Gramercy Capital: management team
With respect to CDO buybacks, a wise man once told me that when you can buy your debt back cheap buy it because you understand your debt better than anybody else – Michael Ashner, Winthrop Realty Trust
THE MANAGEMENT TEAM
It seems that every time someone analyzes management teams it ends up being a pedigree review. I prefer instead to review track records and paper trails because they give a closer understanding of the players’ capabilities and motivations.
Not that Gramercy’s management does not have credentials. Just as an example, CEO Roger Cozzi was the former iStar Financial SFI chief investment officer, a Fortress Investment Group FIG managing director and a Goldman Sachs GS alumni.
However, a closer view of Gramercy’s actions shows survival skills that have been nothing short of amazing. Under the watch of the new management team that took over in October 2008, including among others CFO Jon Clark and President Tim O’Connor former COO of iStar Financial SFI, Gramercy promptly made several right moves. These actions preserved capital and reduced several hundred millions of recourse obligations and tens of millions of expenses while playing near the eye of the credit hurricane.
- Suspended dividends: for both the common and preferreds. Having just closed the AFR transaction it was tough and crashed the stock price. Painful but it was the right move and shows a team willing to face reality.
- Exchanged trust preferreds: Gramercy somehow persuaded the holders to exchange their shares for junior notes bearing 0.25% interest. Then came back to them in 2010 for a par-for-par exchange for CDO bonds that Gramercy had repurchased that were trading at 50% of par. The whole transaction saved at least $75 million in recourse obligations.
- KeyBank facility: Convinced a group of hedge funds which had partnered with KeyBank to accept accelerated repayment of just $60 million on a $175 million facility, for savings of at least $115 million. This was one of the most outstanding negotiations I saw during the crisis.
- Wachovia facility: First Gramercy managed to extend it to 2011 with covenants and recourse obligations waived. Then they paid it only half in cash and the rest in a mezzanine loan participation for a profit. Nice.
- JP Morgan facility: Taking advantage of the crisis, Gramercy negotiated with JPM to retire a $9 million facility for $2 million
- Eliminated management fees: As part of the disentanglement of the relation with SL Green SLG, Gramercy completed its formal management spinoff saving $15 million per year in fees. Most other comparable REITs like Newcastle Investments NCT, Arbor Realty Trust ABR and RAIT Financial RAS still have outside managers siphoning fees.
- Achieved SG&A reduction goals: The management team had a target of reducing corporate SG&A to $28 million and achieved it. This gives me confidence that they will do what is necessary if they lose Realty.
- Repurchased preferred shares: Tendered more that 20% of the preferred shares at a 50% discount to par value plus accumulated dividends at the end of 2010 saving $15 million in recourse obligations.
- Sold joint ventures at book value: Sold at book value several leaseholds to SL Green at book value, generating $89 million in Corporate unrestricted cash. Cash that can take Gramercy very far in the current CRE environment.
- Bought discounted AAA CMBS: One of the reasons that CDOs 2005 and 2006 were well defended, while being in the center of the hurricane, was the fantastic reinvestment in 2008 and 2009 of close to 15% of their assets in CMBSs, rated AAA at their origination, at very high discounts to par to improve the OC/par test. Those CMBSs price have recovered strongly recently.
- Bought back CDO bonds at large discounts: Most buybacks were closed by previous CEO Marc Holliday: $128 million par buybacks with the bulk of it in 2008 at 30 cents on the dollar. But the new team did his share of repurchases, under severe liquidity limitations, buying $61 million par at 45 cents on the dollar and was instrumental in using them used to eliminate the junior bonds.
Financial Alchemy indeed.
However we reach a point of contention: the AFR acquisition of 2008. It is very easy now to dismiss it as hubris; it drained cash and saddled Gramercy with debt at a critical point. But let’s consider a couple of things:
- Negotiated and closed by previous team: The AFR deal was announced in the fourth quarter of 2007 and closed mid 2008. Roger Cozzi became CEO in October of 2008 with the cards already dealt for him while Marc Holliday retrenched to his concurrent job as CEO of SL Green SLG.
- Partially paid in stock: and not only that, Gramercy’s stock at the time was a very overpriced stock at around 13-15x free cash flow. Valuable currency.
- Good price: and the price was very good. Bought at a discount to book value and a very reasonable to the buyer 8% cap rates. If not for the credit crisis they surely would had a chance to keep this stable income stream.
- Provided some cash flow: also the deal was closed with some downside protection since Realty generated more than to $200 million in cash flow to Corporate coffers during the last three years.
- Depreciation preserved capital: This is an important point. To preserve REIT status Gramercy has to pay in dividends a substantial part of its taxable income. Without Realty’s depreciation, Gramercy would have struggled to preserve capital while delevering the recourse obligations. Other REITs preferred to lose their status or find loopholes by paying dividends in stock. Gramercy chose differently, keeping the benefits of its REIT status and avoiding controversial legal issues.
- Not the end of the story just yet: while the whole investment thesis is based on the margin of safety provided by the non-Realty assets while planning for the worst outcome, the result of the Realty negotiations is still uncertain.
The transaction was badly timed and had bad consequences. Though, it had some redeeming factors and we have to balance it against the new team’s vindicating deal streak.
Gramercy also managed to preserve cash, cash that they are now ready to invest. There is a future for Gramercy without the Realty division, and maybe even prosperity, if they manage to protect the Finance Division and its CDOs.
PROTECTING OC TESTS
If Gramercy loses Realty, management’s time will probably be stretched along three main goals. First, to invest its large unrestricted ($196M) and restricted ($141M) cash hoard. Second, to find alternative funding sources to rebuild the REIT. And third, closely related to the reinvestment of the cash hoard, to protect CDO 2006 and heal CDO 2005.
This last goal is crucial in assessing Gramercy’s margin of safety. Failure of an overcollateralization test can cause “phantom income” problems when cash that constitutes income is diverted to pay down debt instead of being distributed to Corporate.
Without being completely exhaustive, since I imagine management is considering other options too, I put forward three ways to protect CDO 2006 and heal CDO 2005.
With CDO 2006 getting to the end of its reinvestment period but with $141 million in restricted cash one option is to invest it to improve the OC test.
So far the best way to improve collateral has been to buy CMBS at a discount. In other words, buy $100M CMBS face value for $70M and you increase the collateral by $30M. CMBSs have recovered in price, the discount decreased, and the collateral positive effect diminished … but there are still opportunities.
Another way is for CDO 2006 to repurchase its own bonds at a discount with its restricted cash. Buying discount CDO bonds would have the same effect as buying discount CMBS; it would increase the par value collateral used in the test.
There are limits to how much CDO 2006 can buy in CDO bonds and CDO 2007 is a dead parrot despite its lighter covenants on this issue. However, there are no restrictions for Corporate’s unrestricted cash.
Gramercy Corporate, among several other mREITs, has chosen to buy back previously sold CDO tranches for cents on the dollar. It makes a lot of sense just as good capital allocation if management is confident in the collateral,.
The math is simple. The average duration of CDOs is like 6-8 years. Let’s suppose it is 7 years and the CDO bonds were repurchased at 50% of par. Using the Rule of 72, that is a 10% annual return plus interests.
And those are unlevered returns! If for example they finance the buyback with a repurchase agreement (repo) backed by the CDO bonds, like Newcastle Investments NCT did last year, those returns can be multiplied by two or three times depending on the repo haircut.
That is jolly good but the repurchase of senior CDO bonds is even more effective for REITs struggling with their OC tests. Along with the outstanding returns and accounting gain, the repurchase can help to recapture diverted cash flows if they cancel enough of those bonds.
This is the redemption strategy, the buyback and cancellation of CDO bonds to reduce the denominator of the OC test.
The redemption strategy is controversial because it is not specifically allowed in the indentures. At the same time, it is not specifically prohibited.
Some senior CDO bondholders liked their accelerated payment of principal so they protested their trustees. Since trustees have many masters to please they kicked the issue to the courts where the Delaware Supreme Court finally upheld the redemption strategy.
Concord (Winthrop Realty Trust) vs Bank of America
I am not going to pretend to be a legal expert and say that this is the last word. What I do know is that, despite fits and starts, cancellations have continued. Maybe not at the brisk pace of 2008 and 2009, where ridiculous CDO bond discounts and some REITs desperation pushed the limits, but still good enough to cure borderline CDOs.
Concluding, the redemption strategy might be controversial and CDO 2006 might be at the end of the reinvestment strategy, but we have seen that they are still useful and powerful. But for the still somewhat concerned, there is a third option.
Even if a CDO is outside its reinvestment period or the CDO bonds cannot be bought at a sufficient discount (or faces legal opposition to cancel them) all is not lost. Gramercy can voluntarily exchange defaulted loans/securities/properties inside the CDO for loans/securities/properties that are current. These new loans/securities should be of an amount at least equal to the defaulted loan/security/property par value plus unpaid interests. The valuation difference is at the expense of Gramercy Corporate.
The math is very favorable because only a couple of quarters of CDOs cash flow can be sufficient to pay for any difference in valuation. And in addition, Gramercy would be taking control of the defaulted loan/security/property.
This strategy is specifically contemplated in the indenture of all CDOs (section 12.1) and it is not limited by the reinvestment period.
Notwithstanding the foregoing, the Collateral Manager (at its option and at any time) shall be permitted to effect a sale of a Credit Risk Security or a Defaulted Security hereunder by purchasing (or causing its Affiliate to purchase) such Defaulted Security or Credit Risk Security from the Issuer for a cash purchase price that shall be equal to the sum of (i) the Aggregate Principal Balance thereof plus (ii) all accrued and unpaid interest (or, in the case of a Preferred Equity Security, all accrued and unpaid dividends or other distributions not attributable to the return of capital by its governing documents) thereon. Notwithstanding anything to the contrary set forth herein, no Advisory Committee consent shall be required in connection with such cash purchase (the “Credit Risk/Defaulted Security Cash Purchase”).
In addition and notwithstanding anything to the contrary set forth herein (and provided that no Event of Default has occurred and is continuing), the Collateral Manager (at its option but only upon disclosure to, and with the prior consent of, the Advisory Committee) shall be permitted to effect a sale of a Defaulted Security or a Credit Risk Security hereunder by directing the Issuer to exchange such Defaulted Security or Credit Risk Security for (i) a Substitute Collateral Debt Security (that is not a Defaulted Security or a Credit Risk Security) owned by an Affiliate of the Collateral Manager (such Substitute Collateral Debt Security, the “Exchange Security”) or (ii) a combination of an Exchange Security and cash, provided that:
(i) (A) the sum of (1) the Principal Balance of such Exchange Security plus (2) all accrued and unpaid interest (or, in the case of a Preferred Equity Security, all accrued and unpaid dividends or other distributions not attributable to the return of capital by its governing documents) thereon plus (3) the cash amount (if any) to be paid to the Issuer in respect of such exchange by such Affiliate of the Collateral Manager, shall be equal to or greater than (B) the sum of (1) the Principal Balance of such Defaulted Security or Credit Risk Security sought to be substituted plus (2) all accrued and unpaid interest (or, in the case of a Preferred Equity Security, all accrued and unpaid dividends or other distributions not attributable to the return of capital by its governing documents) thereon;
(ii) the Eligibility Criteria and the Reinvestment Criteria shall be satisfied immediately after such exchange; and
(iii) the Aggregate Principal Balance of the Defaulted Securities and Credit Risk Securities so exchanged shall not exceed 10% of the Aggregate Collateral Balance as of the Closing (such limitation, the “10% Limit”).
The restrictions to the replacement strategy are standard. The Eligibility Criteria and Reinvestment Criteria are the same as for any acquisition of any loan or security for a CDO. And regarding the Advisory Committee, despite working by unanimity, it does not include representatives of the bondholders just one independent member.
Gramercy Capital is a cash flow positive Graham stock with a dividend trigger event, a good management team to invest the cash, and several other assets to boot, that should provide a good margin of safety and several low-cost options for profit.
PS: There is going to be a part 4 with some loose ends and exhibits