Turning off auto pilot: Why there is more to retirement planning than the “glide-path”

The use of target-date funds within the defined contribution (DC) market continues to click along. The funds now account for roughly 30% of the $5.7 trillion in 401k assets1. Their near uniform selection as the QDIA2 means target date funds, for many younger participants, represent their sole retirement savings investment vehicle.

Target date funds have many redeeming properties, and their initial adoption by plan sponsors represented a good fiduciary decision, if for no other reason than to protect participants from themselves. Plan participants managing their own asset allocations rarely rebalance and seldom make investment decisions based upon prudent risk management.

Participants benefit from target date funds’ simplicity. Knowing when you turn 65 is a lot easier than deciding whether you have a “conservative” or “moderate” risk tolerance - something that is required by target risk funds (which incidentally have lost the QDIA foot race to target date). Target dates offer greater diversification than the prior generation of “60/40” balance funds, and fundamentally, it is reasonable and prudent to use a strategy which moves into lower volatility assets over time.

Having said all that, the investor and sponsor concentration in target dates, and the benign environment in which they have thrived, mask some shortcomings of the design. These portfolios are not dynamic in their asset allocation. We believe that incorporating a dynamic approach to asset allocation can enhance the target date proposition and generate better risk adjusted outcomes for plan participants. In making what is, in our view, the most important decision for a fiduciary, assessing all the available options is crucial.

The good, the bad, and the better

The typical target date “glide-path” approach adjusts risk levels based on the investor’s (or participant’s) proximity to their planned retirement date. Risk is generally managed by lowering equity allocations and increasing bond allocations as the participant gets closer to drawing retirement income.

However, this approach often ignores prevailing market conditions, and as a result cannot, in our view, optimize retirement outcomes. An essential consideration at any time, this is especially pertinent now. Bond yields are again approaching historic lows and the risk of loss from rising rates stands to do much damage to fixed income-heavy near retirement funds.

A dynamic and adaptive approach to asset allocation can draw from a broader set of asset classes than the average target date portfolio. It can also use valuation, cyclical and sentiment indicators to adjust its risk profile. We believe this may be far more effective in improving retirement outcomes, and is an approach that is implementable within a target date fund or on its own. One must not overestimate the complexity of implementing such a strategy.

Looking beyond the glide-path

We have found that asset allocation strategies based on age alone have limited impact upon retirement outcomes, and indeed, can in some circumstances be detrimental.

For any asset allocation strategy to demonstrate its efficacy, it must be able to show an improvement in the retirement outcome for the plan participant. The most commonly used way to measure this effectiveness is via the replacement ratio (RR). RR is a measure of the percentage of pre-retirement income replaced by income drawn from the retirement portfolio.

If we compare a glide-path approach to a static, balanced allocation using historic data, the empirical evidence gathered should show an improved retirement outcome for the glide-path option. The static allocation we looked at in our example averaged the same equity allocation as the glide-path over a 40-year career (using a 60/40 equity-bond split).

However, on average we found that the outcomes of both approaches are comparable. What’s more, if we model the average RR for an investor that moves to a glide-path approach at age 55 (versus one that remains in the static portfolio throughout) the average set of outcomes is actually moderately worse.

Part of the reason that the glide-path outcomes are worse on average than a balanced static allocation is due to “dollar weighted” time in equity. In other words, the participant’s growth driver – the equity allocation - is higher earlier in the career, at a time when contributions are lower in absolute terms.

Informed versus uninformed decisions

Arguably the most impactful shortcoming of the glide-path though, is that an investor is effectively making an active investment decision that is dictated several years ahead of implementation. It is therefore uninformed by prevailing market conditions and market values. Of course, accurate unilateral calls on market timing are extremely challenging. Even so, making measured annual changes to allocations to stocks and bonds based on observed anomalies – in market pricing or macroeconomic conditions – allow portfolios to be rebalanced on the basis of contemporary data and manager skill, not just the time to retirement.

Case in point: our expectations are currently for lower investment returns (for both bonds and stocks) over the medium-term. Approaches tethered to bonds are unlikely to provide returns that will sufficiently replace pre-retirement income. In such a market scenario, there is a justification for maintaining a higher allocation to growth assets for longer, and even through a participant’s retirement date, on a risk adjusted basis. The dynamic allocation must also be used to control risk in the growth-oriented portfolio. It can do so through tactical investments in diversifiers like gold, or reserve currencies, or by using options for protection.

There are of course countries in which a pension fund must still be used to purchase an annuity, and in these locations, a glide path strategy can still be appropriate. However, for those retirees more likely to generate income from their savings rather than buying an annuity - such as in the US – the risk of running out of money in the retirement phase due to a shortfall in growth is more significant.

Dynamic asset allocation - 101

In order to deliver the real returns that a DC participant requires, we believe an active approach that reacts to market conditions is required. Fixed strategic asset allocation is convenient, but inefficient compared with an outcome-oriented, multi-asset approach. By allowing allocation flexibility and using valuation metrics to assess the current state of a portfolio, we believe it is possible to improve the outcomes and significantly reduce drawdowns. We have found that certain long-run measures for a number of asset classes can meaningfully indicate their valuation, with respect to a valuation state being “cheap”, “neutral”, or “expensive”.

For international equities, we have found the Shiller price/earnings ratio to be effective in determining the valuation state. In fixed income allocations, nominal yield is representative of value, but not as informative as term-spread as an indicator of recessionary risk (and therefore wider portfolio risk). However, the predictive quality of each measure (of which these examples are only two) varies with each market state. Part of the investment skill is determining which of these measures is most instructive in which market.

Having determined the valuation state of a market, and factoring in the historic performance of the chosen market following previous instances of this valuation state, we can set our chosen allocation. This allows us the conviction to pursue one of three courses of action:

  • Stay significantly underinvested during expensive market states. Here, the majority of additional value is created by maintaining cash positions when valuations are expensive. The flexibility to exclude certain asset classes can prove valuable.
  • Less diversification is rewarded during cheap market valuations. Investors are rewarded for holding a more concentrated portfolio of growth assets under a cheap valuation state as the probability of loss is lower and expected returns are higher.
  • A traditional fixed strategic asset allocation portfolio can offer adequate performance in neutral market valuation states. A relatively diversified portfolio can offer adequate risk-adjusted performance as long as valuations remains neutral.

Overall, we believe that the typical asset allocation approach followed by US target-date funds does not maximize the opportunity to deliver retirement outcomes, as they ignore market valuations and offer constrained allocation models. By blending valuation indicators, and also allocating to a more diverse set of risk premia, investors can create a portfolio better able to thrive in prevailing market conditions.

1. Sources: & Morningstar

2. Qualified default investment alternative – The US department of Labor states that 401(k) assets “must be invested in a “qualified default investment alternative” (QDIA) as defined in the regulation” signed into law in 2006.

The views and opinions contained herein are those of Schroders’ investment teams and/or Economics Group, and do not necessarily represent Schroder Investment Management North America Inc.’s house views. These views are subject to change. This information is intended to be for information purposes only and it is not intended as promotional material in any respect.