Bank of America - Global Bank Regulation Primer
Bank of America - Global Bank Regulation Primer
is not without cost. Of course, the direct impact is to bank profitability, and we estimate Alastair Ryan >>
Research Analyst
that post-crisis global regulation erased 500bp from bank ROTCEs (returns on tangible MLI (UK)
common equity), pre-tax cuts. And while market dynamics were already shifting, we alastair.ryan@baml. com
believe regulation has opened the door for less-regulated players to take share, diffusing Michael Helsby >>
Research Analyst
rather than reducing risk in key areas, like mortgage lending and trading. Further, greater MLI (UK)
capital and liquidity burden, along with higher regulatory-related operating costs, michael.helsby@baml.com
benefited scale players, concentrating market share at US GSIBs – who appear largely Winnie Wu >>
balanced across different regulatory constraints and have taken advantage of business Research Analyst
Merrill Lynch (Hong Kong)
synergies to deepen wallet share locally and globally. Lastly, while bank clients have winnie.wu@baml.com
largely avoided being re-priced to reflect these burdens, in our view, less liquid and more Futoshi Sasaki >>
turbulent markets could have a marked impact on institutional client returns. Research Analyst
Merrill Lynch (Japan)
What’s inside: A reference book for global rules, key laws futoshi.sasaki@baml.com
Ebrahim H. Poonawala
As in previous versions, we dive deeply into the following global rules: 1) common equity Research Analyst
tier 1 (CET1); 2) supplementary leverage ratio (SLR); 3) supervisory stress tests; 4) MLPF&S
ebrahim.poonawala@baml.com
liquidity coverage ratio (LCR); 5) net stable funding ratio (NSFR); and 6) total loss
Brandon Berman
absorbing capacity (TLAC). We also look into key provisions of the Dodd-Frank Act, such Research Analyst
as the Volcker Rule, resolution planning (“living wills”), and mortgage reform. Further, we MLPF&S
brandon.berman@baml.com
take a look at the US regulatory agency structure, where nine of the ten previously open,
key leadership positions have been filled by President Donald Trump. Christopher Nardone
Research Analyst
MLPF&S
christopher.nardone@baml.com
US Fixed Income Research
>> Employed by a non-US affiliate of MLPF&S and is not registered/qualified as a research analyst under Hima B. Inguva
the FINRA rules. Research Analyst
Refer to "Other Important Disclosures" for information on certain BofA Merrill Lynch entities that take MLPF&S
+1 646 855 6810
responsibility for this report in particular jurisdictions. hima.inguva@baml.com
BofA Merrill Lynch does and seeks to do business with issuers covered in its research reports.
As a result, investors should be aware that the firm may have a conflict of interest that could See Team Page for List of Analysts
affect the objectivity of this report. Investors should consider this report as only a single
factor in making their investment decision.
Refer to important disclosures on page 184 to 185. 11873601
For four years in a row, we have published our global bank regulatory primer as a
handbook for investors to better understand the complex landscape that emerged from
the Global Financial Crisis. While readers can continue to find sections on global
prudential rules and Dodd-Frank (US) legislation, we dive more deeply into the “real
world” impact of regulations to bank strategy and clients, or the real benefits and costs
of regulation as well as the real-time developments of regulatory reform in the US.
New this year, given the fluctuations of the regulatory construct, we have a section
called “Regulation in real time” that outlines the most recent relevant formal
proposals and bills from regulatory agencies and legislators, respectively. Given that the
financial industry has now operated under the post-Global Financial Crisis regulatory
framework for some time, we have also expanded our market and business impact
section, titled “Regulation in the real world”.
As in earlier versions of the primer, we have sought to provide a detailed (but simplified)
explanation of each rule, addressed global and local standards, discussed outstanding
issues for each law/rule/regulation, and outlined specific business impacts.
9. the Department of Labor (DOL) fiduciary rule & SEC Proposals (U.S. only)
The authors of this report are not acting in the capacity of a legal adviser, and the
information contained herein is not intended to constitute legal advice. You should consult
with your legal adviser as to any issues of law relating to the subject matter of this report.
Exhibit 1: UST reform proposals: Banks and Credit Unions Exhibit 2: UST reform proposals: Capital Markets
▪ Raise threshold for participation. Remove SEC requirements that duplicate
▪
Eliminate mid-year DFAST cycle and reduce Public companies financial statement disclosures
Acces to Capital
▪
number of stress scenarios to two. and IPOS Allow companies to "test the waters" with
▪
▪ Reassess underlying CCAR assumptions. potential/qualified investor
CCAR / DFAST ▪ Change CCAR to a two-year cycle. Modify rules that would broaden eligibility for
Challeneges for ▪
Adjust the qualitative assessment to the status as a smaller reporting company
small public
▪ horizontal capital review for banks with less Extend length of time a company may be
companies ▪
Capital Requirements
▪ LCR requirement and a less stringent Rationalize the capital required for
Liquidity
▪
standard for non-GSIBs. securitized products
LCR
Allow for high-grade municipal bonds to be Consider the impact that trading book capital
▪ ▪
HQLA eligible. Securitization standards, such as FRTB, on market activity
▪ Raise threshold for living will requirements. Capital
Recalibrate capital requirements to prevent
Adjust living will submission to a two-year ▪ from exceeding the maximum economic
Resolution ▪
cycle. exposure of the underlying bond
Planning
Subject framework and guidance to public Adjust the global market shock scenario for
▪ ▪
comment. stress testing purposes
Align QM requirements with GSE eligibility High-quality securitized obligations should be
▪ Liquidity ▪
requirements. considered level 2B HQLA
Increase the $103k loan threshold for
▪
application of the 3% points and fees cap. Margin Harmonize margin requirements for uncleared
▪
Derivatives
banks with $10-50bn only need to in assets. the cost it takes to “run” the test.
result of adverse conditions. consequences of such a test.
perform it 1x/year.
Impacted banks over $50bn in assets
The intermediate holding
are required to establish a separately The operating expense of having a
company requirement was Applicable to foreign headquartered
capitalized IHC that would hold all US There are no current alternate completely separate architecture for
Intermediate Holding designed to provide greater institutions that would have aggregate
Federal Reserve bank and non-bank subsidiaries.The proposals that solely affect foreign the IHC within the bank; lower
Companies (IHC) oversight by US regulators of the asset exposure of $50bn or above,
provisions have granted broad powers banks. revenues; burden on leverage ratios
US operations of foreign- domiciled in the US
to the Fed to determine additional from additional liquidity positions
domiciled banks
requirements for IHCs.
For the large SIFI banks (defined as
Requires BHCs to address how >$250bn in assets), resolution plans Operating costs associated with
There are no current alternate
Resolution planning they would properly wind down in must be submitted by July 1st of each Resolution planning: BHCs with over preparing resolution plans; higher
Federal Reserve, FDIC proposals to eliminate banks resolution
("Living Wills") the event of failure without year. Other applicable institutions are $50bn in assets liquidity balances which could be
planning
causing systemic risk required to file their plans by December deployed into higher yielding assets
31st
FDIC, Federal Insurance Orderly Liquidation together to determine whether a Liquidation Fund which helps fund
become a receiver in the received a systemic risk determination planning, burden from additional
Office, Treasury Authority financial company requires the need to liquidation. The CHOICE Act directly
liquidation process of a failing liquidity positions
invoke OLA targets the repeal of Title II
financial firm
Exceptions to the rule includes: 1) Applicable to money centers and The Fed and other regulators are
Imposed restrictions to prohibit or There are three major “costs”
Buying/selling securities if it relates to broker-dealers; however, as written, the planning to eliminate an assumption
restrict the ability of banking stemming from the Volcker rule: (1)
Volcker Rule underwriting a securities offering for a rule is applicable to “banking entities” written into the original Volcker Rule
Federal Reserve, CFTC, entities from engaging in short- higher compliance costs and
customer; 2) Purchasing/selling and “nonbank financial companies” law that positions held by banks for
SEC, FDIC, OCC term proprietary trading and operational requirements; (2) lower
securities pursuant to market-making designated by FSOC. In other words, less than 60 days are speculative and
sponsoring of or investing in liquidity in the bond markets; and (3)
activity; 3) A bank can trade securities Volcker even applies to BHCs with thus banned; overhaul expected by end
private equity and hedge funds. lower revenues in fixed income trading.
to hedge or mitigate their own risk <$10bn in assets. of May.
Source: BofA Merrill Lynch Global Research, Dodd-Frank, Financial Services Committee, CFTC, SEC, Federal Reserve, FDIC, FSOC, OCC
Key acronyms: SIFI = systemically important financial institution; DFA = Dodd-Frank Act; BHC = bank holding company; IHC = intermediate holding company; OLA = orderly liquidation authority; DCO = derivative clearing organizations; DCM = designated contract market; SEF
= swap execution facility; CDS = credit default swap; and ERISA = Employee Retirement Income Security Act
Exhibit 4: Dodd-Frank Cheat Sheet (Titles VII-X)
participants (MSPs)
The DOL is looking to apply a With the 1974 Employee Retirement Industry reports indicate potential
fiduciary standard to a broader Income Security Act (ERISA), the DOL The 5th Circuit Court of Appeals ruled costs for the rule range from $200M
Institutions that advise retirement
Department of Labor Fiduciary Rule group of market participants established standards for private that the DOL exceeded its authority in (DOL) to $5B+ (industry), while the
assets
giving advice related to retirement industry pension plans (both DB and promulgating its Fiduciary Rule revenue impact could be positive or
assets 401k). negative depending on the firm
Source: BofA Merrill Lynch Global Research, Dodd-Frank, Financial Services Committee, CFTC, SEC, Federal Reserve, FDIC, FSOC, OCC.
Global Banks and Brokers | 18 May 2018
Note: Title VII also applies to security-based swaps regulated by the SEC; however these rules are not yet final and therefore not addressed further. In addition, while Title VII encompasses a much longer list of requirements (e.g. reporting, business conduct, etc.), we are
only focusing on requirements imposed on OTC derivatives.
Key acronyms: SIFI = systemically important financial institution; DFA = Dodd-Frank Act; BHC = bank holding company; IHC = intermediate holding company; OLA = orderly liquidation authority; DCO = derivative clearing organizations; DCM = designated contract market; SEF
= swap execution facility; CDS = credit default swap; and ERISA = Employee Retirement Income Security Act
7
Basel regulation cheat sheets: Summary of rules, requirements, and applicability
8
Source: BofA Merrill Lynch Global Research, FSB, Federal Reserve, Basel Committee on Banking Supervision (BCBS), FSA
*There are five different capital constraints in the U.S. stress test
Chinese and Japanese GSIBs align themselves according to Basel standards
UK = United Kingdom, CH = Switzerland, EU = Eurozone (i.e., Basel), SE = Sweden
Table 2: Impacted institutions, businesses and products
Impacted Institutions
Regulation Basel US Europe More Valuable Products Less Valuable Products
Electronic/agency trading
Wealth management
Equity derivatives
Asset management
Securitized products
Advisory
HY credit products
Payments
Common Equity Tier 1 (CET1) All banks All banks All banks Commodities
Clearing
Mortgage servicing
Rates
Non agency MBS
Repo
Higher-risk loans
Agency MBS
Unfunded lending commitments
Electronic/agency trading Equity derivatives
Wealth management Prime brokerage
Asset management Rate
Advanced
Supplementary Leverage Ratio (SLR) All banks All banks Advisory Repo
Capital approaches
Payments IG credit products
Clearing Commodities financing
High yield/ distressed credit Unfunded lending commitments
Largest banks International lending exposure
BHCs with assets Historically low loss content loan
Supervisory Stress Testing All banks covering 70% of Subprime lending
≥ $50bn products (e.g. prime mortgage)
banking assets Non-operational deposits
L/T unsecured funding
Electronic trading
Deposit funding
Wealth management
All G-SIBs (ex- S/T funding
GSIBs (D-SIBs Asset management
Total Loss Absorbing Capacity (TLAC) banks in emerging GSIBs Structured funding
likely) Advisory
mkts) Securitized funding
Payments
HY credit products
Clearing
Agency MBS
Liquidity and credit facilities
G-SIBs (≥$250bn in Lending products to fin'l institutions
Retail deposits
Liquidity Coverage Ratio (LCR) All banks assets) & BHCs All banks Non-operational corporate deposits
Operational corporate deposits
(≥$50bn, < $250bn) Financial institution deposits
Prime brokerage
The current regulatory ST funding
regime places different value
Global Banks and Brokers | 18 May 2018
The 5th Circuit Court of Appeals ruled that the DOL exceeded its authority in promulgating
Department of Labor
Senate 3/15/2018 ▪ its Fiduciary Rule, which was slated to go into full effect July 1, 2019. The DOL has until
Fiduciary Rule
June 13th to challenge this decision in front of the Supreme Court
Fed released a package of three proposals centered on increasing the transparency of its
Enhanced transparency stress testing program, to be applicable to the 2019 DFAST/CCAR process: (1) enhanced
Federal Reserve 12/7/2017 ▪
for 2019 CCAR model disclosure (2) changes to scenario design framework (3) supervisory stress testing
model policy update.
The OCC announced it would begin soliciting public feedback on potential changes to the
▪
Volcker Rule
Federal Reserve,
Volcker Rule reform 8/2/2017 The OCC put forth a set of questions for the banking industry to address, focused on the
Congress
▪ following topics: (1) scope of entities subject to rule (2) proprietary trading prohibition (3)
covered funds prohibition (4) compliance program and metrics reporting requirements
The bill would allow a money market fun, under specified Ordered to be
Consumer Financial Choice and Capital conditions, to elect to operate using a different method of reported to To be voted on No estimate
H.R. 2319
Markets Protection Act of 2017 valuation. If elected, a money market fund shall not be subject to House by House available
specified requirements related to the imposition of liquidity fees. (1/18/18)
Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: TLAC estimated based on publicly available data. NSFR not disclosed, Note: TLAC estimated based on publicly available data. NSFR not disclosed, Note: TLAC estimated based on publicly available data. NSFR not disclosed,
assumed 100% assumed 100% assumed 100%
100% 100%
TLAC NSFR TLAC NSFR
50% 50%
Global Banks and Brokers | 18 May 2018
Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: TLAC estimated based on publicly available data. NSFR not disclosed, Note: TLAC estimated based on publicly available data. NSFR not disclosed,
assumed 100% assumed 100%
13
U.S. Trust Banks
14
CET1 CET1
150% 150%
100% 100%
TLAC NSFR TLAC NSFR
50% 50%
Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: TLAC estimated based on publicly available data. NSFR not disclosed, Note: TLAC estimated based on publicly available data. NSFR not disclosed,
assumed 100% assumed 100%
100% 100%
TLAC NSFR TLAC NSFR
50% 50%
Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: Use MREL disclosure for TLAC Note: Assumes 4.0% minimum (company target)
Chart 10: Credit Suisse Chart 11: UBS
CET1 CET1
150% 150%
100% 100%
TLAC NSFR TLAC NSFR
50% 50%
Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: TLAC as % of leverage exposure. Assumes 5% SLR minimum per FINMA Note: TLAC as % of leverage exposure. Assumes 5% SLR minimum per FINMA
Chart 12: Fixed income, currencies, and commodities Chart 13: Equities
160% WFC 160% WFC
Leverage ratio relative to requirement
(2017)
100% CS UBS 100% UBS CS
90% DBK 90% DBK
80% Stay the 80%
Stay the
Global Banks and Brokers | 18 May 2018
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
15
Fixed income products
16
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
Equities products
Chart 20: Cash Equities Chart 21: Prime Services Chart 22: Equity Derivatives
160% WFC 160% WFC 160% WFC
Lever age r atio r elative to r equir ement (2017)
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
FICC global investment bank league table rank (2017) Equities global investment bank league table rank (2017)
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
17
Fixed income products
18
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
Chart 28: Credit trading Chart 29: Securitizations Chart 30: Commodities
MS MS
160% 160% 160%
CET1 r atio r elative to r equir ement (2017)
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
Equities products
Chart 31: Cash equities Chart 32: Prime Services Chart 33: Equity derivatives
MS MS MS
160% 160% 160%
CET1 r atio r elative to r equir ement (2017)
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank Note: WFC not included in analysis; assumed #13 rank
Global Banks and Brokers | 18 May 2018
19
Regulation in real time: The most recent
2B
global developments
In summary, the recent regulatory proposals and related pending legislation in the
United States appear to favor “tailored” regulation, where rules remain mostly
stringent for the eight U.S. GSIBs but less onerous for regional banks.
Executive order
What follows is the summary of the original February 2, 2017 Executive Order
pertaining to financial regulation, and include our own observations next to each Core
Principle (see Exhibit 8).
Aligning the financial system to support the U.S. economy Progress here focused on regional and community banks
For depository institutions, the three arguably most impactful agencies are the Federal
Reserve, the FDIC (Federal Deposit Insurance Corporation), and the OCC (Office of the
Comptroller of the Currency). All new relevant leadership at these agencies were
confirmed after the Treasury report was published.
For the U.S. capital markets, the primary regulators are: the SEC (Securities and
Exchange Commission), the CFTC (the Commodity Futures Trading Commission), and
state securities regulators. Self-regulatory organizations, or SROs, also are key players in
regulating certain parts of the financial services sector. These SROs include: FINRA
(Financial Industry Regulatory Authority), MSRB (Municipal Securities Rulemaking
Board), and NFA (National Futures Association).
In April 2018, the Federal Reserve published a draft proposal that would amend capital
rules for bank holding companies (BHCs) with $50bn+ in assets. While U.S. SIFIs all have
an overarching binding capital constraint (in other words, the most restrictive of all the
different capital requirements), BHCs with $50bn+ in assets were required to balance
business-as-usual, or “spot” capital requirements, and stressed capital requirements
under the annual DFAST (Dodd-Frank Act Stress Testing) and CCAR (Comprehensive
Capital Analysis and Review) process. In this proposal, the Fed would “merge” the two
requirements for standardized common equity tier 1 (CET1). (A discussion of the
difference between standardized and advanced approaches in CET1 can be found
beginning on page 94).
Under the current regime, the graphic below represents the base, non-stressed capital
requirement for CET1 (see Chart 34).
Chart 34: Non-stressed requirements, CET1
0.0 to 2.5%
Counter cyclical buffer
~1.5%
~1.4%
Planned distributions and b/s growth
Stress test losses
Minimum requirement
4.5%
Under the current proposal, banks would no longer fail the test in a “quantitative”
fashion, which would occur if the minimum requirements set by the Fed under stress are
breached. (A detailed explanation of the DFAST and CCAR process in the US can be
found beginning on page 130.) Rather, the Fed is proposing to replace the capital
conservation buffer of 2.5% with the net amount of capital “burned” in the test – this is
called the stressed capital buffer. In other words, rather than enforcing a quantitative
failure, poor test results can lead to higher “spot” capital requirements.
Key definition
When we refer to “capital burn”, we refer to the Fed’s measurement that takes the
difference between a BHC’s starting capital ratio and the minimum projected over the
duration of the 13-quarter test
We break the new proposed requirements below (see Chart 36). As with the previous
chart, dividend add-on and stress test losses represent the average of the 2017 results. We
have stripped out balance sheet growth and capital return beyond four quarters worth of
dividend payments to calculate average stress test losses.
1.0 to 5.5%
G-SIB surcharge (if applicable)
~ 0.5% Dividend add-on
2.5% to ∞ Stressed capital buffer
Minimum requirement
4.5%
Key definition
The stressed capital buffer (SCB) would represent the net amount of capital
“burned” by a bank during the stress test, and would include four quarters worth of
dividends. Unlike in previous tests, the Fed would allow banks to assume no balance
sheet growth and no other capital return other than four quarters of dividend
payments. This SCB would replace the capital conservation buffer in the current
CET1 framework. It cannot be lower than 2.5%, and has no maximum value.
• Although banks can no longer fail quantitatively, BHCs can still receive an objection
to their capital plans on a qualitative basis. Note that as of February 3, 2017, the
qualitative assessment is no longer applicable to banks less than $250bn in assets.
• The Fed proposes to remove the “soft cap” on dividends that is currently equivalent
of 30% of earnings
• In a concession to the BHCs, the Fed will no longer assume balance sheet growth.
On average, using the 2017 CCAR results, this would “save” 70bp of capital burn.
Note that there has been a notable difference between Fed projected RWAs and
bank projected RWAs in previous tests (see Chart 37)
12.0
10.0
8.0
stressed scenario (%)
6.0
4.0
2.0
0.0
(2.0)
(4.0)
(6.0)
2014 2015 2016 2017
CCAR year
Source: BofA Merrill Lynch Global Research, company data, Federal Reserve
• In another concession to the BHCs, the Fed will no longer assume that banks will
continue to return capital via dividends and buybacks in a stressed scenario, other
than four quarters worth of dividend payments. On average, using the 2017 CCAR
results, this would “save” another 40bp of capital burn.
• The CCAR results are regularly released in June. Under this proposal, the stressed
capital buffer set in June by the current-year CCAR would be applicable beginning in
October of the same year, and would be valid for 12 months. In other words, if
finalized into rule by end of this year, the 2019 CCAR result would determine the
SCB, and therefore, the day-to-day standardized CET1 requirements, for October
2019 through October 2020.
• According to the proposal, the Fed would like to finalize this rule in time for the
2019 DFAST/CCAR cycle.
Leverage component
Under the same proposal, the Fed also introduced the stressed leverage buffer,
applicable to the tier 1 leverage ratio (see Exhibit 12). Like the SCB, the stressed
leverage buffer measures capital burn, or the difference between the BHC’s starting and
minimum projected tier 1 leverage ratio (also including four quarters worth of
dividends). This would replace the requirement under the current capital regime that a
firm demonstrate the ability to maintain capital levels above minimum leverage
requirements on a post-stress basis.
Exhibit 12: Tier 1 leverage ratio formula
Tier 1 Capital
Average total
consolidated assets and
certain off balanc e sheet
exposures
However, there are certain BHCs whose “pro forma” requirement would be above
their public targets for CET1. This is meaningful as investor expectations for
business growth, dividends, and buybacks are set based on public CET1 targets.
Exhibit 13: Pro forma capital requirements under the Fed’s proposal
I=
J=
Key A B C D E= C - D F G H max [E-G- K L M=K-J N=L-J O P=O-J
A+B+F+I
H or 2.5%]
Proj min Stess test
Add: Less: Less: Stressed Current Spot CET1
Minimum GSIB CET1 ratio CET1 ratio losses Pro forma CET1 ratio Current Target vs. CET1 ratio
Planned RWA DFAST div capital min vs pro ratio vs pro
req surcharge (4Q16) (DFAST (DFAST req "target" min req pro forma (4Q17)
dividends growth payout buffer forma forma
'17) '17)
GSIBs
JPM 4.5% 3.5% 12.2% 9.1% 3.1% 0.6% 0.8% 1.1% 2.5% 11.1% 11.0% 10.5% -0.1% -0.6% 11.8% 0.8%
C 4.5% 3.0% 13.0% 9.7% 3.3% 0.3% 0.8% 0.4% 2.5% 10.3% 11.3% 10.0% 0.9% -0.3% 12.1% 1.7%
MS 4.5% 3.0% 16.7% 9.4% 7.3% 0.5% 0.5% 1.0% 5.8% 13.8% 10.5% 10.0% -3.3% -3.8% 16.1% 2.3%
GS 4.5% 2.5% 14.0% 8.4% 5.6% 0.2% 0.5% 0.5% 4.6% 11.9% 10.5% 9.5% -1.4% -2.4% 11.9% 0.0%
WFC 4.5% 2.0% 10.8% 8.6% 2.2% 0.6% 0.8% 1.3% 2.5% 9.6% 10.0% 9.0% 0.4% -0.6% 12.0% 2.4%
BK 4.5% 1.5% 11.3% 11.2% 0.1% 0.7% 0.8% 1.2% 2.5% 9.2% 9.0% 8.5% -0.2% -0.7% 11.7% 2.6%
STT 4.5% 1.5% 10.9% 7.4% 3.5% 0.6% 0.5% 1.4% 2.5% 9.1% 10.0% 8.5% 0.9% -0.6% 10.8% 1.7%
Regionals
BBT 4.5% 0.0% 10.0% 7.9% 2.1% 0.6% 0.7% 1.3% 2.5% 7.6% 10.0% 7.0% 2.4% -0.6% 10.2% 2.6%
CFG 4.5% 0.0% 11.1% 7.7% 3.4% 0.3% 0.6% 0.5% 2.5% 7.3% 10.0% 7.0% 2.7% -0.3% 11.1% 3.8%
CMA 4.5% 0.0% 11.1% 9.4% 1.7% 0.3% 0.7% 0.5% 2.5% 7.3% 10.0% 7.0% 2.7% -0.3% 12.0% 4.6%
FITB 4.5% 0.0% 10.3% 8.0% 2.3% 0.4% 0.7% 0.7% 2.5% 7.4% 10.0% 7.0% 2.6% -0.4% 10.5% 3.1%
HBAN 4.5% 0.0% 9.6% 7.0% 2.6% 0.6% 0.6% 0.7% 2.5% 7.6% 9.5% 7.0% 1.9% -0.6% 10.5% 2.9%
KEY 4.5% 0.0% 9.4% 6.8% 2.6% 0.4% 0.6% 0.5% 2.5% 7.4% 9.5% 7.0% 2.1% -0.4% 10.0% 2.6%
MTB 4.5% 0.0% 10.7% 7.9% 2.8% 0.5% 0.7% 1.0% 2.5% 7.5% 9.5% 7.0% 2.0% -0.5% 11.0% 3.5%
NTRS 4.5% 0.0% 11.5% 10.9% 0.6% 0.5% 0.8% 1.1% 2.5% 7.5% 8.5% 7.0% 1.0% -0.5% 12.6% 5.0%
PNC 4.5% 0.0% 10.0% 8.0% 2.0% 0.5% 0.8% 0.8% 2.5% 7.5% 8.5% 7.0% 1.0% -0.5% 9.6% 2.1%
RF 4.5% 0.0% 11.1% 8.2% 2.9% 0.4% 0.6% 0.7% 2.5% 7.4% 9.5% 7.0% 2.1% -0.4% 11.1% 3.7%
STI 4.5% 0.0% 9.4% 7.1% 2.3% 0.4% 0.7% 0.7% 2.5% 7.4% 8.5% 7.0% 1.1% -0.4% 9.7% 2.3%
USB 4.5% 0.0% 9.1% 7.6% 1.5% 0.6% 0.7% 1.2% 2.5% 7.6% 8.5% 7.0% 0.9% -0.6% 9.0% 1.4%
ZION 4.5% 0.0% 12.1% 8.5% 3.6% 0.3% 0.8% 0.3% 2.5% 7.3% 10.5% 7.0% 3.2% -0.3% 12.2% 4.9%
Source: BofA Merrill Lynch Global Research, company data, Federal Reserve
Note: Stress test losses based on 2017 DFAST results which compares the CET1 ratio at 4Q16 to the Fed’s projected minimum CET1 ratio under the severely adverse scenario. Planned dividends based on actual 2017
capital ask. “Target” CET1 ratio denotes management’s publicly disclosed optimal CET1 ratio (except CMA which we est.)
Below, we look at the estimated tier 1 leverage requirement under the Fed’s proposal
(see Exhibit 14 and Exhibit 15).
As we will discuss later in the report, the proposed tier 1 leverage requirement using
the stressed leverage buffer may be more binding than the proposed CET1
requirement using the stressed capital buffer for MS and STT. Among non-GSIBs,
MTB’s excess capital also may be more restricted by new tier 1 leverage
requirements.
We believe, however, this proposal would drive more volatility in spot capital
requirements. After all the stress test itself is inherently volatile; while the stressed
capital burn has generally narrowed since the CCAR’s inception in 2011, the 2018 test
appears to be more challenging than 2017. (Please see out previously published notes
here and here). An example of how volatile the test is can be seen in the Exhibit below,
which looks at what JPM’s SCB would have been in previous tests – with a range of
2.5%-4.0% (see Chart 38). Such volatility could lead to wider capital “buffers” imposed
by management teams to account for the dynamism of the stress test – which would be
negatively viewed by investors.
Chart 38: JPM CET1 requirement including the stress capital buffer shows volatility of methodology
~12.0%
~11.0%
~10.5%
4.0%
3.0% 2.5%
Based on 2017 DFAST results, regional banks could harvest, on average, ~200bp
more in excess capital when comparing potential pro forma requirements under SCB
to their own public CET1 targets
The proposal would replace the current 2% leverage buffer with another buffer equal
to 50% of the bank’s GSIB surcharge, in line with the Basel proposals from
December 2017. This results in a lower capital requirement for all US GSIBs. All US
GSIBs are comfortably above the current SLR minimum, and this proposal can
potentially offer increased balance sheet flexibility. However, since the SLR would
not be the ultimate binding constraint for any US GSIB in a new capital
regime, the amount of excess capital for shareholder distribution unlocked
by this proposal is limited.
Chart 39: Current vs. proposed SLR minimum requirement for US GSIBs
8.0% Current SLR min.
requirement (US),
Pro forma SLR min. requirement (US)
7.0% 5.0%
6.0%
5.0%
4.0% 1.75% 1.50% 1.50% 1.25% 1.25% 1.00% 0.75% 0.75%
3.0%
2.0%
1.0%
0.0%
JPM C MS GS BAC WFC BK STT
SLR min. requirement (3%) Proposed buffer (50% of GSIB surcharge) SLR ratio (1Q18)
Source: BofA Merrill Lynch Global Research, company data, Federal Reserve
The proposal also addresses the 6% requirement at the bank sub, or covered IDI.
Similarly, covered IDIs would be required to maintain a leverage buffer equal to 50% of
the GSIB surcharge (applicable to the BHC) over the 3% SLR minimum. This compares
to the current 6% threshold to be considered “well capitalized.” If applied, this would
free an average of 260bp of capital at the bank sub or IDI level (see Exhibit 16). This is
not insignificant for client activity, as many BHCs warehouse a significant amount of
client activity in their IDIs.
Exhibit 16: Pro forma supplementary leverage requirements for covered IDIs
KEY C=AxB A B D E F=Ex50% G=3%+F H=B-D I=B-G
($bn) T1C Leverage Bank Current GSIB 50% of New Current Pro forma
capital exposure SLR req surcharge surcharge minimum excess excess
JPM 184 2,759 6.7% 6.0% 3.5% 1.8% 4.75% 0.7% 1.9%
C 127 1,901 6.7% 6.0% 3.0% 1.5% 4.50% 0.7% 2.2%
BAC 151 2,065 7.3% 6.0% 2.5% 1.3% 4.25% 1.3% 3.0%
WFC 143 2,010 7.1% 6.0% 2.0% 1.0% 4.00% 1.1% 3.1%
GS 25 346 7.3% 6.0% 2.5% 1.3% 4.25% 1.3% 3.1%
MS 15 168 9.1% 6.0% 3.0% 1.5% 4.50% 3.1% 4.6%
STT 16 234 7.0% 6.0% 1.5% 0.8% 3.75% 1.0% 3.3%
BK 20 296 6.7% 6.0% 1.5% 0.8% 3.75% 0.7% 3.0%
Est. excess tier 1 capital ($bn) 95 255
Est. change in excess tier 1 capital (pro forma standard vs. current) ($bn) 160
Source: BofA Merrill Lynch Global Research, company data, FFIEC call reports (bank subsidiary).
That said, regulators are considering applying the eSLR standard as a capital buffer
requirement. In other words, a 3% SLR would still be considered “adequately
capitalized,” but there would no longer be a threshold for the IDI to be considered “well
capitalized.” Instead, the eSLR standard would be applied to a covered IDI alongside the
existing capital conservation buffer in the same manner that the eSLR standard applies
to GSIBs. Thus under this alternative approach, GSIBs and their IDIs would be required
to maintain a leverage buffer equal to 50% of the GSIB surcharge or the GSIB HoldCo of
the covered IDI over the 3% SLR minimum.
The Department of Labor’s (DOL) Fiduciary Rule which was finalized in April of 2016
could potentially be struck down. The 2016 rule was slated to go into full effect July 1,
2019 (some parts took effect in 2017); however, in March of 2018 the 5th Circuit Court
of Appeals ruled that the DOL exceeded its authority in promulgating its Fiduciary Rule.
The DOL has until June 13th to challenge this decision in front of the Supreme Court,
though most see this as unlikely, including us, given the DOL and Department of
Justice’s (DOJ) lack of action in appealing the 5th Circuit's opinion since it was given in
early March. However, more recently, the SEC has put out its best interest proposals,
which we expect more likely to get finalized over time.
In December 2017, the Federal Reserve released a package of three proposals centered
on increasing the transparency of its stress testing program, to be applicable to the
2019 DFAST/CCAR process. We summarize them below:
• Enhanced model disclosure. The Fed would provide participating BHCs more
detail about the structure of the supervisory models (e.g., actual equations
calculating credit losses), provide estimated loss rates by loan or asset
category, and publish hypothetical portfolios of loans and assets with
accompanying hypothetical losses, essentially giving BHCs greater insight to
the Fed’s loss forecast.
In the 2017 DFAST, the average difference between the credit provision modeled by
the Fed and the credit provision modeled by the BHC (in the “co-run” model) was
2.5%, with a range of 0.1% to 7.8%. As such, we believe enhanced disclosure on
PPNR, or pre-provision net revenue (net operating income before credit, essentially)
could be more meaningful. The average difference in PPNR between what was
modeled by the Fed and what was modeled by BHCs in the 2017 DFAST was 28%,
with a range of -35% to 98%.
In today’s stress test regime and under the current proposal, participating banks do
not and will not privately receive any more stress test-related information than what
is publicly disclosed in the results.
In August 2017, the OCC announced that it would begin soliciting public feedback on
potential changes to the Volcker rule. In a notice published in The Federal Register, the
OCC noted that “there is broad recognition that the final rule should be improved both
in design and in application.” Specifically, the OCC put forth a set of questions for the
banking industry to address, focused on the following topics:
• Scope of entities subject to rule. As we detail later in this report, banks <$10bn
in assets are subject to compliance. The set of questions set forth under this
subject matter appear to be seeking input on limiting the scope of application.
The 60-day rebuttable presumption is a rule that states that if banking entity sells a
position or transfers the risk within 60 days, dealer must demonstrate that it did not
purchase or sell the security for short-term trading purposes.
The bill would allow a money market fun, under specified Ordered to be
Consumer Financial Choice and Capital conditions, to elect to operate using a different method of reported to To be voted on No estimate
H.R. 2319
Markets Protection Act of 2017 valuation. If elected, a money market fund shall not be subject to House by House available
specified requirements related to the imposition of liquidity fees. (1/18/18)
Economic Growth, Regulatory Relief, & Consumer Protection Act (S. 2155)
Status: Passed by Senate on March 2, 2018. Pending in the House. News reports
indicate passage before the U.S. Memorial Day holiday (Monday, May 28, 2018).
This bill would lift the asset threshold that defines a US SIFI to $250bn from $50bn (see
Exhibit 18). Banks below $250bn in assets would no longer have to go through the
DFAST/CCAR process, or submit living wills. (We explain what living wills are beginning
on page 160). However, the Fed would have the ability to apply tougher standards for
banks between $100bn and $250bn in assets.
This would provide the Federal Reserve with sole rulemaking authority to exclude
community banks (defined as those with $10bn in assets or less) from Volcker Rule
requirements.
• 5/15/18: The Fed and other regulators are planning to eliminate an assumption
written into the original Volcker Rule law that positions held by banks for less
than 60 days are speculative and thus banned; overhaul expected by end of May
• 4/19/18: Fed Governor Lael Brainard said she supports the banking agencies’
ongoing efforts to streamline the Volcker Rule; agencies are focusing on ways
to “tailor the Volcker compliance regime to focus on firms with large trading
operations and reduce the burden for small banking entities”
• 4/17/18: Fed’s Quarles agrees the Volcker Rule hurts capital markets
• 4/13/18: House passes bill to give the Fed exclusive authority to implement the
Volcker rule (H.R. 4790)
• 3/21/18: The House Financial Services Committee holds a markup on eight bills,
including a measure to exclude community banks from the requirements of the
Volcker Rule
This bill has three primary proposals. First, it looks to eliminate one of the three legally
required stress test scenarios (base case, adverse, severely adverse). Second, it would
bar the Fed from objecting to capital plans on qualitative reasons. This is in direct
contrast to the Fed’s SCB proposal, which reaffirms its authority to object on qualitative
reasons. And third, it would look to reduce the frequency of the co-conducted tests.
Again, this appears to be in direct contrast with the Fed’s SCB proposal, where the
proposal was written with the implication that the stress test would remain an annual
exercise.
Key provisions in this bill directly contradict the Fed’s SCB proposal
This bill would require regulatory agencies to take into account risk profiles and
business models of institutions when taking regulatory action.
This bill would bar regulatory agencies from establishing a capital requirement for
operational risk based solely on past events. It would require the operational risk capital
requirement to be based primarily on current activities and forward-looking
assessments of potential losses. On the surface, this bill would drive meaningful change,
as US GSIBs hold $1.9tn in operational risk RWAs, representing an average of 30% of
(50)
WFC STT
(100)
JPM
(150)
(200) USB Standardized CET1
(250) is low er, binding
(300)
15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0%
Operational risk as % of total RWAs
Source: BofA Merrill Lynch Global Research, company data, SNL Financial
Consumer Financial Choice & Capital Markets Protection Act of 2017 (H.R.
2319)
Status: Ordered to be reported to the House on January 18, 2018
This bill would allow money market funds, under specified conditions, to elect to operate
using a different method of valuation. If the election is taken, a money market fund shall
not be then subject to specified requirements related to the imposition of liquidity fees.
This bill would modify the definition of a High Volatility Commercial Real Estate loan
(“HVCRE”). The HVCRE classification requires higher levels of capital held against the
related loans, relative to “standard” CRE loans. Below are the banks in our coverage
universe, ranked by total CRE (not HVCRE) exposure to total loans (see Chart 41).
Chart 41: CRE loans as a % of total loans
35%
31%
30%
26%
CRE as % of total loans
25%
20%
16%
15%
15% 14% 13%
11% 10% Median, 10%
10% 9% 9% 8%
10% 8% 7%
5% 3%
0%
MTB ZION BBT KEY CMA PNC WFC JPM USB HBAN CFG FITB RF STI C
The bill includes provisions that would eliminate regulatory authority to designate firms
as SIFIs. It also would look to repeal the Volcker Rule. Further, it would subject the CFPB
(Consumer Financial Protection Bureau) to the discretionary appropriations process and
reduce its regulatory authority.
• Which assets are eligible for inclusion in an Advanced Internal (IRB) model
• And finally, the minimum ratio of a Standardized model which a bank’s IRB models
can deliver was set at 72.5%
As a reminder, in the U.S., the Collins Amendment requires that the binding CET1
constraint for each institution would be the lower value under the standardized or
advanced approach. As a result, U.S. banks have already been unable to take advantage
of using a model-based calculation of RWAs to incur a lower requirement.
Buy-to-let mortgages
For UK banks, the rise in Buy to let capital requirements shown in Table 4could limit the
benefit of potential adoption of advanced models over the next two to three years.
Table 4: Income producing Real Estate: Risk Weight
Risk weight - Basel II 35% 35% 35% 35% 75% 75% 75%
LTV Bands - 2015 proposal ≤60% >60-80% >80%
Risk w eight - 2015 proposal 70% 70% 70% 90% 120% 120% 120%
LTV Bands - Final proposal ≤50% >50-60% >60-80% >80-90% >90-100% >100%
Risk w eight - Final proposal 30% 30% 35% 45% 60% 75% 105%
Change vs. Basel II -5% -5% 0% 10% -15% 0% 30%
Change vs. 2015 proposal -40% -40% -35% -45% -60% -45% -15%
Source: Basel Committee, BofA Merrill Lynch Global Research estimates
Chart 42 shows that banks will see the proportion of their assets eligible for advanced
internal models fall from 80% to below 60%. This comes from the slightly obscure
reasoning that on assets where defaults are so low as to be difficult to model
accurately, a much higher minimum default probability must be assigned.
This will of course reduce the opportunity for banks to “optimize” models and reduce
Risk Weights accordingly. However, for genuinely low risk assets, banks are likely to lose
share for years to non-bank competitors.
Chart 42: Under final Basel proposals, foundation IRB model doubles in importance (%)
90
80
70
60
50
Current
40
Final
30
20
10
0
Advanced IRB Foundation IRB Standardised, other
-1
-2
-3
-4
Source: Basel Committee QIS
The operational risk improvement in turn will only occur if all national supervisors use
their discretion to set the internal loss multiplier to one. That is, the new rules state
clearly that a bank’s operational risk requirement will be influenced by a “Loss
Component”, of 15x average annual operational risk losses incurred over the previous 10
years. A bank with high losses “is required to hold higher capital due to the
incorporation of internal losses into the calculation methodology”.
See the below example from the Basel committee’s paper. Note the final column shows
that while credit RWA sees an increase, there is an offset from portfolios that do not
have a standardized approach, in particular operational RWA. Operational RWA may well
increase for banks, but we see the magnitude as much lower for banks where there have
already been large “add-ons” for the banks already.
Table 6: Basel Committee example of how the output floor will operate in aggregate
72.5% of standardized Delta vs. pre-
Pre-floor RWA Standardized RWAs floor RWA
Credit risk 62 124 89.9 27.9
o/w asset class A 45 80 58 13
o/w asset class B 5 32 23.2 18.2
o/w asset class C (not
modelled) 12 12 8.7 -3.3
Market risk 2 4 2.9 0.9
Operational risk (not modelled) 12 12 8.7 -3.3
Total RWA 76 140 101.5 25.5
Source: BIS, BofA Merrill Lynch Global Research estimates
Market RWA (fundamental review of the trading book), CVA (move to standardized
measure of counterparty credit risk or SA-CCR) and the operational RWA changes also
had the implementation deadline moved to coincide with the other Basel 4 reforms on
1st January 2022.
The conclusion of the Fundamental Review of the Trading Book (FRTB) shifts the
calculation of market risk from a Value-at-Risk approach to Expected Shortfall. It also
requires that internal models are approved at desk level. The basic, standardized
calculation of market risk has also been overhauled.
• A new standardized approach. This will capture the risk more comprehensively,
making it a more credible fallback option to the internal models.
• Shift from value-at-risk (VaR) to expected shortfall (ES). In simple terms, whereas
VaR may show that “19 days out of 20 I will not lose more than $X,” an expected
shortfall approach says “on the one day in 20 that I do lose more than $X, how
much is that likely to be?”
• A new liquidity horizon factor. Basel acknowledges that a standard 10-day horizon
in the old framework was too standardized to be valid for different instruments.
• Redefinition of the trading book / banking book boundary. The intention here is to
remove subjectivity to where banks book business.
The consultation proposes changes to both the standardized and internally modeled
approaches.
Standardized: lightening up
The new standardized approach to Market risk was set according to sensitivities. This is
an estimate of how much a bank’s trading portfolio would change for a certain assumed
move, for example a 1bp move in interest rates. The methodology uses a set of defined
correlation assumptions to provide a diversification benefit across different risks.
In March 2018, the BCBS published a consultation paper that sought to fix some of the
potential problems with the new standardized approach. These were generally designed
to fix some potential flaws in the original proposal and to make the standardized
approach a more credible fall-back option.
1. Liquid FX pairs. More liquid FX pairs were given a benefit in risk weights.
However, the committee realized that while for instance USD/EUR and
4. Risk weights reduced. The committee discovered that the original proposal
was generating higher RWA inflation than intended. The committee has
therefore proposed to reduce several risk weights: general interest rate risk by
20-40% and equity and FX risks by 25-50%.
The original changes allow a regulator to force a bank back onto the standardized
approach if the internal models do not work. This led to some fear that such an
automatic move could introduce a cliff-effect to bank capital requirements.
The new proposal would use a traffic light system with green for a pass, red for a failure
and fall-back onto the standardized approach, but with an amber group for those models
that have not met all the requirements but not at such a level to necessitate an
automatic fall-back. The desks in the amber group will have an additional capital charge
between the internal model and standardized model.
As with credit risk, there is currently the option to run with either a standardized model
(which is based upon gross income, also known as SMA), or internal models called the
advanced measurement approach, or AMA. Many of the large global banks use internal
models for the majority of operational RWA. In the US, the Federal Reserve Board had
initially set “floors” for its eight GSIBs, or minimum operational risk RWA.
Internal models had underestimated operational risk, with litigation risk costs being
materially larger than such models had predicted.
Basel now suggests a single, standardized methodology, removing the ability of banks to
model operational risk internally. The model has two elements:
• A business indicator component (BIC), which is a standardized factor that sums
three components ((i) interest, leases and dividends, (ii) services and (iii) the
financial component. Each is given a coefficient which increases the larger the
bank’s revenues are.
Below, we show that European IB operational RWA have already been increasing
significantly in the past seven years.
Chart 44: Operational RWA inflation at European IBs (USDbn, FX neutral)
350
DBK CS UBS BARC
300
250
200
150
100
50
0
2011 2012 2013 2014 2015 2016 2017
Source: BofA Merrill Lynch Global Research, company data
impact
15%
13%
11% 15.93%
9%
11.26%
7% Regulatory related
5%
2006 Lower rates Higher Remixed loan Higher capital Balance sheet Lower Higher fee Higher Regulatory 2016
cash/sec portfolio levels growth provisions income expenses (ex- costs
balances reg)
Note that we estimate that US corporate tax reform is ~200bp additive to bank
ROTCEs
We have been asked many times why the revenue impact from regulation is not more
negative. We note that a likely unintended consequence of regulation is concentration.
The largest three US banks now hold 36% of deposits vs. 19% pre-crisis (see Chart 46
and Chart 47). And of course, a bigger balance sheet means bigger spread revenues,
even in a challenging rate backdrop. Further, as we discuss in detail later in this section,
US banks have also gained significant market share globally in investment banking and
trading over the last five years.
Top 3, 19.3%
Top 3, 35.7%
Rest, 64.3%
Rest, 80.7%
Key definition
Here’s an example of why binding constraint is a crucial concept. Let’s take JPM, for
example (see Chart 48). Under the BHC-level supplementary leverage ratio, or SLR, JPM
would have excess capital equivalent of $48.5bn under the current regime. Meanwhile,
under its tailored spot CET1 requirement, JPM would only have $20bn in excess capital.
As such, JPM’s binding constraint is the CET1 requirement, and, specifically, the GSIB
surcharge requirement within CET1. (We go into the components of the CET1
requirement in detail beginning on page 94).
Chart 48: How to compare bank compliance with regulatory requirements (example: JPM)
CET1
At 100% , a bank 150%
is fully compliant A ratio under
with its regulatory 100% represents
minimum, and a shortfall and
implies efficient could negatively
100% impact future
use of capital or
funding TLAC NSFR strategy. Banks
that fall short on
regulatory ratios,
50% regardless of
time to
A ratio too far compliance,
beyond 100% typically feel
represents pressure from
ex cess, under- shareholders to
utilized capital or more
immediately
funding, weighing LCR SLR close the gap.
on returns
1Q17 1Q18
Key definition
Chart 49: We est. US GSIBs have ~$115bn of excess capital today (6% of market cap) as of 1Q18
Under current regulatory framework
Current binding constraint:
Spot CET1 Spot CET1 Spot CET1 Spot Tot Cap Stressed T1L Stressed SLR Stressed T1L
14%
13%
5%
4%
2% 1%
0%
Source: BofA Merrill Lynch Global Research, company data, SNL Financial
Note: Excess capital converted into common equity tier 1 capital for comparability purposes.
Of course, as we laid out in the previous section, the Fed recently proposed new rules
governing both spot and post-stress capital requirements. Note that banks would no
longer quantitatively “fail” CCAR, but the amount of capital burned in the test would be
part of spot requirements. This would have the impact of shifting binding constraints
for certain institutions.
Interestingly, based on the 2017 DFAST results, GSIBs would have $25bn less in
excess capital under the proposed regime than in the current regime
Chart 50: We est. US GSIBs would have ~$90bn of excess capital today (5% of market cap)
Under Fed-proposed requirement changes
Potential binding constraint under new Fed proposals:
Stnd CET1 Stnd CET1 T1L Total Cap Stnd CET1 T1L Stnd CET1
11% 11%
5%
3% 3% 2% 1%
Source: BofA Merrill Lynch Global Research, company data, SNL Financial
Note: Excess capital converted into common equity tier 1 capital for comparability purposes
100%
TLAC NSFR
50%
European GSIBs currently
ex hibit the narrow est
buffer on SLR v s.
minimum requirements
LCR SLR
1Q17 1Q18
75%
70%
65%
60%
55%
50%
2010 2011 2012 2013 2014 2015 2016 2017 2018
US Europe
To illustrate the tension between the two requirements, we examined the banking
products that are considered “more valuable” and “less valuable” under CET1 or SLR
(see Exhibit 19). Repo is a perfect example of conflicting product-level value. Under
CET1, repo is considered a “more valuable” product given its low risk weight. However,
under SLR, repo becomes less valuable, as the exposure is not “discounted” by its low
risk profile.
If a firm is constrained by post-stress test capital minimums rather than spot, this does
not change product-level value (e.g., GS, who is currently constrained by post-stress
SLR). However, the CCAR overlay could be an exacerbating factor in terms of capital
management. For example, a management team may hold more capital against
exposures and/or in product hurdle rates than what is required by spot capital minimums
to take into account the volatility of the stress test results.
In the graphic following this paragraph, we group global GSIBs relative to their binding
constraints. Note that certain firms, like MS, are not bound by SLR but by tier 1
leverage. We grouped the GSIBs by type of binding constraint: risk-sensitive (CET1, total
risk based capital) and risk-insensitive (SLR, tier 1 leverage).
CET1 SLR
CCAR
(US Only)
More Valuable Products
• Historically low loss
content loan products
(e.g. prime mortgage)
Note that in the U.S., NSFR is not finalized and therefore not binding. Meanwhile, also
for U.S. banks, resolution planning – also known as the “living will” process – can be the
binding constraint for liquidity over LCR.
Our section on LCR begins on page 139, NSFR on page 147, and TLAC on page 120. Our
discussion on resolution planning begins on page 160.
Exhibit 20: More- and less-valuable financial products under LCR, NSFR, and TLAC constraints
$1,000
$800
$600
$400
$200
$0
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
On Balance Sheet Credit Card Loans Credit Card ABS
Unsecured consumer loans (ex-credit card) have also grown on balance sheet, at a 4%
CAGR since 2010 in the US (see Chart 55). Given shrinking home equity portfolios, a
handful of institutions have partnered or purchased “fintech” companies that originate
loans digitally in order to provide a pipeline of consumer loans. Examples include loans
for debt consolidation, home renovation, life events (weddings), elective surgery, etc. Of
course, others, like GS, have built their own technology and capabilities to originate
unsecured consumer loans.
Chart 55: US Consumer loans ex credit cards have grown at a 4% CAGR since 2010
850,000
Consumer loans (ex credit cards)
Consumer loans ex credit cards ($mn)
750,000
700,000
650,000
600,000
4Q10
1Q11
2Q11
3Q11
4Q11
1Q12
2Q12
3Q12
4Q12
1Q13
2Q13
3Q13
4Q13
1Q14
2Q14
3Q14
4Q14
1Q15
2Q15
3Q15
4Q15
1Q16
2Q16
3Q16
4Q16
1Q17
2Q17
3Q17
4Q17
Source: FDIC
Auto loans have been recently de-emphasized. But, this is another example of a 100% risk
weight asset that had contributed strongly to loan growth until recently (see Chart 56).
First, we look at overall fixed income and equities market share relative to CET1
excess (see Chart 57and Chart 58).
Chart 57: Fixed income trading market share vs. CET1 compliance Chart 58: Equity trading market share vs. CET1 compliance
MS
160% 160%
CET1 ratio relative to requirement (2017)
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
Further, we look more deeply at global market share for the top 12 banks (under
BofAML coverage) in the following CET1-constrained products: 1) equity derivatives;
2) securitized products; 3) credit (although high yield is more impactful to CET1 than
investment grade); and 4) commodities.
Given shrinking client wallets in cash equities, equity derivatives have been a
significant contributor to equity revenues of late (see Chart 59). JPM, GS, and C all rank
in the top 3 globally and all have excess capital under CET1 under the current regime.
150%
140% UBS CS
WFC
130% C
GS
120% JPM HSBC
SocGen
110% DBK
BNP BARC
100%
90%
80%
70%
Stay the cour se Potential r etr enchment
60%
0 2 4 6 8 10 12 14
EQUITY DERIVATIVES: Global investment bank league table r ank (2017)
In securitized products, the story at the top is more interesting (see Chart 60). JPM
and C are in the top three, and we of course similarly expect these institutions to defend
their turf. DBK also ranks in the top 3 globally, but has less excess capital and of course
has publicly stated that it would shrink its footprint outside of Europe – suggesting it
has potentially share to give. Interestingly, MS has significant excess CET1 capital and
falls just outside the top five globally, in our view, implying a potential market share
opportunity here.
Chart 60: Securitization revenue market share vs. CET1 compliance
MS
160%
CET1 r atio r elative to r equir ement (2017)
150%
Mar ket shar e oppor tunity
140% UBS
CS
WFC
130% C
GS
120% JPM HSBC
110% DBK
BARC SocGen
BNP
100%
90%
80%
70%
Stay the cour se Potential r etr enchment
60%
0 2 4 6 8 10 12 14
SECURITIZATIONS: Global investment bank league table r ank (2017)
In credit, the story at the top is also similarly interesting (see Chart 61). DBK ranks
second globally (tied with JPM), but again could be market share contributors outside of
Europe. C and GS round out the top 5, and C in particular could be another aggressive
competitor here by using balance sheet excess to drive market share higher, in our view.
While we do not have trading market share isolated to high yield – where CET1 is most
meaningful – we looked at recent share (isolated to BofAML coverage universe) of high
yield issuance as a proxy. Not surprisingly, C ranked in the top 3 (see Exhibit 22).
150%
Mar ket shar e Goldman Sachs 4 72,350 8.74%
140% UBS oppor tunity Barclays 5 61,891 7.48%
CS
WFC Deutsche Bank 6 53,962 6.52%
130% C
GS Credit Suisse 7 53,246 6.43%
120% JPM HSBC
Morgan Stanley 8 52,413 6.33%
110% DBK Wells Fargo 9 44,276 5.35%
BARC SocGen RBC Capital Markets 10 34,887 4.21%
BNP
100% UBS 13 15,574 1.88%
90% Source: BofA Merrill Lynch Global Research, Bloomberg
80% Note: Data represents three year time horizon (2014-2017)
Stay the Potential
70% cour se r etr enchment
60%
0 2 4 6 8 10 12 14
CREDIT: Global investment bank league table r ank (2017)
And lastly, in commodities, we don’t expect major shifts to the top five global players,
all of whom are in positions of CET1 strength (see Chart 62. Note that GS’s
restructuring of its commodities business is largely completed, though we can see
ongoing changes based on market conditions.
Chart 62: Commodities trading revenue market share vs. CET1 compliance
160%
CET1 r atio r elative to r equir ement (2017)
150%
MS Mar ket shar e oppor tunity
140% UBS CS
WFC
130% C HSBC
GS
120% JPM
SocGen
110% DBK
BNP BARC
100%
90%
80%
70%
Stay the cour se Potential r etr enchment
60%
0 2 4 6 8 10 12 14
COMMODITIES: Global investment bank league table r ank (2017)
Exhibit 23: Many assets experienced record jumps in recent years from calm to stress
10 Since '28 Since '87 Since '90 Since '98 Since '73 Since '75 Since '62 Since '89 Since '86
Largest and 2nd
8 largest in history
Largest and 2nd
6 largest since 1973
# of stdevs relative to 100d MA
2
SPX SX5E GBPUSD Copper
0
VIX US USDJPY 10y UST Bund
-2 Financials yields
Largest in history
-4
27-Jun-16
24-Jun-99
14-Jan-15
06-Jun-96
09-Feb-18
05-Feb-18
27-Feb-07
24-Feb-12
10-Mar-76
19-Mar-91
19-Mar-12
12-Feb-90
02-Mar-94
11-Mar-14
24-Aug-15
08-Aug-11
31-Aug-98
06-Aug-90
10-Aug-11
24-Aug-15
08-Aug-11
17-Sep-01
14-Nov-16
11-Nov-05
07-Aug-00
02-Nov-73
10-Sep-14
15-Aug-95
21-Nov-12
13-Aug-08
28-Sep-76
14-Nov-16
11-Nov-16
26-Oct-87
14-Oct-03
28-Apr-06
13-Apr-87
30-Aug-72
16-Apr-93
23-Sep-11
14-May-40
19-Oct-87
10-Oct-79
20-Jul-92
Source: BofA Merrill Lynch Global Research. Daily data of the SPX from 30-Dec-27 to 2-Dec-15, VIX from 2-Jan-90 to 2-Dec-15, SHCOMP
from 31-Dec-90 to 2-Dec-15, USDJPY from 31-Mar-71 to 2-Dec-15, EURCHF from 31-Mar-99 to 2-Dec-15, USGG10YR from 30-Mar-62 to 2-
Dec-15, GDBR10 from 31-Mar-89 to 2-Dec-15 and XAU from 31-Mar-20 to 2-Dec-15. We measure this one day dislocation as the magnitude
of the level relative to its trailing 100d moving average in standard deviations. If we see records set on subsequent days, we only record the
largest as representing that event. For USGG10YR, we use the intraday move on 15-Oct-14.
One of the drivers of market volatility, in our view, has been the significantly lower
market liquidity today, especially when compared to pre-crisis levels. All the
aforementioned rules have had a meaningful impact on bank/broker balance sheets –
particularly trading portfolios – which have declined meaningfully since the financial
crisis (see Chart 63). Rationalization of certain products has directly impacted the
liquidity for those assets. We expect the impact on market liquidity to continue as
recent proposals from the Fed imply limited regulatory relief for US GSIBs.
Chart 63: Bank trading portfolios have declined since the crisis
16
14 -46%
12
($ in trillions)
10
8
6
4
2
0
2009
2010
2011
2012
2013
2014
2015
2016
SLR is particularly punitive to repo exposure, and leverage exposure as calculated in the
SLF requirement feeds into each GSIB’s related surcharge calculation for CET standards.
Along with funding rules (LCR globally, resolution planning in the US), these could have a
negative impact on large dealers’ ability to intermediate between cash lenders (such as
Note that the size of the primary dealer repo market has already shrunk by nearly 50%
since its peak in 2008; while primary dealer debt trading volumes have declined 30%
over the same period (see Chart 64). A further reduction in the availability and an
increase in the cost of dealer repo would likely translate into lower trading volumes and
subsequently, revenues.
That said, given material progress on leverage capital by certain global firms,
especially in the EU, repo availability has steadied and we expect it to remain steady
from here.
Chart 64: Repo market and dealer volumes have significantly declined
$5,000 Primary Dealer Balance Sheet and Trading Volumes Declined ($bn) $1,600
$1,500
$4,500
$1,400
$4,000 $1,300
$3,500 $1,200
$1,100
$3,000 $1,000
$2,500 $900
$800
$2,000
Period of stability $700
$1,500 $600
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Primary dealer repurchase activit y (Liability, lhs) Total fixed income volume (rhs)
Source: BofA Merrill Lynch Global Research, New York Federal Reserve
Note: 3m moving average
Issuance has been robust, particularly in Treasuries and corporate debt. However, while
the market has grown, the capacity to intermediate this incremental volume has
declined. After all, regulations have constrained dealers’ ability to trade those securities,
as overall dealer assets have declined (see Chart 66).
Mar-13
Jan-04
Jan-09
Jan-14
Jul-06
Jul-11
Jul-16
May-07
May-12
May-17
Nov-04
Sep-05
Nov-09
Sep-10
Nov-14
Sep-15
- 0%
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Dealer Fix ed Income Inventories (LHS, $bn) Coporate Bond Fund Assets (LHS, $bn)
Ratio of Real Money FI Holdings to Dealer FI Holdings (RHS) Dealer Corp Inventory (LHS)
Dealer Inv./Corp Bond Funds (RHS)
Source: Federal Reserve Source: Bloomberg, ICI Global
From a product standpoint, these rules could have continued adverse consequences for
market liquidity in U.S. Treasuries, agency debt and MBS. These products are not only
balance-sheet intensive under SLR (products have low risk weight), but comprise the
bulk of U.S. repo collateral.
Jan-15
Jan-16
Jan-17
Jan-18
May-13
May-14
May-15
May-16
May-17
Sep-13
Sep-14
Sep-15
Sep-16
Sep-17
First, we look at overall fixed income and equities market share of twelve
international banks relative to SLR excess (see Chart 68 and Chart 69).
Chart 68: Fixed income trading revenue mkt share vs. SLR compliance Chart 69: Equity trading revenue mkt share vs. SLR compliance
160% WFC 160% WFC
Leverage ratio relative to requirement
(2017)
100% CS UBS 100% UBS CS
90% DBK 90% DBK
80% 80%
Stay the Potential Stay the Potential
70% cour se
70% cour se r etr enchment
r etr enchment
60% 60%
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
FICC global investment bank league table rank (2017) Equities global investment bank league table rank (2017)
Source: BofA Merrill Lynch Global Research, company data, Coalition Source: BofA Merrill Lynch Global Research, company data, Coalition
On a product basis, we looked at the following products most constrained by SLR where
we can obtain market share data: 1) equity derivatives; 2) prime services; 3) credit
(converse to CET1, investment grade is more impacted than high yield); and 4) rates.
The market share dynamics in equity derivatives is even more interesting when
observed through the SLR lens (see Chart 70). French banks BNP and SocGen have more
excess SLR than excess CET1; both fall just under the top five, implying that both firms
could be continue to be notable competitors in this area.
Chart 70: Equity derivative trading revenue market share vs. SLR compliance
160% WFC
Lever age r atio r elative to r equir ement (2017)
150%
140% C
JPM BARC
MS HSBC
130%
120% GS SocGen
110% BNP
Mar ket shar e oppor tunity
100% UBS CS
90% DBK
80%
70%
Stay the cour se Potential r etr enchment
60%
0 2 4 6 8 10 12 14
EQUITY DERIVATIVES: Global investment bank league table r ank (2017)
Within credit, investment grade would be the most impacted. Again, we expect JPM and
C to defend its overall market share (IG + high yield) in credit, with GS and MS
potentially using their excess SLR capital to gain market share in investment grade.
Again using issuance as a proxy for overall investment grade market share, WFC could
be an interesting contender here to move market share (see Chart 72). While WFC is
under an asset cap with the Fed, cash and securities comprise a significant amount
(45%) of earning assets, which implies to us that it could continue to compete with
client business – especially in wholesale banking – without breaching the asset cap.
Chart 72: Credit trading revenue mkt share vs. SLR compliance Exhibit 24: Investment grade bond issuance league table
Revenue
160% WFC Investment grade bond issuance Rank ($mn) Share (%)
Lever age r atio r elative to r equir ement (2017)
In rates, the top four players are firmly entrenched US banks (JPM, C, MS, GS) with
significant excess leverage capital (see Chart 73). Again, DBK ranks fifth and could
potentially be a market share contributor outside of Europe.
150%
Mar ket shar e oppor tunity
140% C
JPM BARC
MS HSBC
130%
120% GS SocGen
110% BNP
100% CS UBS
90% DBK
80%
70%
Stay the cour se Potential r etr enchment
60%
0 2 4 6 8 10 12 14
RATES; Global investment bank league table r ank (2017)
Next up: Fed “quantitative tightening” + increase in net new US Treasury supply
We think there are three distinct factors that have driven material amounts of liquidity
build in US banks. First, there has been slower loan demand and less supply in certain
products (e.g., anything residential mortgage related), leading to loan growth that is slower
than in previous economic recoveries. Second, the accommodative backdrop around the
world has led to bank clients storing more cash at the banks in the form of deposits (see
Chart 74). Third, as we will discuss in depth beginning on page 139, new rules governing
liquidity coverage (LCR) also drove material build up in high quality liquid assets (HQLA)
among GSIBs and impacted US regionals (above $50bn assets) (see Chart 75).
Chart 74: Deposit growth is inversely correlated with rising short rates Chart 75: HQLA has grown 86% since ’09 vs. loan growth of 14%
20% -6% 200% +86%
-5%
15% -4% 180%
-3% 160%
(index = 2Q09)
10% -2%
(Inverted)
-1% 140%
5% 0% +14%
1% 120%
0% 2% 100%
3%
-5% 4% 80%
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
2009
2010
2011
2012
2013
2014
2015
2016
2017
Core Deposits YoY (lhs) Fed Funds YoY (rhs) HQLA Proxy Total Loans
Source: BofA Merrill Lynch Global Research, Federal Reserve, Bloomberg Source: BofA Merrill Lynch Global Research, company data, SNL Financial
Data: HQLA proxy includes cash at central banks, UST securities, and municipal securities
We think that US banks already have limited incremental appetite for more HQLA. After
all, Treasury securities and cash have been dilutive to net interest margin. Ginnie Mae
yields have been more reasonable, but some banks have been hesitant to extend
duration today in a rising rate backdrop in the US. Further, growing balance sheet
through HQLA is generally margin dilutive and consumes SLR capital. And under the
current regime, 3 out of the 8 GSIBs are bound by risk insensitive capital measures,
which is a disadvantage to low-risk assets like Treasury securities.
Chart 76: Annual Fed balance sheet normalization expectations Chart 77: Monthly Fed balance sheet normalization expectations
450 50
400 45
350 140
133 35
300 30
250 127 25
115
200 20
15
150
267 10
100 229
197 174 5
50 0
Apr '18
Oct '18
Apr '19
Oct '19
Apr '20
Oct '20
Apr '21
Oct '21
Jan '18
Jan '19
Jan '20
Jan '21
Jul '18
Jul '19
Jul '20
Jul '21
0
2018E 2019E 2020E 2021E
In our recent report titled, “Not all deposits are created equal: Potential impact of Fed
portfolio unwind”, we go through in detail the mechanics and potential impact of the
Fed’s balance sheet reduction. In the exhibits below, we lay out the mechanics of what
happens when the public (bank clients) buys new Treasury issue (see Exhibit 26) and
when the banks buy the new Treasury issue (see Exhibit 27). Note that when Treasury
securities mature from the Fed balance sheet and Treasury pays the Fed back, the Fed
destroys the proceeds from the Treasury – leading to balance sheet shrinkage. Of
course, the Treasury then issues new bonds to replace what has matured.
Public withdraws
deposits to buy Public buysnewly
Treasuries issued Treasuries
Public
Do higher stress test loss content for certain loans limit bank appetite?
Before we dive into this, we do want to recognize up front that economic growth in the
US during this recovery has been slower than previous 35 years (see Chart 78). Of course,
this impacts loan demand, which impacts on-balance sheet growth. Further, given that
residential mortgage lending drove the previous crisis, exposure reduction due to the run-
off of legacy loans and higher underwriting standards followed (see Chart 79).
Chart 78: GDP growth following periods of recessions
10.0%
8.0%
Avg: 4.0% Avg: 3.6%
6.0% Avg: 2.6%
US real GDP (YoY)
Avg: 2.0%
4.0%
2.0%
0.0%
(2.0%)
(4.0%)
(6.0%)
1980
1981
1982
1983
1985
1986
1987
1988
1990
1991
1992
1993
1995
1996
1997
1998
2000
2001
2002
2003
2005
2006
2007
2008
2010
2011
2012
2013
2015
2016
2017
Economic recovery
Source: BofA Merrill Lynch Global Research, Bloomberg, Bureau of Economic Analysis
1% 1% 1%
0%
-1% 0%
-2%
-3% -2%
-3%
-4%
-5%
-5%
-6% -5%
1-4 Family Total Construction HELOC Other cons C&I CRE Credit card
Let’s now look at the loan level stress test losses and the potential impact to growth for
the participants of the stress test (see Exhibit 28).
Exhibit 28: Results of the Fed’s stress test have led to reduction in certain loan exposures, in our view
Avg. projected loan losses Fed's results Company-run (bank models) results Delta: Co-run vs. Fed Historical industry results Delta: YoY loan growth
(% of average loans) 2014 2015 2016 2017 2014 2015 2016 2017 2014 2015 2016 2017 Peak Avg 14/13 15/14 16/15 17/16
First lien mortgages 4.3 3.2 3.5 2.4 2.4 2.4 1.6 1.7 (1.7) (0.7) (1.8) (0.7) 2.5 0.4 -3.0% -13.5% 0.0% -8.6%
Junior liens and HELOCs 7.7 6.3 6.1 4.2 5.1 4.7 4.4 4.2 (2.7) (1.5) (1.8) 0.1 3.1 0.5 -3.1% -5.8% -5.7% -8.6%
Commercial and industrial 5.0 5.0 5.9 5.9 4.0 4.1 4.6 4.3 (1.3) (1.2) (1.4) (1.6) 2.7 0.9 6.3% 5.8% 4.9% 2.0%
Commercial real estate 8.4 9.0 7.6 7.7 5.4 5.0 4.6 4.9 (3.0) (3.8) (3.0) (2.8) 2.0 0.3 2.5% 8.2% 6.6% -22.0%
Credit cards 14.3 12.8 12.9 12.8 15.9 16.4 13.1 14.2 0.3 2.9 0.2 1.4 13.2 4.4 6.0% 11.5% 4.7% 13.4%
Other consumer 5.9 5.7 5.9 5.9 4.2 3.5 6.0 4.6 (0.9) (1.4) 0.2 (1.3) 8.0 0.9 6.1% 5.4% 3.2% 4.7%
Other loans 3.0 3.0 3.5 3.7 2.6 2.4 2.2 2.2 (0.2) (0.4) (1.3) (1.4) - - 12.8% 18.4% 18.4% 9.5%
Source: BofA Merrill Lynch Global Research, Federal Reserve, company data
Note: Data set includes US banks that participated in the Fed’s annual stress test
As a reminder, the stress test requires the Fed and the banks to run separate
models. A bank’s model will largely determine the capital plan presented to the Fed
for non-objection, and the Fed’s model will determine the non-objection or objection
to a bank’s capital plan.
We dive more deeply into the DFAST/CCAR stress test process beginning on page 130.
Generally speaking, while we think the stress test does have some impact to bank
appetite, it is not likely the dominant factor in growth. When looking at residential
mortgage losses – both first lien and junior lien/home equity – the stress test does
appear to have a negative impact, when taken in isolation (see Chart 80). Further, new
mortgage rules under Dodd-Frank likely further curbed bank enthusiasm for this product.
∆ betw een Fed and co-run projected loan losses (lhs) YoY loan growth (rhs)
Source: BofA Merrill Lynch Global Research, Federal Reserve, company data
That said, we would mention a few other factors that are not insignificant to growth.
Following cycles, we observe a “recency bias” – in other words, banks tend to clamp
down underwriting standards for the loan product that had the highest loss experience
in the previous cycle. And of course, loan balances – particularly in home equity, not an
especially popular product among consumers at the moment – are impacted by
continued run-off of legacy portfolios.
In fact, given a largely absent secondary market for loans that do not conform to
agency (e.g., Fannie Mae, Freddie Mac) standards, banks have actually been more
aggressive at originating and retaining jumbo mortgage loans.
Second, the growth in commercial and industrial (C&I) loans are the strongest counter-
argument for CCAR results heavily influencing bank appetite. The Fed has consistently
forecasted higher losses in C&I, at a rate of ~1.5% higher than the average bank
forecast. And yet as we mentioned earlier, banks have consistently grown C&I loans (see
Chart 81). Commercial real estate (CRE) is another counterpoint, where losses projected
by the Fed were consistently higher, but this was also an asset class where loans were
growing until recently (see Chart 82).
Chart 81: Despite high loss assumptions, banks grow C&I loans Chart 82: CRE growth recently impacted by mkt conditions, not DFAST
2.0% 7.0% 2.0% Fed loss assumptions 10.0%
better than the banks'
5.0% 1.0% models 5.0%
Fed loss assumptions
1.0% better than the banks' 3.0% 0.0% 0.0%
models
1.0% (1.0%) Fed loss asasumptions (5.0%)
0.0% worse than the banks'
Fed loss asasumptions (1.0%) (2.0%) models (10.0%)
worse than the banks'
(1.0%) models (3.0%) (3.0%) (15.0%)
(5.0%) (4.0%) (20.0%)
(2.0%) (7.0%) (5.0%) (25.0%)
2014 2015 2016 2017 2014 2015 2016 2017
∆ betw een Fed and co-run projected loan losses (lhs) ∆ betw een Fed and co-run projected loan losses (lhs)
YoY loan growth (rhs) YoY loan growth (rhs)
Source: BofA Merrill Lynch Global Research, Federal Reserve, company data Source: BofA Merrill Lynch Global Research, Federal Reserve, company data
That said, many bank management teams have complained about high loss rate
assumptions for small business loans. While we cannot verify this independently from
publicly-available stress test results, we do note that the commercial loan outstanding
for loans less than $1mn have declined since 2010, at a 1% CAGR (see Chart 83).
Source: BofA Merrill Lynch Global Research, FDIC (share of total loans on bank balance sheets)
Note: Yellow bars denote periods of economic recovery
Capital return impact: US dividend payout lower vs. history, global peers
Of course, capital return is the part of US bank capital management that has been most
directly impacted by CCAR. Previous to this process, banks did not have to undergo such
a formal process for approval (or, in regulator parlance, “non objection”) for shareholder
return. That said, over time, total capital payout to shareholders by CCAR participants
has increased, peaking at 91% in 2017 (See Chart 84).
Chart 84: Total payout has continued to increase throughout each CCAR cycle
100%
91%
80% 75%
CCAR capital payout
(BofAML coverage)
62%
60% 54%
40%
20%
2014 2015 2016 2017
However, given the “soft cap” of 30% on dividend payouts (as a percentage of earnings),
dividend yields and payouts are still on the lower end of historical ranges. In terms of yield,
the sector was about in-line with the S&P 500 dividend yield in 2017 – pre-crisis (we look
at 2006), the banks were higher than the broad market (see Chart 85). In terms of payout,
the banks have the lowest LTM (last twelve months) payout across S&P sectors at 29%
(see Chart 86). This is also well below the pre-crisis dividend payout average of 45%.
200%
4%
S&P 500
Health Care
Utilities
Banks
Financials
Cons. Staples
Materials
InfoTech
Telecom
Cons. Discr.
Real Estate
Energy
0%
Cons. Discr.
Health Care
Financials
Inf o Tech
Cons. Staples
Utilities
S&P 500
Telecom
Industrials
Materials
Real Estate
Energy
Banks
2006 2017
Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: Universe consists of S&P 500 constituents Note: Universe consists of S&P 500 constituents
Further, on a global basis, US banks’ dividend yields lag that of their most of their global
developed market peers (see Chart 87).
Chart 87: Most US bank dividend yields currently lag European and Asian GSIB peers
4.0%
Non U.S. Median,
3.9%
3.0%
Dividend Yield
1.0%
0.0%
ZION
WFC
CMA
CFG
MTB
KEY
BBT
GS
BK
HBAN
C
PNC
FITB
USB
JPM
STT
STI
MS
RF
4Q17: 3.31%
3.70% Historical: 3.50%
3.50%
3.30%
3.10%
2.90%
2.70%
2.50%
1Q84
2Q85
3Q86
4Q87
1Q89
2Q90
3Q91
4Q92
1Q94
2Q95
3Q96
4Q97
1Q99
2Q00
3Q01
4Q02
1Q04
2Q05
3Q06
4Q07
1Q09
2Q10
3Q11
4Q12
1Q14
2Q15
3Q16
4Q17
Source: BofA Merrill Lynch Global Research, FDIC
As we showed in the previous section discussing the Fed’s balance sheet reduction,
bank HQLA (high quality liquid assets) as a proportion of earning assets has grown
significantly over the last five years. Since HQLA typically yields less than a loan, this
change in composition is dilutive to asset yields.
Another aspect to the liquidity rules is that these rules inadvertently render certain
liabilities more “valuable” than others, in our view. As a direct response to the liquidity
issues that arose during the Global Financial Crisis, the LCR was essentially designed for
banks to have enough liquidity to withstand a 30-day stress on its liabilities. As such,
certain deposits were classified as more prone to leave the bank under a stress scenario
than others (see Exhibit 29). In general, the outflow assumption in LCR equals the
amount of HQLA a bank has to hold again every $ of that deposit category. The more
liquidity a bank has to hold against a deposit, it likely means a narrower margin earned
on this deposit relationship.
Exhibit 29: Most valuable deposits according to LCR regulation
"Run-off"
Deposit Category
assumption
Most 3 - 5%
valuable
"Stable" retail deposits (checking accounts)
Less stable retail deposits (excess cash in wealth management accounts) 10%
Least
valuable
Financial institution deposits (hedge funds, asset managers) 100%
Excess deposits typically drain out of the system when short rates rise (see Chart 89).
Recently, we have seen US deposit growth slow. Given where absolute level of rates are,
we expect competition for excess cash at consumer accounts to pick up – potentially
leading to an inflection point in deposit betas (see Chart 90). Given the higher value
liquidity rules place on consumer deposits, we expect LCR-impacted US banks (above
$50bn in assets) to be particularly competitive – and note that it is these larger banks
that typically set the tone for deposit pricing in any given market.
(Inverted)
30%
-1%
5% 0%
20%
1%
0% 2% Largest
3% 10% incremental
-5% 4%
0%
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
100
125
150
175
200
225
250
275
300
325
350
375
400
425
450
475
500
525
25
50
75
Core Deposits YoY (lhs) Fed Funds YoY (rhs) Fed Funds Rate (bp)
Source: BofA Merrill Lynch Global Research, FDIC, Bloomberg Source: BofA Merrill Lynch Global Research, SNL Financial
Liquidity rules could also impact demand, liquidity for certain securities
Given that LCR and NSFR (note that NSFR is not final in the US) encourages banks to
hold lower-risk liquid securities, the market may observe more tempered bank demand
for agency debt, agency MBS, corporate bonds, and municipal bonds relative to U.S.
Treasuries and Ginnie Mae securities.
For example, as we explain in greater depth beginning on page 139, LCR rules caps
Level 2-designated securities at 40% of total HQLA. Agency debt and MBS would be
considered Level 2. As such, while most banks have excess LCR today, we still believe
that banks may prefer to reallocate maturing cash flow to Level 1 securities, such as
Ginnies and Treasuries. We note that banks tend to be most active in the 2-5 year
portion of the yield curve.
2007 2016
Credit Union
Credit Union 5%
3% Nonbanks
23%
Banks
40%
Nonbanks
55%
Banks
74%
What we also examined in this analysis is the balance sheet size of each originator,
which is represented by the size of the bubble in the chart. Of course, many nonbank
originators have an originate-and-sell (mostly to government agencies) business models,
which does not require a large balance sheet. However, if credit does turn, we point out
that during the most recent financial crisis, the US government agencies did “put back”
loans back to the banks that may have violated securitization agreements. This caused
banks to build what is called repurchase reserves, which of course require capital.
That said, mortgage credit availability is still well below pre-crisis levels (see Chart 94).
This is likely because while underwriting standards have tightened dramatically across
the industry, the private-label mortgage-backed securities market is still in rebuilding
mode (see Chart 95). This contrasts with agency MBS originations, which has grown
steadily since prior to and through the Global Financial Crisis (see Chart 96).
Chart 94: Mortgage credit availability well below pre-crisis levels
1,000
Mortgage credit availability index (NSA,
900
800
700
3/2012 = 100)
600
500
400
300
April '18, 177.9
200
100
0
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Source: MBA
Note: The MCAI is calculated using several factors related to borrower eligibility (credit score, loan type, loan-to-value ratio, etc.). These
metrics and underwriting criteria for over 95 lenders/investors are combined by MBA using data made available via the AllRegs® Market
Clarity® product and a proprietary formula derived by MBA to calculate the MCAI, a summary measure which indicates the availability of
Mortgage credit at a point in time.
$2.4
$7.0 $6.5
$2.5 $6.2
$0.0 $0.0
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Source: SIFMA Source: SIFMA
We think this is likely partly due to risk retention requirements, which may be delaying
the revitalization of the private label mortgage securitization market. Securitizers are
required to retain at least 5% credit risk, unless all of the mortgages that collateralize
the securitization meet the definition of a QRM, or qualified residential mortgage. (Note
under this rule, the QRM definition is closely aligned with the QM, or qualified
mortgage, definition under Dodd-Frank).
Also, note that retaining whole residential mortgage loans could be less
constraining to balance sheet capacity than the risk retained post-
securitization. Under the standardized approach, a residential mortgage
loan has a risk weight of 50%. Meanwhile, depending on credit rating and
subordination, risk weights for retained securitizations vary from 20-
1,250%.
Note that dealer fixed income inventories (including: commercial paper, U.S. Treasury
securities, agency & GSE backed securities, munis, and corporate & foreign bonds)
significantly declined (by 37%) by 2010 from its 2008 peak (see Chart 97). However,
inventories have dropped another 28% since December 2013, when the Volcker rule
came into effect.
Jan-05
Jan-06
Jan-07
Jan-08
Jan-09
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Jul-04
Jul-05
Jul-06
Jul-07
Jul-08
Jul-09
Jul-10
Jul-11
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
Jul-17
Dealer Fixed I ncome Invent ories (lhs, $bn) Ratio of Real Money FI Holdings to Dealer FI Holdings (rhs)
$176
180
$161
160
$139
140 $133 $131
$126 $122
$121
120 $112
100
80
2009 2010 2011 2012 2013 2014 2015 2016 2017
The results of these rules on balance sheet exposure are significant, as we showed
previously. Trading portfolios are down 46% from 2011 peaks. Meanwhile, as we
showed previously, the primary dealer repo market has been cut in half since 2008,
while primary dealer debt trading volumes have slid 35% over the same period.
7,000
50,000
6,000 45,000
40,000
5,000
35,000
4,000 30,000 -98%
25,000
3,000
20,000
2,000 15,000
10,000
1,000
5,000
0 0
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Source: BofA Merrill Lynch Global Research, SIFMA Source: BofA Merrill Lynch Global Research, Federal Reserve
As a result, industry credit trading revenue is down 70% since the 2010 peak (see Chart
101).
Chart 101: Credit trading revenues down 70% since 2010 peak
$25,000
Credit trading rev enues
$20,000 dow n 70% since 2010 peak
Qtrly credit trading revenue
$15,000
(annualized) ($mn)
$10,000
$5,000
$0
($5,000)
($10,000)
4Q10
1Q11
2Q11
3Q11
4Q11
1Q12
2Q12
3Q12
4Q12
1Q13
2Q13
3Q13
4Q13
1Q14
2Q14
3Q14
4Q14
1Q15
2Q15
3Q15
4Q15
1Q16
2Q16
3Q16
4Q16
1Q17
2Q17
3Q17
4Q17
A recent study attempted to focus on the impact of Volcker. In December 2016, the Fed
released a staff paper on the implementation of the Volcker Rule, with the aim of
isolating it from other factors (such as Basel III implementation) and its impact on bond
market liquidity. While non-Volcker impacted dealers tended to “step up” and provide
greater liquidity during times of stress, the study concluded that bonds are generally
less liquid when liquidity is most crucial. After all, Volcker-impacted dealers are still the
primary liquidity providers in the market.
The Fed study on corporate bond liquidity focused on times of stress, which they
defined as when a corporate bond rating was downgraded. We thought one of the most
interesting parts of the study compared share of corporate bond volume during times of
stress between Volcker-impacted dealers and non-Volcker-impacted dealers, and
compared this share pre-rule and post rule.
90%
80% 75%
70%
60%
50%
40%
30% 25%
20%
7%
10%
0%
Pre-Volcker Post-Volcker
Notably, when isolated from other regulatory variables, Volcker has also helped push
impacted dealers into (lower-margin) agency businesses, which does not require
inventory-taking, as a dealer would typically have an identified counterparty to
immediately offset a trade with a customer. Pre-Volcker, agency trading comprised 12%
of Volcker-impacted dealer’s trades, on average (see Exhibit 30). This has almost
doubled, to 23% as of 1Q16, in a post-Volcker era. Meanwhile, agency trading used to
comprise 50% of a non-Volcker-impacted firm’s trading volume (see Exhibit 31). This
has been cut almost in half, to 29%.
Exhibit 30: Volcker-affected dealers (as of 1Q16) Exhibit 31: Non-Volcker-affected dealers (as of 1Q16)
Pre-Volcker Post-Volcker Pre-Volcker Post-Volcker
Agency,
12% Agency,
23% Agency,
29%
Non-
agency, Agency,
50% 50%
Non- Non-
Non- agency, agency,
agency, 77% 71%
88%
Further research supports the assertion that corporate bond market liquidity has
declined. The Center for Financial Stability estimates that corporate bond market
liquidity has declined 46% since its peak in 2008. Meanwhile, a study by PwC found that
a reduction in market liquidity had been accompanied by a 40% increase in bond market
volatility. Taken together, this could mean larger price swings and higher costs as assets
become harder to buy/sell. That said, we think that shift in the bond market have been
driven by the leverage rule on European banks as well as from Volcker.
The shift of products to be traded electronic typically comes with more transparency, tighter
spreads, and lower revenue, though volumes can increase to provide some offset. The fixed
income markets were traditionally known as a marketplace where dealers would match
buyers and sellers, typically over the phone. Clients would reach out to several dealers, but
lacked transparent information on the bids and offers in a certain security. Dealers earned
attractive ROEs as the spread (bid-ask) earned on a product would be high, albeit with lower
volume. This eventually led to electronic communication networks (ECNs) creating central
limit order books where clients could see outstanding bids and offers by market participants.
As markets turn transparent, more competition requires dealers to price their products more
competitively to win business, which in turns leads to lower pricing (tighter bid-ask spreads).
The fixed income market is different than the equity market, and is generally less liquid
with many more CUSIPs. That said, equity market trends over the years can be used as
potential blueprints for the future of fixed income markets. On this point, equity market
pricing (bid-ask spreads) has dropped roughly 70% from 2001, but for the fixed income
market, it is still early days as many products are still less than 50% electronic (see
Chart 103).
As we look forward, further electronification of products will likely create some revenue
headwind for the industry, partially offset by market growth and/or a pickup in volatility.
Importantly, while electronification is a headwind for the industry, firms with scale and a
strong technology footprint can still gain share and produce attractive margins and
returns in this increasingly electronic landscape.
Chart 103: State of electronification in various asset classes
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
2012 2015
Source: Bank of International Settlements and BofA Merrill Lynch Global Research
45% 41%
40% 37%
35%
30%
24%
25%
20% 18%
15% 15%
15% 12%
8% 9%
10% 6%
5%
5% 2%
0%
US Equity US Bond Non US Non US Total US Total Non
Equity Bond US
2017 2004
As assets move from active to passive, the level of turnover tends to fall, weighing on
dealer trading activity. This trend has been more pronounced in equities, while fixed
income managers are still outperforming passive funds, which is driving still healthy
active inflows. However, if we use the equities universe as an example, when managers
start making outsized returns, this drives more entrants into the field, leading to
overcapacity and eventually weaker overall performance. This can then drive more of a
shift to passive, which can weigh on turnover and trading revenues. That said, we do
think there will always be a sizeable active market given market inefficiencies.
10
8
(Bn shares)
4
61%
2 52% 56% 55%
51% 49% 49% 49% 49%
35%
21% 26%
-
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
HFT ADV Other ADV
Source: BofA Merrill Lynch Global Research, TABB Group
The fixed income market is much larger than the equity market. Some areas have a
much broader impact to global markets and economies, such as FX and Rates. While the
fixed income market is very different than the equity market in size, fragmentation, and
liquidity, we believe the regulatory transformation process could have some similarities.
Thus far, as mentioned, the OTC market is shifting to a cleared environment. Moreover,
some products are already electronic, and we expect more products to become
electronic over time. These market structure changes can provide benefits to users in
terms of costs and capital efficiencies.
That said, these shifts can also present new risks. Among these “new risks” are:
reduced liquidity in times of stress, less regulated entities, technology issues,
fragmentation, and concentration risks at clearinghouses. This is not an exhaustive
list, of new risks, in our view.
We think one of the major issues in the fixed income market is that regulatory change
has occurred much faster than market structure has been able to adapt. In order to
minimize market disruption, regulatory and market structure changes may need to occur
more in tandem. It may help for new liquidity providers to be regulated to ensure
liquidity in times of stress – although this could mean increased costs.
Importantly, we believe regulators are aware of the new risks from regulation. The U.S.
Treasury department has recently become more engaged in these efforts and is seeking
input from a wide variety of market participants on changes in Treasury market
structure, the implications for market functioning, and risk management policies and
practices. Mandatory reporting of transaction in U.S. Treasury securities was recently
agreed to and FINRA firms are required to report trades to the Trade Reporting and
Compliance Engine (TRACE).
The combination of reduced depth of the fixed income markets and the significant
growth in outstanding debt over the past few years can present heightened risks to
investors (see Chart 106). The rate backdrop – recently normalizing but still low – has
spurred a “reach for yield,” which has benefited bond fund AUMs and encouraged record
issuance of corporate debt.
The recent dip in ownership was driven by money market regulatory reform, which
shifted assets out of prime funds and into treasury funds. These users and products
generally have daily redemptions versus institutional assets and products invested in
are generally stickier and longer term in nature. Hence, we think this could present
increased risks to both investors and the market
Chart 106: Significant growth in debt outstanding Chart 107: Drop in holdings due to decline in money market ownership
Balance sheets of both financial and non-financial firms Balance sheets of both financial and non-financial firms
45,000 20%
40,000 18%
35,000 16%
30,000 14%
Total Debt ($bn)
12%
25,000
10%
20,000
8%
15,000
6%
10,000
4%
5,000 2%
0 0%
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
Source: BofA Merrill Lynch Global Research, Federal Reserve (Z.1 report) Source: BofA Merrill Lynch Global Research, Federal Reserve (Z.1 report)
Higher funding and capital costs will mean that dealers will need to see larger market
dislocations before being able to justify stepping in to take advantage of mispricings
during times of stress. Further, even this may not allow some dealers to step in,
depending on firm-level regulatory ratios/constraints at the time of dislocation.
Given lower inventory, dealers will be less exposed, and in fact may be able to make
more money on wider spreads. Unfortunately, in our view, investors may see bigger
losses in times of stress due to dealers’ reduced ability to provide liquidity. That said,
this could shift to some degree depending on the outcome of Volcker reform.
The negative regulatory view on short-term funding could put pressure on primary
dealer clients, particularly those using leverage. We expect client financing costs and
thereby investment returns to come under some pressure (potentially impacting hedge
funds, structured products, levered vehicles, and certain mutual fund and pension fund
strategies). Over time, lower returns can impact investor’s allocation decisions and the
growth outlook for these products, including for hedge funds (see Chart 108). We also
expect some areas that relied on short term financing to potentially shrink under the
new rules.
$3,500
Assets Under Management ($bn)
$3,000
$2,500
$2,000
$1,500
$1,000
$500
$0
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Source: Barclay Hedge
Note: US hedge fund data
We expect banks to do what they can to continue compress their legacy derivative
exposures (collapsing/netting old trades) – particularly if an institution is still falling
short of SLR requirements at either the hold co or bank sub. Going forward (particularly
for CDS), we expect risk/exposure management and systems to place more emphasis on
the notional values of derivatives contracts, which were previously only tangentially
relevant. Additionally, we could see innovation, with exchange traded products launched
and/or products designed with similar attributes but lower maximum losses (notional
exposures).
Under current SLR rules, it is worth noting that the gross up of collateral (both received
and posted) is somewhat controversial, given that it has the effect of dis-incentivizing
the holding of collateral, despite the obvious benefits to the soundness of the business
of holding collateral. However, as we discussed in a previous section, a recent Basel
proposal could lessen the pressure on this issue. We expect this to gain traction over the
next year, which should create some benefit.
Another area of recent focus has been the potential for centrally cleared Treasury repos.
Centrally cleared repo would bring together a broader range of market participants and
facilitate transactions through a single clearinghouse. This would allow for more
transactions between cash investors, such as money funds, and cash borrowers, like
dealers or hedge funds. It would also reduce single name counterparty risk of cash
borrowers. Other benefits of central clearing include reducing the implications of a
single counterparty default, allowing for increased netting capacity, and enhancing
overall repo market liquidity and availability.
That said, there still lies uncertainty with centrally cleared repo
However, numerous questions are outstanding regarding centrally cleared repos such as:
• How to reduce concentration risk in single counterparty that could be vulnerable
should a single or multiple large members default?
A recent study conducted by the Office of Financial Research found that expanding U.S.
Treasury repo central clearing to non-dealer counterparties would reduce risk exposures
81% for dealers. However, this would also increase risk exposure for a central
counterparty by as much as 75%.
The benefits of central Treasury clearing would then depend on the cost of non-cleared
repo activity vs. the additional funds required to guarantee centrally cleared trades. Any
move to centrally cleared repo would reduce the cost of dealer intermediation in the
repo market and compress the spread between GCF (general collateral financing) and
tri-party repo rates (see Chart 109).
Chart 109: Spread between GCF and tri-party repo rates
2.5 70
60
2
50
1.5 40
1 30
20
0.5
10
0 0
2014 2015 2016 2017 2018
Spread (5D MA, RS, bps) BNY Tri-Party (LS, %) GCF (LS, %)
Overall, due primarily to SLR, we expect some business to move to exchange traded
products (though this will depend on the users need), some to move to unregulated
firms (though clients generally want to clear with strong counterparties), wider spreads
across derivative contracts, and significant efforts to net/compress exposures. The
impact to clients would likely include less capacity, higher pricing for hedging, more
volatility, and lower investment returns over time.
To demonstrate this shift change, we looked at five measures of capital and liquidity as
of year-end 2017, and compared these to 2007. (Keep in mind that today’s regulatory
ratios are not comparable generally, as these rules did not exist pre-crisis.)
We looked at the following for the largest U.S. banks under coverage, including now
non-existent companies in the 2007 statistics: 1) tangible common equity/assets
(TCE/TA); 2) short-term wholesale funding/assets; 3) cash/earning assets; 4) trading
assets/earning assets; and 5) loan/deposit ratios. We found that in each of these ratios,
banks are much better positioned today versus in 2007 (see Chart 110).
Low er ratio
is better
2017 2007
For example, banks have increased their TCE/TA by 460bp to 8.0% as of 4Q17 versus
3.1% in 2007. Moreover, they have more liquidity, with cash/earning assets of 30% in
2017 versus 21% in 2007. Funding profiles have also improved with lower short term
wholesale funding/assets of 17% vs. 29% and loan/deposit ratios at 77% as of 4Q17 vs.
106% in 2007. Banks are also carrying less inventory, as trading assets have declined to
19% of earning assets in 2017 vs. 26% in 2007
In this section, we explore current agency vacancies, and provide some insight to the
regulatory agencies that are most responsible for carrying out Dodd-Frank.
What we think could be most crucial for bank investors, is how the Vice Chair of
Supervision Randy Quarles will interpret the following: the administration of the Dodd-
Frank Act stress test with regards to the CCAR overlay, the Volcker Rule, and the “gold-
plating” of Basel prudential regulatory standards for U.S. banks.
The CFPB operates as an independent agency funded by the Federal Reserve. Mr.
Mulvaney was named acting director of the Bureau in November 2017, following the
resignation of former director Richard Cordray. President Trump is expected to name J.
Mark McWatters, the current chairman of the National Credit Union Administration, as
his CFPB nominee in June. If named the official nominee, McWatters would have to be
confirmed by the Senate Banking Committee, followed by a vote in the Senate.
The FDIC is managed by a five-person Board of Directors who are appointed by the
President and confirmed by the Senate (no more than three can be from the same
political party). The current Chairman is Mr. Martin J Gruenberg and the Vice Chairman is
Mr. Thomas M. Hoenig. Mr. Gruenberg’s term ended in November 2017, but he is
expected to continue serving until a successor is in place. President Trump has
nominated Jelena Williams, most recently an executive vice president and chief legal
officer of Fifth Third Bank, to replace Mr. Gruenberg as Chairman of the FDIC. Mr,
Hoenig announced his retirement from his post as Vice Chairman of the FDIC in April
2018.
The FHFA is a member agency of the Financial Stability Oversight Council (FSOC). The
director of the FHFA, Mr. Melvin L. Watt, was confirmed by the Senate in December
2013 and is serving a five-year term, ending in January 2019. The current Trump
administration has placed focus on GSE reform for 2019.
The Board of Governors, Reserve Banks, and the Federal Open Market Committee
(FOMC) are the key entities of the FRB. The seven members of the Board of Governors
are nominated by the President and confirmed by the Senate. The seven member board
has two vacancies. The Chairman and the Vice Chairman of the Board are named by the
President from among the members and are confirmed by the Senate. On February 5,
2018, Mr. Jerome Powell was sworn in as the Chairman of Federal Reserve Board,
replacing Ms. Yellen who served as Chairman for four years. Mr. Quarles replaced former
(de facto) Vice Chair of Supervision Daniel Tarullo).
The Council supervises and regulates bank holding companies over $50bn in assets and
systemically risky non-bank financial companies. The Council consists of 10 voting
members and 5 nonvoting members and brings together the expertise of federal
financial regulators, state regulators, and an independent insurance expert appointed by
the president. The Chair of the Council is Steven Mnuchin.
In November 2017, Joseph Otting replaced Thomas Curry as Comptroller of the Currency
after Mr. Curry’s more than five year term at the helm of the Office of the Comptroller
of the Currency. Recently, Mr. Otting has voiced his support for industry deregulation as
a means to encourage bank lending
The OFR is currently run by Director Ken Phelan, who replaced previous OFR Director
Richard Berner in January 2018.
CET1 requirements globally have mostly been set. Two potential changes are on the
horizon, however. First in April 2018, the Federal Reserve published a draft proposal that
would amend capital rules for bank holding companies (BHCs) with $50bn+ in assets,
one that would reset capital minimums annually based on each BHC’s current year stress
test (or CCAR) results. Specifically, the proposal would replace the 2.5% conservation
buffer with a stressed capital buffer (“SCB”) equal to the amount of capital “burned
down” in the binding scenario of the test. This would result in the same or higher capital
requirements for impacted U.S. banks.
Second, in December 2017 the Basel Committee published the final regulatory
standards in its post-crisis Basel III reforms. Although final on an international basis,
national regulators may consider applying these updates. Separately, in March 2018 they
proposed changes to the market risk methodology. (See section above for additional
detail).
Based on today’s standards, the lion’s share of GSIBs comply with transitional phase-in
requirements, and given current levels, are all on track to exceed or meet 2019 fully
phased-in minimums (see Chart 111, Chart 112, and Chart 113). As we will explain in
this section, there is no uniform ratio standard for all 30 GSIBs, as each bank’s standard
depends on its surcharge.
Chart 111: North American GSIBs Chart 112: European GSIBs Chart 113: Asian GSIBs
Nordea
MS CCB
RBS
C HSBC
SMFG
ING
WFC StanChart
ICBC
DBK
JPM
UBS
MUFG
Unicredit
GS
CS
Mizuho
RBC BARC
CA
BOC
STT BNP
SocGen
BK ABC
Santander
0.0% 5.0% 10.0% 15.0% 20.0% 0.0% 5.0% 10.0% 15.0% 20.0% 0.0% 5.0% 10.0% 15.0%
The simplest way to think about the minimum for each GSIB is the minimum
requirement of 4.5% + capital conservation buffer of 2.5% + each bank’s surcharge. In
essence, the GSIB CET1 minimum is 7.0% + the bank’s surcharge. That said, this
minimum could increase for U.S. BHCs if the Fed replaces the capital conservation
buffer (currently 2.5%) with its proposed stressed capital buffer (equivalent to CET1
“burned down” in U.S. CCAR stress test: minimum of 2.5%-no maximum) and/or includes
a countercyclical buffer (0-2.5%). In a later section, we will discuss how exactly each
individual institution’s surcharge is calculated (see Chart 114).
Chart 114: Common equity tier 1 buffer composition
20.0%
Fed has proposed a stressed
18.0%
CET1 to risk-weighted assets
Investors may have heard the terms “transitional” or “fully phased-in” to describe CET1
standards. While we have seen the market impact banks to comply sooner rather than later
by assigning discounts to shares of banks that have lagged behind, GSIBs technically have
until 2019 to comply with the full 7.0% + surcharge standard (see Chart 115). Starting in
2014, GSIBs have been under a transitional period (see chart). In 2015, banks technically
only had to comply with the 4.5% minimum capital requirement. Beginning in 2016,
standards began to include part of the “fully-phased in” or full 2019 requirement. Each
year adds another 25% of the buffer and surcharge into the compliance requirement, until
the entirety of both are fully phased-in in 2019. In this example, JPM in 2018 will have to
comply with 4.5%, plus 75% of the 2.5% capital conservation buffer (1.88%) and 75% of
its current 3.5% surcharge (2.63%).
8.0% 7.5%
6.0%
6.0%
4.5%
4.0%
4.0%
2.0%
0.0%
1/1/14 1/1/15 1/1/16 1/1/17 1/1/18 1/1/19
Source: BofA Merrill Lynch Global Research, company data, Basel Committee
Global regulators have the option to add yet another buffer during frothy times
Note that global regulators have the optionality of adding a “countercyclical capital
buffer” should they determine frothiness in credit growth. (Note: This would be in
addition to the proposed stressed capital buffer in the U.S. and capital conservation
buffer in both Basel and U.S. approaches). The countercyclical capital buffer is a
potential expansion of the capital conservation buffer that takes into account the macro
financial environment in which large, internationally active banks function.
As of September 8, 2016 the Federal Reserve published the framework that will apply to
setting the countercyclical capital buffer. On December 1, 2017 the Federal Reserve
reaffirmed setting the U.S. countercyclical capital buffer at 0% and stated that it will
review the amount at least annually. The countercyclical capital buffer can be increased
when policy makers see an elevated risk of above-normal future losses. The
countercyclical buffer can be set at up to an additional 2.5% of risk-weighted assets
(RWA) subject to a 12-month implementation period. Although the implementation can
be shorter at the Fed’s discretion, the rule text states, “However, economic conditions
may warrant an earlier or later effective date.”
In other words, if U.S. regulators determine excess growth in the system, the CET1
standard could comprise of the 4.5% minimum + 2.5% capital conservation buffer +
GSIB surcharge + up to 2.5% incremental countercyclical buffer. So, if Bank A’s
surcharge is 3.5%, the resulting minimum CET1 in this case would be 13%, before
the potential replacement of the capital conservation buffer by the proposed
stressed capital buffer (incremental to 2.5% with no maximum)
Meanwhile, Europe already has a countercyclical buffer in effect for certain countries.
According to the European Systemic Risk Board (ESRB), the countercyclical buffer for
the Sweden and Norway is already set at 2.00%, with 0.5% for Slovakia and the Czech
Republic.
+ +
Qualifying minority interests
Market Risk weighted assets
-
Adj. AOCI +
- Operational Risk weighted assets
DTA from carryforwards (Advanced approaches only)
-
Adj. Intangible Assets
= =
Common Equity Tier 1 Capital ÷ Total Risk weighted assets
=
Common Equity Tier 1 ratio
Source: BofA Merrill Lynch Global Research, Basel Committee.
Note: For the numerator a key adjustment should add an adjustment for DVA for liabilities and derivatives (aka own credit). DTA = Deferred
tax assets
The difference between the two can be most felt in the calculation of risk-weighted
assets, or the denominator. Very simply, standardized risk weights are set by regulators
and are the same across all banks. Meanwhile, under advanced approaches, regulators
require banks to use internal models (taking into account credit experience and internal
ratings) to calculate risk weights.
For example, under the standardized approach, Bank A and Bank B will both incur a
100% risk weight on corporate exposure (see Exhibit 34). However, under the advanced
approaches, each bank’s models will determine a risk weight that is very specific to each
individual institution’s portfolio – which could be lower or higher than the standardized
risk weight. Additionally, the advanced approaches take into account operational risk,
which we’ll discuss in more detail in the RWA section of this report.
Standard Advanced
RWA based on
Secured financing collateral haircut
approach
US PSEs (including municipal
20 or 50%
securities) Determined by the
Residential mortgage loans* 50% likelihood of default
Home equity loans 100% and recovery given
HVCRE loans 150% default, based on firm
CRE loans (term + construction) 100% experience (could be
Auto loans (consumer) 100% below or above
C&I loans 100% 100%)
RWA based on
potential future
FI Trdg: Currencies & gold
exposure and
counterparty
Source: BofA Merrill Lynch Global Research, Federal Reserve
Asterisk denotes “prudentially written”
The advanced approaches are applicable to institutions with over $250bn in assets or
balance sheet foreign exposure of greater than $10bn. For U.S. based banks, the
approach that results in the lower capital ratio will be used as the binding constraint as
ruled by the Collins Amendment of the Dodd Frank Act. While the binding constraint
varies across the banks, generally the standardized approaches have become the more
restrictive of the two approaches (see Chart 116).
The Collins Amendment in the U.S. requires banks to use the lower of the
standardized or advanced ratios as the binding constraint.
As of 1Q18, there was an average difference of 80bp between a U.S. GSIB’s CET1 ratio
under advanced vs. under standardized. That said, many banks are now bound by the
standardized approach, driven mostly by balance sheet optimization and model
improvements.
Chart 116: CET1 Ratio, Standardized vs. Advanced (as of 1Q18)
16.0% 15.5%
Source: BofA Merrill Lynch Global Research, company data, SNL Financial
AOCI: A positive today, but with negative impacts as rates rise (U.S. banks)
While mostly relevant for U.S. based banks, accumulated other comprehensive income
(AOCI) is also included in the numerator. AOCI includes unrealized gains and losses on the
available-for-sale (AFS) securities portfolio. This inclusion applies to banks that are
considered “advanced approaches” by regulators. Institutions that are not under the
“advanced approaches” (those under $250bn in assets or $10bn in foreign exposure) can
make a one-time, permanent election to “opt out” of including AOCI in the calculation.
AOCI served as a boost to capital ratios as rates remained low, leading to an increase in
the value of banks’ bond portfolios. However, as long interest rates have risen, we have
seen the negative impact on capital as bond portfolios are marked lower through AOCI,
eroding unrealized gains as well as creating unrealized losses. These gains and losses
are added or subtracted from the numerator. We plotted Citi’s AOCI on a quarterly basis
from 2Q13 through 1Q18 against the average 10-year U.S. Treasury yield to show
readers the relationship between the two (see Chart 117).
Chart 117: Example: AOCI has benefited C’s regulatory capital in a declining rate backdrop
$4,000 3.0%
$2,000
2.5%
$0
($2,000) 2.0%
($4,000)
1.5%
($6,000)
($8,000) 1.0%
2Q13
3Q13
4Q13
1Q14
2Q14
3Q14
4Q14
1Q15
2Q15
3Q15
4Q15
1Q16
2Q16
3Q16
4Q16
1Q17
2Q17
3Q17
4Q17
1Q18
YoY change in AOCI ($mn, lhs) Avg. 10yr UST Yield (rhs)
• Deferred tax assets (DTAs), but only those arising from temporary differences.
DTAs that can be included in CET1 are those created, for example, from the timing
difference between reporting a provision for credit losses in an income statement and
actually charging off the associated loan. Of course, a handful of banks saw their DTAs
increase from this timing difference during the financial crisis. Allowable DTAs are those
that cannot be realized through net operating loss carrybacks. Additionally, these
allowed DTAs are net of any related valuation allowances (also incurred by a handful of
institutions, which wipes the DTA out of one’s equity base) and DTLs.
Exhibit 35: Limitations included in CET1
+
Minority Interest 10%
+
Mortgage Servicing Rights 10%
=
Aggregate 15%
Source: BofA Merrill Lynch Global Research, BCBS
• Deferred tax assets created from operating losses and tax credit carry
forwards, net of associated DTLs.
Exhibit 36: CET1 Deductions
Deductions
Goodwill
20.0
(ex . tax credits,
DTA balance ($bn)
10.0
(ex . provision / Capped to 10% of
5.0 charge off timing) $11.0
CET1
0.0
DTA Balance
80%
60%
40%
20%
0%
ICBC
WFC
BOC
CCB
ABC
ING
SocGen
GS
BK
Nordea
MUFG
BARC
HSBC
C
StanChart
RBC
BNP
DBK
RBS
UBS
JPM
Unicredit
STT
MS
CA
CS
Mizuho
SMFG
Santander
-20%
Source: BofA Merrill Lynch Global Research, company data, SNL Financial.
Credit RWA is also assigned for non-loan assets, though this is calculated differently.
Examples include:
• “Vanilla” securities (U.S. Treasuries, U.S. government sponsored
entities/agency RMBS, publicly-sponsored entities, munis). RWAs assignations
are similar to loans. U.S. Treasuries have a 0% risk weight, U.S. GSEs/agency RMBS
have a 20% RWA, and U.S. PSEs/munis have a 20-50% RWA.
• Secured financing transactions. The RWA for SFTs are typically calculated based
on the underlying collateral, with a haircut. Counterparty type and tenor of the
transaction also factor into the risk weight.
• Asset- and mortgage-backed securities. RWAs are based not just on the quality
of the underlying assets, but on the structure of the security.
• Credit and other trading inventory. RWAs are calculated by multiplying the total
exposure with the PFE (or potential future exposure), which ranges between 0-20%.
PFE is determined by tenor, with the shortest tenor having the lowest multiplier
and the longest tenor having the highest multiplier. Every product has its own PFE,
scaled toward the perceived inherent risk of each. For example, currencies and gold
have a PFE range between 1-8%, while equity swaps have a PFE range of 6-10%.
Important to note that the PFE has a multiplier not a haircut effect. As such, the higher
the PFE, the riskier the asset. After calculating the PFE, the counterparty type is then
also taken into account to finalize the risk weight.
Note that European banks have historically retained residential mortgage exposure on-
balance sheet, whereas U.S. banks have sold qualifying conforming mortgages to U.S.
government-sponsored entities Fannie Mae and Freddie Mac. Holding low RWA loans
like retail mortgages can be advantageous to the CET1 calculation.
• Incremental risk, which measures default and migration for non-securitized credit
products
• Comprehensive risk, which captures all price risk for correlation trading portfolios
• Specific risk add-ons, which takes into account the value and risk weighting of long
or short positions in debt, equity, and securitization products
Due to the qualitative nature of such activities, there had traditionally been few analyses
done in identifying and measuring such risk. However, following the financial crisis,
regulators have determined that credit and market risk were not entirely sufficient in
capturing the risks faced by financial institutions.
There have been three general sets of approaches in calculating operational risk: 1)
basic indicator approach based on total revenues, 2) standardized approach based on
business segment revenues, and 3) advanced measurement approaches based on a more
detailed internal risk framework of each bank. Banks are allowed to apply their own
definitions of operational risk after following a set of guidelines issued by the Basel
committee.
In the Dec. 2017 proposal, the BCBS also set forth a new standardized measurement
approach (SMA) for operational risk. First, the SMA allows national regulators to decide
whether to require institutions to include historical operational-risk losses in the
operational-risk capital calculations. The new SMA would also recognize three rather
than five business-size categories for measurement.
GSIB surcharges
The CET1 ratio is relatively consistent with historical measures of risk-adjusted capital
ratios, with of course refined and tightened standards. Since the global financial crisis,
regulators have been concerned about the impact that failure of one or more systemically
important financial institutions could have on the broader financial system. In an effort to
address the spillover risks of GSIB failure, the BCBS adopted an additional CET1 capital
surcharge for such banks. As mentioned, the 30 largest global banks are required to add a
CET1 surcharge in addition to the 4.5% minimum CET1 requirement and the 2.5% CET1
Capital Conservation Buffer.
A U.S. bank holding company that is designated a GSIB under the established
methodology will be required, on an annual basis, to calculate a surcharge using two
methods (“method 1” and “method 2”) and will be subject to the higher of the two
surcharges.
The Method 1 framework, created by the BCBS, compares a bank’s activities grouped
into five categories of systemic importance, with the aggregate global figures of all
banks in a specified sample. Method 2 only applies to the eight U.S. GSIBs. This method
replaces “substitutability” (one of the five Method 1 categories) with “reliance on short-
term wholesale funding” and employs a fixed calculation approach.
Should a GSIB becomes subject to a higher surcharge, the higher surcharge would not
become effective for a full year (e.g., a higher surcharge calculated by December 31,
2017 would not become effective until January 1, 2019). However, if a GSIB’s score
would result in a lower surcharge, it would be subject to the lower surcharge beginning
the next calendar year (e.g., a lower surcharge calculated by December 31, 2017 would
become effective January 1, 2018).
• inter-connectedness
• complexity
• cross-jurisdictional activity
• substitutability
The Method 1 calculation considers each bank’s exposure on a relative basis, with the
size of each GSIB compared to the size of the largest 75 banks globally. For example, if
industry growth is flat and a GSIB is growing above this level, its size indicator would
increase year over year. The converse is also true. We break down what comprises each
category below (see Exhibit 39).
2.0%
Deutsche Bank =
• Intra-financial system assets
HSBC =
Interconnectedness • Intra-financial system liabilities
• Securities outstanding Bank of China ▲
Barclays =
BNP Paribas ▼
• Notional amount of OTC derivatives China Construction Bank ▲
• Trading and AFS securities
1.5%
Complexity Goldman Sachs =
(excluding HQLA)
• Level 3 Assets
ICBC =
Mitsubishi UFJ FG =
Wells Fargo =
Agricultural Bank of China =
Cross- • Cross-jurisdictional claims
Bank of New York Mellon =
Jurisdictional Activity • Cross-jurisdictional liabilities
Credit Suisse ▼
Groupe Crédit Agricole =
ING Bank =
• Payments Activity Mizuho FG =
Substitutability • Assets under custody Morgan Stanley =
• Underwriting activity
Nordea =
Royal Bank of Canada 1.0% new
Source: BofA Merrill Lynch Global Research, BCBS
AFS = available for sale. HQLA = high quality liquid assets Royal Bank of Scotland =
Santander =
Société Générale =
Standard Chartered =
State Street =
Sumitomo Mitsui FG =
UBS =
Unicredit Group =
Source: BofA Merrill Lynch Global Research, FSB, BCBS
As of November 2017
The score of each category is determined by taking JPM’s exposure in that category
(e.g., complexity), converting this into euro, and dividing it by the sum exposure of the
75 largest global banks. The scores across the five categories are then averaged. Once
averaged, this final score is compared to a schedule, which brackets a range that
corresponds to each institution’s CET1 surcharge. In this example, JPM’s score of 465
corresponds to a 2.5% surcharge.
The U.S. GSIB surcharge took effect on January 1, 2016. The current surcharge is based on
the systemic indicator scores as of December 31, 2016 (includes the aggregate global
indicator amount, i.e., denominator) while the STWF score will be based upon the last
twelve month average of a bank’s risk weighted assets. For the surcharge calculated in
2017 and thereafter, GSIBs will compute their STWF score using average daily amounts.
Source: BofA Merrill Lynch Global Research, company data (FR Y-15 filings), Federal Reserve
Once each bank’s short-term wholesale funding (STWF) weighting is calculated, this
weighting is then divided by its last twelve month average risk-weighted assets then
multiplied by 350 (the fixed conversion factor).
Exhibit 43: Short-Term Wholesale Funding Weighting (STWF)
Remaining Remaining Remaining Remaining
Maturity of Maturity of Maturity of Maturity of
Component of short-term wholesale funding
30 days of 31 to 90 91 to 180 181 to 365
less days days days
Secured funding transaction secured by a level 1 liquid asset 25% 10% 0% 0%
Secured funding transaction secured by a level 2A liquid asset;
Unsecured wholesale funding where counterparty is not a financial entity;
Brokered deposits and brokered sweep deposits provided by a retail customer or counterparty; 50% 25% 10% 0%
Covered asset exchanges (level 1 liquid asset for a level 2A liquid asset);
Short positions (borrowed security: level 1 or level 2A liquid asset)
Secured funding transaction secured by a level 2B liquid asset;
75% 50% 25% 10%
Covered asset exchanges and short positions (other than those described above)
Unsecured wholesale funding where counterparty is a financial entity
100% 75% 50% 25%
Any other component of short-term wholesale funding
Source: Federal Reserve
Where a bank falls within a range can show investors how much, or how little, room
each bank has to grow its balance sheet/increase exposures before broaching the next
surcharge. Conversely, it can also show how far/close each institution is from lowering
their surcharge.
In examining where the biggest U.S. GSIBs fall at YE17, JPM and GS, are all at the
higher end of their ranges (see Exhibit 44). This suggests that these banks need to be
careful about growing exposures, and are likely closely considering the impact of a
higher GSIB surcharge. On the other hand, WFC and STT appear to be at the low end of
its range – implying potential for balance sheet growth.
Source: BofA Merrill Lynch Global Research, company data, Federal Reserve
During 2016, the ECB changed the way it looks at the pillar 2 element of RWA. This was
partly to make the additional tier 1 market more stable.
The pillar 2 element of the requirement is now split into two parts:
• Pillar 2 requirement (P2R)
Pillar 2 requirement
This element is a mandatory requirement and is used to address risks not captured in
pillar 1. This is the level that is key for banks’ maximum distributable amounts (MDA)
which informs whether banks may pay dividends to shareholders or AT1 holders. This
element is disclosed by the banks.
Pillar 2 guidance
This element is an additional buffer that banks are expected to meet. Falling below this
level (but above the pillar 2 requirement) has no effect on the banks’ ability to pay AT1
dividends and reduced legal action. This element does not have to be disclosed, but a
small number of EU banks have chosen to disclose.
While the term “guidance” may imply that this is optional in some way, the ECB has
been clear that it expects the banks to comply with the pillar 2 guidance.
“…for an individual bank with a 2016 Pillar 2 requirement (P2R) of 11%, after this
clarification, all other things being equal, it may have a P2R of 10% and a Pillar 2
guidance (P2G) of 1%. But, as only the P2Rs are published (because they are
relevant for calculating the maximum distribution amount for dividends and
coupons), it looked as if “the capital demand” (P2R +P2G) had been reduced, from
11% to 10% in my example. But this is not correct; we still expect banks to
comply with the Pillar 2 guidance as well. So it may look like it is decreasing, but
it’s not.” – SSM Chair Danièle Nouy, interview with La Repubblica 30 January
2017
A bank breaching the pillar 2 guidance level of capital could result in increased non-
public supervisory action to improve its capital levels.
2. Higher RWA as a result of other Basel initiatives or the ECB’s TRIM project (see
below)
6%
4%
7% 7% 7%
2%
0%
Erste Bank KBC Permanent TSB
Source: Company data, BofA Merrill Lynch Global Research. Note: * Minimum includes 4.5% minimum + 2.5% Capital conservation buffer. **
Pillar 2 G may or may not reduce by 2019
TRIM aims to assess whether the internal models currently used by banks comply with
regulatory requirements and whether they are reliable and comparable. The ECB has
inherited supervision of the Eurozone’s major banks from a larger number of national
regulators. New internal models require the approval of the regulator. With a new
regulator, it therefore seems logical that the ECB wants to check what it inherited and
that the models currently in use are fit for purpose.
The key differences we can see versus the Basel 4 measures are:
• Basel 4 penalizes low-risk lending by imposing a floor
Timing
TRIM is expected to last through 2019, although given the objectives of the project, it
does seem reasonable to assume that this is a one-off project.
Some studies by the EBA have in the past shown that a large proportion of the
variability in risk weights from individual banks has been explained by historical losses.
I.e., quite often the different risk weights the banks use can be explained by different
risk levels.
Some Irish banks have signaled some first round effects of the TRIM project with higher
RWA in 2016-2018.
“It is important to note that while the project aims to reduce unwarranted
variability in RWAs across banks, a general increase in RWAs is not the intention.
Nevertheless, TRIM could result in increases or decreases in capital needs for
individual banks.” – ECB
Obviously, banks have to be compliant across six broad pieces of prudential regulation,
and also, there are some products (e.g., wealth management, etc.) that are universally
more valuable across all constraints.
Table 8: More and less valuable products under an RWA constraint
CET1: More Valuable Products CET1: Less Valuable Products
Electronic/agency trading Equity/fixed income derivatives
Wealth management Securitized products
Asset Management HY credit products
Advisory business Commodities
Payments business Mortgage servicing rights
Futures clearing Non agency MBS
Rates Higher risk loans
Repo (collateral dependent)
Agency MBS
Countries with the highest minimums, of 5%, include: U.S. and Switzerland.
Meanwhile, countries with the lowest minimums include: some countries in the EU,
Japan, and Canada.
China is at a 4% minimum, while the UK requires a 3.25% minimum + 35% of the GSIB
surcharge. Many countries within the EU also require an additional leverage ratio
minimum for banks which are considered domestic-systemically important banks
(DSIBs): e.g., Netherlands at 4%.
Conversely, unlike with CET1, there are a handful of banks that are still below their
future, Basel 4 fully-loaded minimums, such as DBK and the Swiss banks (although we
note the Swiss are now above the 3.5% CET1 leverage ratio they are also held to). While
each bank has until 2022 to comply, this current shortfall nevertheless informs current
plans, as has been evidenced in recent years and months. As such, we think this
particular rule will continue to have measurable impact on market share dynamics in
trading, with institutions holding excess SLR relative to local rules best positioned to
consolidate market share.
Chart 121: North American GSIBs Chart 122: European GSIBs Chart 123: Asian GSIBs
StanChart
WFC ICBC
RBS
C HSBC
CCB
Unicredit
JPM Santander
BOC
Nordea
MS
BARC
ABC
UBS
STT
CS
MUFG
BK ING
CA
SMFG
GS SocGen
BNP
RBC Mizuho
DBK
0.0% 5.0% 10.0% 0.0% 2.0% 4.0% 6.0% 8.0% 0.0% 2.0% 4.0% 6.0% 8.0%
Numerator Denominator
On B/S assets
+
Off B/S derivatives
Common Equity Tier 1 exposures
+
Off B/S securities financing
transaction
+ +
Off B/S unfunded lending
commitments
Additional Tier 1 Capital +
Off B/S standby letters of
credit
= =
Basel III Tier 1 Capital ÷ Total Leverage Exposure
=
Leverage Ratio
Source: BofA Merrill Lynch Global Research, Financial Stability Board, Federal Reserve
Derivative Assets Mark to market value of derivatives less collateral received and Included as reported under US SLR, under
On Balance Sheet Exposures
(GAAP) enforceable counterparty netting agreements BCBS ex netting collateral
Net value of SFT, allows netting of certain collateral and master Included as reported under US SLR, included ex
On Balance Sheet Exposures SFT (GAAP)
netting agreement netting under BCBS
Collateral Financial collateral given/received to/from lender/borrower in Remove the value of securities received &
Posted/Received exchange for security recognized as an asset by transferor
Netted SFT under GAAP
Master Netting Netting agreements enforceable in the event of a default or a
Netting disallowed under BCBS
Agreement (MNA) legal judgment
Maximum of 0 total fair value of the securities and cash lent to Included under BCBS using current exposure
Counterparty Exposure Counterparty Exposure
the counterparty minus cash and securities received with no add on
Source: BofA Merrill Lynch Global Research, Financial Stability Board, Federal Reserve
Note: Only cash netting under certain conditions can occur
Source: BofA Merrill Lynch Global Research, Financial Stability Board, Federal Reserve
Other Off-B/S Uses a "standardized approach" for converting into Can apply the lower of two applicable CCFs to
Exposures RWAs rather than the uniform 100% CCF method provide a sommitment on an off balance sheet item
Switzerland mirrors the U.S. hold co minimum of 5%. The majority of the EU currently
has a 3% minimum but is likely to follow the Basel 4 reform of adding a GSIB surcharge.
The UK requires an additional buffer equivalent to 35% of the GSIB surcharge.
Since January 2015, banks have been required to disclose their leverage ratios, but are
not required to comply with the rule until January 1, 2018. In the case of the Swiss
banks, the compliance date is January 1, 2020. That said, the implications of the
framework are being felt today, as banks adjust their business models.
It is measured in both average (of end of each month for each quarter) and end of each
quarter. The minimum in the UK is set as 3.25% minimum pillar 1 plus 35% of the GSIB
RWA buffer, plus the countercyclical buffer. For BARC this is equal to 3.975% on a fully-
loaded basis (1st January 2019 in the UK). The GSIB and countercyclical buffer must be
held in CET1, plus 75% of the minimum. Again, for BARC this means a minimum
CET1/leverage ratio of 3.16%.
The introduction of a “blunt” leverage ratio in Basel 3 signaled a major change for the
large European wholesale banks. It has significantly changed the economics of many
“normal” banking products as well as the more naturally leverage-intensive products
within an investment bank.
While no longer a new concept in Europe, implementation has been slow and some
banks remain below fully-loaded minimums. With return on assets generally lower in
Europe (largely reflecting the lower risk on EU balance sheets) it has been significantly
more difficult for banks to adjust to this new reality.
The European Union has so far proposed a 3% leverage ratio minimum although many
banks and some national regulators have imposed higher minimums. Given the
requirements around pillar 2, the majority of banks within the EU are in any case
constrained on RWA rather than leverage.
The methodology differs to that of the Basel requirements slightly, but importantly
allows banks to calculate it using the period-end exposures rather than the three-month
average. In line with Basel requirements, the ratio has been disclosed since 2015.
We would also note that Europe and Basel look set to form additional leverage ratio
requirement for GSIBs once there is international agreement.
Large Swiss banks (UBS and CS) will be required to reach a 3.5% CET1/leverage ratio
and a 5% tier 1/leverage ratio
On a RWA basis, banks will still need 10% CET1, but an extra 4.3% of additional tier 1.
TLAC requirements come on top, with both banks needing 2x their tier 1 requirements,
i.e., 10% of leverage, or 28.6% on RWA.
Swiss regulators indicated that there would be penalties (i.e., increased requirements)
for banks that expanded aggressively or became more complex. Similarly, there may be
a benefit in the form of lower TLAC requirements if they become simpler or smaller.
We understand that the 5% minimum tier 1 leverage ratio is effectively a true minimum
of 3%, plus a 2% buffer. Additionally, the 3% must be CET1. On RWA, an 8% minimum
has a 6% buffer added to it to get to the 14% Tier 1 requirement. If banks dip into the
buffer, then they must inform the regulator on what steps they are taking to re-
establish it.
25%
20%
15%
10%
5%
0%
Old New Old New
Leverage RWA
Source: FINMA
As a result, even banks with excess SLR have a reduced ability to retain inventory to
facilitate client activity. As we discuss in the “The benefits and costs of regulation”
section, this has and will continue to have a lasting negative impact on secondary
market liquidity. While competitive factors have precluded GSIBs from fully passing this
on to their clients, we think investors nonetheless have and will likely continue to see
higher costs to transact.
Table 11: More and less valuable products under leverage ratio
Leverage Ratio: More Valuable Products Leverage Ratio: Less Valuable Products
Electronic/agency trading Equity / FICC derivatives
Wealth Management Prime Brokerage
Asset Management Rates
Advisory Repo
Payments IG Credit products
Futures clearing Commodities financing
High yield / distressed credit Unfunded lending commitments
Financial deposits
In November 2015, the Financial Stability Board (FSB) issued its final rules for Total
Loss Absorbing Capacity (TLAC). In December 2016, the Federal Reserve Board finalized
the TLAC framework in the U.S.
Unlike with funding regulation, institutions with significant deposit funding fare the
worst under TLAC.
Chart 125: United States GSIBs Chart 126: European GSIBs Chart 127: Switzerland GSIBs
60% 40% 9.2%
35%
50% 9.0%
30%
TLAC % of Leverage
TLAC % of RWA
20%
30%
15%
8.6%
10%
20% 8.4%
5%
10% 0% 8.2%
ING
SocGen
Nordea
BARC
HSBC
StanChart
DBK
RBS
BNP
CA
Santander
0% 8.0%
MS GS BK C JPM WFC STT UBS CS
Source: BofA Merrill Lynch Global Research Source: BofA Merrill Lynch Global Research, company data Source: BofA Merrill Lynch Global Research, company data
Note: BofAMLe using publicly available data (as of 1Q18) Note: Data as of 1Q18 Note: Data as of 1Q18
TLAC rules at the individual country level will also need to distinguish which legal entity
must hold TLAC-eligible capacity, an important discussion for banks without a U.S.-style
holding company structure.
As with the other rules, there is a phase-in period to compliance. Global banks must
meet a minimum ratio of at least 16% of RWA and a Leverage requirement of at least
6% by 2019 (see Exhibit 54). Emerging market GSIBs will have additional time to
comply, with a required TLAC/RWA ratio of 16% (and 6% TLAC/Leverage exposure) by
January 1, 2025 and a full compliance by 2028.
For example, if Bank X based in the U.S. were to have a subsidiary of material size in
Brazil, the Brazilian government would not have to worry that Bank X has kept all loss
absorbing assets in their home country, allowing their Brazilian subsidiary to dissolve
without absorbing any losses.
The U.S final requirement sets out two minimum ratios outlined below. Per the final
rule, all of the requirements phase in on January 1, 2019.
• Based on RWAs
U.S GSIBs are required to measure outstanding LTD/TLAC against both the RWA- and
TLE-based required ratios and comply with the most binding one. Important to note, a
GSIB in excess of the total TLAC (plus buffer) requirement could have a shortfall under
the LTD requirement, or vice versa, depending on the composition of its outstanding
TLAC.
How was the U.S. final rule different from the U.S. NPR?
The Fed published the TLAC NPR (Notice for Public Rule Making), the “proposal” in
October 2015. The biggest difference between the proposal and the U.S final rule
published in December 2016 included:
• The requirement for eligible debt/TLE is fixed at 4.5% (see Exhibit 55).
• Unsecured.
• “Plain vanilla”:
4. Not include payment acceleration clauses other than “certain date” put,
insolvency or payment default related clauses (if issued after December 30,
2016).
Note the final rule grandfathers eligible debt issued under foreign law and/or containing
ineligible acceleration clauses if it was issued before December 31, 2016. This is
positive for U.S. GSIBs since nearly all of the outstanding debt issued by GSIBs includes
ineligible acceleration clauses, and a significant portion is under foreign law, non-
grandfathering such debt would have been market disruptive and result in very high
costs to banks given tenders and replacements, exchanges or obtaining bondholder
consent to amend ineligible indentures tend to be very expensive actions.
Eligible debt with remaining maturities of less than one year does not qualify. Debt with
remaining maturities between one to two years still qualify, but at a 50% haircut.
Although eligible external LTD maturing between one and two years would be subject to
a 50% haircut for purposes of the external LTD requirement, such eligible debt would
count at full value for purposes of the external TLAC requirement.
Ineligible debt includes subsidiary level debt, debt maturing in less than one year,
structured notes (including “principal protected notes”), trust preferred securities
(TruPS), “survivor put” debt, and both foreign law debt and debt containing ineligible
acceleration clauses if issued after December 2016.
Eligible debt with 1-2yr remaining maturity must take a 50% haircut
Haircut
(will count at full value for external TLAC).
TruPS
Source: BofA Merrill Lynch Global Research, company data, Federal Reserve
Note: GSIB surcharge for LTD requirement is the greater of Method 1 or Method 2
Unsurprisingly, given the structure of their business model, MS and GS are well above
the LTD requirement (see Chart 128 and Chart 129). C, JPM and WFC are moderately
above at the time of writing.
Chart 128: RWA-based LTD vs. requirement (as of 1Q18) Chart 129: TLE-based LTD vs. requirement (as of 1Q18)
40% 14% Minimum
Minimum 12.3% 12.2%
Outstanding eligible external LTD as % of
34.3% requirement,
35% requirement
12% 4.5%
30% 26.7%
total leverage exposure
10%
25%
8%
RWAs
20% 6.3%
6% 5.5%
13.5% 4.9% 4.6%
15% 3.9%
10.0% 9.7% 9.6% 9.1% 4%
10%
Source: BofA Merrill Lynch Global Research, company data, Bloomberg, Federal Reserve Source: BofA Merrill Lynch Global Research, company data, Bloomberg, Federal Reserve
Note: Above data represents BofAML estimates based on publicly available data for ease of Note: Above data represents BofAML estimates based on publicly available data for ease of
comparability comparability
The U.S. final rule removes the additional 2019-2022 phase-in period for the RWA-
based requirement and incorporates a 2% external TLAC TLE-based buffer.
Exhibit 56: External TLAC RWA-based ratio formula
Category Components
The U.S. final rule requires both an external TLAC RWA-based buffer (2.5% + method 1
surcharge) and an external TLE-based buffer (fixed at 2%) to be held in addition to the
18% RWA/ 7.5% TLE requirement. On the other hand, FSB deducts the 2.5% capital
conservation buffer and Basel III GSIB surcharge (known as Method 1 in the U.S.) from
qualifying capital without including additional buffers above its 18% RWA/ 6.75% TLE
requirement.
Exhibit 57: External TLAC Buffer
MS GS C JPM WFC BK STT
2.5% buffer 2.5% 2.5% 2.5% 2.5% 2.5% 2.5% 2.5%
Method 1 GSIB surcharge 1.0% 1.5% 2.0% 2.5% 1.5% 1.0% 1.0%
External TLAC RWA-based buffer 3.5% 4.0% 4.5% 5.0% 4.0% 3.5% 3.5%
External TLAC TLE-based buffer 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0%
Source: BofA Merrill Lynch Global Research, company data. As of 4Q17
The calculation of a bank’s actual external buffer itself is complicated (see Exhibit 68).
The central idea is that the CET1 (RWA-based) or Tier 1 capital (TLE-based) used toward
meeting the external TLAC requirement cannot be used to fill the external TLAC buffer,
both under RWA and TLE requirements.
Bear with us: to calculate a bank’s external buffer, investors need to start with the
minimum TLAC requirement, which as a reminder is 18% of RWAs and 7.5% of TLE.
Then, subtract from this percentage the ratio of qualifying AT1 capital and eligible LTD
as a % of RWAs or just the ratio of eligible LTD as a % of TLE depending on which of
MS, GS, JPM, C and WFC all met the external total TLAC (plus buffer) requirement at the
time of writing (see Chart 130).
Chart 130: TLAC by bank (as of 1Q18)
60% 55.0% Additional TLAC/RWAs
Qualifying LTD/RWAs
50%
40.9%
Jan '19 min + buffers
40%
30% 27.4%
24.3% 23.5% 23.4%
22.0% 22.5% 23.0% 22.0%
21.5% 21.5% 21.5%
20%
10%
0%
MS GS BK C JPM WFC STT
Source: BofA Merrill Lynch Global Research, company data, Bloomberg, Federal Reserve
Note: Above data represents BofAML estimates based on publicly available data for ease of comparability
The shortfall between actual and the requirement determines payout limitations, as
detailed by the schedule below (see Exhibit 59). For example, if it were 2019 today, a
bank with an external buffer at 80% of the requirement would be restricted to paying
out just 60% of earnings in the form of dividends and buybacks.
Exhibit 59: Payout ceiling (TLAC buffer level vs. requirement)
External TLAC buffer
Payout ceiling RWA-based buffer TLE-based buffer
level vs. requirement
C: 139% C: 216%
GS: 276% GS: 281%
No payout ratio limitation > 100% JPM: 110% JPM: 185%
MS: 385% MS: 312%
WFC: 133% WFC: 294%
60% ≤ 100% , > 75%
40% ≤ 75% , > 50%
20% ≤ 50% , > 25%
0% ≤ 25%
Source: BofA Merrill Lynch Global Research, company data, Bloomberg, Federal Reserve. Data as of 5/2018
RWA-based
(I) External TLAC buffer level 68,672 5.3% CET1 not used to meet (G)
(J) External TLAC buffer requirement 51,507 4.0% RWAs x 4.0% (2.5% + Method 1 surcharge)
Maximum external TLAC payout ratio No limit Check payout ratio calculation
External TLAC buffer shortfall (surplus) (17,165) -1.3% (J) - (I)
TLE-based
(K) External TLAC buffer level 132,023 5.9% Tier 1 not used to meet (H)
(L) External TLAC buffer requirement 44,915 2.0% TLE x 2.0%
Maximum external TLAC payout ratio No limit Check payout ratio calculation
External TLAC buffer shortfall (surplus) (87,108) -3.9% (L) - (K)
Denominators
RWAs 1,287,675
TLE 2,245,743
Source: BofA Merrill Lynch Global Research, company data, Federal Reserve, Bloomberg
Notes: Total shortfall numbers assume minimum CET1 capital ratios exceed TLAC RWA-based buffer requirements.
Originally, MREL was going to require all banks in the EU to hold 8% of assets, un-
weighted, in capital and bail in-able debt.
However, the current MREL rules have now set the requirement according to a RWA
denominator.
The Single Resolution Board (SRB) is responsible for setting the MREL requirements for
most of continental Europe, but has so far not published these at a single entity level.
The calculation is known, but the perimeter of the RWA could be different, as the ESRB
can take into account perimeter changes according to the bank’s living will.
The SRB also states minimum levels of the numerator be in the form of subordinated
instruments:
• G-SIBs minimum of 13.5% of RWA + Combined Buffer Requirements
MREL is applicable to many banks in the EU, not only the systemically important ones.
The bonds that qualify as MREL eligible are an issue of national discretion. For example:
• In Germany, a law makes existing unsecured debt eligible for MREL. This is why
DBK has an unusually high TLAC ratio. With new legislation expected in the
autumn, DBK is likely to be allowed to issue more “senior preferred” bonds
over time.
• In France and some other European countries, banks have begun issuing
“senior non-preferred” (aka SNP or Tier 3) bonds which are subordinate to
existing senior unsecured debt.
• Banks with existing (or new) holding company structures are gradually issuing
holding company debt to meet MREL/TLAC requirements.
Indirectly, given that CET1, surcharge, and SLR limitations are the base of the ratios,
businesses directly affected by these rules are indirectly impacted by TLAC.
Table 12: More and less valuable products under TLAC
TLAC: More Valuable Products TLAC: Less Valuable Products
LT unsecured funding Deposit funding
Electronic/agency trading Short-term Funding
Wealth management Structured funding
Asset Management Securitized products
Advisory High yield credit products
Payments
Futures clearing
Agency MBS
On April 11, the Fed and OCC proposed replacing the 4.5% TLE-based LTD requirement
with 2.5% plus 50% GSIB surcharge requirement and the 2% TLE-based TLAC buffer
with 50% G-SIB surcharge. We estimate that the proposed changes effectively lower
the total leverage exposure (TLE) based LTD and TLAC requirements by 25-125bp for
banks that are TLE TLAC constrained - C, JPM, STT and BK.
In a nutshell, the annual DFAST and CCAR process requires banks with over $50bn in
assets to demonstrate sufficient capital to withstand a hypothetical, highly stressful
operating environment, while executing on its submitted capital plan. DFAST is a
forward-looking quantitative evaluation of the impact, while CCAR is the overlay created
by the Fed that inputs capital planning and non-objections/objections to these plans by
regulators.
As a reminder, DFAST and CCAR are not the same process. DFAST is the process
mandated by law. The Dodd-Frank act requires stress testing annually and for there
to be three scenarios under which to stress the banks. The rest of the CCAR process,
including qualitative objections and non-objections described below, was designed by
the Fed however is not law.
The DFAST results, which come out a week prior to the CCAR results, include the
previous year’s capital plan (without stock buybacks and other repurchases), not the
current year’s “capital ask”, like CCAR. Note however, both CCAR and DFAST include an
estimate of projected revenues, losses, reserves, pro forma regulatory capital ratios, and
any other additional capital measures. In addition to plans to distribute dividends and/or
buyback stock, a capital plan can also include potential acquisitions, divestitures, and/or
more general balance sheet actions (e.g., calling debt). The test is divided into two
portions:
• Quantitative assessment of a bank’s post-stress capital adequacy; and a
• Qualitative assessment of the bank’s capital plan, and importantly, the practices
and processes used by the banks to assess its capital needs. This part of the test
has been an issue for banks that have received objections or conditional non-
objections, more so than the quantitative, or “mathematical”, part of the test.
That said, participating banks are allowed to revise their capital plans between the
release of the DFAST results and the CCAR results (i.e., “mulligan” clause), as the DFAST
results can be informative of how the CCAR quantitative results will turn out. For
example, during the 2016 test, MTB adjusted its original capital plan.
“This enhanced transparency will bolster the credibility of our stress tests and help the
public better evaluate the results,” Vice Chairman for Supervision Randal K. Quarles said.
“The proposed changes will also generate valuable insight from stakeholders and we
look forward to it.”
Moreover, the Fed is proposing to modify its framework for the design of the annual
hypothetical economic scenarios. The modifications are set to improve transparency and
to further promote the resilience of the banking system throughout the economic cycle.
In particular, the revisions include more information on the hypothetical path of house
prices as well as notice that the Board is exploring the addition of variables to test for
funding risks in the hypothetical scenarios.
Source: Federal Reserve. One asterisk denotes BHC is excluded from qualitative portion of CCAR. Two asterisks denote BHC is subject to
counterparty default. Three asterisks denote BHC is subject to global market shock and counterparty default
As noted above, passing the severely adverse scenario is the binding constraint for
capital plans. Although the Fed provides quarterly, progressive parameters for all 16
domestic factors and 12 international factors within its supervisory macroeconomic
scenarios, the table below provides the peaks or troughs for each factor (see Exhibit 64).
Exhibit 64: Severely adverse scenario shocks
Starting Stressed Ending Relative stress More/less
Point Level Point 2017 2018 Δ Severe
Domestic
Dow Jones Index 27,673 9,689 20,168 -50% -65% (15bp) More
Real GDP growth 2.70% (8.90%) 4.50% (1,060bp) (1,160bp) (100bp) More
Unemployment rate 4.10% 10.00% 8.60% (530bp) (590bp) (60bp) More
10-year Treasury yield 2.40% 2.40% 2.40% (140bp) 0bp 140bp Less
Mortgage rate 3.90% 6.00% 4.70% 70bp 210bp 140bp More
Market Volatility Index (VIX) 13 62 14 205% 376% 171bp More
CRE Price Index 279 167 181 -35% -40% (5bp) Similar
House Price Index 194 136 143 -25% -30% (5bp) Similar
BBB corporate yield 4.00% 8.10% 5.00% 240bp 410bp 170bp More
International
Euro area real GDP growth 2.30% (5.20%) 2.30% (840bp) (750bp) 90bp Less
USD/euro exchange rate 1.20 1.07 1.14 -12% -11% 1bp Similar
Developing Asia real GDP growth 5.90% (1.50%) 6.60% (610bp) (740bp) (130bp) More
Japan real GDP growth 1.80% (11.40%) 1.30% (1,030bp) (1,320bp) (290bp) More
Yen/USD exchange rate 112.70 99.30 100.10 -8% -12% (4bp) More
U.K. real GDP growth 1.40% (5.10%) 2.60% (800bp) (650bp) 150bp Less
USD/pound exchange rate 1.35 1.27 1.29 -9% -6% 3bp Similar
Source: BofA Merrill Lynch Global Research, Federal Reserve
• 10yr Treasury rate drops to 2.4% and remains constant (vs. dropping to 0.8% last
year and steadily rising to 1.8%)
• The decline in real estate (residential and commercial) prices begins with a steep
drop off, reaching double digit declines starting in the second quarter of the
forecast period
In December 2017, the Fed approved a modification of the scope of the global market
shock component of the supervisory stress test to apply to a firm that has aggregate
trading assets and liabilities of $50bn or more, or equal to or greater than 10% of total
consolidated assets. The following IHCs will become subject to the global market shock
beginning in CCAR 2019: Barclays US LLC, Credit Suisse Holdings (USA), Inc., DB USA
Corporation, HSBC North America Holdings Inc., UBS Americas Holding LLC, and RBC
USA HoldCo Corporation. In CCAR 2018, in lieu of the global market shock component,
these IHCs will be subject to interim market risk components in the supervisory adverse
and severely adverse scenarios.
The global market shock for the severely adverse scenario incorporates a
sudden sharp increase in credit risk, a steepening of the U.S. yield curve, and a
general selloff of U.S. assets relative to other developed countries. Notable
differences in the 2018 severely adverse global market shock scenario relative
to last year include;
• Rise and steepening of the U.S. yield curve
That said, in February 2017 the Fed removed the qualitative component for BHCs with
$50-250bn in total assets. This is a positive in two ways. First, it streamlines the
“procedure” for CCAR participation. Second, the element of risk of getting an
objection on a qualitative basis is clearly removed
For BHCs over $250bn, the entire qualitative process remains opaque, though the
assessment is crucial. BHC’s with total assets of $250bn or greater have the equivalent
of a “to-do” list from regulators each year – not disclosed publicly – that address
identified deficiencies in the capital planning and risk management process. For
example, feedback from management teams suggests that regulators are particularly
interested in the “systems readiness” of these BHCs. Think of this like the Y2K
phenomenon, where systems could not process beyond the year 1999. The Fed appears
to want to make sure those BHCs can handle various severe economic stressed in real-
time.
BHCs with $50-250bn in assets are the immediate beneficiaries. The Fed’s proposal
may not only result in increased capital return to shareholders, but could also befit banks
to reallocate “dead weight” capital sitting in Treasury securities to revenue-generating
businesses.
That said, we believe this poses an interesting decision tree about organic and inorganic
growth at firms above $200bn in assets. In other words, do these institutions stay
below the threshold for some time, or leap over it to scale the cost of additional CCAR
burden?
As we note previously, the amount of additional capital distributions a BHC can make
during a capital plan cycle was reduced from 100bp to 25bp of a firm’s tier 1 capital (as
of the most recent first quarter FRY-9C regulatory filing). For example, a BHC requesting
an additional buyback from the Fed in February 2018 would be restricted to <25bp of its
tier 1 capital as of 1Q17. Banks do not have to be re-stressed to receive a non-objection
from the Fed.
To qualify for these additional distributions, the Fed states that the BHC would need to
hit the following standards:
• Remain well capitalized pro forma for this additional request. Well-capitalized
is defined as at least 6% tier 1 capital ratio and at least 10% total risk-based
capital ratio
• Fed has 15 days to object or not object, upon receiving the notification
• In the Euro area, concentrations and levels of problem assets – key attributes of
stress tests - are a part of the annual supervisory process. Periodically, the
authorities run system-wide stresses but these have not tended to be binding for
most banks. More recent stress tests are used to help set the SREP capital
requirement for the year ahead.
• The UK has developed its stress testing to the most complete degree and
embeds it in annual bank capital requirements.
The test will cover 44 banks, equal to more than 70% of banking assets.
As well as an adverse set of macroeconomic factors set by the ECB (e.g. real GDP
growth of -1.2% in ’18, -2.2% in ’19 and +0.7% in ’20, 19% fall in residential house
prices, 20% fall in CRE), the test will focus on four areas:
• Abrupt and sizeable repricing of risk premia and resulting tightening of financial
conditions in Europe
• Adverse feedback loop of low bank profitability and low economic growth
The test also has a conduct risk element whereby banks must estimate conduct and
litigation risk costs on a stricter basis than current accounting rules allow.
UK stress testing
We believe the UK system is most interesting. It is in some ways more evolved than
CCAR in that it seeks to be predictable and symmetrical. For example, in a period where
asset prices move further away from historical averages relative to incomes, the stress
test is designed to show a greater potential drawdown in asset prices. But in a
recession where asset prices fall against historical averages, the opposite is also
intended to take place and the stress test be less onerous. In this way, the Bank of
England characterizes itself as having a consistent risk appetite and seeks to provide
clarity that it will be balanced in its capital requirements.
The 2018 stress test for the UK will be the same as the 2017 test in that it is more
severe than the global financial crisis and in the opinion of the FPC, it also encompasses
a range of outcomes that could be related to Brexit.
The UK is likely to include the ring-fenced entities separately in the 2020 stress text
onwards.
The macroeconomic forecasts are similar to the EBA test (UK GDP start to trough of -
4.7%, residential house prices -33%, CRE -40%. Similar to the EBA test, it also allows
for an element of conduct risk costs.
In Japan, there is no current requirement for regular stress tests by the authorities.
Although the financial authorities have also started to argue the need for this to be
introduced, no specific protocol has been established and schedules are yet to be
decided.
CCAR serves as a check for banks to benefit from accrual accounting when it comes to
lending decisions. What do we mean by this? In today’s economic backdrop, it is likely
that losses across all credit categories will remain de minimis, or at least in our view
below cycle averages. As such, losses and higher provisions may not impact a bank’s
EPS power and valuation for a few years. That said, CCAR has the impact of “marking to
market” a bank’s credit risk profile every year. After all, the Fed can certainly apply very
high stressed losses to a portfolio they consider risky, or object to an institution’s
capital plan for qualitative reasons should it feel that its risk profile is changing for the
worse.
Table 14: More and less valuable products under supervisory stress
Stress Tests: More Valuable Products Stress Tests: Less Valuable Products
Historically low loss content loan products International lending exposure
Subprime lending
Global trading operations
This is applicable to all GSIBs and all banks in Europe. In the U.S., large regional banks
above $50bn in assets are also subject to this rule, but with modified (less stringent)
requirements. Like with NSFR (the net stable funding ratio, which we will discuss in the
following section), this rule distinguishes between “good” liabilities and “less good”
liabilities. Moreover, given that banks are required to hold HQLA against certain
Impacted banks are mostly in compliance with this ratio today (see Chart 131 and Chart
132). That said, this ratio is highly volatile, and we think this could have two specific
negative impacts on bank net interest margins and ROEs: 1) the aforementioned build-
up of liquid assets today; and 2) potentially aggressive competition for retail deposits,
which, like the NSFR requirement, are more highly valued by this regulatory rule.
Chart 131: Liquidity coverage ratio by GSIB (as of 1Q18)
250% Americas Europe Asia
150%
100%
50%
0%
ICBC
WFC
CCB
ABC
BOC
ING
SocGen
GS
BK
Nordea
MUFG
BARC
HSBC
C
StanChart
RBC
RBS
DBK
UBS
BNP
JPM
Unicredit
STT
MS
CS
CA
Mizuho
SMFG
Santander
Source: BofA Merrill Lynch Global Research, company data
Chart 132: Liquidity coverage ratio U.S. regional banks (as of 1Q18)
160%
144% Regulatory
140% 126% requirement, 100%
120% 113%
>100% >100% >100% >100% >100% >100% >100% >100% >100%
100%
80%
60%
40%
20%
0%
ZION
CMA
CFG
MTB
KEY
BBT
HBAN
FITB
PNC
USB
STI
RF
In the following section, we walk through the components of the numerator and
denominator:
≥ 100%
• Includes certain residential mortgage-backed securities (RMBS) in Level 2B • Corporate debt securities are not included in Level 2A assets
assets subject to a 25% haircut
• Certain corporate debt securities may qualify as Level 2B assets subject to a
• Includes certain A+ to BBB- corporate debt securities in Level 2B assets 50% haircut
subject to a 50% haircut
Source: BCBS, Federal Reserve Board, FDIC, OCC
Qualifying assets for HQLA: Level 1 assets, Level 2A assets, and Level 2B assets (see
Table 16 for U.S. and Table 17 for EU). Clearly, excess cash at central banks, and U.S.
Treasuries would be considered the most liquid type of asset. The combination of Level
2A and 2B assets cannot exceed 40% of HQLA, after accounting for haircuts. Lastly,
Level 2B assets are capped at 15% of HQLA.
Table 16: U.S. HQLA rules summary
Category Level Includes Haircut
Central bank reserves, USTs, foreign withdrawal reserves, and certain U.S.
Level 1 None
gov’t/ sovereign entity securities
U.S. GSE securities and certain sovereign entity securities (those that have
Level 2A 15%
no higher than a 20% risk weight)
Certain publicly traded corporate debt, publicly traded common stock (not 50% (cannot exceed 15% of
Level 2B
ETFs) and certain municipal securities total stock of HQLA)
Source: Federal Reserve Board, FDIC, OCC. Note Munis are HQLA eligible only in the Fed rule. The OCC and FDIC rules do not include munis in HQLA
Below, we provide a graphic for what typically constitutes HQLA for an institution, using
C as an example (see Chart 133).
$400 bn 90%
Available Cash
$350 bn 80%
US Treasuries 70%
$300 bn
60% 84%
$250 bn Foreign Govt Level 1 Assets
50%
$200 bn Level 2 Assets
40%
US Agency/Govt
$150 bn
Gtd 30%
$100 bn IG Corp/Equities 20%
$50 bn 10% 16%
$0 bn 0%
Expected cash outflows are calculated by multiplying the outstanding balance of various
liabilities and off-balance sheet items by “expected run-off” or outflow rates. (A similar
approach is applied for inflows; however, inflows are subject to a cap of 75% of
expected outflows). The higher the “run-off” rate, the higher the amount included in the
outflow calculation. For example, a 5% outflow rate is applied to deposits classified as
“stable” (see Exhibit 69). Conversely, a 100% rate applies to short-term borrowings.
Total Net
= +
Cumulative Cash Maturity Mismatch
Cash Outflow
Outflows ― Capped Cumulative Cash Inflows
Add-on
Amount
lesser of
Instruments with no
maturity date are
Cumulative Cash Inflows
generally excluded
from cash inflows Net Cumulative
and Peak Day Maturity
minus
75% of cumulative cash outflows Outflow Net Day 30
Cumulative Maturity
Outflow
LCR = 120%
Source: company filings
Note: Foreign Govt. includes securities issued or guaranteed by foreign sovereigns, agencies, and multilateral development banks
This next chart provides the publicly disclosed LCR breakdown for UBS (see Exhibit 71).
LCR = 136%
1
. Includes cash and balances with central banks and other eligible balances as prescribed by FINMA
2.
Calculated after the application of haircuts and accordance with FINMA requirements
Source: company filings
Note (1): Includes cash and balances with central banks and other eligible balances as prescribed by FINMA. (2) Calculated after the application of haircuts and accordance with FINMA requirements
145%
Liquidity coverage ratio
-3%
140%
135% 144%
141%
130%
125%
Current Pro forma
On December 19, 2016, the Federal Reserve published their final U.S. LCR public
disclosure requirements. Starting April 1, 2017, required firms started to disclose their
consolidated LCR every quarter according to the U.S. rules. Required disclosures include
the Firm’s average LCR for the quarter, amount of HQLA by category, and net outflow
amounts including from retail deposits and derivatives, and a qualitative discussion of
material drivers of the ratio, changes over time, and causes of such changes.
Frequency of Calculation Monthly basis 1H15: monthly basis; 2H15 on: daily basis 2016: monthly basis
We think the most impacted product under LCR will be deposit funding (see Table 18).
Under LCR rules, deposits are assigned liquidity values based on expected behavior under
stress, the type of deposit and the type of client. In other words, the final U.S. rule
prioritizes operating accounts of consumers (i.e., retail and commercial banking deposits)
and corporations, while assigning a 100% outflow assumption to balances of financial
institutions. In particular, retail deposits likely have very low outflow assumptions. This
leads us to believe that retail deposit competition – and deposit betas – could be greater
even before interest rates normalize in earnest. As mentioned in the NSFR, some non-
traditional deposit gatherers (GS) are already offering significantly higher rates on
deposits than traditional deposit gathering GSIBs.
That said, any products that have been traditionally funded short – such as secured
financing, or prime brokerage/repo – could also be impacted. This is where the
discussion on LCR sensitivity fully kicks in. LCR could be more of a binding constraint
than is appreciated by the market, given the sensitivity of the ratio to small shifts in
funding.
Table 18: More and less valuable products under LCR
LCR: More Valuable Products LCR: Less Valuable Products
Retail deposits Liquidity and credit facilities
Operational corporate deposit Lending products to financial institutions
Nonoperational corporate deposits
Financial institution deposits
Prime brokerage
Source: BofA Merrill Lynch Global Research
On the U.S. side, the FDIC, the Federal Reserve and the OCC jointly released NPRs
(notice of proposed rulemaking) to outline the proposed local interpretation of the rule.
The NSFR, as proposed, is applicable to both the holding company and the bank
subsidiary. Interestingly, these rules were not pushed down to the broker/dealer level
given lack of jurisdiction by both the FDIC and Fed. This is important to note, given that
most short-term (and “least stable”) funding sources for universal banks are often
housed at the broker/dealer subsidiary. Similar to LCR (liquidity coverage ratio), a
modified, or less strict, version is applicable to banks with $50-250bn in assets.
The NSFR could be most impactful on the traditional investment banking monoline
model, which typically holds a significant amount of short-term funding in matched
trading books and has historically been light on deposit funding. But, given that the U.S.
proposal is similar to Basel, we don’t expect a significant shortfall. In addition, U.S.
proposals appear to be limiting how much “excess” stable funding that are typically
housed at bank subs can be counted for the holding company’s requirement, which we
talk about in more detail in the following sections.
We estimate that most impacted banks are at 100% today or higher (see Chart 135).
Moreover, while the U.S. proposal is modestly tougher than the Basel rule, we think the
current U.S. NPR should ease concerns that the NSFR will be a material, incremental
catalyst for further liquidity reduction in the repo market and areas of fixed income.
Chart 135: Net stable funding ratio as of 1Q18
Many banks do not disclose the NSFR but merely indicate that they are “in compliance” with the ratio
160%
Americas Europe
140% Regulatory
requirement, 100%
120%
100%
80%
60%
40%
20%
0%
WFC
ING
SocGen
GS
BK
Nordea
HSBC
BARC
C
StanChart
RBS
UBS
DBK
JPM
BNP
STT
Unicredit
MS
CS
Santander
Source: Company data, BCBS, Federal Reserve. Although applicable, NSFR has not been disclosed by Chinese or Japanese GSIBs
Note: many banks do not disclose the NSFR but merely indicate that they are “in compliance” with the ratio. As such, we assumed 100%
The regulatory minimum of this ratio is 100% by January 2018. Basel also
requires banks to meet the requirement on an ongoing basis although they will only be
reported quarterly. Below, we break down the major components of the numerator and
denominator.
• 90-95% ASF factor: Retail deposits, deposits under <$250k. Retail deposits
count between 90-95%, regardless of maturity and collateralization. Brokered
deposits and sweep accounts (regardless of counterpart) would fall in this category,
so long as the deposit is covered by deposit insurance. In the U.S., this is currently
$250,000. CDs over a year in maturity also fall in this category.
• 10% RSF factor: Loans (secured) maturing in less than 6 months, loans
secured by Level 1 assets, and loans to banks and other financial sector
entity.
• 15% RSF factor: Level 2A liquid assets, unsecured loans <6 months in
maturity. Level 2A assets include U.S. GSE (government sponsored entities)
securities and certain sovereign issued securities, so long as the risk weight is 20%
or less. In Europe, ECAI 2 covered bonds (A+ to A-), non-EU covered bonds rated
ECAI 1, and corporate bonds rated ECAI 1 are included.
• 50% RSF factor: Level 2B liquid assets, loans maturing between 6-12
months (retail and wholesale), PSEs (i.e., munis). Level 2B liquid assets include
certain publicly traded corporate debt and publicly traded common stock (not ETFs,
or exchange-traded funds), as well as municipal securities.
• 65% RSF factor: Retail mortgages >1 year in remaining maturity, loans
maturing >1 year. Retail mortgages that belong in this category cannot have a risk
weight of more than 50%, while other loans over one year in maturity cannot have a
risk weight of greater than 20%.
• 85% RSF factor: Retail mortgages and wholesale loans with higher risk
weightings, non-HQLA securities. Retail mortgages with over 50% risk weight
and wholesale loans with over 20% risk weight fall in this category. Meanwhile,
wholesale and other consumer loans with a risk weight of greater than 20% fall
under here. All other securities that do not qualify as HQLA (high quality liquid
asset, as defined by the LCR) also fall in this category. These include: cash,
securities, or other assets posted as initial margin for derivative contracts; cash or
other assets that would contribute to the default fund of a central counterparty
(CCP); unencumbered securities with remaining maturity of >1 year and exchange-
traded equities that do not qualify as HQLA; and physically traded commodities,
including gold.
Liabilities and capital line items Factor ($mn) Asset line item Factor ($mn)
Retail/ small business deposits 90.0% 278,010 Cash 0.0% 0
+ +
Corporate deposits 50.0% 333,000 Loans and advances to banks 50.0% 121,285
+ +
Deposits from banks 0.0% 0 Loans 80.0% 528,467
+ +
Short-term borrowings: Securities:
< 6mos 0.0% 0 Level 1 5.0% 7,955
6mos to ≤ 12mos 50.0% 18,047 Level 2 40.0% 58,403
+ +
Long-term borrowings 100.0% 237,938 Net derivatives 100.0% 67,957
+ +
Trading liabilities 0.0% 0 Other assets 100.0% 184,474
+ +
Regulatory capital 100.0% 178,091 Off-b/s exposure 5.0% 26,143
Available Stable Funding (BofAMLe) 1,045,086 Required Stable Funding (BofAMLe) 994,684
NSFR = 105%
Source: BofA Merrill Lynch Global Research, company data
BofAMLe calculation based on publicly disclosed data
That said, there was language in the proposal that implied that holding companies will
be precluded from utilizing the full funding synergy of having both a stable deposit-rich
bank subsidiary and a broker/dealer subsidiary with short-term funding for full
compliance to the rule. While the limitations were not yet explicitly set and language in
the proposal is vague, it appears as though U.S. regulators are setting the stage to limit
the amount of “excess” stable funding a bank sub can count toward the holding
company ratio.
To provide clarity to this, let’s say bank holding company A had two subsidiaries: a bank
sub with an NSFR of 130% and a broker/dealer with an NSFR of 75%. If the amount
over 100% at the bank sub that can count toward holding company compliance is
limited, then NSFR has to be managed at the parent to address the broker/dealer
contribution shortfall.
While the rule is naturally challenging for monoline investment banks, this proposed
limitation would be more incremental for money-center banks – whose rich deposit
bases have been supporting its RSF calculations. Solutions could include issuing more
Additionally, the new funding regulations also have the impact of valuing deposits
differently. Given the high ASF factor, the NSFR places a high premium on retail
deposits – which might require banks to pay up for this type of funding in a rising rate
environment, or even today, as we’ll discuss below. Conversely, rates on financial
institution deposits will likely exhibit a low beta in a rising rate environment, given their
punitive treatment under the NSFR (as well as the LCR and GSIB surcharge).
Table 19: More and less valuable products under NSFR
NSFR: More Valuable Products NSFR: Less Valuable Products
Retail deposit funding ST funding
Rates (Treasury, agency) Financial institution deposits
Nonoperational corporate deposits
Equity derivatives
Prime brokerage
Repo
Non agency MBS
Municipal markets
Credit products
Structured products
Source: BofA Merrill Lynch Global Research
provisions
The mission statement of the Dodd-Frank Act is as follows: (1) address risks to the
stability of the U.S. financial system; (2) end too-big-to-fail bailouts of large,
complex financial institutions; (3) increase transparency and regulation for certain
complex financial instruments; and (4) strengthen protections for consumers and
investors.
There are sixteen provisions (Titles I-XVI) within the 849 pages of Dodd-Frank
that address the primary goal of the legislation. For the purpose of this report, we
focus on those that we view have been the more impactful to industry returns, bank
stock performance, and the economy:
We explain these provisions and its impact on the industry and the economy in this
section. Further, we also explore current proposed alternatives to Dodd-Frank.
• The Financial Stability Oversight Council (FSOC), which is chaired by the Treasury
Secretary
Alternate proposals
There appears to already be momentum in Washington to raise the threshold in the
definition of SIFI, as defined by Dodd-Frank. The Economic Growth, Regulatory Relief,
and Consumer Protection Act – or more commonly referred to as the Crapo Bill – would
change the regulatory framework for banks with less than $10bn in assets and for banks
with over $50bn in assets. At the time of us writing this report, the bill has already been
passed in the Senate and has been sent to the US House of Representatives for voting.
That said, Dodd-Frank appears to give the Federal Reserve broad scope in the design of
and consequences of such a test, noting that the Fed “may develop and apply such other
analytic techniques as are necessary to identify, measure, and monitor risks to the
financial stability of the United States”. As such, the robust framework (e.g., CCAR 2018
has 28 variables to stress) has been created by the Fed, and not by Dodd-Frank.
Mostly through Regulation Y, the Federal Reserve then established the CCAR as a
“complementary” exercise to DFAST. The Fed itself notes that DFAST and CCAR are
“distinct testing exercises”. Of course, the Fed has taken Title I several steps further by
evaluating individual capital plans by each participating BHC for objection or non-
objection.
Applicability
The Fed is required to administer DFAST to banks above $50bn in assets. Each bank
holding company is also required to conduct its own stress testing. The frequency
requirement is twice a year for BHCs above $50bn, and once a year for BHCs above
$10bn. This said, CCAR does not exist for BHCs below $50bn. Therefore, DFAST is not
typically seen as a constraint for capital planning and distribution for institutions below
$50bn in assets.
The FBO rule is a combination of a legal requirement set out in Dodd-Frank and rules
set out by the Federal Reserve, the regulator for bank holding companies and IHCs in the
U.S. Branch assets are not included within the IHC, although some different U.S. rules do
also apply to branches. The Fed has interpreted the Dodd-Frank Act and translated it
into actual metrics and rules for the banks.
The FBOs are required to comply with U.S. capital rules, consistent with U.S. BHCs. The
leverage ratio requirements were effective January 1, 2018 (see Exhibit 77).
Exhibit 77: IHC Requirements
IHC Requirements
US leverage ratio of > 4% (tier 1 capital / US GAAP
All IHCs
assets)
IHCs > 250bn in assets Supplementary leverage ratio of > 3% (not 5%)
Required to establish a US risk committee and a US chief
risk officer
For major foreign banking
Enhanced risk management
organizations with non branch
Enhanced liquidity risk management
US assets > $50bn
Locally binding liquidity coverage ratio and liquidity stress
tests
Source: Federal Reserve, Dodd-Frank
The foreign IHCs are currently going through their first public CCAR process (i.e.
required to file capital plans). The banks took part in the CCAR process in 2017, but this
was not publically disclosed, at the time.
Applicability
The formation and adherence to this rule would apply to foreign institutions that would
have, in aggregate, asset exposure of $50bn or greater, domiciled in the U.S. (Dodd-
Frank Act (section 165(a) 1).
European institutions that would be subject to IHC requirements: Credit Suisse, UBS,
Deutsche Bank, Barclays, HSBC, BNP Paribas, Santander, and BBVA.
That said, the applicability of which rule (IHC vs. FBO) to which kind of foreign banking
organization is rather complex. Please see Exhibit 78 below.
Exhibit 78: U.S. rules for FBOs change significantly depending upon the global size, U.S. presence and legal structure
Size of FBO Large Large Reasonable
US presence Large Limited Limited
FBO US assets >$50bn <$50bn >$10bn
FBO Global assets >$50bn >$50bn <$50bn
Yes (unless non-branch assets Yes (unless non-branch assets
IHC needed? No
<$10bn) <$10bn)
Same as US BHCs, plus annual
Capital requirements Same as US BHCs No
capital plan
Yes...but can be home regulator
Stress test Same as US BHC Same as US BHC
version
US liquidity requirements for IHC
Yes, report results of internal test
Liquidity and branches + liquidity stress No
to Fed
tests.
Yes. Committee must have one
Risk management independent expert + US chief Yes Yes, if publically listed
risk officer required
Source: Federal Reserve, BofA Merrill Lynch Global Research
IHC = intermediate holding company. FBO = federal banking organization
o This has meant the IHC assets have fallen 27% since 3Q16 (the first
public reporting date), while assets in the branch have increased 37%.
o The CET1/leverage ratio of the IHC has improved from 3.3% in 3Q16
to 4.6% by FY17.
"Cost" of regulation
The cost of the IHC requirement comes in several forms:
2. Lower revenues from either higher funding costs or negative carry on liquidity
portfolios;
4. New for 2018 is the potential for these structures to incur additional US tax
burdens through the BEAT tax.
Operating expenses
There was a significant cost in setting up the IHCs. Ongoing costs should not be
underestimated, either: additional IT infrastructure, reporting requirements, data quality
improvements as well as the cost of an additional board and risk committee can all add-
up to be a significant drag on earnings. Further, bank management teams have
suggested new legal structures seem to cost banks more than expected.
While not IHC specific, some banks have given some numbers on the costs of moving to
new legal structures:
• Credit Suisse had reported annualized costs of almost CHF0.5bn in 2015 from its
legal entity program, which consists of the IHC, the Swiss legal entity and the
formation of a service company. It expects such costs to become minimal by 2018.
• A Bloomberg article from June 2016 suggested a cost of between $0.1-$0.5bn per
bank in legal, technology and other compliance costs. Therefore, in our view, it may
be reasonable to assume that setting up a separate legal structure could translate
into incremental cost that falls in the aforementioned range.
“…in particular as it relates to our markets business we have in the IHC, there is
also quite some internal transactions from a group's point of view, but from an
IHC, it's between IHC and rest of group, which creates leverage in the IHC which
we don't necessarily have to leave in place.” – Deutsche Bank Deputy former
Chief Executive Officer and CFO Schenck, capital raising presentation, 6th
March 2017
“What you have in there, for example, assume that the capital is actually
intercompany funding because it reflects how the IHC is actually funding which
actually attracts capital itself. And the second component is the SRU [Strategic
Resolution Unit], a large SRU actually sits in the IHC. Therefore, that, A, it's partly
generating, mostly generating the losses which you actually have seen in that
filing; and, B, as the SRU wind down, that's true, which we talked about already,
that will itself radically improve the financial performance of the IHC, additionally
to what Brian's talked about in terms of the projections for the Global Markets
business.” – Credit Suisse CFO Mathers, Strategy update, Dec ‘16
Alternatively, the European banks may fund directly in the U.S., but we expect this would
be more expensive.
• Most IHCs are materially better capitalized than the overall European consolidated
groups.
Total assets 141,413 140,699 148,248 273,486 139,136 157,927 87,321 128,294
Tangible assets 132,075 128,250 147,267 268,594 133,031 155,386 82,074 122,089
Tangible equity 14,776 8,276 6,961 17,683 12,432 11,557 7,538 17,305
Applicability
Resolution planning, or “living wills”, are applicable to bank holding companies with
$50bn in assets or greater and nonbank financial companies designated by the FSOC for
supervision by the Federal Reserve and the FDIC. More restrictive planning and
oversight are placed on the large SIFI institutions.
In assessing resolution plans, the Fed and FDIC evaluate seven key areas:
1. Capital
2. Liquidity
3. Governance mechanisms
4. Operational capabilities
7. Responsiveness
We discuss the major objectives for each of those core areas and how the agencies are
evaluating the financial institutions with regards to each of those areas (see Exhibit 80).
Exhibit 80: Evaluation of resolution plans
Description Objective for each key area Evaluation by the Agencies
Firms must provide sufficient capital to material entities to Determined whether the firm had enough resources to
1) Capital ensure that they can continue to provide critical services as recapitalize or support all entities needed to execute its plan
the firm is resolved under its scenario
Appropriately forecast size and location of liquidity needs;
Firms must be able to reliably estimate and meet their
2) Liquidity incorporation into everyday process; size, positioning of
liquidity needs prior to, and in, resolution
resources
Firms must have an adequate governance structure capable
3) Governance mechanisms of identifying the onset and escalation of financial stress Firms' impact of plan on broader financial system
events in sufficient time
Firms must maintain significant operational capabilities and Possess fully developed capabilities related to managing
4) Operational capabilities
engage in regular contingency planning collateral, have management information systems
Simplification of legal entity structure to facilitate orderly Development of criteria to achieve structure and actionable
5) Legal entity rationalization
resolution options to wind down or sell discrete operations
6) Derivatives and trading Understanding the impact of trading activities at major dealer Completeness of supporting analyses in the context of
activities firms and its impact on the broader financial system broader resolution plans
Firms should take agency guidance into account for
7) Responsiveness Compliance with prior feedback in developing their new plans
developing future plans
Source: BofA Merrill Lynch Global Research, Federal Reserve, Dodd-Frank
Once each institution has submitted its “living will” plans, the Federal Reserve Board
(FRB) and FDIC review the documents and inform the institution whether there are any
deficiencies or shortcomings to their resolution plans. In April 2017, the Fed/FDIC
Institutions have two years to remedy such requirements. If, following a two-year period
beginning on the date of the imposition of such requirements, a firm still has failed to
adequately remediate that deficiency, the agencies, along with the FSOC, may jointly
require the firm to divest certain assets or operations.”
Cost of regulation
The cost associated with resolution planning stems from personnel and systems costs,
reduction of legal entities, and additional liquidity requirements.
For example, JPM noted that it has added $50bn in incremental liquidity as a result of
complying with resolution planning. While other SIFI banks have not disclosed the
amount of liquidity, they have added liquidity since receiving feedback on their living will
submission and each bank’s liquidity requirement will vary depending on their legal
entities structure, we believe that they have all needed to increase their liquidity levels
even if they were compliant with LCR rules.
It is hard to separate liquidity build stemming from compliance with the liquidity
coverage ratio (LCR, a Basel prudential requirement) vs. what is required from resolution
planning requirements. However, since 2010, cash and securities have grown as a
percentage of earning assets, particularly at large SIFIs (see Chart 136). This has come
at significant cost to net interest margins, which would be naturally lower with greater
liquidity requirements.
Chart 136: SIFI banks have built up greater cash and securities balances
30.0% 29%
Cash and securities as % of earning
29.0%
+3% in sev en
28.0% y ears
assets (%)
27.0%
26%
26.0%
25.0%
24.0%
2010 2017
Source: BofA Merrill Lynch Global Research, company data, SNL Financial
Note: population includes U.S. financial institutions with over $50bn assets
Some liabilities
Funding Sources
transferred to bridge
(Equity wiped out)
In this scenario, the bridge company would first receive the assets from the distressed
bank, along with certain senior liabilities. Existing shareholders would be wiped out. By
establishing a new entity with a stronger balance sheet, critical bank operations
including ATM withdrawals and payments processing would be allowed to continue until
the bridge company can be orderly terminated. The bridge company would also be able
to support funding/liquidity needs through private market sources and the government.
How a bridge financial company would be funded: the Orderly Liquidation Fund
One of the most controversial parts of OLA involves the Orderly Liquidation Fund (OLF),
a fund created by the U.S. Treasury from which the FDIC can borrow from to support the
liquidation of a failing firm. Proponents of this rule note that the fund would not require
the use of taxpayer dollars, as it requires the Treasury to be reimbursed in full for any
losses that are incurred. Moreover, without the OLA, proponents note there is a greater
That said, OLA does not solve the issue of moral hazard as the Fed may face significant
pressure to authorize taxpayer bailouts in a distressed environment. Opponents of the
rule counter that the rule reinforces the notion of too big to fail as banks would have
support from the government in those situations. Additionally, the establishment of the
liquidation fund requires costs that can be deployed into other uses.
Applicability
The OLA process is applicable to all financial companies (referred to as “covered
companies” in Dodd-Frank) for which a “systemic risk determination” has been made.
This includes bank holding companies, banks, broker/dealers, insurance companies, and
other nonbank institutions. OLA was designed to supersede federal laws that were
designed more narrowly to conduct the liquidation and receivership of supervised banks,
insured depositories, and broker/dealers (previously handled by the FDIC and the
Securities Investor Protections Corporation (SIPC)).
Alternate proposals
The current administration may view Title II as unnecessary. Treasury Secretary Steven
Mnuchin has said, “...if we have proper regulation, a lot of the need for Title II also goes
away.”
That said, many of the larger institutions are concerned that a replacement would result
in more draconian regulation. Some Republicans have voiced their desire to replace the
OLA with a new chapter in the Bankruptcy code. In particular, Rep. Jeb Hensarling (R-
Tex.) and Secretary Mnuchin have discussed their support for changes to the bankruptcy
code that would be sufficient to prevent widespread damage in a distressed scenario.
Opponents of OLA are targeting the Orderly Liquidation Fund, which would effectively
limit the ability for OLA to function. Given that the OLA has an impact on congressional
budgets, the removal of the fund could be passed with a simple majority vote (i.e.,
“reconciliation” process). Rep. Hensarling’s Financial CHOICE Act calls for the repeal of
Title II.
"Cost" of regulation
According to the Congressional Budget Office (CBO), ending the Orderly Liquidation
Fund would reduce the deficit by $15.5bn over a 10-year period starting from 2018-
2027 driven by a reduction in both direct spending and revenues of $19.8bn and $4.3bn,
respectively (see Chart 137). We note that the Orderly Liquidation Fund is sourced from
assessments charged by the FDIC on the large financial companies. Additionally, while
there could be some cost relief to individual banks associated with a change in OLA, we
note that banks may need to establish and source their own fund if the OLF were to be
repealed.
$19.8 $4.3
Estimated deficit reduction from the
$20
$15.5
$15
$10
$5
$0
Revenue Expenses Net deficit reduction
Written broadly, we emphasize that Dodd-Frank itself does not establish capital and
leverage limits, but empowers the Federal Reserve to do so.
We have dedicated an entire section to these post-crisis Basel rules, and how local
regulators have interpreted them across the U.S., Europe, and Asia, beginning on page
93.
Alternate proposals
There are no numerical standards set in Dodd-Frank. Thus, the power granted to the
Federal Reserve to set regulatory standards is significant.
Rep. Hensarling’s Financial CHOICE Act appears to propose an alternative to the broad
power vested by Dodd-Frank to the agencies and also an alternative to compliance to
Basel-based rules governing capital and liquidity. The CHOICE Act proposes an “off-
ramp” option from the Dodd-Frank and Basel III regimes – provided that an institution
maintains a leverage requirement of at least 10%. Given where supplementary leverage
ratios (SLR) are today (and note the SLR calculates a smaller denominator than in the
CHOICE Act’s definition of leverage), this would not appear to be the desired option for
most U.S. GSIBs (see Chart 138).
Chart 138: 1Q18 SLRs vs. 10% potential leverage minimum under the CHOICE Act
12.0 Min. req. under
CHOICE Act, 10%
Supplementary leverage ratio (%)
10.0
7.9%
8.0
6.7% 6.5% 6.3% 6.0% 5.9% 5.7%
6.0
4.0
2.0
0.0
WFC C JPM MS STT BK GS
"Cost" of regulation
We believe that the Collins Amendment has positioned U.S. banks in better shape versus
EU banks for “Basel 4”, due to their existing standardized floor when determining RWA
calculations for regulatory purposes. Under the Collins Amendment, U.S. banks must use
the lower of either the standardized or advanced approaches as their binding constraint.
On the other hand, EU banks do not have that restriction. This has led to arguments
between global regulators over the newly proposed Basel rules around operational risk
calculations which would negatively impact EU BHCs more than US BHCs.
As for the direct cost of higher post-crisis capital and liquidity standards to the
economy and industry return, we provide in more detail beginning on page 134.
Note that proprietary trading itself, on average, represented less than 15% of
aggregated U.S. SIFI revenues during the five years ended 2010 (see Chart 139).
Chart 139: Total “proprietary trading” revenue est. of 5 largest U.S. BHCs as % of total revenues
20%
"Proprietary trading" revenues at the 5
18%
largest U.S. BHCs as % of total revs
16%
14% Median, 13%
12%
10%
8%
6%
4% (negativ e rev enues)
2%
0%
Applicability
The applicability of the Volcker Rule is more sweeping than meets the eye, as alluded to
by the number of agencies involved in enforcing Volcker. In theory, the restriction on
proprietary trading is most applicable to money center banks and broker/dealers. That
said, as written, the rule is applicable to “banking entities” and “nonbank financial
companies” designated by FSOC. In addition, the rule applies to the U.S. operations of
foreign banks. In other words, Volcker is felt more broadly down the size spectrum,
impacting even community banks with <$10bn in assets.
In addition, because of the rule’s broad scope, even small banks that aren’t in the
business of underwriting or market making have had to incur large costs just to prove
that they aren’t engaging in proprietary trading. For instance, community banks must
review their investment portfolios to determine whether they are purchasing or selling
any securities for a “trading account.” Of course, most banks, regardless of size, may
purchase securities to hedge balance sheet positioning (mostly interest rate risk).
The agencies appear to also recognize the broad scope of this rule. On December 18,
2014, the Fed announced that it would give financial institutions an additional one year
(deadline: July 21, 2016) to conform investments in and relationships with covered funds
that were in place prior to 2013. On July 7, 2016, the Fed granted financial institutions
an additional one-year extension of the conformance period (July 21, 2017).
On January 14, 2014, regulators approved an interim final rule to permit financial
institutions to retain interests in certain collateralized debt obligations (CDO) backed
primarily by trust preferred securities (TRuPs) from the investment prohibitions of the
Volcker Rule. Under the interim final rule, the agencies permit the retention of an
interest in or sponsorship of covered funds by banking entities if certain qualifications
are met.
1. Value-at-risk (VaR)
2. Position limits
Entities with $50bn in consolidated trading assets must report the metrics on a monthly
basis. Smaller entities are required to report these metrics on a quarterly basis.
3. Management framework
4. Independent testing
6. Sufficient records
Entities with greater than $50bn in assets, as well as foreign banking entities with total
U.S. assets of $50bn or more, are required to adopt an enhanced version of the basic
six-element compliance program. Regulators require more detailed policies, limits,
governance processes, and independent testing and reporting. Further, the CEOs of
entities with greater than $50bn in assets must attest that the banking entity has a
program in place to achieve Volcker compliance.
Moreover, the test of “ownership interests” was similarly broader than what the
industry expected. Volcker defines ownership interest in a fund as any equity,
partnership, or other similar interests. Potentially ensnared in “other similar interests”
were characteristics of a standard cash flow waterfall of a securitization. As such,
certain debt investments/structured products were included, such as debt tranches of
collateralized debt obligations (CDOs), collateralized mortgage obligations (CMOs), and
collateralized loan obligations (CLOs) may be considered to be ownership interests.
Among the investments restricted are CDOs backed by trust preferred securities,
typically issued by banks and insurance companies. These types of investments were
particularly popular with regional banks prior to the crisis. (Some have been
grandfathered, as discussed on previous page).
“Cost” of regulation
We think there are three major “costs” stemming from the Volcker rule:
It is very difficult to isolate compliance and control costs in general for banks, much less
costs related to Volcker compliance. However, a study by the Federal Reserve Bank of
9.0%
noninterest expenses
8.0%
7.0%
6.0% Highest-rated banks
5.0% Other banks
4.0%
3.0%
2.0%
<$100mn $100 to $250mn $250 to $500mn $500mn to $1bn $1 to $10bn
Bank Asset Size
While again very difficult to quantify in aggregate, we believe Volcker has also likely had
a negative impact on community bank spread income and capital as well. Without the
existence of Volcker, TRUPs CDO investments proved problematic during the financial
crisis, as underlying issuers would miss interest payments to preserve capital. That said,
regional and community banks typically had to sell restricted instruments at a loss,
triggering after-tax charges that impacted capital. Further, the yields on these
instruments were typically attractive, especially in light of a low rate environment, and
therefore impacted spread income negatively when sold.
Please refer to our discussion on page 76 on Volcker’s potential impact on bond market
liquidity.
Prior to the financial crisis, derivatives trading was executed in the over-the-counter
(OTC) market. One of the contributors to the financial crisis was the lack of
transparency and liquidity for certain complex FICC products. To address this issue,
Dodd Frank implemented requirements that certain derivatives must be traded
electronically. In particular, Title VII of Dodd Frank requires certain derivatives to be
cleared by a clearinghouse and executed on an electronic execution facility.
Note: Title VII also applies to security-based swaps regulated by the SEC. However,
these rules are not yet final and therefore not addressed further in this report. In
addition, while Title VII encompasses a much longer list of requirements (e.g. reporting,
business conduct, etc.), we are only focusing on requirements imposed on OTC
derivatives.
Regulation of swaps
Post the financial crisis, regulations were passed around OTC/swaps products, including
margin, clearing, and trading of the instruments. However, as rules have gone into effect
and there has been time to review the pros and cons of the regulations, there is the
potential for some modifications ahead, which we discuss in this section.
Margin requirements for uncleared swaps are part of the Commodity Exchange Act
(CEA) in the Dodd-Frank Act which required regulators to adopt initial margin (IM) and
variation margin (VM) requirements for certain swap dealers (SDs) and major swap
participants (MSPs). The final rules would establish initial and variation margin
requirements for SDs and MSPs but would not require SDs and MSPs to collect margin
from nonfinancial end users.
In December 2015, the CFTC adopted its own version of the uncleared swaps rule which
covers swap entities not supervised by the aforementioned prudential regulators (non-
bank CSEs) if they are registered with the CFTC. The SEC is expected to adopt its own
rule for all non-bank CSEs that trade in security-based swaps which are required to
register with the SEC.
As part of the final rule published on August 2016 (effective date October 1, 2016),
regulators adopted interim final margin rules providing an exemption for uncleared
swaps entered for hedging purposes by qualifying commercial end users, small banks,
treasury affiliates acting as agents, and certain cooperative entities.
Uncleared margin requirements will be phased-in over a 5-year period for initial margin,
while currently all in-scope market participants are expected to post variation margin.
However, U.S. regulators have, either through formal no-action relief or examiner
guidance, allowed non-compliant counterparties to continue trading subject to certain
conditions. These conditions include, but are not limited to, good faith efforts to ensure
full compliance with variation margin requirements, as soon as possible after the
original March 1, 2017 effective date of the variation margin requirements as well as
requiring any in-scope transactions executed after March 1, 2017 to become fully
compliant with margin requirements by Sept. 1, 2017.
Below, we lay out the initial margin (IM) and variation margin (VM) requirements, along
with date of required compliance (see Exhibit 83).
Applicability
Given the complexity of applicability, we thought it would be clearer to explain
applicability in a graphic, from a November 2015 publication from Davis Polk, on the
follow page (see Exhibit 84).
Is Counterparty B:
No SEC or CFTC • a commercial end user; Yes
CSE: covered swap Does Counterparty B have material sw aps
Global Banks and Brokers | 18 May 2018
1. Swap entities;
2. Non-bank SIFIs
4. Government agencies Fannie Mae, Freddie Mac, or the Federal Home Loan
Bank
5. An investment fund
7. Insurance companies
"Cost" of regulation
Some derivatives participants currently already set aside some initial and variation
margin. However, assuming the requirements were fully implemented (no phase-in
period), the CFTC stated that U.S. swap entities would be required to set aside a
significant amount of additional margin to comply, which would lead to lost revenue
assuming firms could earn a return on the IM.
Exhibit 85: U.S. participants would need to set aside additional margin to comply
($ in bil) FRB CFTC
Estimated IM Margin Requirements $315bn $800bn
FRB annual opportunity cost $760-410mn NA
CFTC annual opportunity cost between 25-160bp NA $320mn -2.05bn
Source: BofA Merrill Lynch Global Research, Federal Reserve, CFTC
Swaps clearing
Requirements, interpretation, & enforcement
Title VII of the Dodd-Frank Act also established a new regulatory framework for swaps,
and the requirement that swaps be cleared by derivative clearing organizations (DCO).
The Commodity Exchange Act (CEA), as amended by Title VII, now requires a swap:
1. To be cleared through a DCO if the CFTC has determined that the swap, or
group, category, type, or class of swap, is required to be cleared, unless an
exception to the clearing requirement applies;
Note that Dodd Frank introduced swap execution facilities (SEFs) to increase
transparency in the OTC derivatives markets. SEFs are regulated platforms that execute
swap transactions with thorough bid/offer information (i.e. trade details). Such
platforms must register as a SEF or DCM with the CFTC. SEFs can list products for
trading, following product approval rules governed by the CFTC. SEFs must file a “made
available to trade” determination with the CFTC if a product is listed that is subject to
clearing mandate. Pre-trade transparency that should generate more competition and
pricing improvements is the goal of self-trading, since the market becomes more
transparent.
As mentioned, the creation and enforcement of swaps rules is divided between the SEC
and CFTC. The SEC has authority over “security-based swaps,” and the CFTC has
primary regulatory authority over all other swaps. The two commissions share authority
over “mixed swaps,” which are security-based swaps that also have a swaps component.
In November 2012, the CFTC set the timeline for mandatory clearing for certain interest
rate and credit default swaps, which made up approximately 90% of the swaps market.
Clearing for Category 1 entities began March 11, 2013, Category 2 on June 10, 2013,
and Category 3 on September 9, 2013.
Applicability
The CFTC initially required that swaps in four interest rate swap classes and two CDS
classes be cleared (see Exhibit 86 and Exhibit 87).
Exhibit 86: Initial CFTC interest rate swap classes
Specification Fixed-to-Floating Swap Class
1. Currency U.S. Dollar (USD) Euro (EUR) Sterling (GBP) Yen (JPY)
2. Floating Rate Indexes LIBOR EURIBOR LIBOR LIBOR
3. Stated Termination Date Range 28 days to 50 years 28 days to 50 years 28 days to 50 years 28 days to 30 years
4. Optionality No No No No
5. Dual Currencies No No No No
6. Conditional Notional Amounts No No No No
• Certain overnight index swaps (OIS) denominated in AUD and CAD, as well as U.S.
dollar-, euro-, and sterling-denominated OIS with termination dates up to 3 years
Congress enacted the G-20 swaps execution reforms in the Dodd-Frank Act by requiring
that swaps transactions be traded on regulated platforms called swap execution
facilities. As CFTC Chairman Giancarlo states, it “grafted into its SEF rules a number of
market practices from highly liquid futures markets that are antithetical to episodically
liquid swaps trading. The CFTC limited methods of swaps execution in a misinformed
attempt to re-engineer the market structure of swaps execution.”
In regards to swap dealer capital, the Chairman is looking to introduce a more principles
risk-based capital requirements, taking into account risk mitigating features of
offsetting swap positions between pairs of counterparties and posted margin. Finally
with the end user exemption, the Chairman is looking to extend the same end user
exemption to small financial end users (that don’t carry systemic risk) as it was provided
to commercial end users.
• FSOC has the authority to issue a stay to the CPFB for any rules it deems a threat
to financial stability with a 2/3s vote
The CFPB is headed by a director appointed by the President, for a term of 5 years.
Applicability
The CFPB covers banks and credit unions with more than $10bn in assets in addition to
mortgage lenders, loan modification and foreclosure relief services, private education
lenders, and payday lenders.
"Cost" of regulation
While the creation of the CFPB added to banks’ compliance and risk-related costs, it is
difficult to clearly isolate the incremental cost of the CFPB. However, we note that
according to the CFPB, the agency has transferred $11.8bn to U.S. consumers from
financial institutions since its inception, primarily through principal reductions/debt
cancellation ($7.7bn), followed by monetary compensation ($3.7bn) as well as relief as a
result of supervisory activity ($400mn).
Moreover, the CFPB’s laws and new rules have also led to increased costs. We flesh out
costs of mortgage regulation as outlined by Dodd Frank, in the section discussing Title
XIV beginning on the next page.
The Durbin Amendment is enforced by the Federal Reserve, which has the power to
regulate the amount of interchange fees banks can collect. As the law is specific around
the amount that institutions are allowed to charge, a change in leadership at the Fed
would unlikely be enough for relief. Instead, a full repeal of the Amendment would likely
be needed before banks would be able to realize any relief from interchange restrictions.
Applicability
The Durbin Amendment applies to all financial institutions with over $10bn of assets.
Alternate proposals
A full repeal of the Durbin Amendment would require significant Congressional support,
which we view is unlikely given that the law itself had bipartisan support. For example,
the retail industry has benefitted the most from Durbin and may likely challenge any
attempt to repeal the Amendment.
Unlike the other Dodd-Frank laws, the Durbin amendment specifically lays out the
maximum interchange fee and transaction cost a bank can provide. As such, change
in regulatory agency leadership may be insufficient to provide any change to
interchange restrictions.
"Cost" of regulation
We found that there was a $2bn decline in interchange fees across the industry in 2011-
2012 following the enactment of the Durbin Amendment (see Chart 141), with some
banks noting greater than 40% decline in interchange fees. This translates to 15bp in
returns that could be generated if that revenue were to be returned. The industry has
been able to recover some of the loss fee revenue by increasing rates on other products.
$37
$36
2011 2012 2013 2014 2015 2016 2017
Source: BofA Merrill Lynch Global Research, company data, SNL Financial
Note: population includes U.S. financial institutions with $10bn or more in total assets
• Appraisal activity regulation and oversight: The Federal Reserve Board (FRB),
the Office of the Comptroller of the Currency (OCC), the Federal Deposit Insurance
Corporation (FDIC), the National Credit Union Administration Board (NCUA), the
Federal Housing Finance Agency (FHFA), CFPB
• Subtitle C – High Cost Mortgages. This law defines a “high cost mortgage”,
including but not exclusive to a first mortgage that is 6.5% higher than the average
Prime offer rate, or a second mortgage with an interest rate that is 8.5% higher
than the average Prime offer rate. This subtitle also requires such borrowers to
receive pre-loan counseling, and prohibits balloon payments. Late fees and other
points/fees on such loans are also limited by this subtitle.
Applicability
The QM rules apply to creditors including depository institutions and mortgage banks.
However, certain mortgages do not qualify as QM. The rule is subject to borrowing
limits with loans over a certain level ($424k or $636k in high-cost areas) excluded from
QM status and thus the legal protections. Other types of lending such as interest only
are also excluded from QM status.
"Cost" of regulation
While banks likely beefed up mortgage standards prior to the QM rule going into effect
in 2014, we note large banks have reduced their exposure to the mortgage market in
recent years as QM was discussed and set as law (see Chart 142). The most dramatic
shift has been at WFC, where its share has fallen to 7.6% in 2015 vs. 12.9% in 2012. We
note that had the largest banks kept their market share in 2015 at 2012 levels, the
banks would have accreted between 2bp-22bp in additional ROTCE in 2015, assuming
they sold the loans for a 2% gain on sale (see Chart 143).
Chart 142: Mortgage market share Chart 143: Impact to ROTCE, if applied ’12 mortgage mkt share to ’15
14.0 12.9 0.25%
HMDA Mortgage market share (%)
12.0
0.20%
10.0
7.6
Impact to ROTCE
8.0
0.15%
5.6
6.0
4.5
4.0 0.10%
2.2
1.6 1.8 1.5
2.0 1.0 0.9
0.05%
0.0
WFC JPM C USB PNC
0.00%
2012 2013 2014 2015 WFC USB JPM PNC C
Source: BofA Merrill Lynch Global Research, company data, SNL Financial Source: SNL Financial, Note: Mortgage data from Home Disclosure Act
Note: Mortgage data from Home Disclosure Act; Data for banks based on largest operating subsidiary Note: Impact based on 2% gain on sale of additional loan originations, expense adjusted (75%
efficiency ratio) as well as tax-effected (35%).
While larger banks have reduced their exposure to the market, mortgage banks have
helped to fill the gap. For example, Quicken loans increased its share to 4.1% in 2015
versus 3.2% in 2012. That said, the mortgage market in 2015 was still below 2012
levels at $1.8 trillion vs. $2.1 trillion in 2012.
2.5
Funded mortgage loans ($tr)
QM rule
2.0 issued
QM rule
1.5 effectiv e
1.0
0.5
0.0
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
The DOL has also noted that it believes existing rules have insufficient scope, given
changes in where retirement assets are held (rapid growth in 401ks and IRAs) as well as
how retail investors receive advice. Since about 30% of the retirement industry is DB,
this segment is already governed under the fiduciary standard, as well as many 401(k)
IRAs
33%
Other retirement
assets
40%
Defined contribution
plans
27%
For those advisors who do not currently fall under the DOL fiduciary standard, they are
still subject to regulation and oversight, but the standard they are required to meet is
generally a “suitability” standard. Although the definitions of each are subject to some
interpretation, the primary difference between the two is that the fiduciary standard
requires an advisor to manage plan assets prudently and with undivided loyalty to the
plans and their participants (ERISA section 404a), while an advisor falling under the
suitability standard must give advice that is suitable for the client (though interests do
not necessarily have to be fully aligned).
What’s next?
The Commission will seek public comment on the proposed rules & interpretations for
90 days. Following the 90 days, along with a period to interpret the comments, we
would expect final rules.
BofA Merrill Lynch Research Personnel (including the analyst(s) responsible for this report) receive compensation based upon, among other factors, the overall profitability of Bank of America
Corporation, including profits derived from investment banking. The analyst(s) responsible for this report may also receive compensation based upon, among other factors, the overall
profitability of the Bank’s sales and trading businesses relating to the class of securities or financial instruments for which such analyst is responsible.
Officers of MLPF&S or one or more of its affiliates (other than research analysts) may have a financial interest in securities of the issuer(s) or in related investments.
BofA Merrill Lynch Global Research policies relating to conflicts of interest are described at https://go.bofa.com/coi.
"BofA Merrill Lynch" includes Merrill Lynch, Pierce, Fenner & Smith Incorporated ("MLPF&S") and its affiliates. Investors should contact their BofA Merrill Lynch representative or
Merrill Lynch Global Wealth Management financial advisor if they have questions concerning this report. "BofA Merrill Lynch" and "Merrill Lynch" are each global brands for BofA
Merrill Lynch Global Research.
Information relating to Non-US affiliates of BofA Merrill Lynch and Distribution of Affiliate Research Reports:
MLPF&S distributes, or may in the future distribute, research reports of the following non-US affiliates in the US (short name: legal name, regulator): Merrill Lynch (South Africa): Merrill Lynch
South Africa (Pty) Ltd., regulated by The Financial Service Board; MLI (UK): Merrill Lynch International, regulated by the Financial Conduct Authority (FCA) and the Prudential Regulation Authority
(PRA); Merrill Lynch (Australia): Merrill Lynch Equities (Australia) Limited, regulated by the Australian Securities and Investments Commission; Merrill Lynch (Hong Kong): Merrill Lynch (Asia
Pacific) Limited, regulated by the Hong Kong Securities and Futures Commission (HKSFC); Merrill Lynch (Singapore): Merrill Lynch (Singapore) Pte Ltd, regulated by the Monetary Authority of
Singapore (MAS); Merrill Lynch (Canada): Merrill Lynch Canada Inc, regulated by the Investment Industry Regulatory Organization of Canada; Merrill Lynch (Mexico): Merrill Lynch Mexico, SA de
CV, Casa de Bolsa, regulated by the Comisión Nacional Bancaria y de Valores; Merrill Lynch (Argentina): Merrill Lynch Argentina SA, regulated by Comisión Nacional de Valores; Merrill Lynch
(Japan): Merrill Lynch Japan Securities Co., Ltd., regulated by the Financial Services Agency; Merrill Lynch (Seoul): Merrill Lynch International, LLC Seoul Branch, regulated by the Financial
Supervisory Service; Merrill Lynch (Taiwan): Merrill Lynch Securities (Taiwan) Ltd., regulated by the Securities and Futures Bureau; DSP Merrill Lynch (India): DSP Merrill Lynch Limited, regulated
by the Securities and Exchange Board of India; Merrill Lynch (Indonesia): PT Merrill Lynch Sekuritas Indonesia, regulated by Otoritas Jasa Keuangan (OJK); Merrill Lynch (Israel): Merrill Lynch Israel
Limited, regulated by Israel Securities Authority; Merrill Lynch (Russia): OOO Merrill Lynch Securities, Moscow, regulated by the Central Bank of the Russian Federation; Merrill Lynch (DIFC):
Merrill Lynch International (DIFC Branch), regulated by the Dubai Financial Services Authority (DFSA); Merrill Lynch (Spain): Merrill Lynch Capital Markets Espana, S.A.S.V., regulated by Comisión
Nacional del Mercado De Valores; Merrill Lynch (Brazil): Bank of America Merrill Lynch Banco Multiplo S.A., regulated by Comissão de Valores Mobiliários; Merrill Lynch KSA Company, Merrill
Lynch Kingdom of Saudi Arabia Company, regulated by the Capital Market Authority.
This information: has been approved for publication and is distributed in the United Kingdom (UK) to professional clients and eligible counterparties (as each is defined in the rules of the FCA
and the PRA) by MLI (UK) and Bank of America Merrill Lynch International Limited, which are authorized by the PRA and regulated by the FCA and the PRA, and is distributed in the UK to retail
clients (as defined in the rules of the FCA and the PRA) by Merrill Lynch International Bank Limited, London Branch, which is authorized by the Central Bank of Ireland and subject to limited
regulation by the FCA and PRA - details about the extent of our regulation by the FCA and PRA are available from us on request; has been considered and distributed in Japan by Merrill Lynch
(Japan), a registered securities dealer under the Financial Instruments and Exchange Act in Japan, or its permitted affiliates; is issued and distributed in Hong Kong by Merrill Lynch (Hong Kong)
which is regulated by HKSFC is issued and distributed in Taiwan by Merrill Lynch (Taiwan); is issued and distributed in India by DSP Merrill Lynch (India); and is issued and distributed in
Singapore to institutional investors and/or accredited investors (each as defined under the Financial Advisers Regulations) by Merrill Lynch International Bank Limited (Merchant Bank)
(MLIBLMB) and Merrill Lynch (Singapore) (Company Registration Nos F 06872E and 198602883D respectively). MLIBLMB and Merrill Lynch (Singapore) are regulated by MAS. Bank of America
N.A., Australian Branch (ARBN 064 874 531), AFS License 412901 (BANA Australia) and Merrill Lynch Equities (Australia) Limited (ABN 65 006 276 795), AFS License 235132 (MLEA) distribute
this information in Australia only to 'Wholesale' clients as defined by s.761G of the Corporations Act 2001. With the exception of BANA Australia, neither MLEA nor any of its affiliates involved in
preparing this information is an Authorised Deposit-Taking Institution under the Banking Act 1959 nor regulated by the Australian Prudential Regulation Authority. No approval is required for
publication or distribution of this information in Brazil and its local distribution is by Merrill Lynch (Brazil) in accordance with applicable regulations. Merrill Lynch (DIFC) is authorized and
regulated by the DFSA. Information prepared and issued by Merrill Lynch (DIFC) is done so in accordance with the requirements of the DFSA conduct of business rules. Bank of America Merrill
Lynch International Limited, Frankfurt Branch (BAMLI Frankfurt) distributes this information in Germany and is regulated by BaFin.
This information has been prepared and issued by MLPF&S and/or one or more of its non-US affiliates. The author(s) of this information may not be licensed to carry on regulated activities in
your jurisdiction and, if not licensed, do not hold themselves out as being able to do so. MLPF&S is the distributor of this information in the US and accepts full responsibility for information
distributed to MLPF&S clients in the US by its non-US affiliates. Any US person receiving this information and wishing to effect any transaction in any security discussed herein should do so
through MLPF&S and not such foreign affiliates. Hong Kong recipients of this information should contact Merrill Lynch (Asia Pacific) Limited in respect of any matters relating to dealing in
securities or provision of specific advice on securities or any other matters arising from, or in connection with, this information. Singapore recipients of this information should contact Merrill
Lynch International Bank Limited (Merchant Bank) and/or Merrill Lynch (Singapore) Pte Ltd in respect of any matters arising from, or in connection with, this information.
General Investment Related Disclosures:
Taiwan Readers: Neither the information nor any opinion expressed herein constitutes an offer or a solicitation of an offer to transact in any securities or other financial instrument. No part of
this report may be used or reproduced or quoted in any manner whatsoever in Taiwan by the press or any other person without the express written consent of BofA Merrill Lynch.
This document provides general information only, and has been prepared for, and is intended for general distribution to, BofA Merrill Lynch clients. Neither the information nor any opinion
expressed constitutes an offer or an invitation to make an offer, to buy or sell any securities or other financial instrument or any derivative related to such securities or instruments (e.g.,
options, futures, warrants, and contracts for differences). This document is not intended to provide personal investment advice and it does not take into account the specific investment
objectives, financial situation and the particular needs of, and is not directed to, any specific person(s). This document and its content do not constitute, and should not be considered to
constitute, investment advice for purposes of ERISA, the US tax code, the Investment Advisers Act or otherwise. Investors should seek financial advice regarding the appropriateness of
investing in financial instruments and implementing investment strategies discussed or recommended in this document and should understand that statements regarding future prospects
may not be realized. Any decision to purchase or subscribe for securities in any offering must be based solely on existing public information on such security or the information in the
prospectus or other offering document issued in connection with such offering, and not on this document.
Securities and other financial instruments referred to herein, or recommended, offered or sold by BofA Merrill Lynch, are not insured by the Federal Deposit Insurance Corporation and are not
deposits or other obligations of any insured depository institution (including, Bank of America, N.A.). Investments in general and, derivatives, in particular, involve numerous risks, including,
among others, market risk, counterparty default risk and liquidity risk. No security, financial instrument or derivative is suitable for all investors. In some cases, securities and other financial
instruments may be difficult to value or sell and reliable information about the value or risks related to the security or financial instrument may be difficult to obtain. Investors should note that