If real life doesn't mirror asset-liability management models, then ALM models should be reconstructed to reflect reality.
That's what Ortec Finance — a specialist risk management consultant based in Rotterdam, Netherlands — set out to do more than a decade ago.
The firm's newly launched “dynamic scenario generator” is as much about sharpening existing asset-liability management tools as it is a new way of monitoring investment risk within a broad set — up to hundreds — of different risk factors. The aim is to better prepare institutional portfolios to meet future liabilities, both in the short term and the long run, said Hens Steehouwer, head of research at Ortec.
Depending on the application of traditional scenario models, “what you typically see is that investors might use different models or different calibrations of the same model to determine the long-term vs. the short-term risk management strategies,” Mr. Steehouwer said. “What we've done is to come up with one framework that does both. That's important for consistency in defining your investment strategy and monitoring the strategy.”
Using a set of properties within scenario analysis that is more consistent with historical data can help to fine-tune investment decisions. DSG “does not predict but compares investment decisions under different economic environments, and therefore helps investors prepare for what might be coming their way,” said Lucas Vermeulen, London-based head of Ortec Finance U.K.
Already available in the Netherlands and launched in the U.K. earlier this quarter, DSG will be available elsewhere including the U.S., Canada, Australia and South Korea within the next few months. So far DSG is being used by institutional investors in portfolio construction, investment strategic planning and risk monitoring, Mr. Steehouwer said. Clients include some of Europe's largest pension funds, including PGGM, which manages the e103 billion ($134 billion) Pensioenfonds Zorg en Welzijn, Zeist, Netherlands.
“A lot of other (dynamic financial analysis) models are built for valuation and are based on finance theories,” Mr. Vermeulen said. “Therefore, they're not concerned about reality, but creating conditions to put a value on an (investment) instrument. … DSG started from empirical evidence rather than finance theories.”
One example is of the “25-sigma event,” which theoretically was only supposed to occur once every 100,000 years but ended up happening for several days in a row during the 2008-'09 financial crisis. “When reality set in, investors were shocked,” Mr. Vermeulen said.
Since the crisis, institutions including pension funds have put more emphasis on managing risk in the short term. “That puts a lot of long-term investors in a more difficult position because they have to navigate between long-term objectives and short-term constraints,” Mr. Vermeulen said.
Andrew Slater, Ortec's head of pension risk management for the U.K. added that DSG is “a new way of creating realistic scenarios as the foundation for asset-liability modeling.”
At Ortec, Mr. Steehouwer has been working to improve ALM and scenario models since he joined the company in 1996. Diving into the quality of the scenarios used by Ortec's ALM model at the time, he determined the methodology needed “to bear more resemblance to what was happening in the world around us,” both in the short-term and the long term, he said.
A key part of his work rested on a relatively unknown Ph.D. thesis published in Germany in 1984 titled “Digital Filter Design and Its Use in the Analysis of Economic Time Series.” The thesis addressed frequency domain methods to disentangle historical data into different components — a long-term component and short-term component.
Ortec's “first draft of the new model” was available in 2007 and took another two years to perfect before the company began introducing it to Dutch clients. In 2010, the model further evolved to provide analysis on a monthly basis from the previous annual timetable. “Having a good long-term strategy is no longer good enough,” Mr. Steehouwer said. “Investors need to monitor the strategy on a short-term horizon to see whether it's still the right one for them.”
Within DSG, liabilities are considered at the members' level, taking into account life events such as marriage, disability and other changes affecting retirement cash flows for individuals.
“We see a lot of possibilities on the (defined contribution) side” as well as the defined benefit sector, Mr. Slater said. “The traditional approach to lifestyle funds leaves a lot (of room) for improvement. For example, it's very static when what is needed is a dynamic model that reflects what happens in real life.”
From a corporate sponsor's point of view, such business decisions as mergers and acquisitions might be included in the model to help determine the company's ability to contribute to the pension fund.
“The aim is to provide more flexibility in portfolio management,” Mr. Vermeulen said, “which we think will result in the ability to better manage funding, investment and liability risk in the context of the (company's) balance sheet.”