- Balance risk management and conviction
- Diversify at the asset class level, too
Professional stockpickers have a patchy record: research by Morningstar published earlier this year suggested 90 per cent of funds bought more losers than winners between 2013 and 2023.
But this doesn’t necessarily mean they lack skill. What if they are just buying too many shares, diluting high-conviction plays with filler stocks? The academic paper Best Ideas (Antón, Cohen and Polk, 2009) suggests that while fund managers’ favourite picks often beat benchmarks, performance trails off after about the top 20 stocks they hold.
That insight brings to mind a perennial question for private investors: what is the optimum number of shares to own, and what level of concentration makes a portfolio too risky? The research paper's findings also raise an intriguing possibility: to take fund managers’ best ideas and put them to use in a tidier, nimbler portfolio.
Fund managers’ top picks
The IC Ideas Farm pages publish selections of fund managers’ best ideas. Grouping funds by their Morningstar categories. The most recently disclosed top holdings are compared with benchmark weights and the most ‘overweight’ positions are rated as best ideas.
One flaw is that sometimes a stock can have a very good run and end up looking like it’s still a favourite idea, when in reality it’s an ‘old news’ winner and the manager needs to do a little pruning. On the whole, however, these lists will reflect stockpicking convictions.
In the case of UK large-cap funds, our Ideas Farm process most recently highlighted names including Gamma Communications (GAMA), Bellway (BWY), 4imprint (FOUR), Serco (SRP) and Moneysupermarket.com (MONY) as popular overweight positions. Globally, Microsoft (US:MSFT), Amazon (US:AMZN), Dutch semiconductor business ASML (NL:ASML) and French luxury conglomerate LVMH (FR:MC) were favoured.
Does this make these companies buys? Not necessarily. Nonetheless, it’s worth exploring the rationale for conviction investing and juggling this with the portfolio theory that, by contrast, emphasises diversification.
The ‘original quant’
The oft-repeated quote about diversification being the only free lunch in investing is attributed to economist Harry Markowitz, who died last year.
Markowitz developed the Modern Portfolio Theory in the 1950s, for which he later won a Nobel prize. The gist of his thinking is that an investor’s ideal portfolio is one that offers the maximum level of return for a given level of risk. Markowitz created a mathematical framework for this, known as mean variance optimisation.
He also became a mentor to Bill Sharpe, creator of the Sharpe ratio, which has been widely adopted in the investment industry as a method for measuring an asset’s risk-adjusted return.
In an obituary in the Financial Analysts’ Journal last year, Yale University professor William Goetzmann noted that Markowitz’s theory changed the way in which the investment industry operated, moving from a bottom-up approach of analysing individual securities to a top-down view of portfolio construction. He described Markowitz as “the original Wall Street quant”.
Modern Portfolio Theory encouraged a broadening of investors’ portfolios and ultimately gave rise to a passive investing industry that has worked wonders for many. Passive funds that cheaply track broad market indices such as the S&P 500 or the FTSE 100 have proved to be the ultimate diversification tools, ensuring that investors avoid managers that overcharge for funds that underperform. Their share of the funds market has more than doubled over the past decade, to 41.2 per cent at the end of March, according to Morningstar Direct.
Yet there are many who feel that passive funds’ growth has had a distorting effect. Speaking on the Money Maze podcast last year, David Einhorn, founder of New York-based hedge fund Greenlight Capital, argued that flows to passive vehicles have “bifurcated” markets, leading to a “wasteland of companies where literally nobody is paying attention and nobody is following”.
The outcome can arguably be seen in the S&P 500, where the Magnificent Seven group of tech stocks now comprises around a third of the index’s market capitalisation. For a value investor like Einhorn, the opportunity lies in the “washed out” shares, whose low prices offer enhanced yields. A growth investor, on the other hand, might want to stick with the companies that now dominate the index given their track record and potential to maintain above-average returns.
And yet, if an investor chooses a fund to match either style, they run the risk of buying into a vehicle with holdings so diverse that it is effectively a closet tracker. C Thomas Howard, chief executive of Athena Invest, a consultancy specialising in behavioural finance, says the active fund managers with whom he interacts typically profess to having between 10 and 20 strong investment ideas. By contrast, the Best Ideas paper found that the average fund holds 160 stocks.
Howard applied the Best Ideas methodology to a series of stocks deemed to be managers ‘best’ and ‘worst’ picks, defined as the shares held most and least often by the best US active equity funds between 2013 and 2022. The best ideas all outperformed the net return of the equally weighted S&P 500 index, while the worst ideas all underperformed.
If a fund manager only considers 10-20 of their ideas to be their best, “that’s all you should invest in”, Howard argues.
He blames the industry’s reliance on tools such as the Sharpe ratio for contributing to the “myopic loss aversion” that affects both institutional and DIY investors alike – ie, the fear of short-term losses overrides the desire for (and often, therefore, the achievement of) long-term gains.
The Sharpe ratio measures returns in terms of their standard deviation, which is a measure of volatility, rather than risk, Howard says, adding: "If you’re a long term investor, you shouldn’t care about short-term volatility.”
How many stocks should you own?
If you believe it is possible to beat the stock market, it stands to reason you should choose individual shares and not just buy a tracker. Overweight positions are necessary to see more benefit from the best-performing stocks.
But it's worth noting that the marginal benefits from diversification start declining after a certain number of stocks are owned.
Work by the economist Joram Mayshar in 1979 emphasised the trade-off between diversification benefits and the cost of owning each additional share in a portfolio. His study aimed to show the optimal number of stocks an investor should hold before the trade-off between reducing risk and higher costs is no longer attractive. Costs may have come down since then, but the method is still useful in helping active stockpickers set a rule of thumb for how many shares they should work with.
Because it’s interested in looking at utility as a function of risk versus cost, the Mayshar model assumes a level of volatility, returns and correlation between all stocks. That is a useful starting point for the examples we use in this feature. Our own calculations have allowed for a 30 per cent annualised volatility for individual stocks: a realistic figure when working towards the goal of showing the benefits of diversification. In real life the stocks investors would select would differ from the examples we include. We are simply interested in how many shares we can manage, although other models should inform the best risk-versus-reward decision for weighting stocks in a portfolio.
Variables specific to the investor include the amount they have to invest, what it costs to buy an additional stock, and their attitude to risk. The latter is denoted by the coefficient ‘theta’: in short, this is a number, typically from 1-10, where lower values mean the investor is prepared to expose their portfolio to greater risk.
For our example, we assumed a £7.50 transaction cost and worked with a theta of 3.5, equating to an investor with a medium to moderate risk tolerance. If the correlation between stocks is around 0.8 and if our fairly risk-tolerant investor had £100,000 to invest, these outputs suggest the ideal number of shares would be around 20 holdings. That’s manageable and, interestingly, chimes with the Anton, Cohen and Polk study on fund managers' optimum number of best ideas.
In markets with more dispersion, there will be greater benefit to diversification. Adjusting the assumed correlation between stocks to 0.5 means the model suggests an investor would receive maximum utility from owning 32 shares. Beyond 30 shares, however, the incremental declines in portfolio volatility are measured in single basis points. That suggests it would be less worthwhile adding more holdings above this level, especially if the opportunity cost is losing returns by not focusing on the best ideas.
Investors with larger pots will be less affected by the marginal costs of trading – but they still don’t want to turn their portfolio into an expensive closet tracker by over-diversifying. One could argue, therefore, that investors who want to gain diversification benefits beyond their 20 favourite shares should invest in a supplementary exchange traded fund (ETF) tracking the market benchmark.
For those starting out on their investing journey, the impact of costs is more telling, and the Mayshar model suggests that a moderately risk-tolerant investor with £10,000 would gain most utility from holding six or seven stocks in a high-correlation environment, or 10 stocks if there was wider dispersion. Of course, no investor with a limited amount of money with which to absorb losses should take on too concentrated a risk, so some brute force diversification with a market tracker would also make sense here.
That said, the potential to be handsomely rewarded for strong convictions is one of the great things about investing. The compromise for smaller pots is to go half and half, albeit a common-sense rule about the minimum size for individual shareholdings is also helpful here.
For example, if half the small investor’s £10,000 was in an index ETF and the other half was earmarked for stock picks, the Mayshar model suggests splitting the 'active' portion between seven holdings in a more dispersive environment. Be that as it may, it makes sense to set a rule to hold no less than £1,000 in any single holding (so that the potential returns are able to move the needle for the portfolio as a whole). So five high-conviction ideas plus the tracker represent a good compromise for a relatively new investor.
Of course, there are diversification principles to consider even before delving into the equities portion of a portfolio. Investors should consider allocation levels to the likes of bonds, other assets and cash for liquidity requirements. That caveat made, shares do the heavy lifting in any portfolio and ought to be the biggest asset class.
Working best ideas into a portfolio of shares
Let's return to our relatively risk-tolerant investor and their £100,000 equities allocation. For the purposes of this example, we'll continue to treat stocks as reasonably correlated with one another (which was true for much of the 2010s, when easy money meant a lot of co-movement, but may not be as true now). In this case, around 20 stocks would give the best trade-off between diversification and costs.
Let’s say the really good ideas were just the nine UK and international stocks we mentioned at the top of this article. Alongside them you might choose to hold ETFs tracking the FTSE 100 and the MSCI World indices. Some of the larger companies are going to be well represented in the trackers, so additional holdings in the likes of Microsoft, Amazon and LVMH make for ‘overweight’ positions.
To these nine, you might add a riskier idea with the potential for higher growth; say, a small-cap stock. For the sake of an example, we’ll include Franchise Brands (FRAN), a franchised industrial vehicle and plant hire business, on the strength of it now being an overweight position for three of the UK small-cap funds surveyed by our Ideas Farm process.
The point is to invest within the overall framework to manage risk, but still have a meaningful position to profit from. Our stock ideas each warrant an investment of at least 2 per cent of equity funds, and we give each an upper limit of 5 per cent.
In the case of US mega-caps we'll be more exposed because of their weightings in benchmarks. Microsoft accounts for over 4.5 per cent of the MSCI World index and in our optimised portfolio (which contains some tracker fund exposure) we’ll end up holding more like £7,600 in Microsoft shares. That’s a confident position but not excessive. Still, it’s something to watch.
Restrictions applied, we set our solver application the task of creating an equities portfolio that minimises modified value at risk (MVaR), or the threshold below which we wouldn’t expect daily returns to fall on 99 per cent of occasions. The results are set out in the table, which shows we’d have just over two-thirds of equity investments in the trackers, leaving a decent tilt towards active choices. Of course, with more good ideas, the minimum weighting to trackers can be reduced further.
The right asset allocation
For our final considerations, let's assume this investor is following the traditional 60/40 portfolio model, with 60 per cent of their assets in equities (although this may well be too cautious for many IC readers with long-term time horizons). That gives us an extra £67,000 to add to our £100,000 in equities. For the rest of our hypothetical investor’s £167,000 – if we’re following a variation of the 60 per cent equities portfolio – the holdings in alternative assets and bonds add another layer of diversification.
As governed by the Investors’ Chronicle Alpha moderate risk asset allocation model, we have 5 per cent in gold and 34 per cent in government bonds – mostly US Treasuries, although UK investors may want to hold more in gilts due to currency risk. Tactically, decisions also need to be made regarding interest rate risk: bond funds with a shorter average duration will lose less value if base rates surprise to the upside, and vice versa.
All the same, if we keep 5 per cent of the overall portfolio in longer-duration gilts then we lock in a decent yield and will not miss out if inflation falls and rates are cut. As it stands, inflation is proving stubborn, so it remains sensible to hold a decent portion of fixed-income allocations in low-duration funds. For dollar holdings, this could mean short-duration US Treasury Inflation-Protected securities (Tips). But it is also worth moving some of the dollar-denominated allocation over to shorter-duration gilts – a risk-reducing move for investors counting their returns in pounds.
Such a move retains US exposure, but is mindful of America’s bloated fiscal position, potentially a source of future market shocks. Earmarking some capital for a mid-duration strategic bond fund would also allow an investor to take advantage of a manager’s skill in picking a blend of corporate, government, regional and asset-backed debt issues that could outperform in uncertain times.
More broadly, there is an important place for active decisions in successful investing, but it isn’t a case of either/or. If you can get a smart asset allocation and some passive holdings working for you, that will free up your mind to make the most intellectually and financially rewarding stock selections.
