RANKING THE ANALYSTS: Methodology based on stringent confidential processes
The industry’s biggest players were used to conduct survey
The rankings and other results presented in this survey are drawn entirely from a confidential questionnaire process conducted with the domestic institutional clients of SA stockbroker firms.
The structure of the survey is developed by the Financial Mail in conjunction with research house Intellidex. The survey is conducted through an online questionnaire process. The Intellidex team collects the results and analyses them to determine the rankings and ratings.
The intention of the survey is to capture each institution’s house view of the brokerage services it obtains, rather than the views of its individual fund managers. However, the online questionnaire makes it possible for multiple individuals at an institution to complete different sections. This allows for specialists to contribute their insights in their relevant sectors.
Before compiling the questionnaire, Intellidex asked all institutional brokers to submit a list of their analysts and the sectors they covered in the past year. This year 34 firms submitted lists, three more than last year. Those names and sectors were then placed into lists on the questionnaire.
In the case of firms that did not submit nominee analysts, institutional investors were encouraged to insert the names of analysts into the questionnaires and the resulting rankings in this survey reflect the firms’ performance on that basis.
Questionnaires were completed by 41 institutional respondents before the deadline. Respondents included asset managers, life insurance companies, some wealth managers and hedge funds. We attempted to include all of the largest fund managers by assets. Based on information submitted, we calculate domestic assets under management by participating firms to be R4.82 trillion. These figures are not directly comparable to last year’s figure as fewer firms participated this year. Only 21 firms provided data on brokerage and they spent a total of R663m.
For each research sector, respondents were asked to rank the five best firms and, separately, the five best analysts, using the following criteria: quality of written fundamental research in terms of content and value; accuracy of forecasts and buy/sell recommendations; useful ideas; effective communication, both written and verbal; and consistency over the past 12 months.
Points were then awarded to each analyst according to the rankings, ranging from five points for first place to one point for fifth. The analysts’ rankings are derived from the votes specifically for individuals, while the firms’ rankings are derived from the votes for firms. Votes for individual analysts were not transferred to the firms.
In the sections covering institutional sales teams, corporate access, dealing and back-office administration, the respondents were asked to nominate and rank the firms they considered to be the best 10. There are no rankings for individual traders or sales specialists.
Results were weighted in the first instance by commission paid to the sell side. We also included weightings by domestic assets and unweighted results. This ensures the survey correctly reflects achievement in the industry according to its own business objectives.
The commission weightings were determined from two sources. We asked brokers to place their clients into different brokerage "buckets". We also asked the institutions to tell us how much they paid in brokerage — broken down into equity, fixed interest and derivatives or other, and split into research and execution. On both the sell and buy sides, some firms would not disclose this information, but most did.
The brokers’ buckets were added together to get an indication of the total paid by each institution. This was compared to the institution’s disclosure, where we had it. Where we had no information, assets under management and our own market insights were applied to give a subjective weighting. All institutional respondents were categorised into 10 bands, which were used to weight their responses. A similar process was used for the AUM (assets under management) weightings.
All of the sectors, equity dealing, sales teams, corporate access and administration tables are presented without further amendment. However, in the case of the derivatives and fixed-interest securities dealing tables, an additional weighting was applied to firms that indicated they make extensive use of derivatives and fixed interest research. The assets under management and commission weightings were adjusted to give greater weight to those firms that indicated that derivatives and fixed interest trading was a greater priority for them, both in the quantity of assets they hold in those asset classes, and in their indication of value derived from their brokers. This attempts to cater for the distortion that arises from the fact that most brokerage and assets are connected to equity instruments.
In the case of the overall rankings for research, an additional layer of weightings was applied. Equity sectors were weighted according to their market capitalisation as at December 31 2017, using the capped Swix index constituent weightings. The rankings with this weighting applied are presented in the column, "Equity sectors weighted by market cap".
Nonequity sectors, such as quantitative analysis and risk management, were weighted according to the institutions’ feedback on how important these nonequity sectors were to them. Investment strategy and both international and domestic economics carried the largest weightings of the nonequity sectors. We then gave the equity sectors a 70% total weight and the nonequity sectors a 30% total weight to determine the results presented in the "All sectors weighted" column.
We will continue to develop the methodology in future surveys.
Read the full report here.