This research compares the performance and viability of two prevalent alternative indices, equal and fundamental, with a traditional market capitalisation weighted index (MCWI). The paper assesses the viability of the three strategies in terms of both investment capacity and trading requirements. We find that as fund size and, consequently, transactions costs increase the difference in returns between alternative and traditional indices decreases to the point where no significant outperformance exists. We also consider alternative index implementation and find that alternatives are not viable for large funds, since execution shortfalls induce tracking error. We conclude that the traditional, market capitalisation weight index will remain popular for its simplicity, vast investment capacity and low inherent implementation costs.
The reliance of traditional indexing on market price bestows both theoretical and practical benefits. In an efficient market, prices reflect all information and, hence, a market capitalisation index (which relies on price to find a firm’s total market capitalisation and, therefore, sets its MCWI weight) is theoretically optimal. Furthermore, by investing in proportion to a stock’s market capitalisation, portfolio rebalancing occurs automatically as prices change, leading to very low trading costs when replicating the index. However, various theoretical mechanisms exist which may subvert efficient pricing and hence traditional MCWI indexation.
Recently, alternatives to traditional market capitalisation weighted indices have been proposed. These alternative indices do not weight their constituent stocks by market capitalisation but instead use various alternative, theoretically motivated, weighting procedures to do so. Here, we focus on two alternative indices: the first, equal weight indexing (EWI) weights each stock in an index equally; and, the second, fundamental indexing (FI) weights each stock using a measure of ‘economic size’.
Both performance and viability are essential for a successful alternative index. Without a performance advantage over standard MCWI there is no reason to use alternative indexation. Also, overcoming any implementation issues is critical in ascertaining whether any outperformance is practically viable. If simulations of alternative indexes report outperformance but do not properly account for all transaction costs or ignore implementation issues, then any reported outperformance may be neither genuine nor achievable.
We assess the performance and viability of FI, EWI and MCWI, after accounting for all transaction costs using different rebalancing frequencies, trade size and firm size.
Although prior studies find that alternatives outperform traditional indices, they omit important transaction costs known as market impact costs, which mean that any reported outperformance may be neither genuine nor viable.
Market impact is the cost associated with adversely moving the stock price as a result of trading in the stock. The cost of market impact increases nonlinearly as an index’s assets under management, small firm ‘tilt’, rebalancing frequency and, consequently, trading increase. It is not directly observable and thus difficult to estimate accurately, even though it makes up the greatest proportion of total transaction costs when rebalancing an alternative index.
Market impact costs are particularly severe for alternative indexes as they are not anchored to price and therefore sacrifice the automatic, price dependent, rebalancing of MCWI. This results in the need to rebalance the many small stocks in the index replicating portfolio frequency so as to accurately track the index. This generates portfolio turnover, especially in small stocks, which increases transaction costs in general, and market impact costs in particular, ultimately eroding performance.
Rather than assume no, or constant, market impact costs we implement a simple but more realistic, nonlinear, market impact cost model to assess the effects of costs on alternative index performance.
Approximate breakeven transaction costs are obtained for each strategy which give an insight into the required transaction costs for the superior return performance of the EWI and FI strategies to be fully eroded.
This is the first time that percentage holding and volumes required have been analysed for these strategies.
As expected, given the tilts of alternatives to size and value risk factors, the EWI and FI strategies generate higher returns compared to MCWI. However, as the fund size and consequently transaction costs increase, the outperformance is reduced to the point where it is no longer statistically significant. In effect, the additional transaction costs associated with rebalancing the alternatives completely erodes the higher returns realised from their increased size and value tilts.
We conclude that, particularly when performance evaluating the performance of alternatives, we should not ignore their implementation. We find that EWI strategies are a viable investment technique as long as the size of the fund remains relatively small.
Investment in an FI portfolio has investment capacity and is a viable option at current, moderate, investment levels. However, the results suggest that the MCWI strategy will remain popular due to its simplicity, vast investment capacity and low costs.
This Working Paper was produced by the CSIRO-Monash Superannuation Research Cluster a collaboration between the CSIRO and Monash University, the University of Western Australia, Griffith University and the University of Warwick in the United Kingdom. In addition, the Cluster engages on an ongoing basis with a range of industry supporters, government agencies and industry peak bodies who assist in providing guidance and feedback to researchers, providing data, and in disseminating outcomes. The purpose of the Super Research Cluster is to examine issues pertaining to the future of Australia’s superannuation and retirement systems.