By Joao Frasco
There is a common misconception that a multi-managed fund should, at best, give you average or median performance. However, we can point to examples of multi-managed solutions that have topped their categories, out-performing the best of single-managed active funds.
To explain this, let us first understand what a multi-managed fund is and what it is not.
Multi-managed versus fund of funds
A multi-managed fund does not need to be a fund-of-funds, as these are two quite different things. A fund-of-funds has to invest in other funds, which means that it needs to select from the universe of available funds. It may select from funds with vastly different mandates, as long as the fund in aggregate meets the fund-of-funds specific regulatory requirements.
A fund-of-funds can be multi-managed, but does not need to be. It could just combine various funds with different mandates from the same manager.
A multi-managed fund can be a direct fund (and often is), that provides segregated mandates to each of its underlying managers. Each of these managers therefore invests directly in the underlying securities on behalf of the fund.
This allows the fund to invest not only in managers that have funds, but also in other managers that may not have funds of their own (they may be managing assets for pensions funds or life companies, for instance). The manager could also provide mandates to the managers that make them different from any other funds they may have. For example, a manager could have a fund that invests in global and local property, but the multi-manager could provide a segregated mandate that invests locally only.
The above implies that a multi-managed fund could theoretically out-perform a universe of funds just on the basis of being able to invest in managers and mandates outside of what is available in the universe.
A multi-managed fund can out-perform the universe of available funds for other reasons as well.
Let us assume a simple case where only three funds exist in the universe, and the multi-managed fund only invests in two of them. Let us assume further that all three funds have a period of performance where they are in first place, and a period in which they are last. On average, over time, the funds may all have similar performance, and hence their individual ranks do not really matter.
A multi-manager that just selects randomly from this universe should also not expect to out-perform, as it simply has exposure to two of the three funds that have each only produced average returns.
However, a multi-manager with perfect foresight would be able to select the two best-performing funds for each future period, and hence not participate in the funds that under-perform. It should be obvious that in this impossible case, the multi-managed fund would out-perform all three funds.
Philosophy and Process
Now while we will concede that we do not have perfect foresight, we do spend a lot of time researching managers and their funds and capabilities. This is the foundation of a multi-managed fund, and what we need to do is to construct a fund made up of the ‘best’ funds in our estimation. (By ‘best’ we mean both in terms of performance and providing diversification).
While these funds may not end up being the best over any given period, we may be able to include the better funds and exclude the worse funds over time just enough to out-perform the average and/or median fund in any given short-term period. If we are able to do this consistently over time, our funds’ performance and ranking should rise.
It is very easy to get caught up (and caught out) in simple heuristics, like assuming that multi-managed funds will on average only give you average performance. While over the short term it is easy for a single manager fund to be number one, it is equally easy for it to be last.
For multi-managed funds, this is nearly impossible over the short-term. And, over the longer-term, multi-managed funds can provide average-beating returns and better.
Joao Frasco is CIO at STANLIB Multi-Manager.