Many behavioural studies have shown there are several traits and biases that can impede us from making reasonable decisions about everything from what to eat to how to invest. Understanding these biases and considering whether they may be negatively impacting decisions can be beneficial when implementing long-term investment plans. These studies show, in general, people have asymmetric risk profiles and fear losses more than the expectation of gains by at least a 2:1 margin[1]. Interestingly, and perhaps not surprisingly, this ratio increases substantially as people approach retirement.

American psychologist and economist, Daniel Kahneman, who won a Nobel Prize for his work challenging the prevailing assumption of human rationality in modern economic theory has stated, ‘If you have an individual whose objective is to maximise wealth at a certain future point in time, then loss aversion is very bad because loss aversion will cause that individual to miss out on many opportunities.’

This loss avoidance trait stands in contrast to a basic investment principal, that investors need to accept higher risk (and higher potential for near-term losses) in order to achieve higher returns over the long term, particularly during market sell-offs. When faced with losses, rational decision-making can become impaired by the emotional desire to avoid more losses.

There are a wide range of cognitive biases that can impact retirement plans, some are listed below:

Confirmation bias

Confirmation bias is the natural human tendency to seek information that confirms an existing point of view or hypothesis. This can lead to overconfidence if investors keep seeing data that appears to confirm the decisions they have made. This overconfidence can result in a false sense that nothing is likely to go wrong, increasing the risk of being blindsided when something does go wrong.

Information bias

Information bias is the tendency to evaluate information even when it is useless in understanding a problem or issue. Investors are exposed to an array of information daily, and it is difficult to filter through this and focus on the relevant information. In general, investors would make superior investment decisions if they ignored daily share price movements and focused on prices compared to the medium-term prospects for the investments. By ignoring daily share price commentary, investors would overcome a dangerous source of information bias in the investment decision making process.

Loss aversion bias

Loss aversion is the tendency for people to strongly prefer avoiding losses than obtaining gains. The loss aversion effect can lead to poor and irrational investment decisions, where investors refuse to sell loss-making investments in the hope of making their money back. Investors fixated on loss aversion can miss investment opportunities by failing to properly consider the opportunity cost of their investments.

Anchoring bias

Anchoring bias is the tendency to rely too heavily on, or anchor to, a past reference or one piece of information when making an investment decision. For example, if you were asked to forecast a stock’s price in three months’ time, many would start by looking at the price today and then make certain assumptions to arrive at a future price. That’s a form of anchoring bias – starting with a price today and building a sense of value based on that anchor.

How do we try and overcome the biases when building retirement portfolios?

The objective based nature of Lonsec’s Retirement portfolios means there is a greater focus on absolute rather than relative performance. Additionally, the portfolios have been constructed to manage risks, including:

  • Market and sequencing risk
  • Inflation risk
  • Longevity risk

Some investment strategies that can assist in controlling for these risks include:

Variable beta strategies can vary equity market exposure by allocating to cash in periods where equity market opportunities are perceived to be limited due to expensive valuations, or where market downside risk is considered high.

Long / Short – Active Extension (also known as 130/30 funds) utilise a broad range of strategies including short selling and adjusting the net equity position for performance enhancement, risk management and hedging purposes.

Multi-asset real return funds invest in a wide range of asset classes, with the managers having considerable flexibility in the type and percentage of asset classes allocated to. Typically, these funds will seek to limit downside risk, while also targeting a real return i.e. a CPI + objective.

Real assets such as property and infrastructure, commodities and inflation linked bonds can assist in managing against inflation risk.

When constructing the Retirement portfolios, Lonsec takes a building block approach by assigning a role for each fund – yield generation, capital growth and risk control.

The yield component of the portfolios generate yield, or a certain level of income from investments that have differing risk return characteristics. The capital growth component is designed to generate long term capital growth, with limited focus on income, and is more suited to early retirees. The risk control component is critical for retirement portfolios and is designed to reduce some of the market risks in the yield and capital growth components. It is important to note that the risk control part of the portfolios will not eliminate these risks but aims to mitigate them. Asset allocation and diversification are also important ingredients in managing the overall volatility of the portfolios.

The Retirement portfolios can assist in managing the risks that impact retirees, however it is important to note that none of these strategies provide a guaranteed outcome. The range of products that offer certainty of income or capital protection such as annuities has increased in recent years, in recognition of Australia’s aging demographics and demand for greater certainty in retirement. Separate guidance on the use of annuities is available from Lonsec.

 

[1] Gachter, Johnson, Herrmann (2010). Individual – level loss aversion in riskless and risky choices. Columbia Business School

The Australian equity reporting season was weaker than expected with a third of companies having their FY20 earnings downgraded. At the sector level, Healthcare and Resources had impressive results whilst the Financials sector reported poor earnings. Dan Moradi runs through the key themes that emerged over the reporting season.

August was a weak month for equity markets with continued concerns over trade tensions between the US and China. While equities sold off bonds, ‘bond proxy’ assets such as property generated positive returns. For the year to the end of August Australian bonds, as represented by the Bloomberg AusBond Composite 0+ Yr Index, returned an impressive 11.2%, outperforming Australian equities. Despite bond yields being at low levels and valuations generally indicating that bonds are overpriced, bonds have continued to provide diversification to equities in downmarket periods.

We continue to believe that bonds play an important diversification role within managed portfolios, despite valuations remaining high. From a portfolio perspective, the primary role of fixed income assets is to provide diversification, with the secondary role being to generate income. Lonsec’s portfolio construction process employs a fixed income allocation within the portfolios to provide a diversified exposure to duration, credit and absolute return bond strategies. We caution against ‘chasing yield’ in the defensive part of the portfolio as most high yielding bond assets tend to be correlated to equities, which work well when equity markets are performing well, but on the reverse will move directionally with equities in an equity market pull-back. In our most recent Asset Allocation Investment Committee, we did not change our allocation to bonds but we increased our exposure to ‘bond proxy’ assets such as global REITs providing some more defensiveness to the portfolios.

One of the lasting achievements of sports like golf and horse racing is the way the handicapping system can make it possible for athletes to engage in a close and enjoyable competition despite obvious differences in ability, experience or personal characteristics.  In Thoroughbred horse racing, the older steeds would otherwise win most races, hence the ‘weight-for-age’ system, whereby each horse carries a small additional saddle weight based on statistical analysis of the historical differences in ability by age.  The system has been working well for more than a century – so why can’t we do the same for measuring super investment returns?

In January this year, the Productivity Commission handed down its final report to the Federal government on its investigation into the superannuation system.  One of the terms of reference for the inquiry was the following:

‘The Productivity Commission should develop criteria to assess whether and the extent to which the superannuation system is efficient and competitive and delivers the best outcomes for members and retirees, including optimising risk-adjusted after fee returns.’

It is of course widely accepted in the superannuation industry that there is a trade-off between risk and return.  Moreover, there is a general consensus that performance comparisons should be done on the basis of products with a similar risk profile.

The most common risk metric tracks the amount of change in return over time (technically, it is the ‘standard deviation’ of the net investment return measured over a specified period, also known as ‘volatility’).  It should be noted that this is a measure of ‘market’ risk only – arguably a young investor saving for retirement should be more concerned about the risk of losing her capital (‘capital risk’) due to poor quality asset selection, but there are currently few metrics for this.  A ‘Standard Risk measure’ must be displayed on MySuper Dashboards  in accordance with the Corporations Act, and the SRM is essentially a Market Risk metric, expressed as the expected number of negative returns over a 20-year period, based on modelling of the fund’s investment strategy.

A long-standing proxy for Market risk has been the ‘Growth Ratio’ or the percentage of funds allocated to ‘Growth’ type assets, such as shares, property and, more recently, infrastructure.  There has also been much discussion about the methods used by funds to report the percentage of growth assets, particularly in relation to unlisted asset classes like property, infrastructure and hedge funds.

How good is the handicapping system used within our industry?  Despite some limitations, there seems to be a reasonable level of consistency across the different metrics, and returns are consistent with the expected risk / return trade-off.  The following graph is based on median rolling returns and volatility (risk) over the three-year period to 30 June 2019:

With the exception of ‘High growth’, the relationship between median risk and return is in line with expectations.  Even so, there is still a potentially wide dispersion of risk / return combinations within each classification. Is there a way to apply a handicapping system to enable a fair comparison across and within the risk profile categories?

The Productivity Commission was tasked with optimising ‘risk-adjusted’ returns.  The ‘Sharpe Ratio’ is one such metric – it attempts to adjust for the trade-off between risk and return (technically, it is a measure of excess portfolio return over the risk-free rate relative to its standard deviation).  It does this by giving a lower weight to excess return earned from taking additional risk.

We ran a test race with just five horses, based on the SuperRatings Option Type classifications, and here are the results:

Horse

Place

Risk-adjusted Return*

Actual return*

Market risk*

Balanced (60-76)

1

1.60

8.80

4.14

High Growth (91-100)

2

1.30

9.25

5.59

Conservative Balanced (41-59)

3

1.19

6.58

2.90

Growth (77-90)

4

1.13

9.68

5.30

Capital Stable (20-40)

5

1.05

4.81

2.16

Source: SuperRatings FCRS 30 June 2019
Risk-adjusted Return: Sharpe ratio
Actual Return: 3-year rolling net return % per annum
Market Risk: 3-year rolling Standard Deviation % per annum

On this basis, ‘Balanced’ was the horse to be on.  ‘High Growth’ can run faster, but the risk trade-off is too high (on this metric) to justify the extra speed.

What is ‘Conservative Balanced’ doing in third place?  It’s the second slowest horse in the race, but it gives its jockey a smooth enough ride to offset the performance difference.  This seemingly anomalous result reflects the current investment environment, with very low risk-free rates of return.  This option was able to achieve a higher excess over the risk-free rate per unit risk than any of the other options bar Balanced and High Growth.

Is Risk-adjusted Return (RAR) then a sensible metric for comparing different investment strategies?  One major weakness is that RAR takes no account of an individual member’s risk characteristics and assumes that the risk / return trade-off is the same for all members. Clearly this is not the case – younger members are less likely to be impacted by short-term fluctuations in market asset values, whereas members approaching retirement can be affected by sharp declines in market value immediately prior to retirement (‘sequencing risk’).  In the context of the table above, a member approaching retirement might value the stability offered by ‘Conservative Balanced’ enough to accept the lower return, whereas a younger member would (should!) be unhappy about the underperformance, given that market risk is not really an issue.

This suggests that a ‘weight-for-age’ handicapping system would make more sense for MySuper performance comparisons.  In fact, many fund trustees have already put this into practice via a ‘Lifecycle’ strategy, whereby each member’s risk exposure varies with age, becoming more conservative as retirement is approached. However, it is difficult to see much standardisation occurring in this area, given that there is not even consensus within the industry regarding the value of the Lifecycle approach, let alone the appropriate risk exposure for each age.

Despite these limitations, there is one group of members where there is broad general agreement across the industry.  Our analysis was based on MySuper default options – by definition, the investment strategy that will apply to superannuation fund members who do not make an active investment choice – generally those members who are less ‘engaged’, and typically the younger members for whom retirement is in the distant future, and accumulated super is relatively small.  As already indicated, market risk should be less of a concern for this group (although managing expectations after a share market crash can still be a challenge for fund communication strategies).

This graph shows 3-year rolling returns versus risk (standard deviation over 3 years) for all SuperRatings Balanced (60-76) options.  The highlighted options are the Top 10 by Risk-adjusted return (Sharpe Ratio).  These are the options that performed the best for a given level of risk over the 3-year period to 30 June 2019.

Risk-Adjusted Returns therefore have some value in highlighting the best performers for a chosen level of risk, but even within the constraint of just Balanced options, there is a wide range of risk experience.

The SIS Act and Regulations require that trustees must formulate an investment strategy taking into account at least ‘the membership profile (for example, members’ age and expectations, occupational profile …)’.  The investment strategy decision is likely to be a bigger determinant of performance outcomes than the fund’s ability to manage assets.  Thus we have the current situation where the trustee of one fund can decide that the appropriate default level of risk for a 50-year-old is a Balanced option, while the trustee of another fund can quite legitimately decide that a Capital Stable option has the appropriate level of risk.  There is potential here for uninformed commentators to label a fund as a ‘dud’ or underperformer simply on the basis of ‘first past the post’ outcomes, when the underlying reality is that the trustee is managing efficiently to its stated investment strategy.

Einstein is reputed to have said ‘Everything should be made as simple as possible, but no simpler’.  In the world of superannuation investment performance measurement, there may be no black and white answers, and we will have to wait and see before we place any bets on future performance measurement systems.

One of the continual challenges in running an investment committee is ensuring that it operates effectively and facilitates sound decision making. There are several ingredients that can assist in this process, but the most fundamental are the governance structures you have in place. At a minimum your investment committee should be governed by a charter outlining things such as the role of the committee, its membership, record-keeping, quorum requirements, and the voting structure for determining investment decisions.

But governance is just the foundation of an effective investment committee. One critical element is of course to ensure that you have the right people on your committee (we’ve discussed this in more detail). Another is to establish among your members an appropriate structure for your discussions so that they stay relevant and focused on the issues at hand. Many of us have sat on investment committees where the conversation moves off track or thought processes shoot off on tangents. This can be unavoidable at times given the breadth of the subject matter, and it’s important not to end a discussion prematurely because the person speaking is presenting a different view. Getting the most out of your investment committee means balancing the diversity of voices with the need to have a structured discussion and agenda that facilities disciplined and controlled decision making.

To help achieve this, we believe having a clear model as a starting point to facilitate discussion is important. For example, if you’re operating an investment committee focused on asset allocation, then it’s important to clearly identify the time horizon that the investment decision is based on, as well as the relevant metrics that contribute to asset allocation decisions, whether they include asset valuations, business cycle and economic inputs, or sentiment indicators. Ideally, there will be a reference model capturing the various inputs your committee considers to be relevant.

Having a starting reference point such as a model ensures that any discussion can be framed against what the committee is trying to achieve and the information it considers important in making decisions. This ensures that discussions are focused and not skewed by the latest headlines or anecdotes, and that members can reach a clear resolution before moving on to the next agenda item. As part of such a process, all committee members should understand how their model works, what it says (and doesn’t say), and what its key inputs and sensitivities are.

Lonsec continually reviews its own internal investment committee process to ensure that it operates efficiently, captures all relevant information, and reflects our core beliefs about managing money. We also work with clients to assist them in structuring their investment committee as well as being an external member on their investment committees, bringing not only our research knowledge and portfolio expertise, but understanding of the intricacies of investment committees.

In light of the new KiwiSaver contribution rate changes, which came into effect on 1 April 2019, SuperRatings utilised their Net Benefit model to quantify the actual impact on a member’s KiwiSaver balance.

The Taxation (Annual Rates for 2018-19, Modernising Tax Administration, and Remedial Matters) Bill introduced contribution rates of 6% and 10%, in addition to the 3%, 4% and 8% rates previously offered, providing members with greater flexibility and allowing more tailored financial plans. The 6% rate also bridges the gap for members currently contributing the minimum who don’t have the means of contributing 8%.

SuperRatings analysed the difference in outcomes using their Net Benefit methodology, which aims to show the dollar amount credited to a member’s account. SuperRatings’ Net Benefit methodology models investment returns achieved by each scheme over a seven-year period, as well as the fees charged. The analysis uses a scenario of a member that has a salary of $50,000 and a starting balance of $20,000 and a tax rate of 17.5%.

SuperRatings’ analysis shown in the chart below, indicates that a member contributing 3% into the median Conservative Fund would have generated a balance of $38,261 over the 7 years to 30 June 2018, whereas a member contributing 6% would have a balance of $49,272, a difference of over $11,000.


Source: SuperRatings

Evidently, additional contributions coupled with the benefit of compounding can have a significant impact on members’ account balances over the long term. In addition to supporting members to select an appropriate contribution rate, helping members to choose a suitable fund type continues to be an important determinant of member outcomes.

Thankfully my kids have moved on from their ‘Frozen’ phase and the tunes of ‘Let it go’ are well and truly buried away in the back of the DVD cabinet. As professional investors, one of the biggest challenges we face is when to ‘let it go’. When we make an investment into a stock or managed fund the investment rationale is clear, attractive valuations, positive earnings growth, solid investment team, appropriate investment style. However, what happens when our investments don’t follow the course we anticipated and perform poorly? An even more difficult decision is when to let go of a ‘winner’?

Behavioural factors play a big role in terms of how people react to events and the subsequent decisions they make. The belief that things will turnaround, the comfort of the pack (we all go down together), ‘falling in love’ with an investment. Such emotions impact all of us even the most experienced investor. The main line of defense to minimise the impact of behavioural factors in a decision making process is to always point back to your investment philosophy and the underlying process which underpins that philosophy. If your overall philosophy is one of generating returns with lower downside risk than the market do the underlying investment align to this philosophy? have they provided downside protection? if not, why? (are there cyclical reason for this or is there something structural impact the return profile). If an investment has provided this type of return profile what have been the factors contributing to this e.g. certain sector or country exposures, and do you expect these factors to work in the future? If we use a managed fund as an example it is important to look out for any changes to how the manager is managing money which may be reflected in a change in the risk and return profile of a fund. Is there a change in how the manager positions their investment approach to what they communicated a few years ago?

The main forum for our manager and stock decisions for our managed portfolio are our Manager and Security Selection Investment Committees.  The committees are made up of senior members of our Research and Investment Consulting teams, our CIO as well as our external experts. Decisions to ‘let an investment go’ are made via the committee process. Investment recommendations are supported by qualitative and quantitative analysis. If we use managed funds for example this would include meeting with the manager (outside of the formal annual review process) focusing in on the issues at hand and targeted quantitative analysis which may provide a clue as to where the problem rests, an example being where a manager has taken stock-specific risk which is uncharacteristic of the manager.

Important information: Any express or implied rating or advice is limited to general advice, it doesn’t consider any personal needs, goals or objectives.  Before making any decision about financial products, consider whether it is personally appropriate for you in light of your personal circumstances. Obtain and consider the Product Disclosure Statement for each financial product and seek professional personal advice before making any decisions regarding a financial product.