Felix Salmon

Why we should ban reverse converts

Felix Salmon
Jun 18, 2009 15:01 UTC

I had no idea what a can of libertarian worms I was opening when I suggested that the sale of reverse convertible bonds should be banned. I dealt with one line of objections last night — the silly idea that since reverse convertibles are basically a put-writing strategy, and since Warren Buffett writes puts, then reverse convertibles must be kosher.

But now a more invidious objection has cropped up, as hilariously encapsulated by Daniel Indiviglio:

Its usefulness must be clear to someone — otherwise there would be no market demand for the product, and one never would have been sold.

I must admit I laughed out loud on reading that. Reverse convertibles are sold by banks as a way of generating cheap capital and hefty fee income. No one in the history of the world has ever approached their stockbroker with a bright idea that what they really want to do is lend a bank a few thousand dollars at a high interest rate, but would be happy to receive depreciated Apple stock instead of their money back if AAPL does badly over the next few months.

Indiviglio and his fellow-travellers like Vincent Fernando have been defending reverse converts on the grounds that (a) you can view them as sophisticated options-writing strategies and that (b) sometimes sophisticated investors actually want to enter into such strategies.

But people, these things are called reverse convertible bonds. The options-writing part of the deal — which is the beating heart of these instruments — is generally downplayed to the point of invisibility, while all of the emphasis is on the juicy “coupon” being paid by the bank. (Which is really the bank buying a cheap option from the investor.)

Here’s Fernando:

I don’t think we should be banning products just because we ourselves believe that others are fools for buying them. Rather, I think it’s healthier if we encourage a culture of skepticism towards any investment opportunity rather than banning cherry-picked products completely and implying that the approved products are “safe”. We trust almost anyone to have the brains to drive a car at deadly (upon impact) speeds, yet not to do a little homework towards a financial product?

You need to think hard when you invest substantial sums of money and you shouldn’t give much weight to the words of the guy selling you the product. For any product. Too big of a financial decision for the average person? I doubt it. We let people choose their career types, their schools, their houses, etc. We don’t say the average Joe isn’t sophisticated enough to buy a house. And we shouldn’t.

I think we’re seeing right now what happens when the average Joe buys a complex financial instrument (a subprime mortgage) which he doesn’t understand. The whole point of the Consumer Financial Protection Agency is to prevent predatory lenders from talking naive individuals into products which are egregiously unsuited to them. Similarly in this case, stockbrokers should be prevented from persuading naive individuals to start writing underpriced put options while kidding themselves that they’re just buying a high-interest short-dated bond.

Mike at Rortybomb has a great post on all this, with a couple of very sensible principles:

People are Poor with Estimating Tail Risk; They Underestimate it

People should be discouraged, though not prevented, from profiting by taking on excessive tail risk. If they do want it, they should be made very aware of what they are doing.

People are Poor at Understanding Embedded Options

When embedded options are hidden inside financial products, they should be made very clear, and if reasonable, broken out into another product.

In general, we shouldn’t assume that most Americans can, will, and should “do a little homework” before buying something from their friendly stockbroker. This is why it’s so good and so important that stockbrokers are soon going to have a fiduciary duty: they won’t just be able to blithely sell nuclear waste to their clients any more. But more to the point, most Americans aren’t remotely financially literate enough to understand this stuff. They’re not good at math, they barely understand the difference between a stock and a bond, and they certainly should have no business playing in the options market. (Similarly, an enormous number of investors in the USO exchange-traded oil fund should never have bought it because they think they’re just buying a proxy for the spot oil price, and they don’t have a clue what contango is. If you don’t understand the dynamics of contango, you have no business buying USO.)

People who live and breathe the financial markets are more than welcome to do all manner of sophisticated things in those markets if they’re so inclined. As Mike says, if you want to short a straddle on AT&T stock, go right ahead. But retail-facing financial instruments should never embed such bets, because retail investors, as a rule, lack the sophistication necessary to be making such bets in the first place.

Fernando looks at those investors and says it’s “their responsibility to be skeptical buyers”. No. It’s the stockbroker’s responsibility to act in the investor’s best interest, and it’s the government’s responsibility to prevent the sale of products which can end up being extremely damaging to those who buy them. As Elizabeth Warren famously says, no one has a problem with the government banning the sale of dangerous toasters, and dangerous financial products cause much more damage than dangerous toasters ever do.

The fact is that reverse converts, in particular, are the kind of product which can cause a great deal of harm; what’s more, they exist entirely so that banks can use their stockbroking arms to rip off their clientele. Banning them would do much less harm than selling them. So let’s ban them, and their ilk.


Our money manager purchased converts for my wife’s accounts, so some comments from my (narrow) experience:

1) The interest rate usually ranged from 15% to 25%
2) The price drop to trigger the put was usually 15% or so (sometimes more)
3) We did OK until the fall of 07, when we were put a couple positions, and we subsequently pulled everything from that money manager (Long story)
4) The biggest downside that I observed was that gains were taxable as interest income but losses were taken as long term (or short term) losses, and this was an expensive mismatch

Net/Net – I agree with Felix. I was much more educated in these products than the average person (I actually read the prospectus and I have traded options), but I had no way of making an evaluation of these at the time – I just took the MM’s word for it.

Posted by Patrick | Report as abusive

UK banks are too big

Felix Salmon
Jun 18, 2009 13:25 UTC

Peter Thal Larsen picks up on a potentially-explosive quote from Bank of England governor Mervyn King last night: that “if some banks are thought to be too big to fail, then, in the words of a distinguished American economist, they are too big”. King, of course, is the UK’s top bank regulator, which means that some of the biggest banks in the world ought to be feeling rather worried right now: too-big-to-fail banks dominate the UK banking system.

King does in his subsequent thoughts back off a little from the implication that TBTF banks should simply be chopped up into small-enough-to-fail pieces. Instead, he seems to think that something along the lines of Glass-Steagall might be in order:

It is not sensible to allow large banks to combine high street retail banking with risky investment banking or funding strategies, and then provide an implicit state guarantee against failure.

I think this doesn’t go far enough: both high street retail banks and risky investment banks can be TBTF, and blowups can happen in retail banks as well as investment banks. Just look at Northern Rock.

My bright idea is a simple cap on the size of any bank’s balance sheet: I’ve proposed $300 billion in the US, and the number would presumably be smaller in the UK. That’s a lot easier than trying to create highly-complex capital adequacy standards which fluctuate according to perceived systemic risk. And a lot harder to game.


I would limit the amount of risk-based capital that they could employ, and make sure that the actual tangible capital was twice that. Use a scheme like the insurers use. It works well for us — few defaults since it was instituted.

Oh, and the insurance regulators don’t allow for soft types of capital…

PS — Okay, AIG was an exception, but they hid what they were doing with securities lending… a hole in the formula, I think. Had most of the life insurance subs been independent, most would have failed. Also the mortgage insurance subs, but for different reasons…

Wednesday links translate quite well

Felix Salmon
Jun 17, 2009 22:31 UTC

Chittum reviews Tett

Translate the white paper! Mike O’Sullivan’s Financial Reform Glossary

Kedrosky calls California “Latvia by the Pacific“: “while California won’t default, for practical purposes it should”

The new regs: “more like a Windows Service Pack upgrade than a whole new architecture… FinReg Vista?

Ahmadinejad sucks at Photoshop

Autistics better at problem-solving

“The current generation of managers running Starbucks has just figured out that its stores do not brew enough coffee

Great stuff from Rortybomb on the white paper. Banning prepayment penalties is a great idea.

Ultrasilliness from Pesek: “Let’s assume for a moment that these bonds are real”. Er no, let’s not. ‘Cos they’re fake.

More on the reverse-convert scam

Felix Salmon
Jun 17, 2009 22:26 UTC

Nick Schulz and Philip Guziec are both pushing back against my assertion that reverse convertible bonds are a scam, and they’re both wrong.

The argument is similar in both cases: by buying a reverse convert, you’re essentially selling a put option. Selling a put option can make sense, sometimes. So what’s the problem?

There’s a bunch of problems. For one thing, as Guziec concedes, these instruments aren’t sold as an option-writing strategy. He was offered one once:

I declined the offer as I would any inbound call from a broker, but I did say, “So, you’re asking me to write a put on company XYZ.” I recall that my comment generated a sheepish response.

For another thing, if, pace Guziec, you actually bothered to price out the bond by working out how much it would cost to replicate it in the options market, you would never buy the bond, since it will always be much cheaper to just do the trade yourself.

But in any case, as JH points out, pricing these things is pretty much impossible, especially when they have “knock-in” characteristics which make them path-dependent (don’t ask).

As for the specific arguments, Schulz seems to think that buying a reverse convert is tantamount to investing like Warren Buffett:

They are so much like insurance that it explains why Mr. Geico himself, Warren Buffett, has been selling put options. So if insurance is pointless then Kwak has a point.

Well, yes. But individuals should be buyers of insurance, not sellers of insurance. If you’re Warren Buffett, you can sell insurance. But most of us shouldn’t go near that business — which is essentially the business of pocketing a relatively small premium up-front in exchange for promising to pay out a large amount of money if things go wrong. Nearly all insurers go bust eventually, and writing insurance is not a remotely sensible thing for any individual investor to do.

Guziec takes a similar tack:

We are actually quite fond of this type of transaction, as the market often offers huge insurance premiums to sell the downside on fundamentally sound companies, often with economic moats. This high-volatility environment actually raises the value of the put options while lowering the value of many sound companies, making put-selling strategies particularly attractive. In fact, even superinvestor Warren Buffett has been selling puts recently.

Again, yes but. Buffett hasn’t been selling short-dated puts on individual companies, he’s been selling long-dated puts on stock market indices. There’s a world of difference. (Oh, and he’s lost billions of dollars on those deals.)

If the reverse converts were tied to indices rather than individual stocks, I might hate them a little bit less. But I’d still hate them, if only because neither Schulz nor Guziec mentions the other toxic part of the deal: you’re taking not only stock-market volatility risk, but also the unsecured credit risk of the issuing bank. In all of these cases you’d be better off just buying short-dated debt of the bank in question. And if unsecured bank debt is too risky for you, you certainly don’t want to go anywhere near a reverse convert.

The fact is that if these things made sense, they’d be much more popular, and they’d be bought by relatively sophisticated investors who are comfortable trading options. In reality, they’re quite uncommon, they’re generally sold by banks desperate for capital, and they’re generally sold to investors who have no idea what, really, they’re buying.

But the good news is that stockbrokers, if the regulatory reforms go through in their present form, are going to have a fiduciary duty for the first time. And that alone should put an end to these things.


Kyle, I agree with most everything in this and your previous comment, with one exception. It is my understanding that reverse convertible securities, unlike most converts, are almost always written on a different underlier than the issuer. For example, say Company X (e.g. Citi) issues a reverse convert where the stock into which the instrument is convertible is that of Company Y (e.g. Microsoft). In the parlance of converts, this is sometimes called an “exchangeable” feature.

The upshot is that your suggested “good reason for this product to exist” (that it bakes-in an expectation that these subordinated debtholders will get converted to equityholders should the company get in hot water) doesn’t much apply but to a small subset of the population of reverse converts. But should we hope this subset grows for the reasons you espouse? Maybe. Though I’m not sure there’s much of a difference between a issuer-indexed reverse convert and a mandatorily redeemable preferred stock. There is one nuance: the preferred stock at least has a liquidation preference, whereas an issuer-indexed reverse convert would effectively have no claim in liquidation but to the residual (as it would have then converted to common equity).

The recipe for effective regulation

Felix Salmon
Jun 17, 2009 22:02 UTC

There’s no doubt that Barack Obama gives a good speech:

Where there were gaps in the rules, regulators lacked the authority to take action. Where there were overlaps, regulators lacked accountability for their inaction.

One his plan is put in place, Obama says, the Federal Reserve will be held accountable for regulating any firms which pose a systemic risk. Not the Financial Services Oversight Council, which is more of an advisory shop which will help the Fed keep tabs on what’s going on.

This is the good part of the plan, I think: making it hard for regulators to pass the buck or claim that they had no authority to intervene. On the other hand, as we saw with the OTS and AIG, sometimes regulators are just simply out of their depth, and the sheer number of regulatory institutions which is going to exist is so incredibly large that, statistically speaking, it’s inevitable that many if not most of them will end up falling down on the job: I don’t think it’s possible for any country to have more than two or three really effective regulators. (One is hard enough.)

So there might be new authorities, and there might even be new accountability — although accountability will be to Congress, which is a recipe for political infighting rather than transparent benchmarks. But I’m not holding my breath that all this change will actually be particularly effective.


On the other hand having a multiple of regulators increases the chances one at least will note the shenanigans the Wall Street “Masters of the Universe” are trying to pull. I recall that in the run-up to the Iraq invasion the US had many different intelligence shops, but the only one that actually got it right that Iraq had no WMD was the little bitty one the State Department had. I’d think the situation with financial oversight would be analogous.

Posted by borisjimbo | Report as abusive

Introducing prize-based savings to the US

Felix Salmon
Jun 17, 2009 16:56 UTC

Matthew Bishop is tweeting from the Global Financial Literacy Summit in Washington today, featuring not only Ben Bernanke but also Sheila Bair. This is particularly intriguing to me:

FDIC chair Sheila Bair is looking at creating ‘prize-based savings’ to turn some of the money people spend on lottery tickets into savings.

Prize-based savings, such as the UK’s premium bonds, make a lot of sense: they’re particularly good at appealing to segments of the population who have no savings plan currently and who are heavy lottery players. But they also tend to run straight into legal obstacles: South Africa’s million-a-month account, for instance, was closed down last year on the grounds that it contravened the National Lotteries Act.

If prize-based savings are currently illegal in the US — and they almost certainly are — then it makes a lot of sense to use the new financial regulation legislation to make them legitimate. Maybe the birth of the Consumer Financial Protection Agency is the perfect opportunity to regulate such things effectively.


I’m glad to hear that someone is thinking about incentives in a rational manner. Take another arena: voting. Imagine how many more people would vote if a ballot was also a ticket for a prize.

Posted by jonathan | Report as abusive

How much do chief risk officers talk to each other?

Felix Salmon
Jun 17, 2009 16:35 UTC

Algonaut asks whether the Financial Services Oversight Council will have a direct line to banks’ chief risk officers; I’m sure the answer is yes. But I also think that won’t be enough. What I’d love to see — and this could be put in place directly by the major banks, without the need for any legislation at all — would be a regular formal meeting of all the big banks’ chief risk officers, where they can talk about all the systemic risks they’re worried about which require coordinated response. Does anything like that exist? Is there some way in which the FSOC or the Fed could use its moral suasion to make it happen?

Update: It turns out that the IIF has a Markets Monitoring Group, chaired by Jacques de Larosière, which meets 2-3 times a year with the aim of “bringing together observations and assessments of various developments to build a systemic picture of current risks and their potential negative impacts and seeking to mitigate those risks by encouraging member firms to take the Group’s findings into account in their risk management and collaborating closely with the official sector”. (From page 108 of this document.) Chances of it doing any good at all? Very slim, I’d say, but then again I’m biased against the IIF so I’ll admit I’m not an impartial observer.


Well, there is the ABA’s annual Compliance Conference, at which senior regulators and bank compliance officers meet to discuss regulatory topics…

Posted by Eric Dewey | Report as abusive

Why the SEC won’t merge with the CFTC

Felix Salmon
Jun 17, 2009 15:56 UTC

David Sunstrum emails with an interesting idea:

I know the big surprise of the last couple weeks was how the SEC would survive Obama’s regulatory shakeup. But the paper repeatedly states (bottom of p. 6 for instance) an intention to harmonize regulation between futures and securities. Could they be paving the way for a merger of the CFTC and the SEC in the not-too-distant future? Did he not want “Obama abolishes SEC” headlines?

I’m not holding my breath on this one. Here’s what the paper says:

We recommend that the CFTC and the SEC complete a report to Congress by September 30, 2009 that identifies all existing conflicts in statutes and regulations with respect to similar types of financial instruments and either explains why those differences are essential to achieve underlying policy objectives with respect to investor protection, market integrity, and price transparency or makes recommendations for changes to statutes and regulations that would eliminate the differences.

While it’s conceivable that the “recommendations for changes to statutes and regulations” might comprise an outright merger of the CFTC with the SEC, I’d put the probability at roughly zero, for three main reasons:

  1. All institutions have a natural tendency to self-perpetuation, and government institutions particularly.
  2. The CFTC is principles-based, while the SEC is rules-based. That’s like oil and water: you can’t comfortably merge two institutions which are run so very differently.
  3. The report has to go to Congress, where the SEC is governed by the House Financial Services Committee and the Senate Banking Committee, while the CFTC is governed by the House and Senate Agriculture committees. Any merged entity would lose the oversight of the Ag committees, which are very attached to the power and influence which comes with being able to regulate the futures market. What’s more, Congress feels zero responsibility for having in any way caused or enabled the crisis, and therefore feels zero need to reform its own outdated practices, like having the House Agriculture Committee oversee the CFTC.

Basically, any momentum for an SEC-CFTC merger has been lost, and the two agencies will remain separate for the foreseeable future. It’s a pity, but it’s in line with the generally-disappointing nature of the reforms being announced today.

The US move to principles-based regulation

Felix Salmon
Jun 17, 2009 15:24 UTC

Remember the regulatory arbitrage whereby clever use of securitization could reduce the amount of capital that banks needed to hold without reducing the amount of risk on their balance sheet? Well, the white paper wants to put an end to such shenanigans:

Risk-based regulatory capital requirements… should minimize opportunities for firms to use securitization to reduce their regulatory capital regulatory capital requirements without a commensurate reduction in risk.

On the one hand, this is blindingly obvious stuff, but it’s also well worth codifying somehow, given the failures we’ve seen over the past year or two.

And here’s the language forcing banks to keep skin in the game when it comes to securitizations:

The federal banking agencies should promulgate regulations that require loan originators or sponsors to retain five percent of the credit risk of securitized exposures… The federal banking agencies should have the authority to specify the permissible forms of required risk retention (for example, first loss position or pro rata vertical slice) and the minimum duration of the required risk retention. The agencies should also have authority to provide exceptions or adjustments to these requirements as needed in certain cases, including authority to raise or lower the five percent threshold and to provide exemptions from the “no hedging” requirement that are consistent with safety and soundness.

All of this caused no little debate in the increasingly-lively Reuters Commentary newsroom, where I seem to be in a minority of one when it comes to the rules-based vs principles-based debate.

There’s no doubt that the white paper tilts quite strongly in the direction of principles-based regulation — which I think is a good thing. If you have rules, you invariably create a mini-industry of hundreds if not thousands of lawyers trying to come up with ways of getting around those rules while remaining within the law. This is not helpful. A principles-based approach, by contrast, if enforced by regulators who haven’t been completely captured by the people they’re regulating, tends to be much more constructive. (And if the regulators have been captured, it makes no difference either way.)

Didn’t a principles-based approach fail quite spectacularly in the UK? Yes — but the UK financial system was so big and so leveraged that in the face of a financial and economic crisis of this magnitude, failure was inevitable whatever the regulatory structure. I’m still of the opinion that at the margin, a principles-based approach will be more helpful and less harmful than a lawyered-up rules-based approach, and one of the good things about this white paper, in my view, is that it does build a certain amount of flexibility into the system, rather than just trying to construct new rules for financial companies to navigate their way around.


“enforced by regulators who haven’t been completely captured by the people they’re regulating”

But Felix, they will be captured by their own practice and rulings. Letter for letter, word by word, their work will build a mountain of written texts that all the lawyers and bankers will collect and study. And they will start each letter with “Since you said that x was okay in your letter of you must okay our y since it share very much the same basic structure as x”. And the regulators will never admit a mistake, and they will not be able to recruit those who are clever enough to spot the difference between x and y in the short time at disposal to write the answer.

But then I’ve always been pessimistic about the chance of reform changing anything if the job still is done by the same people as yesterday.

Posted by Gaute | Report as abusive