« Connecting the eurozone crisis to the Canadian economy | Main | Maybe the Bank of Canada won't raise interest rates on June 1 »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I think we are a little too confident and self-righteous about this. Are we really so sure we will never need a hand ourselves?

By all means, make any such tax dependent on risk and transparency. If a bank wants a low rate, it must submit to international audits of risk. But I think a transaction tax used to establish a fund for future bail-outs is a good idea. It could reduce speculative, high volume, risky trading, provide an incentive to submit to auditing, and create a bail-out fund at the same time.

A bank tax and regulations limiting leverage are not mutually exclusive.

Having insurance in place for when things go off the rails is considered prudent in many areas. I think that it's prudent in banking and that there should be a way to design a bank tax system that wouldn't create too many perverse incentives.

Wouldn't a bank tax create a moral hazard similiar to a government bailout? I can envision this insurance destabilizing rather than stabilizing the system. I think having an OFI style organization with broad powers to limit banks in each nation is far more effective.

I think Germany is pressing for this tax because it doesn't have the ability to "print money" and inflate its way out of a problem. You may say it can order around the ECB, but unlike Canada this ability is an open question for Germany.

ISTM that Canada has always assumed and continues to assume that if we have a banking problem, we'll have to pay for it ourselves. I can't think of any circumstance that would require the Government of Canada to ask a foreign country to provide us funds for a bank bailout; I believe that at worst the Bank of Canada would inflate its way out of the problem.

If Canada's proposed solution is convertible bonds to shore up a failing bank's equity, does this mean that Germany is averse to having shareholders pay for bank rescue? Are German bank equity levels really that weak?

I just don't see how a bank tax helps.

Suppose that, like me, you buy the story that banking and finance have failed miserably in its function of productively deploying capital in favour of running a casino and rent extraction racket. And suppose as well that they really have captured their regulators and the political process. In this circumstance, it seems very likely to me that the tax will just be passed on us plebs, resulting in a further decline in the amount of capital available for productive deployment while doing nothing to end the dysfunction in the system.

Now, if I was a German banker, and the politicians said to me: "Politics mandates we do something to slap you down. Take your pick; tougher regulation or a tax". What would I pick? I'm pretty sure I'd pick a tax, knowing that I'd just raise prices to compensate. I would NOT pick tougher regulation because that would eat into my implicit (or explicit) gov't guarantees and my over sized salary and bonuses.

Given the size of Deutsch Bank, UBS, BNP Paribas, etc... The cynic in me suspects the Merket et. al. are just doing what their banker/financier masters are telling them to do.

Does the revenue from the tax go into a single global fund, or does each country have its own fund?

If, as I suspect, it is the latter, there is no reason for Canada to oppose the tax. The reason for coordinated international action is so that no one country becomes a low-tax haven for banks.

You're juxtaposing two different articles and implying they're connected when they're not. The first says Chancellor Merkel wants a tax, the other that German *banks* oppose tighter regulation (well, duh!). Nowhere is it stated that Merkel opposes tighter regulation.

I have to judge by her action. And she isn't putting much public effort into drumming-up support for tighter regulation. She IS making lots of noise about the bank tax. Now, to be fair, it could just be that the media is only reporting the tax angle, but I haven't seen anything on any of the blogs about regulation either.

And that doesn't change my reasoning about why it seems to me to be, at best, useless.

Stephen, although I have been, and remain, an advocate of the bank tax, I think your post has highlighted one of its major weaknesses. The bank tax is not a regulatory tool, but a means to fund the 'insurance' the industry needs. The Lehman Brother's failure demonstrated the need for an international structure to rescue and/or liquidate in an orderly fashion distressed international financial institutions, and I think that is a goal that Canada should pursue, but not at the cost of weakening existing regulations.

I think you're right to point out that the Germans look like they're trying to get others to fund the bailout mechanisms for their own largest banks. The bank tax should, at best, supplement the strengthening of bank regulation, not replace it.

Of course, even with a global bank tax, what's to stop country A from saying "O.K., but we're going to bring in a special corporate tax rate for "Special Financial Institutions" (i.e., corporations that pay the bank tax) which, golly gee, is 1% less than the corporate tax rate for other corporations? Or better yet, what's to stop country A from providing for a tax credit for any bank tax paid (meaning that it reduces their general corporate tax liability, rendering the bank tax meaningless (banks would pay the same taxes they'd pay before the bank tax came into effect, it would just be papered differently). If Merkel thinks having a uniform international bank tax will prevent there from being low-tax havens for banks in those countries which can afford to forego the revenue, she's kidding herself.

What are the chances of and the effects of Canada being put on some type of global tax haven blacklist if it doesn't go along with this. Could Canada get kicked out of the OECD which tends be the international agency in charge determining what is "unfair" tax competition. Is there any advantage to Canada actually being a member of the OECD? If the US and Europe puts a tax on lets Canadian oil to retaliate for us not having a bank tax wouldn't that effectively only be raising taxes on American and European consumers.

Interestingly enough the entire Swiss political establishment is begging the parliament to pass a new treaty with the US to turn over information on wealthy Americans hiding money or else be put on a OECD blacklist of "unfair" countries primarily determined by the EU and the US. Yet in Canada the entire political establisment is against a bank tax even though in the near future not having one could very well be viewed by the OECD as "unfair" tax competition.

I don't see how the Swiss situation bears any resemblance. Swiss banks were caught running schemes to help customers evade US income tax. The US suffered a real and material loss in that case.

Canada on the other hand is simply arguing that since our banks didn't fail, didn't need bailouts, and we have always regulated as if we are on our on IF one of the Big 5 fails, the tax is superfluous.

Apples and Oranges.

Tim, we're not talking about Canada changing anything. Up until recently, bankers were complaining that Canada was a hopeless backwater. Now suddenly we're the next Caymans? C'mon.

If politicians like Merkel want to setup an international slush fund to periodically hand over to bankers when they make stupid mistakes, that's there problem. I'd prefer to have nothing to do with it. All it's going to do is provide and incentive for big banks to take big risks. The contingent convertible bonds thingy that Canada is proposing is a much better idea.

Anyway, if Canada has a comparative advantage in bank regulating, then how about the OSFI offer it's services (for a fee) to the other OECD members? Might be a way to get rid of our budget deficit.

"A planned cap on bank leverage would not make the sector safer, said a German banking lobby on Friday, adding heavyweight support to a growing campaign."


Kostas (and others): I don't have any problem with the ideas of regulatory changes, a resolution authority (funded somehow) or with international co-operation. The problem is the idea of a global fund; it's like asking the three little pigs to pay equal shares for wolf insurance.



Don't our banks obey a leverage ratio of 20:1 and are still universal? Seems to me that the only thing that can truly guarantee solvency is cold, hard capital in the form of equity. "Riskiness" or the lack thereof implies an ability to foresee the future; as the Credit Crisis has shown, that ability is lacking in many major institutions.

Yup. And German banks are particularly odious, with assets many multiples greater than German GDP, no credible deposit insurance, and a lack of a clear commitment on the part of the ECB to provide adequate bailouts. We are seeing capital flight out of euro-denominated deposits as a result.

That last puts it really well, Stephen. Everyone will build straw houses.

Stephen, you're right, there's no point to a globally funded resolution authority unless every country's bank regulations are solidly built. There is a need for a global resolution authority, but right now it appears that Flaherty's tactic of focussing on regulations is the better choice.

But your three little piggies analogy is particulary apt ... in the end the third pig was forced to rescue the first two, and if another megabank implodes Canada will be forced to pay a price whether there's a global bank tax or not.

I have to agree with tyronen. The fact that German banks (like all banks everywhere) are opposed to the notion of a cap on their leverage is a dog bites man story that could be placed in a giant file of corporations of all sorts opposing regulations of all kinds since the day the first corporation drew (legal fictional) breath, and doesn't have anything to do with the merits (or lack thereof) of the bank tax.

Somehow (because of the distance involved?) the German government and the German banks get merged in your post into a single entity of 'Europeans' that is pushing a global bank tax yet has no interest in preventing bank bailouts from being needed. Never mind that you could level the same bizarre accusation of hypocrisy against any jurisdiction that has ever pushed for a regulation that its corporations opposed and never mind the long record of Europeans pushing for stronger global regulation of banking - the German banks oppose regulation so everything the European/German governments have said is just a ruse. OK then.

Is the point of the "global" nature of the tax to prevent domicile shopping, or to create a global fund? It seems you can support the first goal without the second.

RSJ: keep in mind that this is Canada we are talking about. Firstly, it's just about impossible to setup shop in Canada due to foreign competition limitations and other limits on bank activities. But even supposing a foreign bank can get over that, why would they be at an advantage? They'd be trading no tax for a stronger regulator and a political reality where they hold comparatively little influence.


I'm not sure the OECD could "kick" Canada out, and in any event, the OECD has no real authority so what if it could? And in any event, as I noted in my earlier post, Canada could impose a bank tax, then fiddle with its other taxes to undue the effect of the bank tax (for example, by providing a credit against income tax for bank tax paid - our tax system does that with a number of taxes).

It is not as impossible as one might think any more for foreign banks to do business in Canada. First the securities side of the business is under provincial control and hasn't had any ownership restrictions for almost a generation(1985?) and thus almost everyone major bank in the world owns a Canadian securities dealer i.e. Goldman Sachs Canada inc, Deutsche Bank Securities Canada, Nomura Securities Canada all of which operate on the exact sames basis as domesticaly owned dealers.

In terms of bank charters many foreign banks have Canadian Chartered Bank subsidiaries regulated by OSFI i.e. Citibank Canada, Societe General Canada, UBS Bank Canada(there is even a Wal-Mart Bank Canada in the process of starting up). Conceivably there are some limations in the size and market share of foreign banks in Canada but no one has ever enforced them. I guess the questions is whether foreign banks would want to change the corporate structure to make there parent company/bank Canadian. The current government has strongly encouraged at least non bank foreign companies to change there corporate charter to Canada. Outside of Tim Horton's which Jim Flaherty was wildly excited about no other company that has made that move yet and with added regulation of banking I kind of doubt any banks will.

Funny, too, because Tim Hortons HQ consisted of a filing cabinet in a lawyer's office in Connecticut. I'm not aware of any high-profile companies substantially moving operations to Canada to take advantage of current and expected tax advantage.

This is in response to Angelo's comment in the other thread, about how he was surprised that Canada was cautioning about moral hazard in the creation of a resolution authority which would have the power to immediately wipe out equity holders. My response would be that moral hazard is not limited to the equity holders. If bondholders are going to be backstopped by a bailout fund, that would enable banks to borrow at below market rates (as many were able to through implicit government guarantees). The way to prevent, or at least ameliorate this moral hazard is to make sure that bondholders will feel pain due to bank failure, through a mechanism such as contingent capital.

One question I have is what impact this contingent capital would have on the use of preferred shares as capital.

Andrew F,

Interesting question. If contingent capital was designed as debt that was convertible into common shares, you'd end up in a situation where bondholders ended up behind existing preferred shareholders in terms of preference. There's no reason why you couldn't do that, I suppose, although it's a bit of an odd result. I suppose you have convertible capital that was convertible into new preferred shares (which preferred shares would rank ahead of all other classes of existing preferred shares) which, I might have though (without knowing much about bank capital requirements), would have the same effect on common share holders.

One of the interesting flaws of the new supposedly "tough" financial bill in the US is that most of the provisions such as "Volcker Rule" don't apply to foreign subsidiaries of US banks. So while Goldman Sachs won't be able to engage proprietary commodities trading in the US it could simply move these operations to its Goldman Sachs Commodities Canada subsidiary in downtown Calgary under the jurisdiction not of the federal government but the Alberta Securities Commission. The head of the Alberta Securities Commission has already come out essentially called many of the new rules in the US on derivatives, hedge funds and commodities trading as politically motivated so I don't think Goldman has much to worry from the ASC.

Someone from the US actually argued to me that the hotshot proprietary energy traders at Goldman would never give their homes in the Hamptons to live in some cow town called Calgary something I don't necessarily agree with especially if their line of work in the US faces complete prohibition. I also don't see what direct influence the G20 has with provincial securities regulators unless we create a new G20+13 made up of the G20 plus the 13 provincial and territorial securities commissions.


Although the Feds can (and, by all accounts will in short-order) override provincial securities regulation with their own. So, if the Federal government were to sign off on a G-20 proposal for financial regulation, the provinces wouldn't be able to say boo about it.

I know, I know, Alberta and Quebec are trying to fight that proposal on jurisdictional grounds, but that's a loser argument. The Feds clearly have jurisdiction in this area, either under section 91(2) of the Constitution Act, 1867 (dealing with the regulation of trade and commerce) or under the federal government's residual POGG powers. And the Federal paramountcy powers would mean that the federal legislation would take precedence.

"It would appear that the Europeans are more interested in finding people to pay for their bank bailouts than in preventing them from happening in the first place."

Kinda the same way the Americans approach drilling for oil...

Re Bob Smith:

I agree with you 90%. 90% of what the provincial securities commission do is within federal jurisdiction under the consitution and has been since the dawn of Confederation. The federal govt's power is even further strengthened given its powers to sign treaties and regulate foreigners or foreign companies. The 10% percent that is legitamately within the power of the provinces deals with almost irrevelant issues such companies in Alberta selling stock only to residents of Alberta. Yet in the great soap opera of Canadian history the provinces have used this 10% of legitimate power along pure political muscle to control 100% of a vitally important area of Canadian commerce and life.

The political aspect is no less relevant either. The National Energy Program was fair and square constitutional Peter Lougheed I think would even agree with that and was enforced by not just Trudeau and Turner but for several years under Brian Mulroney. Yet the political consequences of the NEP a generation ago have more than anything else created the politics of today and have determined why the current PM is PM and not someone else with more political savvy but might have not been on the "right" side of the NEP in Alberta. I suspect given weak kneed statements by Quebec Liberals such as Stephane Dion and Flaherty himself we will end up with some of opt out ala the CPP/QPP in which case I suspect Ontario might be the only opt in. Furthermore the federal government will effectively be letting provinces opt out of something which is actually in the federal govt's jurisdiction.

The embarassment of provincial securities regulators is due to the Wicked Stepfathers of Confederation, the Judicial Committee of the Privy Council.

There main federal "catch-all" powers in the Constitution are POGG and Trade & Commerce. The main provincial catch-all is Civil Law.

The Judicial Committee sided with Quebec in the 1920's in choosing Civil Law over POGG in a series of cases. This muddied the waters for generations, though the Supreme Court has recently shown a sympathy to the federal position.

I expect the new Canadian Securities Commission to roll out in short order.

In my previous post I was going to mention the embarassement of still using the Judicial Committee of the Privy Council as case law. They did make one sound decission though in the 1920s when Quebec argued Air Traffic Control and Aviation should be a provincial responsibility LOL. The federal govt was able argue that only it could sign the international treaty creating the predicissor to what is now known as ICAO and thus under international law the federal govt would be responsible from the standpoint of other countries for managing aviation. Quebec in return somehow got to obtain the permanent headquarters of ICAO in Montreal for this hijinx.

I did find an obscure clause of the Bank Act which I believe actually covers the issue of who regulates Goldmans Sachs' Canadian subsidiares. The following:

522.27 A foreign bank or entity associated with a foreign bank shall, at the times and in the form specified by the Superintendent, provide the Superintendent with the information that he or she may require.

Effectively this clause of the Bank Act gives OSFI prudential oversight over any Canadian entity such as the aformentioned Goldman Sachs Commodities Canada even if the associated foreign bank doesn't directly due business in Canada as a bank. Even before Goldman becoming a bank holding company during the crisis it would have been covered by this provision by owning banks in Europe and an FDIC insured industrial loan company. Quebec and Alberta have actually recently argued that this prudential authority OSFI already has over foreign and domestic banks and associated entities actually negates the reasons for a National Securities Commission.

I worry about this "contingent capital" idea, and I hope someone can correct me if there is something about it I don't get.

Bank A issues contingent capital, which is a bond when issued. Bank B buys the bond. It expects to get the amount back later plus interest, which is presumably higher than for a regular bond because of the risk of conversion. Bank B likes the high yield and buys a lot of those bonds.

Bank A gets into trouble and triggers the bonds to convert to shares, which are suddenly trading at a fraction of face value because bank A turns out to be owning a lot of worthless securities. Now bank B is in trouble too.

This just seems way too much like the credit default swaps business. I don't think that contingent capital would be anywhere near as good as cold, hard, capital, whether that is high reserve requirements or a bail-out fund. I think we should have both of the later, and forget the contingent capital.

Bank B would trigger it's conversion. It may or may not matter. Of the top of my head, the problem seems to be if there are some subsets of the graph of asset/liability relationships that are strongly connected and for which a blow-up implies insolvency all around, then we are definitely in trouble.

The relationships could be tracked and the effects of triggering the conversion modelled though I haven't heard any talk of doing so.

Yes. Credit default swaps were supposed to be making the financial system more robust. Instead, they wound up as more ingredients in the soup of toxic assets. No one knew what they were worth.

Contingent capital seems to me to have exactly the same potential. Suppose rumours start that a bank is about to trigger conversion. What will its contingent capital bonds be worth? And if financial institutions are all holding each others' bonds, nobody will know whose bonds are worth anything.

This seems blindingly obvious to me in the light of the past couple of years. But why is it not obvious to the people who dreamed up this idea?

Or are they smarter than I think in some way I don't understand?

My understanding is that pricing CDS was not the problem. The contracts specify what the payout if a triggering event occurs. The problem was that the sellers didn't have sufficient capital to cover the payouts triggered by a "credit event" - that's what happened to AIG. Some people also argue that naked CDS are a problem because it's the equivalent of being able to take out fire insurance on your neighbours house; it gives you an incentive to burn down the house.

I don't think it matters what the bonds trade for around the time of conversion. The conversion rate is set when the bond is issued. I think the bigger problem is likely to be the risk premium demanded by buyers of coco bonds. In some respects they kinda get the worst of both worlds in the event of a bankruptcy.


This was my point -- that we already had sufficient tools in the form of a receivership process that can convert bondholder claims to equity claims. We can already do all of this.

The real issues are of political power and the network patterns that you point out. If there was one change I would make, it would be to create separate ownership structures between broker/dealers, investment banks and depository instutions.

Only the first of these should be allowed to hedge. The latter must be sinks for risk. Or another way of putting it would be that if you make money from the spread between short and long term rates, than this income must come at the expense of you absorbing risk.

If the financial institutions weren't hedged by everyone else, then one of them could collapse without bringing the whole house down. But as long as they can hedge away risk, then the government will step in and save bondholders by necessity, since the failure of the institution is not an option, and in order for the institution to function it must have access to the bond markets.

I think the idea is that contingent capital would be more unequivocal and expeditious than receivorship. It can be done over a weekend, without the government sinking hundreds of billions into the banks.

A bail-out fund is an absolute worst-case. The moral hazard means that such a fund could never be big enough. At least the necessity for contingent capital would tend to put a damper on leverage ratios since, at some point, it would be cheaper to issue new common share equity rather than contingent capital. Seems to me it serves a purpose very much like preferred shares, except preferred shares are never transformed into common shares in the event of a 'default' (dividend cut) on the preferred share. It seems to me that it would be much more elegant for solvency problems to be resolved with equity holders being wiped out and (at least junior) bond holders being transformed into the new equity holders. It means any company worth salvaging doesn't spend two years in bankruptcy proceedings, and allows the bondholders to auction off the company if it has no value as a going concern.

I don't agree that a bail-out fund needs to create a moral hazard problem. The key is in the rules that govern what happens if the fund is used.

If that happens, the management of the financial institution should be kicked out. The shareholders should not receive anything, but depositors should be protected. Money from the fund should be used to make the institution solvent again, and it should be then sold off. New owners, new management.

All this should be clearly spelled out in rules that all players understand. Who the managers are who will be held accountable. That they get no severance pay if the fund is used. Any other rules to make sure managers who run an institution into the ground don't stay in power and don't get to ride off into the sunset with bags full of cash.

There would also need to be rules that would force institutions to take a bail-out with all these conditions, if they are insolvent but still trying to operate. We don't want zombie banks like they had in Japan.

If the rules are right, a bail-out fund financed with a bank tax is a very good option.

Depositors are already covered by things like FDIC in the US and CDIC in Canada, and deposits should be insured through deposit insurance. I don't think bond holders should be backstopped, as this creates a moral hazard problem that is perhaps worse than that with equityholders.

You're right about the CDIC and FDIC. They already solve part of the problem, and basically do it using a bank tax funding protection for depositors. The problem is they don't go far enough. They didn't prevent the financial collapse in the U.S. because they didn't cover those companies and the things they were into.

Not that I think they should have covered them. Most of that sort of activity is completely non-productive and the goal should be to regulate it out of existence. But having a backstop for the next time regulation fails is not a bad thing.

Your moral hazard problem already exists, because financial institutions know that governments are very likely to bail them out next time. Saying they will not is not credible because of the huge damage that would result if they ever made good on the threat. A bail-out fund would not create more of a problem, and rejecting the idea will not solve it. It needs to be solved through rules that force managers to take responsibility for their companies and inflict pain on them personally if they fail.

That's why we need contingent capital requirements so that the threat to not bail out failed banks is credible. We need to eliminate this moral hazard, and a bailout fund would institutionalize it.

I don't think a bailout fund could ever be big enough, and keeping such a vast sum of money locked up for contingencies would be a rather unproductive use of the capital. I expect that such a bailout fund would go the way of Social Security and be shoveled into the US federal debt. It solves precisely nothing that just generally raising taxes wouldn't do better.

RSJ: "... that we already had sufficient tools in the form of a receivership process that can convert bondholder claims to equity claims. We can already do all of this"

I don't know enough about the legal machinery know whether this is true or not. The official line (which I don't entirely buy), at least in the US, seems to be that nobody had that kind of authority over the big bank holding companies. In Canada, we don't know if we can handle the doomsday scenario; we've never really faced it. Even during the Depression.

Andrew, I don't think the contingent capital idea will work in a big crisis like the one we've just seen. It would work for the failure of a small bank somewhere. But think about who is going to buy those bonds. It's other financial institutions. In a big crisis, they will all be holding each others' bonds and will all be in trouble at the same time, as we have just seen.

The notion seems to be that buyers of the bonds will be policing the risk of the institutions whose bonds they are buying. I don't buy that. It will be monkey see high returns, monkey buy. The idea that these guys are more sophisticated than that doesn't fit the facts.

They should just keep the bailout fund locked up as ready cash. It will reduce the effective money supply, of course. But the central bank can just create more to bring the economy up to capacity. In a crisis, banks slow lending, reducing the money supply. That's when money would flow out of the bail-out fund and pump it up again.

The moral hazard problem is easily solved as I explained earlier. Also, contingent capital creates the same problem. Management has an incentive to take risky, high-return gambles and hide the risk from bond purchasers.

Paul, I'm certainly no expert, but I'm not seeing where you think the problem is. If a bank gets into trouble - meaning it's Tier 1 capital ratio is too low - then the bond conversion will be triggered. This will help re-capitalize the bank and get it out of trouble.

So say Bank A holds Bank B's CoCo bonds, and Bank B holds Bank A's CoCo bonds. Bank A and Bank B both make lots of stupid decisions (e.g. sub-prime NINJA loans) and now they are in big trouble. The conversion is triggered. What happens? Bank A is recapitalized, Bank B is recapitalized. Bank A's goes from holding bonds in an insolvent bank to holding equity in a solvent bank. That sounds like an improvement. Similarly for Bank B. I don't think it affect their Tier 1 capital.

I think where it gets dicey is if the bank can't be saved. Then the holders of the CoCo bonds get the worst of both worlds. They get risky bonds that convert to equity and get wiped out right around the time you want to be first in line to pick over the corpse. If investors believe the bank can get into a situation where it converts and can still go bankrupt, their going to want some serious risk premiums.

So it seems to me that the issue will be designing the system so that a bank is extremely unlikely to fail if it does convert. And the only way to do that is regulation. Which raises the question: if you we can regulate and preclude failures, then why bother designing a system that's resistant to failures?

As I said - I'm out of my depth on this one.

I don't claim to be an expert either and if I've got it wrong, I hope somebody explains where. But the problem I see is that, in your situation, the two banks wouldn't wind up recapitalized. Where would the money come from to do that? Both banks wind up holding shares in each other and because both are undercapitalized and in trouble, those shares will be worth very little, so they remain undercapitalized.

If it's just a one bank failing and its contingent capital is held by a other banks that are solvent, those banks will wind up owning the failed one and they have an incentive to put in the capital and get it going again. That works. But it won't work in a big crisis like we've just seen.

Paul, I agree that we definitely need someone who knows more than us to weight-in...

I think the idea is that it's the banks OWN common stock that counts towards tier 1 capital, whereas the shares they might hold in other banks don't count against tier 1 capital, nor must they have loss provisions for the bonds and equity they hold (as opposed to loans they have made) - but I may be wrong on that.

I think it works like this if the bank is short $1 million (for argument's sake) in tier 1 capital, it would trigger conversion of $1 million of bonds to $1 million in equity and then they're not short anymore. The money comes from eliminating the claims of the bond holders and turning then into equity claims. It'd be like your credit card company forgiving what you owe them. In a sense, you are richer because your future income doesn't have to go to the credit card. You can spend it however you like.

If Bank A is undercapitalized by $1 million and triggers a conversion of contingent capital bonds, then their liabilities are reduced by $1 million, their Common Stock (Shareholders Equity) is increased by $1 million, improving the capital ratio and diluting existing shareholders. Say Bank B owned that $1 million in bonds. Their assets would be reduced by $1 million less than mark-to-market value of the equity, while their Retained Earnings (Shareholders Equity) is reduced by the same amount as they book the loss. Now, if Bank A has value, then the loss Bank B incurs is less than $1 million. So if Bank B was just barely adequately capitalized before Bank A triggered a coversion, Bank B will be short strictly less than $1 million in capital, and probably substantially less.

So if Bank B must then convert contingent capital bonds to equity, and Bank A holds these bonds, the feedback will not necessarily blow up and leave both banks insolvent. But they could shrink precipitously. The same problem we saw in 2008 results if either bank runs out of contingent capital and equity before the bank is solvent. That is, assets < (liabilities - contingent capital bonds). If we look at AIG, for instance, it had about $1 trillion in assets in 2008, and required somewhere in the range of $150 - $200 billion in additional capital to become solvent once again (please double check this). So even if this event were to happen again, requiring contingent capital of 20-30% of assets ought to be able to absorb most or all of the failure, and AIG blew its brains out with fairly risky behaviour. With some better regulation (say, increasing capital ratios of CDS, perhaps by putting them through a central exchange with margin requirements), that risk could be reduced and so in turn the contingent capital requirement. I suspect the risk premium that would be paid on contingent capital bonds would tend to raise the bar on acceptable assets and thus keep balance sheets smaller than otherwise. Also, I could imagine that banks would elect to have higher capital ratios in an attempt to reduce the premium they'd need to pay on their contingent capital. And of course, the common shareholders would have a whole lot of incentive to watch risk very closely, since there is a quick, unequivocal process by which their holdings become worthless.

It may be that the government might feel the need to bail out a bank that fails so spectacularly that it blows through all its contingent capital. But at least the size of the bailout will be reduced considerably by the shock absorber of contingent capital, as a systemic failure would result in a substantial reduction in the size of balance sheets across all banks. And since it will likely be just a liquidity crisis, rather than a true solvency crisis, the government may elect to recapitalize through the issuance of new warrants.

Thanks, Andrew and Patrick. I think you have at least convinced me that contingent capital is not as bad an idea as I thought it was. I think I understand better how it works.

But, while I no longer think it would add to a crisis as I did before, I am still not convinced it would do much or anything to solve one.

Suppose it's 2008 in the U.S. again, only all those financial institutions have contingent capital. They also hold lots of dodgy mortgage-backed securities. A few institutions start to trigger conversion. When that happens, they are, as you explained, off the hook as far as paying the bondholders. But what about those bond holding institutions? They were counting on repayment of the bonds to pay their bills, and now it won't happen. They have the shares, which they can sell. But if this is widespread and everyone is trying to raise money by selling shares in institutions that have converted and that will be presumed to be in trouble, who will be buying and what will those shares be worth?

I still think a bail-out fund is a good idea. Just set up rules that ensure that managers take a big hit personally if the institution needs rescuing, and the moral hazard problem goes away.

Paul - you seem to be describing, in part, a liquidity crisis (everyone wants to sell at once). The textbook prescription for that is for the central bank to provide the liquidity (take illiquid assets as collateral and lend short term to other banks). This prevents the firesale death spiral.

As for the bond holders; yes they do loose the flow of payments from the bond when it converts to equity, but they aren't wiped out. The equity has value and they'll get the upside if/when the bank recovers. Better to have shares in a bank that is a going concern than to be a bond holder picking over the corpse of a dead bank.

Paul Friesen,

Contingent capital would help maintain bank solvency. When the conversion kicks in, the bank's liabilities are reduced. This allows the bank room to sell off assets (say to pay down liabilities that are due) and still maintain required capital ratios. If the bank owns another bank's contingent capital that gets converted, it doesn't have to result in a firesale of the newly acquired equity; the bank can sell other assets to raise the cash it needs. Also, other banks will not be the only holders of the contingent capital. In the US at least, other major investors in high yield debt include insurance companies, pension funds, endowments, hedge funds, and mutual funds. Whether proposed levels of contingent capital would have been sufficient in the 2008 crisis is another matter.

You said, "Just set up rules that ensure that managers take a big hit personally if the institution needs rescuing, and the moral hazard problem goes away."

I don't think this is correct. Banks operate at very high leverage. This means that the creditors have far more at risk than the owners or the managers. The managers, especially, can reap many years of large bonuses before the risk-taking causes an implosion and sinks the company. Even in the absence of a bailout of the equity holders, the managers are in a position to benefit by making big bets with other people's money. The moral hazard problem is therefore mainly the result of bailing out creditors, which reduces creditors' incentive to monitor and restrict the risk taking of the owners/management. http://cafehayek.com/2009/10/how-moral-hazard-works.html

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad