In this article:
The only guarantee in life is that nothing is guaranteed. If the members of defined benefit pension plans don’t know this already, they may find out soon. In the UK, one in six private DB schemes are severely underfunded and at risk of needing rescued.
The act of balancing today’s assets against tomorrow’s liabilities is tricky at the best of times. But the managers of DB plans, which promise a level of retirement income based on working pay, face uniquely difficult challenges.
Most of today’s DB pension plans were created decades ago, when people were only projected to live into their early 70s, and the world had strong economic growth and high interest rates. Now, DB sponsors must provide for longer-living pensioners in a low-yield world.
This is leaving many DB pension schemes underfunded, and companies, retirees, employees and taxpayers could be impacted by these shortfalls. The UK is a good example, where aggregate DB schemes have been in deficit since 2011. By 2017, only 12 per cent of DB schemes in the UK were open to new members, and no FTSE 100 company has a DB scheme that is open.
Source: Pension Protection Fund/Pensions Regulator, based on aggregate s179 funding position in PPF7800 index
The UK is not alone. A 2017 stress test among the EU’s Institutions for Occupational Retirement Provision (IORPs) found its DB and hybrid schemes had, on average, insufficient assets to meet pension liabilities. Employers worldwide are switching to defined contribution (DC) plans, in which individual members bear the investment risks, as they are more cost-effective and less risky than DB schemes. In the seven biggest pensions markets, DB assets fell from 67 per cent to 51 per cent of the total (DB plus DC) between 1997 and 2017, according to Willis Towers Watson.
DB schemes will need to adapt by exploring new investment models and strategies. Many still use high rates for discounting liabilities. This, together with low interest rates and demographic challenges, make it difficult to achieve adequate returns without significant risk-taking. For many, it's a race against time.
The consequences of being underfunded
Companies with pension shortfalls may need to cut wages or future benefits of current workers. In 2000 to 2015, wages in the UK fell from 87 per cent to 83 per cent of total compensation, while pension contributions increased. Half of these contributions went to servicing the deficits in closed DB schemes, rather than deferred compensation for the current employees.
Some countries have regulations that guarantee payments for retirees, which means taxpayers or future retirees could end up footing the bill. The US and Canada have compelled private sector workers to pay for public sector retirement schemes, especially in places with large shortfalls and broader fiscal stress such as California and Illinois. The European Union’s MEP pension scheme had a €326 million actuarial deficit as of early 2018, which is guaranteed by the Parliament and ultimately taxpayers.
In the UK, the Pension Protection Fund (PPF) can take over scheme assets if corporates go bankrupt. It would continue to pay benefits but cap payments for future retirees based on an earlier salary, leaving them with less than expected.
The PPF is funded by a levy on all eligible DB schemes, which distributes the risk but is subject to scrutiny and controversy. For example, more than 100 companies have used controversial pre-pack administrations to offload £3.8 billion of pension liabilities to the PPF, according to a 2017 report in the Financial Times. After offloading the liabilities, the assets are sold as a going concern, often to existing directors or owners. This allows the shareholders to remain relatively unscathed, while reducing benefits for members and costing the DB schemes which are still solvent.
As populations age and birth rates drop, a shrinking pool of new workers must pay for a growing group of retirees. Overall this means incoming contributions are smaller compared to the liabilities being paid out. Many pension schemes are no longer receiving inflows, so cannot absorb market shocks as they approach their payment obligations.
Population ageing also has implications for economic growth, which in turn affects long-term investment returns. Real economic growth is driven largely by labour force growth and productivity, both of which fall as populations age. Because of these challenges, we see the UK’s potential real GDP growth slowing from around 3 per cent in 2020 to 2 per cent by 2040.
Finally, calculating how much money is needed today to pay for future benefits requires pension managers to make assumptions for retirees’ lifespans and investment returns. Longer lifespans mean longer payout periods on DB liabilities and higher total payouts. In 1983, the UK government predicted that life expectancy would be 80 years in 2014. By 2014, that estimate was revised up to 86 years, and each one-year increase adds around 3 per cent to the value of schemes’ liabilities.
The discount rate as a hurdle rate for returns
Another tricky calculation involves the long-term interest rates used to discount future payout obligations to the present day. There is some flexibility as to the exact rate used, but it should be in line with the long-term growth rate of investments, preferably erring on the side of caution. This is assuming the scheme is fully funded, as underfunded schemes would require a higher return than the discount rate.
For example, if a scheme owes £105 in a year's time, and only has £100 today, it would need an investment return of 5 per cent to meet its liability. It should only use that discount rate if there’s high confidence in achieving such returns. Otherwise more contributions may be needed to plug future shortfalls.
On the other hand, using too low a discount rate relative to the likely returns could also be misleading. When yields fall, DB schemes see a small increase in their assets as the value of bonds they hold goes up, but a larger increase in their liabilities as the discount rate goes down. That’s because the liabilities typically have longer duration than the bond holdings, with cash outflows continuing many years longer than those produced by a bond portfolio. The UK government estimates that for every £1 increase in assets due to falling gilt yields, there is a corresponding £5 increase in liabilities.
Abnormally low interest rates are often cited as a sign that pension liabilities are overvalued, and that the shortfalls could easily shrink if gilt yields increase. The caveat is that in practice, many schemes do set discount rates based on expected returns on their investments rather than tied to UK gilt yields. This lower interest rate environment has made it impossible for many schemes to be fully invested in gilts and fully hedged for inflation (which many liabilities are linked to).
Low returns in a low rate world
When global central banks reduced interest rates to record lows in response to the 2008 financial crisis, they caused bond yields to plummet, reducing pension fund returns. In many developed countries such as the UK, real yields on sovereign bonds went negative, for maturities out to 50 years.
Source: OECD, 2016
Source: Fidelity International, Merrill Lynch 28 September 2018
This low-rate environment means many schemes need to hold more risky assets to earn returns in line with their discount rates. For example, in the UK the average Local Government Pension Scheme (LGPS) discount rate is around 4.5 per cent. Based on our 10-year projections for capital markets, the portfolio would need to hold at least 70 per cent in growth assets to earn this level of return. This is far from the current reality.
Instead, DB schemes in the UK have de-risked significantly since 2006, with the proportion of assets allocated to equities falling from 60 per cent to 29 per cent in 2017. Meanwhile, the proportion of assets invested in fixed interest bonds climbed from 28 per cent to 56 per cent. This can be explained by many pension schemes reaching the stage of negative cash flows, and reducing volatility to ensure meeting their pay-out obligations, as well as investor risk aversion caused by the shock of the 2008 financial crisis.
However, de-risking also pushed down discount rates, which increased the size of schemes’ liabilities on paper. This fuelled a vicious cycle of corporate schemes becoming more underfunded, requiring more deficit repair contributions and, for many, closure to new members.
Source: Fidelity International, July 2018
While all DB schemes are unique with different needs, in the UK they can generally be classified into three situations:
- Plans that will become cash flow negative within five years
- Plans that are already cash flow negative, and need income that’s below 4 per cent of assets, this income level can typically be produced by a diversified asset allocation
- Plans with cash flow requirements higher than 4 per cent of assets and that must be met by selling off some assets in addition to investment income
Chart 5: Typical UK pension scheme liability curve
Source: Fidelity International
The many ways forward
Each scheme must strike its own balance between risk, return and cash flows. Underfunded schemes or those with more aggressive discount rates can’t de-risk fully and need to generate more cashflow while still maintaining growth. To achieve this, they could hold more high-yield securities, emerging market debt, high-dividend equities or absolute return strategies, for instance. They can also add more niche assets to diversify and protect against large drawdowns.
Fully funded schemes with conservative discount rates have more room to de-risk. They could do so by investing in less risky assets, such as investment grade debt, or by diversifying their asset allocation. However, care should be taken to ensure any changes do not jeopardise the scheme’s ability to meet its expected return in this low rate environment.
Investors could also use hedging instruments to mitigate downside risk, but it’s not a free lunch. More protection usually comes with a greater drag on returns when markets are positive (see How to Protect Against Black Swans). Pension schemes which buy options on a long-term recurring basis may need to reduce their discount rates to correspond with lower expected returns. Therefore, they could benefit from having a plan for adding risk back into the portfolio or removing the hedges.
There are also ways to help DB managers reduce their liabilities and costs, none of which are easy or popular with scheme members or regulators. These include closing DB plans in favor of DC plans, offering lump sum settlements to current beneficiaries (i.e.- the fund sponsors buy out their liabilities) or transferring the liabilities to other parties such as insurers.
Schemes can purchase longevity swaps and transfer to third parties the added risks of scheme members living longer than expected. In such contracts, the scheme would make fixed payments based on mortality assumptions for members, in exchange for floating amounts from the swap seller based on the actual mortality rates.
Besides these, schemes are increasingly delegating their investment management and administrative functions, and smaller schemes are consolidating to improve economies of scale. When outsourcing to an external investment manager, sponsors may be hoping to improve investment results, lower fees or reduce demands on internal resources.
Most pension trustees in a poll said they would use outsourcing, and about half said external investment managers would become more common. More than 40 per cent believed consolidation would “significantly” improve the efficiency of DB pensions in the UK.
The deficit problem won’t go away anytime soon. The ILC-UK estimates that in 2060, around 3 million people will still be receiving money from private DB pensions, with over 1 million remaining by 2070.
While there is much a DB manager can do to redress the balance, from boosting their risk appetite to cutting costs, the structural demographic factors that deepened the funding issues in the first place will simply not go away.
Perhaps the only guarantee is that there will be no quick fix.