, just as importantly, we seize VW’s test cars, so they can’t actually pass EPA emissions testing, and thus can’t legally sell any new models here anyway. So even some sort of workaround of VW selling cars to dealers *before* those cars touch US soil doesn’t really help, although I guess it means the dealers could keep getting shipments of 2016 cars. (And, hilariously, the laws that dealers created to stop competition means that VW probably *couldn’t* sell cat to a third party and have their third party sell them to dealers.)Report

  5. Avatar Tod Kelly says:

    I have heard a lot about this lately and there’s on important question that no one who reports on it (at least that I have seen) either asks or answers. I am hoping someone here will know.

    What was in it for Volkswagen to have their cars do this?

    If the software in the car was able to make the emissions system do what the law said it needed to be done, what was the point of not having it just do that all the time? Is the Volkswagen headed up by some villain from the Captain Planet cartoon?

    I get how greed gets tempts people into breaking the law to bypass a lot of regulations, but not being a car guy or an engineer I cannot for the life of me see what Volkswagen thought the upside to doing this would be.

    Anyone know?Report

    • Avatar LeeEsq says:

      Volkswagen probably thought that this was a lot cheaper than having to meet emission standards. If emission standards grow tougher because of concerns over climate change or other environmental and public health reasons, than Volkswagen might have decided it was for increasing costs and decreasing profits. A computer trick might have been seen as cheaper.Report

      • Avatar Tod Kelly says:

        Why would it have been cheaper? They put the systems that would make in legal in all of the cars. In fact, they developed a system that would do it two different ways. How is that cheaper than just making a system that only does it one of those two ways?Report

        • Avatar Oscar Gordon says:

          Because hardware costs money. Software is a simple decision tree:

          if (dynoDetected())

        • Avatar Fred says:

          I think it is about performance and fuel economy. The emissions controls reduce engine power. Buyers want more power so there is a competitive advantage to selling a car with a reduced emissions control.

          Edit: Now I see Stillwater has answered the question with more clarity than I am capable of.Report

      • Avatar Barry says:

        “Volkswagen probably thought that this was a lot cheaper than having to meet emission standards. ”

        And the executives making the decisions figured that they’d have moved on, and be more likely to be firing their replacement for being in charge of that department than being fired themselves.

        The top management figured that it’s not like megacorps are ever harshly punished for such things (or worse, that the top executives do time), so the risk-reward calculus favored doing that.Report

        • Avatar Stillwater says:

          Not to disagree with anything you wrote, Barry, but from what I’ve been able to figure out VW chose the cheater chip not because it was the cheapest solution to the problem, but rather because the problem could not be fixed. They couldn’t engineer a car hitting the performance spec AND the clean diesel spec. So they deceived people into thinking they had.

          One thing I read on the nerd forums about this very issue was why, given all the R&D funding from major car companies, no one else could come close to meeting the “clean diesel” standard. Which gets into something (I think) TFrog wrote about utpthread: that the code installed on the computers for these types of vehicles (well, all vehicles, I’m led to believe) is pretty much where the game is at these days, and that stuff is guarded by the fullest measures consistent with the law.Report

    • Avatar Stillwater says:

      what was the point of not having it just do that all the time?

      Performance. The little TDI cars are famous for getting high forties/low fifties MsPG, even while getting pinned to the seat when you hit the high part of the torque curve.Report

      • Avatar Stillwater says:

        And to answer your next question: the best story I heard about this is that VW was deep into the new TDI engine build when the new EPA specs hit and they couldn’t re-engieer the existing engine to meet spec. So, cheater chip. The alternative, which they used in a few of the models (surprisingly, actually) would be to install the urea injectors (etc) to all their passenger cars.Report

        • Avatar Morat20 says:

          If that was the case, then why the chips? Nobody goes through that much effort to develop and install something that risky if there’s no payoff.Report

          • Avatar Stillwater says:

            The price of a urea injection system, for one. The cost of reengineering – during production! – the engine compartment to accommodate the urea system for seconders. The politics of “clean diesel” for three.

            That’s what I’ve read on forums full of TDI NERDS anyway.Report

            • Avatar trizzlor says:

              @stillwater “The politics of “clean diesel” for three.”

              I think this is a big point that gets lost when going from the car forums to the political forums. The few auto world articles on this that I’ve read think it could be the end of consumer diesel in America, which means VW was having a lot more impact than just their TDIs. If you can sell the idea that diesel cars are clean, cheap, and efficient – even in a small number of models – that idea can motivate the whole consumer diesel market, spur innovation, bring new players in, etc. And VW was going full Don Draper on CleanDiesel. Just like all those academic scandals, my guess is that VW was hoping that cheating today was a temporary stop-gap until the technology caught up with the regulation requirements tomorrow. It’s doubly ironic that they got busted because a consumer advocacy group was so excited about diesel cars they decided to do some road demos.Report

    • Avatar nevermoor says:

      My understanding is that their system could not deliver the required level of emissions while performing like a car you would want to drive. This is because it didn’t have a sub-system other diesel systems have that injects uric acid.

      The upside is that they were able to sell cars without this system for quite some time.Report

    • Avatar Burt Likko says:

      The video Vikram embedded describes the issue as one of performance, rather than principally one of economy. The vehicles can comply with emissions laws — but the engines don’t perform as well. Translated: when the car is in “compliance mode”, it would drive like it doesn’t have any guts. Doesn’t need them: when it’s on the rack, actual acceleration doesn’t matter because the car is stationary. When the car knows it’s out on the open road, it gives the driver some pick-up-and-go when the gas pedal is pressed.Report

      • Avatar Stillwater says:

        Exactly,. As I wrote below, tho, our 2012 TDI was tested without the cheater chip and still passed spec (by a lot). Which makes me wonder something: my best guess (from limited evidence to be sure) is that a very high percentage (95+) of these cars will pass emissions if they’re tested without the cheater chip under conditions which apply at actual emissions facilities. What to do then?Report

        • Avatar Burt Likko says:

          Seems like then there’s no harm to that specific consumer, or so it would seem: the vehicle complies with the law even though the “cheater chip” is there. That vehicle should re-sell for the same amount as projected, since it doesn’t need a performance modification to be legal. No harm to the individual consumer.

          Thus do we get into the minutiae of how the legal profession chisels away the economic incentives that statutes and regulations are intended to create! Again, I blame the lawyers.Report

          • Avatar Stillwater says:

            See, I love the lawyers!

            One other question regarding the EPA testing regime: IF it’s the case that these vehicles pass emissions inspections without the cheater chip, can the EPA STILL claim that the cars “fail to meet inspection requirements”? That is, are the EPA specs based on emissions at the testing facility, or emissions in real world driving conditions? (Cuz that’s how this whole story broke, actually: people hooking up emissions equipment to TDIs while they were driving around town.)Report

            • Avatar oscargordon says:


              I can’t recall, do you live in CA? If not, that might be why you passed.Report

              • Avatar nevermoor says:

                Bye Bye nationwide class action! Each car must be individually tested without the cheater chip! No predominance!Report

              • Avatar Stillwater says:

                I think the requirements that these engines must meet are EPA mandated, so I’m under the impression that they’re national (???). In any event, the test was performed in CO.

                ALso, as yet there’s only been one test about this stuff done in the US (outa Virginia, I think) and I believe it was limited to ’14 and ’15 models. (FWTW.)Report

              • Avatar Oscar Gordon says:

                Emissions are state by state. EPA sets the baseline, states can accept that or dial it up.Report

              • Avatar Stillwater says:

                Good. Given what you wrote (that the EPA sets the lowest level operative in any state), the complaint against VW is that they fail to meet the EPA level. So CA VW drivers might be in a different shitstorm situation than we are here in CO. I don’t know, however, if the CO limit is the same as or different from the EPA limit.Report

              • Avatar aarondavid says:

                Well, most car manu’s work on the 49 state model, in other words one car model will be sold everywhere but CA, and then others will be modified to pass CA’s more stringent emission standards. Externally the cars are no different, but the emissions systems would be set differently. Here is a run down from the CA DMV. At this point, that would be covered by the engine management chip for the on board computer. Very easy fix.Report

              • Avatar Michael Cain says:

                IIRC, what’s actually in the Clean Air Act is that the states can adopt the EPA standard, or they can adopt the tougher California standard, but that’s all. States can’t make up something else that’s tougher than California, or somewhere between the EPA and California, unless Congress gives explicit permission.Report

      • Avatar nevermoor says:

        Not sure how that disagrees with me. You can’t sell a car with no performance (at least as well as VW sold its cars).

        The way other companies solved that problem was with the injection material. The way VW solved it was the cheater chip.Report

    • Avatar Jaybird says:

      It was pointed out that Germany is likely to lose more money from this incident than from the Greek thing.

      That’s one hell of a gamble, right there.Report

    • Avatar Michael Drew says:

      I was wondering this myself, and, beyond not really having anywhere to ask it, I also assumed it was one of those rare truly-too-stupid-to-ask questions that I was just being singularly slow on. So I’m grateful to Todd for putting it out there. I can’t say I really understand any of the answers that had been offered, either.Report

      • Avatar Oscar Gordon says:

        To understand why you need to understand that diesel engines are not gasoline engines. A car with a diesel 2 liter 4 cylinder engine will, without any modification, be slower to accelerate than the same gasoline engine. You can fix that by using a turbocharger, but that kills the fuel economy. Emissions controls further limit acceleration.Report

        • Avatar Michael Drew says:

          Since the whole story was limited to the diesel line (in a sense VW caught yooge break there, whoa), right there I was sort of guessing the answer was over my immediate head (not my ability to understand, but beyond my then-current knowledge), and likely more fine-grained than was going to be covered in stories covered from a business angle rather than a technical angle. So thanks for that.

          So that’s the answer to why not make it conform all the time rather than just in tests, Tod: if they did, it would kill the acceleration of the engine all the time, rather than just in emissions tests (when no on cares about acceleration anyway). So it would be a car with low highway punch.Report

          • Avatar Michael Drew says:

            …So then the follow-up: how were they able to get away with it at all, for even a week or a day? Wouldn’t other manufacturers look at the marketing claims re acceleration and emissions and say, “Uh, nope. You can only get there in diesel by cheating the test.” Or is that exactly what happened? Or is VW’s engineering so supposedly superior that other makers were just saying, “Damn. Wish we could do that.”?Report

            • Avatar Oscar Gordon says:

              The more I think about it the more I’m betting it was an open secret, or widely held assumption that they were cheating somehow, but no onewas sure how nor interested enough to find out (or keeping quiet because they’re cheating too & don’t want draw attention) .Report

              • Avatar Michael Drew says:

                Also: is anyone really competing that hard with them in the American diesel space anyway? Daimler-Chrysler via Mercedes? Other major diesel players in the states? Maybe it was simply a straightforward German agreement to play that way in America.Report

              • Avatar DavidTC says:

                That is basically what my brother says, that clean diesel (Or at least clean diesel with any sort of performance) is a lie, and everyone’s cheating in different ways.

                Although probably not as badly as VW…everyone else at least admits they need urea systems. But you put the same device on the back of any clean diesel vehicle and drive it around, and you’d get results wildly different than the lab results.

                You don’t even have to use a ‘defeat device’. Give the car some extra power on an incline, or on turns, or after long accelerations. Emissions tests don’t test that.Report

              • Avatar Oscar Gordon says:

                So back in my Navy days, I was a turbine tech. My turbines, being turbines, could burn all sorts of stuff, but we primarily fed the something in the diesel family, be it kerosene, JP-4, JP-7, or bunker fuel (diesel fuel – marine). When at ACU-5, they got one of the JPs. When deployed, they generally got bunker unless we needed to refuel from the LHA (helo carrier), then we got a JP. Every once in a while we’d load up some kerosene if that was what we could get (usually from a source on a remote shore).

                In order of refinement, kerosene is damn near as pure as you can get (it’s clear), the JPs are a light beer, and bunker looks like Guinness, it has so much crap in it. Guess which fuel had me cleaning the turbines & changing fuel filters more often?

                If we really want “Clean Diesel”, we should just stop playing around and configure everything to burn something on the order of a JP.Report

            • Avatar Stillwater says:

              …So then the follow-up: how were they able to get away with it at all, for even a week or a day?

              I was looking into that as well, and I think (admittedly an incomplete bit of research) that the EPA pretty much relied on the emission testing program to determine compliance with the regs. Which makes Oscar’s comment just above probably even more likely.Report

          • Avatar Oscar Gordon says:

            Or it would have they highway punch without the fuel economy that makes diesels attractive.

            Basically it’s marketing hype running up against engineering realities.Report

      • Avatar Michael Drew says:

        …Also, I got to Oscar’s comment/joke about the computer program above and apparently left off, then came back thinking I had read all the responses to Tod. But I clearly hadn’t. So sorry to rehearse this again. Everything after that point answered the question clear as day, I just didn’t see it.Report

    • Avatar Snarky McSnarksnark says:

      Because, when the cars are tuned to meet emissions standards, the fuel mileage drops precipitously. As does the performance (diesel engines, are not by nature, fast-accelerating). Their big business opportunity came with the high gas prices of 2007. The TDI had the highest MPG ratings of any car sold in America, and was, by all measures, a larger, more fun-to-drive car than the other cars in its mileage class. The fuel efficiency of the TDI Golfs / Rabbits was the primary thing that made them exceptional.

      Coupled with a 10-year drive to make Volkswagen the biggest car company on earth, the company made a all-out push to gain world market share. And largely succeeded–it came close to doubling its unit sales between 1998 and 2014.

      I hope they fry.Report

  6. Avatar Stillwater says:

    Couple of random thought:

    1. My wife and I own an aught 12 Golf TDI and had an emissions test performed after the scandal broke. THe guy tested it once with the cheater chip operative and once with it disabled, resulting in readings that were far below the maximum allowable each time (tho they were different). Given that, and the limited testing done on vehicles scooped up in the suit (which appears, right now, to be everything post-tighter EPA regs) I wonder how big a problem this will actually be down the road.

    2. I’m curious as to what VW will propose as a fix for this problem. I’d be surprised (like, VERY!) if it goes beyond replacing the cheater chip with the “EPA spec” override chip, which will reduce performance in those vehicles.

    3. I’m increasingly frustrated by our overly strict diesel car emissions requirements given that diesel light trucks (including pickups) can and do legally spew that shit, and I’m especially irritated when I read about the VW TDI passenger vehicles in other countries which get 80 mpg and up. (Not to mention their and other manufacturers smaller diesel engines for pickups and SUVs). Errrgggghhhhh.

    4. Regarding fines and such, it seems to me that putting extra effort into cheating should entail even stiffer fines, but like you say, this “cheat” took no effort at all. And without excusing VW’s behavior, I’d like to see the extent to which other car companies have engaged in this, myself. I think it would be informative. Also, just to be clear, I think Federal penalties are appropriate, but I also think VW is gonna get hit a lot harder by class action suits.Report

    • Avatar Burt Likko says:

      I also think VW is gonna get hit a lot harder by class action suits.

      How would you articulate a typical plaintiff’s damage claim? Loss of resale value?Report

      • Avatar Stillwater says:

        From what I’ve read it’s loss of resale value as well as (something like) false advertising.Report

        • Avatar Stillwater says:


          SEATTLE – National consumer-rights law firm, Hagens Berman, filed a class-action lawsuit against Volkswagen’s U.S. manufacturer and distributor, and is continuing to expand its national suit alleging that the automaker deceived consumers by using software to cheat emissions tests, allowing its “CleanDiesel” vehicles to emit nitrogen oxides (NOx) at levels 40 times higher than legal limits during normal use.Report

      • Avatar Morat20 says:

        That and wouldn’t there be a case for fraud?

        I mean the vehicles were sold as meeting EPA requirements and clearly did not. And now it’s entirely likely that they will not be allowed to pass their next inspection (and thus be road legal) if and until VW performs a fix.Report

        • Avatar Doctor Jay says:

          I endorse this. This seems like a massive case of fraud.Report

        • Avatar Burt Likko says:

          That’s the thing I’ve been wondering about. An element of fraud is “detrimental reliance,” so what’s the detriment? The vehicle functions. A generalized harm to the environment is not the same thing as harm to the individual consumer. Pretty much the government gets to sue for generalized harm to the environment.

          That’s why I was thinking it might be diminishment in resale value — the fix, when it’s eventually devised, will apparently diminish vehicle performance because the way the engine is designed, there’s less acceleration when the software complies with emissions standards, and therefore lower the resale value.Report

          • Avatar nevermoor says:

            PAGA claim on behalf of Californians whose life expectancy has been reduced by the pollution!

            17200 claim by anyone engaged in the sale of any other brand of car!

            The resale issue is also certainly valid.Report

          • Avatar Morat20 says:

            No it doesn’t. The vehicle must function as a street legal vehicle. It was sold as one, not as a decorative vehicle for your garage.

            Failing to meet EPA standards means it is not street legal, therefore it is not functional. Selling me a watch that keeps perfect time, if there are 65 seconds in a minute, is not a functional watch no matter how precise and reliable it is.

            Now, upon a recall wherein VW fixes them to meet emissions standards — then the case would shift towards fraudulent performance claims (if they merely disable the device and thus lose performance) and resale loss, which is why the “add a urea system” is considered pretty likely.

            Even those it will cost thousands per car, what will result is a car that is street legal with the same performance and same resale value (roughly), and the only thing the purchaser is left with a need for a lifetime supply of Clean Blue or whatever they call that urea stuff, which is pretty cheap.Report

            • Avatar Stillwater says:

              which is why the “add a urea system” is considered pretty likely.

              Putting aside issues of being able to actually install the system, it seems to me VW has two choices as far as this issue goes: install the urea system at great cost (I’ve heard it’s about 6 grand to retro-fit) to maintain performance, or install the override chip fulltime (at prolly just a few hundred bucks) and pay people cash monies for lost performance. I really have a hard time believing their gonna go with the retro-fit, myself, but we’ll see.

              Problem is, either way, they’re gonna get hit with class action suits that they’ll surely try to settle outa court, or perhaps even get out front of by offering TDI owners hush money up front. So they’re gonna be on the hook for cash no matter what, it seems to me.

              But back to my earlier point, I’m really curious if the EPA regs are based on the type of test done at emissions facilities or not. For some reason, I can see that being a weak spot in all this “fine em!” business. That is, if the vehicles pass emissions without the chip, and if the EPA spec only requires that the vehicles pass emissions in the highly controlled envirnonment of a testing facility, then what happens to the Gummints case? Does anyone know? Bueller?Report

              • Avatar Stillwater says:

                Regarding that last paragraph:

                More simply put, I haven’t read anywhere (ANYWHERE!) that VW vehicles without the chip fail standard, controlled emissions tests (that information might be included in the University of Virginian study which broke this story open, btw, but I haven’t read about it). What I hAVE read is that the real world emissions are radically different than the controlled-test emissions.

                So I’m just curious about how the EPA determines it’s baseline on emissions requirements.Report

              • Avatar Stillwater says:

                Well, after looking into this a bit more, it appears that the EPA probably requires test data for real world conditions but gets lazy and defaults to the using the emissions … uh …. emitted {clears throat} during the controlled test in a facility as it’s requirement. Just a guess, given the lack of anything definitive. So I haven’t figured out if the EPA actually requires emissions standards to apply to in REAL real world driving conditions or the faux real world conditions created in the lab. Alsotoo, all you CAians, they do roadside tests AND take roadside pictures of emissions!!Report

            • Avatar Burt Likko says:

              Well, yeah, I see that gets us to the mandatory repurchase. But only for the vehicles that don’t actually pass a practical road test. (VW pays for the vehicle-by-vehicle road tests, of course.)

              If a vehicle passes a practical road test notwithstanding the cheat chip, then it is a street legal vehicle because it actually emits fewer emissions than the legal limit, and the only problem is the code embedded in the cheat chip, which VW ought to be able to figure out a fix for and implement in what’s become standard recall procedure.

              In exchange for which, the government doesn’t do what @vikram-bath suggested in the OP. Seems reasonable to me.Report

      • Avatar Francis says:

        How much would VW have charged for a car with the performance characteristics that the car has once it’s street legal.

        ETA: I don’t think you can limit it to resale impacts. An owner can drive his car until it rusts, but he’s still not getting what he paid for. There’s no way that these cars will be allowed to stay on the streets. Major fixes, likely having significant impacts on performance, are coming. So the owners have been defrauded by the difference between what they actually paid and what they would have paid to buy the car with the new characteristics.Report

        • Avatar Stillwater says:

          How much would VW have charged for a car with the performance characteristics that the car has once it’s street legal.

          From what I’ve gathered on the nerd forums (god I love those guys) the argument is very close to what you’ve articulated up there, coupled with the idea that VW TDI buyers paid a premium price to get high performance coupled with clean diesel tech. So, the argument goes, they clearly wouldn’t have paid that much had they known the facts.Report

      • Avatar Saul Degraw says:

        Loss of Resale Value, Cost of Replacement, Maybe you need to rent a car for a while during the fix, another resale.

        There is also pure fraud and deception. VW marketed their car as having a feature that it did not. Wouldn’t plaintiffs have bought another car if they knew the truth.Report

        • Avatar Burt Likko says:

          Are you talking about tort damages?

          “Oh! I stayed awake at night, dozens of nights, literally shaking in anger and unable to sleep, because I knew Volkswagen had screwed me over and sold out the atmosphere! I lost weight! I couldn’t have sex with my spouse; I was so depressed by the whole thing my libido was depressed! Here’s all the bills from my pshrink and my therapist and my spiritual counselor who helped me get through that difficult time. Strangely, my insurance company hasn’t paid them all yet, something about a ‘coverage review.'”

          Ugh, let’s spare the legal system that disgrace. I think that the consumer claims, especially the ones arising under California law, should be just fine to make the individual buyers whole.Report

        • Avatar nevermoor says:

          How do you certify those classes?Report

          • Avatar Saul Degraw says:

            By state and claim. There are already a lot of lawyers with claims out there and many of them are much more experience. I know a class action was filed in Vermont. My guess is that there are class actions in every state at this point.

            Or you can do it through the MDL like a pharma or antitrust case.Report

            • Avatar nevermoor says:

              I meant, how do you define a non-individualized class on these theories.

              Rental cars I guess, but you probably have a numerosity problem there. The others? Write me the class definition and we’ll see.Report

    • Avatar DavidTC says:

      Wait, I don’t understand how the ‘cheater chip’ can be turned off, or why people are calling it a ‘cheater chip’?

      The computer in the car had a bunch of code. It would activate one set of rules when the car was being emissions tested, and another set when it wasn’t. It’s not a ‘chip’….I mean, it is one, but without it, the car doesn’t work at all. Talking about turning it off is like talking about turning off a computer’s ‘Microsoft Office chip’…that’s not how computers work, outside of 1980 science fiction.

      Have testing places figured out how make the car thing it’s *not* being emissions tested (I.e., figure out the rules for the test) and test it that way?Report

      • Avatar Stillwater says:

        It’s a sensor that goes to the back wheels. Unplug the sensor, and you’ve unplugged the “trip” that kicks in the cheater chip.Report

        • Avatar DavidTC says:

          Ah, okay, so it *is* that easy to disable.

          Man, seriously, if I were writing code to trigger an illegal emissions system, I’d make it a bit less obvious than that, and err towards ’emissions test’. (Rather have some people accidentally occasionally have mushy acceleration than accidentally have it fail the test.)

          I’d say it’s only not an emissions test if the back wheels are moving *and* the steering has moved back and forth some *and* if nothing is plugged into the computer *and* if the car has changed speeds a few times.

          But who knows how much space their code had to fit in, especially when they were putting two entirely different rulesets into space designed (Or had to look like it was designed) for one.Report

      • Avatar Stillwater says:

        Oh, and why they call it a “cheater”… There were two types of mixes programmed into the computer: one when the sensor detects all the wheels moving, one when it finds that only the front wheels are moving.Report

        • Avatar DavidTC says:

          I wasn’t asking why people were calling it a cheater, I was asking why people were acting like there was some independent ‘cheater chip’ installed, instead of what was actually a cheater *subroutine* inside the ECU computer. (And if you disabled *that* computer, the car simply wouldn’t work.)

          People were talking like this was some hardware thing, which could be physically bypassed. It’s a software thing. (Although, apparently, you can easily make it not think it’s not an emissions test via a hardware sensor.)Report

          • Avatar Michael Cain says:

            People were talking like this was some hardware thing…

            The Clean Air Act makes it illegal to include a “defeat device”. Back in the 1970s, when the auto makers first cheated on the emissions tests, they were indeed putting in extra hardware that did the deed. Today, of course, those companies all follow one of Cain’s Laws™ that says “To the extent the budget and project requirements will allow, put the tricky parts in software.”Report

          • Avatar Brandon Berg says:

            Can’t they just update the firmware?Report

            • Avatar Brandon Berg says:

              I just realized that I don’t know whether firmware is called that because it’s somewhere in between software and hardware, or because it’s made by the firm that makes the hardware.Report

              • Avatar Morat20 says:

                Firmware is software held in non-volatile memory — that is, power loss won’t lose it. Some (not all) can be altered, generally in a PITA fashion (you reflash it or replace the ROM outright).

                It’s not hardware because it’s not using circuits to create logic paths, but it’s not considered software because it’s non-volatile and possible unalterable entirely.

                Much of a car’s software is really firmware — all the bits that make the car “go” are firmware, for certain.Report

              • Avatar Brandon Berg says:

                Firmware is software held in non-volatile memory — that is, power loss won’t lose it.

                So is regular software—it’s stored on a disk. Or do you mean that rather than being loaded into RAM, the CPU reads it directly from the ROM storage with load instructions? I never really thought about that before.

                Anyway, I know what firmware is, more or less. I just hadn’t thought about the origin of the term before. I always thought the “firm” in “firmware” referred to the firm that produced the hardware, rather than to the fact that it’s sort of a middle ground between hardware and software.Report

              • Avatar DavidTC says:

                Or do you mean that rather than being loaded into RAM, the CPU reads it directly from the ROM storage with load instructions?

                Well, that was sorta originally how it worked, but in reality, a lot of firmware is copied into RAM if the device has it, because ROM is slow. For a while, PCs used to always do this, called ‘Shadow BIOS’. (Now PCs don’t generally used their BIOS once the OS loads, and that feature can cause problems, so they don’t bother.)

                At this point, what firmware is, vs. software is, is a little vague. It used to be ‘software that is impossible, or at least very hard’, to alter…but uh, it’s *not*, anymore. Almost all ROM is simply flash memory, exactly like a USB device.

                So now the definition of firmware usually comes more to ‘Anything that is not exposed as *files*, and is instead just occupying a memory location the CPU can access.’. I.e., you cannot browse to your BIOS/UEFI in My Computer and see the files on it. So it has to be somehow different, right?

                But it turns out the only reason it’s not visible as a drive is, uh, the OS doesn’t want to show it that way! I mean, yes, it’s a different hardware API, but there’s no reason it couldn’t be shown as a drive. (Well, you couldn’t see ‘files’ on it anyway, as it doesn’t have a file system, but it’s not showing up as a block device. Mainly so you won’t insanely try to format it.)

                And then look at a phone. There is a single flash memory drive in a phone. It has at least one boot partition that isn’t ever altered, another partition with iOS or Android/Linux on it that (supposedly) can only be changed during an OS upgrade, another partition with installed programs on it, and some data partition.

                It doesn’t really have any ‘ROM’, as in, a dedicated chip. And everything is a file system, even if, for example, the boot partition doesn’t remain mounted. It simply uses part of its flash memory as things the user can change, and part as things it can’t. (If it did have a separate chip, it’s would be for the ‘hold reset to restore from there to recover the phone from’ feature. Although often *that’s* just another partition on the flash drive, switched to during boot!)

                And it doesn’t run *anything* from ‘flash memory’. (Except the microscopic amount of code required to find, copy into memory, and run the boot loader.)

                Which of this counts as firmware and which is just software? Who knows? Do phones even have ‘firmware’?

                Firmware, at this point, is basically whatever comes with the device and *the device itself* treats as ‘stuff you shouldn’t mess with in flash memory’.Report

              • I always thought the “firm” in “firmware” referred to the firm that produced the hardware, rather than to the fact that it’s sort of a middle ground between hardware and software.

                I assumed exactly the opposite. I wonder what that says about us.Report

            • Avatar DavidTC says:

              Oh, and to answer this question, yeah, they can just update the firmware…which will result in completely crappy performance.Report

  7. Avatar Jaybird says:

    It is resolved: If Volkswagon wishes to ever do business in the US again, it must first build an automobile plant in Flint, Michigan.Report

  8. Avatar Doctor Jay says:

    Being a computer guy, I was shocked the last time I had a vehicle with emissions problems, because I learned that the standard emissions test in CA now does not consist of hooking up a sensor to your tailpipe, but instead just plugging a computer into an onboard computer and having it tell you.

    The problem I was having was that the onboard computer was broken, and the fix was to get a new one.

    This was last winter. I said to my family that it was just a matter of time before somebody started hacking that device and offering people a way around the test. This is, in fact, really easy. Just report different numbers when queried, and all will be well.

    Basically this is a really bad way of testing something. That onboard computer can’t be trusted, since it belongs in the hands of an adversary.

    The good news is that there is some really good roadside testing in the works, and we can go to that. it’s sort of back to the future.Report

    • Avatar Michael Cain says:

      The good news is that there is some really good roadside testing in the works, and we can go to that.

      Yep. Colorado puts roadside testing vans here and there along the Front Range. Locations are published in advance. Interstate on-ramps are popular because (a) lots of cars and (b) cars are accelerating. IIRC, drive by and pass twice in the period from one year to two months before your emissions sticker expires and you’re good.

      On the down side, Colorado’s roadside testing doesn’t currently apply to diesel vehicles.

      On the really down side, gasoline vehicles less than eight model years old, and diesel vehicles less than four model years old, aren’t required to be tested. The assumption is that the car makers aren’t cheating.Report

      • Avatar Kazzy says:

        How exactly do the roadside tests work?Report

        • Avatar Michael Cain says:

          Low-power lasers at two frequencies (one IR and one UV, I think) are pointed across the road at a reflector. As you approach the measurement point, radar measures your speed and acceleration. Just as you pass, the strength of the reflected laser light is measured. In combination, all of those indicate how much carbon monoxide, unburned hydrocarbons, and nitrous oxides were emitted. As you move away, a camera records your license plate number. The statistical match between roadside and dynamometer testing is very good. The calculated values are skewed in a way that fails rather than passing borderline cases.

          The value of any of the emissions testing is a problem. Forcing the auto makers to design to tougher standards has had a much bigger benefit than identifying broken cars. OTOH, since Denver is always going to be borderline on ozone and particulates [1], there’s the benefit that if Colorado is implementing its own plan, the EPA doesn’t get to come in and impose whatever they decide will “fix” the problem.

          [1] High altitude makes any sort of combustion less efficient with more byproducts. The Denver Cyclone and some other weather patterns can keep air pollution bottled up. The metro area can exceed its bad-particulate days for the year if there’s a big fire in the next state and the wind is blowing the wrong way.Report

    • It’s just like having a corporation audited by people that same corporation pays to do the audit. And that’s never been a problem.Report

  9. Avatar DensityDuck says:

    This is also going to screw over Porsche and Audi, which are owned by the same people as Volkswagen (there’s some complicated interaction regarding which entities own how much of what brand, but as far as anyone outside the deal cares they’re all the same people.)


    And this commercial sure is embarrassing now…Report

  10. Avatar DavidTC says:

    Incidentally, my brother is an auto mechanic currently employed at a Volkswagon dealership, we talked about this recently, and he’s said two things that have stuck with me.

    First, he said that VW is probably not going to get away with changing the software, they’re going to have to add a urea injection system, which is going to be a huge hassle. (I actually asked him how he felt about this, and he’s like ‘Hey, I’m a mechanic, not a salesman. Do you have any idea how much work this is going to bring us?’)

    Second, he said that he, and a lot of other people, suspect that *all* cars behave differently under emissions testing than on the actual road. Especially diesel cars. Not to quite this extent, but enough to make a mockery of ‘clean diesel’ once the EPA actually starts testing this. He pointed out that you don’t really have to ‘cheat’ there…the computer could, for quite logical reasons, give the car more power during turns or longish acceleration…and the emission tests *doesn’t do those things*.

    With gasoline cars, this doesn’t quite matter as much, because emissions and MPGs are usually lined up in the same direction, you want to use the gas as completely as possible, and MPG testing is getting pretty smart, so car companies have an incentive to reduce emissions. With diesel, however, the emissions don’t really impact MPGs, but they do impact performance. He said, basically, that clean diesel is a lie (Or you’re driving a damn tractor.), and everyone’s going to realize it pretty soon.

    At least, that was what I understood.Report

    • Avatar Road Scholar says:

      The problem, as I understand it, is that it’s basically impossible to simultaneously optimize a diesel engine to emit acceptable levels (for admittedly arbitrary values of “acceptable”) of particulates and nitrous oxides. It’s a matter of combustion temperature. The temperatures you need to attain to burn up the particulates is high enough to trigger combustion between the atmospheric O2 and N2, thus nitrous oxides.

      The previous generation solution to this problem was, as I understand it, a metallic filter device in the exhaust stack that actually caught and held the particulates. When you’re really cranking out power, like going up a long hill, the exhaust temps would get high enough to burn those off. It works sort of like the catalytic converter in your car. Occasionally, if you ran for extended periods at low power, enough gunk would accumulate that you would have to run a “regen” cycle, basically a really high idle, to burn it off. The thing is, whenever you were running hot enough to regen you were also putting out a lot of NOx, but on average both values met spec.

      Then they tightened up the specs. So my current ride (’14 model) still has the regen system but also has the urea system. It also runs about 25 degrees hotter than the older system. And that’s the whole point of the urea. It allows the engine to run hot enough to minimize the particulates by neutralizing the NOx produced as a result.Report

      • Avatar DavidTC says:

        Yeah, that’s sorta how I understood it.

        With gasoline, you can have an ideal gas engine that combusts everything completely, and incidentally gets the best possible gas mileage *and* the best possible performance. It’s a complete win-win-win. Now, this not being an ideal world, we can’t actually make the engine doe that, there are *still* trade-off being made, but there is some sort of hypothetical target out there.

        With diesel, not so much. Like you explain, there isn’t even an optimal way to reduce emissions…you trade one for another. And, from what I understood from my brother, running the engine hot enough to burn off the particulates is actually moving in the wrong direction for increased MPGs. A huge part of the increased efficiency of diesel *is* the lower temp. (Which is why they do the whole ‘regen’ thing instead of just running hot the entire time.)Report

        • Avatar Oscar Gordon says:

          The hotter you run, the more energy you dump to the atmosphere as waste heat. If we could wrap a diesel in some really good insulation & keep the heat contained so the operating temperature was very carefully maintained…

          But that adds material, and complexity, etc.

          As you said, tradeoffs. Engineering is all about tradeoffs.Report

          • Avatar Joe Sal says:

            The only reason energy is dumped to the atmosphere is because there is a limit on the power stroke. If you had a magic engine with no limit of stroke, you could continue to harvest power (chamber pressure) on the piston until the cylinder pressure reached ambient pressure.

            Now if that engine could be perfectly insulated and weigh next to nothing and fit in your pocket….Report

            • Avatar Oscar Gordon says:

              That’s not the only reason, but it is the big reason. I was more alluding to why we don’t want to run the engine hot all the time. We could, but adding more fuel or running at higher rpms, but that wastes fuel. Or we could do more to control the engine temperature, but that adds complexity or cost (in materials, or maintenance, etc.).

              There are all sorts of ways to make diesels work the way marketing claims, but doing so adds cost & complexity.

              The common piston engine is a mature tech and we are really straining against it’s limits. Which is why there is a lot of work being done on alternate configurations, to see if a paradigm shift can happen.

              Hrmmm, perhaps another idea for a post?Report

  11. That said, the fines for not picking up dog poop are pretty hefty. I think this is a nod to the fact that violations are unlikely to be detected.

    They will almost certainly be detected, and far too often by someone wearing sandals.Report

  12. Avatar Michael Drew says:

    What the system cannot handle, however, is capable people going out of their way to evade enforcement and experiencing a high likelihood of wild success. In such cases, getting caught becomes a near impossibility, and there is then nearly no (dis)incentive to break the law.

    This is essentially the entire dynamic of major-player financial regulation. But the degree to which noncompliance on Wall Street swamps regulators’ ability to punish it I’m going to guess dwarfs what’s the case in this space. (Could be wrong there, but it’s at least comparable.)

    So should this be the case on Wall Street?:

    On those occasions that people are caught doing such things, they shouldn’t get to play the game anymore. They are a threat to the whole system. It is too difficult to give them the level of monitoring required to make sure they follow the rules when they have demonstrated a willingness to circumvent whatever little monitoring we do have in place.Report

  13. Avatar Michael Drew says:

    Only tangentially related, I saw this car today. (Sorry for the flash, it was just getting dark & I wasn’t sure.)

    I call the license plate smug there, but I admit I chuckled a little.Report

  14. Avatar Jaybird says:

    A story from the 90’s, apropos of nothing, really. I drove a little 80’s Pontiac LeMans and had to take it into the car place to pass emissions testing. I failed the test. I started freaking out because, hey, I was, like, 23 or something and could not afford to fix stuff up. The mechanic told me “hey, it’s no big deal. You get one free retest. Just get on the highway and drive 65 MPH up to Monument and then turn around and drive back here.”

    I did so and, when I got back, my emissions test passed with flying colors. The numbers plummeted.

    And all I had to do to pass the test was drive 65 MPH for a half hour. I’m sure that the car would have failed again within a week of driving around town stop and go.Report

  15. Avatar gingergene says:

    The CEO of Peanut Corporation of America just got 28 years for shipping salmonella-contaminated peanut butter. His brother got 20 years and his QA manager got 5. PCA killed fewer people than VW is estimated to have killed with its increased emissions.Report

  16. Wow, amazing blog layout! How long have you been blogging for? you make blogging look easy. The overall look of your site is wonderful, as well as the content!Report