DC: Steering a smarter path to retirement
The turbulence that investment portfolios have experienced this year is likely to have given many defined contribution (DC) pension schemes pause for thought. How can they best protect their savers from volatile markets, while also generating returns that maximise what are often inadequate contributions?
The traditional answer has been a glidepath in the final years before retirement that “de-risks” the portfolio by gradually shifting the mix toward bonds (typically gilts) and away from equities and other growth assets.
This is meant to provide a certain level of insurance against big falls in the market at a time when savers can least afford them.
However, the new pension freedoms and longer lives mean that many savers may wish to remain invested “through retirement” to allow them to take an income from their investments after work stops.
In these circumstances, the protection provided by a glidepath may come at the wrong time and prove expensive in terms of forgone investment returns. Is there a better way? To answer that question, we first needed to establish reasonable targets for what a pension should provide.
Given that it is likely to be supplemented from other sources, such as the state pension, we would argue a sensible expectation is that, on average, our main goal should be to replace at least 50% of work earnings. However, it is in the nature of investment income that it will fluctuate, so as a minimum we would hope to achieve replacement income of at least 35% of work earnings. This would form our “downside goal” (see chart).
Having set the targets, we looked at what retirement outcomes could be expected from different strategies.
We started by comparing a fixed allocation balanced between equities and bonds against a glidepath approach that gradually ratcheted down the equity allocation from the age of 55 to retirement at 65.
We then tested them over a 40-year saving period using a random mix of actual stockmarket returns from the past 90 years or so.
The results were striking. Neither strategy hit either of the targets, although the “balanced” approach came close at the average level. So while a glidepath may provide a smoother ride in the final years, the cost in terms of missed performance is clear. We would suggest that an investment manager should be able to make better choices.
On the, admittedly highly simplified, basis that a manager has three options open to them with a portfolio – to buy, hold or sell – even a manager with no skill should get the decision right a third of the time.
By the same token, a middling to good (though not necessarily exceptional) manager should be able to make the right decision from half to two-thirds of the time.
Using these assumptions, we ran the same tests to see how close we got to our income replacement targets.
Replacement ratio is level of final work earnings matched by pension income. Static balanced assumes a fixed 60% equities/ 40% bonds portfolio over the entire period. Other outcomes relate to a simplified “dynamic” strategy based on the percentage of time a buy, hold or sell decision involving 15% annual adjustments to the balanced strategy proves correct. Outcomes are based on 40-year lifetimes sampled 1,000 times from annual returns drawn from actual market history 1928–2014. Top diamond on each line is best 10% of results; square is median result and bottom diamond is bottom 10% of results. Equities are the S&P 500; bonds are 10-year US Treasury bonds. Source: www.damodaran.com and Schroders.
As one would expect, the outcome depended on the skill of the manager.
At one end of the scale, the unskilled manager produced outcomes almost identical to the balanced approach: just shy of the main target (49%), but well short of the minimum (27%).
However, at the upper end – where the manager gets their decisions right 60% of the time or better – both targets were met.
What this says to us is that choosing a strategy with a fixed asset allocation, whether balanced or using a glidepath, is effectively taking a bet on markets over a lengthy period that no-one can predict.
Having a manager of even modest ability to steer the portfolio along that long road must make more sense. And, of course, there is nothing to stop a manager charting a more intelligent course along a glidepath in those sensitive final years, when the portfolio might still benefit from an insurance policy to counteract potentially disastrous bear markets in the years leading up to retirement.