Risk & Portfolio Construction: ‘All equities are black boxes’
In January 2011 the PNO Media Pension Fund terminated its US and European equity enhanced-index mandates with Barclays Global Investors and brought the assets in-house. However, it did not go passive – or at least, not in the sense in which most of us would understand the term.
Instead, it took the largest 200 stocks from the MSCI Pan-Euro index and all the stocks from the Russell Top 200 index and equally-weighted them, and now rebalances each portfolio once a year. In November 2011, it set up a similar equally-weighted benchmark for a ‘passive’ emerging markets portfolio with Robeco. All three together account for around one-fifth of the fund’s equity allocation alongside active mandates with FDA, Aberdeen Asset Management, Principal Global Investors, and Lupus Alpha.
PNO Media’s board always agreed that they wanted a part of their equity portfolio to be managed passively, but fund director Jeroen van der Put recalls the philosophical discussions about what exactly they meant when they said ‘passive’.
First, they decided that a passive strategy should reflect the broad economy in its diversity. While the market cap-weighted index arguably reflects the broad market, the dominance of a handful of stocks at the top ensures that it does not reflect the broad economy. That provided the impetus to move to an enhanced index.
But the second requirement was that it should not have to trade too much and that, when it does trade, it should not risk a negative impact by following market momentum or being forced to buy or sell with the market, mechanically. The market cap-weighted index scores well in terms of volume of trades, but not in terms of the negative impact of its trades – which tend to happen when constituents enter or exit the index. The enhanced index improved on the ‘dumb trading’ aspect, but introduced excessive, costly turnover.
The equally-weighted portfolio, PNO Media decided, is a good compromise on the turnover problem, a clear solution to the diversification problem, and offers the rebalancing effect to counteract the tilt to momentum in the market cap-weighted portfolio. And the case is sealed, Van der Put suggests, by a third requirement.
“You could argue that you are passive if you are not taking a view on expected returns from the market and that you consider all those equities out there as if they were black boxes,” he says. “That leads you towards weighting those equities equally, and to the maximium level of diversification. We thought that was particularly important in a world where black swans come along much more frequently.”
Van der Put talks about the “maximum level of diversification”, but the equally-weighted portfolio is a crude – albeit pretty effective – way to realise that goal. Risk parity and maximum diversification practitioners would point out that equalising capital allocations does not equalise risk allocations (because lower-volatility stocks are weighted equally with higher-volatility stocks), or maximise diversification (because no account is taken of correlations and common factors). While Van der Put acknowledges this – and adds that the fund is looking at the risk-parity concept at the whole-portfolio, multi-asset level – he also draws attention to the problems associated with the optimisation necessary to implement these alternative approaches.
“Going in the direction of risk parity or maximum diversification involves forming opinions about volatility and correlations that are based on historical observations that may not be certain to persist into the future,” he reasons.
Similarly, PNO Media does not ‘correct’ for the ‘active’ sector weights that appear against the market cap-weighted index – in Europe the largest underweights are energy at six percentage points, and healthcare, while industrials are overweight by about five percentage points and utilities by about 3.5 percentage points – nor for the considerable tracking error against that index.
“It’s a question of your starting point,” Van der Put says. “If you look at the world through market cap-weighting glasses then you will see high tracking error if you go down the equally-weighted route. If you put on equally-weighted glasses you see the high tracking error of market cap-weighted indices. What counts is whether you have an efficient portfolio or not.”
But there was another, equally important reason for choosing equal-weighting: low cost and simplicity.
“The ESG policy we have means that complexity would be introduced if we started using an external benchmark that would then have to be tailored for us,” Van der Put explains.
“We wanted to keep our solution simple and cheap, so that we could easily integrate our ESG requirements. We looked at things like minimum variance, too. But the question is always about the balance between the added value and the added complexity. While some of the trustees on our board are CFOs, they are not professional in this area: from the perspective of the ‘market standard’ equally-weighted portfolios represent something different, and that is something we have discussed in a lot of detail with our trustees, but they are confident in this process because they clearly understand the problem of companies being like black boxes.”
There are still one or two questions that can be asked about PNO Media’s decisions in this area. Some might consider it strange that, given that one of the advantages of the equally-weighted portfolio is that it corrects the egregious large-size bias inherent in the market cap-weighted portfolio, the fund has elected to apply the new approach to only the 200 largest companies in each respective market. Van der Put counters that the objective wasn’t to capture the small-cap effect – which it does through its separate active smaller companies mandate with Lupus Alpha – but simply to create an intuitive, cheap, low-turnover and diversification-maximising approach to passive equities.
Van der Put also explains how the move to equal-weighted passive equities forms part of the three-step strategic development of its investment programme: first, the fund has begun to replace some of its equities allocation with alternatives; second, it has introduced some strategic thematic tilts into its portfolio, such as emerging markets and high-yield bonds; and third, it has explored more efficient implementation for some benchmarks – which is where equal-weighting comes in.
If anything, the market cap-weighted portfolio makes more sense in equity markets than it does in bond markets – where it tends to leave investors allocating the most capital to the most indebted borrowers. Does PNO Media therefore have plans to extend its experiment with alternative benchmarks into its fixed-income portfolio?
“It does have our attention, but it is not currently our top priority,” Van der Put concedes.