Chapter 9: The government's response to the Lakanal House fire

Investigations immediately following the fire

¶9.1 Following the Lakanal House fire,⁹³⁰⁹³⁰ Phase 1 Report chapter 8.1. BRE was asked by DCLG to investigate the incident under a standing contract known as the "Investigation of Real Fires" project.⁹³¹⁹³¹ Crowder {BRE00043716/2} paragraph 5(a). It was to form part of BRE's continuing project monitoring fires of special interest. Mr Martin Shipp and Dr David Crowder carried out the investigation. Initially, Mr Shipp led the investigation⁹³²⁹³² Martin {Day256/30:19-25}. and supervised Dr Crowder's work,⁹³³⁹³³ Shipp {BRE00047594/9} page 9, paragraph 54. but later Dr Crowder took over as point of contact for the investigation.⁹³⁴⁹³⁴ Martin {Day256/30:25}-{Day256/31:1}. The police and LFB carried out their own investigations. They had statutory powers to control the scene of the fire and pursue relevant prosecutions and therefore BRE could examine the site only with their permission.⁹³⁵⁹³⁵ Martin {Day256/32:12}-{Day256/33:1}.

¶9.2 The purpose of BRE's investigation for DCLG was to establish whether the fire could be traced to a defect in the Building Regulations or the guidance in Approved Document B (and if so, whether there any changes to either of them should be recommended) or whether there had been a failure on the part of those carrying out the refurbishment to follow the guidance and comply with the regulations.⁹³⁶⁹³⁶ Crowder {BRE00043716/2} page 2, paragraph 5(a); Martin {Day256/26:5}-{Day256/27:15}.

¶9.3 The version of Approved Document B that applied to the refurbishment was the edition published in 2000 with amendments made in 2002.⁹³⁷⁹³⁷ Crowder {Day229/227:2-7}. See Chapter 6 for a summary of the requirements relating to the construction of external walls. Unlike the 2006 version, which provided that all insulation material used in the external wall of a building over 18 metres in height should be of limited combustibility, the 2002 version of Approved Document B required only that insulation used in ventilated cavities in external walls should be of limited combustibility. Accordingly, the guidance did not advise that the insulation in composite panels of the kind used at Lakanal House should be of limited combustibility. It did, however, advise that if the building were over 18 metres in height the external surfaces of the walls should be classified Class 0.⁹³⁸⁹³⁸ {CLG10000740/91}.

¶9.4 BRE first attended the scene at Lakanal House on 6 July 2009, three days after the fire.⁹³⁹⁹³⁹ Crowder {BRE00047668/24} page 24, paragraph 89. Dr Crowder said that he attended the site on about another six days that month.⁹⁴⁰⁹⁴⁰ Crowder {BRE00043716/27} page 27, paragraph 88 (b); Crowder {Day229/229:17-24}. During that time BRE's investigation included examining and removing one heavily damaged composite window panel and exploring the internal construction of the building. At a later stage, further composite panels were removed from the site for testing.⁹⁴¹⁹⁴¹ Crowder {Day229/230:11}-{Day229/231:15}. In the first week, contact between DCLG and BRE was mainly between Mr Shipp and Brian Martin, but thereafter Dr Crowder began to communicate directly with Mr Martin about the investigation.⁹⁴²⁹⁴² Crowder {Day229/230:1-10}. BRE reported regularly to Mr Martin about the progress of the investigation.⁹⁴³⁹⁴³ Crowder {BRE00043716/27} page 27, paragraph 88(b); Crowder {Day229/232:4-20}. Mr Martin also attended the site himself on one occasion to gain a better understanding of the building and what had happened.⁹⁴⁴⁹⁴⁴ Martin {Day256/34:11-20}; Crowder {BRE00043716/22} page 22, paragraph 76; Crowder {Day229/233:9-18}.

132

¶9.5 A Fire of Special Interest (FOSI) report was prepared by BRE, a draft of which was sent to Brian Martin on 16 July 2009. The final report was sent on 10 August 2009.⁹⁴⁵⁹⁴⁵ Crowder {BRE00047668/24} page 24, paragraph 91. A draft of the 16 July 2009 report is at {BRE00036261/3}, another at {BRE00036265/5} and an earlier draft from 7 July 2009 at {CLG00001693/2}. In that report BRE identified the fact that the external wall had included "polymeric panels" in various locations,⁹⁴⁶⁹⁴⁶ {BRE00032286/196}. i.e. that the panels had been made of a type of plastic that was combustible. At that stage, however, their exact composition had not been determined.⁹⁴⁷⁹⁴⁷ Crowder {Day230/6:19}-{Day230/7:4}. BRE also said that falling burning debris, probably from both the façade itself and the contents of Flat 79 on floors 11 and 12, had ignited fires lower down the building on floors 5 and 7. At that early stage it had not been possible to determine from the physical evidence precisely what was the source of the debris which had ignited the contents of each flat.⁹⁴⁸⁹⁴⁸ {BRE00032286/202}; Crowder {Day230/10:17}-{Day230/11:4}. Under the heading "Potential Implications for the Building Regulations" the report drew attention to the external spread of fire as a matter of potential importance.⁹⁴⁹⁹⁴⁹ {BRE00032286/203}. No explanation was offered, however, for the spread of fire and the report contained no information about the construction of the external wall. No conclusions were reached about whether it followed the guidance in Approved Document B or complied with the Building Regulations. At the time the report was prepared BRE had not tested any samples of the window panels.⁹⁵⁰⁹⁵⁰ Crowder {Day230/4:25}-{Day230/5:3}. As Dr Crowder explained, BRE was unable to provide a more detailed report in the time available based on its initial investigation. Unfortunately, the investigation remained incomplete.⁹⁵¹⁹⁵¹ Crowder {Day230/8:20-25}; Crowder {BRE00047668/36} page 36, paragraph 153; Crowder {Day230/10:11-14}.

¶9.6 On 14 July 2009, just 11 days after the fire, Brian Martin told someone who had offered to help DCLG with its investigation into the fire that, based on the information he had received so far, he did not think there would be any need for changes to Approved Document B.⁹⁵²⁹⁵² {CLG10003915} exchange between Brian Martin and Glenn Horton of Locke Carey, a firm of fire safety consultants. Mr Martin told us that his preliminary view, based on the reports he had received from BRE, was that the problem lay mainly with the original construction of Lakanal House. Although he recalled a discussion with BRE about external fire spread, he could not remember when it had occurred or what had been said.⁹⁵³⁹⁵³ Martin {Day256/38:8}-{Day256/41:14}.

¶9.7 Although BRE's investigation was then still at a very early stage, on 28 July 2009 Mr Martin sent an email to Mr Shipp stating that he was satisfied that there would be no need for BRE to visit Lakanal House again and that any further visits would have to be funded by someone else.⁹⁵⁴⁹⁵⁴ {BRE00043742}. Mr Martin told us that he had discussed the matter with his colleagues in the Fire Resilience directorate and that they were happy for BRE to drop the investigation.⁹⁵⁵⁹⁵⁵ {BRE00043742}. After it received Mr Martin's email BRE produced its final report into the incident on 10 August 2009.

133

¶9.8 Dr Crowder said that he had been disappointed⁹⁵⁶⁹⁵⁶ Crowder {Day230/17:1-5}. by DCLG's decision to curtail BRE's investigation because he had thought it important⁹⁵⁷⁹⁵⁷ Crowder {Day230/16:12-19}. and had wanted it to be completed.⁹⁵⁸⁹⁵⁸ Crowder {Day230/9:2-7}. He had wanted to understand better what had happened and why,⁹⁵⁹⁹⁵⁹ Crowder {Day230/10:21-22}. including the time at which the fire had spread vertically and the cause of the downward fire spread, which he said had not been common at the time.⁹⁶⁰⁹⁶⁰ Crowder {Day230/11:5-16}; {Day230/4:15-23}. When the investigation was shut down Dr Crowder had not begun to review Approved Document B⁹⁶¹⁹⁶¹ Crowder {Day230/12:13-23}. and he did not know how Mr Martin had come to the conclusion that there was no need for any changes to it.⁹⁶²⁹⁶² Crowder {Day230/22:5-15}.

¶9.9 According to Mr Martin, the decision to discontinue BRE's investigation was taken by officials in DCLG after discussion with Sir Ken Knight,⁹⁶³⁹⁶³ Martin {Day256/64:12-21}. Sir Ken Knight was the Department's Chief Fire and Rescue Adviser at the time. then the Chief Fire and Rescue Adviser. The decision proceeded on the basis that since the police and the LFB were pursuing their investigations, it was better to leave it to them, given the nature of the incident and its seriousness.⁹⁶⁴⁹⁶⁴ Martin {Day256/58:8-21}; {Day256/59:3-24}. That was consistent with Dr Crowder's understanding of the position.⁹⁶⁵⁹⁶⁵ Crowder {Day230/19:10-20}. Mr Martin also said he was concerned about whether the department's resources might be exhausted in the investigation of one incident if too much time was spent on the Lakanal House fire.⁹⁶⁶⁹⁶⁶ Martin {Day256/60:16-23}.

¶9.10 Although we accept that those were some of the reasons for the decision to terminate BRE's investigation, we consider that the decision was premature and find it hard to understand, given that its fundamental objective had not been achieved. Although the police and the LFB might have been expected to find out whether the Building Regulations and Approved Documents had been complied with, DCLG could have had no confidence, based on what had been investigated thus far, that they would or could reach a conclusion about whether the regulations and guidance were themselves satisfactory. Indeed, we note that, when BRE's investigations were terminated, there had been no substantial dialogue between its investigators and the police.⁹⁶⁷⁹⁶⁷ Crowder {Day230/16:20-25}. We also consider that conserving resources should not have been a priority in circumstances where six people had died and completion of BRE's investigation might prevent more deaths in the future.

¶9.11 Although Mr Martin told us that he thought that discussions about the downward fire spread at Lakanal House had taken place with BRE at an early stage, he also said that Sir Ken Knight had taken the view that it was not unusual for falling debris to result in secondary fires.⁹⁶⁸⁹⁶⁸ Martin {Day256/45:1-12}. The latter view seems to us to be consistent with DCLG's lack of interest in learning lessons from the incident, which is evident in Mr Martin's correspondence shortly after the fire⁹⁶⁹⁹⁶⁹ {CLG10003915} on 14 July 2014 Mr Martin told a third party that "Based on the snippets of info I've had so far I don't think there's any need for changes to ADB." and in his evidence to us.⁹⁷⁰⁹⁷⁰ Martin {Day256/47:10-22}; Martin {Day256/56:22}-{Day256/57:1}. He was clearly of the view that the 2006 edition of Approved Document B had dealt with the combustibility of the façade more effectively than the 2002 version⁹⁷¹⁹⁷¹ Martin {Day256/49:7-18}. and it is apparent that there was little appetite on his part or that of the department for revisiting the guidance, despite the fact that it did not specifically refer to the danger of fire spreading downward or the creation of burning droplets.⁹⁷²⁹⁷² Martin {Day256/51:15-19}.

134

¶9.12 We accept that Dr Crowder and Mr Shipp⁹⁷³⁹⁷³ Crowder {Day230/17:20-24}. were disappointed that their investigation was not able to reach a conclusion. However, it is a matter of some concern that there is no formal record of BRE's expressing that dismay to the government, let alone formally advising the government that it was important for more investigative work to be done. In circumstances where BRE thought that its investigation needed to be completed to understand the incident fully, it is surprising that it did not say so clearly in a letter to the department. Although in its report of 10 August 2009 BRE drew attention to some matters that called for further investigation,⁹⁷⁴⁹⁷⁴ Including communications between the LFB and the occupants at Lakanal - see {BRE00032286/203}. that did not include the spread of fire across external walls.⁹⁷⁵⁹⁷⁵ See {BRE00032286/203} and contrast the wording at penultimate and antepenultimate paragraphs. While Dr Crowder explained that omission as a poor choice of language,⁹⁷⁶⁹⁷⁶ Crowder {Day230/14:11-22}. that does not do justice to complete silence on the subject. The episode suggests a lack of independence on the part of BRE and a failure to provide robust advice to the government on reasonable steps necessary to ensure safety from fire.

Sir Ken Knight's report to the Secretary of State

¶9.13 On 30 July 2009 Sir Ken Knight submitted a report to the Secretary of State on the questions arising from the Lakanal House fire in which he said that the problem of fire spreading over external walls was not significant and that initial enquiries did not suggest that it had occurred in a new way.⁹⁷⁷⁹⁷⁷ {HOM00045791/21} paragraph 39. Sir Ken told us that at the time the Lakanal House fire was not seen as a fire in an external wall because the cladding system itself had not caused it.⁹⁷⁸⁹⁷⁸ Knight {Day245/165:8-18}. However he also accepted that the downwards spread of fire had been unusual⁹⁷⁹⁹⁷⁹ Knight {Day245/170:20-23}. and "very much a one-off".⁹⁸⁰⁹⁸⁰ Knight {Day245/187:7-16}. He also accepted in hindsight that it would have been helpful if, as the government's Chief Fire and Rescue Adviser, he had asked for an investigation to determine the extent to which the external window panels at Lakanal House had promoted the downward spread of fire.⁹⁸¹⁹⁸¹ Knight {Day245/172:18}-{Day245/173:14}. We agree.

Further testing by the LFB of the external wall panels

¶9.14 Following the termination of its investigation for DCLG, BRE was engaged by the police and the LFB to assist them with their investigations into the fire.⁹⁸²⁹⁸² Crowder {BRE00043716/28} page 28, paragraph 88(d); Crowder {Day230/28:7-10}. That included carrying out reaction to fire tests on some of the external wall panels taken from the building after the fire. In particular, in November 2009, surface spread of flame tests were carried out on three types of the external wall panels with insulating cores in accordance with BS 476-7. Three different colours of insulating foam had been found on site and tests were carried out on all three.⁹⁸³⁹⁸³ Crowder {Day230/24:19}-{Day230/25:15}.

135

¶9.15 The panels performed poorly in those tests, most of which were directly observed by Dr Crowder.⁹⁸⁴⁹⁸⁴ Crowder {Day230/80:21}-{Day230/81:1}. They all achieved only Class 3, meaning that the flames spread a considerable distance across them.⁹⁸⁵⁹⁸⁵ BRE's Investigation of Real Fires Report dated 17 November 2009 {BRE00032286/243}. They were therefore incapable of achieving national Class 0, which requires Class 1 to be achieved in a test under BS 476-7. In addition, incandescent spalling was observed during all test runs and specimens were flaming strongly at the end of the tests. During some of the tests flaming debris was observed on the floor.⁹⁸⁶⁹⁸⁶ {BRE00032286/242} section 4.2. Those observations were worrying because they showed that the combination of materials in the panels was dangerous.⁹⁸⁷⁹⁸⁷ Crowder {Day230/82:11-14}. The material behaved so badly that BRE feared its equipment might be damaged if it carried out fire propagation tests in accordance with BS 476-6.⁹⁸⁸⁹⁸⁸ Crowder {BRE00043716/8} paragraph 18. It is to be noted that no further reaction to fire tests (including testing to BS 476-6) were ever carried out on the panels from Lakanal House.

¶9.16 Dr Crowder reported the results of the tests to Mr Martin in November 2009. Although contact between BRE and DCLG on the subject of the Lakanal House fire had diminished by this time, he recalled a telephone call on about 18 November 2009 with Brian Martin during which it was likely that they had discussed the results of the BS 476-7 tests.⁹⁸⁹⁹⁸⁹ Crowder {Day230/26:23-25}. He said that he had told Mr Martin that during the tests the panels had produced burning embers which had fallen out of the test rig.⁹⁹⁰⁹⁹⁰ Crowder {Day230/30:20}-{Day230/31:1}. Dr Crowder regarded the downward spread of fire as unusual.⁹⁹¹⁹⁹¹ Crowder {Day230/34:24}-{Day230/35:1}. That information was provided informally and was never committed to writing, despite the fact that it was important and BRE had a contract with DCLG to inform it of any failure in the regulations or in compliance with them.⁹⁹²⁹⁹² Crowder {Day230/32:12}-{Day230/33:6}.

¶9.17 Although Dr Crowder did not recall the details of his telephone call with Mr Martin, he did recall having been told, either by Mr Martin or Mr Shipp, that the problem with the external wall panels at Lakanal House had been one of a failure to follow the guidance in Approved Document B or to comply with the Building Regulations. It was not that the panels had been compliant with the Building Regulations but had nevertheless produced burning debris.⁹⁹³⁹⁹³ Crowder {BRE00043716/22} page 22, paragraph 78. As far as he could recall, he had been told by Mr Martin or Mr Shipp that if the panels had been Class 0, they would not have produced burning debris during a BS 476-7 test, but his recollection of that was not very clear.⁹⁹⁴⁹⁹⁴ Crowder {Day230/31:2-8}. Mr Martin could not remember having told Dr Crowder that the problem with the panels was one of non-compliance,⁹⁹⁵⁹⁹⁵ Martin {Day256/71:1-17}. although he agreed that it was something he was likely to have discussed with him. On the whole, we think it likely that Dr Crowder's recollection is reliable and it is consistent with an unwillingness within DCLG to accept that there was anything fundamentally wrong with the 2002 version of Approved Document B. It is troubling that the department should have reached a conclusion of any kind without having taken expert advice based on a full understanding of what had occurred.

136

¶9.18 BRE was not asked to provide a report on the policy implications of the downward spread of fire at Lakanal House.⁹⁹⁶⁹⁹⁶ Crowder {Day230/51:3-6}. It drew the department's attention informally to the problem of fire spreading downwards and debris falling from burning panels,⁹⁹⁷⁹⁹⁷ Under the Investigation of Real Fires contract there was an obligation on the BRE to maintain informal, direct and regular contact with DCLG policy officers - see {BRE00000951/44} second paragraph. but the department does not appear to have given any serious consideration to the implications that might have had for the regulatory regime. In the years before the investigation into the Lakanal House fire, fire safety staff at BRE were not allowed to hold direct conversations with the department's officials outside formal meetings attended by senior managers.⁹⁹⁸⁹⁹⁸ Crowder {Day230/43:5-19}. However, there do not appear to have been any formal meetings to discuss the downward spread of fire or the danger posed by falling debris following the Lakanal House fire.

¶9.19 Brian Martin gave Dr Crowder the impression that since the external wall and other aspects of the construction of Lakanal House had not complied with the relevant regulations and guidance, he doubted that there was any problem with the guidance itself.⁹⁹⁹⁹⁹⁹ Crowder {Day230/87:19}-{Day230/88:7}; {Day230/89:11-15}. Further, Dr Crowder's impression was that Mr Martin had decided that there were other explanations for the downwards fire spread at Lakanal House which did not point to any fundamental problem with the regulatory regime.¹⁰⁰⁰¹⁰⁰⁰ Crowder {Day230/41:2}-{Day230/42:13}. We understood Dr Crowder to mean that because burning debris might have fallen from the flats that had caught fire, it could not be proved that falling debris from the exterior wall panels had caused the downward spread of fire. A number of times during their evidence both Dr Crowder and Mr Martin gave that as an explanation for the failure to investigate the downward spread of fire.¹⁰⁰¹¹⁰⁰¹ Martin at {Day256/49:7}-{Day256/50:6}; Crowder {BRE00043716/22} paragraph 78; Crowder {Day230/39:14}-{Day230/41:11}. See also {CLG10005528/1} email dated 4 March 2013 from Brian Martin to Anthony Burd and others. However, that overlooks the worrying behaviour displayed by the composite panels in the BS 476-7 tests. Once that evidence had become available, there was no reasonable excuse for the department's failure to obtain further advice on whether the Building Regulations and the guidance in the Approved Documents were sufficiently robust. Its failure to do so was a serious abdication of responsibility.

BRE's Investigation of Real Fires report: November 2009

¶9.20 BRE covered the Lakanal House fire in the Investigation of Real Fires Report which it produced on 17 November 2009 but it made no recommendations for changes to the Approved Documents or for further investigative work.¹⁰⁰²¹⁰⁰² {CLG00019142/9} item 23; {CLG00019142/16}. In addition, the report contained the familiar phrase that the findings from the project reaffirmed the overall effectiveness of the Building Regulations and Approved Document B in providing for the safety of life in the event of fire.¹⁰⁰³¹⁰⁰³ {CLG00019142/3}. Despite being one of the authors of the report,¹⁰⁰⁴¹⁰⁰⁴ Crowder {Day230/37:23}-{Day230/38:8}. Dr Crowder could not explain why it did not refer to the downward spread of fire spread at Lakanal House or to the fact that Approved Document B gave no guidance on how that might be prevented.¹⁰⁰⁵¹⁰⁰⁵ Crowder {Day230/38:17}-{Day230/39:12}. He did say, however, that Martin Shipp had decided what went into the the report.¹⁰⁰⁶¹⁰⁰⁶ Crowder {Day230/37:15}-{Day230/39:2}.

137

Failure to learn wider lessons from the Lakanal House fire

¶9.21 On 14 December 2009, five months after the Lakanal House fire, Commissioner Dobson wrote to Sir Ken Knight to express the LFB's concern about the panels forming part of the external walls of Lakanal House and the fact that similar panels on other buildings could present a risk of fire spreading across or through the walls. In his letter the Commissioner explained that BRE's tests had demonstrated that the composite panels used at Lakanal House did not comply with the requirements of the Building Regulations in relation to the spread of fire over external walls or the guidance in Approved Document B on how those requirements might be met.¹⁰⁰⁷¹⁰⁰⁷ {LFB00104291}. The tests to which he referred were the BS 476-7 surface spread of flame tests to which we have referred above. The Commissioner went on to say that it was the LFB's understanding that the type of panel tested had been supplied by more than one company. He suggested that either Sir Ken Knight or the department should write to housing providers across the country advising them to check that external wall panels in high-rise housing stock met the correct specification and to include such checks in fire risk assessments for the relevant properties. The LFB was concerned that a fire similar to that which had occurred at Lakanal House could occur at any number of buildings if similar panels had been used.¹⁰⁰⁸¹⁰⁰⁸ Dobson {Day210/122:20}-{Day210/123:11}.

¶9.22 Sir Ken responded on 22 December 2009 saying that the Metropolitan Police had recognised that the results of the tests carried out by BRE had tended to show that such panels did not comply with the requirements of the Building Regulations or the guidance in Approved Document B. However, he thought there was insufficient information at that time to warrant alerting housing authorities or property owners generally to the matters raised, in particular because it had not yet been determined whether the ability of the exterior panels at Lakanal House to resist the surface spread of flame had been specified by the housing authority and, if so, whether panels of the kind specified had been fitted.¹⁰⁰⁹¹⁰⁰⁹ {HOM00046018}.

¶9.23 In a letter dated 22 December 2009 addressed to the chief executives of local housing authorities in England, Terrie Alafat, the Director of Housing Delivery and Homelessness at DCLG, said that the department had been made aware of concerns about window and cladding systems used on high-rise blocks and reminded them of what had become section 12 of Approved Document B. If they had any doubt about the application of, or compliance with, the regulations she advised them to seek the assistance of the local Building Control body.¹⁰¹⁰¹⁰¹⁰ {LFB00041215}.

¶9.24 In our view, that letter was an inadequate response to the problem exposed by the Lakanal House fire because it did not encourage local housing authorities to take active steps to ensure that their high-rise residential buildings complied with the Building Regulations. Sir Ken said that he would have expected them to carry out checks on receipt of that letter if they were not immediately aware of the materials used in the external walls of their high‑rise buildings,¹⁰¹¹¹⁰¹¹ Knight {Day246/4:5-19}. but we regard that as unrealistic. When asked why the department had not adopted the approach advocated by Commissioner Dobson, Sir Ken initially said that it was not possible to refer specifically to defects at Lakanal House while a police investigation was continuing,¹⁰¹²¹⁰¹² Knight {Day245/199:2-4}. but the contemporaneous correspondence between Sir Ken and the department shows that the Crown Prosecution Service was willing to consider requests for disclosure in the interests of public safety.¹⁰¹³¹⁰¹³ {CLG00019120/2}. In any event, it would have been possible for the department to ask housing authorities to carry out checks on the external walls of their high-rise buildings without referring specifically to Lakanal House. We think it should have done so and that the decision to send out a very broad message was a misjudgement.¹⁰¹⁴¹⁰¹⁴ {CLG10004194/3}.

138

¶9.25 There is no evidence that any further attempt was made by the department to understand the implications of the tests carried out by BRE in order to discover whether the risk posed by similar panels was likely to be repeated elsewhere.¹⁰¹⁵¹⁰¹⁵ {CLG00019120/2}; Knight {Day245/207:18}-{Day245/208:3}. Whether it should take a more active approach was therefore not considered at any stage, despite concerns raised by the LFB at a meeting on 21 April 2010, at which Mr Martin was present, about the quality of construction and checking of newly built and newly refurbished residential properties.¹⁰¹⁶¹⁰¹⁶ {LFB00025654/13}. Following a significant number of fires during the previous two years that had exhibited an unusual spread of smoke or fire, the LFB had become worried that the regulations covering the construction and approval of building work were not being enforced with sufficient vigour to ensure that new residential buildings in London were fit for their purpose.

¶9.26 Mr Martin candidly accepted that the department had concentrated on the need for amendments to Approved Document B and had not concerned itself with whether there might be existing high-rise buildings with combustible insulation materials in their external walls.¹⁰¹⁷¹⁰¹⁷ Martin {Day256/78:11}-{Day256/79:9}. Mr Martin's explanation was that the department had been concentrating on improving building standards for the future and did not consider examining the effect of previous standards to be part of its function.¹⁰¹⁸¹⁰¹⁸ Martin {Day256/79:10-25}. In our view that was not a responsible approach to take to a fire in which six people had died.

¶9.27 A detailed investigation into the fire performance of composite panels with foam cores of the kind used at Lakanal House was never carried out,¹⁰¹⁹¹⁰¹⁹ Crowder {Day230/62:8-12}; {Day230/74:12}-{Day230/78:3}. primarily because BRE's investigation was terminated prematurely.¹⁰²⁰¹⁰²⁰ Crowder {Day230/79:19}-{Day230/80:13}. Such an investigation might have been of benefit to the building industry,¹⁰²¹¹⁰²¹ Crowder {Day230/78:4-11}. although Dr Crowder did not think that high-pressure laminate panels with foam cores were widely used.¹⁰²²¹⁰²² Crowder {Day230/78:15}-{Day230/79:18}. That view may have been based on the absence of other fires involving similar panels,¹⁰²³¹⁰²³ Crowder {Day230/82:15}-{Day230/83:16}. but that was scarcely a sound basis for drawing a conclusion of that kind. At all events, BRE did not advise DCLG that it might be important to investigate the prevalence of that particular kind of panel.¹⁰²⁴¹⁰²⁴ Crowder {Day230/79:19}-{Day230/80:13}.

The inquests into the deaths at Lakanal House

¶9.28 The hearings at the inquests into the deaths at Lakanal House took place between 14 January and 28 March 2013. Both Dr Crowder and Brian Martin gave evidence. Before they did so, the coroner heard evidence from Mr David Walker, a chartered building surveyor who had been appointed as expert to the inquests. Mr Walker's initial view (which was wrong) was that the composite panels in the window sets on the external wall of the building had to satisfy certain requirements in relation to fire resistance (i.e. resistance to the passage of heat and flame), as well as having surfaces classified Class 0 for reaction to fire. When seeking to apply the guidance in Approved Document B, Mr Walker initially told the coroner that the panels beneath the bedroom windows should have had 30 minutes' fire resistance,¹⁰²⁵¹⁰²⁵ {INQ00015064/102} lines 4-6. but he changed his opinion during cross-examination to say that the requirement was 120 minutes.¹⁰²⁶¹⁰²⁶ {INQ00015065/103} lines 13-16. Finally, after reflection overnight, he said that the panels were not actually required to be fire resistant at all.¹⁰²⁷¹⁰²⁷ {INQ00015066/29} line 11. It seems that, when giving evidence to the inquests, Mr Walker struggled to interpret the provisions of the Building Regulations and Approved Document B. The coroner commented during Mr Walker's evidence that she found it extraordinary that a document intended to be used by contractors, subcontractors and others wanting to do building work was, in her words, "impenetrable".¹⁰²⁸¹⁰²⁸ {INQ00015065/120} lines 14-17.

139

¶9.29 Although both Dr Crowder and Mr Martin quite properly sought to correct the misunderstandings which had arisen as a result of Mr Walker's evidence, aspects of their evidence relating to the external wall of Lakanal House were not as clear as they should have been, as we explain below. Mr Martin was defensive about the clarity of Approved Document B and dismissive of suggestions that it could be improved. Those are important matters because they are consistent with the longstanding pattern on the part of both the department and BRE of failing to grapple with problems relating to the guidance in Approved Document B on the construction of external walls. As a result, the coroner and the jury were not as well informed about the limitations of Approved Document B, and Class 0 in particular, as they should have been.

Class 0 and combustibility

¶9.30 There were a number of passages in Dr Crowder's evidence at the inquests in which he confused Class 0 with limited combustibility. In particular, when explaining the meaning of Class 0, he indicated that the surface of the material "should be of limited combustibility".¹⁰²⁹¹⁰²⁹ {INQ00015064/8} lines 4-5; Crowder {Day230/116:16}-{Day230/17:23}. That was fundamentally erroneous (and, to be fair, Dr Crowder admitted to us that it had been a mistake), because a material that was not of limited combustibility might still be able to satisfy Class 0 if it achieved the required results when tested in accordance with BS 476-6 and 476-7.¹⁰³⁰¹⁰³⁰ Crowder {Day230/116:16}-{Day230/117:23}. In circumstances where the coroner was also told that a material of limited combustibility was capable of burning, but not liable to burn unless under an imposed heat source,¹⁰³¹¹⁰³¹ {INQ00015064/7} lines 14-15; Crowder {Day230/110:3-13}. anyone listening to the evidence might well have obtained the impression that products with Class 0 surfaces normally perform adequately in fire so as to satisfy Regulation B4(1) of the Building Regulations. The fact that some products may perform very badly was never explained. Much of the confusion appears to have been due to a failure to distinguish clearly between the reaction of a product to fire and the ability of its surface to resist the spread of fire.

¶9.31 In addition, when Dr Crowder was asked what difference it would have made if the surfaces of the composite panels at Lakanal House had achieved Class 0, he said that he would not have expected them to ignite and burn on their own.¹⁰³²¹⁰³² {INQ00015064/11} lines 11-12; Crowder {Day230/118:2-15}. At a later stage he also said that if the panels had been Class 0 they should not have burned.¹⁰³³¹⁰³³ {INQ00015064/20} line 25; Crowder {Day230/121:7}-{Day230/122:7}. Both of those statements failed to reflect the fact that both the department and BRE (including Dr Smith and Dr Colwell, though not Dr Crowder¹⁰³⁴¹⁰³⁴ Crowder {Day230/183:2-14}.) knew by that time that products with Class 0 surfaces could perform very poorly in fire, as the cc1924 tests had shown.

140

¶9.32 A similar confusion between Class 0 and limited combustibility was made in Dr Crowder's written report to the Metropolitan Police on the Lakanal House fire, in which he said that the recommendation to use Class 0 materials meant that walls in common and circulation spaces should be of very limited combustibility.¹⁰³⁵¹⁰³⁵ {BRE00005881/18}; Crowder {Day230/122:24-25}. He admitted that that had been a poor choice of words.¹⁰³⁶¹⁰³⁶ Crowder {Day230/122:17}-{Day230/123:8}. We think it was more than that. It was the perpetuation of a confusion that had bedevilled official thinking about Class 0 at least since the Knowsley Heights fire.

¶9.33 Misleading statements of a similar kind were also made by Mr Martin in the course of his evidence to the inquests. In his witness statement he said that the 2002 version of Approved Document B had introduced the principle of limiting the combustibility of external walls of tall buildings to reduce the danger of fire spreading up the face of the building, which was related to the need to use Class 0 products for walls above 18 metres in height.¹⁰³⁷¹⁰³⁷ Martin {CLG00019202/2} page 2, paragraph 7. That implied that Class 0 limited the combustibility of materials used in external walls, although materials which have a Class 0 surface can be highly combustible, as demonstrated by the composite polyethylene cored panels used in the cc1924 test. Mr Martin told us that Class 0 was one of a range of classifications of combustibility,¹⁰³⁸¹⁰³⁸ Martin {Day256/125:19}-{Day256/127:1}. but that is technically incorrect. The main tests which are relevant to Class 0, including BS 476-6 and 476-7, are tests which measure fire propagation and surface spread of flame respectively. They are not tests which measure the combustibility of a material; those are BS 476-4 and BS 476-11.

¶9.34 In his evidence at the inquests Mr Martin said that if the panels caught fire they probably were not Class 0¹⁰³⁹¹⁰³⁹ {INQ00015070/51} lines 15-16. and that a Class 0 material would burn but not very much.¹⁰⁴⁰¹⁰⁴⁰ {INQ00015070/72} line 25. He also said that the non-combustible classification was only slightly more stringent than Class 0.¹⁰⁴¹¹⁰⁴¹ {INQ00015070/57} lines 21-22. When he gave evidence to us, Mr Martin accepted that some of those answers were imprecise, although he again referred to what he described as a "sliding scale of combustibility"¹⁰⁴²¹⁰⁴² Martin {Day256/145:2-19}. and explained that he might not have understood some of the limitations of Class 0 at the time.¹⁰⁴³¹⁰⁴³ Martin {Day256/141:8}-{Day256/142:24}. We do not accept that explanation, which is directly contradicted by what Mr Martin and those at BRE undoubtedly knew as a result of the cc1924 tests in 2001.¹⁰⁴⁴¹⁰⁴⁴ As to which see Chapter 7. Indeed, Mr Martin accepted that he probably should have made the connection between Class 0 products and the fires in the UAE the year before, in which composite materials had burned ferociously.¹⁰⁴⁵¹⁰⁴⁵ Martin {Day256/146:11-16}.

¶9.35 Although neither Dr Crowder nor Mr Martin was asked in terms to comment on the suitability of Class 0 as the sole criterion adopted in Approved Document B for panels to be used in the external wall of Lakanal House and high-rise buildings generally, we find it surprising that neither of them thought it appropriate to draw the coroner's attention to what they knew to be its limitations.

141

¶9.36 Nor did either of them draw attention to the way in which the fire had spread downwards, which in their experience had been an unusual feature of the incident.¹⁰⁴⁶¹⁰⁴⁶ Crowder {Day230/4:15-21}. It is unfortunate, therefore, that the unusual way in which the fire had spread and the absence from Approved Document B of any guidance on how to prevent it was not considered during those proceedings.

Approved Document B

¶9.37 Although Mr Martin had been asked to include in his witness statement for the inquests a description of the extent to which requirements for the fire resistance of external walls and the use of Class 0 panels had been revised in the 2006 edition of Approved Document B,¹⁰⁴⁷¹⁰⁴⁷ Martin {CLG00019202/4} page 4, paragraph 17 (question (d)). he did not include in his response any reference to the fact that the 2006 edition stated that all insulation materials in external walls should be of limited combustibility.¹⁰⁴⁸¹⁰⁴⁸ Martin {Day256/131:18}-{Day256/132:20}. Nor did he mention the addition of the word "filler" to what had become paragraph 12.7 or explain what that word had been intended to capture. Mr Martin accepted that in hindsight it would have been useful to identify the key textual changes to Approved Document B as well as changes in its structure and format.¹⁰⁴⁹¹⁰⁴⁹ Martin {Day256/131:21}-{Day256/134:24}.

¶9.38 It was suggested to Mr Martin during questioning at the inquests that it might be desirable for Approved Document B to be revised so that it was clearer to those who had to use it. Mr Martin disagreed and said that enquiries he had received suggested that most professionals in the industry seemed capable of applying it without too much difficulty,¹⁰⁵⁰¹⁰⁵⁰ {INQ00015070/43} lines 22-24. and that, having not received any indication to the contrary, he did not think that any revision to Approved Document B was necessary.¹⁰⁵¹¹⁰⁵¹ {INQ00015070/43-44} lines 25-1. As we see below, however, in fact Mr Martin did know that Approved Document B was not a model of clarity.

¶9.39 Although Mr Martin thought that Approved Document B was straightforward, the coroner, who was not unfamiliar with the construction industry, clearly found it difficult to understand. Despite what he told her, by March 2013 Mr Martin had become well aware that there was widespread confusion in the industry about the meaning of parts of Section 12 of Approved Document B dealing with the construction of external walls. In late 2010 he had been told that a former senior head of building control for several local authorities who was advising the LFB had misunderstood functional requirement B4 and the guidance in Approved Document B in a way that left him "horrified".¹⁰⁵²¹⁰⁵² {LFB00052135/4}; {HOM00046917}; Martin {Day256/87:13}-{Day256/88:18}. In January 2013, just a few weeks before he gave evidence at the inquests, Mr Martin also told a colleague who had asked for his advice on the use of panels that were not made of materials of limited combustibility, that Approved Document B needed to be read two or three times to work out what it meant.¹⁰⁵³¹⁰⁵³ {CLG00019193/1}.

The jury's verdicts

¶9.40 Verdicts were returned by the jury in the Lakanal House inquests on 28 March 2013. They included two important findings: first, that the composite panels beneath the bedroom windows were not Class 0, contrary to the requirements of Approved Document B; secondly, that even if they had been Class 0, that would not have prevented the fire from spreading from Flat 65 to Flat 79, although the spread of fire within Flat 79 would have been slower.¹⁰⁵⁴¹⁰⁵⁴ {INQ00015079/40} lines 14-18. Those findings should have been a warning to the department that building professionals were not aware of, or were misinterpreting or ignoring, Approved Document B. They should also have served as a warning that the effectiveness of the Building Regulations and Approved Document B to protect life were in question. The department ought to have undertaken further investigations to find out whether the use of unsafe panels in the external wall of Lakanal House was an isolated incident or had been due to systemic ignorance or misunderstanding of the Regulations and Approved Document B. However, the department failed to undertake any such investigation.

142

The coroner's rule 43 letter

¶9.41 On 28 March 2013, the coroner wrote a letter to the Secretary of State for Housing, Communities, and Local Government, The Right Honourable Eric Pickles MP, now Lord Pickles, under rule 43 of the Coroners Rules.¹⁰⁵⁵¹⁰⁵⁵ {CLG00001870}. The letter addressed four separate matters: (1) fire safety, firefighting and search and rescue, (2) fire risk assessments pursuant to the Fire Safety Order, (3) retro-fitting sprinklers in high-rise residential buildings and (4) the Building Regulations and Approved Document B.

¶9.42 In relation to the last matter she said:

"During these inquests we examined Approved Document B (2000 edition incorporating 2000 and 2002 amendments) ("AD B"). I am aware that AD B has subsequently been amended, and believe that a further amendment is due to be published soon. The introduction to AD B states that it is " ... intended to provide guidance for some of the more common building situations". However, AD B is a most difficult document to use. Further, it is necessary to refer to additional documents in order to find an answer to relatively straightforward questions concerning the fire protection properties of materials to be incorporated into the fabric of a building. It is recommended that your Department review AD B to ensure that it • provides clear guidance in relation to Regulation B4 of the Building Regulations, with particular regard to the spread of fire over the external envelope of the building and the circumstances in which attention should be paid to whether proposed work might reduce existing fire protection • is expressed in words and adopts a format which are intelligible to the wide range of people and bodies engaged in construction, maintenance and refurbishment of buildings, and not just to professionals who may already have a depth of knowledge of building regulations and building control matters • provides guidance which is of assistance to those involved in maintenance or refurbishment of older housing stock, and not only those engaged in design and construction of new buildings."

¶9.43 On receipt of that letter, and in the light of the observations of the coroner at the inquests and the jury's verdicts, the department could have been in no doubt that the guidance in Approved Document B, particularly on the construction of external walls, had been found to be lacking in clarity. Even though specific problems had been identified at the inquests, any official paying proper attention should have appreciated that the coroner's concerns were not limited to matters of structure and language but extended to the substance of the guidance on the construction of external walls, which was regarded as inadequate. That should have prompted an immediate review of that part of the guidance, particularly since the department already knew that there were existing concerns about the suitability of Class 0 as an indicator of the suitability of materials for use in external walls and that some building control officers were having difficulty in interpreting the guidance.

143

Conclusions

¶9.44 In our view the department's response to the Lakanal House fire was inadequate. The BRE's investigation was prematurely and unreasonably curtailed and at every stage, even including the inquests, officials displayed a complacent and short-sighted attitude towards learning wider lessons from the fire. That included lessons about the adequacy of the regulatory regime itself. Although we recognise that the Lakanal House fire had particular features that were not directly relevant to what occurred later at Grenfell Tower, there were lessons that could and should have been learnt from it which might have improved the robustness and clarity of the regulatory regime before the Grenfell refurbishment took place.

¶9.45 In particular, we consider that more should have been done to investigate the propensity of composite panels to contribute to the downward spread of fire, the adequacy and clarity of the statutory guidance on the construction of external walls, including the suitability of Class 0 as a criterion and whether the use of materials liable to create flaming droplets when exposed to fire ought to be regulated. The Lakanal House fire also provided an opportunity to consider whether more general problems existed in the construction industry, including a failure within it to understand or apply the regulations and guidance and a resulting stock of tall buildings with external walls containing combustible materials.

145