Health & Safety Sentencing Guidelines – One Year on
Higher fines and greater consistency – that was the promise when new guidelines for sentencing health and safety offences came into force just over a year ago (on 1 February 2016). Well they got it half right.
According to a survey of sentences done by IOSH and Osborne Clarke LLP, in 2016 nineteen companies received fines of a million pounds or more compared to just three in 2015 and none in 2014. Perhaps even more startling, the authors calculated that the total income from the highest 20 fines in 2016 (£38.58 million) was higher than the total fine income for the 660 prosecutions successfully brought by the HSE in its reporting year of 2015/2016.
As the new guidelines started to bite, fines soared. £3 million for ConocoPhillips after gas leaked into a turbine hall (the guideline was not applied but the Court of Appeal said it should have been, and that the fine would have been at least this high if it had), £166,000 for Ineos Chlorvinyls (upheld on appeal) for polluting a canal with no lasting damage and very little effect on wildlife even in the short term. £1.8 million for G4S Cash Solutions for failing to mitigate the risk of legionella in its water systems; £5 million for Merlin Attractions following the Alton Towers accident. And, last month Havering Borough Council was fined £500,000 when a worker was badly injured, despite a hefty discount for being a public body and Thames Water was fined £20 million after sewage leaked from some of their plants.
One of the key areas of difficulty that is emerging is how to deal with very large companies. The guideline gives detailed starting points and ranges for companies with turnovers of £2 million or under (“micro”) up to those with a turnover of “£50 million or more” (“large”). A note at the top of the table suggests that where the defendant’s turnover “greatly exceeds” the threshold for large organisations “it may be necessary to move outside the range to achieve a proportionate sentence”.
The Guideline then invites the court to consider proportionality again at Steps Three and Four where the judge should “step back” and review the fine based on turnover to see that it fulfills the objectives of sentencing. At Step Three there is a check on “whether the proposed fine based on turnover is proportionate to the overall means of the offender” but that step also involves consideration of whether the fine is “sufficiently substantial to have a real economic impact” in order to bring home to the company the seriousness of the offence.
It makes for a very hard job for practitioners advising large companies who want to understand what the fine is likely to be on a plea, and whether or not it is worth fighting the case in purely economic terms. The lack of any specific starting points or bands apart from the general “over £50 million” category, leaves judges with a very wide discretion and a requirement to balance a large number of competing factors. Given that the Court of Appeal has told advocates that citing other first instance decisions, or even Court of Appeal cases, is bad practice (R v Thelwall  EWCA Crim 1755 at paragraphs 21-23) consistency is going to be hard to achieve.
In the ConocoPhillips case, the judge at first instance did not apply the guideline (wrongly, the Court of Appeal found) as the first ineffective sentencing hearing had been before it came into effect. Had he done so, the Judge would have had to balance the fact that a significant number of workers were put at risk of death or serious injury by the gas leak, against the company’s excellent health and safety record. He would have had to weigh up its co-operation with the investigation, what were said to be “systemic” failures, and of course the need for the fine to have “real economic impact” given the company’s turnover of £4.8 billion. Treacy LJ, rejecting the company’s appeal against the fine said for a company of this size the fine “could well have been greater” but did not set a figure, simply holding that £3 million was not manifestly excessive.
Compare that case with the Thames Water case. Their turnover was round £2 billion in 2014 and 2015 with an operating profit of around £600-700 million. The leaks were from a number of facilities, took place over a long period and caused considerable damage to businesses and the environment. The fine was £20 million. In passing sentence at Aylesbury Crown Court, HHJ Sheridan was reported to have said that “it should not be cheaper to offend than to take appropriate precautions” a sentiment which could introduce a whole new set of criteria into the sentencing process.
What level of fine does have a “real economic impact” on a company of the size of ConocoPhillips, Thames Water or Merlin Attractions? Would the £20 million handed out to Thames Water have been manifestly excessive in the ConocoPhillips case? If the fine has to have real economic impact, it is difficult to see what role mitigation (previous good record, one-off failing, no guilty intention etc.) can play unless fines for those lacking those factors go well over the real economic impact threshold.
Then there is the question of when judges will feel the need to “move outside the range to achieve a proportionate sentence”. In the Merlin Attractions case, HHJ Michael Chambers QC found that culpability was “high” and the harm category was “1”. He also found that thousands of people had been exposed to death or serious injury over the long period during which The Smiler ride was operated in an unsafe manner, and further that the offence had been the cause of a number of very serious actual injuries. The company had a turnover of around £400 million, however he did not feel the need to go outside the guideline to pass a proportionate sentence.
It is still comparatively early days for the Guideline. It may be that as more cases go through the Court of Appeal the way that very large companies are to be treated, and other disputed parts of the new system, will become clearer. However as things stand it would be a brave practitioner who is prepared to stick his neck out and give anything like the sort of specific predictions that companies may wish for when deciding whether or not to go to trial.