One of the reasons that pension funds’ interest in 130/30- type products and strategies has been slower to develop than expected is that these strategies have exposed the weakest link in the investment process - the ability to select suitable stocks for shorting.
The problem can be simply stated. On the buy side, research analysts and portfolio managers who are good at picking ‘buys’ are not necessarily good at picking ‘sells’.
On the sell side, most brokers’ recommendations will be different degrees of buy or neutral.
Sell side research, in particular, is systemically optimistic. Jim Connor, a partner and investment management specialist at Morse Consulting, says broker research is inevitably skewed towards buys and neutrals.
“If you look at the ratio of sells to buys you’re typically finding that most people are talking stocks up not talking them down. They’re only going to push the boat out and say somebody is a sell on the basis that they probably made a recommendation in the past that it’s a buy or a hold.
“It’s not universally true, but essentially broker research is more geared to long. So if you’re on the fundamental side, you’ve got to figure out an extension to your process that identifies suitable shorting candidates.
“It’s a mind-shift thing that not everybody will be equipped to deal with easily and quickly. As a result we may see some bad decisions made on the shorting side. So things like stop loss limits are absolutely crucial to make sure you don’t expose a client to these decisions which could go wrong.”
Connor says a key question consultant actuaries should ask is whether a fundamental manager’s investment process geared up to identify good shorting candidates. “130/30 is much more than ranking a bunch of stocks, picking the top stocks and investing in the best 10 and then going to the bottom of the list and shorting the 10 stocks at the bottom because the stocks at the bottom of the list may not be simple short candidates,” he says.
“In the long-only world you would have under weighted the bottom 10 stocks relative to benchmark, but that’s different to saying that you believe that they’re a downside play.”
The simplest solution is to rely on quant models to pick suitable shorts. It is estimated that some 80% of the 130/30 market has been captured by investment managers like State Street and Barclays Global Investors that use quantitative investment techniques
For quant managers, the 130/30 process is a logical extension of the work done by their quant models, capitalising on all the information that is available on stocks in the investable universe.
Rick Lacaille, CIO State Street Europe, says 130/30 enables quants to use all the information that is on the table. “Most of the early running has been from quants because they’re able to look at the whole universe of stocks and they’re just as likely to be effective in finding overvalued securities as finding undervalued. So the process tends to be symmetrical.”
Yet Lacaille suggests there is room for the fundamental approach. “Fundamental managers have a different set of challenges but a different set of opportunities. If they really have got deep insights into a smaller number of companies and have a higher level of confidence in those decisions the 130/30 allows them to leverage up those insights.
“That clearly is not as optimal as a strategy where you deliver a lot of value from the short side but nevertheless it’s a legitimate way of approaching 130/30.”
One advantage quant houses have over fundamental managers is that it is easier for them to explain to prospective clients how they identify candidates for shorting.
Christian Pellis, head of institutional business Europe at Threadneedle Investment, an active manager which has developed its own 130/30 product, says “The quant approach is easier to explain because it is analytical. Fundamental is about skill, it is about people, it is about how you look at a stock.
“It is not always a reason to go underweight because a stock is bad. It may be overvalued. That is something you can’t model. You need people to decide that.”
Gunnar Miller, co-head of global research and head of equity research and sector fund management at RCM, the active management arm of Allianz Global Investors, agrees that a paucity of ‘sell’ recommendations has been the Achilles heel of fundamental managers and brokers alike. “Traditionally, people on both sides of the fence have only concentrated on ideas for long-only portfolios.”One way that fundamental managers can demonstrate that they can handle the shorting involved in 130/30 strategies is by developing a ranking system for securities, he says.
RCM laid the groundwork for a shorting capability four years ago through a system of vote distribution whereby stocks are rated on a range of 5:1 votes, from strong outperformer to strong underperformer. It was also the first asset manager to buy into Starmine, a system for ranking analysts, so that it could measure the value that its research produced.
With four years of data to demonstrate its shorting skills, Miller says RCM is ready to implement 130/30 strategies. “Once you get to a four year track record you start to separate the wheat from the chaff, the luck from the skill. So our ‘sell’ signal is actually very credible.”
Miller says one of the challenges analysts and portfolio managers face is shorting stocks in a market where prices are generally rising, “A research analyst may put a sell recommendation on a stock on the basis of its underperformance relative to the group of stocks to which it belongs. If the group is expected to go up 10%, the analyst’s buy recommendations will be the ones that are expected to go up 15 to 20% and the sell recommendations are the ones that are expected to go up 0 to 5%.
“But should you short the stock that you expect to go up by 5%? That’s where some of the first pushback comes. The fund manager may say if it’s a sell idea it should be a stock that goes down absolutely.
“A lot of fund managers are now saying if we lose money on a short that goes up 5%, that’s life. We’re not going to remain lopsided and throw out potential sell ideas simply because they have some element of expected upside within a research sector pair trade.”
Having a hedge fund operation in Frankfurt also provides a reservoir of shorting expertise on which 130/30 managers can draw, Miller says.
“Balancing an in-house hedge fund with other groups has always been a difficult proposition, but with the advent of 130/30 all of a sudden we have a ‘go-to’ person.”
The problem with long-only fundamental managers extending their stock picking from buys and holds to sells is that they are being asked to change their game, says Griff Williams, institutional product strategist at Pioneer Investments.
“When clients look at a fundamental manager’s list of buys and sells they often find that there’s a big section of the universe unresearched because, for whatever reason, the manager has decided that those stocks do not meet the criteria required by the long-only process, hence they simply get neglected.
“For the fundamental manager to then turn around and say we’re now going to pick those up is not very convincing. And if they need to build a quant model to initiate coverage, they usually have to start from scratch.”
Williams says the solution is to combine the analytical strengths of quant model with the intuitive skills of the fundamental manager. “The way that we do it is through a hybrid approach whereby we have an existing long-only fundamentally based portfolio. We add to that a long/short portfolio and we implementthe long-short component through a total return equity swap.
“If you think of two integrated moving parts, one of the moving parts is the quant and that’s where all the short ideas are. This is then combined with an existing fundamental long only portfolio.”
In this way quant and fundamental can work together, the one providing the other with the information it needs, says Williams.” It’s a circular process. The fundamental managers and analysts are constantly feeding their experience and their ideas into the quant side and vice versa. It also solves the problem of universe coverage for short selection because the shorts are selected by the quantitative process which encompasses the entire universe.”