Pendock: With fewer holdings, in my experience, then you had more chance of knowing absolutely everything that was going on in those companies. So, then you reduce the chance of surprises. Further to that, if we can say, “Well why don’t we replicate that with a team of thousands” and there are companies with 5,000 analysts and guess what? There’s analysis paralysis, there’s bureaucracy, there’s political in-fighting, you have analysts who are paid to try and become PMs, they’re trying to break their stocks to the PM and it goes back to culture. So in my experience there is definitely a sweet spot for the number of stocks and it tends to be lower than the industry norm, because the industry norm tends to look the wrong way back to volatility.
Arthur: They’re managing a business model and their business risk as much as the client.
Pendock: Yes, which is often reverse-engineered to fit the consultant questionnaire.
Cantara: We manage a global equity diversified portfolio of 90 to 100 stocks, as well as a concentrated portfolio of 25 stocks. So, it really depends on how you build that concentrated portfolio. For us, it was always the names that we had the highest conviction in. What we believed was the stability of the earnings growth of that company and so, that drives a portfolio which doesn’t care about sector relative weights. So there’s big over-weights and that drives an absolute focus on, “Why do you own these names? What is the philosophy that’s driving that?” which again is an extremely important element in understanding, “Does the concentration lead to more volatility?” and in the case of the portfolios that I’m referring to at MFS, there was really no discernible difference in the volatility measures between the 25 stock portfolio and the 95 stock portfolio.
Arthur: The issue is performance tends to come relatively short and sharp and then a prolonged period of neutral to negative and so any asset owner with a one to three-year time horizon will almost certainly sack their value manager before the next uptick. So, you don’t know how long the cycle is to your point earlier, so you don’t know when it would pick up. You could still be wrong by a period of time, so your problem is performance and value comes in a short sharp shock period and it will come post a prolonged period of quite possibly gently negative performance. So any screen of historic performance will rule out value. So, having done your management selection exercises, all your managers will have shown negative five-year performance probably.
Pendock: You can’t rank one through five, one through six, the duration risk. People might have totally different ways of managing it and it’s what do we think is better? What are people looking for? And so much of what came out of that is that past performance as we know is just a factor; it’s not saying, “Well we can’t have them, they’ve done badly” so people are understanding far more. Cantara: Often when we’re talking to our contact staff at any pension fund, there’s a real understanding of that, but then board structures and oversight and others that are coming in that might have a different view of, “Well, we need to look at the three-year numbers” or, “We need to look at certain metrics”.
Lindenberg: An asset manager said something interesting to me yesterday which is that knowing when something is expensive is three times as valuable as knowing when something is cheap. It ties in with this theme of “winning by not losing”. If my value manager is under performing in an environment where all other asset classes are going up”, I’m not very worried about that. But the value of having something in your portfolio which helps protect against severe drawdowns and the impact these can have on long-term returns can be very valuable. It ties in with the discussion around concentration risk and having a high conviction manager; the key question is how this correlates with the wider risks within the portfolio? “Do I mind a manager having only 10 stocks in their portfolio when it’s a good diversifier for all of my other risks?” It’s important to assess risk holistically and focus on building an overall strategy that is robust to different market environments.
Pendock: I keep hearing stories about value managers who have been beaten up by their clients over the last five years and it’s, “Well hang on, if you bought a value manager and they’re not keeping up with the S&P 500 and whatever – well of course that’s because they’re doing their job”, so there’s also a bit of an onus on the asset owner not to torment the managers.
How can inflation be managed in portfolios?
Arthur: I do feel there’s an index-linked and correlated short-term inflation risk; there are assets that are correlated to long-term inflation, such as infrastructure, long lease property, ground rents; the problem is they are illiquid, they’re governance cost high and there are queues to get in and queues to get out because of the illiquidity. But, there is still to me, within reason, an interesting long-term inflation linkage, not short-term inflation linkage, because I see that as what pension funds should be looking at.
Lindenberg: It’s worth putting inflation in the context of the wider portfolio risk, particularly if you’re an institutional investor with long-dated liabilities, often a rising inflation environment would mean rising interest rates which is beneficial from a funding perspective. In many cases a pension fund might also have caps on the inflation-linked benefits which means that liabilities don’t increase one-for-one with rising inflation. So, in the context of the big risks facing institutional investors, inflation is often relatively low down the priority order, compared to say interest rates, equities and so on – and there are some quite simple and effective hedging strategies that you can use to remove inflation risk altogether.
Greening: Well, it’s important as a measure of liabilities and ultimately the game is about trying to figure out how you’re going to be able to pay pensions over 30, 50, however many years into the future, when you are going to have to cope with the long-term inflation.
How useful is volatility as a measure of valuation? Is it too short term?
Cantara: Any models of volatility measures are backward looking, so it doesn’t tell us anything about what’s actually going to happen and sometimes that gets confused in the messaging.
Pendock: But, then there are risk measurement models which use volatility as VAR models etc. that then get it wrong.
Cantara: You compound the errors within the model.
Pendock: Exactly until it breaks and it goes the other way and everyone goes, “We didn’t see that coming” and now everyone’s using it again because it worked so well last time.
Greening: It boils down really to whether you believe in the stuff that you bought. Because if you do end up with a series of equities in companies that look as if they have long-term prospects of growth and are in a good market. Emerging markets is something that we have focused on more recently because we see that as a potential hedge against all of the costs that increased longevity and having an older population is going to impose, both on the fund and on the employer. So, emerging markets is a potential way of being able to invest in areas where there is a good prospect of growth and where the population dynamics are different.
Pendock: I agree, it’s a hedge against QE, it’s a hedge against distorted pricing and the risk premium tends to be better. It’s more like the early 90s now in many parts of emerging markets, particularly LatAm where you’ve got country self help and reinforcing civic society, the institutions, the events going on in Brazil; it’s all about that rather than a leveraged play on Chinese GDP through commodity prices.
Cantara: When you think about volatility within emerging markets, a lot of it revolves around governance and now we’re having those discussions in the developed market. The biggest sources of political risk are in the US – Europe and who’s worried about the regimes in many of the emerging markets? It’s a little bit ironic.
Arthur: In the past, a lot of time has been spent analysing economic data, building lead indicators to help define how you’re looking at the economy going forward and the shift has been relatively gradual. In the era we’re coming into potentially, volatility is being supplied by political risk; so, in theory, we’re going to have greater diversification of returns and therefore scope for active managers to add more value. But, if it’s going to be defined by your ability to analyse political risk and political outcome, then I don’t know whether fund managers are set up to achieve that. Because they’ve spent a long time analysing economic risk and economic outcome, but not binary political outcome.
