Bank Stress Testing and

Comprehensive Capital Assessment and Review

(CCAR)
Peter Cai
VaR -- Important But Often Overlooked Basics
Important VaR Parameters

Design and implementation of a risk model invariably need to consider the following important parameters.
• Simulation methods
− Historical simulations (empirical distributions) are mostly commonly used
− Parametric simulations are used to give modeling flexibility for:
• Normal or fat-tailed distributions
• Different correlation regimes
• Look-back window
− Past performance is no guarantee of future returns, but past volatility and correlation patterns give strong
indication of current volatility
− Time weighting techniques are used to be more reflective of recent markets, but need to be balanced
against too much pro-cyclicality (e.g. if current volatility is low, we may potentially understate risks)
• Risk horizon
− This is the most overlooked, yet the most important parameter
− Common choices include 1 day and 10 days. Capital planning-related stress testing or economic capital
(EC) modeling often use 1-year horizons
− What does one mean by 1-year VaR: confusion between risk horizon and lookback window

3
Historical Simulation VaR

Historical simulation VaR is the most widely across the financial services industry. It is often criticized for the
following deficiencies:
• Limited look-back window to derive a rich empirical distribution
− Lack of historical extreme events
− Lack of correlation breakdown under extreme events
− Losses beyond confidence interval bounds (tail risk) not considered
− Unconditionality (most use equal-weighting of observations) on current markets, which can be partially
addressed by using time-weighting
• Short risk horizon (1 day to 10 days) seen in most VaR models limits the usefulness to only liquid risks
− Generally VaR is more useful for portfolios of more liquid macro markets (such as general interest rates
and FX) and plain-vanilla flow products
− For more illiquid risk factors (e.g., securitized credit and OTC exotics) longer risk horizons should be
considered; however, availability of long historical data series may be in question

Different VaR methods lead to very different results for the same risk portfolio. Lack of standardization makes
comparison across products, groups and companies difficult

4
VaR Disclosure from Major Banks

Major banks disclose 1-day VaR, with variations in modeling and scope. Broader derisking trends are very
clear, indicated by declining levels of VaR over the years.

160
$MM Average 95% VaR
140
120
100 2011
2012
80
2013
60 2014
40 2015
20
0
MS GS JPM BAC Citi DB UBS CS
Current Vol Exponentially Historical Historical Monte Carlo Monte Carlo Historical Exponentially
adjusted weighted (1 Year) (3 Year) (> of 3yr or (5 Year) weighted
1m vol)
Notes:
Reported VaR adjusted to 95%: BAC, Citi, DB report 99%. CS reports 98%. GS, JPM , BAC, UBS and CS exclude CVA from disclosed VaR. DB is unclear about treatment of CVA in disclosed VaR.
MS and Citi disclose VAR both including and excluding CVA. All values are shown ex-CVA where available. All VaR Values are in USD.
5
Stress Testing for Bank Capital Planning
Stress Testing Basics
Portfolio/ Business/ Bank Capital Central Bank
Book Bank Planning Stress Tests

“How bad would the P/L be for the exotic equity derivatives book if equity markets were to sell
off by 30%?”
Where is this risk managed: first and second lines of defense (i.e. revenue generating trading
desk or investment strategy; and risk management group covering that unit)

• To address the shortcomings of “normal” markets and lack of extreme tail events associated
with VaR, one can stress a portfolio or book using hypothetical scenarios
− Such a book/portfolio level stress test generally represents limited scope
− It generally addresses limited number of risk factors, typically within one asset class (e.g.,
sharp drop in equity market prices and attendant volatility spike)
• Greek approximation (delta, gamma, vega) commonly used in VaR may not be adequate to
capture the nonlinear nature of derivatives, especially for out-of-the money options

7
Stress Testing Basics
Portfolio/ Business/ Bank Capital Central Bank
Book Bank Planning Stress Tests
“How much would the entire bank lose if the 2008 crisis were to repeat?”
Where is this risk managed: CFO/CRO
− Bank-level stress tests cover all assets (trading positions, loan portfolios and investments)
− Economic scenarios covering a comprehensive set of risk factors are commonly used, spanning all
asset classes
• Historical stress events, or tweaks based on historical stress events, are most commonly used
• Hypothetical scenarios conditional on a small set of factor shocks – requiring the modeling of
correlations – are also used
• Given the disparate liquidity profiles associated with banks’ diverse assets, liquidity-adjusted horizons
(e.g., 1 week for liquid exposures and up to 1 year for illiquid exposures) are important
• Risk factor specification
− Strike a balance between too many (on the order 100,000 for a typical large bank) and too few
− Thoughtful consideration for functional form (normal or log normal) and price representation
(price, yield, or spread) is essential
− Examples: S&P prices are modeled log normal; (low) interest rate yields log normal; and credit
8
spreads often log normal (especially for wider credit spreads)
Stress Testing – Bank Capital Planning
Portfolio/ Business/ Bank Capital Central Bank
Book Bank Planning Stress Tests

“Would the bank survive if something like the 2008 crisis were to occur?”
Where is this risk managed: CEO/CFO/CRO
● Firmwide capital planning (known as ICAAP, or Internal Capital Adequacy Assessment and
Planning), as a Basel 2 pillar, has become a focus for large banks over the past decade
− Using internal models and processes, the bank asks the question: Would the bank have
enough capital and other resources to weather a multiyear storm of historic severity?
• ICAAP involves full modeling of bank’s balance sheet over a multiyear period (e.g., risk
horizon = 2-3 years)
− In additional to stress losses in asset value (capital depletion), banks need to consider
reduced ability to generate earnings (insufficient capital replenishment to offset the
depletion)
− Consideration of macroeconomic repercussions that would impact institutions outside the
bank itself

9
Bank Capital Ratios

Two Key Ratios On Which All Major Banks Are Measured

Common Equity Tier 1 Ratio [Risk-based] Why it’s important:

CET1 includes common equity
Common Equity Tier 1 only, making it more stringent
Capital (CET1) than Tier 1 Capital
Risk Weighted Assets
(RWA)

Tier 1 Leverage Ratio [Leverage-based] Why it’s important:

Total assets are more objectively
calculated than risk weighted
Tier 1 Capital assets, making it more
(T1) standardized and transparent
Total Assets

10
 Bank Capital Ratio Constraints

Four external constraints (Basel and CCAR), and two internal ones (ICAAP)

Pre-stress loss Post-stress loss

Risk-Based CET1/RWA ≥ (CET1 – EC)/RWA ≥ (CET1 – CCAR)/RWA ≥

10.5% 6% 4.5%

T1/Assets ≥ (T1 –EC)/A ≥ (T1 – CCAR)/A ≥

Leverage-Based
4% 5% 4%

Basel Regulatory ICAAP CCAR Stress

Minima (Internal Capital; EC)

RWA = Risk Weighted Assets; Required Economic Capital is often implemented as a form of bank-wide stress test
Regulatory minimum figures are from JPM’s recent disclosure
11
Leverage-Based Capital
● Leverage Ratio is defined as (Tier 1 Capital) / (Balance Sheet Assets)
● Balance Sheet Assets refer to all assets carried at current value, with no consideration for riskiness. For example, a
newly executed credit default swap may be carried at its market value – close to zero
● Does not include risk factors, but is more rule-based and transparent
● However, Basel and the Federal Reserve Board (FRB) are moving toward a regime known as Supplemental
Leverage Ratio (SLR), where the denominator will take into account off-balance-sheet exposure such as OTC
derivatives
− For example, a newly executed credit default swap will be carried at its market value (close to zero), plus a
measure to account for potential future exposure expressed as a percentage of notional
• Furthermore, a final rule known as Enhanced Supplemental Leverage Ratio (eSLR) that strengthens the agencies’
SLR standards, requiring a 2% buffer above the minimum SLR requirement of 3% for applicable banks.

12
Risk-Based Capital

● RWAs represent risk-adjusted asset value

− Risk model based, therefore not standardized
− Traditionally, RWA is calibrated to an asset that attracts an 8% risk haircut:
$1MM of RWA = $1MM of risky assets with an 8% haircut by Basel rules
− In Basel I dating back to 1980s, BB Corporate debt was charged an 8% haircut. So $1MM
worth of cash BB corporate bonds had an RWA of $1MM, and capital usage of $125K
− Current Basel RWA methodology considers market risk inherent in derivatives. For
example, a sold protection on $1MM notional of the same BB corporate debt attracts an
approximate RWA of $1MM as well, despite the small to near-zero market value
− If an AA corporate bond has half of the volatility of BB (resulting in a haircut of ~4%),
then $1MM worth of AA corporate bond has an RWA of ~$500K
− Basel has progressively moved towards more model-based RWA calculations, so the above
rule-based haircut approach is no longer used for the vast majority of market and credit
risks, but the risk weighting concepts remain

13
 RWA Methodology Evolution (Basel)

Basel evolved from simple rules to complex models over 1990s and 2000s, but the tide is now toward
reducing complexity

Past Current Future…

(MS & GS) (2016-…)
Regulated by Regime shift to
SEC tighter regulatory Rapid Changes and Regulatory Uncertainty Fundamental Review of
(Consolidated environment Trading Book Changes Basel
Supervised under Basel
(Effective 2019) 4?
Entities - CSE) the Federal VaR 2 5/3 Stressed VaR 1. Revised trading and
Reserve Board banking book boundary
VaR (FRB) and market risk models
Basel
2 Basel Incremental Comprehensive & approaches
Market Risk VaR 1 Risk Charge Risk Measure 2. Expected shortfall
(Trading Book) (reapproved) (IRC) (CRM) 3. Market and
Standardized
instrument liquidity risk
Charges
Securitization Standardized 4. Suitableness of the
Standardized Charges internal model of the
Charges
Charges trading desk
Credit Risk Basel
(Banking Book) Incremental 1
Default Risk Standardized Advanced IRB Constraints for some
Charges Advanced IRB Basel internal ratings-based
Charge (IDRC) 2
(enhanced)
Basel Basel (IRB) models
2 3
Internal Model IMM
IMM
Method (IMM) (enhanced)

Advanced Basel
Internal Ratings
2 Securitization CVA Capital
Based (IRB) Framework Charge

Operational Basic Indicator Standardized

Basel
Risk Approach (BIA) 2/3 Advanced Measurement Measurement
Approach (AMA) Approach (SMA)

14
Basel 3 RWA Disclosure
• RWAs are often not comparable between US and European banks due to implementation differences in
national capital frameworks despite the uniform “Basel 3” label.
• Banks have been aggressively managing downward the usage of RWA over the last 5 years. Levels for years
before 2014 would have been significantly higher.

1800 2014, 2015, & 1H 2016 Total RWA ($Bn)

1600
1400
1200
1000
800
600
400
200
0
14 15 1H16 14 15 1H16 14 15 1H16 14 15 1H16 14 15 1H16 14 15 1H16 14 15 1H16 14 15 1H16
MS GS JPM BAC Citi DB UBS CS
Credit Risk RWA Market Risk RWA Op Risk RWA

Credit Suisse Operational Risk Bond

• On May 26, 2016, Credit Suisse issued Sfr 220mm ($224.7mm) of notes to help insure against certain operational risks, such as computer system
1
disruptions, cybercrime, and regulatory compliance failures. The initial plan is to sell Sfr 630mm but had to be downsized due to lack of market interest.
• The major risks within the bond have a high probability to be correlated with Credit Suisse’s stock
• Credit Suisse remains liable for the first Sfr 3.5bn of op risk losses it suffers in any year
• New Basel SMA is likely to replace the AMA and all other existing approaches to op risk capital, which does not give benefits for capital relief on banks
buying insurance for risk mitigation

15
Introduction to Central Bank Stress Tests
Portfolio/ Business/ Bank Capital Central Bank
Book Bank Planning Stress Tests

“Would the US financial system collapse under a shock of historic severity?”

• Primary objective of central bank stress tests
− Macro-prudential: To assess the financial system’s ability to continue providing financial
services, without government assistance, following a specified shock
− Micro-prudential: To determine whether a major bank can continue providing financial
services, without government assistance, following a specified shock
• In early 2009, in the depth of the crisis, the FRB ordered 19 large financial institutions to
conduct a first such system-wide stress test known as SCAP (Supervisory Capital Assessment
Program)
− It was designed to be “deliberately stringent,” with the outcome that an additional $75Bn
capital were needed for these banks, provided under the TARP program
− SCAP was viewed as a success, restoring the confidence in the US financial system
• European banks are still recovering from lack of decisive action early on

• Subsequently, the stress test has been renamed Comprehensive Capital Assessment and Review
(CCAR), and administered annually to a broader set of banks 16
CCAR As A Critical Regulatory Instrument
Post-crisis, CCAR has become a critical regulatory tool for the Federal Reserve. Banks devote large
amounts of resources (in the thousands for large banks). Recently foreign banks that have large
operations in the US and some large insurers are being folded into the same regulation.

2017 vs. 2016 severely adverse scenario: more severe U.S. economy downturn, unemployment rate at
10%, larger decline in commercial real estate prices, no negative short-term U.S. Treasury rates
CCAR 2011 CCAR 2013 CCAR 2017
CCAR 2015
• Option to resubmit capital • Every bank
• Modeling differences
• CCAR as a plan passed for
between FRB and banks
supervisory tool for • Two supervisory stress first time
are narrowing
approval/rejection of scenarios since
• Further emphasis on
bank capital actions • Emphasis on potential financial
process and governance
Bear Lehman qualitative objection crisis

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

SCAP 2009 CCAR 2012 CCAR 2014 CCAR 2016

• Scenario to consider • Considers negative
• First reference to • CCAR results largest counterparty US interest rates
“bank stress tests” publicly disclosed default • More severe downturn
following crisis • Emphasis on banks’ in US and a global
own scenario design recession
• CFO attestation

17
 CCAR As A Critical Regulatory Instrument
CCAR is at least as critical a regulatory tool in the US as long-standing Basel regulations.
“CCAR and Dodd-Frank Act stress tests have shown the significant supervisory value of conducting coordinated
cross-firm analysis of the major risks facing large banks.”
"Over the six years in which CCAR has been in place, the participating firms have strengthened their capital
positions and improved their risk-management capacities.”
– Daniel K. Tarullo, Member of the Board of Governors of the Federal Reserve Bank

Macro- and micro-prudential stress testing is being adopted internationally following the CCAR success.
The ECB’s Asset Quality Review (AQR) and Stress Test is now a key building block of the Eurozone
Banking Union. The biennial examination covers over a 100 banks.
In 2017, stress testing and resolution expectations will continue to ease for smaller banks and stop
rising for the largest ones. Overall move toward greater transparency.

(CET1-CCAR)/RWA ≥
Risk-Based CET1/RWA ≥ (CET1-EC)/RWA ≥ 4.5%
Capital 10.5% 6%

T1/Assets ≥ (T1 – EC)/Assets ≥

Leverage-Based
Capital
4% 5% (T1 – CCAR)/Assets ≥
4%
Basel Regulatory Basel ICAAP (Internal
Minima Capital)

CCAR Stress

18
CCAR Components
CCAR evaluates income reduction and asset losses under multi-year stress scenarios, and helps the
FRB determine whether and to what extent the banks can pay dividends or buy back stocks

Net Income
Stress Loss
Stressed Net Income Credit Lending
Regulators
Stressed Trading, Counterparty,
Market Approve?
and Investments
Net Income
Available Operational Litigation and
Capital Operational Stress
Loss Proposed Capital
Beginning of Available Actions
Capital Plan Capital
Period Available Capital
Available Pre-Capital
End of Capital Plan
Action Period
Capital
Beginning of
Capital Plan
Period Is this enough
capital? If so, how Is this enough
much can be capital?
dividended out?

19
2016 CCAR and Capital Plans
CCAR stress results are disclosed by the banks and the FRB. The banks’ own calculations often
differ from the FRB’s, creating discussions on quality of the banks’ submissions and lack of
transparency around FRB’s own internal processes
FRB Calculation for JPM 63
-83 -13

163 -78 -13

130
163
130

FRB Calculation –
All 33 CCAR Banks
399 -617
-617
-~ -156
(more than entire
1,180 1180 pre-crisis capital (payout ratio is 1/3 of
base) net revenue) ~806
806

Beginning Available Capital Net Stress Capital End Available

(CET1) Revenue Loss Actions Capital
(3x compared w/ pre-crisis) (2-3x compared w/ pre-crisis)
1. Sources: Beginning available capital, net revenue and stress loss are from FRB published results and JPM DFAST disclosure.
2. Capital Actions include dividends on common shares and buybacks. Debt and share issuances are excluded. 20
CCAR Outcomes for Major US Banks

There are 4 tiers of CCAR outcomes for each bank’s capital plan. No major bank in the US has had a
perfect track record.
Non-Objection with Conditional Non-
Non-Objection Objection
Adj. Capital Actions Objection

better worse

2012 2013 2014 2015 2016 2017

Goldman
Sachs
Bank of
America
Citi

JP Morgan

Morgan
Stanley

(Peter’s tenure in MS. ) 21

 CCAR Scenario Components

CCAR Stress scenario has two main components: an instantaneous Global Market
Shock (GMS) applied to trading inventory, counterparty exposure (with the largest
counterparty defaulting) and investments, followed by a 9-quarter evolution of a very
severe recession that hits the loan book and revenues/expenses.

Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9

Macroeconomic Scenario
Exposure Stress Methodology
(credit risk/real economy):
Evolves through the 9 quarters of
Trading and
the capital planning horizon and is Global Market Shock
applied to revenue/expenses, and Investments
lending exposures
Global Market Shock +
Counterparty
Default of Large
Default
Counterparty
Macroeconomic
Lending
Scenario

Global Market Shock

(financial market risk):
Is applied to trading, investment
and counterparty positions

As of date for Global Market Shock is

kept secret until the instructions are
released
22
CCAR 2016 Scenario Assumptions

Key scenario assumptions and severity of losses designed to be at least as severe as 2008

• Macroeconomic scenario is similar to 2008 in severity

• Global market shock (loosely modeled after second half of 2008) does not allow for risk mitigating
management actions
• Instantaneous default of a large financial institution counterparty & absence of risk mitigating
actions assumed

US Real GDP level Unemployment Rate (%) House Price Index Level
100 15 190
100 183
10
90 2008 Actuals 90
10 170
10 10
89
2016 FRB 84 7
80 Severely Adverse 5 150
150 139
146
77 5
70 0 130
Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9

Dow Jones Total Stock Market Index Level Global BBB Bond Spread Market Volatility Index Level

30,000 Global 800 +511 bps shock 100

Market
600
20,000 Shock 580
11,183 12033 400 310 410 50 32.20
11826 16,180
10,000 31.10
24.40 22.80
-30% 200 240
190
shock 27.30
0 0 0
Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q0 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9

Macroeconomic Scenario 2008 Actuals Global Market Shock

23
CCAR 2016 Global Market Shocks
Equities and Traded Credit

Bond and CDS spreads widen significantly

2016 2015 2016 2015

Severely Severely Severely Severely
Adverse Adverse Adverse Adverse
Traded Equities Agencies & Munis
S&P -29% -29% US Residential Agency Pass-Throughs 170 bps 150 bps
Euro Stoxx -27% -27% US Commerical Agency 200 bps 190 bps
Hang Seng -35% -35% Muni Bonds (BBB) 419 bps 372 bps
Portugal -29% -29% Credit Spread Levels
Ireland -40% -40% CDX (IG) 162% 157%
Italy -33% -33% ITRAXX (IG) 155% 194%
Spain -24% -24% Bonds (BBB) 229% 413%
Japan -34% -34% Bonds (BB) 223% 318%
Greece -48% -48% Bonds EM (BBB) 199% 422%
MSCI EM Eastern Europe -68% -68% Single Name CDS (BBB) 187% 202%
Private Equity Loans (BBB) -23% -23%
Funded PE: Energy (US) -43% -43% Loans (BB) -34% -27%
Funded PE: Banks (US) -62% -19% Securitized Products
Funded PE: Real Estate (US) -63% -40% Non-Agency RMBS (BB) -92% -63%
Dividend Expectations - 3Y Cash Non-Agency CMBS (BB) -86% -65%
US -30% -30% CLO (BB) -87% -87%
Europe ( and others) -50% -50% ABS Auto (Non-IG) -19% -28%
Japan -57% -57% ABS Other (BB) -47% -55%
Large shocks on Equity Indices and all *See slide 7 on risk
types of Private Equity specification and Securitized products shocks reflect an extremely
functional form* severe drop in housing prices 24
CCAR 2016 Global Market Shocks
FX, Interest Rates and Commodities

FX Markets receive severe 2016 2015 2016 2015

shocks across many A deflationary period in the US
is implied by these shocks Severely Severely
exchange rates
Severely Adverse Severely Adverse Adverse Adverse
FX Interest Rates
US Pairs EUR 2y swaps + 28 bps -8 bps
EUR -11% -11% EUR 10y swaps - 87 bps - 34 bps
JPY 17% 17% USD 2y swaps - 70 bps - 47 bps
AUD -27% -27% USD 10y swaps - 168 bps - 116 bps
Non-USD Pairs USD 2y Govt - 85 bps - 42 bps
AUD/ JPY -37% -37% USD 10y Govt - 164 bps - 109 bps
JPY/ MXN 55% 46% GBP 2y Govt + 16 bps - 54 bps
EUR/ GBP 21% -8% GBP 10y Govt - 95 bps - 97 bps
FX Vol - 3 month USD Inflation 2y - 5 bps - 42 bps
USD/ MXN + 25 pts + 25 pts USD Inflation 10 - 113 bps - 109 bps
USD/ CNY + 4 pts + 4 pts USD Vol 10y 50% -4%
JPY/ AUD + 17 pts + 17 pts GBP Vol 10y 54% -1%
EUR/ USD + 13 pts + 13 pts EUR Vol 10y 44% 2%
Commodities Sovereign Credit
WTI (Spot) -68% -68% France Sovereign Credit + 177 bps + 139 bps
Heating Oil (Spot) -56% -56% Germany Sovereign Credit + 84 bps + 61 bps
Natural Gas (Spot) -57% -57% UK Sovereign Credit + 86 bps + 36 bps
Gold (Spot) -5% -5% Brazil Sovereign Credit + 494 bps + 191 bps
Silver (Spot) -35% -35% Greece 1y Government Bond + 1538 bps + 1263 bps
WTI Vol 1 88% 178% Italy 2y Government Bond + 26 bps + 345 bps
1
Natural Gas Vol 21% 75% Portugal 2y Government Bond + 24 bps + 619 bps
1
Power (North East) Vol 67% 39% Spain 2y Government Bond + 10 bps + 192 bps
The euro zone countries receive substantial shocks
Turmoil in commodities markets 25
 2016 CCAR Results
Historical Trends Published by the FRB

Stress Losses – Trading, Investments, Counterparty & Other Losses ($Bn)

Trading, Investments &
Counterparty Losses Other losses (including Fair Value loans)

Morgan Stanley Goldman Sachs Bank of America 2 Citigroup JP Morgan Chase

2014 2015 2016

Stress Losses – Loan Loss Provisions ($Bn)

AFS/HTM1

Loan Loss Provisions

Morgan Stanley Goldman Sachs Bank of America Citigroup JP Morgan Chase

2014 2015 2016
1. AFS – Available for Sale, HTM – Held to Maturity
26
2016 CCAR Results – Post Planned Capital Actions
Minimum Stressed Ratios Over Planning Horizon – Published by the FRB

Tier 1 Common Ratio (%)

Thanks to the CCAR
mechanism, FRB is able to Regulatory Minimum (4.5%)

keep more and more capital

in the US banks. Key
capital ratios have been
improving over the years,
generally in excess of Basel
requirements. US banks are
perceived to be healthier MS
2014 2015 2016
GS BAC Citi JPM

than the European *Basel I + 2.5 based capital approach RWA

counterparts.
Tier 1 Leverage Ratio (%)

Regulatory Minimum for

2016 CCAR (4%)*

MS GS BAC Citi JPM

2014 2015 2016 27
Where Is CCAR Going?

CCAR is still young and will continue to evolve; there is more uncertainty given the
political backdrop and the expected turnover in FRB chairmanship and membership
Hal Scott and John Gulliver have a piece in the WSJ reiterating why the Federal
Reserve's stress tests could be illegal. Scott and Gulliver are with the Committee on
Capital Markets Regulation — a group comprising senior executives from 34
financial institutions and trade groups. They say the Fed is likely to increase the
tests' required regulatory capital by an average of 57% and note that stress testing
costs each bank $150 million to $250 million.

Daniel Tarullo, the Central Bank’s top financial regulator, resigned in April 2017,
after President Donald Trump took action to start easing regulations on the financial
industry (reviewing Dodd-Frank, and delaying a rule requiring financial advisors to
provide advice in client’s best interest). In the meantime, major revisions are
unlikely until new Fed governors are confirmed.
Homework
Homework

30
Homework

31
Homework

32
Homework

33
Acknowledgements

• The original content of this presentation borrowed heavily from “CCAR 101,” prepared by Kalpana
Telikepali and Alyssa Barzideh of Morgan Stanley in 2014. I also benefited greatly from discussions
with many Morgan Stanley colleagues. Mary Wu of Global Atlantic Financial Group and David
Zhang of University of Michigan provided updates in 2016.
• Dr. Shreve reworked the homework assignment into its current form, adding helpful guidance and
removing ambiguity.
• Anne Chen, a Columbia senior majoring in Economic-Statistics, helped with revising and updating
content in 2017.

34