|Shares Out. (in M):||66||P/E||0||0|
|Market Cap (in $M):||3,871||P/FCF||0||0|
|Net Debt (in $M):||0||EBIT||0||0|
|Borrow Cost:||Available 0-15% cost|
Short - BOK Financial Corporation (Nasdaq: BOKF)
Share Price: $58.28
Market Capitalization: $3,871 million
Total Assets: $31,476 million
We believe BOK Financial (“BOKF”) provides an attractive short opportunity for the following reasons:
With the extreme and persistent decline in oil prices, we believe regional banks with energy exposure are at high risk for significant loan losses
BOKF has one of the most concentrated energy portfolios in the country (19.4% of total loans on an outstanding basis and 24.0% on a fully committed basis)
We have been able to compile a comprehensive list of BOKF’s exposure to public energy borrowers, and BOKF’s 2.89% allowance for loan losses in their energy portfolio (“ALL”) severely understates actual losses we believe will be incurred over the next 1-2 years
The depressed commodity environment will also have a significant indirect effect on BOKF’s commercial and residential real estate loan exposure in oil and gas concentrated economies, particularly in Oklahoma, which comprises 30% of BOKF’s loan portfolio
Losses on energy-related loans alone imply 30-40% downside before even factoring incremental losses in related to general economic downturn/weakness in Oklahoma and Texas, which we believe adds at least an incremental 7-10% of downside for 40-50% total downside
BOK Financial (“BOKF”) is a regional bank focused in an eight state market with a primary focus in Texas and Oklahoma
BOKF has 152 total branch locations and is the largest financial institution in the state of Oklahoma with 15% of the state’s total deposits
BOKF is a comprehensive commercial and consumer lender with total assets of $31.5 billion, loans of $15.9 billion and deposits of $21.1 billion as of 12/31/15
It’s lending profile is comprised of $10.3 billion of commercial business loans, $3.3 billion of commercial real estate loans and $1.9 billion of residential mortgages
Commercial loans by geography and type are shown below. Oklahoma and Texas represent ~70% of commercial business loans and ~45% of commercial real estate loans.
Commercial energy exposure is notably 19.4% of the total loan book.
BOKF’s revenue by business line is shown below and commercial loan breakdown by type and geography are shown below
BOKF historically has had ROA of ~1.1% and ROE of ~10.5% respectively.
Tangible book value per share is $42.51 and the current share price is $58.28 for a 1.37x P/TBV ratio.
We believe the market significantly underestimates future loan losses in BOKF’s energy loan portfolio
While, BOKF’s share price has declined ~20% since the end of Nov. 2015, this doesn’t seem to be due to its energy loans as it has actually outperformed the regional bank index (3.3% total outperformance since Jan. 2015)
Company had $5.6 billion of energy-related commitments and $3.1 billion outstanding (62% utilization) as of 12/31/2015
~82% of these are oil and gas exploration and production companies (“E&P”), which are some of the most troubled loans in the industry
Many E&Ps were spared in 2015 due to hedged production and large undrawn revolver balances, but are starting to see many of these unravel in 2016
Energy-related loans represent 19.4% of total BOKF loans outstanding on a drawn basis, but almost 27% on a fully committed basis – fully committed amount important as we believe energy-related borrowers (particularly E&Ps) will draw significantly on revolvers as they continue to burn cash in 2016
Already seeing evidence of significant revolver drawdowns. Energy loan exposure grew 9.1% from 3Q2015 to 4Q2015.
BOKF’s allowance for loan losses (“ALL”) on energy-related loan portfolio is just 2.89% as of 12/31/2015
Company’s reports a “criticized” loan balance of $325.7 million or 10.5% of its outstanding energy loan balance.
Loans are classified as “criticized” when default is determined to be “probable”, which the SEC guides to be approximately 75%-80% likely
A 2.89% provision means that of the loans they believe are “probable” defaults, the estimated recoveries are roughly 72%
BOKF justifies their loan loss reserve in their latest earnings call with the following:
“On Slide 21, we provide some additional details around our energy portfolio that validate, for us, that our 2.89% loan loss reserve to outstands in the energy portfolio, is appropriate given our loss history and portfolio composition. As Stacy showed, 82% of our energy portfolio is first lien, senior secured, reserve based lending, which we believe is a sweet spot in the energy sector. We have no exposure to unsecured high-yield debt or equity positions in the E&P companies, nor do we have any higher risk offshore relationships. As a result, while many investors appear to have accepted a 5% loan loss reserve to outstandings for energy exposure as a de facto standard, we believe our reserve accurately reflects the risk profile of our energy portfolio.”
In the prior slide of the earnings presentation, the Company shows an Energy Portfolio Stress Test, which the Company says allows them to guide to 2016 loan loss provisions of $60-80 million based on a “punitive” oil price deck
Sell-side analysts asked multiple questions on the earnings call regarding the Company’s energy loan exposure and loan-loss provision and were clearly content with the responses as they are modeling company’s current provision of 2.9%
Note that even if the Company provisioned at the high end of their guidance ($80 million of additional losses in 2016) their ALL would only increase to 5.4% assuming that borrowers drew nothing on their revolvers in 2016. If they fully drew their revolvers the implied ALL would be 3.0%.
We were highly skeptical of these energy provisions and guidance given the 75% decline in oil prices and therefore sought to better understand the Company’s exposure
In order to do this, we screened for credit agreements of all energy-related public equities and credits where BOKF was a lender
We were able to identify 38 public E&P issuers where BOKF was part of the bank group and this represented $1.05 billion in BOKF commitments or roughly 23% of the Company’s E&P commitments (complete list is shown in the appendix)
For those of you who are familiar with current distressed E&P names this is a who’s who list of troubled issuers, and we believe that the Company’s private “mom-and-pop oil co.” borrowers are likely materially worse
To be clear, based on the Q&A on earnings calls and in investor conferences we believe that very few market participants, if any, have compiled this list of public borrowers, which is why most are underwriting to Company guidance. In fact, we believe that if the market were provided access to this list the equity would rerate considerably
Some might argue, as management does, that because of their seniority in the capital structure, 1st liens/revolvers are unlikely to be impaired. However, if you look at some of the most recent energy bankruptcy precedents (Energy & Exploration Partners, Swift Energy, and others) 1st liens and revolvers have been equitized. The equity value implies recoveries well below par. Furthermore in most cases these bankruptcies require DIP financings/rights offerings. Either BOKF will have to participate in these, or their previously “highly secured” facility will be primed by those who do
In fact, many of the public issuers we identified already have revolvers quoted in the 60-80 cent context. Banks are trying to unload their revolver exposures for fear that they will be restructured to post-reorg equity that they don’t want on their balance sheets as heavily risk-weighted capital
Incurred Loss vs. Expected Loss Model for ALL
Given our view of credit quality in BOKF’s energy book, we were obviously surprised my management commentary that they are “properly provisioned as of 12/31/2015 with an allowance for loan losses of 2.89%”
Additionally, management has suggested that their “stress test” plays a role in determining loan loss provisions
Given all this, how could it be possible that the ALL is still so low?
We believe there is actually some misunderstanding regarding the accounting for ALL in the investment community
Banks currently calculate the ALL based on IAS 39 which is known as the “incurred loss model”
In this model, banks “can only recognize an allowance for credit losses when a credit loss even occurs and changes in creditworthiness of borrowers are not recognized until such a credit loss event occurs (typically when a payment default occurs)”
In other words, BOKF cannot book an ALL until the borrower has defaulted
If you listen to the 4Q15 earnings transcript carefully management confirms this
“There is some mobility to book qualitative reserves, but under the incurred loss model required by current accounting regulations for all banks, reserves must be supported by analysis”
Therefore, BOKF may in fact be adequately provisioned at 2.89% as of 12/31/15, but this does not mean that this amount is representative of the actual lifetime losses that will be incurred on this portfolio, or even management’s view of losses for that matter
However, effective as of January 1, 2018 IASB will transition from the “incurred loss model” to the “expected loss model under IFRS 9 and ALL will have to be calculated based on “expected” future credit losses
Revolver Cash Collateral
Another concept we believe is misunderstood by investors, is the collateralization of energy-related revolving facilities
E&P revolvers generally have a provision whereby lenders can redetermine their borrowing bases semi-annually based on their estimated value of the underlying oil reserves.
Therefore, many borrowers have recently been drawing down these facilities in anticipation of either a redetermination downward or an impending bankruptcy filing where use of revolvers may also be impaired
Many lenders are comforted by the logic that even if these businesses draw their revolvers at least they will have a lien on the cash drawn
Unfortunately, this may not necessarily be the case
Even if cash and deposit accounts have been pledged and comprise part of the collateral package it is not good enough to merely be a secured creditor. The liens must be perfected, which is unlikely to occur if the issuer defaults
Current distressed E&P precedents (e.g. Midstates Petroleum, SandRidge Energy, Linn Energy) suggest that cash collateralization will be a significantly and contested issue as it relates to BOKF’s revolver recoveries
Additional Upside from General Macro Weakness in Oklahoma/Texas
We believe there is also significant incremental downside in BOKF’s non-energy loan portfolio in Texas and Oklahoma whose economies will also be significantly affected by the decline in oil and gas prices.
PWC did a study several years ago, which ranked states by the impact of oil & gas exposure on employment, labor income and “value added” (their proxy of GDP). The results are shown below, and the takeaway is that Oklahoma (#2) and Texas (#4) has about 16% of employment and 23% of GDP tied to the oil and gas industry.
The recessionary impacts of this oil and gas cycle have not fully manifested yet, but we believe they will start to catch on over the next few months. The news media has already started to take notice, and this should be a helpful catalyst to rerate BOKF’s stock. A few sample articles are linked below.
Our analysis of the potential impact weakness in the Texas and Oklahoma economies is shown below. Commercial and Residential Mortgages represent $2.6 billion of BOKF’s outstanding loans and commercial business loans in those geographies represents another $3.8 billion.
We believe loan losses could conservatively amount to somewhere between 3.75% and 5.0% relative to the current ALL of 1.4%
This leads to a conservative additional $3-$5/share of downside in the stock incremental to the energy write-downs.
In order to estimate share price downside, we sensitized our estimate of realized loan losses on BOKF’s energy portfolio. We think that losses in the 10-12% context are reasonable given where many of the Company’s revolvers are already trading and the fact that Energy loan losses during the commodity price drop in the 1980s were at similar levels.
We also believe that in addition to the decline in value of the energy portfolio itself, the Price/Tangible Book Value multiple of BOKF would de-rate after the provisions/writeoffs we are expecting.
Our valuation output using this methodology is shown below. The first chart excludes the impact of OK and TX commercial loan exposure outside of energy, while the second chart incorporates this incremental downside.
Note that we also included the effect of borrowers incrementally drawing on outstanding revolvers
We thought that this P/TBV valuation was helpful in illustrating how the intrinsic value of BOKF will be impaired as a result of this commodity down-cycle. If you viewed valuation on a P/E basis you would see the same affect with additional provisions and write-offs impacting earnings over time. Clearly, if provisions/writeoffs are greater than expected, as we anticipate they will be, the stock will trade off considerably.
For example, when BOKF announced an increase in their expected provision in Jan. of this year from $3.5-$8.5 million to $22.5 million the stock lost ~$300 million of its market capitalization (or about $4.50/share).
In other words, a $16 million increase in provision guidance resulted in a 19-fold decline in the market cap
To put this in context, we believe BOKF to be under-provisioned by something in the ballpark of $450 million