Pension Risk Management: Best Practice in De-risking Corporate Pension Funds
What is risk? Generally, it can be defined as the potential for not meeting a prescribed objective. The investment objective of a pension plan is to provide an adequate level of income over the life of the pension member following retirement. The associated pension risk is, therefore, the risk that there is a shortfall in these pension entitlements. This question of pension risk is likely to become more significant for pension members in Asia over the next decade or so, as a shift to defined contribution (DC) pension schemes manifests across the region.
There are two different types of pension structures: defined benefits (DB) and DC. DB plans are pension programmes in which the pension entitlements (the benefits) are specified in advance by the pension provider on entering the pension plan. The pension provider, for example a corporation or government, is essentially providing an undertaking (either explicitly or implicitly) to secure these benefits to its members. Pension risks under a DB arrangement can arise from a number of sources. For example, there are the risks associated with the institution failing to meet its obligations – i.e. a corporation facing financial difficulties or governments facing fiscal deficits.
This institutional risk can be reduced by ‘fully’ funding the pension obligations through a segregated pool of money (a pension fund) that is set aside to provide for the pension entitlements, but this will create an investment risk associated with the pension fund. Funding a pension plan through a segregated pension fund imposes an obligation on the corporation to make good any potential shortfalls between assets and liabilities. The corporation is exposed to the pension fund’s market fluctuations and is responsible for the pension entitlements.
In contrast, DC plans are programmes in which the pension member is exposed to the market fluctuations of the pension fund and is responsible for the pension entitlements. In a DC programme, regular pre-defined contributions accumulate in the DC funds and, on retirement, the accumulated value of the fund (plus investment returns) is used as the basis of the retirement income programme through purchased annuities, return of a lump sum amount or other investment arrangements.
Given that the investment risk is borne by the pension member rather than the corporation for DC plans, it is no surprise that DC is the largest growing segment of the market and remains one of the most important ways of reducing pension risk from the corporate perspective. As noted last year in a State Street Vision Paper, ‘Pensions Strengthening the DC Model for the Future’, the growth of DC assets is now outpacing that of DB globally. Among the seven largest pension markets in the world – the US, the UK, Australia, Canada, Switzerland, Japan and the Netherlands – DC assets have grown at approximately 6.4% per year over the past 10 years, while DB assets have grown at only 1.6% per annum. As a result, DC assets now represent 42% of total pension assets in those markets, up from 40% in 2004 and 32% in 1999. However, the shift toward DC is not happening at the same speed, nor taking the exact same shape, across all markets worldwide.
This trend towards DC in the developed pension markets is, in part, driven by regulatory and structural changes. During the period until early 2000, corporate DB plans benefited from strong market returns resulting in large surpluses (i.e. assets exceeded estimated pension liabilities) in corporate DB pension funds. These large surpluses created pension contribution ‘holidays’ that allowed corporations to waive payment contributions into DB plans for the year, and thereby improve on their net income; at the same time, if there happened to be a pension deficit, then accounting rules allowed these to be disclosed as a contingent liability. All this changed in the middle of the last decade, when some corporate DB plans began to show deficits and new accounting standards required these deficits to hit company’s bottom line.
In Asia (generally a less mature pension market), these changes have not been as important; in addition, a generally conservative investment focus has shaped pension policies, and as a result DC models have been slower to catch on. Australia, with its superannuation funds, is furthest along the curve to becoming a completely DC-dominated market. Still, the trend toward DC and the accompanying changes to pension risk profiles has significant consequences for pension holders in Asia. It also has consequences for governments in the region, many of which are under pressure to ensure sufficient funding for a decent quality of life in retirement for ageing populations.
The role of DC as a mechanism for transferring investment risk from corporations to individual pension members seems obvious: corporations should favour this idea whereas individuals should prefer DB. This may be a valid proposition, but there are some grey areas. For example, DC programmes may also be attractive to individuals as they provide more potential flexibility, portability and individual empowerment to pension members. A key question for pension holders is this: do you really want to rely on a corporate DB contract that spans, possibly, the next 70 years?
Corporations also need to exercise some caution before fully embracing DC. While DC has allowed risk to be taken off the balance sheet, other forms of corporate risk are accentuated. As DC plans are usually set-up by the corporations, there is a fiduciary and reputational risk element that will persist. This type of risk can be reduced through the adoption of best practice in the provisioning of DC services. Important considerations include:
Therefore, while DC goes a long way towards to reducing balance sheet risk, corporations need to be aware of the new form of risk that has been taken up. A good illustration of this is lifecycle, or target date, retirement funds. These are programmes in which the equity allocation gradually declines over time so that, on retirement, the allocation is mainly in bonds. These are viewed in the industry as a good overall solution for retirement plans and have been widely used as default options in DC plans, but recent attention has focused on their performance during the recent period of market stress during the global financial crisis.
The moral of this story for corporations is this: DC programmes can provide a good solution for reducing corporate balance sheet risk, but companies should be aware of the associated duties and fiduciary risks. Unfortunately, fiduciary risk, just like investment risk, also tends to be skewed with fat tails – no one is going thank you for doing a good job, but you will be blamed if anything goes wrong.
There are other compelling reasons for the popularity of DC worldwide, which could also assist the spread of such programmes in Asia. The retirement of the ‘baby boomer’ generation will begin to put a lot of strain on the ratio of those working to those retired. Also, the current low levels of bond yields is putting a strain on achieving nominal return targets; at the same time, the perception is that inflation risk is increasing, so real returns may also be impacted.
Clearly DC is the future for the pension markets, but what about the existing DB plans? DB plans continue to be sizeable and point to important risk issues that need to be addressed. Some of these issues are also relevant to the DC space. The main focus when considering DB risk is the risk of a pension shortfall. This risk has two components: the risk that pension asset values are lower than projected, and the risk that pension liabilities are higher than projected.
Looking at the asset dimension, strategic asset allocation has been a useful tool in structuring policy asset allocations that are consistent with the DB plan’s overall return and risk objectives. The key element in this trade off is that a high equity allocation might result in higher projected asset returns, but with higher pension fund volatilities. Most recently, developments have begun to focus on dynamic strategic asset allocation, or dynamic strategic hedging, as a means of implementing a policy allocation framework more systematically. These types of ideas have been imported into the DC world as asset allocation programmes for target date retirement programs.
An obvious source of risk in the asset portfolio is market risk, particularly given the current perception that market volatilities have been increasing over the past few years. One way to hedge this risk is through a purchased option strategy, but this has, generally, been viewed to be an overly expensive proposition to hedge against the broad equity risk exposure. To overcome the cost issue, portfolio insurance techniques were previously used to avoid the payment of explicit option premium (although an implicit volatility cost is still present). Nowadays, these types of ideas are not so often used for hedging broad equity risk exposure; instead, users of hedging techniques have tended to favour purchased options to hedge against very large equity market declines (so called ‘tail event’ risk).
Pension liability risks provide additional challenges and relate essentially to the potential that the pension liability is bigger than expected. For a DB plan, the principal liability risks include longevity risk (pension members in retirement draw out more money than anticipated); inflation risk (inflation linked pensions are bigger than anticipated due to a hike in inflation); and re-investment risk (future bond yields are too low to meet income projections). Some of these risk factors, such as inflation and re-investment risk, can be handled through standard asset liability matching techniques. For example, inflation risk can be managed by holding inflation linked bonds in the asset portfolio, or other assets such as property rental yield that may provide an inflation hedge.
Other liability risks are more intriguing: how do you hedge against the risk of longevity? This is becoming an issue for DB plans, and indeed for DC plans that are attempting to use the DC fund to purchase annuities on retirement. Given improvements in affluence (and, thereby, lifestyle and medical treatments), this is fast developing into an important issue for DB plans that has resulted in significant market innovation. It is understandable that DB plans might want acquire a hedge against longevity risk; the natural short provider of longevity hedging is from insurance companies as a hedge against their life policy portfolio – the hedge arises because a shorter than expected life expectancy will create more claims on life policies that are offset by gains from a short longevity position (and vice versa). Instruments now include securities and swaps based on life expectancy. The main challenge for this market is that it is not very large compared to the size of the underlying DB liabilities.
Looking further ahead to a pension world largely dominated by DC, the broad challenge for the financial services industry is to work closely with other stakeholders – including policymakers – to deliver better risk outcomes for retiring workers and also for corporations and governments. It will be particularly interesting to watch this process play out across Asia, where the pressure to provide a decent living for aging populations is perhaps greater than anywhere else.
This document may contain certain statements deemed to be forward-looking statements. All statements, other than historical facts, contained within this document that address activities, events or developments that Hon Cheung expects, believes or anticipates will or may occur in the future are forward-looking statements. These statements are based on certain assumptions and analyses made by Cheung in light of his experience and perception of historical trends, current conditions, expected future developments and other factors he believes are appropriate in the circumstances, many of which are detailed herein. Such statements are subject to a number of assumptions, risks, uncertainties, many of which are beyond his control. Please note that any such statements are not guarantees of any future performance and that actual results or developments may differ materially from those projected in the forward-looking statements.