As one might expect, the process of changing portfolio benchmarks can be particularly challenging, requiring extensive planning and consideration in terms of execution. Perhaps the most important issue here is whether an investor actually needs to change benchmark index. The current environment is one of significant change because of the implementation of free float in both the FTSE and MSCI indices. Both events appear to involve considerable cost to the end user from a transactional perspective given the relatively high levels of turnover within the indices, with MSCI-based investors suffering higher levels of turnover than FTSE investors in the free float shift.
The final implementation of free float by all the major index providers, with Stoxx having moved to free float in September 2000, means that from a European perspective at a broad level there is little to differentiate the benchmark indices. In fact, given that the MSCI index move also incorporates the expansion of the universe from the current 60% target coverage to 85% the similarities are even greater.
The main issue for investors when it comes to benchmark transition should not be the turnover and transaction costs. This is because for the most part investors have available to them a wide range of programme trading techniques designed to facilitate large-scale portfolio transactions at low cost in terms of both direct and indirect costs. The most important indirect cost during benchmark transition is market impact resulting from trade execution.
The issue of turnover from one benchmark to another can be a misleading statistic in that generally such trades can be implemented for low single-digit charges in terms of costs in basis points. This applies to even relatively large portfolios. The largest cost relates to the opportunity cost in performance terms if the benchmark transition is not just a customer event but a broader event for all benchmark users. A good example of this would be the Stoxx/Euro Stoxx rebalance during the move to free float in September 2000. One can argue that the real cost to the user was not the trading cost but more importantly the market impact due to the flows as a vast range of investors had to implement to the shift.
In the Stoxx rebalance, turnover was high at over 28% in the Euro Stoxx 50 and 25% in the Stoxx 50. However, the trading part of the cost of the free float adjustment was extremely low, especially for clients that chose to execute on either an agency or risk basis. The implicit cost was more accurately reflected in the performance of the new index versus the old. At a Euro Stoxx 50 level the free float adjusted index outperformed the old fully market capitalisation weighted index by over 3.25%. When one just considers the main stocks with the largest increase in weight relative to those with the largest decrease in weight due to free float the relative performance was over 20%. Because the Stoxx index change had a short implementation period, with there being only two and a half months from announcement to index change the market impact was noticeable.
This scale of market impact from free float adjustment has been instrumental in other benchmark providers planning a more orderly transition process. The FTSE indices shift in the middle of last month was the outcome of a process that has been determined since August 1999. In the event, there was no really significant volume as the indices rebalanced. We believe that such a long period allowed many funds to manage the transition efficiently from both a trading and implementation perspective. So the degree of market impact that characterised the Stoxx rebalance was avoided. In the main FTSE benchmark indices turnover is also lower than in either the Stoxx or forthcoming MSCI change, with turnover in Europe of only 15.1%.
The MSCI transition to free float, although officially scheduled as being a two-stage process, with the first phase at the end of November 2001 and the final adjustment at the end of May 2002, is likely to be a major event. Due to the desire of investors to avoid the main transition cost of market impact we expect many funds to wish to trade to the new provisional index that is already available before the main specified index change dates. The availability of full benchmark data for the new index means that investors can manage the change procedure efficiently and with a degree of timing that suits their needs. This avoids investors being tied into a given index change event which can have a high level of market impact.
As mentioned previously there are now relatively few reasons for investors to wish to change portfolio benchmarks. The main benchmark shift still remains for those investors that have yet to diversify from their domestic equity market to the broader pan-European or Euroland based benchmarks. For investors considering Europe there is the interesting issue of the fact that MSCI and FTSE create European benchmark indices by amalgamating the individual country indices, whereas Stoxx considers Europe to be a single region and effectively amalgamates pan-European sectors to form the benchmark. Investors in the Stoxx indices therefore have more limited scope for country asset allocation but a broader choice for sector strategies, together with list sector futures products.
Perhaps the main role of benchmark transition comes from investors needing to shift from a European mandate to a global framework. At this level again transaction issues in terms of implementation and execution costs should not be significant for investors who access the liquidity of global programme trading firms. The highest costs that investors face are invariably the costs of the benchmark itself and the process of changing or managing internal risk management and reporting systems. At the moment, given the relatively low usage of the Dow Jones Global indices there is still a significant shift from Stoxx to either MSCI or FTSE as investors pursue non-European investment strategies. The main cost though is not related to trading and transaction costs but index and benchmark fees.
Ultimately the fact that the majority of indices are now effectively constructed on broadly similar lines in terms of general levels of coverage and methodology removes the main rational for changing benchmarks. Investors should be more concerned with the on going turnover and management of changes within the indices. In the past few years it has been the trend for sizeable cross-border transactions that has increased benchmark turnover. As indices settle down in a post-free float world this may be a more important factor and here investors will benefit from access to global programme trading houses that can manage not just regional flows, but global trade flows. Ultimately for investors looking to manage the cost of benchmark turnover and rebalancing costs access to liquidity will be the most important factor and programme trading firms with access to global client and exchange flows are best placed to minimise costs to investors.
Nizam Hamid is director derivatives and portfolio trading research at Deutsche Bank in London