Science in Society Archive

Volkswagen & Others Cheat Because Regulators Are Toothless and Worse

Volkswagen has been caught cheating on emissions by a civil society organisation; it is not the first time that the motor industry have cheated, nor is it only the motor industry that cheats. Regulators are simply not protecting public health and interest Prof Peter Saunders

Programmed to cheat on emissions

A modern car has an on-board computer that tracks what the car is doing and tunes the engine appropriately. In its latest diesel models, Volkswagen has been programming the software so that if the car is being tested, which it can determine from clues such as the engine running but the steering wheel not being moved, the engine will be set to reduce the emissions of the nitrogen oxides NO and NO2 (referred to collectively as NOx) to below the maximum legal level. On the road, the engine reverts to its normal setting, which gives better performance and fuel economy but very much higher emissions, possibly as much as 35 times the legal limit [1].

Volkswagen will have to recall and refit eleven million cars in Europe and America. They are likely to incur very large fines. Their customers may sue because when the engines are modified to bring the emissions down to the legal limits, the performance and the fuel economy may be significantly below what they were promised when they bought their cars. The CEO has already resigned, and other executives may follow.  There may be prosecutions in Germany. And the company’s reputation has suffered a serious blow.

The emissions kill 9 500 a year in London alone

There’s been a lot in the media about the cost to Volkswagen but little about the more serious issue, the cost to the rest of us. The regulations were not introduced to annoy the manufacturers. They are there because the nitrogen oxides are known health hazards. It is estimated that high levels of NOx lead to 9 500 premature deaths every year in London alone [2].

Most of the NOx comes from diesel vehicles, and while that’s not just Volkswagens, Skodas, Seats and Audis, a lot of it is. It has been shown that under city driving conditions a small diesel car produces four times as much NOx as a large bus or lorry [3]. (The most commonly used NOx reduction technology, injecting urea into the exhaust gases, is not fitted in diesel cars on grounds of cost and convenience.)  Even allowing for the fact that the comparatively lax European limits result in unsafe NOx levels even when they are adhered to, there can be little doubt that people have died because cars were emitting far more NOx than the regulations allow.

It is especially worrying that the cheating was not discovered by the regulatory agencies, the Environmental Protection Agency (EPA) in the US and various national agencies in Europe. Instead, it was a small NGO, the International Council on Clean Transport (ICCT) that found out what was going on. All it took was a bit of curiosity - someone wondering why apparently similar cars could meet stricter standards in the US than in Europe - and some simple on-the-road tests [1].

It’s not as if no one had ever heard of this sort of thing before. Manufacturers have been cheating on the tests for years, pretty much for as long as there has been testing. In the 1970s, Ford was fined for performing constant maintenance on its cars during testing to improve their performance, and Volkswagen and Chrysler were caught installing devices to shut down the pollution control systems. In the 90s, GM paid a $45 million fine and recalled nearly half a million Cadillacs that switched off emission control systems whenever the air-conditioner was being used, and long haul truck manufacturers paid nearly $1 billion for using software to optimise the performance of diesel engines during testing, much the same as Volkswagen was doing this year.  

There are also what have become accepted practices within the industry, such as taping the doors and grilles to improve the aerodynamics, using special lubricants to reduce friction in the engine, recharging the battery between tests so that the alternator is never operating, and so on. That’s why the consumer magazine Which finds it can never reproduce the manufacturers’ results for fuel economy. You might argue that this wouldn’t matter if the difference were the same for every car they test (though it isn’t) but it would be far better if the manufacturers got into the habit of providing honest information. If there is a culture of cheating, then the industry will always be looking for ways of thwarting the regulations. Like bankers, really, or corporations like Amazon who never seem to pay any tax.

Volkswagen does appear to have cheated on an unusually large scale, though we can’t be sure because we don’t know how often they and other manufacturers have got away with it. If they were caught out every time, they probably wouldn’t do it. Volkswagen also ignored warnings from as far back as 2007 that ought to have tipped them off that what they were doing was not as likely to remain secret as they were hoping.  It seems they had not been able to tune their new VA189 engine both to meet its specification and conform to the regulations [4]. That must have made it all too tempting to fake the tests rather than throw away all that effort and start again.

Not just Volkswagen

It’s not just Volkswagen and it’s not just motor manufacturers. The same sort of behaviour can be seen in many other industries -- and of course in the banks as well.

A common feature is the weakness of the regulators. The EPA should be ashamed that it was not they but a small NGO that discovered what Volkswagen were up to. Where was the EPA when this was going on? They seem to have seen their role as purely bureaucratic: the manufacturer would have its vehicles tested at a testing station of their own choosing and pass the numbers to the EPA who would tick an appropriate box. Had the EPA been doing the tests themselves, someone might have wondered about the results just as John German of the ICCT did [1].

The EPA has collaborated with the biotech industry in playing down evidence for harm (see e.g. [5, 6] Genetically Modifying Genes and Scientific Evidence, SiS 67).  The European Commission is no better,  the re-approval of glyphosate herbicides is to be based on a review carried out largely by the Glyphosate Task Force, a consortium of chemical and biotech companies including Monsanto “joining forces and efforts in order to renew that glyphosate registration with a joint submission” [ 7, 8] Scandal of Glyphosate Re-assessment in Europe, SiS 63).

The regulatory agencies do not generally do their own testing; they rely on the assurances they are given by the manufacturers. Yet even if what is supplied to the regulator is strictly speaking correct, it can be totally misleading, as in the case of pharmaceutical drugs. There are many ways of accomplishing this; a common one is to carry out a number of clinical trials and report only the one that gives the desired result. This could be countered by insisting that all clinical trials must be registered in advance and the results reported, but while the leading medical journals favour this, the pharmaceutical companies have so far successfully resisted it. Pharmaceutical companies have also been slow to reveal harmful side effects that became apparent after the drug was on the market. (See [9] Vioxx, A Mercky Story SiS 43; [10] The Depressing Side of Medical Science, SiS 39 [10]; [11] Statins: Flawed Studies, False Advertising and Lack of Transparency, SiS 67; and many more).

Commercial Confidentiality

Manufacturers rely heavily on ‘commercial confidentiality’ to prevent independent scientists and sometimes the regulators as well from seeing the evidence on which a drug was approved. Yet without details of the protocol and the results, it is impossible to have confidence in the outcome, any more than we can trust emissions measurements if we do not know how the testing was done and in particular whether so-called defeat devices were used.

Commercial confidentiality should be thought of in terms of a bargain, just as patenting is. When you invent something, you do not have to tell anyone what it does or how it works, but anyone can copy it. If you don’t want this to happen and if you have no way of keeping your invention secret, you can protect it by taking out a patent. You will be given a 20 year monopoly, but in return you must provide a detailed description that anyone can ask to see.

Similarly, while companies are entitled not to let their competitors know what they are planning or how their new developments are getting on, if there are regulations to be satisfied or a licence to be obtained, then in return for the authority to go ahead, the evidence should be publicly available.

To conclude

Industries cannot be trusted to set their own exams and mark their own scripts. The conflicts of interest are too great. They should obviously have a significant role in setting up the regulations, but they must not be permitted to dominate the process. The members of the panels and boards that take the decisions should not be mostly either from the industry or closely connected with it. It may seem natural for governments to leave the decisions to such people, and they do have relevant expertise (though not necessarily in health and safety issues) but they also have their own agendas.

Recently, for example, the governments of France, the UK, and Germany have been lobbying the European Commission on behalf of the motor manufacturers to keep loopholes in the test procedures that would allow carbon dioxide emissions to be 14 % above what the regulations permit [12]. Reducing carbon emissions may not be important to the motor industry and its friends but it is for the rest of us, which is why we cannot leave them to frame the regulations.

The regulatory agencies must play a more active role in implementation than many of them have so far. This would not require building new laboratories and hiring large numbers of staff. Most of the necessary equipment and expertise can already be found in universities and research institutions. But the work would be commissioned by the regulators rather than the manufacturers, and the staff would have proper career paths within their institution; they would not be looking forward to later employment with the companies they are being now paid to regulate. It would be part of their job to think of and ask awkward questions, as the ICCT did and the EPA did not. It is encouraging that a suggestion along these lines has recently appeared as an editorial in the journal Nature [13].

Where the nature of the industry means that companies have to undertake most of the actual work – the Food and Drug Administration (FDA) can hardly run large scale clinical trials – the regulatory agencies should still be closely involved and the details of the tests and the raw data should be available to independent scientists.

This need not cost much more than the present system. The testing still has to be done; it should not be that much more expensive to do it properly. Corporations might find it inconvenient in the short term, but they would get used to it. Volkswagen, faced with a bill that is expected to be at least $11 bn and possibly a good deal more [14], and a devastating blow to their reputation, would be far better off if cheating on the test had not looked such an attractive option.

Above all, we must never lose sight of the fact that the regulations Volkswagen has been evading are there because they save lives.

Article first published 26/10/15


  1. “Meet John German: the man who helped expose Volkswagen’s emissions scandal.” Rupert Neate, The Guardian, 26 September 2015.  27/09/15
  2. “The Volkswagen emissions scandal explained.” Gwyn Topham, Sean Clarke, Cath Levett, Paul Scruton and Matt Fidler,, 23 September, 2015.
  3. Institute of Transport Economics. “Diesel cars have high emissions in real traffic.” Norwegian Centre for Transport Research, 2015, accessed 22 October 2015,
  4. “What’s the fix for Volkswagen’s ‘Defeat Device’? Experts unsure.”  Paul Eisenstein, CNBC News, 28 September 2015.
  5. Druker SM. Altered Genes, Twisted Truth. Clear River Press, Salt Lake City UT, 2015. ISBN 978-0-9856169-1-5 (hardcover), 978-0-9856169-0-8 (softcover).
  6. Saunders PT. Genetically modifying genes and scientific evidence. Science in Society 67 48-49, 2015.
  7. Swanson N and Ho MW. Scandal of Glyphosate Re-assessment in Europe.   Science in Society 63 8-9 (2014).
  8. Glyphosate Task Force (2013). The European Glyphosate Task Force.
  9. Saunders PT. Vioxx—A Mercky story. Science in Society 43, 7-8, 2009.
  10. Saunders PT. The depressing side of medical science. Science in Society 39,  50-51, 2008.
  11. Rosch PJ. Statins: Flawed studies, false advertising and lack of transparency. Science in Society 67 36-39 (2015).
  12. “UK, France and Germany lobbied for flawed car emissions tests, documents reveal.” Arthur Neslen, The Guardian, 24 September, 2015.
  13. Testing times (editorial). Nature 2015, 526, 5.
  14. “Cost of Volkswagen’s emissions scandal? At least $11 billion, analysts say. Friedrich Geiger, Wall Street Journal, 29 September, 2015.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a duck?

There are 2 comments on this article so far. Add your comment above.

Rory Short Comment left 27th October 2015 04:04:27
Very few, if any, industries exist as a public service, they exist to make money. Therefore they have zero motivation to follow rules or practices which will cost them money even if the rules and practices will benefit the general public. Consequently the monitoring of their adherence to such rules and practices simply has to be done by people who have absolutely no interest in the industry concerned.

algimantas k bronisas Comment left 4th November 2015 05:05:00
the regulations and penalties for the producers of potentially harmful substances are "toothless"by design.....they are written by the corporate law firms of the producers then given to the legislators.....who without any research ,work,or even reading them......make them the virtually unenforceable and obfuscating laws theyre designed to be and ........when the harmful substances begin causing evident illness and economic becomes a very difficult,costly and long process to prove toxicity and the resulting health damage and economic losses.........meanwhile the toxic substance producing corporations show clear bottom line profit for selling tons of their products worldwide.....this now becomes a cost-benefit analysis proof for government to defend the violating corporations.......unfortunately many scientists are buying into this myopic view instead of seeing the big picture where even the flap of of a butterflys wing vibrates throughout the the section on Exposure Control