Citigroup Is Perfect for Investors with a Multi-Year Approach

Includes: BAC, C, GS, JPM, MS, WFC
by: Amit Chokshi, CFA

About two weeks ago I issued my January commentary which discussed some earnings and economic issues along with why I felt Citigroup (NYSE:C) was an interesting investment. My discussion regarding C led to a number of comments that questioned why one would be long this bank.

I discussed the long case for C on Bloomberg's Taking Stock program Monday and felt it was worth presenting some more information on the merits for investing in C.

The most important factor when investing in a security is valuation and C is cheap. C trades for just 0.7x Tangible Book Value ("TBV") which is much lower than the multiples of its peers who trade at a healthy 1.3-1.7x TBV. C was the worst positioned relative to its peers heading into the crisis and as a result deserved a discount to its peer group. However, investors should really focus on the future with C and I believe that there are a number of positive factors occurring within the bank that should allow it to be valued more closely to its peers.

Click all images to enlarge


Investors are currently preoccupied with headline numbers without really looking under the hood. Consequently, the investors and press are focusing far too much time on the income statement and GAAP earnings. However, what will drive C's valuation is improving asset quality and that is clearly happening based on the Company's latest earnings release.

For some brief background, in Q109 C management reorganized C into two holdings: Citicorp ("CC") and Citi Holdings ("CH"). CH would house the worst of the Company's assets which would be wound down over time while CC would represent the "good" part of the Company. Based on this reorganization, CH assets were $898B at the start of 2008 (C provided a breakdown restating asset classifications after the reorganization in Q109 back to Q108) and were steadily wound down each quarter. As of Q409, CH assets stand at $547B, a decline of 40% since Q108 and roughly 24% from Q408. As Table II illustrates, CH accounts for less than 30% of C's assets as of year-end 2009 compared to 37% at year-end 2008.


Chart I illustrates why this matters. For fiscal 2009, CH accounted for 34% of C's total revenues but a monstrous 78% of the Company's credit losses. However, with CH now a smaller portion of C's total assets, total credit losses should decline as 2010 progresses. It's not just the decline of the CH holdings but the decline of some specific toxic assets within CH that positions C to realize credit losses of reduced magnitude.


During the worst of the crisis, when C secured TARP and preferential treatment from the US government ("USG"), it also entered into an agreement whereby a pool of $301B in assets was ring-fenced by the US taxpayer. This Special Asset Pool ("SAP") was the worst of the worst of C assets and consisted of $189B in consumer loans (first and second mortgages), $28B in corporate loans (CRE, leveraged loans), $30B in securities tied to Alt-A, SIVs, CRE, and other underlying assets, and $54B in unfunded commitments tied to second mortgages, CRE, and other loans.

As of Q409, the SAP portfolio has been whittled down to $154B, a decline of 49%. More importantly, C scaled the SAP book down by 15% in Q409 alone and took just about $1B in losses from this effort. Reducing toxic assets by $28B and realizing just $1B in losses is actually an impressive feat considering the overall asset quality in the SAP. If C can continue to unwind this portfolio at a rapid clip in 2010 and recognize losses in this range, that would bode very well for the Company.

While the SAP accounts for 28% of CH's total assets, CH's Brokerage & Asset Management Assets are 6% of CH's $547B assets ($35B) and C's consumer lending segments which include mortgages, student loans, credit cards, auto loans, personal loans and other loans are 65% ($358B) in total assets. The press and skeptics are now looking for the next shoe to drop and are looking at CRE or consumer credit tied to credit cards as possible catalysts.

This is certainly a problem for the broader economy but in the case of C it won't be anywhere near the magnitude that the financial crisis in 2008-2009 was. The reason is because when one looks at CH's breakdown of assets, the numbers are just too small. For example, student loans within CH assets are $31B while personal loans total $25B. This would require a massive loss rate beyond the reduced marks C has already taken on these assets to impact C's capital standing.

That is something investors may not realize. When C was receiving government assistance, management pored over its books and aggressively marked down their assets. The common, skeptical saying is that "we have no idea what things are valued at" and management is "hiding the bodies" but provide little granular evidence to support this. The reality is that there was a significant shift in psychology where in 2006, bank investors could not entertain the notion that any asset would go bad while in late 2008, bank executives believed every asset was a potential zero. Under the umbrella of government assistance, C was incentivized to aggressively mark down assets and it did so with regards to CH.

For example, while CH had $898B in assets at its peak and has reduced these troublesome assets to $547B, a number of these assets were wound down at levels above the marks on C's books. In fact, C completed approximately $10B of asset sales in its highly toxic SAP book that were at or above marks in Q409. These are actual sales, not accounting mechanisms to obscure values so there is clear tangible evidence that some of the most toxic securities may in fact have been marked below their true market values.

Another action C took that indicates it may have been overly conservative with writing down asset values was its transference of roughly $75B in assets from the SAP to CC. Of those assets, nearly $40B consisted of mortgages. So C management, after freeing itself from exceptional government assistance, was comfortable enough to transfer these assets back to CC. A skeptic would say it's just window dressing but that statement would make no sense. The fact is that shifting potentially dangerous assets to CC makes no sense and offers no benefit to C unless C management expects those assets to perform. C already has CH to dispose of bad assets so it's clear that the only reason C is bringing those assets to CC is because the quality of those assets is good.

This notion was further supported by the performance of C's second mortgages in Q409. In Q409, second mortgages demonstrated positive trends in credit losses and delinquencies. With second mortgages, C was proactive in addressing this problem as discussed in its Q409 conference call:

John Gerspach: Second mortgages and I think we mentioned this last quarter as well, we were early on pushing that book. We’ve cut lines dramatically and again we did that fairly early in the crisis and so that’s just been something we’ve been working down over time. It’s a book that perhaps has burned out. Now having said all of that, it's still subject to the same caveats that I would put on some of the other comments that I made on credit in North America in that it's ultimately going to be subject to the overall US economy as well as the impact of some of the other factors including HAMP.

What Gerspach is essentially saying is that C was focused on triage when its restructuring was taking place. Another example of the shift in psychology with respect to aggressively marking down assets can be seen in C's non-performing assets ("NPAs"). NPAs have been showing signs of moderating with Q409 NPA levels of $34B, basically flat compared to Q309. However, investors are keeping an eye on consumer non-accrual loans which increased from $18B in Q309 to $19B in Q409. This is a concern but what should be noted is the decrease in corporate non-accrual loans from $15B to $14B as shown in Table III. What is particularly interesting about C's corporate non-accrual loans is that over 66% of these are current with the borrower paying their interest and any other credit payments.


This is another example whereby C management was fearful of many of these loans going bad so rather than wait for that to happen, C aggressively lumped corporate loans it felt were questionable into the NPA category. During the worst of the crisis, 66% of corporate non-accrual loans remained current so if one believes the absolute worst of the economic crisis has passed, then a case can be made for a good chunk of these current loans moving from the NPA to performing category. This reversal could total up to $9B (67% of corp non-accrual loans) over the course of subsequent quarters, more than enough to offset potential increases in consumer non-accrual loans. Further, if C management has been as conservative with its consumer non-accrual loans as its corporate non-accrual loans, then it could be possible that many of the consumer NPAs are still current as well. In addition, early stage delinquencies are improving in consumer credit segments in the 30-89 days past due category, indicating the inflow of major problems is slowing.

Based on this evidence, C may see reversals by late 2010 whereby loan loss reserves may be significantly reduced. As of Q409, C has set aside $36B for loan losses. This represents 6% of total loans and current levels of net charge-offs and credit trends suggest credit losses may be peaking for C. If one can review C's overall asset quality improvements and credit trends as increasingly positive, then the survivability of C at this point is not in question.

If investors are comfortable with C being a long-term survivor, the next question is what drives C's valuation improvement. It's already established that C is trading at a significant discount to its peers, particularly on a P/TBV metric. The poor valuation may be due to a number of reasons. The first reason -- asset quality -- can be addressed with the points mentioned above. The credit and asset trends are improving so P/TBV should improve in that regard.

CH is becoming a smaller portion of C overall and as 2010 progresses, its contribution should continue to decline. Currently, CH provides negative value to C and it is clearly illustrated in C's latest earnings release. While CC generated EPS of $0.52 (using total outstanding shares at period end instead of average), CH generated a loss of $0.29. As CH becomes a smaller contributor to C, the Company's overall earnings power should be less obscured whereby a higher multiple is ascribed to CC and a lower to CH. Focusing solely on CC, a valuation range of $4.00-$6.00 would be reasonable but CH creates a drag on the stock. However, as CH winds down, the valuation drag should be reduced. Further, a $4.00-$6.00 share price would put C closer to its peers in terms of P/TBV.

If investors are concerned with poor asset quality then the concern could be for more dilutive equity issuances. This is not a likely prospect in my opinion. C was comfortable repaying $20+B in capital to the USG through a share issuance and it makes little sense that the Company would do that and then come back with hat in hand to investors. C has indicated that it will be issuing just $15B in financing (notes, term debt) in 2010 and has enough liquidity to redeem up to $50B in debt coming due this year.

The skeptic would say "management missed it all, why believe them?" That's a valid concern but one can actually look through the filing data and see that C is well capitalized. In fact, C could be considered over-capitalized which is why it trades at a discount to peers. C's Tier I Capital Ratio, Common Ratio, and Tangible Common Equity ("TCE") Ratio are 11.7%, 9.6%, and 10.9%, respectively. This was due to C de-risking itself over the past two years to shore up its balance sheet.

What this means is that while C's peers have been deploying capital, C has been hoarding it to get a handle on asset losses. Now that the Company seems to have a handle on loss trends, C should start to reduce its capital ratios and put money to work. For example, if two banks have $100 to invest and one deploys 80% of the $100 and the other invests just 40%, if both investments generate the same return, then the more conservative bank is not utilizing its $100 - or book capital - as effectively as the bank that deploys 80% of its capital. There's clearly that fine line between bank leverage and reserving for losses but all else equal, banks do not do shareholders a big service by being over capitalized. And again that represents a shift in investors' views over the past two years when banks were painfully undercapitalized in a highly levered environment to now, when capital ratios are very high.

C management has admitted that its liquidity levels are anywhere between 2-3x normal levels. Banks make money by deploying capital into interest earning assets and high liquidity levels prevent C from doing this. However, as 2010 progresses and bad assets continue to burn off, C will be in a position to reduce its liquidity and place that capital into interest-bearing assets. This will result in higher Net Interest Margins and Efficiency Ratios for C, driving earnings.

The skeptics would come back and say "nobody's lending!" That's true in the US and that actually benefits C relative to some of its peers. Unlike some other TBTF banks, C is truly a global institution with 68% of its revenues generated outside of North America. So if one looks at C's credit trends as well as broader financial credit trends globally, international segments are outperforming. More importantly, while the US may still be dealing with levels of overcapacity, the same may not be said in emerging markets where demand may be much stronger. As a result, as time progresses and bad assets roll off, C's earnings power can continue to improve.

Investors should also keep in mind that C can repurchase shares and crank up its dividend which I expect to happen by late 2010 and into 2011. At current prices, buybacks would be highly accretive and would help reduce C's liquidity levels. This would result in a higher ROE and smaller share base which could make the case for a higher share price in the coming year.

These are just a few reasons that I think C makes an attractive long for longer-term investors that can stomach some volatility. The fundamentals and capital position of C are now very strong and it has a tremendous international presence which should outperform the struggling US economy. Second, valuation levels are depressed due to some fundamental issues but also possibly trading/artificial reasons due to concern regarding the USG disposing of its stake over the course of the next year. For investors that can take a multi-year approach, C could generate a high rate of return with limited downside risk.

Last, I would also note that I believe I have a good understanding of C's moving parts. I spend a fair amount of time on due diligence and I had focused on Downey Financial and discussed this as a short opportunity when it was over $70 per share and loved by the Street. This is simply to illustrate the point that I don't take the risks of bank stocks lightly but believe we've had a big shift in investor sentiment. Two years ago, banks were overlevered and setting aside a low amount for potential loan losses. We have pulled a 180 since that time with banks now carrying much lower leverage and setting aside reserves that are likely much higher than what realized losses will be.