A few months ago I heralded the MCDX as a potential game changer in the muni market. I'm afraid its failed to live up to its potential, and arbitragers are likely to step in and push it into oblivion.
For those who did not read the original article, the MCDX is credit default swap (CDS) index of 50 municipal credits. By buying (or selling) the index, you are in effect buying (or selling) equal portions of 50 different protection contracts. If one of the credits within the MCDX defaults, the buyer of protection delivers a qualified obligation of the defaulted credit to the seller of protection. In return the seller of protection pays 100% of the face value. The par amount of the bond delivered is equal to 1/50th of the original notional amount.
The current 5-year MCDX spread is 86.25bps. This means that in order to buy $10 million in default protection against the 50 names in the MCDX, investors make the equivalent of $86,250 in annual payments, assuming a new contract were created today.
Historically, defaults on municipal credits have been very rare, particularly in comparison to corporate credits. According to Moody's, there was only 1 default of a "general obligation" (GO) municipal (meaning a state or local government with taxing authority) from 1970 to 2006, and that default was cured in 15 days. Among investment-grade, non-GO municipal bonds, only 0.29% defaulted within 10-years of issuance, according to Moody's. This compares with 2.09% of investment-grade corporate bonds.
However, the weak economy generally, and declining property tax collections specifically, could result in unusual pressure on municipal credits. Yet, the particular credits within the MCDX are among the safest in the market. Property taxes are usually collected by states, but distributed primarily to cities, counties and school districts. While other revenue sources, from income and sales taxes to toll collections are likely to be impacted by the current economy, those revenues are likely to follow a more typical recessionary pattern, which municipalities have weathered in the past. The MCDX includes only 4 cities (New York, Los Angeles, Phoenix, and Columbus, OH), one county (Clark County, NV), and one school district (Los Angeles).
So perhaps municipal default rates will rise in the future. But wouldn't we expect corporate default rates to rise as well? Compare the MCDX with the CDX IG, an index of 125 investment-grade corporate credits. The CDX IG closed on Wednesday at a spread of 144bps.
Using standard recovery assumptions for both indices (80% for munis, 40% for corporates), one can calculate the expected default rate based on each index's current spread. The graph below shows the cumulative expected default rate based on current spreads for both indices.
As you can see, the MCDX is trading at levels that imply nearly 20% of the municipals in the index will default within 5 years. Again, it seems likely that municipal credits will be more stressed than in years past, but given that the credits within the index are only mildly exposed to property taxes, a 20% default rate seems unlikely.
In addition, because of the stronger recovery expected in municipals, the current spread levels imply greater default levels for municipals than investment-grade corporate bonds. This is tough to imagine. We know consumers will be pinched, but given a choice between paying their taxes (and avoiding jail) versus buying goods from corporate America, I think its obvious which way people will go.
So why not sell protection on the MCDX versus a half as large long protection position in the CDX IG? The trade would be slightly positive carry, and your only bet would be that losses among the 50 municipals are less than half of the 125 corporates in the CDX IG. Or if recovery is similar to historic norms, then merely that municipal defaults will be about the same as corporate defaults. Whatever your view of the economy, this should be a relatively easy trade.
I think at some point, arbitragers will put this trade on, and it will expose a lack of deep liquidity in the contract. Talking to various traders, it looks like much of the trading in the MCDX has been macro hedgers, not betting on munis in particular, but using municipals as a means of hedging against a disaster event. But at current levels, the hedge is too expensive, and given a little positive momentum, it will be exposed as such.
The current 5-year MCDX spread is 86.25bps. This means that in order to buy $10 million in default protection against the 50 names in the MCDX, investors make the equivalent of $86,250 in annual payments, assuming a new contract were created today.
Historically, defaults on municipal credits have been very rare, particularly in comparison to corporate credits. According to Moody's, there was only 1 default of a "general obligation" (GO) municipal (meaning a state or local government with taxing authority) from 1970 to 2006, and that default was cured in 15 days. Among investment-grade, non-GO municipal bonds, only 0.29% defaulted within 10-years of issuance, according to Moody's. This compares with 2.09% of investment-grade corporate bonds.
However, the weak economy generally, and declining property tax collections specifically, could result in unusual pressure on municipal credits. Yet, the particular credits within the MCDX are among the safest in the market. Property taxes are usually collected by states, but distributed primarily to cities, counties and school districts. While other revenue sources, from income and sales taxes to toll collections are likely to be impacted by the current economy, those revenues are likely to follow a more typical recessionary pattern, which municipalities have weathered in the past. The MCDX includes only 4 cities (New York, Los Angeles, Phoenix, and Columbus, OH), one county (Clark County, NV), and one school district (Los Angeles).
So perhaps municipal default rates will rise in the future. But wouldn't we expect corporate default rates to rise as well? Compare the MCDX with the CDX IG, an index of 125 investment-grade corporate credits. The CDX IG closed on Wednesday at a spread of 144bps.
Using standard recovery assumptions for both indices (80% for munis, 40% for corporates), one can calculate the expected default rate based on each index's current spread. The graph below shows the cumulative expected default rate based on current spreads for both indices.
As you can see, the MCDX is trading at levels that imply nearly 20% of the municipals in the index will default within 5 years. Again, it seems likely that municipal credits will be more stressed than in years past, but given that the credits within the index are only mildly exposed to property taxes, a 20% default rate seems unlikely.
In addition, because of the stronger recovery expected in municipals, the current spread levels imply greater default levels for municipals than investment-grade corporate bonds. This is tough to imagine. We know consumers will be pinched, but given a choice between paying their taxes (and avoiding jail) versus buying goods from corporate America, I think its obvious which way people will go.
So why not sell protection on the MCDX versus a half as large long protection position in the CDX IG? The trade would be slightly positive carry, and your only bet would be that losses among the 50 municipals are less than half of the 125 corporates in the CDX IG. Or if recovery is similar to historic norms, then merely that municipal defaults will be about the same as corporate defaults. Whatever your view of the economy, this should be a relatively easy trade.
I think at some point, arbitragers will put this trade on, and it will expose a lack of deep liquidity in the contract. Talking to various traders, it looks like much of the trading in the MCDX has been macro hedgers, not betting on munis in particular, but using municipals as a means of hedging against a disaster event. But at current levels, the hedge is too expensive, and given a little positive momentum, it will be exposed as such.
Like many indices/products that are created and never attract liquidity the MCDX is just fancy packaging of a weak idea...
ReplyDelete"... I think at some point, arbitragers will put this trade on, and it will expose a lack of deep liquidity in the contract..."
Yeahhh... you're right!
Great graph of default for CDX versus MCDX... cool...
I've heard several dealers pitching this trade, but I'm suspicious that they're just looking for someone to sell MCDX protection.
ReplyDeleteOne thing I haven't gotten a satisfactory explanation of is how the tax-free nature nature of munis affects the basis trade. I would think that on the long side, there would be no issue. But it seems like it would be unprofitable to short bonds and sell CDS to arbitrage CDS spreads that are too wide.
The technicals are also worrisome if a lot of people set this trade up. In the last few months, there was a similar trade in LCDX vs HY as LCDX priced in extremely high loan defaults. Eventually the relationship came back into line, but LCDX has underperformed again as people look to unwind their trades. And I think LCDX is liquid compared to MCDX...
cate I agree 100%
ReplyDelete"a choice between paying their taxes (and avoiding jail) versus buying goods from corporate America"
ReplyDeleteSkipping property taxes will not land you in jail, but it will result in a lien, which won't matter much to an "home owner" if the mortgage is upside down and in default. Property taxes never stop accruing and that results in further downward price pressure on the seller - the bank.
GO bonds get paid by an amount levied against the property tax base. A falling total assessed evaluation simply raises the tax rate automatically. Short of a victory by the devil at Armageddon, it is hard to conceive of GO's issued by most municipalities defaulting. Pay raises will stop, construction will stop, soe cops and firemen will be laid off before the bonds go into default.
The problem is that anytime congress is in session no man's life liberty or property is safe.
"But it seems like it would be unprofitable to short bonds and sell CDS to arbitrage CDS spreads that are too wide."
ReplyDeleteOne cannot short a municipal bond...
One cannot short a municipal bond...
ReplyDeleteBut there are many muni etf's that can be.
Correct.
ReplyDeleteAlso some synthetic options like an MMD Rate Lock...
Another problem with the MCDX is that it hasn't been acting like the cash muni market. The MCDX has drifted wider consistently since inception, where muni/tsy ratio has moved up and down. So the obvious user of the MCDX (dealers hedging their muni book) is getting a bad hedge.
ReplyDeleteI expect people will start putting on shorts on the MCDX, which will expose the poor liquidity in the contract and eventually cause them to discontinue creating new series.
I would be willing to bet that there are very few dealers using MCDX to hedge their muni book. While they're may be some proprietary dealer books using the vehicle as a partial portfolio hedge, there are definitely no dealer flow desks, institutional or retail, using this vehicle solely because of its day-to-day inefficiency in tracking muni cash, which has to be marked to market daily and therefore creates a horrific basis inefficiency (similar to using IR Swaps[BMA or LIBOR] during the recent credit crisis).
ReplyDeleteI would contend that a majority of the players in MCDX are crossover, or non-typical muni market players, institutions looking to play the muni market from a macro perspective, expecting municipal tax receipts to fall precipitously with the current housing and economic environment. These same institutions will be the first ones to reverse their bets and short MCDX to oblivion.
AI - I too would be very interested in seeing some form of liquidity data on this product.
I think when the MCDX was conceived, the thought was longs would be hedgers and shorts would be non-tax paying entities that wanted access to the muni market. But its clear that shorts haven't been in the market. And whether or not anyone is actually using it as a hedge, it isn't working, and its getting damn expensive.
ReplyDelete