Financial Webring Forum brings to my attention a Globe & Mail article titled What happens to rate reset prefs in Basel III?:
John Nagel at Desjardins Securities has been watching the issue closely, trying to get clarity from the Office of the Superintendent of Financial Institutions. He has a vested interest in the outcome because the Desjardins team invented the structure.
At the moment nothing has been decided, but Mr. Nagel said the last he heard, a contingent capital clause was being considered for all new rate reset issues. As a reminder, contingent capital simply means a security type that will convert to common equity when things get rocky.
As far as he knows, outstanding rate reset issues will be grandfathered under Basel III and will count as Tier 1 capital and equity. Going forward, though, Mr. Nagel thinks prospectuses for these issues could have a section, possibly called the Automatic Exchange Event, that describes how preferred shares are exchanged into common equity.
However, this type of “trigger event” would only happen if OSFI declares the financial institution “non-viable” and Mr. Nagel suspects it’s unlikely that will happen in Canada.
“If a bank or an institution was in trouble, long before it would be declared non-viable they would halt trading and OSFI would say ‘Fine, you’re merging with BMO or RBC,” he said. If a merger occurred, the distressed institution’s preferred shares would then become obligations of the acquirer.
No moral hazard here, no way, not in Canada!
The critical “point of non-viability” at which Mr. Nagel believes conversion will be triggered is in accord with Dickson’s speech in May, most recently referenced in PrefBlog in the post A Structural Model of Contingent Bank Capital. The recent BIS proposals insist on some conversion point, setting the point of non-viability as the floor limit, as discussed in BIS Proposes CoCos: Regulatory Trigger, Infinite Dilution.
As I have discussed, many a time and oft, I think that’s a crazy place to have the conversion trigger. It may help somewhat in paying for a crisis, but it will do nothing to prevent a crisis. S&P agrees.
In order to prevent a crisis, the conversion trigger has to be set much further from the point of bankruptcy; the McDonald CoCos are an academic treatment of a model I have advocated for some time: there is automatic conversion if the common price falls below a pre-set trigger price; the conversion is from par value of the preferreds into common at that pre-set price. I suggest that a sensible place to start thinking about setting the trigger price is one-half the common equity price at the time of issue of the preferreds.
Tory’s published a piece by Blair W Keefe in May, titled Canada Pushes Embedded Contingent Capital:
A number of concerns arise with the use of embedded contingent capital.
First, it is likely that the conversion itself could cause a “run” on the troubled bank: effectively, the conversion means that the bank is on the eve of insolvency and the conversion does not create any additional capital; it merely improves the quality of the capital. As a practical matter, it will likely be essential for the government to immediately provide funding to the bank; however, with the former holders of subordinated debt and preferred shares being converted into holders of common shares, the government could replenish the subordinated debt rather than being required to replenish the Tier 1 capital, which occurred in the financial crisis. Therefore, it should be less likely that the government would suffer a financial loss.
This echoes my point about prevention vs. cure.
Third, the cost of capital could increase significantly for banks, particularly if the new capital instruments are viewed as equity – given their conversions in times of financial difficulty to common share equity – rather than debt instruments. OSFI is sensitive to this concern and is the reason why OSFI is advocating a trigger that occurs on the eve of insolvency (rather than earlier in the process) when the holders of subordinated debt and preferred shares would anticipate incurring losses in any event.
In other words, OSFI thinks you can get something for nothing. Ain’t gonna happen. Either we’ll raise the cost of capital for the banks, or we’ll do this pretend-regulation thing for free and then find out it doesn’t work. One or the other.
Seventh, if the embedded contingent capital proposals are adopted, how will those requirements need to be reflected in the Basel III capital proposals? Similarly, what treatment will rating agencies give to contingent capital? If the triggering event is considered remote, rating agencies may not give “equity” credit treatment for the instruments.
Finally, with any change of this nature, market participants worry about the unexpected consequences: Will hedge funds or other market participants be able to “game” the system? Will the conversion features create more instability for a bank experiencing some financial difficulty? Could the conversion create a death spiral of dilution? and so on.
Ms. Dickson’s beloved “Market Price Conversion” formula will almost definitely create a death spiral. While fixed-price conversion may create multiple equilibria (which the Fed worries about), I see that as being the lesser of a host of evils. Gaming can be reduced if the conversion trigger is based on a long enough period of time: my original and current suggestion is VWAP measured over 20 consecutive trading days. It would be very expensive to game that to any significant extent, and not very profitable. On the other had, if the conversion trigger is a single share trading below the conversion price … yes, that presents more of a problem.