PrefBlog

Performance of the HIMI Indices for May was:

The Claymore Preferred ETF may be viewed, with caution, as a proxy for the S&P/TSX Preferred Share Index. Caution is required due to tracking error – the ETF will deliver performance according to what it actually holds and how well it is able to do things like reinvest its dividends, which is not the same thing as an index return. I suspect that tracking error this month will be larger than it will be in the future, since it has issued a great many units. Additionally, the fund’s NAV will be reported after MER of 0.45% p.a.

Be that as it may, the NAV on 5/31 was $19.44; the NAV on 4/30 was $19.91. Therefore, NAV Performance for May 2007 for CPD was -2.36%, net of MER (which will be just under 0.04% monthly).

Diversified Preferred Share Trust, DPS.UN is the main competitor of CPD. It doesn’t publish month-end NAVs, but the May 30 NAV was $22.55, while the May 2 NAV was $23.11, so its performance for this four-week period was -2.42%, net of its MER of about 0.04% for the period. The corresponding figures for CPD are $19.48, $19.90, -2.11%. Ouch! I presume that DPS.UN underperformed due to its greater weight in BCE issues, but that’s merely speculation.

It should also be noted that the HIMI Indices are prepared using the closing bids, which can be a very different thing from the closing price. When averaged over a lot of issues the difference should be minimal, but you can’t tell until you rip apart the data.

Caution should also be used in interpreting the differences between the various HIMI Indices. I will suggest that the lousy performance in Ratchet and FixFloat have a lot more to do with nervousness about BCE’s credit than the intrinsic performance of those investment classes – both of those subindices held entirely BCE for the period.

One very interesting thing that happened this month is that a lot of the yieldCurvePremiumLiquidity disappeared, as shown in this graph. I interpret the change in the premium as reflecting a desire by some holders, at least, to get out of the sector in size and quickly; such holders might simply sell their most liquid holdings to adjust portfolio exposures; this will affect the prices of these issues; hence, liquidity will become a lot less expensive. The PerpetualDiscount index is the most liquid of all the sub-indices – it’s dominated by recent issues, apart from anything else – and thus a portion of the decline in this index might be attributed to this factor rather than the intrinsic characteristics of the investment.

Such a hypothesis gains some support from examination of the changes in the yield curve, which I found a little surprising. The long-end hasn’t moved by nearly as much as one might have expected. Note that this graph is of the TAXABLE curve and refers to SPOT YIELDS … therefore, the x-axis shows the yield one might expect on a “stripped dividend”, after tax.

Attribution analysis is tricky and very implementation dependent. I might return to this topic on the weekend.

This entry was posted on Friday, June 1st, 2007 at 7:16 pm and is filed under Index Construction / Reporting. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.