Description

At 62 cents per share, Tronox "B" shares (TRXBQ) are a compelling risk-adjusted return opportunity that should play out over the next few months.  At current run-rates, the equity is trading at less than one times free cash flow and the enterprise is trading at about four times EBITDA.  Putting even a modest six multiple on run-rate EBITDA ($175mm), as compared with an average multiple of between nine and ten for its comparables, yields a stock price of about $8.00 per share, or over 12 times the current price.  Unfortunately, the Stockholder currently faces enormous obstacles to actually realizing this value because Tronox is in the process of emerging from bankruptcy and, without a deep-pocketed champion to help cash out the senior claims in the capital structure, recovery depends more on negotiating with the Debtor and Unsecured Creditors than on the intrinsic value of the assets.  Nonetheless, intrinsic value should have a significant if not central role in determining the outcome of this negotiation, and even very unfavorable results from a fairness perspective should offer significant upside from 62 cents per share.

Note that this write-up describes a bankruptcy situation focused on the lowest, most levered position in the capital structure.  Investment should therefore be sized with eyes wide open, factoring in the possibility that there may be no distribution to shareholders upon emergence.  This write-up is in no way intended to advocate, solicit, encourage or discourage votes for any potential plan of reorganization for the company. 

Tronox is no stranger to the VIC - the stock was recommended at $17.00 per share and at $7.00 per share, and the bonds were recently written up at 112% of par.  Please review the VIC write-ups from April 3, 2006, December 19, 2007, and April 9, 2010 for a comprehensive discussion of the company, its titanium dioxide production and distribution businesses and the non-core assets it owns.  I will focus my discussion here on the process to emergence in front of us, a comparable valuation analysis supporting the view that the intrinsic value of Tronox's equity is very large in relation to its current trading price, and an expected value analysis based on the probabilistic outcomes of how value will be distributed upon completion of the stakeholder negotiations.

Bankruptcy Considerations and Timeline

Within the next few weeks, the company will unveil its Plan of Reorganization (the "POR"), file its Disclosure Statement, and begin the solicitation process for votes in its favor.  The company has until July 12^th to file the POR, and I'm guessing that emergence will be targeted for October.  The framework for this plan is already known, as a Plan Support Agreement (the "PSA") was filed on the docket in late December 2009.  This framework anticipates a bondholder backstopped rights offering of $105mm to prepay certain environmental claims, equitization of existing unsecured debt, and canceling the equity with no recovery. 

Because the industry and the company's profitability had been improving rather dramatically up to the filing of the PSA, and things have continued on a nice trajectory since then, some recovery to the equity is highly likely.  We can keep a pulse on the company's performance by accessing the monthly operating report (the "MOR") it files on or near the 21^st of each month on Pacer and also under the "Court Documents" section on http://www.tronox.com/reorganization/index.htm. 

If the Equity Committee maneuvers correctly, recovery to the stock could be very substantial.  The Debtor has made references on the docket about a desire to work with the Equity Committee to craft a "fully consensual" plan with the support of all stakeholders.  While I have no insight into how these discussions are proceeding, an analysis of the value of the company's assets (presented below) suggests that there is a very strong argument for significant equity recovery.  Further, the Equity Committee has two sophisticated institutional investors within its ranks that are highly familiar with bankruptcy situations, which should be very helpful in driving the outcome toward a fairer treatment for equity.  Note that while this write-up ultimately estimates an expected recovery based on a probabilistic estimate of negotiated outcomes, I personally believe the rightful recovery for Tronox's shareholders would be for them to maintain their full pre-petition ownership of Tronox's assets.

Tronox's bonds are currently trading near par, but significant volume has traded above 120 cents on the dollar before the most recent correction.  In my experience, it is very rare for the bankruptcy process to offer no recovery to a class of claimants (or interest holders) when the market has priced in an above par recovery to a widely traded public security just ahead of it in the capital structure.  Recent examples include the equity of Smurfit Stone Container (SSCCQ) and Chemtura (CEMJQ).  In the case of Smurfit, the bonds were trading in the low-80s and the equity (SSCCQ), which had no official committee to represent it, has been given a recovery currently valued in the market at about $55mm.  In Chemtura (CEMJQ), the bonds have been trading between 95 and 116 and the equity is being offered a package of equity and rights currently valued in the market at $170mm (though here additional recovery may be in the offing as the equity committee potentially delivers a more advantageous alternative plan).

I believe the company will offer a minimal recovery to Tronox's pre-petition shareholders in its initial POR, which will be filed within the next two weeks.  Debtor's management teams have fairly strong incentives to set the bar low for the value of an enterprise just prior to emergence for a variety of reasons, including the fact that they would rather have their option and/or equity grants struck at lower valuations and their performance expectations similarly set at lower levels.  Release of the POR may offer a better opportunity to purchase the stock than there is now, but it is difficult to say whether this expectation has already been fully factored in with the stock down to 62 cents from $1.40 in mid-May. 

Once the initial POR is filed, it will be incumbent on the equity holders (led by the Official Equity Committee) to spend the balance of July and August forcing out a higher recovery.  Some of the methods they might employ include:  1) coming up with the money to pay some or all of the pre-petition claims, 2) using litigious tactics to delay or invalidate the POR and/or Disclosure Statement, or 3) exercising "moral suasion" on the stakeholders in general and the Debtor in particular, whose fiduciary duty to the equity increases as solvency becomes more apparent.  While #1 would likely be the most effective means of getting equity its fair due, a combination of ALL of these would hopefully be deployed toward this goal.  The ability to use ANY of these depends upon valuation, as #1 requires value to be had if anyone is going to put up money, #2 likely involves a valuation fight before the judge, and #3 involves convincing the Debtor and possibly other stakeholders that valuation supports solvency and an equity recovery.  The following valuation analysis strongly supports the view that the company is solvent and that the equity has substantial value.

Valuation Analysis

Here I use more conservative assumptions than those advanced by Vincent in his write-up on April 9, 2010.  Where Vincent used a $175mm normalized EBITDA estimate for the company's productive assets and a $540mm claims pool, I am using $160mm in EBITDA plus I add an extra $60mm to the claims pool for post-petition interest and other liabilities for additional conservatism.  I should mention that run-rate levels actually support Vincent's $175mm EBITDA estimate. 

Where we largely differ is in accounting for the very significant level of excess working capital at the Debtor.  In fact, none of the research reports I have read regarding this situation make any adjustment for, or even mention, this anomaly.  In my conversations with various parties observing this case, I have not heard a reasonable explanation for why this massive value driver should be ignored.  This working capital adjustment is discussed in detail later in this section.

I have selected Kronos Worldwide Inc. ("Kronos", ticker "KRO"), Huntsman Corporation ("Huntsman", ticker "HUN") and E.I. du Pont de Nemours & Company ("Dupont", ticker "DD") as the most obvious valuation comparables.  Of the six major players in the titanium dioxide manufacturing and distribution business, these are the three with easily referenced public market valuations. 

Because Huntsman and Dupont are much more diversified, Kronos is the only available comparable that is a "pure play" titanium dioxide producer whose productive capacity is directly measurable against Tronox.  Kronos is therefore the only one that we can use to estimate the value of Tronox's capacity.  Kronos has the added benefit of being approximately the same size as Tronox in terms of capacity and revenue.  Furthermore, Kronos' chairman has recently been purchasing Kronos stock on the open market.  We can reasonably expect that the market valuation of Kronos is fair to conservative, since presumably a highly informed insider would use his own money to purchase shares at prices below or up to the company's fair value, but not above it.  As presented in detail later in this section, Kronos' annual capacity is currently valued at $3,040 per metric ton.  Applying this valuation to Tronox's capacity implies an equity value of $402mm, or $9.80 per common share.

We can also derive Tronox's value by estimating what multiple of cash flow it should be worth.  As shown later in this section, the "headline" enterprise values of Huntsman, Dupont and Kronos are currently trading at 7.7 times, 7.3 times and 11.6 times their respective estimated 2010 EBITDA generation.  The average of these multiples is 8.9.  Further, their "apples to apples", "adjusted" enterprises are currently valued at 8.4, 8.9, and 12.4 multiples of 2010 EBITDA, or 9.9 times on average.  Applying a more conservative 7.0 multiple to Tronox's estimated 2010 EBITDA of $160mm implies an equity value of $397 million or $9.68 per share, while using a lower, 6.0 multiple still gives a valuation of $237mm, or over $5.78 per share.

The following discussion details the steps used to compute the estimates above.  I also include additional insight as to why these estimates are likely very conservative.

A.  Working Capital Adjustments   

An accurate estimate of Tronox's value requires a baseline "apples to apples" measure against its peer group.  The most glaring adjustment to be made in this respect relates to the extremely large amount of excess working capital (i.e., current assets less current liabilities) that Tronox currently carries relative to the selected comparables.

It is not unusual for a company operating under Chapter 11 to experience an initial jump in working capital as suppliers demand more rapid payment from the debtor and customers take extra time to remit their payments to allow for increased goods inspection against less certain warranties, or for other reasons.  However, because the Debtor was require to fully fund the replacement DIP it negotiated with Goldman Sachs, a secondary jump in Tronox's working capital levels has persisted since the December MOR.  Note that working capital consists of short-term, high turnover accounts, and any excess should be quickly converted to cash once operations normalize upon emergence. 

As shown below, the peer group tends to hold working capital at a fairly consistent operational level of about 102% of each company's most recently reported quarterly revenue.  The table also presents the adjustment to be added to or subtracted from each comparable's enterprise value to normalize it to reflect the value of the business as if it held industry average working capital levels.

The WC Adjustment line suggests, for example, that Huntsman is holding $4mm less working capital than would be expected based on its peers.  By increasing our estimate of Huntsman's implied enterprise value by $4mm, we are acknowledging the company should incur an extra $4mm of debt and purchase that small amount of additional working capital to maintain its earnings power.  

While the adjustments for the comparables in the above table are quite small, it is clear that Tronox currently holds working capital well in excess of the standards of its peer group.  At March 31, 2010, the Debtor held $623mm in working capital against $170mm of revenues reported for the first quarter of 2010.  It may be that this extreme amount is the result of inter-company accounts that should be offset, which would in any case reduce the liability burden of the company dollar-for-dollar as analyzed herein.  Under the industry standard of 102% of the revenue reported in the first quarter, the Debtor only needs $174mm of working capital for normal operations.  Note that this is somewhat validated by the fact that Huntsman only requested $300mm of working capital in the asset purchase agreement that defined its stalking horse bid for Tronox.  Since Huntsman at the time was not competing against other bidders, my presumption is that this represented a higher than necessary amount in favor of Huntsman. 

Altogether, this step in the analysis suggests that the Debtor currently holds, at a minimum, $323mm (based on Huntsman's bid), and reasonably $448mm (based on industry averages), of excess working capital that can be monetized and used to reduce its debt and other obligations and still operate normally.  Based on this, we conservatively estimate that $300mm should be added to the valuation we derive for the Debtor's business when determining the value of the Debtor as a whole.

B.  Other Adjustments

Since Tronox's reorganization must account for all of its legacy liabilities upon emergence, the valuation of its comparables must be adjusted to reflect their major unfunded future cash obligations.  Most prominent among these is the estimated underfunding of each company's pension and OPEB plans, as well as their environmental reclamation reserves.  These obligations are presented below.

C.  Capacity Valuation Estimate 

Kronos is our sole comparable for estimating Tronox's value using production capacity.  Kronos' enterprise value is presented below.

With the market valuing Kronos' 532,000 metric tons of capacity at $1.617 billion, we arrive at a capacity valuation estimate of about $3,040 per metric ton.  Tronox currently has 370,000 metric tons of operational capacity.  At $3,040 per metric ton, Tronox's operational capacity is worth $1.125 billion. 

This estimate has several degrees of embedded conservativism, since Tronox's facilities appear superior to Kronos' for technological and geographical reasons.  All of Tronox's operating facilities use the chloride manufacturing process, as opposed to only 75% of Kronos' capacity.  The chloride process typically has lower manufacturing costs than the sulfate process due to newer technology, higher yield, less waste, lower energy requirements and lower labor costs.  Further, about half of Kronos' production serves the slower growth European markets.  By contrast, less than 25% of Tronox's capacity distributes to Europe.  Tronox's Hamilton, Mississippi facility serves the Americas, including the burgeoning Latin American markets, while its Tiwest joint venture in Kwinana, Australia focuses on the much faster growing Asian markets.  Asian demand for titanium dioxide is expected to grow at a 7% average rate over the next decade, versus an estimated 1-2% growth rate in Europe.

Adding $300mm in monetizable excess working capital to the $1.125 billion estimate for the company's operational capacity yields a valuation estimate of $1.425 billion for the Debtor.  This implies that Tronox's equity value is equal to $402mm (or $9.80 per share) after subtracting $450mm in liabilities subject to compromise, $423mm in postpetition debt, and a $150mm funding requirement for environmental obligations.

This valuation estimate attributes zero value to the estate for its electrolytic and boron facility in Henderson, Nevada, its idled facility in Savannah, Georgia, and its facility in Uerdingen, Germany.  Given the company's limited disclosure on the status of these assets, I hesitate to estimate what, if any, value these could bring to the estate under an alternative plan.  I note, however, that Vincent's bond recommendation included a conservative $25mm estimated additional value for the just the excess land at Henderson.  Tronox also owns a 50% interest in Tiwest's mining operation, the value of which is not accounted for here.

D.  Cash Flow Valuation Estimate

 The following table presents the market valuation of Tronox's comparables as a multiple estimated 2010 EBITDA for each company:

On October 7, 2009, Tronox disclosed EBITDAR projections that trended from $130mm in 2010 to $179mm in 2013.  Since then, the industry has announced seven titanium dioxide price increases and Tronox's most recent MORs imply annualized EBITDA in excess of $175mm.  Assuming 2010 EBITDA actually comes in at $160mm, and applying a 7.0 multiple (versus the 8.9 "headline" comparable average and 9.9 "adjusted" comparable average estimated above), we arrive at a $1.12 billion enterprise value for the company's operating assets.  Adding $300mm in monetizable excess working capital yields a valuation estimate of $1.420 billion for the Debtor.  This represents an equity value for Tronox of $397mm, or $9.68 per common share, after subtracting $450mm in liabilities subject to compromise, $423mm in postpetition debt, and the $150mm funding requirement for environmental and other obligations. 

Again, this valuation estimate excludes any value for the Henderson, Savannah, and Uerdingen facilities.  Henderson and Savannah have apparently been given up to the EPA and CERCLA claimants under the PSA, but it's difficult to even guess what these might be worth if recaptured by the estate under an alternative plan.  We know that the effectively idled Savannah plant is generating negative EBITDA and that these negative numbers are flowing through the MORs, but the company provides no information on the actual amount of cash burn this represents (this does, however, imply that EBITDA should increase when and if Savannah is fully divested). 

Because of the company's terrible disclosure policy, it's difficult to say what cash flows from Tiwest are reflected in the MORs.  While it is known that Tronox markets all of Tiwest's production, it is unclear whether the margins from its mining operation are embedded in these income numbers or if they are retained at the nondebtor joint venture level.  In response to a written inquiry by a savvy shareholder, the company has agreed to provide information about the performance of its nondebtor subsidiaries, which it will hopefully provide in the next few weeks.

Finally, it is important to recognize that Tronox's assets are very unique.  High construction costs, difficult permitting, and long lead times represent enormous barriers to initiating any greenfield or brownfield projects.  Tronox is one of only six major producers in the global market for this product that has few, if any, substitutes.  In its most recent 10-K, Kronos states, "Worldwide capacity additions in the TiO₂ market resulting from construction of new plants require significant capital expenditures and substantial lead time (typically three to five years in our experience).  We are not aware of any TiO₂ plants currently under construction and we believe it is not likely that any new plants will be constructed in Europe or North America in the foreseeable future."

Expected Value Estimate

Given all this, I have outlined the following set of negotiation outcomes and assigned my best guess at the probability of each:

The expected value here is approximately $1.80 per share, or almost three times the current share price. 

Note that there is only a very small probability of losing money from the current share price.

Catalyst