Time For Citigroup To Sell Mexico

Summary

• Citigroup should sell Banamex Mexico.
• A U.S. G-SIB bank is not the natural owner of a Mexican bank.
• This will unlock substantial shareholder value with minimal damage to the overall franchise.
• In this article, I will make a compelling argument why Citi should sell Mexico.

Citigroup (NYSE:C) CEO Michael Corbat has vehemently denied in the past that his company has any intentions of selling Banamex Mexico. I think it is high time for Citi's board and CEO to reconsider this position.

Image source.

Citi is not the natural owner of the Mexican franchise

Citi is designated as a U.S. G-SIB and as such is subject to substantial regulatory burdens that do not apply to smaller, less complex banks. As such, it must hold higher capital, leverage and liquidity ratios and is subject to more rigorous stress tests and restrictions on its capital distributions.

As such, comparatively to its Mexican peers - the returns on the Mexican consumer businesses are diluted by Citi's G-SIB status.

This is perhaps best explained by looking at a recent presentation by Barclays (BCS) CEO James Staley:

As can be seen from above slide - Barclays Africa delivers strong ROE levels from a country view; however, when translated to BCS group results, reported ROE is diluted by a staggering 8.3%. Barclays management, of course, only disclosed this slide when it announced the intention to sell the Africa business.

Similarly in Citi's Mexico case - a local/regional owner will not be burdened by G-SIB capital buffers and other regulatory requirements. Simply put, Citi Mexico consumer business should be reporting higher ROE and valued much higher if held by a different owner.

But what about synergies of the global consumer banks?

Retail/Consumer banking businesses are largely managed locally and the synergies are somewhat limited (primarily technology sharing and some product development costs). As such, the term a "global consumer bank" is a misnomer in my view - this is in stark contrast, of course, to a global corporate business (which clearly is a globally-managed business).

The evidence also confirms this - considering Citi's LATAM rather high efficiency ratio at 61% for 2015, it does not suggest Citi is benefiting from substantial synergies.

But the key reason is CCAR

Citi's binding capital constraint is CCAR as opposed to G-SIB minimum regulatory capital ratios - as such, it is likely to run with a CET1 ratio closer to 13% by the end of 2016.

This is the "denominator" problem Corbat is concerned about, and a topic I covered in a recent article.

Having gone through Citi's 10-K and 2015 CCAR results, it is clear to me that the Mexico consumer business consumes a disproportionate amount of CCAR capacity (i.e. generates significant loan losses under CCAR stressed scenario).

Citi's consumer loan losses for FY2015 are shown below:

As can be seen from above, Latin America loan losses are substantially higher than Asia and North America (3.06% for retail and a staggering 9.57% for Cards). In a CCAR stressed scenario, one would expect forecasted loan losses to increase somewhat exponentially.

Considering the 2015 CCAR results:

As can be seen - Citi "Other Loan" losses (which includes international real estate loans) reflects a stressed loss of $5.1 billion. Given the Asian real estate portfolio is of pristine credit - most of that $5.1 billion would likely relate to Mexico.

Extrapolating the above charts - I estimate Latin America consumer business to result in around $6-$9 billion of CCAR stressed losses.

Citi naturally does not disclose the details around CCAR and specific loan assets' forecasted losses - but undoubtedly, LATAM consumer is a disproportionately negative contributor.

Any other ramifications?

Other points to note and consider:

• The DTA will not be impacted
• Citi's G-SIB capital buffers may be further reduced (an additional benefit)
• Regulators will certainly applaud
• The firm will become simpler and less complex (lower conduct/regulatory risks)
• Should prompt a special large capital return request once complete

What should the selling price be?

The Forbes article quoted above suggests a $40 billion price tag; I think that is a wildly exaggerated number (around $20 billion+ sounds about right, but I have not done the full homework as yet). Clearly, it is likely to be a multiple of tangible book value, perhaps further utilize DTA and naturally extremely accretive in the context of current valuation.

Final thoughts

Corbat's strategy has always been focusing on ensuring all businesses can earn above their cost of capital. I am not certain this is the right question to ask when it comes to Banamex - to my mind, the real question should be, why should a U.S. large G-SIB bank be its natural owner?

I acknowledge, there just isn't sufficient publicly available data to conclusively ascertain the full impact of selling Banamex - but from what is known and available (specifically around CCAR impact), there appears to be very little economic sense to keep it.

In the context of Citi's valuation and uncertainty around capital returns to shareholders - I strongly believe Corbat and the board owe shareholders clear and transparent answers to these questions.

As a side but related point, KBW analysts have recently issued a report arguing for a complete split of Citi between its consumer and corporate divisions. I do not believe KBW's proposal necessarily makes sense for shareholders for various reasons. I plan to further discuss this in my next article.

I cover U.S., European, Asian, Canadian and Australian large-cap banks, identifying long and short opportunities. If interested in the topic, I encourage you to sign up as a "real-time follower" or message me if interested in a specific banking name.

Disclosure: I am/we are long C, BCS. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.