Disruption is happening all the time. And it will happen in the active investing business too. What is active investing? Anything which starts from intraday trading to mutual funds to hedge funds can be bundled as active investing. Anything which is designed to conserve capital (even if it does not) and deliver absolute returns is Active. Anything non-standardized using fundamental analysis, quantitative, technical, behavioral or based on interdisciplinary studies also would come under the same classification. Anything not Passive can also be called Active. Anything held for a periodic rebalancing of 3 months and higher (say up to 60 months) would be more towards Passive.

The Clear Distinction

The clear distinction between the two could be absolute or relative returns, indexed or non-indexed, and shorter or longer holding periods. So where is the disruption? Since Passive is traditionally indexed while Active is considered the domain of experts. Because absolute money management is considered a skill set which needs product knowledge and expertise, Active style is considered primarily discretionary, and generally not indexable. This is where we have the potential of disruption.

An Apple cannot become an Orange

So if we could build a strategy which is absolute return, works cross assets, cross regions, cross risk preferences and we can Index it; what will happen if such an indexing approach framework was possible? Active money managers could have a better benchmark to compare returns. Every strategy would not by default be benchmarked to S&P500 or any popular index. Out of the large passive universe there, it’s easy to pinpoint where all your Active strategy outperformed. Your Active strategy will invariably outperform a lot of Passive me too universe. But how realistic and accountable is it? It’s not. Maybe the accountability would work out in favor of the Active models, which have to keep defending themselves against S&P 500. In the long term we are all dead, so how can you compare Active with Passive over the long run and say Active underperforms Passive most of the time. You can compare an apple with an orange and say the apple is not as good as the Orange or vice versa, but the comparison is redundant.

One Risk Preference?

Active and Passive are two different risk preferences. It’s like the school of thought which says Value is better than Growth. I have a risk preference, what if my risk preference is not about holding value (reversion) for the long term, but playing with growth (momentum) for a shorter term. It’s this risk preference the investing community has muddled up. The fruit market is for everybody, somebody needs apples, somebody needs oranges and somebody grapes. If we don’t sell Palinka to the wine drinker, why do we undersell the Active as Stone Age to sell the glorious Passive and vice versa? No customer surprise, delight, satisfaction here. “This is the only solution”. We disagree.

It’s all about Design

I have been interacting with two friends, one is pursuing a doctorate in architecture and the other owns a fashion label working out of the vortex in Paris. I am looking at extending the idea of Jiseki to fashion and even architecture. I have been telling them that fashion data is predictable. You can predict the color of the next season, the style, the growth region, basically the preference. While speaking to the architect, she innocuously mentioned, “Mukul it’s not about architecture, it’s about design”. And it suddenly hit me. Are we not all in the business of design? The design comes from data universality; designing fashion, architecture, financial innovations etc. The design is an ever-evolving symmetry, moving from inefficiency to efficiency and evolving with Time. The financial industry has to learn it too. And the superior design has to be natural, so if a certain data design can drive fashion, architecture, web data, it can drive financial innovations too; it does. Extreme reversion is a phenomenon working across natural data systems.

Designing the Active

So how do we design an Active index, which gives the Active investing a framework to benchmark and defend itself from a simplistic comparison with the S&P500? What’s your strategies’ volatility, holding period, return preference. If you build a US 30 which selects 30 stocks from the US 500 with a holding period of average 60 days or an India Active 10 which selected 10 stocks out of India top 100, with 50% of the universe coming from Banking, Technology, Pharma and Metals with a volatility less than 20% annualized, you need a benchmark different from the S&P 500. You need a benchmark that is similar to your Active strategies’ risk preference. How will it help you? Apart from the benchmarking, it brings in accountability to any Active strategy out there which desires to adopt better design practice and wants to be transparent about its asset selection process. Sooner or later a better design matching a risk preference has to emerge. This is why the need for Active Indexing.

Why did it not happen?

First; because of the false perception that absolute return can only be discretionary, as it can only be skill based rather than the system based or in other words, skill can’t be quantified. Second; quants were narrow in their data focus, now that they have stepped out to focus on interdisciplinary data, they are ready to work across fundamental, technical, quantitative, behavioral or internet data etc. Third; geographical bias kept us focused regionally and asset focused, we did not think that what we use in India could work across the globe, or what works across the globe could also work in Brazil. Fourth; love for status-quo. “It’s not my job to bring accountability to my sector.” “Who will do it?” Fifth; the need to question ethics. Insider trading has been rampant even outside the USA. It is easier than building frameworks or bringing accountability. Sixth: lack of thinking about universal laws. Could one strategy work across risk preferences, regions, assets etc.?

The Design Variables

Before we design we need to understand the multi-variables in the Active strategy that could be summarized by the Active Index. Average holding period for components; quality of components; classification of components, role of cash (to have it! not to have it! or how much to have!), volatility and drawdown limits, monitoring frequency (intraday, end of day, weekly, monthly…); leverage or non-leveraged; expected return. Assuming these are the broad categories for an Active strategy, an Active index will have to skim through these variables and look for an efficient design.

Considering we are looking at a benchmark that is tradable, standardized, and customizable, is a good proxy, we are looking at the quality universe (blue chip), generally not the 500 but the top 100 to select from. The role of cash is essential for an active portfolio strategy, to have the cash to allocate when to go cash when to go all in. Cash also plays an essential role to deleverage, fasten or slow down the strategy. Hence an Active Index should have cash as an essential variable element. Holding period cannot be too short, as it conflicts with execution for big money. Active Indexing should be designed for at least a billion dollar scalability. The holding period cannot be too long as then Active style starts blurring into Passive (longer holding period).

Can design encompass skillset?

Now there are two ways to look at it. “A good design can never outperform a stock picking expert skillset” or “I am open to looking at good design that helps me understand and appreciate that an interdisciplinary quant strategy can lay down a framework for me to follow and benchmark”; “It could change my business as lack of accountability has harmed Active more than benefited it”. We are no more in a world of, “My strategy is smarter than you”. It’s all about “Can we really build an Active Index?”

A design that could achieve absolute return while keeping the active investing variables in mind could be only because of superior selection and an understanding of natural laws. Moreover, an Active model that makes an absolute return and is standardized is no more market timing, but a selection framework that understands value; growth; momentum; reversion. And when an interdisciplinary quant system can do that, how different is it from a fundamental value selection system. Is selection not ultimately about ‘good in good out’? And how can “good” be different for fundamental and quantitative?

Can Active Add or Subtract?

Behavioural finance has questioned the investor's ability to add or subtract, bringing out the gaps in investing and explaining how few understand probability. Of course, the same idea can be extended to Active investing, which should invariably look at matching expectations vs. risk preference, rather than lose itself in the winners to losers ratio, or in any other design variable. This also leaves the field of investing open for quants who can match expectations with risk preference, irrespective of the fact that a few like Montier (Behavioural Investing) may consider quants as peripheral to the investing industry.

Active Quant

We did our bit to build an Active Index framework that can be used globally, is a cross-asset and can we tweaked for a risk preference. This, of course, happened thanks to real money which seeks good design and ask questions. Can I reduce the number of annual trades? Can I just focus on blue chips? Can I reduce portfolio volatility further? Can I get similar returns with half the trades?  Can I mix value with growth? Can I have weekly exits and monthly entries? 

Design should reduce noise in an active life and do more for the industry than what has been attempted till now. When an Active manager complains about lack of sleep, the design of his bed becomes less important than the design of his strategy.