As the name – passive management- indicates, people tend to think it’s a super easy job to manage an index fund. Afterall, once the index is created, the manager only needs to buy all the securities in the index or, more particular under the circumstance, the benchmark, and assign the weights as defined in the benchmark too.
In reality, it’s not that easy and passive. It is even claimed as a part art, part science job.
With respect to replicate the constituents at the same weight in the benchmark, there are four strategies: full replication, stratified sampling(linear optimization), sampled optimization(quadratic optimization), and blended approaches.
First, full replication. There are three major issues to tackle in full replication – trading cost, corporate actions, and dividend reinvestment.
The index performance is calculated as if there is no transaction cost, while in reality, trading cost is unavoidable for fund managers.
When corporate actions take place, the indexes assume that the investor receives the cash value by using either an intrinsic price or the price of the security on the first day it trades. In the case of non-U.S. securities, however, it may take days or even weeks before the settlement of the new security allows the stock to be sold. Additionally, intrinsic prices (a security’s theoretical value if the parts equal the whole) often are not equal to actual value once the securities trade.
Dividends are paid with different lag times across the holdings of an index, to calculate index value timely, indexers apply “smoothing” method where one-twelfth per month, theoretical dividend without any cash drags are assumed. Not to mention there is tax rate implemented in dividend income, which varies in different countries.
Second, Sampling instead of full replication. The concept of sampling is straightforward: bucket holdings based on size, industry, geography etc. and then take samples from each of them to form the final index list. In actual operation, numerous problems such as which sectors to apply, number of stratums, weight given… have to be gauged and decided.
Third, Optimizer. It uses computer models that measure the historical interrelationship of several risk factors to glean how the mix of these factors impacts security movements.
Lastly, blending approach. An attempt to attain the most optimized strategy by blending the above three strategies.
With respect to the trading strategies, there are lot of intricacies. for instance, as they constantly buy and sell slices of their portfolio, using the cash flow to smooth the implementation of minor portfolio rebalances, PMs possess a high skill in cash flow management. On the other hand, PMs are also involved in managing securities lending together with another dedicated team in asset management firms to bring a steady alpha in.
The following diagram depicts a holistic view how a PM manages an index fund.
Overall, one can see that managing an index fund or ETF is not as easy as people usually would think. The tracking error is prevalently used as a metric to assess whether the portfolio manager does a good job or not.