team, the group that terms the root cause of the problem, reviewed the claims relating to the cobalt nondeployment, but ultimately decided not to pursue it. why does the product investigations not pursue this matter at the time? >> that's a very good question. and the answer is this was some of those things passed off to another agency. mr. operation was keeping track of it. the other investigators weren't following up with regard to it. they were gathering information, if you will. but that's where they went with it. >> so when you say that it was kind of handed over somewhere else, you're referring to the product -- field performance assessment division? >> yes. >> because according to again our information, it says afz te proceed duct investigators declined to investigate, the responsibility for being tag e tracking the claims was assigned to the field performance assessment division. do you consider this unusual, would this be typical in a situation where there has been an obvious issue that has come to light and it just couldn't of be passed off to another -- and i guess i would like to know,

inconsistent recollection as to whether the product investigations group became involved in the cobalt air bag nondeployment issues at this stage. everest reports in april 7 the fda transitioned the cobalt matter to the p.i. group where it was taken on by an engineer. document in the file indicate that he was working on the issue and a may 42007 status review planning worksheet states that he was scheduled to present on an issue described as cobalt air bag discussion item. he had no recollection of the involvement. i mean, right up to the point where they had an answer, and now this guy doesn't even remember working on it. was that pretty frustrating from a trial lawyer's perspective? >> one of the key problems we found is the lack the lack of documentation which led to lack of accountability. and i think a classic example of that is what happened in 2005. when we went back to find out why did they close the investigation into the cobalt issue. and we found ourselves in a position where there were no notes with regard to the matter. everybody at the meeting pointed to somebody else in the meeting as having

team, the group that terms the root cause of the problem, reviewed the claims relating to the cobalt nondeployment, but ultimately decided not to pursue it. why does the product investigations not pursue this matter at the time? >> that's a very good question. and the answer is this was some of those things passed off to another agency. mr. operation was keeping track of it. the other investigators weren't following up with regard to it. they were gathering information, if you will. but that's where they went with it. >> so when you say that it was kind of handed over somewhere else, you're referring to the product -- field performance assessment division? >> yes. >> because according to again our information, it says afz te proceed duct investigators declined to investigate, the responsibility for being tag e tracking the claims was assigned to the field performance assessment division. do you consider this unusual, would this be typical in a situation where there has been an obvious issue that has come to light and it just couldn't of be passed off to another -- and i guess i would like to know,

where gm investigators analyze this tread data to attempt to identify or exexplain airbag nondeployment in cobalts and -- is is that not correct? >> the answer is i cannot give you a number of where that was done. >> and the federal regulators also conducted analysis of the early-warning report data but were not able to separate the weak from the chaff, so to speak and identify the defect? >> the issue of the airbag, we noted that there were these nondeployments and gm's response to that was to begin the administration under mr. sprague, to keep apprised of what was taking place. there were no major further discussions about that issue until 2013. >> it seems that part of the problem here is that early-warning report data provided, are reported in 23 broad categories. and in the case of this defect, the early-warning data provided spanned several categories including engine, airbags and category of "other." and they are able to request more detailed information from auto manufacturers for individual warranty claims and field reports but it is difficult to know what is -- what to request given the

inconsistent recollection as to whether the product investigations group became involved in the cobalt air bag nondeployment issues at this stage. everest reports in april 7 the fda transitioned the cobalt matter to the p.i. group where it was taken on by an engineer. document in the file indicate that he was working on the issue and a may 42007 status review planning worksheet states that he was scheduled to present on an issue described as cobalt air bag discussion item. he had no recollection of the involvement. i mean, right up to the point where they had an answer, and now this guy doesn't even remember working on it. was that pretty frustrating from a trial lawyer's perspective? >> one of the key problems we found is the lack the lack of documentation which led to lack of accountability. and i think a classic example of that is what happened in 2005. when we went back to find out why did they close the investigation into the cobalt issue. and we found ourselves in a position where there were no notes with regard to the matter. everybody at the meeting pointed to somebody else in the meeting as having

witnesses have inconsistent recollection as to whether the product investigation became in the cobalt airbag nondeployment issues at this stage. the abridged report said april 7 the cobalt airbag letter to the ti group or was taken on by an engineer. documents indicate he was working on the issue in the may 4, 2007 status review presentation worksheets states he was scheduled to present on an issue described as cobalt airbag discussion item. they had no recollection of the involvement. they were right up to the point where they had an answer and now this guy does name remember working on it. was that frustrating from a trial lawyer's perspective? >> one of the key problems we found is the fact that the of documentation which led to a of accountability. a classic example of that is what happened in 2005 when they went back to find out why did they close the investigation to cobalt issue and we found ourselves in a position where there were no notes with regard to the matter. everybody at the meeting point to somebody else in the meeting as having responsibility for having close the matter that but we could

cobalt ignition switch was redesigned in '06, right? and testing documents from that time show that the torque of the redesign switch was still below specifications and yet after this change, the reported incidents of nondeployment in these vehicles dropped dramatically. when we look at that and we read those documents and the chairman mentioned, we have been through a million pages of documents and 15,000 pages from ntsa. we are not sitting idly, we are taking action. i want you to go back through this and elaborate that something could still not meet specifications and be acceptable for safety and i'd like to hear from you when it is okay to deviate from specifications and people in the process not be aware of this. >> well, i think when you look, as you start developing something, you have a design specification. what's most important in the testing we are doing now and have done in the past and are doing in a more broad fashion now, relates to the actual performance of the part and how it operates in a subsystem, in a broader system and how it operates in a vehicle. so, as we design now, we are validating that the part level with the new organization put together called the product integrity organization, th

something gm failed to do during over seven years of investigations into the mystery of cobalt airbag nondeployment. ms. barra, you sought this internal investigation of the ignition switch recall and you have publicly acknowledged how troubling its findings are. your company has cooperated with this committee investigation, and i thank you for that. you have taken corrective action by changing procedures and trying to remove roadblocks to make sure safety concerns come to light. base ed on this report, though, there are no easy fixes for the kind of systemic cultural breakdowns and fundamental misunderstoodings that permitted gm engineers not to suspect a safety problem when cobalts were stalling due to a faulty ignition switch. the possibility that these problems are pervasive and cultural deeply concerns me. concerns us all. we learned monday that gm has announced yet another recall. it's 39th since january. this one is hauntingly similar to the cobalt ignition switch recall. the ignition switch in certain buicks, chevies and add lacks inadvertently moves out of the run position if the key has t