Description
Sunterra announced late last Thursday that its CEO is on paid leave and its CFO resigned, at a time when its is delinquent in filing its 10-Q and looking for an auditor after dismissing the accounting firm Grant Thorton. The recent sell-off provides an opportunity to buy Sunterra at a 63% of adjusted book value, 73% of adjusted tangible book, an adjusted EV/EBITDA multiple of 4.1 and adjusted FCF yield of 12%.
The reason for making the adjustments stems from the unique nature of time-share operators and how GAAP accounting distorts certain aspects of its business.
Sunterra sells, finances and manages vacation time shares, which they call vacation interests. They went bankrupt several years ago because they were not paying attention to credit quality and the loans that they had financed and securitized blew up. New management was brought in and has been turning the operations around and putting them on sound footing.
At $7.77, there are 20.0 million shares outstanding, $19.3 million in cash and $364.2 million of debt, bringing total EV to $500 million (I will adjust it later). The debt consists of $161.3 million of secured credit facility, $106.4 million of on-balance-sheet securitized notes (no off b/s debt), $1.5 million of notes payable and $95.0 million of 3.75% converts due 3/29/2024. The converts were struck @ $16 and are putable in 2011. Book value is $9.65/sh, with tangible book at $8.20. However, there are two time-share accounting rules that understate book.
The first issue deals with mortgages on the balance sheet. In the last 10-Q as of 12/31/05, Sunterra had $244.5 million of mortgages and contract receivables, net of $48.3 million of allowances. That allowance is 16.5% of mortgages and contracts, but the actual charge-off runs around 3% - 4%. They wrote off $10.3 million of bad loans in FY05. The reason that the allowance is so high is that they are not allowed to assume any recovery on a defaulted loan. Even if you assume charge-offs double, there is still a $1.20/sh in “excess” allowance that could be added back to book value. The defaulted time-share unit gets reassigned to unsold inventory and, from p. 61 of the 2005 10-K, recovered vacation interests was $9.0 million. This brings us to the 2nd issue.
Unsold Vacation Interest, net, on the balance sheet represents the lower of fair value or cost. Cost typically runs at 25% to 30% of retail value. Management has indicated that cost runs about $2,800 per vacation interest (note that excludes land development) but that replacement cost is closer to $3,800 per vacation interest. Most of Sunterra’s resorts are mature, so any appreciation in the resort and undeveloped land is not reflected on the b/s. Assume the unsold inventory, including land, is worth 15% more, which is still below 50% retail value, and that adds $1.58/sh.
That gives us an adjusted book of $12.43/sh and tangible book of $10.98.
On a ttm basis, revenues were $428 million and EBITDA was $73 million. EBITDA is expected to go up this year because they had been spending about $11-$12 million per year for a European sales force and they are in the midst of drastically reducing it. I think $80-$82 million of EBITDA going forward is reasonable. Now I am going to adjust the EV for the debt. The line of credit and securitized debt are collateralized primarily by mortgages and contracts receivables. I have little doubt that Sunterra could sell these obligations at par. The line of credit is with Merrill Lynch and has a rate of 1-month LIBOR + 1.50%, which was lowered last July from LIBOR + 2.25%. The average rate that Sunterra is receiving is 13.9% with the typical loan averaging between 7 to 10 years. 7% spreads are not too shabby. The financing rates charged by Sunterra have been fairly steady, so they are not really influenced by rising interest rates.
Getting back to adjusted EV, assume that they sell the net amount of mortgages and reduce secured debt accordingly, EV is now $255 million. I will adjust EBITDA by subtracting $17 million (7% spread X $244.5 million of debt/loans) to get $63 million. Thus, Sunterra is trading at an adjusted EV/EBITDA for 2006 of 4.1, which includes management and rental fees that deserve at least an EV/EBITDA multiple of 10 (see what multiples HOT, MAR, HLT and CHH trade at).
Free cash flow generation now is very powerful. Using EBITDA minus maintenance cap ex, I get $62 million or an amazing 40% yield. I derive that from $73 million of ttm EBITDA and $11 million of cap ex (they spent $11.5 million in FY05 and expect to spend in $11.0 million in FY06). I will add a caveat as to the number. Time-share accounting follows the old home builder accounting, where building costs are run through the income statement as they are incurred. This distorts things because they look lousy when building new resorts and fantastic when done. It does make it easy to determine maintenance cap ex, though, which is the entire cap ex line on the cash flow statement.
An important adjustment has to be made to FCF for the fact that Sunterra has to build or buy replacement vacation interests. Fortunately, there is a very good proxy for replacement costs right in the income statement: vacation interest cost of sales. Subtracting the $44 million of FY05 vacation interest cost of sales brings us to FCF of $18 million, for a 12% FCF yield.
I will note that there are three phases for how Sunterra makes money off a resort. There is the initial sale of a vacation interest. Then they earn interest from financing the sale. Finally, they derive an annuity-like income stream from rental and management fees for running the resort, which totaled $68 million in FY05. The beauty of them shuttering the European sales force is that they are still making money without having to sell any more vacation interests. I have heard that they were offered $70 million for the European operations in 2001 during bankruptcy and it was rejected by the creditors. Rumor has it that the recent interest in the European operation was for around $80 million, and again, it was declined. Europe represented 22% of revenues and 38% of units.
The time-share industry is very fragmented, as evidenced by the fact that as small as Sunterra is, it is #2. Cendant’s Wyndham is #1. The industry fragmentation has allowed them to make selective acquisitions in recent years to replace inventory. Buying existing resorts is accretive for Sunterra. Someone else has already taken the hit, both in time and money, for the initial build-out and Sunterra can leverage its network, as they sell vacation points that allow access to a variety of resorts. Sunterra also has an additional 25% of potential vacation interests by adding units to existing resorts.
Having a large footprint that offers choices makes their vacation points system more attractive than a fixed location. They have 93 locations to choose from, of which 38 are located in Europe and 59 in America. They own 59 locations (34 Europe and 25 N.A.) and the remainder are affiliations. Sunterra claims 300,000 ownership interests. What separates Sunterra from two other publicly traded time-shares, SVL and BXG, is that Sunterra operates “fly to” resorts where the other two are “drive to” resorts. The typical Sunterra consumer is a boomer with an annual salary of $80K. For some good info, check out the Wyndham slide show:
http://www.wyndhamworldwide.com/documents/WYN_Roadshow_Presentation.ppt#2
By the way, Wyndham operates 140 time-share resorts and claims 750,000 ownership interests, but is 95% located in the US.
So why is Sunterra so cheap?
They have had a series of events occur that look bad when strung together, and even worse via poor communication.
It is not that they are poor operators – they have made the right moves – but they apparently subscribe to the Courtney Love School of Publicity. Ready, fire, aim.
The main controversy involves the European operations, specifically Spain. In December, a former employee made allegations of accounting irregularities. The board of directors was made aware of the issue, but there is some debate as to whether the auditor, Grant Thorton, was notified. What is clear is that Sunterra subsequently paid $3.1 million to Spanish authorities related to withholding taxes and reserved $0.9 million for potential penalties. Since the error with the withholding took place over several years, they will have to restate financials back to 2002.
The board then made a decision to dismiss Grant Thorton, not because of the above incident, but because they did not feel that Grant Thorton was handling the intricacies of time-share accounting well. A lot of it relates to SFAS No. 152, which had a significant impact upon the time-share industry. Unfortunately, the timing sucked.
In the press release announcing that Grant Thorton had been dismissed, Sunterra said there were no disagreements and this is where the he said/she said issue arose. The board’s Audit and Compliance Committee is investigating the incident. The board is also in the process of searching for a new auditor.
On June 22, 2006, the company issued a press release which the news sources headlined by “Sunterra puts CEO on paid leave; CFO resigns.” While the CEO, Nicolas Benson, would take a paid administrative leave while the investigation continued, he would remain on the board. What the headline failed to mention was that the CFO, Steven West, had accepted a job with another company. Board member James Weissenborn was appointed as interim CEO. Weissenborn is a managing partner of Mackinac Partners, a consulting firm that is currently providing management services to Sunterra’s European operations. The current controller and CAO, Robert Krawcyzk, was promoted to CFO. I don’t see this as a major impact to operations. When the stock tanked the next day, management announced an investor conference call to discuss matters and set the date for Wednesday June 28th.
All of this news looks bad, but if you parse it out, the accounting issue is relatively small and has largely been addressed. Putting the CEO on paid leave makes sense because it should allow the Audit and Compliance Committee to make a thorough and independent investigation. It sounds like the board is doing the right thing. The fact that they replaced the CFO with the CAO indicates that the board is comfortable with the accounting and this will speed the ultimate completion of financials. Being four years removed from bankruptcy also lessens the chance for fraud.
Valuation: If you value the time-share business at an adjusted EV/EBITDA multiple of 6.0, that business is worth $13.70/sh. The quick and dirty evaluation of the finance business is that they earn $20 million net interest, minus $12 million of charge-offs, minus $2 million SG&A (just a guess), 40% taxes and a 12 P/E multiple comes to $2.15/sh. I think this is a $15-$16 stock.
Additional resource: Michael Axon at CRT Capital Group. His analysis of Sunterra was very helpful in peeling back the layers of this onion.
Risks:
- levered to the consumer
- dilution from converts, warrants and stock options
- more p.r. miscues
- delisting and further delays in filing financials
Catalyst
- selling below tangible book
- accounting issues are overblown
- resolution of investigation
- hiring an auditor
- filing of financials
- clarification from June 28, 2006 conference call
- forced selling ends from Russell 2000 deletion