The discipline of ‘stock picking’ is the bread and butter of most active portfolio managers. It involves deep company and industry research, poring over reports and analysis as well as conducting on the ground meetings with management, competitors and suppliers across both emerging and developed markets. Each individual investment opportunity is arrived at only after countless hours spent meticulously assessing the strengths and weaknesses of the business model, its return profile and reinvestment opportunities, as well as the starting valuation at which we are able to acquire that particular future cash flow stream.

Selecting the right stocks is crucial. However, it’s only one part of the investment process. A collection of well-chosen stocks is a good start, but they have to then be cobbled together into a larger portfolio. And deciding on each individual stock’s weighting commensurate with its risk-reward profile versus the market’s current valuation of that profile, as well as relative to the other stocks in a given portfolio, requires a different skill set.

A focus on the 'real' risk

Portfolio construction is key to determining both investment returns and the volatility of those returns. Yet, it all too often appears as an afterthought for many active managers. I believe portfolio construction should manage the ‘real’ risk to investors; specifically, the risk of a permanent loss of capital. The amount of capital that managers allocate to a particular stock should reflect both the maximum amount of capital we’re willing to lose if we’re wrong and the probability of that negative outcome occurring.

Over the last decade or so the fixation on tracking error, active money and other risk management measures which are implicitly determined by the nature and construction of the indices against which portfolios are judged, has led many investment managers to neglect the ‘real’ risk which their investors are attempting to mitigate.

Stocks are often allocated capital within a portfolio in a manner that reflects their respective weight in an index, or in order to balance other perceived risks resulting from the fund being over or under-weight a particular sector, geography or investment theme. In my view, this can lead to portfolio structures and investment outcomes that are not aligned with the objectives of end investors.

Conviction is key

I believe the strength of a manager’s conviction in the underlying business and its risk/reward profile should provide the starting point for a position size. I find it useful to begin with a thorough understanding of the potential downside for each stock and, subsequently, the range of payoffs to the upside should our investment thesis prove correct – as well as the probability of each outcome being realised.

Ultimately, the size of each position within a portfolio should reflect the real risk to investors: how much money could they lose if we are wrong? Those stocks with the least potential downside and a relatively high degree of certainty on a positive outcome receive more capital than those which might handsomely beat our hurdle rate, yet have an equal chance of significantly underperforming it.

Intra-portfolio dynamics matter too

An additional layer of the portfolio construction process takes into account both exogenous risks, as well as the interplay between all of the stocks within the broader portfolio, considering factors such as the correlation of cash flow drivers. This helps mitigate both absolute risk as well as the portfolio’s relative risk versus its index or universe, measured via such metrics as tracking error, active money and beta.

A well-constructed portfolio should be inherently highly diversified, even a relatively concentrated one of around 40 stocks (see table below). In my experience, a focus on sustainable returns, attractive entry points that compensate for risks, and low correlations between different stocks within the portfolio both enhances its diversification ratio and dampens volatility compared to the wider investment universe (which can include hundreds or thousands of names).

Source: Fidelity International, for illustrative purposes only, 29 December 2017

The resulting portfolio might look markedly different from the index against which its performance will be judged; however, it’s one which better reflects the real risks faced by end investors.