Edhec-Risk
Institutional Investment - June 28, 2010

More pension funds are asking for downside risk protection - an interview with Thibaud de Vitry

In this month's interview, Thibaud de Vitry, Global Head of Investment Solutions with AXA Investment Managers, discusses the results and potential implications of the recent EDHEC-Risk Institute survey on the asset-liability management practices of European pension funds, which was carried out as part of the EDHEC-Risk Institute "Regulation and Institutional Investment" research chair sponsored by AXA IM, and his hopes and expectations for the future research work from the chair.


Thibaud de Vitry

EDHEC-Risk Institute recently carried out a survey on the asset-liability management practices of European pension funds as part of the "Regulation and Institutional Investment" research chair sponsored by AXA Investment Managers. The overall theme of the report is that pension funds should favour rule-based strategies over discretionary ones. Is this a position with which you agree?

Thibaud de Vitry: Yes, we agree with the use of rule-based strategies for the purpose of pension fund risk budgeting. Pension funds operate within a very special accounting and prudential framework that requires adapting risk budgets in order to manage both short-term solvency and volatility objectives as well as long-term return objectives such as indexation.

However, while we are proponents of a rule-based approach for risk budgeting, we believe part of this risk budget can be allocated to judgemental medium-term asset allocation, thereby creating some out-performance potential over liabilities within a strict, risk-controlled framework.

One telling example concerns Dutch pension funds, which are subject to minimum levels of solvency under FTK regulations. We have defined a minimum liability hedge ratio based on a rule which takes into account the specific solvency and capital requirements of each fund. Our portfolio managers can then decide to be at the minimum or above, depending on the interest rate outlook – for example we tend to be at the floor in the current context, but at a different point in the cycle we could decide to hedge more than the minimum required.

One of the findings of the survey is that most respondents have a restrictive view of the risks they face: prudential risk (the risk of underfunding) is managed by only 40% of respondents, accounting risk (the volatility from the pension fund in the accounts of the sponsor) by 31%, and sponsor risk (the risk of a bankrupt sponsor leaving a pension fund with deficits) by less than 50%. Are these results surprising to you?

Thibaud de Vitry: Yes and no. These numbers are averages, and therefore hide a less uniform reality across pension funds.

The three risks you mention are common to many funds, but with varying degrees of importance. For instance, sponsor risk is irrelevant for sector and public pension funds, and prudential requirements are much stronger for Dutch pension schemes than they are for British ones. Conversely, accounting is key for British schemes, many of which have listed sponsors. Therefore, the objectives and ensuing perception of risk are very specific to each fund, and to each country.

These differences are the reason why, when we initiate a relationship with a new pension fund, we spend a lot of time assessing the specific requirements of the fund, as well as establishing a risk map to identify those risks that need to be pushed higher up on the agenda.

The survey also indicates that pension funds are relatively reluctant to invest in potentially illiquid assets such as hedge funds, private equity and infrastructure, even though, as long-term investors with no need to worry about short-term liquidity, they are in a good position to invest in these assets and take on liquidity risk. What do you think accounts for this reluctance?

Thibaud de Vitry: We witnessed during the crisis a growing focus on transparency and quantitative risk management. This has led to an increasing use of short-term measurement approaches such as marking to market all assets - and sometimes liabilities -, or instantaneous stress tests as well as quantitative measures such as Value at Risk (VaR), Conditional Value at Risk, Economic Capital, volatility, etc.

Finally, not only is applying this framework to less liquid asset classes a challenge - and not always an adequate approach – but both risk budgeting and the rebalancing of risk budgets are difficult exercises to implement. All of these factors may explain why some institutions appear to be reluctant to invest in potentially illiquid assets, although we believe that the illiquidity premium can be leveraged within the asset allocation of a pension fund.

This being said, some of the larger institutions which have developed an expertise in these particular asset classes have been able to implement a specific risk framework adapted to their specificities. Telling examples are APG and Alpinvest, as well as endowments such as Yale or Harvard, all of which are known to be successful long-term investors in these asset classes.

A good illustration of how specific, yet feasible, risk management is, insofar as these asset classes are concerned, is the fact that AXA Private Equity uses a proprietary database of 550 funds and 7,500 portfolio investments to better assess the risk/return expectations from an investment, and values investments in its fund of funds by looking through each individual invested company. In parallel, our asset allocation and financial engineering teams have developed strategic asset allocation calibrations to model the long-term behaviour of these asset classes, within a pension fund framework but also under Solvency II – all of which clearly shows that it is possible to create a sound risk framework for such asset classes.

A further conclusion of the survey is that risk-controlled strategies, which insure against a fall in funding ratios below the required minimum, make it possible to forgo some of the upside potential of a performance-seeking portfolio in exchange for downside protection. 50% of pension funds are fully aware of these strategies, but only 30% use them. Do you think that institutional investors will be more receptive to this approach in the future?

Thibaud de Vitry: Overall, we have witnessed a real change in the last few years, with more pension funds asking for downside risk protection. The classic example during the crisis has been the use of option portfolios (“collars”) to mitigate all or part of the downside equity risk. These structures have been popular with pension funds, but we have also seen some interest in dynamic strategies such as portfolio insurance, which work better when the underlying does not have standard derivatives (e.g. hedge funds). We expect these strategies to become increasingly popular as a consequence of the current trend towards “outcome orientation”. Indeed, the investment portfolios should aim to reflect the ultimate goals of the pension funds, which are to provide long-term indexation with sound solvency risk management.

Along the same lines, we have also witnessed an increase in the demand for risk-controlled, total return strategies that can also contribute to a liability-driven objective.

The next research work from the "Regulation and Institutional Investment" research chair will be on the impact of international accounting standards on the financial management of corporate pension funds. What are your hopes for this new research?

Thibaud de Vitry: We hope that this research will help the finance departments of the corporate sponsors of pension schemes to better understand the impacts their pension fund can have on their accounts, as well as think of different solutions to help them mitigate the unwanted accounting volatility. Similarly, we hope it will be of interest to pension fund trustees, who are exposed to their sponsors through a complex relationship related to accounting – indeed the sponsor consolidates the surplus or deficit in his account, but generally does not have control over the assets. However the sponsor does have some influence through the premium policy, as well as the negotiation of potential guarantees.

Optimising the management of a pension fund’s assets requires understanding the context within which each stakeholder operates, and after reviewing the overall role of the prudential and accounting framework, we would like to focus on the same topics from the sponsor’s point of view.

At AXA IM we have worked extensively with our internal AXA client to redefine investment solutions for balance sheet assets in the IFRS environment, and we now hope to have the same approach for pension funds!



About Thibaud de Vitry

Thibaud de Vitry is Global Head of Investment Solutions and Member of the Executive Committee, AXA Investment Managers. He was appointed Global Head of Investment Solutions in September 2007 and is responsible for Solutions’ Engineering, Structuring and Development, relationships with the AXA Insurance Companies, Asset Allocation Management and Insurance Investment.

Thibaud joined the AXA Investment Managers Group in March 1998 as Managing Director of Quantitative and Structured Investment, in charge of creating, developing and managing structured and alternative investments.

In 2000, he was appointed Global Head of Operations during which time he created the transversal function, before taking over the positions of Chief Operating Officer of Securities Investment Management and Global Head of Insurance Investment in 2002.

Prior to joining the AXA Investment Managers Group, Thibaud was Head of Structured Asset Management at Crédit Agricole Asset Management and earlier Actuary, Head of the Technical Department at Legal and General in Paris.

He graduated with a degree in Finance and Insurance from “Ecole Nationale de la Statistique et de l’Administration Economique” (French school of Statistics and Economic Administration) in 1989. He has been a member of the “Institut des Actuaires” (Actuarial Institute) since 1989.




URL for this document:
http://www.edhec-risk.com/Interview/RISKArticle.2010-06-28.1241

Hyperlinks in this document: