Assessing the case for diversification
Share
Over the past year many managers discovered that diversification did little to protect the value of their portfolios. The problem is not a flaw in modern portfolio theory, but a trend among some managers to invest in insufficiently uncorrelated assets.
Even before we learn the difference between a stock and a bond, the first investment rule has already been learned: do not keep all your eggs in one basket. While the concept of diversity is intuitive, why is it that this method of risk management is so misapplied?
Rather than using diversity as a method to increase returns for a given level of risk, many investors and wealth managers make mistakes. The consequence is that they performed worse than they might have done over the past year. One reason is that many portfolios were “diworsified” when they should have been diversified.
In 1952 Harry Markowitz published a paper entitled “Portfolio Selection”, earning him a Nobel Prize in 1990 and setting his place in the economic history books as the father of modern portfolio theory. Markowitz concluded that, if different assets or stocks do not move completely together – that is, have a correlation coefficient of less than one – then combining multiple stocks or assets in a portfolio will increase returns for a given level of risk.
This has been the bedrock of portfolio management and fund structure ever since. But market turbulence has caused some investors to lose more money than they would have expected from a supposedly “well-diversified” portfolio.
Diversification is a good thing, but like most good things you can have too much of it. Just as the enjoyment (or “marginal utility” to an economist) of every subsequent mouthful of chocolate cake decreases, so does the risk reduction benefit of adding further assets to a portfolio.
Furthermore – if you pass a certain point – rather than receiving any enjoyment from “a little more”, you do more harm than good and potentially end up regretting your hunger. Investors are better off holding fewer investments that are genuinely diversified, than a larger number of investments that are not.
Perhaps we need to pay more attention to the numbers of individual assets within a portfolio. At the point where no further diversification benefit is derived, the portfolio is of optimal size. The addition of further stocks simply adds unnecessary costs and dilutes the manager’s ability to manage the entire portfolio.
At this point “diworsification” starts to reduce returns and many investors would be better advised to buy the benchmark return through low-cost exchange-traded funds.
So what is the ideal number of assets within a portfolio or fund? Studies reach differing conclusions; some argue that even with 100 stocks investors still benefit from diversifying further, while others argue that 10-20 is the ideal number. The correct answer is – rather unsurprisingly – that it depends. It depends on the asset being considered and the risk/return objectives of the investor. Apart from the more obvious factors such as market liquidity and trading costs, it depends on hard to quantify factors such as the increased “monitoring costs” associated with having to spread attention more widely.
Just because the costs of managing a broader range of investments are not easily quantifiable, it does not mean they should not be considered. After all, who is likely to perform better? Manager A with a broad portfolio of 100 stocks, or manager B with 30-50 stocks (which probably provides more than enough diversification) that are more tightly monitored? The problem in the financial markets is too much information, rather than too little, and having a narrower range of investments to monitor provides a valuable filter.
If costs of trading are low, if the manager does not need to monitor the performance of individual companies, then the ideal portfolio size will be larger. This might sound remarkably like a market index fund. What is certain is that as portfolio size increases to a point where it is at its maximum – that is, owning every stock in the index – it becomes harder for an active manager to outperform through stock selection. This point will depend on the types of holdings within a portfolio and the style of management used.
Given that many portfolios have become “diworsified”, what is the solution? Alternative investments such as hedge funds, private equity, commercial property and commodities have all been mooted as portfolio diversifiers, but many of these spectacularly failed to do what they were supposed to over the past year. Where there is commonality, there is a danger of “diworsification”.
Sometimes the commonality can be generic; such as the large quantities of debt used to enable hedge funds to achieve their 20% performance fees. Sometimes the commonality can be more specific; as within a single private equity fund of funds, where one of the sub funds buys an asset from another sub fund, within the same fund of funds.
When we emerge into the post credit crunch world, the successful investors will be the ones that understand they are looking in the rearview mirror when looking at “best performers” and find assets that increase diversity, not “diworsity”. To do this, some will need to discover asset classes like unleveraged venture capital, life settlements and managed futures.
People can be their own worst enemy. We are programmed to be biased by simple rules of thumb. But the most simple rule of investment is not a strict rule.
There comes a point where the benefits of diversity and good active management become negated. Investors and their advisers should spend more time assessing the factors underlying the relative performance of different assets.
Although the concept of modern portfolio theory is not broken, the way we apply it needs to be reassessed.
As John Maynard Keynes said a long time before Markowitz was born: “It is a mistake to think that one limits one’s risk by spreading too much between enterprises about which one knows little.”
