The Amazing Shrinking 114

Federal Motor Vehicle Safety Standard (FMVSS) 114 Theft Protection and Rollaway Prevention began its life in 1969 as solely a theft protection standard in response to a wave of amateur car thieves and the attendant mayhem on the road. By the late 1980s, that rule began to tackle the problem of rollaways caused by the ability to inadvertently shift the gear selector into non-Park positions with the key removed from the ignition. In the mid-2000s, the agency began another round of rulemaking to address the advent of keyless ignitions. But the standard has not been amended in 17 years, and during that time it has drifted far from its historic intent, providing the public with little theft protection or rollaway prevention. Whether the issue is deadly rollaways in keyless ignition vehicles or a tsunami of auto thefts, NHTSA has found no recourse in FMVSS 114 – the standard it wrote to address these hazards. Be it rulemaking, compliance, or enforcement NHTSA has consistently reacted to issues that underscore the standard’s weakness by declining to use its authority to uphold the standard’s intent. That is especially true if identified safety hazards can be blamed on drivers, rather than designs which encourage human errors without adequate countermeasures.

In 2021, Congress broke a decade-long impasse to update the standard with the passage of the Infrastructure and Jobs Act. The law compels NHTSA to pass a Final Rule requiring vehicles with keyless ignitions to install an automatic engine idle shutdown to prevent carbon monoxide poisonings. Ford and GM have been doing this beginning in 2012. Chrysler had one vehicle in 2018. Toyota followed suit in the last few years. According to the Fall Unified Agenda, in which various federal agencies publish their schedules of rulemaking activities, NHTSA expects to publish a new FMVSS 114 Notice of Proposed Rulemaking in June.

Last week, in response to the surge in amateurs stealing Hyundai and Kia vehicles with inadequate theft protection, Safety Research & Strategies submitted a petition for further rulemaking, to strengthen the anti-theft compliance test provision, which is virtually un-fail-able. The petition outlines the current Kia Boyz continuing mayhem, history of the standard, the basic and vague requirements of the current anti-theft test, and compares them to the specific and more rigorous requirements of the tests for immobilizers – which are optional for manufacturers.

We can’t say why NHTSA has abandoned FMVSS 114, but we can chronicle its descent into near-irrelevancy.

Theft Protection?

In the late 1960s, many amateur car thefts never had to break a window or jimmy a lock – they simply slipped into the driver’s seat of a vehicle with the keys dangling in the ignition and took off. Yet, joyriding posed a significant safety hazard, because thieves were not only stealing cars, they were getting into crashes and causing injuries to themselves and others — and at a disproportionately higher rate than vehicles that were not stolen. According to an April 1968 Federal Register Notice proposing FMVSS 114:

The evidence shows that cars operated by unauthorized persons are far more likely to cause unreasonable risk of accident, personal injury, and death than those which are driven by authorized individuals. Further, the incidence of theft, and hence the risk of accidents attributable thereto is increasing. According to a recent study by the Department of Justice there were an estimated 94,000 stolen cars involved in accidents in 1966, and more than 18,000 of these accidents resulted in injury to one or more people. On a proportionate basis, 18.2 percent of the stolen cars became involved in accidents, and 19.6 percent of the stolen-car accidents resulted in personal injury. The same study predicted that automobile thefts in 1967 total about 650,000; about 100,000 of these stolen cars could be expected to become involved in highway accidents. Comparing these figures with statistics for vehicles which are not stolen, the approximate rate for stolen cars would be some 200 times the normal accident rate for other vehicles. Thus, a reduction in the incidence of auto theft would make a substantial contribution to motor vehicle safety. It would not only reduce the number of injuries and deaths among those who steal cars, it would also protect the many innocent members of the public who are killed and injured by stolen cars each year

In response, NHTSA proposed to establish a theft protection rule. From the beginning, the agency’s goal was to remind drivers, via an audible warning, that they had left the key in the ignition and to make it difficult for thieves to activate the engine “within a short period of time.”

Some automakers questioned whether NHTSA was properly using its authority, or had a solid factual basis for its proposal.  Others supported the idea, but found NHTSA’s execution to be totally wrong. Nonetheless, there was consensus for the objective of thwarting the casual thief from nicking a car, and manufacturers were quite firm that they didn’t want to be forced to use a specific anti-theft technology to do so. They argued, persuasively, that technology was always evolving, and that it was necessary to stay nimble to keep one step ahead of criminals who can eventually figure out how to overcome a system.

The agency agreed that it would be best to keep the language of the technology neutral, and allow automakers to use, develop and innovate any strategy that satisfied the regulatory language. The Final Rule simply requires:

Each vehicle must have a starting system which, whenever the key is removed from the starting system prevents:

(a) The normal activation of the vehicle’s engine or motor; and

(b) Either steering, or forward self-mobility, of the vehicle, or both.

Although the final rule was not technically prescriptive, there were nine years of discussions about the best ways to accomplish this. And, even back then, the agency understood that insufficient shielding of ignition wires was a factor in the speed at which a thief could breach the ignition system. In May 1979, a constituent of then Sen. Ernest Hollings noted that cars were easier to steal due to manufacturers’ change from “a heavy metal casting steering-column (at a weight savings of only 1 ½ pounds) to a molded-plastic housing which no longer adequately protects the ignition lock.” In a 1971 docket, NHTSA wrote:

After a review of the state of the art in lock design and the difficulty of articulating performance standards for stronger locks, the NHTSA has tentatively concluded that it would be more effective to approach the problem of the susceptibility of locks to tampering by limiting the utility of removing the lock. Consequently, a new requirement is being proposed which requires the ignition systems to be inoperative if any part of the ignition lock is removed. To further protect the ignition system, the agency proposes also that the wires which activate this system shall be shielded so that they cannot be directly contacted from within the passenger compartment. The shielding could be provided by the vehicle structure or by other means. The agency is considering establishing a requirement that would necessitate the use of metal or other similar strong shielding materials which would have to be cut by special cutting tools before access to the ignition wires could be gained.

At the time, the idea that the ignition lock should be designed to resist removal was supported by major automakers such as GM, Ford, Mercedes and VW. And NHTSA noted that it was evaluating a study by the National Bureau of Standards of tensile, torque and extraction testing on current ignition lock systems to determine if it would propose specific lock retention and system operations performance standards in the future. The agency never did propose an ignition shielding requirement, and theft protection rulemaking never advanced any further.

NHTSA’s Office of Vehicle Safety Compliance (OVSC) lays out the procedures for its contracted testing labs performing FMVSS 114 compliance tests, the primary tests associated with the theft prevention aspects of rule state:

“With the key removed from the starting system, attempt to start the vehicle engine or motor. If the vehicle is equipped with an advanced key system, it may be necessary to move the electronic key device outside the minimum effective range for loading the electronic key into the vehicle starting system.”

“Determine if the steering wheel locks as a result of removing the key from the starting system by rotating the wheel in both directions.”

“Determine if forward self-mobility is prevented whenever the key is removed from the starting system. If the vehicle is equipped with an advanced key system, it may be necessary to move the electronic key device outside the minimum effective range for loading the electronic key into the vehicle starting system.”

So, test labs have to try to start a keyed vehicle without placing the metal key in the ignition slot, or, in the case of a keyless ignition vehicle, move the key fob out of vehicle range and attempt to start the vehicle by depressing the Start/Stop button, then try to  move the steering wheel and determine if vehicle can move forward under its own power.  No tools or methods for accomplishing this are prescribed. How any automaker could actually fail to comply is beyond The Safety Record’s comprehension. The procedure only partially addresses the second part of the requirement, that a FMVSS 114 compliant vehicle must also prevent “either steering, or forward self-mobility, of the vehicle, or both.”

While OVSC’s lab test procedures states they aren’t intended to limit or restrain its contractors from devising and using any techniques or equipment to obtain the required compliance test data,  and the document isn’t an endorsement or recommendation for using a particular test method, the practical reality is the opposite. Is a contracting test lab really going to devise a different test method based on an historical review of the intent of the standard? Not likely.

But maybe they should review a legal interpretation from 2004 from then-NHTSA Chief Counsel Jacqueline Glassman who recognized the intersectionality of the two-part requirement in a response to an unidentified automaker, who requested agency guidance on its engine control immobilizer module and the requirements of FMVSS 114. Glassman agreed that the system the automaker described would be compliant with FMVSS 114 because if an attempt was made to circumvent the ignition lock, including through “hot-wiring,” the immobilizer prevented engine starting without the key.

Immobilizers, defined by NHTSA as “an anti-theft device that combines microchip and transponder technology with engine and fuel immobilizer components that can prevent vehicles from starting unless a verified code is received by the transponder,” are not required, but most automakers use them.  They showed up in the 1980s and today, according to the Highway Loss Data Institute, 97 percent of MY 2021 model cars have immobilizers, compared to 26 percent of MY 2021 Hyundai/Kia models. In contrast to FMVSS 114’s meaningless compliance test, if a manufacturer chooses to use an engine immobilizer anti-theft technology, there is specific criteria that the immobilization system must be designed to meet so that it cannot easily be defeated to allow forward self-mobility by either disrupting the voltage or by using tools. These criteria enumerate 18 common tools auto thieves are known to rely on: Scissors, wire strippers, wire cutters and electrical wires, a hammer, a slide hammer, a chisel, a punch, a wrench, a screwdriver, pliers, steel rods and spikes, a hacksaw, a battery operated drill, a battery operated angle grinder; and a battery operated jigsaw

Enter the Kia Boyz. With one viral 2021 TikTok video, every issue associated with the safety hazard of quick and easy car thefts identified and debated more than 40 years ago has been borne out in the ongoing crime wave involving MY 2010-2021 Hyundai and Kia vehicles with traditional metal keys. Thieves can just rip off the steering column shroud, pull out the ignition cylinder, and using the end of a USB cable, grip the ignition cylinder to twist it, starting the engine. The whole deal can go down in about 30 seconds.

In the span of approximately a year, Hyundai/Kia’s decision to forego basic anti-theft features resulted in 14 crashes and eight deaths, according to a NHTSA press release in February. But surely, these are undercounts, because in March and April alone, the news media reported another 24 such crashes, resulting in four more deaths (including a six-month-old boy), 28 injuries, nine of which were said to be serious injuries, and one house fire, in addition to structure damage and other vehicle damage, including two police cruisers, caused by drivers of stolen Hyundai/Kia vehicles.

Hyundai/Kia now facing a demand from 17 state and Washington, D.C. attorneys general to launch a recall, and lawsuits from many municipalities, including Buffalo, Milwaukee, Rochester, Columbus, Madison, St. Louis, New York City, Cleveland, Seattle and San Diego, auto insurers, and multiple class actions. One entity not holding Hyundai/Kia to account is NHTSA. NHTSA didn’t open a defect investigation or press Hyundai/Kia to launch a formal recall with its notification and accountability requirements. NHTSA took no enforcement action at all, because it didn’t believe that the theft protection safety standard was actually enforceable. In response to inquiries from Safety Research & Strategies, a NHTSA media representative stated that the agency could not act more forcefully, because “the standard does not define normal activation.”

If that’s the case, it’s high time the agency either define normal activation or re-fashion a meaningful compliance test that addresses FMVSS 114’s historic intent.

Rollaway Prevention?

FMVSS 114 pivoted to the rollaway problem in 1988, in response to reports of crashes and injuries caused by steering wheel lock-up when the key is inadvertently removed, and rollaways caused – primarily – by children moving the gear selector from Park to Neutral. By 1990, the agency published a Final Rule requiring vehicles with automatic transmissions that have a Park position to have a key-locking system that prevents removal of the key unless the transmission is locked in Park or becomes locked in Park as the direct result of removing the key. The agency noted that this amendment accomplished several goals at once – it didn’t diminish the theft prevention aspect of the rule, it prevented steering wheel lock up if the key was inadvertently removed from the ignition, and it prevented inadvertent shifting of the gear selector, because the key’s status was now linked to the Park transmission position.

The final rule addressed the safety need for transmission shift locks, with the agency calculated that there are “roughly 400 to 800 relevant injury producing transmission lever shifting accidents each year.” NHTSA also emphasized that being able to move the gear selector out of Park without the key in the ignition posed a significant safety risk and that the agency has a “special obligation to reduce injuries involving children.”

Nothing much changed on the rulemaking front for the next 15 years. But ignition systems began to undergo a significant transformation, as automakers introduced keyless ignition vehicles into the fleet. No longer did the driver need a metal key to lock and unlock the doors, or even to start the vehicle. The key was now an invisible electric code stored in or removed from the ignition module, delivered by a plastic key fob. In a 2002 interpretation letter, NHTSA Chief Counsel Glassman warned an unnamed automaker that this new system would disrupt a driver’s relationship with the key:

We observe that if the ‘Smart Key’ device remained in the car. e.g. in the pocket of a jacket laying on the seat, a person would need only turn the ignition switch knob to start the engine. It appears to us that, with systems of this kind, there would be, in the absence of some kind of a warning, a greater likelihood of drivers inadvertently leaving a ‘Smart Key’ device in the car than with a traditional key. This is because the driver must physically touch a traditional key, unlike the “Smart Key” device, as part of turning off the engine. You and/or the vehicle manufacturer may wish to consider whether there are any practicable means of reducing the possibility of drivers inadvertently leaving their ‘Smart Key’ devices in the car.”

Glassman got it partly right – but didn’t address other serious problems that emerged with keyless ignition vehicles – which facilitated the omission of critical steps in the vehicle exit behavior sequence: driver’s exiting vehicles with the key fob, vehicle in Park, but without shutting engine off, creating a carbon monoxide hazard or shutting the engine off and exiting without placing the gear selector into Park, creating a rollaway hazard. Both hazards were largely addressed with mechanical key designs that prevented the removal of the ignition key unless the engine was off and the gear selector in Park.

The year 2005 was a FMVSS 114 watershed for a couple of reasons. First, although NHTSA still had not promulgated a rule to prevent rollaways by requiring a brake to shift interlock (BTSI), then-New York Sen. Hillary Clinton filed the Cameron Gulbransen Kids and Cars Safety Act, which would force the agency to do so. (In 2006, 19 major automakers scrambled to head off a regulation by announcing a voluntary agreement to install brake to shift interlock. That was swell of them, but in 2008, the Clinton’s bill passed, and the Final Rule was published in March 2010.)

Second, in 2005, NHTSA attempted to amend FMVSS 114 to address the advent of keyless ignition, by revising the definition of the key to include an invisible electronic code. This seemingly created an FMVSS 114 compliance problem, because these new systems allowed the driver to shut the engine off in a drive gear, and walk away with the fob. In a keyed vehicle, that metal key cannot be released from the ignition slot unless the transmission is in Park or automatically moves into Park when you remove the key. Not to worry though, NHTSA told automakers how to get around the rule in a Federal Register Notice: “We also explained that systems using an electronic code instead of conventional key would satisfy the rollaway prevention provisions if the code remained in the vehicle until the transmission gear is locked in the “‘park’ position.’”

This “cheat” immediately made FMVSS 114’s rollaway prevention provisions less effective.

Most manufacturers designed their vehicle electrical architectures so that if the driver turned off the engine without placing the transmission in “Park,” the ignition state would quietly move to the Accessory position to keep the electrical power on and the electronic “key” was maintained in the ignition module. These designs typically included a dashboard text message and/or indistinguishable chime that was allegedly intended to alert the driver to put the transmission into Park. In many cases, drivers exiting with the fob with the gear selector not in Park and the engine off, never saw the message as they turned their bodies toward the door, or failed to hear the chime or if they did, had no way to discern its meaning, and exited without knowing their vehicle was unsecured and the power was still enabled, draining the battery, to boot. Automakers resolved the dead battery complaints by installing battery saver features that would eventually turn off the power, but did little to protect the departing driver or nearby pedestrians from rollaways.

In 2009, NHTSA asked the industry to convene a SAE sub-committee to write a voluntary standard to address the human factors issues created by keyless ignitions. But the result was pathetic: it blamed drivers for manufacturers poor design choices and didn’t offer much guidance on countermeasures beyond than vague recommendations for warnings.

Dissatisfied, the agency opened a rulemaking in December 2011 to amend FMVSS 114 to address panic stops during an unintended acceleration, and the rollaway and carbon monoxide safety problems created by keyless ignitions. The proposal to solve for the latter were very loud 85-decibel warnings, if the driver exited without placing the transmission in Park or turning off the engine. Predictably, manufacturers loathed this idea, and excoriated NHTSA for even suggesting such a thing without rigorous data to determine a safety need. When NHTSA had proposed to hire a consultant to do some human-machine-interface investigations into keyless ignitions, the industry objected again. NHTSA hasn’t touched the rulemaking for a dozen years.

In 2014, apparently having forgotten that they told manufacturers how to meet technical compliance with FMVSS 114, while leaving the vehicle able to freely roll, NHTSA OVSC began testing 34 recent model-year vehicles to determine if push-button systems allowed the vehicle to be turned off in a gear other than park, or the key fob to be removed from a running vehicle with no warning to the driver, or allowed vehicles to be restarted without the key fob present. They found that you could turn off the engine with the transmission in drive, leave with the fob, and easily push the vehicle into a roll. The agency officially launched the compliance probe in January 2014 with an Information Request seeking sought a host of details related to manufacturers’ keyless ignition systems, ranging from the electronic architecture of the system, when the electronic code is purged from the system and the audio and visual telltales used to alert the driver when he or she has exited the vehicle. NHTSA also asked for complaint data and the safety information manufacturers provide to their customers about keyless systems.

After automakers explained exactly how their vehicles met the rollaway standard using the Accessory work around, the agency closed the probe with no findings of non-compliance. For example, in her report regarding potential non-compliances in Kia vehicles, the safety compliance engineer who conducted the investigation, noted: “Each vehicle was started with the push button control and the transmission selection control was placed in Drive. The starting system was deactivated with the push button control and the key fob was removed from the vehicle. We verified that the vehicle was not in Park by pushing it.”

After conversations with Kia, NHTSA’s OVSC learns: “The information and test data provided by Kia indicates the vehicles listed above meet all requirements of FMVSS No. 114. Regarding SS .2.1, if the vehicles’ starting system is deactivated when the transmission is not in Park, the starting system will be in the accessory position, the key (electronic code) has not been removed, and the transmission control is not required to be in Park. In addition, drivers are provided audible and, for some vehicles, visual warnings about the key and transmission position.”

NHTSA eventually told automakers they needn’t bother to answer the question of how many complaints they had logged.

In 2015 and 2016, NHTSA opened three separate rollaway-related defect investigations into vehicles with novel shifters, with spotty results.

The first, PE15-030, in August 2015, investigated rollaways after intended shifts to PARK in 2014-2015 MY Jeep Grand Cherokees vehicles with monostable-style shifters. The monostable is a T-handle style shifter with a similar appearance of a traditional mechanical T-handle shift selector. But monostable designs are activated by depressing a button on the handle and moving the handle rearward or forward to achieve a transmission position; the handle always returns to the centered/neutral position. The design provided no tactile feedback, so drivers trying to achieve Park were inadvertently selecting Reverse or Neutral. In April 2016, FCA brought the investigation to a swift close by recalling Jeep Grand Cherokees, Chargers and 300s with the monostable shift levers. FCA blamed drivers for failing to respond to the vehicles’ warning strategies, but nonetheless installed Auto Park, which automatically moved the electronic monostable transmission selector into Park when the driver failed to do so, and the ignition switch was pressed to turn of the engine, or if the driver attempted to exit the vehicle without the Park engaged.

The second two defect investigations were opened on the same day in December 2016, into vehicles with rotary style electronic shifters:  one covering MY 2013-2016 Ram 1500 and MY 2014-2016 Dodge Durango, and the MY 2012 through 2014 Land Rover/Range Rover Evoque and Jaguar XF vehicles. In both cases drivers complained that the vehicles had rolled after the drivers had placed the transmission in Park – or thought they did. In August 2020, NHTSA closed the investigation into the Jaguar / Land Rover vehicles, after declining complaints and claiming it found no mechanical or electronic fault that would prevent Park from being engaged; this was likely case of driver error, the agency said. While FCA initiated a series of service campaigns to add automatic shift to Park software updates, the investigation into the FCA vehicles is still open six years and five months later.  

So there you have it, after nearly 40 years of regulatory progress, FMVSS 114 begins its descent into a theft protection and rollaway prevention safety standard that does neither. Most of the time NHTSA has an opportunity to enforce it or improve it, the agency backs off, and the public is less safe because of it.