Julian Adams, who used to head up wholesale insurance at the regulator, agreed. He said: “The analysis is quite compelling and would have put Northern Rock as our highest risk firm in December 2006.”
Northern Rock failed between August and September 2007. Sigismund’s ‘Harm Metric’ did not go online until August 2007. He argued, however, that had it gone live earlier, he could have predicted the financial crash.
In an ‘Executive Harm Update’ sent at the end of May 2008, Sigismund wrote that Royal Bank of Scotland was at the ‘Top of the Harm list’.
“The update also contained a chart which indicated that the magnitude of RBS’s undercapitalisation was £28.8bn,” the tribunal heard.
While the watchdog knew RBS needed additional capital, having raised £12bn via a record rights issue a month earlier, Sigismund’s data suggested £12bn would not be enough. RBS failed in October 2008.
Sigismund also warned of the undercapitalisation of around 150 regulated firms in December 2007. “We note at this point that other firms, which the claimant identified as at imminent risk of failure, did not,” the tribunal said.
‘Considerable disagreement’ between supervisors
Around the time of the financial crash, banks were required to hold capital at least equivalent to 8 per cent of their risk weighted assets.
Sigismund had argued that the regulator should also use his metric, which judged capitalisation based on either market, credit, operational or liquidity risk.
But not everyone felt this method worked in practice. As the tribunal observed, there was “considerable disagreement” among supervisors as to the suitability of using his ‘Harm Metric’.
The regulator’s former chief economist, Peter Andrews, said banks were undercapitalised “for a long time” and that there was "no other alternative".
“If the FSA were to declare, as the claimant has suggested, that the majority of banks were undercapitalised, this would have destroyed market confidence and led to the huge economic costs of bank runs,” he said.
The tribunal did, however, say this illustrates how one statutory objective of the FSA/FCA “might be in conflict” with another.
Former chair of the FCA, Charles Randell, went as far as to suggest there was nothing “genuinely new” or particularly useful for the FCA in the ‘Harm Metric’.
Sigismund has since argued the “underuse” of his metric was down to a “lack of sufficient confidence and understanding of such sophisticated mathematical techniques”.
$200,000 a year spent on data
From May 2007, the regulator began paying for the data which underpinned the ‘Harm Metric'. By 2009, the risk department was spending around $100,000 (£90,575) per annum on data, and by 2013 this cost had doubled to $200,000 (£180,633).
