F&C's Jeroen Wilbrink warns pension funds not to be bedazzled by the 'beauty' of asset and liability management.

Beauty is in the eye of the beholder, and we can become blinded by our own expectations and assumptions, only to be shocked back to reality when things turn out uglier than expected. Asset and liability management's 'beauty' is similar in the sense that trustees and consultants alike need to be vigilant so they are not blinded by their own expectations. Equity risk premiums certainly seem reasonable at 3%, but we should look at what we've achieved in gross terms since the beginning of this century.

Everybody agrees with the simple assumption that any modelling tends to adhere to a 'rubbish in, rubbish out' path, but we never seem to recognise the fact our own variables may be just that - rubbish. Even modelling under various scenarios such as slow growth or high inflation does not deal with events that are not captured in the probability distributions used for forecasting. And forecasting it is.

Such forecasts may not be as unreliable as the Met Office's predictive skills - how often did they get the 'Indian summer' or the 'mild winter' forecasts right? Imagine having to forecast the average rainfall for 2015. The only thing guaranteed is that you will get it wrong and that, even for a good forecaster, you're as likely to estimate too high as too low. The pertinent question is - just how wrong you will be?

Order is the shape upon which beauty depends.  But what if order changes to chaos? Beauty turns ugly. Financial markets can be as unpredictable and ugly as natural disasters. For every earthquake in Japan or New Zealand, flood in New Orleans or Bangladesh, tornadoes, typhoons, etcetera, there is a Russian Crisis, LTCM, WorldCom, Enron, credit crisis or sovereign debt crisis. Which unknown unknowns, to quote Donald Rumsfeld, are still lurking in the shadows?

Against these natural disasters, we insure, budget, undertake detailed contingency planning and diversify to mitigate risk. And so should pension schemes. What is the effect on the scheme if the forecasts are wrong and the unknown unknowns bite the proverbial bottom? How does the funding level behave in another dotcom bubble or a Japan scenario?

Risk management is not just optimising risk and return within an asset liability management (ALM) framework. Once an ALM is finalised, it would be a big mistake to fall in love or make an emotional commitment to the 'planned' outcome. Managers should think back to September 2008 and consider how their scheme would have fared then. True risk management is repeated stress testing of all assumptions and preferably mimicking a number of historical events.

History keeps repeating itself, simply because we expect markets to learn from the past. And if markets don't learn from their own mistakes, trustees will have to do it for them. If the absence of flaws in beauty is itself a flaw, maybe a flawed market is not that ugly after all?

Jeroen Wilbrink is director of F&C Investments' Investment Solutions Group