Tranched EFSF – Or TARP Lite

Posted by on Nov 8, 2011 in Uncategorized | No Comments

The EU is getting closer to having two actual alternatives for EFSF on the table.

Partial Protection Certificate (PPC)

The PPC approach is included as one of the two proposals.  They came up with a name so that is progress.  It looks like payout would be linked to events that are very similar to CDS Credit Events.  It seems that the PPC’s would be issued in conjunction with a bond issue.  The goal would be to reduce the coupon on that issue.  They got rid of all the awkward mechanics of EFSF issuing bonds to be used as collateral – but we assumed that would never happen since it made no sense.

Everything else seems still up for debate – detachable or not, what the actual payout definitions are, the form of the payout (immediate or regularly scheduled), etc.

They make it clear that you would need to own bonds in order to collect on the PPC if they are detachable.  I assume they will change that to something where you need to own the bonds if you own the PPC, otherwise “speculators” will buy the PPC as a short and only buy the bonds when they go to collect.

The First-Loss EFSF detailed analysis from last week still applies <http://www.tfmarketadvisors.com/2011/11/03/the-efsf-first-loss-insurance-is-a-truly-unique-vehicle/> , but my sense is the EU is backing away from this proposal since it wasn’t that well received and offers the least flexibility.

Co-Investment Funds (CIF)

I think CIF’s are the way they are heading, partly because they are extremely flexible and partly because the PPC didn’t seem to do well when it was the only option on the table.

Multiple CIF’s could be established.  Each would be able to buy in the primary or secondary.  Each could be for one country or for multiple countries.  Each could have its own bespoke “tranching” to suit each investor’s needs.  It is clear that they are trying to create something that is a cross between a CDO and TALF. I think it has a better chance of succeeding than the PPC does, but there are still a lot of issues to be addressed.

Each CIF would have some standard features.  It would have “nominal equity capital”.  This is usually a very small sliver of risk, often sold to charitable trusts, so that the rest of the capital structure can be considered debt, and only the small charitable trust has to consolidate the assets of the CIF.

The EFSF would provide the first loss risk (or at least a portion of it).  With this structure, the EFSF would have to deliver cash into the CIF.  It couldn’t work based on just guarantees.  Whether it is more efficient for member states to issue bonds and deliver cash into the CIF or to have the EFSF issue bonds and deliver the cash into the CIF is yet to be determined.  I suspect the cost would be lower if individual countries did it, but that would force those countries to recognize debt, and remove the fiction that all the countries are participating in the EFSF.  In the end, the EFSF will likely issue bonds to fund their portion of the CIF’s because the accounting and headlines sound better and that will offset the extra costs.

How appealing will the CIF be to investors?  What can be done to make it attractive?

Tranching a Single Asset – The problem of low expected recoveries

Creating tranches for a single asset is difficult.  The only way to create value is to create a “senior” tranche that should be covered in event of default.  Basically the risky tranche(s) need to cover down to a likely recovery rate.

Mortgages are one example.  In a “normal” mortgage market, banks liked to lend when the homeowner had a 20% first loss.  The homeowner could achieve that by putting down more equity or buying insurance from PMI (yes they are now in default).  It was the fact that banks had a buffer on home value that made them comfortable lending at cheap rates.

The same principle applies to corporate credit.  By pledging collateral against certain debt, the company can achieve a lower cost of capital, because that investor can look to the assets in the event of default.  The more fungible the assets pledged, the better.

The issue with a single sovereign credit (or unsecured credit in general) is that the expected recovery in event of default will be 40% or lower.  An investor would need at least 60% first loss protection before they would consider themselves isolated from the risk of default.  The reality is, they would probably want to be protected down to a recovery of 30% if not 20% (the rating agencies would likely assume a 20% recovery when providing ratings).

Let’s assume that with enough political pressure and a bit of greed, that a “senior” investor could be found.  A realistic assumption is they would want 70% first loss, so they are providing 30% of the funds to the CIF.  I think they would want to earn what they could make by buying French debt.  So in 5 years, it would be 2%.  German 5 year debt trades at 1%, but I think with only 70% subordination on a single issuer (Spain or Italy) the investor would consider it reasonably safe, but wouldn’t take the risk for less yield than they could get by investing in France.  “Mid-swaps” for 5 years are at 1.90% so in line with the thinking that decently safe investments in Europe would yield 2% for 5 years.

If the goal of the structure is to get a 4.5% cost of funds for 5 years to the issuer, which seems a reasonable goal as it would balance a reasonably low coupon with a decent tenor.  The EFSF will likely want to earn their own cost of funds on their first loss portion.  If they have negative carry, it would complicate things for the EFSF as they would have potential capital calls in the future and they would have to restrict how much debt they could issue now, to account for those future capital calls.  In the end it is just so much simpler for them to charge the CIF their cost of funds, that I think that is what they would do.  Let’s assume they issue where France is trading (mid-swaps + 10) for a cost of 2%.  This is a bit aggressive versus where their 10 year deal priced the other day, but I’m trying to make CIF’s work.

So CIF’s would have to pay 2% on 55% of its capital structure (the first loss is heavily subsidized by the EFSF and the senior debt is priced fairly).  So for every €100 of debt that the CIF buys yielding 4.5%, it would earn €4.50.  It would pay out a total of €1.10 for the first loss and senior debt.  That would leave €3.40 for the €45 of “mezz” debt.  That is a yield of 7.6%.  The mezz debt would need to trade at 7.6% for this strategy to work.

Is that a “good” price for that risk?  It clearly depends on where the underlying bonds trade outright.  If the straight debt is trading at 5% is this good risk/reward?  What if the straight debt is at 6.5%?

It will depend on your assumptions about default and recovery.  If the straight debt was yielding 7.6%, the “mezz” tranche would be a bad investment.  You would get the same return in a no default scenario, but likely do better under any reasonable recovery assumption.  The breakeven recovery would be about 54%.  If you bought outright debt, you would lose 46% of your investment.  If you bought the “mezz” tranche, you would lose 21 points (75 – 54) because the first loss tranche would absorb the first 25 points.  That would be 46.7% of your investment.  A recovery about 54% would favor the “mezz” investment, but reasonable recovery assumptions (40% and below) favor an outright investment.

When you factor in a lack of liquidity and lack of control of the CIF, the yield required to make the mezz attractive has to be even higher.  So, as “straight debt” yields increase, or the cost of capital of senior debt or EFSF debt increases, the ability to get a “good” coupon to the issuing country decreases.   The “mezz” tranche only offers attractive yields to outside investors when the underlying bonds don’t need the structure – why would you go through all of this if you could issue bonds at 5%?  You wouldn’t.

This analysis also ignores initial price of the bonds.  If the CIF is buying “discount” bonds that is helpful, but then it does nothing for the new issues and the roll risk countries are facing.  Investors can buy Italian bonds in the 4-6 year maturity range at 90% of par in many cases.  That already provides a “first loss” protection of 10%.  How exactly the market will respond is unknown, but the more bonds that trade at a discount in the secondary market, the less useful the “first loss” protection is.  The CIF’s could be structured to focus on secondary market discount bonds, but that may do little for the new issue market which is clearly what needs to be helped the most.

The structure is more compelling than the PPC’s but it is hard to determine whether a real senior note provider will step up and whether the yields would be compelling to a “mezz” investor relative to an outright bond investment.  To the extent the participants in CIF’s are just shifting how they purchase Italian or Spanish bonds, then most of the perceived benefit will be reduced by this cannibalization.

Using Multiple Assets To Increase Diversification

Having more than one asset in a CIF would help.  There is some probability that the different issuers won’t default at the same time.  So let’s use Spain and Italy together.  The ratio should be about 3:1 which is about in line with their debt outstanding.  The CIF would own €25 of Spanish bonds and €75 of Italian bonds.

The senior investor has more protection now as both countries would have to default and both would have to have the same low recovery rate.  In the real world, an investor might not give much benefit to the idea that if Italy defaults Spain wouldn’t, or that both wouldn’t have low recoveries.  But since we live in a government pressured market, let’s assume that the senior investor now only demands 60% of first loss protection (they are willing to reduce their subordination by 10%).  I think this is a stretch, but if it happens, it has a big impact on the “mezz” tranche.

Now the cost of first loss and senior debt is €1.30 (2% on 65).  Since the mezz tranche is now only €35, its yield would be 9.1%, a significant improvement.  If somehow only 1 country defaulted (especially if it was Spain) then the “mezz” tranche wouldn’t have a loss.  These numbers are more compelling, but hinge on finding the “senior investor”.  I don’t think there is a real world investor who takes that risk at that price.  I would much rather own French bonds at 2% than 60% subordination risk on a pool of 25% Spanish bonds and 75% Italian bonds.  I don’t think even BRIC’s would be keen on that risk.  They will buy the EFSF bonds to fund the CIF’s but the second loss portion seems too risky on those terms.  The only likely home for those senior bonds (on these terms) would be a government controlled entity.  Either the ECB or some French or German pension funds are the likely “stuffee”.

A couple of months ago, I would have doubted that the leaders of these countries would purposely saddle their pension plans with bad investments.  That they wouldn’t risk more of their citizens future on a plan to keep together a debt riddled zone of countries that is getting worse rather than better.  Now, I have to admit that not only would they be willing to do it, but they might be eager to do it, to “prove” that their plan worked.

CIFs Are Better Than PPCs But Still No TALF

CIFs seem to have a better chance of working, though they will require not only cheap EFSF money at the first loss part of the capital structure, but also some “dumb” money at the senior part of the capital structure.  If they get enough of that, they can create some compelling value for “mezz” investors.  These would then be Wall Street firms or hedge funds (angel investors when buying CIF’s, evil speculators when buying CDS to short).  It would be a big shift of funds from taxpayers to money managers, but that is about par for the course.  The value of the mezz relative to an outright investment decreases as yields on the underlying go up.  This makes the deal easy to sell when underlying yields are under 5%, but not much help as yields go well above 6%.  So it will be trillions in could times and “crickets” in bad times.  Not exactly what you want out of a bailout fund.  They will have to be careful, once again, to conserve their limited resources.  Using too much up front, in relatively stable times, could be a big problem down the road.

This is not TALF.  TALF was a much better deal for outside investors.  The range of assets the investor could choose from was broad.  Most fund managers believed they were “cheap” but couldn’t come up with the capital to invest, or handle the downside.  The assets themselves were good for “tranching” in that they were diversified pools.  In TALF, the government (fed/treasury) provided both the senior financing and the co-invested first loss financing.  TALF was a great opportunity.  CIF’s may create some interesting opportunities, and are at the very least flexible enough, that investors could have a discussion, but they are nowhere near as appealing as TALF was.