PrefBlog

Alan S. Blinder, the Gordon S. Rentschler Memorial Professor of Economics and Public Policy at Princeton University and former Vice-Chairman of the Fed’s Board of Governors, gave a wonderfully informative and chatty speech at the Federal Reserve Bank of Boston conference at Chatham, Massachussets, on October 23, 2009 titled
It’s Broke, Let’s Fix It: Rethinking Financial Regulation.

One quote I simply must highlight is:

After all, regulatory failure on a grand scale was one major cause of the mess.

However, at the moment I am more interested in his thoughts on Contingent Capital than anything else:

I myself am attracted to a particular idea for “contingent capital” suggested recently by the Squam Lake Working Group on Financial Regulation, an ad hoc panel of academic experts. Under the proposal, regulators would have the power, by declaring a systemic crisis, Their idea, which derives from Mark Flannery’s (2005) clever earlier proposal for “reverse convertible debentures,” is to require certain banks to issue a novel type of convertible bond. Conventional convertible debt gets exchanged for equity at the option of the bondholder; and because this option has value, convertible debt bears lower interest rates than ordinary debt. The proposed new form of convertible debt would reverse the optionality by giving it to the regulators instead.

Under the proposal, regulators would have the power, by declaring a systemic crisis, to force holders of these special convertibles (but not holders of other debt instruments) to convert to equity. As in many cases, one key question is price—specifically, how large an interest rate premium would investors demand to cover the risk that their bonds could be converted into equity against their will? If this premium proved to be very large, these new convertibles would be a very expensive form of “capital” that banks might shun, preferring ordinary equity instead. Only experience will tell. —thus giving banks more equity capital (and less debt) just when they need it most. Naturally, the existence of such an option would detract from the value of the bond and therefore would make the interest rate on reverse convertibles higher than on ordinary debt. Indeed, if the requirement was limited to TBTF institutions, as seems appropriate, that higher interest rate on a fraction of their debt would constitute a natural penalty cost for being TBTF. Furthermore, the spread on this new type of debt over regular debt could become a useful market indicator of the likelihood of a systemic crisis.

As in many cases, one key question is price—specifically, how large an interest rate premium would investors demand to cover the risk that their bonds could be converted into equity against their will? If this premium proved to be very large, these new convertibles would be a very expensive form of “capital” that banks might shun, preferring ordinary equity instead. Only experience will tell.

I have previously discussed the Squam Lake proposals, as well as the original Flannery paper; I think they need a little work. My major objections are that:

• It mixes book value with market value; theoretically suspect and leading in times of stress to unpredictable – probably procyclical – results, and
• by incorporating regulatory discretion into the conversion trigger, it unnecessarily introduces regulatory uncertainty into the evaluation of the investment, as well as encouraging regulatory capture and even corruption.

He breaks my heart by advocating credit ratings by government agencies:

But many observers think the fundamental problem lies deeper: with the issuer-pays model. As long as rating agencies are for-profit companies, paid by the issuers of the securities they rate, the agencies will have a natural tendency to try to please their customers—just as any business does. Unfortunately, the most obvious alternative, switching to an investors-pay model, is probably infeasible except in markets with very few investors. Otherwise, information flows too readily, and everyone wants to free ride. What to do? The way out of this dilemma, it seems to me, is to arrange for some sort of third-party payment. The government (e.g., the SEC) or an organized exchange or clearinghouse seem to be the natural alternative payers. In either case, they could raise the necessary funds by levying a user-fee on all issuers.

He also discusses the separation of prop trading from vanilla banking:

For example, the Group of Thirty (2009, p. 28)—hardly a bunch of wild-eyed radicals–recently concluded that, “Large, systemically important banking institutions should be restricted in undertaking proprietary activities that present particularly high risks…and large proprietary trading should be limited by strict capital and liquidity requirements.” That’s not quite a ban, but it’s getting close.

But there is a downside. Roping off “proprietary trading” from other, closely-related activities of banks is not as easy as it sounds. For example, banks buy and sell securities, foreign exchange, and other assets for their clients all the time. Often, such buying and selling is imperfectly synchronized or leaves banks with open positions for other reasons. Does that constitute “proprietary trading”? Furthermore, market-making has obvious synergies with dealing on behalf of clients. Do we want to label all such activities as “proprietary trading”? The point is: There is no bright line. That is why Adair Turner (2009), the chairman of Britain’s FSA, concluded that “we could not proceed by a binary legal distinction—banks can do this but not that—but had to focus on the scale of position-taking and the capital held against position-taking.”

I say – yes. we do want to label all such activities as “proprietary trading”; and the fact that there is no bright line is just something we’ll have to get used to. As previously urged on PrefBlog, I suggest that there be two regulatory regimes – for investment banks and vanilla banks, the former imposing relatively heavier capital charges on long term positions, the latter imposing relatively heavier charges on short term positions. It won’t be perfect, by any stretch of the imagination; but it will allow each type of institutions to make decisions on a tactical basis, according to their marginal value.

One of the things that brought down the investment banks was that they engaged in buy-and-hold strategies, which are more properly the province of vanilla banks, which have (or should have!) the expertise and controls in place to look beyond the next portfolio flip.

The papers of the Squam Lake Working Group , a very distinguished group of academics, are published by the Council on Foreign Relations (which seems rather strange, but there you go), who also publish the periodical Foreign Affairs, which I love but don’t have time to read any more.

The Squam Lake paper titled An Expedited Resolution Mechanism for Distressed Financial Firms: Regulatory Hybrid Securities is of great interest, albeit lamentably short on detail.

Most notably, they propose a double trigger for conversion:

A bank’s hybrid securities should convert from debt to equity only if two conditions are met. The first requirement is a declaration by regulators that the financial system is suffering from a systemic crisis. The second is a violation by the bank of covenants in the hybrid-security contract.

This double trigger is important for two reasons. First, debt is valuable in a bank’s capital structure because it provides an important disciplining force for management. The possibility that the hybrid security will conveniently morph from debt to equity whenever the bank suffers significant losses would undermine this productive discipline. If conversion is limited to only systemic crises, the hybrid security will provide the same benefit as debt in all but the most extreme periods.

Second, the bank-specific component of the trigger is also important. If conversion were triggered solely by the declaration of a systemic crisis, regulators would face enormous political pressure when deciding whether to make such a declaration. Replacing regulatory discretion with an objective criterion creates more problems because the aggregate data regulators might use for such a trigger are likely to be imprecise, subject to revisions, and measured with time lags. And, perhaps most important, if conversion depended on only a systemic trigger, even sound banks would be forced to convert in a crisis. This would dull the incentive for these banks to remain sound.

I don’t like the first trigger, the declaration by regulators that a systemic crisis exists. First, there is more than one regulator, which will lead, at the very least, to delays while simultaneous announcements are arranged and, at worst, to political kerfuffles regarding cross-border banks if there is no widespread agreement.

Secondly, it introduces an element of political uncertainty regarding conversion, which will lead to the political pressure they allude to in their discussion of the second trigger.

Thirdly, I just plain don’t trust the regulators.

It will be noted that the group skims rather lightly over the justification for the first trigger!

The group also suggests using Tier 1 Capital Ratios as a trigger:

What sort of covenant would make sense for the bank-specific trigger? One possibility, which we find appealing, would be based on the measures used to determine a bank’s capital adequacy, such as the ratio of Tier 1 capital to risk-adjusted assets.

I don’t like it, for reasons which have been discussed in my posts Contingent Capital: Reverse Convertible Debentures and Lloyds bank to Issue Contingent Capital with Tier 1 Ratio Trigger?. Tier 1 ratios are too easy for a bank and regulators to manipulate, do not measure the degree of investor confidence in an institution and do not provide a framework for market arbitrage. As a bank’s situation deteriorates, the price response of the hybrid should gradually become more-and-more equity-like, which suggests a market based approach rather than the binary now-it’s-debt-now-it’s-equity paradigm implied by an all-or-nothing conversion based on calculated figures. I have not seen anything yet to shake my belief that a fixed-rate conversion with a trigger based on the trading price of the common is the best solution.

The authors discuss the conversion rate:

In addition to the triggers, this new instrument will have to specify the rate at which the debt converts into equity. The conversion rate might depend, for example, on the market value of equity or on the market value of both equity and the hybrid security. Conversions based on market values, however, can create opportunities for manipulation. Bondholders might try to push the stock price down by shorting the stock, for example, so they would receive a larger slice of the equity in the conversion. Using the average stock price over a longer period, such as the past twenty days, to measure the value of equity makes this manipulation more difficult, but it opens the door for another manipulation. If the stock price falls precipitously during a systemic crisis, management might intentionally violate the trigger and force conversion at a stale price that now looks good to the stockholders. Finally, in some circumstances, a conversion ratio that depends on the stock price can lead to a “death spiral,” in which the dilution of the existing stockholders’ claims that would occur in a conversion lowers the stock price, which leads to more dilution, which lowers the price even further.

An alternative approach is to convert each dollar of debt into a fixed quantity of equity shares, rather than a fixed value of equity. There are at least two advantages of such an approach. First, because the number of shares to be issued in a conversion is fixed, death spirals are not a problem. Second, although management might consider triggering conversion (for example, by acquiring a large number of risky assets) to avoid a required interest or principal payment on the debt, this would not be optimal unless the stock price were so low that the shares to be issued were worth less than the bond payment. Thus, management would want to intentionally induce conversion only when the bank is struggling. The advantages and disadvantages of different conversion schemes are complicated, however, and will require both further study and detailed input from the financial and regulatory community.

It seems that they believe that a problem with the conversion at a fixed rate is the potential for manipulation of the trigger terms by management. This would be avoided with a market-based trigger.