Wednesday, May 07, 2008

MCDX: Once munis start down the dark path...

From the people who brought you the ABX, now comes the MCDX, a basket of municipal credit default swaps (CDS). The index will begin trading on May 6 with three, five, and ten year tenors. Markit set the coupon for the MCDX last Thursday night at 35, 35, and 40bps respectively. It started trading today, and traded wider, closing at 42bps for the 5yr tenor and 48bps for the 10-year.

This is a potential game changer in the municipal market. First, we'll go over what the MCDX is, and then how it might change municipals forever.

The MCDX is going to be very similar to the CDX or ABX indices currently trading. It will represent a basket of 50 equally weighted municipal CDS. You can see the list of credits here. These will be recognized by municipal traders as more or less the 50 largest regular issuers of bonds. There are a few AAA credits in there, but mostly AA and A-rated credits. If rated on Moody's Global Scale, the one where Moody's attempts to match muni ratings with corporate ratings, almost all of these issues would be AAA.

There are 26 "general obligation" issuers. These issuers have the legal authority to levy taxes and have pledged their full taxing power to bond holders. 21 of these are states, the other 5 are local municipalities: New York, Los Angeles, Los Angeles School District, Phoenix, and Clark County Nevada.

There are also 24 "revenue" issuers, who don't have any taxing power. The items in the MCDX are of the "essential service" variety, including water and sewer systems, public power, and transportation. The term "essential service" implies that while the issuer does not have taxing power, the local government would have a strong incentive to ensure continued operation. Tobacco and health care issuers are explicitly excluded from the index.

Here is how the index works. A buyer of protection on the MCDX has essentially bought equal amounts of protection on the 50 names in the index. So a $10 million notional trade in the MCDX is de facto $200,000 in protection on each of the 50 names. Should any of the names default, the buyer of protection would deliver an eligible obligation of the issuer to the seller of protection at par. Markit has provided a list of CUSIPs as examples of eligible obligations. Any bond which is pari passu with the listed CUSIP would be eligible.

So why should you care? To date, trading in municipal CDS has been very light, and with good reason. Default rates of general obligation and essential service municipals are almost non-existent. There is a limited number of large and frequent issuers outside of these two categories. So demand from hedgers for specific names is light. There might be demand from speculators who want to bet on the contagion hitting munis. But such a buyer would prefer to make a generalized bet on municipal credit as opposed to picking out individual credits.

The MCDX solves both these problems. Trading desks who want to hedge against municipal credit spreads generally widening can use the basket as a on-going hedge. It wouldn't really matter if the particular names in the index don't match the names the desk owns, since the hedge is really a macro/contagion position. If California runs into major budget problems, odds are that New York CDS would widen at the same time. Obviously this is a better product for a speculator who wants to bet on a broad municipal contagion. So the MCDX is bound to be a hell of a lot more liquid than the single name market ever was.

The implications for the muni market are huge. First of all, it would seem the MCDX will more or less dictate the price of muni bond insurance. It will also heavily influence the spread between insured and uninsured munis. I've heard some talk that such a product would be another nail in the muni insurance coffin, but not so fast. The muni market will remain retail driven, and mom-and-pop investors don't buy CDS. They will still demand insurance.

The MCDX will also heavily influence how munis trade on a given day, especially in institutional size. If dealer desks start using the MCDX to hedge their books, then the daily movement in the index will become part of their P&L. In other sectors, when traders hedges are up, they are a little more willing to cut the price on their long position. The same will happen in munis. If the MCDX is 3bps wider on the day, traders will be willing to sell their bonds 3bps wider too. Well, maybe 2bps anyway. Traders aren't generous people.

It could also start to chip away at some of the old habits of muni buyers. Today municipals are traded mostly on yield. Even if the Treasury bond market is mildly up on the day, muni traders usually don't mark their positions higher. If the MCDX becomes heavily used as a hedging vehicle, traders will want to quote their offerings in terms of their hedges. Thus you are likely to see offering levels altered more often, and possibly even starting to be quoted on spread.

Right off the bat, it almost has to widen. There are going to be more natural buyers of protection (anyone who has a large muni portfolio) than sellers (speculators). So I wouldn't read too much into the movement of the first month of trading. Given that the natural sellers of the MCDX are probably mostly hedge funds and prop desks, I expect municipals to be permanently more correlated with corporate bonds.

All participants in the muni market should become familiar with the MCDX, even if you have no intention of actually trading it. Like the CDX and the ABX before, it has strong potential to alter the market substantially.


Anonymous said...

"Should any of the names default, the buyer of protection would deliver an eligible obligation of the issuer to the seller of protection at par."

Shouldn't that be the other way around??

Accrued Interest said...

No that's right. The buyer of protection owns a put to the seller at par.

Anonymous said...

What is to stop a seller of protection from selling more protection than they can actually cover? Presumably, a hedge fund can sell 10x protection and collect 480 bps until the trade blows up. Is no one going to make them post margin?

Anonymous said...

also, stupid question on credit default swaps:

Assume that they are a whole class of bond eligible to be put to the CDS seller, what are the factors which make putting one bond better than putting another bond? Is it just this: buy the cheapest bond price-wise (say at 30 or 40) and put it back at 100?

Accrued Interest said...

I believe the MCDX would follow similar rules as CDS. It would be up to the counter-party to set a margin requirement. If any of our readers who really trade this stuff know more detail on this, please post.

Again, think of it like shorting a put in equities. The same issue applies.

Accrued Interest said...

And on delivery: there is a cheapest-to-deliver issue here. I know you can't deliver zero coupon bonds on the MCDX, but otherwise its pretty much anything. In munis the CTD issue will be less clear-cut than in corporates where the issue sizes are larger.

There is also the potential for an auction process where the contracts settle in cash. I honestly don't understand how the auction works, but it is an attempt to prevent a squeeze of cash bonds in the event there are more CDS than cash bonds.

PNL4LYFE said...

I think AI's answers regarding counterparty margin and auction settlement are right on. If someone sells protection on this index (or anything else), they are effectively buying a bond at or near par depending on the index strike. As the credit widens, the CDS MTM goes against them just as if a cash bond they bought at par fell to 90. At this point, the CDS counterparty would require more margin. This wouldn't work in a case where the credit widened massively within a few days, but I would imagine that it's usually more gradual. Also, some dealers have increased their margin requirements on CDS lately.

Any defaults in the index will probably be settled by the auction process. Basically, investors and dealers submit bids and offers and a process is used to determine the clearing price where all CDS (single name and index) will cash settle. Single name CDS holders have the option of opting out of the auction and physically settling bonds. I believe index holders are required to adhere to the auction price. Details of the auction process are at

Anonymous said...

Ok, so I -- like a lot of low involvement muni investors -- have all of my muni portfolio in the form of Vanguard funds, some Nuveen closed-end funds, and a couple of ETFs (MUB, PZA).

As much as I tried to follow your detailed and thoughtful analysis, I was really missing some sort of statement at the end to the effect of "Bottom line: If you're a holder of diversified muni funds, this means _____ for your investment, and you should consider _____." In other words, a little further dumbing down for the luddites in the audience would be greatly appreciated. :)

Accrued Interest said...

If you are going to trade your positions, then it would be worth watching the MCDX. If the MCDX moves wider but your funds don't, then it might be worth trading out of them.

But if you are a guy who is holding munis for the classic reason (you want the tax-exempt income and want a stable off-set to your stocks), then there is nothing to do.

Anonymous said...

Are there any interesting tax angles to the MCDX? Protection sellers have to pay tax on any profits, right?

Anonymous said...

Just curious about your thoughts on this theory. Bloomberg had an article a while back on how the ABX was created to keep the CDO machine moving as the Fed was raising interest rates and squeezing yields. If I understood correctly, ABX helped keep the CDO machine moving by making lower tranches of synthetic CDOs easier to hedge and therefore easier to sell.

So when I heard about the MCDX, I thought, here we go again. Basically it seems to me that synthetic CDOs are not about financing anything, but about pure speculation on both sides -- which becomes dangerous when the index starts to influence the cost of funds for those who are trying to actually do something productive.

Feel free to tell me I'm an uninformed luddite, but is it possible that this new index will be used to keep some of the worst aspects of structured finance in business?

Accrued Interest said...

Well, I wouldn't expect any synthetic CDO's anytime soon. But if your point is that CDS will heavily influence trading, and possibly make it easier to manipulate/obfuscate fundamental value... I'm with you 100%.

Here are my thoughts on that:

Anonymous said...

i completely disagree with your comment that this index will fundementally change the way munis are traded. most dealer desks need to hedge their duration risk not credit risk.

Accrued Interest said...

To be fair... dealer desks haven't been able to hedge credit risk until now. Think of how much better their P&L would have been in late February if the MCDX had been around. I think it would have been much better.

Anonymous said...

it was muni/tsy basis risk that killed them. in my view, the only credit that deteriorated were insured bonds with weak underlyings, and this index wouldn't have helped with that.

Accrued Interest said...

Yeah but if you look at how the individual muni CDS traded at that time, it would have been a useful hedge. JPM sent out a report that showed the most heavily traded muni CDS moved from about 10bps in December to 35bps at the end of February.

Anonymous said...

But will risk management overseeing these trading desks allow traders to hedge interest rate risk with a vehicle designed to primarily reflect fluctuation in credit risk?

A play in monoline insurers' CDS might have been an effective hedge for a long muni cash position earlier this year as well, but just because it might have worked then doesn't mean it's the right hedging mechanism.

Institutional traders want to hedge interest rate risk - buyers, and especially retail buyers want to hedge credit risk, and you've already pointed out that the majority of these (retail) buyers are more comfortable with insurance, not a CDS index. So I'm not sure how this is such a great tool for the traditional, institutional muni trader as a hedge tool.

Defaults are so low in municipals (compared to corporates) that there is a general push to rate them on a global scale (effectively upgrading ratings on all muni issuers). With such low default risk, many issuers have called into question the relevance of muni insurance, and this CDS index appears to be just another form of protection against default risk that is already extremely low.

It will be interesting to see just how much traction the MCDX is able to achieve.

Accrued Interest said...

So far I'm hearing the traction isn't good. Trading in the contract is lite. It will never work as a hedge for anyone if it isn't highly liquid.

Anyway, wouldn't you agree that its impossible to separate municipal interest rate risk vs. credit risk?

Anonymous said...

Theoretically, I don't think it's impossible to separate interest rate risk (or duration risk) from credit risk. Effectively hedging these independent risks is the real world challenge. In munis, it's extremely difficult to achieve a 100% effective hedge against either, but conceptually they can be easily stripped out. Muni insurers exclusively protect (hypothetically anyway) from credit risk, and in theory strip credit risk out for the investor.

The bondholder is still susceptible to interest rate risk however, as he will see depreciating prices in a rising rate environment regardless of the credit enhancement.

On the flip side, the owner of an uninsured bond that defaults is obviously vulnerable to credit risk. The credit risk remains regardless of of how perfectly the position is hedged against the risk of rising rates, or even if the credit event takes place in a falling rate environment (on an uninsured bond).

We haven't even brought basis risk into the picture - which was the real bane of the muni market earlier this year.

Accrued Interest said...

I guess I should have said that basis risk can't really be separated from credit risk.

In other words, basis risk in corporate bonds is all credit risk. But in munis there are special supply/demand factors that aren't related to credit risk. So when you see MMD cut 3bps on a day when the Tsy is 2bps better, is that just basis or is there a credit element? Impossible to say.

Accrued Interest said...

Closed yesterday at 42, 46.5, and 49.

Anonymous said...

I have a municipal bond ladder built of 100% Ct issues, about 50%state and town GO bonds and the rest are hospitals and educational institutions -- including some resets where the auction is failing ( and the rate isn't that great for 30 year paper ). If I were concerned about a muni-meltdown, I assume I should just buy protection, right? My question is -- how do I do it? Are these things listed somewhere? What is the minimum size? ( the total face value of the bonds is $1.9 million )

Accrued Interest said...

You should call your broker and ask him. I'm honestly not sure. You might have to be a QIB.