“Events my dear boy, events,” was British Prime Minister (1957-1963) Harold Macmillan’s response to a question about what was most likely to blow a government off course. Despite this truism, the narrative around global geo-politics is widely accepted, namely that conditions are exceptional, markets are volatile and events are unprecedented.
However, it was ever thus or, as we say in France: plus ça change. We do indeed live in uncertain times and markets are volatile, but looking at financial markets over the course of decades, and even centuries, they always have been and they always will be. In fact according to the IMF, between 1970 and 2011 alone there were 431 financial crises (note 1).
That works out to be roughly 10-11 crises each year. So investors should invest based on facts and without bias, not relying on conventional wisdom where actions are explicitly or implicitly influenced by sentiment.
We think a quantitative and systematic approach to financial markets is the only way to mitigate animal instincts from the investment process – the output is something akin to passive investing with risk controls, although the input is considerably more involved. Strangely, trusting algorithms in financial markets still raises eyebrows, despite the growing popularity of products like ETF’s which are managed by computers.
In other industries, like aviation, turning the algorithms off would be considered suicidal, and in the fullness of time, the inevitable proliferation of technology and data will make algorithmic investing the norm rather than the exception.
Algorithms more involved
Of course algorithms are human constructs and the processes behind their creation and oversight should be scrutinised by investors. For the more complex quantitative strategies (i.e. not index trackers) success is a product of an investment process which is not straightforward. In asset management it’s commonly accepted that performance is a function of cleverly selecting assets and instruments whose prices move in a way that drives profit and/or creates income for the owner.
In systematic firms it’s more complex, with performance in any given year entirely a function of the cumulative actions that create a successful investment process. These range from hiring the right researchers to avoid in sample bias in data analysis, to investing sufficiently in IT infrastructure, which plays a vital role in both collecting data, executing trades and monitoring portfolio volatility. Whilst the algorithms are largely left to trade the markets in what may appear a passive way, the overall process is more involved than any traditional active manager.
Today’s investors are increasingly sensitive to fees and need diversification options beyond fixed interest, where yields are at historic lows and are focused not only on growing their assets but also preserving them. This requires a coherent risk management framework. In traditional asset management, risk is qualitatively defined as a function of the underlying investment.
This may anecdotally be correct but it’s been shown to be manifestly imprecise. For example, an investor in a ‘low risk’ FTSE 100 index tracker may now consider it to be more risky than first thought after the UK’s Brexit referendum at the end of June.
Poorly defined risk levels
This isn’t to say such investments are bad, rather that risk levels are poorly defined. CFM’s strategies are delivered at what we call targeted risk, which means our portfolios are constantly monitored to operate within the risk parameters outlined to our investors.
Lastly, investors want genuinely decorrelated returns. We firmly believe that truly decorrelated returns, or alpha, is by definition limited, and accessing it requires a great degree of skill, investment and constant innovation. This demands high fees and a commitment to innovation, meaning that for some investors true alpha funds will remain off limits.
There’s an alternative which is cheaper, more transparent, liquid and exhibits sufficient decorrelation to be a diversifier – we call it alternative beta.
So in between now and the next crisis, advisers should reconsider whether the risk appetites of their clients match current portfolios. Chances are they’re not as well matched as they think.
1: IMF Working Paper, Systemic Banking Crises Database: An Update’. WP/12/163
TOPICS: beta, correlation, geopolitical risks, volatility