Contents
Dodd Frank Act vs CCAR
Dodd-Frank Act-
bank holding companies with over
$10 billion in consolidated assets to evaluate and report their capital position under baseline, adverse,
and severely adverse scenarios created by the regulators on an annual basis. Forecasting
models to assess their effects
on the firm’s revenues, losses, balance sheet (including risk-weighted
assets), liquidity, and capital position for each of the scenarios
over a nine quarter horizon.
|
CCAR-
large Bank must submit their data to the regulators who
will then execute the supervisory test, the results of which are published annually. CCAR
test requires the banks must
submit their proposed capital action
plans (including changes to dividends, stock buybacks, etc.) to the
regulators for review.
|
1-
Common stock dividend payments are
assumed to continue at the same dollar amount as the average of the
prior-four quarters.
2-
Scheduled dividend, interest or principal
payments for other capital instruments are assumed to be paid.
3-
Repurchases of common stock and
redemptions of other capital instruments are assumed to be zero.
4-
Issuance of new common stock, preferred
stock, or other capital instruments are assumed to be zero, except for common
stock issuance associated with employee compensation
|
Federal
Reserve uses a BHC’s planned capital
actions under its BHC baseline scenario, including both proposed capital issuances and proposed capital distributions,
and assesses whether the BHC would be capable of meeting minimum regulatory
capital ratios even if stressful conditions emerged and the BHC did not
reduce its planned capital distributions.
Includes capital actions under the
baseline scenario, incorporating related business plan changes.
|
Example- if a BHC includes a
dividend cut, or a net issuance of common equity in its planned capital
actions,
BHC’s post stress capital ratios
projected for the CCAR capital analysis could be higher than those projected
for the Dodd-Frank Act supervisory stress tests.
|
|
1.1 10 Key Points of CCAR 2016-
1-
Integration of CCAR assessments into
year-round supervision begins.
2-
Baseline earnings projections must be
reasonable- Fed to
carefully assess the quality of baseline projections and to closely monitor
actual results for deviations from projections. This additional scrutiny is
needed not just to assess forecasting effectiveness but also to assess the
reasonableness of implied dividend payout ratios, and to test for changes that
might prompt required revisions of the capital plan.
3-
The Fed expects consistency with
respect to capital distributions- capital actions to be consistent with the firms’ historical
experience and capital policies. This will discourage BHCs from gaining capital
flexibility, for example, by projecting significant drops in planned capital
distributions in the “out- quarters” of the planning horizon. BHCs are unlikely
to be allowed to address the breach of a minimum capital ratio in one quarter
by lowering capital distributions in that quarter only. Rather, the Fed will
expect to see consistent capital distribution adjustments across all quarter.
4-
Global market shock’s January 4th as-of date provides a rare
opportunity for the six largest BHCs- Stress losses on trading book
positions are double-counted because the underlying positions are stressed
twice, once as part of PPNR stress and once under the global market shock
(GMS). The Fed allows BHCs to use the higher of the two losses, rather than
double counting them, but only under the condition that a BHC can demonstrate
that the stressed position is “identical” under PPNR and GMS.
5-
Negative Treasury yields in the severely adverse scenario may lead to
projected negative deposit rates.
6-
The severely adverse scenario’s unemployment reaches 10% again, but at
a faster pace than last year.
7-
Deep recession and deflation should lead to higher default rates- Includes a deeper real GDP
contraction in the US, and a slower recovery in both the US and in developing
Asia.
8- Real Estate
stress scenario similar to 20155 but prices need to stress on local and
regional stress prices.
9-
Model sensitivity to inflation and interest rates.
1.2 Reports
§ The FR Y-14A
schedules form a semi-annual collection that is primarily used to collect company-run
stress test results.
-
The collection comprises six schedules: Summary, Scenario,
Counterparty, Basel III / Dodd-Frank, Regulatory Capital Instruments, and
Operational Risk.
§ The FR Y-14Q schedules are a quarterly
collection that covers loan and portfolio-level information, operational
revenues
§ The FR Y-14A
schedules form a semi-annual collection that is primarily used to collect company-run
stress test results.
-
The collection comprises six schedules: Summary, Scenario,
Counterparty, Basel III / Dodd-Frank, Regulatory Capital Instruments, and
Operational Risk.
§ The FR Y-14Q schedules are a quarterly
collection that covers loan and portfolio-level information, operational
revenues and expenses, regulatory capital, and reconciliation data.
-
The portfolio-level information covers Wholesale Loan, Retail Loan, Fair Value Option (FVO)/Held for Sale
(HFS) Loan, Securities, Trading, and Mortgage Servicing Rights portfolios.
-
The operational revenues and expenses are collected through the
PPNR and Operational Risk schedules.
§ The FR Y-14M schedules are a monthly collection that covers loan level information of the first and
second lien and credit card portfolios.
Scenarios-
The scenarios start
in the first quarter of 2016 and extend through the first quarter of 2019. Each scenario includes 28 variables.
The variables describing economic development within the United States
include:
Variables in the Scenarios
|
|
Six measures of economic activity and prices (Annual Rate)
|
i.
- Real and Nominal gross
domestic product (GDP);
ii.
- Real and Nominal disposable personal
income;
iii.
- Consumer price index (CPI);
iv.
- Unemployment rate of
the civilian non-institutional population
aged 16 years and over; percent changes (at an annual rate)
|
Four aggregate
measures of asset prices
or financial conditions
|
-
House prices index
- Commercial property prices,
- Equity prices, and
- U.S. stock market volatility;
|
Six measures
of interest rates
|
- The Rate
on the 3-month Treasury bill;
- The yield
on the 5-year Treasury bond;
- The yield
on the 10-year Treasury bond;
- The yield
on a 10-year BBB corporate security;
- The
interest rate associated with a conforming, conventional, fixed-rate 30-year
mortgage; and
- The prime
rate.
|
Variables describing international
economic conditions
|
|
Three variables for
each Country
|
- The
percent change (at an annual rate) in real GDP,
- The
percent change (at an annual rate) in the CPI or local equivalent, and
- The level
of the U.S. dollar/foreign currency exchange rate
|
Four countries or country blocks
|
-
- The euro
area (the 19 European Union member states that have adopted the euro as their
common currency)
-
- The United
Kingdom (GBP)
-
- Developing
Asia (the nominal GDP-weighted aggregate of China, India, South Korea, Hong
Kong Special Administrative Region, and Taiwan)
-
- Japan
|
2.1
Stress
Test Scenarios (2016)-
1-
Baseline- reflects the Federal Reserve’s interpretation of
market expectations
2-
Adverse-
3- Severally Adverse Scenario
Scenarios (Jan-2016)
|
|||
Economic Activity and Prices
|
|||
Baseline Scenario
|
Adverse Scenario
|
Severely Adverse
|
|
The baseline scenario for the United States is a
moderate economic expansion through the projection period.
|
Characterized by weakening economic activity
across all countries. Downturn is accompanied by a period of global deflation.
|
Characterized by a severe
global recession, accompanied by a period of heightened corporate financial
stress and negative yields for short-term U.S. Treasury securities
|
|
Real
GDP growth (United States)
|
Growth 2.5 % per year
|
Growth of 1.25% rises at an annual rate in the second quarter of 2017 to 3 % at an
annual rate by the middle of 2018.
|
Decline in
the first quarter of 2016 and reaches a trough in the first quarter of 2017 that is 6.25 % below the pre-recession peak
|
Unemployment
rate
|
Declines to 4.5 % in the middle of 2017 near that level through the end.
|
Rises
steadily to 7.5 % in the middle of 2017. Declined to 7% by
the end of the period.
|
Increase to 10%
|
CPI
|
Rises to 2.5 % at an annual rate by the middle of 2017, before dropped to 2 % in the first quarter of 2018 and
remaining near that level thereafter.
|
Falling about 0.5% over the four quarters of 2016.
Rise to 1.75% by the end of the period.
|
Rises from about 0.25% at an annual rate in the first quarter of 2016 to about 1.25% at an annual rate by the end of the
recession.
|
Interest Rate Measures (Assumed to
rise Steadily)
|
|||
Short-term
Treasury rates
|
Increase
from about 0.5% to 5.5% by
the beginning of 2019
|
Almost 0 %
|
Fall to – 0 .5% remain
at this level through the end.
|
10-year
Treasury securities
|
From 2.5% to
about 3.75% percent over the same period.
|
Declines to
1.25% early 2016 , rising gradually thereafter to 3%in the first quarter of 2019
|
Drops to about
0.25% in the first quarter of 2016, rising gradually 0.75% by the end of the
recession in early 2017 and about 1.75
% by the first quarter of 2019.
|
The
prime rate
|
Increase in line with short term T-rates and
Mortgage rate in line with long term T-rates.
|
Spreads between mortgage rates, investment grade
corporate bonds and 10-year Treasury yields widening through the end of 2016
|
|
Spreads
|
Yields on
investment-grade corporate bonds and yields on long-term Treasury securities narrow modestly
|
Increases to 5.75 %
|
|
Asset prices or financial conditions
|
|||
Equity
Prices
|
Rise an average of about 4.75%/year and volatility assumed
to remain near its historical average.
|
Fall 25%
through the fourth quarter of 2016, moderate rise in equity market volatility
|
Fall approximately
50% through the end of 2016, volatility at a level attained in 2008.
|
Nominal
house prices
|
Rise an
average of 2.75% per year
|
Moderate
decline; CREP fall 12% through the
third quarter of 2017 and house prices fall 12 % through the third quarter of 2018.
|
Dropping 25% through the third quarter of 2018
|
Commercial
real estate prices (CREP)
|
Rise an
average of 4.25% per year.
|
Falling 30% through the second quarter of 2018.
|
|
INTERNATIONAL VARIABLES (Real GDP
growth) over the scenario period
|
|||
Developing
Asia
|
6.00% per
year
|
Table
|
Mild
Recession
|
United
Kingdom
|
2.25% per
year
|
Recession
|
|
Euro
area
|
1.75% per
year
|
Recession
|
|
Japan
|
1.00%
per year
|
Recession
|
|
2.2 Factors and Impact on Banks-
- The period of deflation generally reduces
nominal household income growth and raise
real effective interest rates. These are conditions that may be expected to
reduce loan repayments and increase
credit losses.
- The lower path of Treasury rates may be expected to reduce pre-provision net revenue (PPNR),
largely through reduced net interest
income.
Controllers usually concentrate on what has already happened inside a company. They
prepare financial statements and other reports based on past activity.
Treasurers focus outward and interact with the bankers, shareholders and
potential investors who provide capital. In some small businesses, the owner, a
controller and an outside accountant might share the financial duties.
.
Stress Test Models
Top down
Approach – Portfolio is segmented into pools of common
risk factors.
Using Factor
Autoregressive Model and Vector Autoregressive Models.
Off- Balance Sheet Activities- Operating Lease, client’s assets under
management, borrowing done through Special Purpose Vehicles.
3.1 Simulation based on lagged default probabilities-
Simulation of forward-looking default probabilities based on
a function of their lagged value is a common starting point for firms early in the
stress-testing process when data and time is limited.
3.2 Merton-based stress testing-
Merton model takes this
form for company (i) –
B = value of debt which must be repaid
at the end of single period,
V = value of company assets, μ = the drift in the return on company
assets,
σ = the volatility of the return on
company assets, and k = the ratio of
(the drift in the return on the total assets of all companies) to the risk free rate r.
Approach to fitting the
Merton approach to macro factors takes these steps:
1-
The
unobservable value of company assets and the unobservable volatility of company
assets are implied from historical stock price movements.
2-
An
index of the value of assets for all companies in the same industry and
geographical region is constructed.
3-
A
regression is run which links the return on the assets of all companies
in that “industry-region” with the
return on the assets of company i. The error
term of the regression is “the idiosyncratic component” of credit risk for company i.
4-
After
the fact, the variation of the “industry
region” asset returns is in turn fit to a specified set of macro factors to
allow factor-specific stress testing of Merton default probabilities.
Drawbacks of Merton Model - approach to stress testing fails normal validation
procedures for three important reasons:
1-
Default
probability estimates developed from Merton procedures fail the “effective
challenge” test.
2-
The
four stage econometric procedure used to fit these default probabilities to
macro factors is a “forbidden model”.
3-
Portfolio
simulation using the Gaussian copula methodology (still distributed by a rating
agency) is widely recognized as inaccurate
3.3 Reduced Form Default Probabilities-
Uses a combination of
financial ratios, stock price history, and macro-economic factors.
Three Methods overview-
Method 1
|
o
Predict future default probabilities from historical default
probabilities.
Lagged 3
month default probability so we have a nesting of equations
|
Method 3
|
Future default probabilities as a function of “assumed known”
macro factors alone.
No lagged explanatory variables in this simple implementation
|
Method 4
|
Future default probabilities as a function of “assumed known”
macro factors and lagged historical company attributes.
Because we have these lagged variables, we need 13 equations.
|
The most common choices for a default probability function is a cumulative
probability function like that of the
normal or logistic distribution.
Logistic
function to predict the default probability at time t based on n functions of
macro factors Xi for i = 1, n:
Econometrically, some
common choices are-
Regression Estimation Strategy- We now need
to fit one relationship for each period for which we need a forecasted default probability at quarter k given macro
factors at time k and our company specific inputs at time 0. Case in which explanatory variables includes the lagged default probability (which is
only “known” at time 0 and historical dates) and lagged financial ratios. Assume that there are n functions of the 28 CCAR macro factors. One such system of
equations that generates the relevant 13 quarters of forecasts as a function of
our time zero inputs and the 28 macro factors Xi specified in CCAR is given
here. The macro factors are not lagged
unless the econometric process reveals that lags are helpful. Negative numbers in parentheses denote
lagged values, which effectively cause the time 0 values to be used as
explanatory variables:
Quarter 1:
PD=f [PD(-1),NI/A(-1), R(-1), X1, X2,….Xn]
Quarter 2:
PD=f [PD(-2),NI/A(-2), R(-2), X1, X2,….Xn]
------
Quarter 13:
PD=f[PD(-13),NI/A(-13), R(-13), X1, X2,….Xn]
Validation via Root Mean square root error by considering out of sample
data.
3.4 KMV Model-
The
model calculates all parameters Return
on individual Loans (Ri), Risk (si) as a
function of expected default frequency (EDFi), loss given
default (LGDi) and correlation (r) and then
all are used to calculate portfolio returns and risk. The model splits risk into systematic and unsystematic and then systematic to
Industry specific risk, country specific risk and common factors leading to
country and industry risk. Also the model overcomes the assumption of the normally distributed loans’ behaviors.
KMV
correlation model uses multifactor model, which reduces
correlations to be calculated to those between the limited numbers of common
factors affecting asset returns. Three-layer
factor structure model-
1- Model decomposes asset returns into systematic
and unsystematic risk.
Historical
returns of each exposure of the each business and T-bill yields of four
countries of government loans are regressed with composite market index. The part
not explained by the regression is the firm specific risk.
2- Composite
market index is constructed as a weighted sum of the firm’s exposure to country
and industry factors.
3- The factor structure the risk of countries and industries is further decomposed into
systematic and idiosyncratic components. Than with the slope coefficients correlation is calculated. Once correlation is
calculated expected return of the portfolio and loan portfolio risk is
calculated.
Formulas-
·
·
·
·
)
·
·
·
·
·
.
–
.
Regression
4.1 General linear methods-
Solve via
Ordinary Least Squares (OLS)
Y
4.2 Log-Linear Model-
Then we have,
4.3 Logit Probability Models-
A
probability model is one where the possible outcomes (“y-values”) are between 0
and 1.
Probability
Models are based on cumulative distribution functions (“cdf”) of probability
distribution functions. The Logistic
Distribution (“Logit”) and Normal Distribution (“Probit”) are the most common
distributions.
The difficulty with a Probability Model is that
we do not know the “TRUE” probability values. We can only observe of the
event happened (Y=1) or it did not happen (Y=0).
The
difference between Logistic and Probit
models lies in this assumption about the
distribution of the errors.
Logistic Distribution:
Steps Involved-
Transform the Probability Model into
a linear model via log transformation
Solving using 1) “grouped data,” 2)
“point estimation,” 3) “maximum likelihood estimates”
Do not need to make adjustment for
Regression Error
4.4 Fractional regression-
4.5 Maximum likelihood estimation –
Maximizes the likelihood function (probability of obtaining that
particular set of data, given the chosen probability distribution model).
Maximizing log likelihood for parameters.
4.6 Non-linear least squares estimation – Using random value of parameters and minimizing the error (In Excel via solver). Analysis if the error term is normally distributed.
Definitions
-
Common Equity Tier 1- Banks core capital, Capital Common equity tier 1 capital is the most
loss-absorbing form of capital. It includes qualifying common stock and related
surplus net of treasury stock; retained earnings.
-
Additional Tier 1 Going-Concern Capital- Convertible Capital Instruments (fixed-income instrument that is convertible into equity ), Perpetual preferred stocks
- The Tier 1 common capital ratio differs from
the Tier 1
capital ratio. Tier 1 capital includes the sum of a
bank's equity capital, and its disclosed reserves and non-redeemable,
non-cumulative preferred
stock. Tier 1 common capital, however, excludes all types of preferred
stock as well as non-controlling interests.
-
Tier 2 capital is supplementary capital (not related to ordinary
business results) revaluation
reserves, hybrid capital
instruments and subordinated term debt, general loan-loss reserves, and undisclosed
reserves.
- Tier 1 Leverage Ratio- to ensure the
capital adequacy of
banks and to place constraints on the degree to which a financial company can
leverage its capital
base.
-
A key difference between
the SLR and the U.S. leverage ratio is that the former takes into account
both on-balance sheet and certain off-balance sheet assets and exposures,
whereas the latter only measures a banking organization’s on-balance sheet
leverage
Capital Conservation Buffer Calculation-
Cushion of 2.5% to deal stress scenarios.
- Types
of payments that are restricted if a bank does not satisfy the Capital
Conservation Buffer requirement: • Dividends • Share buybacks • Discretionary payments on Tier 1
instruments • Discretionary bonus payments.
-
Capital Action- Means any issuance or redemption of a debt or equity capital instrument, any capital
distribution, and any
similar action that the Federal Reserve determines could impact a bank
holding company's
consolidated capital.
-
Capital Distribution- Means any payment by the Bank, whether in cash or stock, of a
dividend, any return of capital or retained earnings by the Bank to its
shareholders, any transaction in which
the Bank redeems or repurchases capital stock, or any transaction in which the Bank redeems,
repurchases or retires any other instrument which is included in the calculation
of its total capital.
- Pre-Provisions
Net Revenue- Net revenue before
adjusting for loss provisions (Amount net aside by a company as assets to pay for anticipated future losses).
- Net
Charge-Offs- Dollar
amount representing the difference between gross charge-offs and any subsequent
recoveries of delinquent debt
- The
trading book is an accounting term that refers to assets held by a bank that are regularly
traded. The trading book is required under Basel II and III to be marked
to market daily. The value-at-risk for
assets in the trading book is measured 99% confidence level based on a 10-day
time horizon under Basel II.
- The
banking book is also an
accounting term that refers to assets on a bank's balance sheet that are
expected to be held to maturity. Banks are not required to mark these to
market. Unless there is reason to believe that the counter-party will default
on its obligation, they are held at historical cost. Value at Risk is calculated at a 99.9% confidence level on a one-year
horizon.
§ Other
comprehensive income is those revenues,
expenses, gains, and losses that are excluded from net income on the income statement. Examples-
Unrealized holding
gains or losses on investments that are classified as available for
sale
Foreign currency
translation gains or losses
Pension plan gains
or losses
Pension prior
service costs or credits
Example Composition of CRE Portfolios Held by CCAR Banks
Two types of commercial real estate loans:
1-
Permanent Loans- Permanent loans are loans backed by existing commercial properties, such
as apartments, office buildings, retail stores, hotels, etc.
Reported in two parts-
a-
Non-farm Loans Non-Residential Loans- source of repayment comes from cash flows
generated by the borrower’s business operations
b-
Multi-Family Residential-
- Construction
loans are loans for
commercial properties under construction.
Commercial Real Estate- Loan
specific Details- geography, property type, and other financial
information such as LTV, DSCR, and coupon rate, etc.
Debt Service Coverage Ratio = Net Operating
Income / Total Debt Service (TDS- Proportion of gross income already spend)
-
Projections of Pre Provision Net
Revenue, Gain and Loss of AFS and HTM securities.
-
Projections of net charge-offs, reserves, and loan
balances, based on composition and characteristics of wholesale and consumer
loan portfolios across:
Wholesale – sector, region, and risk rating
segments
Consumer – loan level, asset class, and
behavioral segments
Regulatory Ratios Requirements
https://drive.google.com/file/d/0Bx3mfFH5R-y3Vm5PSVNib19nVGM/view?usp=sharing
No comments:
Post a Comment