Now that an eligible stock universe has been built based on keyword matching, an important quality control step takes place to make sure only the companies that matter make it into the thematic index.
Consider the case of Long Island Ice Tea Corp, and its now infamous rebranding to Long Blockchain Corp in 2017. After the name change, its share price soared overnight as speculative traders – and most likely algorithmic trading programs – piled in to capitalise on the crypto and blockchain trend. Over the following months, the share price came crashing down and the company was ultimately hit by a subpoena from the SEC.
To prevent such misleading anomalies from appearing in our funds, the index methodology carefully takes relevance into account.
To achieve this, MSCI makes use of four-digit SIC codes*, a granular and standardised way to identify a company’s business activities and associated revenues. A company’s SIC code is selected if it’s directly mapped to a relevant business segment (i.e. keyword hit), and if it’s assigned to business segments of at least two different companies in the eligible universe.
*Standard Industrial Classification (SIC) codes are assigned by the US government to business establishments to identify their primary business. Find out more here: https://siccode.com/page/what-is-a-sic-code MSCI sources raw business segment data from its data vendors. The raw segment data shows company’s business segments, revenue for each segment and segment’s SIC Code. MSCI’s data team then follows MSCI’s standard data quality validation process including dual vendor validation for each company’s segment data.
To make the cut for index inclusion, a company must derive between 25 and 100% of its revenue from activities relevant to the theme in question – i.e. a relevance score between 0.25 and 1.
Making use of SIC codes not only controls for relevance, but can also help identify other relevant segments and associated revenue that may have been missed by the initial keyword mapping.
Working out the relevance score for an Asian mass transit operator
Controlling for false positives Revenues derived from selected SIC codes that haven’t been directly mapped to a relevant segment are considered with a discount factor applied, as revenues attributed to such SIC codes can be prone to overestimation.
Segment names
Selected SIC code?
Discount factor
Attributable revenue
Mass Transit System
4789
Media
4Source: MSCI, Lyxor International Asset Management, as at 29/05/2019. For illustrative purposes only.
