An old thread regarding preferreds has come to life on Financial Webring Forum, with one poster speculating that the Swoon In June is tax-driven – there are two tax changes scheduled to have an effect on preferreds in the next few years, Federal Bad and Ontario Good, netting out to a modest negative.
It is my feeling that the recent decline is driven more by portfolio window-dressing by retail stockbrokers more than any fundamental factor.
If fundamental factors were to blame – or even if they were fundamental factors mis-applied! – I would expect to see that the market would retain some degree of internal consistency.
This is not what’s happening. Deeply-discounted perpetuals are being hit harder than slightly-discounted perpetuals (the discount of market price relative to potential call price, that is) – I gave the example of the CM issues on June 26 and two RY issues more recently.
This has happened to the extent that deeply discounted perps yield more than slightly-discounted perps. This simply should not happen (see the links in the June 26 post) and, I will emphasize, is not due to any historic spread relations or anything like that … the causation is the other way round. Convexity is a pretty basic fixed-income analytical tool, with a negligible effect for normal bonds, but a huge factor in callables (or other type of embedded option).
You can, if you like, make an argument that convexity has zero value and that therefore any perpetual discount from a given issuer should trade at the same yield, regardless of its combination of price and dividend. I might laugh at your pathetic little arguments, but you can make them with some semblance of rationality. But it is not possible to argue that convexity has a negative value and that therefore a discounted perpetual with a relatively high coupon should trade at a lower yield than its lower-coupon, otherwise identical, sibling. There might possibly be other things going on that would lead to that effect – but it wouldn’t be a direct effect and I haven’t been able to come up with any actual evidence that such an effect is even remotely possible.
The fact that it has happened anyway has convinced me that it’s retail window-dressing. We are approaching quarter-end and stockbrokers do not want to remind their clients, yet again, that the new issue they bought at $25 is now trading at $20. So, of the two or three issues in the client’s portfolio, they sell the lowest priced one, regardless of yield.
It’s only a hypothesis and it’s completely untestable … but I have not been able to think of anything else!
The fact that the relative prices are out of whack – and not based on any kind of normal fixed-income analysis – causes me to be very suspicious of the fundamental underpinnings of the overall decline.
If we plug in the projected Ontario tax rate on dividends for 2012 of 26.7% and the projected rate on income of 46.41%, we arrive at an equivalency factor of (1-0.267)/(1-0.4641) = 1.37. This is marginally lower than current, but PerpetualDiscounts now yield 6.04%, while long corporates are about 6.1%. Even the lower equivalency factor results in a spread of about 215bp, well in excess of historical norms and very hard to justify in fundamental terms (although, it should be noted, you can always justify anything you like in fundamental terms).
When the market re-establishes some degree of internal consistency (as far as the pref market ever is internally consistent!) we can have a look and see what the spreads really are. Until then, the situation is too clouded by clear signs of panic to allow any conclusions to be drawn.