How benchmarking can hurt your investment returns

Benchmarking has always been a popular starting point for constructing a portfolio. However, sticking rigidly to the mark could in fact hinder potential returns, writes Stephen Kwa. 

Most Australian investors benchmark their global equity portfolios to the MSCI World ex Australia index.

This, in turn, has encouraged fund managers to use this index as the starting point for building their portfolios, particularly in terms of setting risk controls for individual stocks or sectors and region limits.

In contrast, we are firmly of the belief that investors who slavishly follow market cap weighted indices are actually restricting the future potential returns of their portfolios, effectively leaving money on the table. In this short article, we outline our view on what investors should do to generate higher returns in global equities.

What’s wrong with a cap-weighted index?

The primary issues with capitalisation-weighted indices are:

• it is biased to big stocks, not necessarily the best stocks;
• large cap stocks have historically underperformed;
• cap weighting follows momentum — buying high and selling low; and
• it artificially restricts the investment universe and is typically focused on developed markets.

Cap weighting by definition will overweight big companies. As at 30 April 2010, half of the exposure of the MSCI World ex Australia Index was to the top 150 stocks versus 1,572 in the total index. Put another way, just 10 per cent of the stocks give rise to 50 per cent of the index.

Figure 1 shows just how skewed a capitalisation weighted index is to mega/large cap stocks.

Such a bias towards mega and large cap stocks is fine if such companies actually outperform over time. However, the evidence actually suggests the opposite.

If we had bought the largest 10 stocks in the S&P 500 each year from 1926 to 2004 and held them for a 10-year period, the return would have actually lagged the return on the average stock by 3 per cent per annum.

Next is the problem of momentum. Cap weighting naturally increases exposure to stocks that have risen in price and reduces exposure to stocks that have gone down in price.

In effect, it is a buy high and sell low strategy that is the exact opposite of what a successful value investor should be doing.

With regard to the investment universe, while 1,572 stocks in MSCI World ex Australia may sound like a broad investment universe (especially relative to the 200 or so stocks in the S&P/ASX 200 Index that we are more used to), some managers attempt to broaden the universe further by including another 800 or so stocks from emerging markets (which comprise the MSCI Emerging Markets Index).

However, even this pales in comparison to the more than 15,000 investable stock opportunities on stock exchanges around the world. There is a world of opportunity that is being ignored by most benchmarks and most fund managers.

Three steps to higher returns

If traditional benchmarks and cap weightings are not a sensible place to start when building a global equity portfolio, what should an investor do?

We believe there are three simple steps that strategically focused investors should take to achieve higher long run investment returns:

• be unconstrained — employ fewer constraints and invest in the widest possible universe;
• adopt non-market cap weighting — do not weight stocks based on their market capitalisation; and
• focus on value — valuation focused stock selection strategies outperform in the long term.

Why unconstrained?

Put simply, constraints cost you return. Removing constraints greatly increases the chance of benefiting from the skill of a manager through a greater breadth of investment opportunity. The obvious constraints we suggest removing are:

• universe restriction; and
• benchmark relative risk controls.

Investors should look more broadly than just at large cap companies in developed markets (the domain of the MSCI World ex Australia Index).

There is strong evidence of a ‘small cap effect’, where the excess return to small cap over the past half a century has been estimated as 2.8 per cent per annum for small cap and more than double that figure for micro cap.

Emerging market equities have also enhanced the returns of the broader MSCI World All Countries Index (which had 10.9 per cent exposure to emerging markets at the end of April 2010), outperforming by 0.6 percentage points per annum over the last 10 years compared with the narrower MSCI World Index.

The Japanese equity market bubble illustrates the potential impact of imposing benchmark relative risk controls.

The market constituted a bloated 44 per cent of the world index at its peak in 1989, even though Japan only represented 17 per cent of global gross domestic product.

Benchmark relative constraints forced managers to hold overvalued Japanese shares, whereas a skilful unconstrained manager holding little or nothing in Japan would have benefitted to the tune of more than 2.5 percentage points of performance each year for nearly 15 years.

Unconstrained investment does not imply a disregard for risk, but instead recognises that constraints are a very costly way of managing fund volatility.

Risk for most investors is about absolute risk of loss rather than performance relative to a benchmark.

Therefore, risk should be managed more strategically at the overall fund level on an absolute basis rather than micro managed by over-constraining individual components of the portfolio to some benchmark.

Why non-market cap weighting?

For the reasons outlined earlier, cap weighting results in a long-term drag on performance: big stocks have historically underperformed and cap weighting results in a portfolio that is overexposed to expensive stocks and underexposed to cheap stocks.

Even a very simple equal weighting strategy — the most naive of all weighting schemes — would have handsomely outperformed cap weighting over nearly all time periods for most markets of the world.

Figure 2 shows how an equally weighted version of the MSCI World has outperformed its cap weighted cousin by 3 per cent per annum over time.

Why value investing?

There has been a wealth of academic and practitioner research over the past few decades that support the superior returns to value as an investment strategy.

Our own research suggests value investing has been very effective globally as well as across different regions and sectors, as shown in figure 3.

Stephen Kwa is an investment specialist at Schroders.