Issues are typically selected from new issues for inclusion in the HIMIPref™ analytical universe based on a number of factors:
- Will the issue be tradeable next year? Issue size is an important determinant for this question. For example, a $30-million split share issue will probably be tradeable only for a short period after issue, while a $300-million bank issue will probably be tradeable in size for as long as it exists. The former issue would probably not be included in the universe, absent special factors - there is not much point in performing computations and reporting them if there is doubt regarding the ability of portfolio managers to act on these reports.
- Is the issue of good credit quality, as determined by an external rating agency? HIMIPref™ does not currently perform any meaningful analysis on issues rated Pfd-4 or lower, so there is no point in adding such issues to the Universe - even assuming such shares are issued, which is rare at best. Pfd-3 issues are in something of a grey area: a portfolio optimized according to the HIMIPref™ recommended settings will hold some Pfd-3 issues, but not many of them. It is much more likely that Pfd-1s and Pfd-2s will be selected for inclusion.
- Does this issue have characteristics the Universe needs? There are a number of risk factors in preferred shares that are taken into account in the course of determining the yield curve. It is desirable that each of these risk factors have a sufficient number of data points for both the "present" and "not present" attributes that the statistical analysis required to quantify the effect of these factors has a valid sample. It is also desirable that each of the risk factors should be both investable and avoidable. From time to time, therefore, an issue having characteristics that facilitate such results will be included solely for this reason.
- Will this issue be considered for inclusion in the BMO NB 50 Index? This question is usually redundant, being addressed by the first two points as this index typically includes only large, high quality issues. It is, however, useful to consider the question separately as a check on the validity of the choice made.
As implied by the above statements, there are no hard and fast rules regarding selection for inclusion in the Universe - HIMI makes an informed judgment on the question of inclusion. Client requests are given every consideration.
Issues are sometimes added to the Universe on a back-dated basis; for example, FIG.PR.A was added when the managers of a group of split-share corporations proposed that other, extant issues that were in the Universe be merged into it. Backdated changes to the Universe are, of course, somewhat suspect from the viewpoint of back-testing validity, so such actions are not taken without considerable thought; however, it is not my intention that HIMIPref™ be constrained from a sensible course of action by mere dogma.