Impact of Recent Supervisory Guidance on Capital Planning

This article was written by Jacob Kosoff and Rachel Bryant on the impact of SR 15-18 & SR 15-19 on US Banks.

1. The RMA Journal February 2017 44
ENTERPRISERISK
<$50 BILLION $50-$250 BILLION >$250 BILLION
The Impact of Recent Supervisory
Guidance on Capital Planning

2. February 2017 The RMA Journal 45
S
hutterstock
.
com
certain areas of capital planning may
require enhanced rigor—even for less
complex firms.
To help banks digest this new guid-
ance, this article focuses on the day-to-
day impacts for the large and noncomplex
firms that take part in the Federal Re-
serve’s Comprehensive Capital Analysis
and Review (CCAR), while also providing
useful takeaways for medium-size and
smaller firms.
Key features of both the heightened
expectations and the reduced or tailored
expectations are detailed below, including
governance, risk management, internal
controls, capital policy, incorporating
stressful conditions, and estimating im-
pacts on capital positions.
Potentially Tailored Expectations
Large and noncomplex (LNC) firms oper-
ating under SR 15-19 have an opportunity
to assess the impact of the guidance and
to potentially identify tailored expecta-
tions. SR 15-19 does not appear to reduce
expectations drastically for less complex
large firms, and meaningful pullbacks
in capital planning activities would not
seem the intent of the supervisory guid-
ance. However, there are certain areas,
outlined below, where expectations seem
to have been tailored between the sys-
temically important financial institutions
(SIFIs) and the LNC firms.

1. Scenario Design
LNC firms may currently produce in-
ternal baseline and stress scenarios, which
require significant resources, models, and
production time. SR 15-19 provides relief
in this area by allowing a firm to 1) adopt
the Federal Reserve’s baseline scenario as
its own, and 2) adjust the Fed’s severely
adverse scenario to reflect the firm’s own
activities, outlooks, and idiosyncratic
risks, instead of developing an internal
forecast.
In practical terms, this change presents
both a risk and an opportunity. The risk
lies in the fact that the scenarios may not
be published until a few weeks before the
CCAR submission due date, which has
been the pattern in past CCAR cycles. If
this trend continues, firms will not know
what the scenarios will contain until late
in the production cycle. To complicate
matters further, the scenarios may not
be consistent with the firm’s specific risks
and therefore could require significant
tweaking. Waiting for unknown sce-
narios may crunch an already tight pro-
duction timeline and disrupt the firm’s
governance process.
Despite the opportunities for reduced
costs, curtailing of redundancies, and a
more simplified process, some firms may
decide that the risks outweigh the ben-
efits, considering that the firm’s capital
plan is determined by these scenarios.
2. Use of Qualitative Approaches
versus Quantitative Approaches
Today’s LNC firms might rely heavily
on quantitative approaches for generating
projections. However, these firms may
consider using qualitative approaches,
especially if they can establish that these
simpler approaches produce realistic and
transparent output and that they are sup-
ported by well-documented rationale
based on expert judgment. Importantly,
these qualitative approaches must be well
supported and, although they may reduce
overhead in some ways, they could re-
quire new costs to establish and support
their ongoing use.
Similarly, a firm may use simple as-
sumptions to generate estimates for im-
material portfolios or businesses, which
would imply that the use of simple ratios
or cause-and-effect relationships may be
possible in certain instances. SR letter
BY JACOB KOSOFF
AND RACHEL BRYANT
In the context of capital planning, the
Federal Reserve separates banks into three
tiers based on asset size: 1) banks with
less than $50 billion in assets, 2) banks
with between $50 and $250 billion, and
3) banks with more than $250 billion.
Over the past 12 months, the Fed began
tailoring capital planning expectations for
banks in the second tier, establishing sig-
nificant differences for this group versus
the larger, more complex institutions.
This bifurcation began on December
18, 2015, when the Federal Reserve is-
sued separate supervisory letters with
guidance on capital planning expectations
for large, noncomplex firms with assets of
between $50 and $250 billion (SR 15-19
1
)
and large, complex firms with more than
$250 billion in assets (SR 15-18
2
).
Today, this shift continues as the Fed
publicly discusses
3
and clarifies
4
these
newly tailored expectations. Meanwhile,
in light of additional details provided in
the new SR letters and in public com-
ments made by Federal Reserve officials,

3. The RMA Journal February 2017 46
allows for a less formal risk identification
process as well as less frequent reviews
of material risks.
While this may represent a relaxation of
logistical requirements, the main require-
ment is the same: “A firm should be able
to demonstrate how material risks are ac-
counted for in its capital planning process.
For risks not well captured by scenario
analysis, the firm should clearly articulate
how the risks are otherwise captured and
addressed in the capital planning process
and factored into decisions about capital
needs and distributions.... The risk identi-
fication process at all firms subject to this
guidance should be dynamic, inclusive,
and comprehensive, and drive the firm’s
capital adequacy analysis.”
Thus, the identification of risks, de-
termination of materiality, and capture
of material risks within the scenarios and
results continue to be critical pillars of
capital planning.
6. Use of Payout Ratios in Distribution
Decision Making
Interestingly, SR 15-19 exempts non-
complex institutions from using payout
ratios when deciding how much capital
to distribute through dividends, share re-
purchases, or other means. Meanwhile,
SR 15-18 expects SIFIs to use quantitative
payout ratios in their decision-making
framework. This presents a conundrum
that may prompt SR 15-19 firms to hold
themselves to the higher standard.
While SR 15-19 may allow firms with
15-19 states, “A firm can apply simple
assumptions to generate losses or PPNR
for its non-material portfolios or business
lines…. A LISCC Firm and a Large and
Complex Firm is generally expected to
use quantitative approaches in estimat-
ing losses and PPNR, whereas a Large
and Noncomplex Firm may use either
quantitative or qualitative approaches.”
Overall, while quantitative approaches
remain useful and important for material
portfolios, the guidance provides latitude
for firms when it comes to deciding where
to invest modeling time and resources.
3. Independent Review of Qualitative
Approaches
Guidance allows for the review of
qualitative and quantitative approaches
to differ based on the complexity of the
approach, the impact of the approach’s
output, and the materiality of the port-
folio or business line for which the quali-
tative approach is used. This allows an
opportunity for quantitative approaches
to continue to be independently reviewed
by a model risk management and vali-
dation team, while the independent re-
view of qualitative approaches may be
performed by other areas of the bank.
4. Use of Benchmark Models
Firms should give careful consider-
ation to the benefits of developing and
deploying benchmark or challenger
models, since the guidance states that
noncomplex firms are not expected to
use benchmark models in their capital
planning processes. However, SR 15-19
is not intended to replace SR 11-7,
5
which
mentions benchmarking 11 times. There-
fore, firms should continue to assess the
need for benchmark models in each situ-
ation, as well as assess other methods for
achieving the necessary benchmarking.
5. Formality of the Risk Identification
Process
SIFIs are expected to have a more for-
mal risk identification process, to work
toward greater risk granularity, and to
include quarterly evaluations of material
risks. However, these heightened expec-
tations are not present in SR 15-19, which
assets of less than $250 billion to exclude
payout ratios from their decision-making
framework, doing so may not be advis-
able. Given that distributions funda-
mentally affect the firm’s overall capital
levels, exclusion of payout ratios from the
capital decision-making framework may
prove a difficult or inappropriate choice
for some firms, particularly those seeking
to make material distributions.
7. Full Validation Before Model Use
SR 15-19 states that firms should inde-
pendently validate or review models used
in internal capital planning consistent
with SR 11-7. After devoting multiple
paragraphs to the expectations of these
validations and reviews by a model vali-
dation group, SR 15-19 then recognizes
that not all models can be fully validated
prior to use. This is where the difference
between the sets of banks emerges.
The guidance “expects” SIFIs to
complete full validations prior to use,
while the guidance cautiously allows
for instances when the LNC firms are
expected only to “make efforts to con-
duct a conceptual soundness review”
of material models prior to their use in
capital planning. Therefore, a reading
of the guidance suggests that both sets
of firms should attempt to perform an
independent assessment of the models
prior to use. The SIFIs are expected to
have these completed prior to use, while
LNC firms may have more flexibility.
For example, an LNC firm may simply
While sensitivity analysis has become well established
at the model and portfolio levels, sensitivity analysis
should also be performed at higher aggregated
levels of the capital plan, including assessments
“across the entire firm’s projections under stress.”

4. February 2017 The RMA Journal 47
have model validations in process prior to
use or may only be able to complete the
conceptual soundness review before use
and other components of the validation
follow. Also, SR 15-19 refers to “mate-
rial” models, whereas SR 15-18 refers to
“all” models requiring validation before
use. This gives firms the opportunity to
prioritize validation work and focus on
having, for example, high- and medium-
risk model validations completed prior
to use, while low-risk models may be in
process while the model is in use. This
may require firms to update their corpo-
rate model risk policy to allow for this
situation.
The guidance does refer to this prac-
tice—using a model prior to valida-
tion—as a “model risk management
shortcoming” and recommends using
the model output with “caution” in those
situations. In short, while a firm may, in
certain situations, use a model prior to
the completion of a validation, this ap-
proach is not encouraged and should be
well controlled.
As firms reevaluate the categorization
of their capital planning models, any cat-
egorization of a model as “not material”
will require evidence. Therefore, the time
spent on appropriately understanding
the uncertainty and importance of each
model may increase to ensure appropriate
classification of each model.
Furthermore, leaders of model valida-
tion departments may have to spend a
larger share of their time ensuring that
the criterion to determine materiality is
applied consistently across all validation
teams. To achieve this consistency, firms
may need more rotation of validation
managers and analysts between valida-
tion teams, additional staff training on
appropriately classifying models, and
enhanced classification procedures. For
example, if one validation team considers
materiality relative to the firm’s systemic
risk while another considers materiality
with respect to absolute size, the result
may be inconsistent model classification.
Thus, firms may use SR 15-19 to revisit
their model classification procedures and
outcomes.
8. Other Areas of Potentially Tailored
Expectations
SR 15-19 addresses numerous areas
in which the expectations applied to
large, noncomplex firms may be differ-
ent or less onerous than those applied
to SIFIs, including additional areas not
specifically discussed above. Overall, SR
15-19 reiterates the importance of capital
planning, while enumerating areas where
the different complexities between these
two groups may result in potentially tai-
lored expectations for firms with less than
$250 billion in assets.
TABLE : OTHER AREAS OF POTENTIALLY TAILORED
EXPECTATIONS
Internal audit depth
and breadth
Senior Management
Engagement
Approaches To Operational
Loss Projections
Granularity Of Projections
Use Of External Data
Independent Review
Of Model Overlays
TAKEAWAYS FOR BANKS WITH LESS THAN
$50 BILLION IN ASSETS
While SR 15-19 does not apply to banks with less than $50 billion in assets,
smaller banks may take away a few key points from the new guidance.
First, SR 15-19 reiterates and stresses the importance of stress-testing
fundamentals, including accuracy of reports, reliability of processes,
maintenance of internal controls, effectiveness of internal audit, and
accuracy of reports. These fundamentals are important for all banking
organizations, even those that do not have to comply with SR 15-19.
A smaller institution may benefit from taking a holistic read of the guidance,
since it seems to stress the importance of these and other fundamentals
rather than focusing on complex processes or dictating specific methods
for stress testing.
Meanwhile, SR 15-19 reduces expectations of complexity and, in certain
cases, discusses the risks of conducting an overly complex process.
Embracing the spirit of the letter, a smaller institution might evaluate the
complexity of its processes and note areas where excessive complexity is
hampering essential activities.

Potentially Heightened Expectations
While SR 15-19 tailored expectations in
certain areas, other areas received ad-
ditional attention and detail that firms
may perceive as heightened expectations.
None of these areas is new to capital plan-
ning or regulatory guidance; rather, SR
15-19 provides new details that firms
should fully understand and incorporate
into their processes.
1. Ongoing Capital Planning
SR 15-19 discusses the importance of
capital planning becoming a year-round,
ongoing process that is integrated with
the firm’s regular business activities.
While these expectations may have
been in place before SR 15-19, this
guidance addresses capital planning as
a year-round, ongoing process in a more
explicit manner. Regulators may expect
firms to make a discernable shift toward

5. The RMA Journal February 2017 48
Moreover, SR 15-19 makes clear that
the firm’s standards for managing changes
to the capital planning process should
be codified in internal policy, which is
an expectation that builds on previous
supervisory guidance.
4. Internal Controls
Recent guidance reiterates the impor-
tance of internal controls and describes
various expectations for maintaining
strong internal controls throughout the
capital planning process. After many
years of execution, firms may need to
refresh their internal controls across all
three lines of defense to ensure complete
coverage.
5. Board Reporting
While robust reporting is a long-stand-
ing expectation of a sound capital plan-
ning process, SR 15-19 took additional
steps in detailing the type and depth of
materials to be presented to the board
of directors. Given the specificity in this
regulatory guidance, firms should revisit
their board reporting to ensure compli-
ance with SR 15-19 standards.

Conclusion
Ultimately, SR 15-19 contains many ex-
amples of both tailored and heightened
expectations. Accordingly, firms might
benefit from conducting a gap analysis
integrating capital stress testing into
business-as-usual processes.
2. Firm-Wide Sensitivity Analysis
While sensitivity analysis has become
well established at the model and portfo-
lio levels, sensitivity analysis should also
be performed at higher aggregated levels
of the capital plan, including assessments
“across the entire firm’s projections under
stress.”
Given the importance of sensitiv-
ity analysis and the expectation that
it should be presented to the board of
directors, firm-wide sensitivity analysis
becomes even more important. While
this expectation may create an additional
obligation within capital planning, the
aggregated sensitivity analysis is likely
more digestible and useful for the board
of directors as it reviews and approves
the firm’s capital plan.
3. Change Management
As capital planning processes have ma-
tured across the industry, change manage-
ment has become more critical, simply
because changes can disrupt the estab-
lished process in unexpected ways. Thus,
SR 15-19 expects senior management to
“make certain the firm is identifying,
documenting, reviewing, and tracking all
material changes to the capital planning
process and its components.”
relative to the new guidance to determine
where they are exceeding expectations
and to identify areas where additional
work may be needed.
SR 15-19 is unique: It is the only super-
visory guidance to explicitly differentiate
between SIFIs and firms that are large but
less complex, and it is a welcome addi-
tion to the body of regulatory guidance
because it helps clarify expectations for
banks of both sizes. With this clarified
guidance, the industry can continue to
refine capital planning processes to en-
sure they are sustainable, executable, and
integrated into business practices.
Jacob Kosoff is head of model risk management
and validation and Rachel Bryant is head of capital
planning and stress testing at Regions Bank. They
can be reached at [email protected]. and
[email protected].
Notes
1. See Federal Reserve Supervisory Letter 15-19,
“Federal Reserve Supervisory Assessment of
Capital Planning and Positions for Large and
Noncomplex Firms,” December 18, 2015.
2. See Federal Reserve Supervisory Letter 15-18,
“Federal Reserve Supervisory Assessment of
Capital Planning and Positions for LISCC Firms and
Large and Complex Firms,” December 18, 2015.
3. Available at http://www.bloomberg.com/news/
videos/2016-06-02/fed-s-tarullo-on-bank-regula-
tions-fed-policy-economy.
4. Available at http://som.yale.edu/event/2016/09/
daniel-k-tarullo-member-of-the-board-of-governors-
federal-reserve-system.
5. See Federal Reserve Supervisory Letter 11-7,
“Supervisory Guidance on Model Risk Manage-
ment,” April 4, 2011.
The opinions expressed in this article are state-
ments of the authors, are intended only for
informational purposes, and are not opinions of
any financial institution. Any representation to the
contrary is expressly disclaimed.
SR 15-19 IS UNIQUE: IT IS THE ONLY
SUPERVISORY GUIDANCE TO EXPLICITLY
DIFFERENTIATE BETWEEN SIFIS AND FIRMS
THAT ARE LARGE BUT LESS COMPLEX.