Citigroup Capital Securities XIII – a high yield trust preferred from Citi
This is a writeup for Citigroup Capital Securities XIII, or Citi N’s. Citi N’s are trust preferred obligations of Citigroup. Why write up a pref instead of a stock? This one has some interesting special features.
A trust preferred is basically a subordinated debt obligation of the issuer. It ranks above common equity, above preferred shares but below senior in the capital structure. Coupons on subordinated debt are deferrable for up to five years but cumulative, meaning that if Citi cannot pay the coupon on Citi N’s in any given year, this does not constitute an event of default. If Citi does not pay a coupon for five years, this does constitute an event default. As coupons are cumulative any coupons that were skipped at one point have to be repaid later on. Here, I am simply explaining the structure of the security. Citi has never skipped a coupon on Citi N’s and is unlikely to do so in the future.
Citi N’s pay a coupon of Libor + 637, which means the current coupon is 9.1205%. It is a $25 face security currently trading at 27.70, so the current yield is about 8.2%. 8.2% is a very high yield for a high quality credit like Citigroup. For comparison, JP Morgan and Bank of America Preferred Shares currently yield 5.6%, Morgan Stanley Preferreds yield 5.7% and Goldman Sachs Preferreds yield 5.8%. Trust Preferreds are higher in the capital structure than preferred shares, yet Citi N’s have a significantly higher current yield.
Citi N’s have a final maturity of 2040, but are immediately callable today. If Citigroup decided to call Citi N’s tomorrow, they would fall from 27.70 to something closer to par + accrued, so around 25.50. Citi could call Citi N’s tomorrow and issue new preferred shares with a coupon around 5.6%-5.7%. Why has this not happened yet?
To understand this we have to go back to the financial crisis. On January 15th 2009 Citigroup entered into a loss-sharing arrangement with Treasury, the Federal Deposit Insurance Corporation (the “FDIC”) and the Board of Governors of the Federal Reserve System related to a pool of $301 billion of assets. Citigroup issued to Treasury $4.034 billion of its perpetual preferred stock as consideration for the loss-sharing protection provided by Treasury and $3.025 billion of its perpetual preferred stock to the FDIC as consideration for the loss-sharing protection provided by the FDIC. Treasury’s and the FDIC’s perpetual preferred stock was exchanged for capital securities issued by Citigroup Capital XXXIII on July 30, 2009. On December 23, 2009, as part of an agreement to end the loss-sharing protection, Treasury cancelled $1.8 billion of the $4.034 billion Capital XXXIII Capital Securities it held, and the FDIC agreed to transfer an additional $800 million of its remaining Capital XXXIII Capital Securities to Treasury upon the maturity of Citigroup debt issued under the FDIC’s Temporary Liquidity Guarantee Program. On September 29, 2010, Citigroup modified Citigroup Capital Trust XXXIII by redeeming $2.234 billion of those securities that were owned by the U.S. Treasury Department. Citigroup Inc. replaced those securities with Citigroup Capital Trust XIII in the amount of $2.246 billion with a coupon of 7.875%, payable quarterly. The U.S. Treasury Department then sold all of such securities of Citigroup Capital Trust XIII to the public.
To summarize the important points: Citigroup received a capital injection from the Treasury and the FDIC in the form of Citigroup Capital Securities XXXIII. Part of the debt Citigroup owed to the Treasury was forgiven (“cancelled”). Citigroup Capital Securities XXXIII were converted into Citigroup Capital Trust XIII with a fixed rate of 7.875% which later in 2015 went to a floating rate of Libor + 637. In 2010 the US Treasury sold its remaining interest in Citigroup Capital Trust XIII into the market. Even though Citi N’s have been callable since the fixed to float date in 2015, it has not yet called them and is highly unlikely to call them in the next several years. Why is that?
The key to this lies in the accounting treatment of Citi N’s. At the time of the September 2010 exchange with the Treasury, C applied a fair value mark to the $2.246 billion of Ns, resulting in a lower amount of Tier 1 treatment. In fact, Citi is only counting only $1.366 billion of this security as Tier 1 capital, or roughly 61% of its notional value. The remaining $880 million of Citi N generate the highest form of capital, common equity Tier 1. This is simply the by-product of accounting rules that require the offset of the negative fair value mark on a liability to flow through as an increase in equity. What follows are the main reasons why Citigroup is unlikely to call these securities:
- As Citigroup has elected a complicated fair value accounting treatment under which the discount on Citi N’s amortizes over the next 21 years until final maturity in 2040. Therefore, inevitably, Citigroup would realize an accounting loss equal to the $880million of CET1 generated by Citi N’s if they called Citi N’s today.
- Citi N’s, together with Ally A’s received permanent grandfathering treatment by the Federal Reserve . While all other Trust Preferred’s in the US lose their capital treatment and must be phased out by 2022, Citi N’s and Ally A’s received special treatment and are permanently grandfathered.
- Trust Preferred securities pay coupons whereas preferred shares pay dividends. Coupons are tax deductible whereas dividends are not. Citi N’s are currently generating a tax benefit to Citigroup by shielding about $205 of income per annum
- Finally Citi has repeatedly said on its quarterly fixed income investor calls that they have no plan on calling these securities
I will keep the credit analysis short. As mentioned previously, Citi N’s are subordinated debt obligations of Citigroup. Citi is a well-capitalized bank with CET1 of 11.9% and total capital of 16.6% and a diversified global business. While the last crisis at least partially originated in the banking sector, the next one is very unlikely to put much of a dent into large cap banks. Why? Increased regulation from Dodd-Frank to CCAR have turned the banks into far less risky businesses. HELOC’s, low FICO consumer lending, high LTV mortgages, large bets on CDOs are all problems of the past and now reside within the nonbank financial sector instead of the banks. Whenever the large money center banks want to take on any significant new risk, they have to give long and detailed presentations to the Federal Reserve, explaining their rationale, internal risk limits and other factors behind the decision. Banks are also much better capitalized today. For example, Citi’s Tier 1 ratio was only 7.1% in 2008.
I think a safe level to buy Citi N’s is between 26-26.50. While I don’t expect a call in the next few years, the downside is a manageable -4-6% from these levels. If Citi N’s remain outstanding, receiving six months of coupon can pay for any downside an investor would experience in an unexpected call. While very few fixed income securities can provide ‘equity like’ returns, I believe Citi N with their 8.7% current yield at 26.3 are surprisingly close to the 10%+ type returns of a good compounder.
This may lead to some pushback by readers, but from a portfolio construction perspective one could even argue that there is a case to be made for using limited leverage to target a 10% return on Citi N’s. For example, an investor could invest $100 in Citi N with a current yield of 8.7%, then borrow an additional $25 at 3% generate a return on equity of 10.125%. While some may have a strong aversion to using any leverage at all, this is at least a useful way to illustrate the relative stability and limited downside of this security.
Finally, due to their limited downside and high yield I think Citi N’s are a good place to park some money when equity markets are expensive and there are few attractive stocks. Citi N’s will definitely not deliver the type of upside we have seen so far from stocks 2019, but they would have been a good source of cash in a selloff as they traded no lower than 26.12 in last years selloff and will produce a high and steady current yield over a long time horizon.