Wells Fargo 2010 Annual Report Download - page 135

Allowance for Credit Losses (ACL)
The ACL is management’s estimate of credit losses inherent in
the loan portfolio, including unfunded credit commitments, at
the balance sheet date. We have an established process to
determine the adequacy of the allowance for credit losses that
assesses the losses inherent in our portfolio and related
unfunded credit commitments. While we attribute portions of
the allowance to specific portfolio segments, the entire allowance
is available to absorb credit losses inherent in the total loan
portfolio and unfunded credit commitments.
Our process involves procedures to appropriately consider
the unique risk characteristics of our commercial and consumer
loan portfolio segments. For each portfolio segment, impairment
is measured collectively for groups of smaller loans with similar
characteristics, individually for larger impaired loans or, for PCI
loans, based on the changes in cash flows expected to be
collected.
Our allowance levels are influenced by loan volumes, loan
grade migration or delinquency status, historic loss experience
influencing loss factors, and other conditions influencing loss
expectations, such as economic conditions. We have had limited
changes in our allowance methodology primarily associated with
integration alignment of loss estimation processes between
Wells Fargo and Wachovia. Those changes did not significantly
impact the allowance for credit losses.
COMMERCIAL PORTFOLIO SEGMENT ACL METHODOLOGY
Generally, commercial loans are assessed for estimated losses by
grading each loan using various risk factors as identified through
periodic reviews. We apply historic grade-specific loss factors to
the aggregation of each funded grade pool. These historic loss
factors are also used to estimate losses for unfunded credit
commitments. In the development of our statistically derived
loan grade loss factors, we observe historical losses over a
relevant period for each loan grade. These loss estimates are
adjusted as appropriate based on additional analysis of long-
term average loss experience compared to previously forecasted
losses, external loss data or other risks identified from current
economic conditions and credit quality trends.
The allowance also includes an amount for the estimated
impairment on nonaccrual commercial loans and commercial
loans modified in a TDR, whether on accrual or nonaccrual
status.
CONSUMER PORTFOLIO SEGMENT ACL METHODOLOGY For
consumer loans, not identified as a TDR, we determine the
allowance on a collective basis utilizing forecasted losses to
represent our best estimate of inherent loss. We pool loans,
generally by product types with similar risk characteristics, such
as residential real estate mortgages and credit cards. As
appropriate, to achieve greater accuracy, we may further stratify
selected portfolios by sub-product, origination channel, vintage,
loss type, geographic location and other predictive
characteristics. Models designed for each pool are utilized to
develop the loss estimates. We use assumptions for these pools
in our forecast models, such as historic delinquency and default,
loss severity, home price trends, unemployment trends, and
other key economic variables that may influence the frequency
and severity of losses in the pool.
In addition, we establish an allowance for consumer loans
that have been modified in a TDR, whether on accrual or
nonaccrual status.
OTHER ACL MATTERS Commercial and consumer PCI loans
may require an allowance subsequent to their acquisition. This
allowance requirement is due to probable decreases in expected
principal and interest cash flows (other than due to decreases in
interest rate indices and changes in prepayment assumptions).
The allowance for credit losses for both portfolio segments
includes an amount for imprecision or uncertainty that may
change from period to period. This amount represents
management’s judgment of risks inherent in the processes and
assumptions used in establishing the allowance. This
imprecision considers economic environmental factors,
modeling assumptions and performance, process risk, and other
subjective factors, including industry trends.
133