Peter Elston, CIO, Seneca Investment Management, says too many people have retirement savings invested in poor products…
I abhor poor active management because there are so many people whose retirement savings have been invested in poor products; because it has given the active management industry a bad name; and because the big firms who are the worst offenders (1) continue to peddle the notion that bigger is better. With, for example, football clubs, bigger generally means better quality. This is absolutely not the case with active management; the more interesting investment opportunities tend to be smaller and out of the reach of large firms.
While I abhor poor active management, I adore Gina Miller (on a strictly professional basis I should add). I too get an itch when I think “people are being bullies, or being dishonest or hypocritical” (2). Most big fund management companies for me tick all three of those boxes. There’s plenty of evidence that most actively managed funds fail to beat their benchmark net of costs (3). The implication of this – and the associated headlines – is that funds that beat their benchmark have done a good job, and that funds whose performance is in line with the benchmark have achieved what they set out to achieve. This has to be wrong.
Active managers should aim to beat their benchmark by a wide margin to compensate investors for the risk that they will fail to reach it. Why on earth should an investor accept index performance with risk when they could get, from a passive fund, index performance with no risk? Most single asset class funds have an investment aim or objective that is vague, and a benchmark which is an index (4). Where there’s a benchmark stated, it’s not generally clear what it’s there for, but let’s assume it is for performance measurement purposes. After all the proper definition of a benchmark is ‘a measuring device’ not ‘something to be copied’.
I have a simple solution. Active funds should state the margin by which they aim to beat their benchmark, net of costs. For example, FTSE All Gilts + 2 per cent per annum, or S&P500 + 3 per cent per annum. We do this at Seneca, either explicitly or less formally, accounting for the value we seek to add from active management decisions. And why would performance in line with a benchmark be acceptable when this means that managers give all the outperformance to themselves in fees, once other costs have been paid, leaving nothing for the customer! I have a different approach.
My starting point is to consider the costs for a particular fund, then to aim to produce a multiple of these costs in gross outperformance (alpha). I make sure that we have sufficient tracking error to give our funds the potential to produce this alpha (too little tracking error is in my view worse than too much) then trust our value-oriented investment style and process to achieve it.
What is the multiple? It depends on the total costs, but for all three of our funds, it’s considerably above two. We recently changed the benchmark for our investment trust, the Seneca Global Income & Growth Trust, to:
‘Over a typical investment cycle, the Company will seek to achieve a total return of at least CPI plus 6 per cent per annum after costs with low volatility, and with the aim of growing aggregate annual dividends at least in line with inflation, through the application of a Multi-Asset Investment Policy.’
Benchmark changes are generally met with great scepticism, and rightly so, but in our case we have raised the bar rather than lowered it. The previous benchmark of LIBOR + 3 per cent did not reflect how the trust was being managed. The change won’t mean we have to start jumping higher (we won’t change the way we manage the fund). The bar has been raised to a level commensurate with our process rather than at an inappropriately low height.
I’d argue that the objectives of most actively managed funds are not ambitious enough. Perhaps this is why the active management industry is so despised. Many, including me, think it is still providing a safe harbour for the cowardly.