Research conducted by Edinburgh-based Modelling Design Partners (MDP) and SolveXia in collaboration with Actuarial Post, revealed that few calculations to assess the risk of investments by large financial bodies in the UK are automated.
Most actuarial departments continue to use manual calculations, cutting and pasting data into Excel spreadsheets, a method that most actuaries accept increases the risk of catastrophic blunders, the study reveals.
The London Whale incident cost US investment bank JP Morgan $6.2billion because of a simple arithmetical miscalculation by a London-based quantitative expert while in the Reinhart-Rogoff study, human error produced flawed data that was used by politicians in Europe and America to justify austerity strategies.
Daniel de Bruin, managing partner of MDP, which provides consulting and software solutions to help businesses improve their financial models, said: “Spreadsheets are pervasive in many industries, due to their flexibility and perceived reliability. Actuarial departments are no exception: 90% of respondents indicated that their departments use them.
“Despite their ease of use, spreadsheets are notoriously difficult to audit and control. Frequent manual manipulation of data and equations is often required, which leaves spreadsheets vulnerable to human error.
“The insurance industry uses the same investment instruments as investment banks, within asset-liability hedging strategies. Undetected errors in these spreadsheets may lead to insurance executives not correctly estimating the risks within their portfolios, thereby increasing the risk of extreme losses in relatively ordinary market conditions.”
Subscribers to Actuarial Post were questioned between March 19 and April 7. The opinions of 94 respondents were gathered, including 14 at executive and board level, 18 in senior management, 31 in middle management, 21 analysts, and 10 in other roles, such as contractors and consultants. The majority of respondents are fellows of the Institute and Faculty of Actuaries.
Of those who responded, 51% were from the life insurance industry, 16% were from general consultancies, 10% from the general insurance industry and the remainder were from audit/accounting firms, asset management companies and academic institutions.
Some 90% said their firms used spreadsheets as part of the process to produce regular results while 69% said they were used for major/important calculations. In addition, 76% said that assumptions were stored and updated on spreadsheets.
Only 9% of respondents said they were completely satisfied that their calculation processes were not vulnerable to input error, while 49% said that they were dissatisfied.
Only 12% said they were completely satisfied their methods of calculation were transparent, while only one in ten were completely satisfied at the amount of effort required to run their calculations.
“The results revealed that the most pertinent challenge associated with actuarial processes is vulnerability to input error. Not surprisingly, the majority of respondents also indicated continued concern about regulatory requirements,” said de Bruin.
“Nonetheless, the survey results also show that an increasing number of actuarial departments use process automation. Process automation platforms, such as SolveXia, allow actuaries to have the flexibility of spreadsheets, while reducing the risk of error caused by manual input and also adding transparency and auditability.”
Notes:
The London Whale incident was caused when a London based quantitative expert for JP Morgan miscalculated a value-at-risk (VaR) model for a synthetic credit portfolio when a spreadsheet divided a set of figures by their sum instead of their average, as the modeler had intended. It is likely that this error likely had the effect of muting volatility by a factor of two and of lowering the VaR.
For more information go to http://www.modellingdesign.com/ and http://www.actuarialpost.co.uk/
South African born Daniel de Bruin founded MDP in July 2010 following a career in financial modelling consulting when he worked with some of the world’s biggest financial institutions including Santander, Munich Re, ING, Resolution and the Phoenix group.
MDP helps clients to choose the right technology investments for financial modelling solutions, improving their knowledge in IT strategy, sourcing, and project management.
It helps clients better align IT to their organisation’s financial modelling requirements and ensure their information security procedures comply with regulations.
