I am grateful to my colleague Colm Nugent for sending me a copy of the decision in Baker -v- British Gas Services (Commercial) Limited [2017] EWHC 2302. Amanda Yip QC (sitting as a Deputy Judge of the High Court*), considered the credibility of key witnesses for a defendant.  It is an example of the court assessing witness credibility when the witness evidence at trial differs from the witness statements.


“On paper, the evidence of Mr Moule and Mr Foxall looked strong. Having had the opportunity to assess them in the witness box, I was less impressed.”


The claimant was an electrician who had been seriously injured when he was electrocuted whilst working on a light fitting.  The primary issue was who was responsible for the fitting of the mis-wired light that had caused the accident. The second defendant had fitted the lights and passed tested them twelve years previously.  The second defendant called evidence from the electrician and the person who, it said, had inspected the installation after it was fitted.


26.Mr Foxall had started his electrical apprenticeship in 1965.  He is now retired.  In 2004, he was working for J & L.  His role on this project was  inspection and testing of the installation.  He had been qualified for that role since 1992.  It was his responsibility to test the installation and to issue an electrical installation certificate.  His evidence was that he carefully tested the installation, carrying out earth loop impedance testing at each light fitting and that no fault was found.  He also confirmed that Mr Moule was a careful electrician who would not have made an error of this nature.  The inference, if I accept his evidence, is that the fault could not have been present at the time he tested and approved the installation
28.On paper, the evidence of Mr Moule and Mr Foxall looked strong. Having had the opportunity to assess them in the witness box, I was less impressed.  Of course, I allowed for the fact that they were giving evidence about work they undertook over twelve years ago. In his statement, Mr Moule said that he could not remember the specifics of the job. That would be unsurprising.  Mr Moule said:
“If there had been anything specifically difficult or unusual during the job, I would most likely have remembered.  This was not the case and the installation was completely standard.” …
29.By contrast, in the witness box, he said that he could remember the job because it was a particularly easy one.  That is inconsistent with his statement.  He was asked about timings.  I note the following relevant evidence in his witness statement:
“The first test is undertaken before the installing electricians start the third fix stage. …. When John Foxall had completed the initial dead testing, I would proceed with the third fix. … Following the connection of the lights and on completion of all remaining third fix items, John Foxall would have proceeded to carry out the live tests.” 
In his oral evidence, Mr Moule said that he did not have to wait until the testing was completed before starting work on the final fix.  Rather, he would start work on one circuit while Mr Foxall was testing another.  In my judgment, that is very different from what he describes in his statement.  His explanation of the process which he called “conveyor belting” came about in my view when it became apparent that the time Mr Foxall had been on site would not have allowed sufficient time for the processes described in Mr Moule’s statement.  It is another significant inconsistency between his statement and his oral evidence.
30. Mr Moule told me that he could fit 15 to 20 light fittings per hour.  He would have been working on a ladder or platform and would need to move from one location to the next, taking each fitting to its location.  On his evidence, he would have only three to four minutes per fitting with no pauses in between.  Even allowing for him being an experienced electrician well practised at installing fittings of this sort, that seems a remarkably short time.  Mr Moule’s evidence as to timings also appeared to be contradicted by Mr Ray.  He agreed in cross-examination that the final fix could only start after dead testing was completed and thought that the fitting of the lights was likely to take “all day”.  Even allowing for Mr Ray not being as quick as those who regularly carried out the work in practice, his evidence suggested things would take longer than Mr Moule claimed
33. Contrary to Mr Ferris’s suggestion that Mr Foxall had a careful approach and could be described as “pernickety”, that was far from the impression he gave in the witness box. I made allowance for the considerable time that had elapsed since his involvement and for the fact that he is now retired.  However, I do not think this can explain the multiple inconsistencies between his witness statement and his oral evidence.  He told me initially that he could not remember the contract but later said he could remember undertaking the testing and inspection of the installation.  Asked about the time taken to complete dead testing, he said that he would not have tested every junction box but would just go to the last fitting on the circuit.  In his statement, he said he would test each and every junction box.  In oral evidence, he said that he would start live testing while Mr Moule was still working on other circuits.  That was not something that appeared in his statement.  When pressed as to whether it was safe to do that, Mr Foxall said that he would lock off the other circuit breakers with padlocks to prevent any danger.  Somewhat surprisingly, he suggested that would involve applying 46 padlocks.  There was no mention of that in his statement.  There was a discrepancy as to whether or not he would de-energise the circuit in between testing each fitting.  When asked about this he said there was an error in his statement.  I note that he did not seek to correct that error at the start of his evidence although he did correct another matter.  He then introduced the suggestion that he had used a volt stick at each light fitting (to explain why it was not necessary to de-energise the circuit between tests).  Not only did this not appear in his statement but when I asked Mr Foxall to carefully describe how he went about his testing he made no mention of the volt stick.  When asked about how he would record the readings taken at each fitting, he said he would remember them all and then write them down at the end.  He gave an example of holding eleven readings in his head (although some of the circuits including the relevant one had more than eleven fittings).  In re-examination, he said that it was not necessary to remember all the readings, he just had to write down the last one and check that the resistance was increasing as he moved along the circuit.  However, he had been quite specific about holding all the readings in his memory until he reached the end, even making a joke about being younger at the time. 
34. Mr Foxall also sought to cast doubt on his own timesheet. Mr Foxall spent two and half days at the premises.  It was suggested that he had insufficient time to carry out all the tests in the way that he said he had.  On the Tuesday afternoon and Wednesday, his time sheet records only that he was “testing”.  On the final day (Thursday), it notes “testing + wiring up ASU + fans”.  Mr Foxall suggested his time sheet was probably wrong as he could not recall wiring anything on the Thursday.  It seemed surprising that he could remember the detail of what he was doing so long ago.  His challenge to his own timesheet also seemed wholly inconsistent with the claim that he was a careful and pernickety employee.
35. Mr Foxall also accepted that as the inspector he should have signed the electrical installation certificate personally. He did not do so.  This, he said, was a “mistake”.  The certificate requires the signatures of the designer, the constructor and the inspector of the installation.  Mr Ray signed all three boxes.  He gave his name as the designer although he accepted he had not designed the circuit.  That had been done by his son and Mr Dudley.  Mr Moule and Mr Foxall should have signed the certificate personally but did not, although they were named as the constructor and inspector.  Mr Ray explained the need to sign the certificate and have it sent out to comply with their contractual obligations.  This adds weight to the suggestion that J & L came under some time pressure at the end of the job.  It is also another factor suggesting that Mr Foxall may not have been as careful and pernickety as was claimed.  In short, I did not find Mr Foxall to be a reliable witness.
36. I have to bear my assessment of Mr Foxall in mind when having regard to his test results. It is common ground that a proper inspection as described by him would have detected the wiring error if it was present from the start.  Mr Sykes and Mr Braund were of the view that the evidence suggested that it was unlikely that electrical tests were carried out on all the light fittings after they were connected to the junction boxes given the time available to Mr Foxall.  Mr Jones strongly disagreed with this.  In the joint statement, the experts agreed (at paragraph 3.4.4) that:
“The results of the tests carried out when the installation was commissioned appear credible, but we cannot be sure where when or how the measurements were made.”
37. Mr Jones placed reliance on Mr Foxall’s description of the tests performed and his evidence that his testing was thorough and correct to conclude that the fault cannot have been present at the time of testing. In light of the findings I have made about Mr Foxall’s evidence, the results of his tests have to be viewed with caution.  There is, as the experts have identified, uncertainty as to how the measurements were made.
38. On the basis of Mr Foxall’s evidence, I cannot be confident that he in fact live tested every light fitting as he claimed to have done. The relevant test results are to be found in the column headed: “Maximum measured earth loop impedance” on page B/227.  Again, that sheet has not been signed by Mr Foxall personally as it should have been.  There was a suggestion in the course of the trial that some or all of the earth fault loop impedance results could have been calculated by adding the readings in the column headed “R1 + R2” to the recorded earth fault loop impedance (Zs) for the distribution board.  On the relevant circuit, the measurement is exactly R1+R2+Zs as is the case for most but not all of the other circuits.  Mr Foxall told me that he was aware of occasions when other electricians would complete the relevant column by adding R1+R2 and Zs although he asserted that he would never do this and did not take shortcuts.
39. Having considered all the evidence, I am not persuaded that Mr Foxall thoroughly tested the circuit in such a way as to exclude the existence of the fault from the outset…”


The claimant succeeded against both defendants on a 75:25 basis.There was no deduction for contributory negligence.


  • Now appointed to the High Court.