Human investors still have the edge over robots and algorithms
THERE are myriad approaches to investing in various asset classes locally or offshore. Investors can choose offerings with passive assetallocation techniques or active asset-management styles, or a combination of the two.
At one end of the spectrum is fundamental active management. Here human judgment plays a large role in interpreting all the publicly available information about an investment. It could be qualitative, such as the trustworthiness and competence of a company’s management, or quantitative – for example, looking at a company’s financial results. Either way, the focus is on determining the merits of a potential investment.
Risk and cross-correlations also come into the equation when constructing an efficient actively managed portfolio.
Active asset management seeks to take advantage of the changing investment merits of asset classes against the backdrop of a constantly evolving macro environment and any mispricing of instruments within asset classes. It often also incorporates risk assessment – ensuring that undue risk is not taken in chasing returns. Within this framework, active managers aim to deliver alpha (above-market returns) to investors.
At the other end of the spectrum are passive investment vehicles that mimic market indices.
There are also quantitatively determined passive investments, such as smart-beta funds, which use alternative index-construction rules that take into account factors such as size, value and volatility.
There are two specific advantages to using a passive investment vehicle: it tends to be relatively low cost, and it eliminates the possibility of underperforming a benchmark.
Although they may seem to be completely different, active and passive investment vehicles are fundamentally linked. In fact, passive investment vehicles cannot exist without active managers making market-moving calls on what and when to buy or sell. Passive investments, in turn, can be actively managed or incorporated as part of an active portfoliomanagement strategy.
Active-passive strategies may include using specific asset-class trackers as building blocks in an asset-allocation strategy, or incorporating specific passive instruments, such as an offshore exchange traded fund (ETF), in a balanced, actively managed portfolio with other instruments, such as stand-alone shares.
Globally, passive strategies have seen a pick-up in fund flows, particularly in more efficient developed markets where alpha has become more difficult to come by. This has deeper implications for human active managers, who clearly still have a role to play in the modern world of investing, despite the fact that we are living in a time when robo-advisers and algorithms can make asset-allocation and even instrument-selection decisions for funds and portfolios.
One could argue that the human active manager still has the edge over the robots. The areas in which this is the case include:
• As part of the active-management process, analysts will have contact with the management teams of the companies in which they are looking to invest, or in which they are invested. This can be one-on-one, in smaller groups, on conference calls, or in a larger audience at company results presentations.
Analysing human behaviour.
Managers tend to talk a good talk, and the ability to pick up on little cues – even something as trivial as management “tone” – can be quite telling. In a 2016 Harvard Kennedy School white paper (“Reading managerial tone: how analysts and the market respond to conference calls”), Marina Druz, Alexander Wagner and Richard Zeckhauser assert that a negative tone on conference calls results in stock prices drifting downward. It has also been proved that the market reacts to the use of negative words in earnings releases, according to a 2011 paper titled “When is a liability not a liability?” by Tim Loughran and Bill McDonald.
Moral and ethical considerations.
Since algorithms typically digest only numerical information, a computer cannot effectively grasp the trend towards responsible investing.
Although the hard numbers can be analysed with more vigour, machines fall short when it comes to the softer issues. Incorporating environmental, social and governance (ESG) considerations in the investment process, identifying “funny” accounting and the assessment of management intent still require human interpretation.
In recent years, investors have begun to demand a more ethical approach to investing. They are looking for certainty in management intent and an assurance that ESG factors are properly analysed and accounted for when making investment decisions.
• Algorithms typically use historical information to make an assessment of future returns (in this case, the dependent variable). This is based on the levels or movements of one or more independent variables – for example, gross domestic product growth, the exchange rate and
Looking to the future.
valuations. The human active manager also tends to look closely at history, making reasonable predictions about how changes over time could give him or her a competitive edge.
Thematic investing is also a recent addition to the assetmanagement space. Mega themes, such as globalisation, a more connected society, the impact of resource scarcity and rising rates of obesity, can be assessed and considered when choosing stocks, instruments or markets that have exposure to such themes.
In the evolving world of robotics, it is important to note that investment is both an art and a science. It is in the art of investing that the robo approach may well be found lacking. Mark Appleton is the South African head of multi asset and strategy at Ashburton Investments.