BofA/ML's Jeffrey Rosenberg proves once again why he is one of the best credit analysts on Wall Street. Two months ago, the Bank of American put together a basket of potential LBO names which included Pactiv, Lexmark, Lubrizol, US Cellular, and Harris Corp, duly noted on Zero Hedge. He also proposed ways to play these names, focusing on various CDS strategies, of which by far the simplest one was to buy outright naked CDS on the names. Sure enough, this week Pactiv blew out, on rumors of an Apollo LBO (we hope for the sake of Pactiv's employees, not to mention Calpers, that the deal never materializes) and the names in the basket have widened by 121%-257% since inception. For those who followed Rosenberg's advice and made a 20x annualized return on the recommendation, congratulations. Sure enough, the trade is now closed. Additionally, after it was noted that a consortium of private equity firms was likely to acquire Fidelity Information Services, Rosenberg noted on May 7 that the deal is unlikely to materialize. Subsequent to his note, the deal has now fallen apart. This week, Rosenberg provides an updated LBO basket, as well as several strategies on how to play these, either outright or as pair trade. We are confident that with liquidity soon to become overabundant yet again, that these specific LBO names are set to see their credit spreads blow out as usual.
Assessment of LBO Risk
The Rosenberg approach to quantifying LBO risk is very simplistic: he uses a "feasibility score" based on a benchmark relative value of 6x Pro Forma leverage and 2x Pro Forma Interest Coverage, typical LBO outcomes post new equity infusion. The actual formula is = 6/Pro Forma leverage + pro forma coverage/2. The higher the result pro forma for a potential LBO, the more likely the name to be taken private. To keep the analysis as simple as possible, IRR considerations are completely ignored as there is so much idiot money out there, that one man's meat is certainly another man's poison (typically the latter "man" will have 0% discount window access to "other people's money" so the cost/benefit analysis is virtually irrelevant). The base-case assumptions are as follows.
Our approach to identifying potential LBO candidates focuses on determining which companies can sustain the higher leverage of a typical LBO, in addition to size restrictions. After making assumptions regarding the relative size of the equity contribution in a potential buyout (40%), takeover premium (30%), and expected earnings growth, we can calculate pro-forma fundamentals and compare to those in past LBO transactions. Based on historical averages for large LBOs the typical target becomes leveraged 6x (Debt / EBITDA) and has a coverage ratio of 2x (EBITDA / Interest Expense) following an LBO transaction. We say the company is a feasible LBO candidate if pro-forma credit fundamentals are similar or better than the past successful deals.
We have put together a simple spreadsheet version of the model that relies on Bloomberg data feeds, and can be used to quickly review a given LBO candidate based on our standard assumptions – these can be overwritten as we encourage clients to use their own assumptions. The spreadsheet is available upon request.
We show the results of the spreadsheet model for Pactiv (ticker PTV) in Figure 25. Notice that the pro-forma EV of the trade is $6bn, and the required amount of incremental debt is estimated at close to $2bn.
The final model output is the feasibility score which is based on the pro-forma leverage and coverage ratios. A score of 2.0 or above indicates that an LBO is feasible for the company based on our assumptions. The score for PTV is 3.3, which is substantially above our threshold for LBO feasibility.
Now that Pactiv is out the picture and profits on the original basket of LBO names have been taken, which names should investors focus on? Rosenberg is confident that despite the failure of the FIS LBO, there will be various other large name go-private deals soon announced. As such, he believes that while some of the names that have rallied alongside the initial basket, other names have been largely ignored, whose LBO feasibility score is in fact materially higher than those that have blown out. As such, an unhedged trade would include short LBO risk (buying CDS) in five larger names, specifically CBS, Clorox, Quest Diagnostics, Omnicom and Sara Lee, or alternatively hedging short LBO risk with long LBO risk in names that have already run up, in a convergence pair trade.To wit:
Following yesterday’s Pactiv news we closed out our two short credit risk LBO hedging strategies on smaller names. The apparent collapse of a deal for Fidelity National – what we estimate would have been a $15bn enterprise value (EV) LBO – highlights two isssues. First obviously the difficulties in implementing large LBOs under current market conditions but also clearly that the potential for larger LBOs exceeding $10bn in EV exists. To hedge against LBO risk among larger names with pro forma EV in the $10bn-$20bn range we buy CDS protection on a basket of names that appear on our LBO screen but are not influenced by significant LBO speculation
Took profits on original LBO strategies
Yesterday’s Pactiv news adds to a number of recent LBO events for names with pro forma enterprise value below $10bn, including IMS Health and IDC. We took profits on our two LBO hedging strategies in yesterday’s Situation Room. Recall that our first strategy was to buy CDS protection on a basket of LBO candidates. The second strategy hedged such short risk position in CDS by selling protection on an FTD written on the same basket. The net view expressed in the second trade is to effectively buy CDS protection on the second to fifth default. Accordingly this strategy profits as LBO risk increases broadly across the names in the basket (as opposed to more idiosyncratically for one name).
Our original LBO basket consisted of Lexmark International Inc, Pactiv Corp, Lubrizol Corp, Harris Corp and United States Cellular. CDS spreads on this basket have widened 87 bps to 168 bps since mid March (as of May 17th). Clearly, the names in the basket had a correlated move since inception, with 4 names including Pactiv having widened by 121%-257% of their starting spreads compared with only 25% for the investment grade market (CDX.IG). Accordingly, the FTD in the second trade priced at 60% correlation yesterday compared to 55% at inception. Given this increase in correlation (positive for our FTD) and 90 bps spread widening (negative for our FTD) we estimate that CDS protection on the FTD widened by 206 bps. Thus the net move in the second LBO hedging strategy is a spread widening of 57 bps on our net short position (for our trade we bought protection on $50mn CDS basket and sold protection on $10mn FTD, for the net notional of $40mn and net widening of ($50mn * 87bps - $10mn * 206bps) / $40mn = 57bps).
New LBO hedging strategies
The breakdown of talks for private equity to acquire Fidelity National Information (FIS), in what we estimate would have been a $15bn LBO and on the top-20 list of all time leading LBOs, suggests continued limited ability to complete larger deals. While we are unaware of the precise details, WSJ reported that FIS would not agree to the $32/share price proposed by Blackstone – a 23% premium compared with the close on May 5th. That premium is lower than the historical average of 30% seen in large LBOs suggesting that the economics of large LBOs may still be somewhat stretched. However, clearly the apparent near completion
of a deal suggests the potential for larger LBOs exceeding $10bn in EV.
Judging from the limited reaction to FIS we suggest that the market continues to overestimate the LBO feasibility of well publicized large LBO candidates with pro forma enterprise value of $10bn-$20bn, such as Computer Associates. Thus to hedge against LBO risk among larger names we look elsewhere in the universe of large candidates and select a basket of five names for which our quantitative LBO feasibility screen suggest an LBO is feasible, but that have widened only in line with general market weakness since mid-March. This basket consists of Omnicom Group, Sara Lee, Clorox, Quest Diagnostics and CBS. Our first new trade simply buys CDS protection on an equally weighted basket of these names at 94 bps ($5mm each name). Our target for this trade is a basket spread level of 170 bps while we use a stop loss at 60 bps.
Figure 16 shows our basket along with analysis using our LBO feasibility model – scores above 2.0 indicate that LBOs are feasible. Incremental debt and sponsor equity are also derived from our model. For full details see the section “Assessing LBO Risk” as well as Figure 26 below.
Here is the summary table for those who can't wait to put some virtually zero margin CDS trades on, and are bored of destroying Europe using weapons of mass financial destruction.
Cutting to Rosenberg's recommendation:
- We recommend buying protection on a basket of credits that are a) feasible LBO candidates based on our model, b) have proforma EV above $10bn, and c) have widened roughly in line with the market since the beginning of March.
- This basket consists of Omnicom Group, Sara Lee, Clorox, Quest Diagnostics and CBS, with $5mn notional for each name.
- The current spread on the basket is 94bps. We set the stop loss at 60bps and target of 170bps.
Investors could also trade on the lack of discrimination in the credit market between large and small LBO candidates. We see this below as both credit spreads and LBO feasibility scores for companies that fall within our feasible range vary little with pro forma enterprise values. In contrast, as we have argued above, although large deals certainly are possible there is limited capacity in the market for such deals. While we do not formally put on a trade that suggests selling CDS protection on a basket of larger names subject to significant LBO speculation and buying CDS protection on a basket of smaller well known LBO candidates. We show below a hypothetical example of such a trade using baskets of large LBO candidates. Obviously investors would want to carefully select names for this type of trade where one leg is long risk in LBO candidates, exposing the investor to significant downside risk. Alternatively the baskets could be diversified by including more names.
Figure 22 shows example baskets along with analysis using our LBO feasibility model – scores above 2.0 indicate that LBOs are feasible. Incremental debt and sponsor equity are also derived from our model.
Once again to summarize, here are the three proposed trades, which can either be put on together or seperately:
For readers who are unable to trade CDS, a possible analogous pair trade hedge is to buy the stocks of the companies in the short risk bucket, and to short the long-risk ones, on expectation of relative convergence and a possible 25% short- to mid-term IRR. While we have long-endorsed abstaining from indiscriminate market participation, when fundamentals suggest a reasonably favorable risk/return scenario, a trade certainly would make sense, such as the one proposed by Jeffrey Rosenberg, if for no other reason than his so far relatively impeccable track record to date. On the other hand, should another May 6 event occur, and/or should the Fed, for some ungodly reason, decide follow through with threats of liquidity withdrawal, any equity trade will likely see massive downside, which is why these trade suggestions certainly make far more sense for the CDS realm, as any adverse developments in the market will lead to a beta blow out in spreads independent of market fundamentals.