SlideShare a Scribd company logo
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company
is strictly prohibited
March, 2022
Discussion Document
Data and analytics
innovations in Credit
Portfolio Management
2
McKinsey and
IACPM together
completed a survey
on new
developments in
data and analytics
for credit portfolio
management
The survey had 3
main objectives
Survey range of industry practice on:
Ÿ Emergence of alternative data sources for credit risk
identification, assessment, and monitoring
Ÿ Related evolution of analytics tools
Understand degree to which different data types and
analytical approaches are in use/under consideration
across institutions, geographies, subsectors, and CRE/C&I
portfolio segments
Develop insights on current state and path forward for
participants to incorporate next generation data and
analytics
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 3
44 financial institutions participated in the survey from across
Americas, APAC, and EMEA
Asia Pacific
Americas
Europe Middle East and Africa
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 4
Content Summary of survey results 20 mins
Perspectives on selected topics 30 mins
Questions 10 mins
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company
Summary of results
D Investment and
Strategic Goals
Top investment areas have been data tech and data acquisition. Over the next 2 years, participants expect
a greater role of innovative data and advanced analytics in improving Credit Strategy and Client
Experience
A Trends Over the next 2 years, larger number of participants expect significant increase in the use of internally
developed advanced techniques and new types of external data
F Impact Use of innovative data and/or advanced analytics improves model accuracy , turn-around-time, automated
decisioning and time spend on analysis, with higher benefit observed in the SME segment
E Use cases Machine learning models are primarily gaining traction for Risk Scoring of SMEs and Early Warning across
the board
Innovative external data sources are more used for Corporate segment while SME segment uses more
innovative internal data sources
B Challenges Data quality assessment and talent management are the top challenges for use of both advanced
analytics and innovative data solution
C Climate Majority of the participants believe that Climate and ESG is the next biggest challenge for credit
assessment
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 6
A: Over the next 2 years, a vast majority of participants expect
significant increase in the use of new types of external data and
internally developed advanced techniques
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
Expectation for the next 2 years
In the past 2 years
How has your firm’s data/analytics for credit decisions changed in the past 2 years and how do
you expect it to change in the coming 2 years?
In the past 2 years, over 60% of
the participants have seen an
increase in the:
Ÿ Use of new types of internal
and external data
Ÿ Use of internally developed
advanced techniques
Ÿ Size of data and analytics
team
Over the next 2 years, even
larger % of participants expect
this trend to continue
% participants see increase in trends
72%
74%
Increased use of new types
of internal data
Increased use of internally developed
machine learning and other
advanced analytics techniques
Increased use of vendor-developed
machine learning and other
advanced analytics techniques
Increased use of new types
external data
Change in the size of the
data & analytics team
Change in share of automated
credit approvals
66%
64%
49%
32%
91%
81%
59%
79%
81%
66%
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 7
B: Data quality assessment and talent management are the top
challenges for use of both advanced analytics and innovative
data solution
42
49
14
26
26
40
9
42
7
2
5
63
42
30
28
28
19
14
14
12
2
2
Validation difficulty
Cost of data
Data Quality
Ability to explain
Regulatory
Resources
Skepticism/Probability
Unmet expectations
Risk concerns
Other
Fragmentation
Currently, what are the major challenges faced by your firm that constrain the use of innovative
data or advanced analytics (e.g., machine learning and AI)?
Top 3 challenges for use of
innovative data solutions
Top 3 challenges for use of
advanced analytics solutions
Percentage
Major challenges for use of
advanced analytics solutions
are:
Ÿ Attract, retain and develop
resources
Ÿ Ability to explain
Ÿ Data quality
Ÿ Validation
While for using innovative data,
key challenge in data quality
assessment and talent
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 8
C: 86% of the participants believe that Climate and ESG is the next
biggest challenge for credit assessment
86
58
51
42
9
7
Post COVID-19 model adjustments
and uncertainties
Capital, Provisioning, or regulatory
stress testing model requirements
Climate and ESG
Incorporating Machine Learning
models within regulatory and
risk constraints
Effectively competing with attackers
Other – please specify
27: What are the biggest challenges facing credit risk
and credit portfolio management analytics in the next 2-
3 years?
Percentage
25: For incorporating the impact of climate risk ,are
you using existing loss models with climate shock
applied to input variables? Or are you developing
new loss models to assess it?
Percentage
86% of the participants believe
climate and ESG is the next
biggest challenge for credit
assessment, followed by
COVID-19 adjustments, capital
and regulatory exercise, and
using machine learning models
One third of the participants plan
to use existing credit models to
translate the climate impact to
credit risk and another one third
of the participants plan to
develop new loss models for
climate assessment
35
33
33 Using existing models
Developing new loss models
Not yet exploring
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 9
12%
Yes - Obligor climate risk scorecard
Yes - Adjustments to obligor rating methodology
No
Yes - Climate stress loss scenarios
36%
Yes - Other (1)
39%
21%
18%
Percentage of participants
SME Mid-Market Corporate
34%
21%
59%
21%
28%
54%
17%
39%
20%
22%
27%
21%
52%
39%
15%
CRE
1: E.g., Adjustments to obligor rating methodology and climate stress loss scenarios, but beyond 12 months. Bucketing of risks (geography, industry, property type segments).
N=33 N=29 N=41 N=35
Have you implemented or are planning to implement in the next 12 months any changes to the credit assessment/ adjudication and monitoring models to
capture the impact of climate change? (transition and physical risks)
C: > 50% of participants have implemented/or are planning to
implement climate stress loss analyses in mid-market, corporate,
or CRE space
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 10
D: In the past 2 years, top investment areas have been data tech
and data acquisition – this trend is expected to continue over the
next 2 years
13
5
8
2
6
6
10
4
9
5
6
7
10
9
1
0
Data scientists and engineers
for development and validation
Data acquisitions
Data tech
(e.g. cloud, visualization)
11
Data scientists and engineers
for data quality processes
3rd Party/Vendor models
0
0
Other, please specify 1
25
22
22
20
3rd largest
Largest 2nd largest
15
7
4
2
5
1
9
6
6
8
5
5
7
11
10
1
1
0
20
29
21
0
20
11
Where have you made the most investments in the past 2 years and where do you expect to
invest the most in the coming 2 years?
In the past 2 years, the top
investment areas for participants
were data tech and data
acquisition
And this trend in expected to
continue over the next 2 years
with higher expected investment
Other top investment areas
include talent for both
development/ validation and
data processing
Number of votes Expectation for the next 2 years
In the past 2 years
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 11
E: Machine learning models are primarily gaining traction for Risk
Scoring of SMEs and Early Warning across the board
Early warning indicators
Credit pricing
39%
Risk score/ Rating/
PD model
Stress Testing
IFRS 9/ CECL
44%
44%
39%
32%
37%
68%
24%
56%
46%
7%
10%
5%
29%
5%
12%
2%
0%
0%
0%
0%
0%
5%
0%
0%
Simulation-based
models (e.g.,
monte-carlo
approaches to
economic capital)
Statistical
approaches (e.g.,
linear or logistic
regression or
CHAID)
Machine learning
tree algorithms
(e.g. random
forest, gradient
boosting)
Expert based
qualitative
models
Other machine
learning models
(e.g. support
vector, NLP,
neural network)
44%
Stress Testing
Risk score/ Rating/
PD model
Early warning indicators
41%
Credit pricing 47%
IFRS 9/ CECL 44%
38%
66%
72%
56%
53%
38%
6%
9%
3%
6%
28% 0%
13%
16%
6%
3%
0%
0%
0%
3%
3%
Corpo-
rate
SME
What methodologies are being “used in production”, “validated” or “in pilot” for each of the listed use cases
for the Corporate and SME portfolio
Expert based and statistical
models are most widely applied
approaches across the
spectrum of use cases
Simulation based models are
more widely used for stress
testing
Machine learning models are
getting traction for Risk Scoring,
Early Warning, and Pricing
Percentage
Innovative data sources
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 12
E: Innovative external data sources are more used for Corporate segment
while SME segment uses more innovative internal data sources
Which of the following categories of data are being used in production, under pilot or under
consideration for credit risk management use cases within the Corporate portfolio?
External data sources:
For Corporates, over 50% of the
participants are using, piloting or
considering New media or social
media and 3rd
Party account data, a
higher proportion than for SMEs
Both segments use E-commerce data
at similar rates
Internal data sources:
For Corporates, over 70% of the
participants are using, piloting or
considering Automated client/ issuer
financials and internal credit behavior
data, while for the SME segment in
addition to above 2, internal cross
product data also has large share
External
data
sources
Internal
data
sources
97%
88%
76%
59%
53%
33%
32%
88%
76%
62%
56%
44%
Economic and market forecasts
Credit bureau and rating agency data
Financial market data on issuers
News media or social media
Other high frequency data
3rd Party account data
Automated client / issuer financials
E-commerce data
Internal credit behavior data
Internal cross product behavior
Internal client interaction data
Internal consumer-wholesale crossover data
91%
63%
35%
32%
30%
29%
19%
83%
91%
76%
45%
38%
Corporate SME
Percentage
Innovative data sources
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 13
F. Automated decisions are largely a feature of SME portfolios
SME
18%
18%
16%
34%
19%
11%
72%
8%
4%
Mid-market
87%
13%
Corporate
89%
11%
CRE
None and Not implemented
Less than 10%
11-30%
31-50%
Above 50%
Percentage of participants, where applicable
N= 27 25 31 27
Fully automated decisions for a
material portion of the portfolio
is almost exclusively a feature
of SME portfolios
However, there are pockets of
portfolio with full automation,
even for mid-market and others
Key insights
In the past 3-5 years, what was the typical percentage of automated decisions based on models
in your portfolio?
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 14
F: Where implemented, use of innovative data and/or advanced
analytics typically achieved increased automation
80
20
60
0
10
40
70
50
30
0
14
70
4
7
9
5
37
22 11
23
0
Mid-market
7
0
7
SME CRE
9
Corporate
5
26 27
36
In terms of increasing automation, what benefit have you seen in the past 3-5 years from the
use of innovative data and/or advanced analytics?
63% participants reported an
increase in automation for the
SME segment, followed by Mid-
Market (27%), Corporate (19%)
and CRE segment (14%)
Where implemented, use of
innovative data and/or advanced
analytics typically improved
automation by up to 10%
Higher improvement (11-50%) in
automation is typically observed
in SME segment, which involves
the highest level of direct
automated decisioning
Increased by 11-50%
Implemented but no uplift in automation
Increased by up to 10% Increased by more than 50%
70% 36% 26% 27%
XX%
% of participants
implemented
Percentage of participants that implemented
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 15
F: Where implemented, use of innovative data and/or advanced
analytics significantly improved turn-around-time “TAT” for SMEs
10
40
60
50
70
20
30
0
7
9
12
0
41
3
7
19
30
Mid-market
SME
20
28
30
12
0
Corporate
10
0
13
CRE
63
52
44
56% participants reported
decrease in TAT for the SME
segment, followed by Mid-
Market (43%), CRE (32%), and
Corporate (31%)
Where implemented, use of
innovative data and/or advanced
analytics typically improved turn-
around-time “TAT” by up to 10%
Higher improvement (11-50%) in
TAT is typically observed in SME
segment but also to some extent
in Mid-Market and CRE
segment
In terms of accelerating TAT to decision, what benefit have you seen in the past 3-5 years from
the use of innovative data and/or advanced analytics?
Decreased by 11-50%
Implemented but no reduction in approval time
Decreased by upto 10% Decreased by more than 50%
63% 52% 41% 44%
Percentage of participants that implemented
XX%
% of participants
implemented
Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 16
Agenda
Summary of survey results 20 mins
Perspectives on selected topics 30 mins
Questions 10 mins
McKinsey & Company 17
Perspectives on specific topics : Climate Risk, Next generation data
and analytics, and Scenario planning and forecasting
Climate Risk Next generation data and analytics Scenario Planning and Forecasting
Using cross-product data with help from
AI/ML can drive both revenue growth and
automated credit decision
For larger obligors, availability of analytics
and accessible data is key for turnaround
time reduction, however, doing so requires
treating data quality as more than a
‘regulator-required’ effort
Setting up a modular, portfolio specific
methodology for scenario analyses is
critical, prioritization will depend on
exposure to high-risk sectors
Climate risk impact on portfolio requires
inter-disciplinary skills and mobilization
across credit, front-line, and model risk
management
Building a flexible infrastructure to forecast
and optimize portfolio is more critical than
ever
Rather than waiting for a full-scale solution,
banks would want to establish analytics and
organizational capabilities that enable rapid
‘what-if’ analyses
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 18
Key learnings on the materiality of Climate Risk on Credit, based on
our extensive work with banks
The majority of the risk is in
knock-on impacts that most
banks do not model
Direct damages are immaterial on credit; Knock-on effects can dwarf direct impacts, e.g., 4.5x the impact
of direct 1st order impact for a Muni flood example
Material risk drivers include community deterioration, geographic transition risk, and broader ecosystem
impacts (e.g., insurance cost and availability)
Real estate losses are driven by asset pricing (property values and cap rates), not physical damage
Risk is concentrated in
“pockets” across the
portfolio; banks need to take
a targeted approach
Both physical and transition risk lie in very targeted areas of the portfolio: for example, for a large global
bank we identified that approximately ~15% of their loan book was materially exposed to climate risk
Banks need to perform heatmapping to focus their efforts on the high risk portfolios and risks: even within a
CRE portfolio for a large US bank, we found that majority of the credit impact came from 10% of the portfolio
​
C
urrent
SVA
(estimated)
​
O
il
and
Gas
​
C
ommercial
Real
Estate
​
A
griculture
​
M
anufacturing
​
E
xpected
2030
SVA
​
M
ining
​~700
​
T
ransportation
​
U
tilities
​
R
esidential
​~1100 ​-35%
​Impact on Economic profit1, CAD mn
1. Calculated as the return on invested equity above the return on equity (estimated at 10.2% based on McKinsey CPAnalytics) times invested equity
(allocation based on exposure). Assumes that percentage NII decline = percentage net income decline
Climate is not a “capital”
problem; however it can
have real impact on returns /
economic profit
A ‘CCAR mindset’ of focusing on capital risks will underestimate the business value risk and miss the
opportunity to steer the business
For a North American bank, we identified that 35% of economic profits could erode by 2030 without taking
action on key pockets of climate risk exposures
The “average” impact is
moderate in the near-term,
but there is high degree of
counterparty-level variability
We found that even for high risk industries, the average impact is moderate: For example, in a portfolio of
upstream O&G companies, the impact by 2025 under below 2C scenario was ~7% median impact on EBITDA
However, the difference between winners and losers is stark: in the upstream O&G example, we saw
several counterparties with up to ~40% impact on EBITDA, while there were other companies that saw a
positive EBITDA impact
Climate Risk
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 19
Scenario analyses for corporate and mid-market obligors needs to
be sector-specific: oil and gas example
Module Specific Data
Carbon cost
Upstream Oil &
Gas
Acute impact
Competition
module
Stage 2 profit
revenue, cost
Integration
module
Physical risk
impact
Transition risk
impact Climate-stress
obligor
financials
Supply Chain Oil
& Gas
Cleantech
Scenario selection (e.g. NGFS Scenarios)
Scenario expansion and country downscaling (e.g., damage curve, transition pathways)
i
ii
iii
iv
v
vi
vii
Obligor Financials
Oil and Gas Asset-
Level Data
Clean Tech Market
Shares
Obligor Emissions
Abatement Cost Curves
Methodology isolates individual aspects of scenario analysis through modules that calculate Oil and Gas obligor impacts from:
Ÿ Oil and Gas demand changes across Upstream and Downstream Operations
Ÿ Clean Technology demand changes
Ÿ Carbon Costs
Ÿ Acute Physical Hazard damage costs
Climate Risk
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 20
Physical risk impact Revenue impact Carbon cost impact
Sanitized example : Demand destruction, direct carbon cost, and
market impacts for a specific obligor
Driven by current
business composition,
breakeven cost, and oil
and gas price
Driven by carbon emissions which
drives carbon costs, abatement and
ability to pass cost to customers/
market impacts
Demand
destruction
Market
impacts
Current
valuation
7%
Adaptation
Physical
impact
3%
40%
0%
Demand
creation
50%
Direct
carbon costs
5%
Abatement
30%
Full impact
100%
41%
Delayed transition scenario – 2050
Valuation impact waterfall – average impacts weighted by exposure Takeaway
• Demand destruction, direct
carbon cost and market
impacts are the top 3 drivers
of the climate impact to the
O&G portfolio
• Demand destruction is
related to the business
composition and
breakeven cost
• Direct carbon cost is
related to the carbon price
and the amount of output
that can be sustained
• Market impact is related to
the ability to pass cost to
customers and gain
market share from other
players
ASSUMING NO MANAGEMENT/BANK ACTION
Climate Risk
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 21
Sanitized example: Portfolio level impact for a sample of public oil
and gas obligors
200
1,300
40
300
65
1,700
800
35
25 70
15 30
1,100
400
20
600
45
700
500
50 55
900
1,000
60
1,600
1,200
0
100
1,500
1,400
Cimarex Energy Co.
Exxon Mobil Corporation
Total SE
Petrobras
Scope 1 CO2eq per $M total revenue
Breakeven Oil Price (USD/bbl)
Seven Generations Energy Ltd.
Woodside Petroleum Ltd
BP p.l.c.
Aker BP ASA
Antero Resources Corporation
APA Corporation
Equinor ASA
Conocophillips
Marathon Oil Corporation
Santos Limited
Suncor Energy Inc.
Canadian Natural Resources Limited
Chevron Corporation
Cnx Resources Corporation
Eog Resources, Inc.
OMV Ag
Pdc Energy, Inc.
Pioneer Natural Resources Company
Royal Dutch Shell Plc
500
25
300
600
20
15 70
40
30
800
35 50
900
1,000
400
1,100
45
1,200
1,300
55
1,400
60 65
1,500
0
700
100
200
1,600
1,700
Total SE
Scope 1 CO2eq per $M total revenue
Exxon Mobil Corporation
Antero Resources Corporation
BP p.l.c.
Marathon Oil Corporation
Equinor ASA
Aker BP ASA
Woodside Petroleum Ltd
Suncor Energy Inc.
Cnx Resources Corporation
Canadian Natural Resources Limited
Royal Dutch Shell Plc
Chevron Corporation
Pioneer Natural Resources Company
APA Corporation
Santos Limited
Eog Resources, Inc.
Cimarex Energy Co.
OMV Ag
Pdc Energy, Inc.
Petrobras
Seven Generations Energy Ltd.
Breakeven Oil Price (USD/bbl)
Conocophillips
Impact of each driver on earnings for upstream companies
Bubble size: Market cap 100% of current market cap
2030: Immediate transition scenario 2050: Delayed transition scenario
A B
Note: Scenario analysis assumes no mitigation action by bank or borrowers; Source: McKinsey Energy Insights, Rystad
Earnings increase
Earnings decrease 51-100%
Earnings decrease <=50%
Climate Risk
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 22
Case study: Using a phased framework to develop bespoke scenario
impact assessment tool for CRE portfolios (1/2)
Impact to Net Operating Income Mechanism to translate financials to
expected loss
Identify risks and specify scenarios Assess impact on obligor’s financials Quantify portfolio impact
Identify hazards
Ÿ Flood, hurricane and
wildfire most important,
given the bank’s
footprint
Ÿ Transition risk leading
to GVA impact in
counties with high
fossil-fuel dependency
Downscale scenarios
Ÿ Hurricane wind severity
in every mile across
areas where storms
have been observed
Ÿ Probability of flood and
inundation levels
Additional drivers of credit
worthiness
Ÿ Community and infrastructure
resilience or vulnerability
Ÿ Additional sponsor risk
Ÿ Risk mitigants like levees
Impact to Property Value
Ÿ Decline in property value due to
change in expected impact of
climate events
Ÿ Uninsured damage from physical
risk event
Ÿ Short term revenue decrease due
to business interruption
Ÿ Increase in insurance cost due to
increase in frequency and severity
of physical risk
Ÿ NOI and property valuation impact Loan-
to-value and DSCR of properties,
translated into PD, LGD and EL
Ÿ Detailed sensitivity testing, e.g.:
— Change in cap rate
— Insurance unavailability for specific
hazard like fire, or in a specific state
Climate Risk
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 23
Case study: Using a phased framework to develop bespoke scenario
impact assessment tool for CRE portfolios (2/2)
Sanitized example from US based CRE portfolio of a large bank #1
100
140
2040
9
2030
Current
23
8
2050 Total relative
EL by 2050
Climate impact to EL arranged by deciles
70%
40%
10%
0
20% 60%
0.2
50%
30% 80% 90% 100%
Ø 4.6 0
43.0
1.6 0.6 0.3 0.1 0.1 0
Average
indexed
increase
in EL
100
334
125
72
Total relative
EL by 2050
37
Current 2030 2040 2050
3 2 3
10
5
4
2
Florida
0
1
Ohio
5
New York
1
14
9
California
5
0 1
Uninsured damage
Increase in insurance cost
Uninsured business interruption
Reduced demand for properties
exposed to climate risk
Transition risk & Heat stress
Net Cash Flow decline by transmission channel
With insurance
Without insurance
Evolution of climate shocked Expected Loss
Scenario : Delayed Transition
Evolution of climate shocked Expected Loss
Scenario : Delayed Transition
EL increase by year (starting EL indexed to 100)
Climate Risk
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 24
Deposits data can be used to create an up-to-the-minute estimate of
SMB’s financials
Strategic implications
Risk signals extracted from transactions
Estimated revenues and profits
Business seasonality
Sectoral dependencies
SME financial
position
Multi-banking clients
Loans with other lenders
Interest rate changes by other
lenders
Fintech relationships
(e.g. Paypal)
Competitive
information
Deposits transactions can be analyzed
using a transaction classifier to derive
an estimate of
Ÿ SMB’s financials (e.g. revenues,
revenue growth, and profits)
Using a reinforcement learning and
natural language processing, deposits
transactions are analyzed to provide
structure to unstructured data
In a client application, this method yield
95% transaction classification
accuracy and over 300 risk signals and
competitive indicators, resulting in
substantial improvement in predictive
power of credit risk models
Competition to become the
primary bank will intensify
In turn, banks that host the
deposit account where the
payroll is deposited have a
competitive advantage
Deposit accounts are already
increasingly being bundled
with other financial products
Fees for deposit accounts are
expected to reduce further
Context
Next generation data analytics
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 25
Case example: Leveraging transaction data to transform
credit decisioning
Up to 40%
Accuracy of financial
inflows and outflows
from classifier
95%
Predictive risk signals
– linked to
transaction
classifications
>300
Traditional lending process
More sophisticated understanding
of risks
More customer insights
More predictive power
Reimagined credit decisioning
additional predictive
power in some
subsegments
Illustrative; Client example
Customer provides bank account and
provides simple historic financial
statements
ETB: Transactional data from own
system, used to build synthetic
financial statements
Customer completes credit
application form
Customer completes online application
(pre-populated, leveraging API enabled
data sources)
NTB: Customer provides permission1
to access bank transactions data to
build synthetic financial statements
RM manually reviews data and
performs credit assessment (scoring
model, uses credit bureau, and
historic/ narrow financial data)
Algorithm integrates data and performs
credit assessment (e.g., leveraging
transaction classier to build financial
statements, link to risk signals)
Manual decision communicated to
customer
Decision communicated to customer
(instant time yes)
Next generation data analytics
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 26
AI4DQ
An AI-driven asset deployment driven approach can accelerate data
quality improvement significantly
1 AI toolkit for data 2 Deployment accelerators 3 Training modules
Detection: Ready to implement
package to assess data quality
Correction: Relationship discovery
and anomaly detection to find errors
Repair: AI driven correction through
an open-loop process
Deployable as pipelines that can be
stand-alone for immediate results
and integrated into Data Platforms to
continuously monitor and improve
data quality (e.g., Apache Airflow
integration with Collibra), platform
agnostic deployment
A new way of working, including
roles, talent, an a fast-paced Agile
operating model
Co-development of solutions through
build-operate-transfer to sustain the
impact
When implemented
together these 3
components significantly
accelerate data quality
capabilities
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
AI driven data quality
McKinsey & Company 27
Data Quality can be addressed through modularized tools in three
core areas
Detection Correction Repair
How machine
learning helps
Automatic generation of reports on the profile
of data, inferred relationships between tables
and anomalies, root cause identification to
prioritize upstream interventions
Rule-mining and clustering algorithms
recommend corrections to data quality errors,
quantify confidence, and help estimate
business impact
Validate corrections with experts (open-loop)
confirming only lower confidence
recommendations manually and automated
approval of validated changes
Illustration
Detect top 1th percentile interest rate,
unusually late maturity date as a potential
error
Correct loan type with attribution, country of
origination, and address with 95% confidence
Flag and automatically fix reporting dates that
were corrected by experts repeated
Objectives Measure the quality of each record on a
scale of Low, Medium, High data quality to
help prioritize remediation
Recommend corrections to data quality
errors prioritized by business impact and
measure the confidence (e.g., 95%)
Validate recommendations with experts
and incorporate changes into underlying data
systems feeding reports
Ready-to-
deploy tools
Root cause identification Comparison to 3rd party data
Data relationship discovery Automated DQ rule generation Collaborative workflows
Anomality detection
Entity disconnect identification Free-form text corrections Programmatic data validation
Corrections that integrate attribute,
record, and database signals
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
AI driven data quality
McKinsey & Company 28
High impact examples AI being used to improve data quality
Automated
identification of errors
in CRE loan data
Root cause detection
and correction to
errors in 5M customer
accounts
Accelerating an
enterprise data
transformation with AI
Corrected 5M free-form
occupations to accelerate
AML Customer Risk Rating
using a neural language
model and traditional fuzzy
matching
Automated data quality detection
and correction with AI (human-in-
the-loop) to reduce data
transformation timeline by 30-40
percent
Automated detection of data
quality errors in over 100,000
loans using time series anomaly
detection to detect issues in real
time and prioritize remediation
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
AI driven data quality
McKinsey & Company 29
Scenario planning forecasting approach needs to be tailored to
bank’s footprint flexible to incorporate different driving factors
Full estimates from business drivers
and optimize across BUs
Interest Income
Credit Loss
…
Canada
UK
US
B2
B1
B3
A3
A1
B4
A4
A2
B5
Scenarios from underlying reasons
…
Canada
UK
US
Illustrative
Develop inflation scenarios based on set of
underlying reasons (e.g., supply chain
bottleneck leading to ‘cost-push’ inflation) and
consider incorporation of important macro-
linkages (e.g., currency fluctuation risk)
Business drivers from scenarios
Identify drivers that are likely to be inflation-
sensitive for business portfolios / geographies,
and direction of impact – e.g., inflation may drive
up transaction volume in the short term;
however, inflation may also reduce demand in
the long term
Develop analytics to project the underlying
drivers and the business portfolio financials
Based on projection results, synthesize
implications for strategic decision-making
Scenario Planning and Forecasting
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 30
Uncertainty in economic path require a forecasting approach that is
nimble, flexible, and responsive to emerging risks like inflation
Ÿ Model and analytics developed for relevant
drivers and separately for relevant geography
Ÿ Models capture time trend, macroeconomic
factors, seasonality and latent portfolio factors
in a transparent manner
Ÿ Designed to capture, test and visualize
business actions under different macro-
conditions
Ÿ Structured to generate report on portfolio and
sub-segment level with different ‘what-if’
Ÿ Flexibility to expand current internal scenarios
with new variables
Ÿ Ability to add macro and event driven overlays
Ÿ Ability to refresh scenarios weekly vs. monthly
Versatility
Coherence
Flexibility
Scenario Generation and Refinement
Customized scenarios for portfolio/segments
Ÿ Can be macro-economic or event driven
Ÿ Recent macro-trends (e.g., EPOP ratio) core to design
Business driver forecast through models
Volume, Revenue, and Expense Forecast Models tied to business
drivers (e.g., line utilization) and scenario conditioned
Implementation Engine
One-click solution for aggregation, reporting and visualization
Scenario Planning and Forecasting
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 31
In addition to standard capital or liquidity constraints, exploring
emissions related constraints can help in portfolio alignment
Capital consumption
Carbon emission
Preliminary list of constraints incorporated in the
approach (to be refined based on observations
during design phase)
Product Industry
Carbon
limit
Capital
limit
Approach to incorporate emissions related constraints
BU
Example output by industry at a loan level
Carbon consumption
Economic return
Energy Materials Telecom
Capital
Risk Weighted Assets
Expected Loss
Origination/balance growth
Liquidity coverage ratio
Emissions
Ÿ Include carbon limit based on benchmark scenario
at North America with relevant downscaling to
Canada/Alberta
Ÿ Add constraint for net zero target of total portfolio
emissions
Ÿ Develop functionality to add sector specific targets
and connect with potential sector-specific carbon
intensity metric
Loans out Loans in
Scenario Planning and Forecasting
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 32
Microsegment based forecast models are coded in engine with a
Balance Sheet Optimizer added for ongoing scenario planning
Node 2
Node n
Node 1 Node 1 O/P
Node 2 O/P
Scenario File
Input
internal data
Config File
Overlay File
.
.
.
.
.
.
Node n O/P
Input Pre-processing (e.g., Python)
Run pre-processing code in Python
e.g., overlays, variable transformation
Prepare input
files and data
Data
aggregation
and Optimizer
Aggregation
Aggregate
predictions,
conduct sensitivity
Cashflow Engine
Cashflow calculator
(e.g., Spark)
Run microsegment
level calculations
Strategic
planning
FP & A
Add-on modules
Feed output to
additional modules
per specific use case
Client mgmt.
Architecture
tailored based
on level of
manual
intervention
desired, and
size of input
data
Example forecasting engine architecture for volume and revenue estimation
Scenario Planning and Forecasting
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 33
Content Summary of survey results 20 mins
Perspectives on selected topics 30 mins
Questions 10 mins
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited
McKinsey & Company 34
Questions
CONFIDENTIAL AND PROPRIETARY
Any use of this material without specific permission of McKinsey & Company is strictly prohibited

More Related Content

What's hot

Cloud value in cash management
Cloud value in cash managementCloud value in cash management
Cloud value in cash management
McKinsey & Company
 
Lifting the Barriers to Retail Innovation in ASEAN | A.T. Kearney
Lifting the Barriers to Retail Innovation in ASEAN | A.T. KearneyLifting the Barriers to Retail Innovation in ASEAN | A.T. Kearney
Lifting the Barriers to Retail Innovation in ASEAN | A.T. Kearney
Kearney
 
Seven Forces Reshaping Enterprise Software
Seven Forces Reshaping Enterprise SoftwareSeven Forces Reshaping Enterprise Software
Seven Forces Reshaping Enterprise Software
Boston Consulting Group
 
18th Annual Global CEO Survey - Technology industry key findings
18th Annual Global CEO Survey - Technology industry key findings18th Annual Global CEO Survey - Technology industry key findings
18th Annual Global CEO Survey - Technology industry key findings
PwC
 
PwC’s new Golden Age Index – how well are countries harnessing the power of o...
PwC’s new Golden Age Index – how well are countries harnessing the power of o...PwC’s new Golden Age Index – how well are countries harnessing the power of o...
PwC’s new Golden Age Index – how well are countries harnessing the power of o...
PwC
 
The Decade to Deliver: A Call to Business Action
The Decade to Deliver: A Call to Business Action The Decade to Deliver: A Call to Business Action
The Decade to Deliver: A Call to Business Action
accenture
 
2018 Brand Owner Packaging Survey
2018 Brand Owner Packaging Survey2018 Brand Owner Packaging Survey
2018 Brand Owner Packaging Survey
L.E.K. Consulting
 
Unleashing Competitiveness on the Cloud Continuum | Accenture
Unleashing Competitiveness on the Cloud Continuum | AccentureUnleashing Competitiveness on the Cloud Continuum | Accenture
Unleashing Competitiveness on the Cloud Continuum | Accenture
accenture
 
Global Capital Confidence Barometer 21st edition
Global Capital Confidence Barometer 21st editionGlobal Capital Confidence Barometer 21st edition
Global Capital Confidence Barometer 21st edition
EY
 
2019 CPG Growth Leaders Report
2019 CPG Growth Leaders Report2019 CPG Growth Leaders Report
2019 CPG Growth Leaders Report
Boston Consulting Group
 
MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...
MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...
MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...
EY
 
2018 Local Dynamos: Emerging-Market Companies Up Their Game
2018 Local Dynamos: Emerging-Market Companies Up Their Game2018 Local Dynamos: Emerging-Market Companies Up Their Game
2018 Local Dynamos: Emerging-Market Companies Up Their Game
Boston Consulting Group
 
IP Theft
IP TheftIP Theft
2016 Strategic Hospital Priorities Study
2016 Strategic Hospital Priorities Study2016 Strategic Hospital Priorities Study
2016 Strategic Hospital Priorities Study
L.E.K. Consulting
 
Tech Adoption and Strategy for Innovation & Growth
Tech Adoption and Strategy for Innovation & GrowthTech Adoption and Strategy for Innovation & Growth
Tech Adoption and Strategy for Innovation & Growth
accenture
 
Shaping the Sustainable Organization | Accenture
Shaping the Sustainable Organization | AccentureShaping the Sustainable Organization | Accenture
Shaping the Sustainable Organization | Accenture
accenture
 
Strategy Study 2014 | A.T. Kearney
Strategy Study 2014 | A.T. KearneyStrategy Study 2014 | A.T. Kearney
Strategy Study 2014 | A.T. Kearney
Kearney
 
Power transactions and trends Q2 2019
Power transactions and trends Q2 2019Power transactions and trends Q2 2019
Power transactions and trends Q2 2019
EY
 
A.T. Kearney Consolidation of the US Banking Industry
A.T. Kearney Consolidation of the US Banking IndustryA.T. Kearney Consolidation of the US Banking Industry
A.T. Kearney Consolidation of the US Banking Industry
Kearney
 
Pursuing Customer Inspired Growth
Pursuing Customer Inspired GrowthPursuing Customer Inspired Growth
Pursuing Customer Inspired Growth
Kearney
 

What's hot (20)

Cloud value in cash management
Cloud value in cash managementCloud value in cash management
Cloud value in cash management
 
Lifting the Barriers to Retail Innovation in ASEAN | A.T. Kearney
Lifting the Barriers to Retail Innovation in ASEAN | A.T. KearneyLifting the Barriers to Retail Innovation in ASEAN | A.T. Kearney
Lifting the Barriers to Retail Innovation in ASEAN | A.T. Kearney
 
Seven Forces Reshaping Enterprise Software
Seven Forces Reshaping Enterprise SoftwareSeven Forces Reshaping Enterprise Software
Seven Forces Reshaping Enterprise Software
 
18th Annual Global CEO Survey - Technology industry key findings
18th Annual Global CEO Survey - Technology industry key findings18th Annual Global CEO Survey - Technology industry key findings
18th Annual Global CEO Survey - Technology industry key findings
 
PwC’s new Golden Age Index – how well are countries harnessing the power of o...
PwC’s new Golden Age Index – how well are countries harnessing the power of o...PwC’s new Golden Age Index – how well are countries harnessing the power of o...
PwC’s new Golden Age Index – how well are countries harnessing the power of o...
 
The Decade to Deliver: A Call to Business Action
The Decade to Deliver: A Call to Business Action The Decade to Deliver: A Call to Business Action
The Decade to Deliver: A Call to Business Action
 
2018 Brand Owner Packaging Survey
2018 Brand Owner Packaging Survey2018 Brand Owner Packaging Survey
2018 Brand Owner Packaging Survey
 
Unleashing Competitiveness on the Cloud Continuum | Accenture
Unleashing Competitiveness on the Cloud Continuum | AccentureUnleashing Competitiveness on the Cloud Continuum | Accenture
Unleashing Competitiveness on the Cloud Continuum | Accenture
 
Global Capital Confidence Barometer 21st edition
Global Capital Confidence Barometer 21st editionGlobal Capital Confidence Barometer 21st edition
Global Capital Confidence Barometer 21st edition
 
2019 CPG Growth Leaders Report
2019 CPG Growth Leaders Report2019 CPG Growth Leaders Report
2019 CPG Growth Leaders Report
 
MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...
MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...
MAPS2018 Keynote address on EY report: Life Sciences 4.0 – Securing value thr...
 
2018 Local Dynamos: Emerging-Market Companies Up Their Game
2018 Local Dynamos: Emerging-Market Companies Up Their Game2018 Local Dynamos: Emerging-Market Companies Up Their Game
2018 Local Dynamos: Emerging-Market Companies Up Their Game
 
IP Theft
IP TheftIP Theft
IP Theft
 
2016 Strategic Hospital Priorities Study
2016 Strategic Hospital Priorities Study2016 Strategic Hospital Priorities Study
2016 Strategic Hospital Priorities Study
 
Tech Adoption and Strategy for Innovation & Growth
Tech Adoption and Strategy for Innovation & GrowthTech Adoption and Strategy for Innovation & Growth
Tech Adoption and Strategy for Innovation & Growth
 
Shaping the Sustainable Organization | Accenture
Shaping the Sustainable Organization | AccentureShaping the Sustainable Organization | Accenture
Shaping the Sustainable Organization | Accenture
 
Strategy Study 2014 | A.T. Kearney
Strategy Study 2014 | A.T. KearneyStrategy Study 2014 | A.T. Kearney
Strategy Study 2014 | A.T. Kearney
 
Power transactions and trends Q2 2019
Power transactions and trends Q2 2019Power transactions and trends Q2 2019
Power transactions and trends Q2 2019
 
A.T. Kearney Consolidation of the US Banking Industry
A.T. Kearney Consolidation of the US Banking IndustryA.T. Kearney Consolidation of the US Banking Industry
A.T. Kearney Consolidation of the US Banking Industry
 
Pursuing Customer Inspired Growth
Pursuing Customer Inspired GrowthPursuing Customer Inspired Growth
Pursuing Customer Inspired Growth
 

Similar to McKinsey_Data_and_analytics_innovation_inCPM.pdf

Foundry Data & Analytics Study 2021
Foundry Data & Analytics Study 2021Foundry Data & Analytics Study 2021
Foundry Data & Analytics Study 2021
IDG
 
Views From The C-Suite: Who's Big on Big Data
Views From The C-Suite: Who's Big on Big DataViews From The C-Suite: Who's Big on Big Data
Views From The C-Suite: Who's Big on Big Data
Platfora
 
Who's big on big data?
Who's big on big data?Who's big on big data?
Who's big on big data?
The Economist Media Businesses
 
Data & Analytics Sample Slides_NEW.pdf
Data & Analytics Sample Slides_NEW.pdfData & Analytics Sample Slides_NEW.pdf
Data & Analytics Sample Slides_NEW.pdf
IDG
 
Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...
Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...
Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...
CGAP
 
Infrastructure that can stand the test of time | Accenture
Infrastructure that can stand the test of time | AccentureInfrastructure that can stand the test of time | Accenture
Infrastructure that can stand the test of time | Accenture
accenture
 
The Relevance of Data Analytics in External Audit.pdf
The Relevance of Data Analytics in External Audit.pdfThe Relevance of Data Analytics in External Audit.pdf
The Relevance of Data Analytics in External Audit.pdf
Fiyona Nourin
 
big data analysis concepts by dr. lahe
big data analysis  concepts  by dr. lahebig data analysis  concepts  by dr. lahe
big data analysis concepts by dr. lahe
YaseenMo
 
CIO Tech Priorities 2023_Sample Slides.pdf
CIO Tech Priorities 2023_Sample Slides.pdfCIO Tech Priorities 2023_Sample Slides.pdf
CIO Tech Priorities 2023_Sample Slides.pdf
IDG
 
Value creation with big data analytics for enterprises: a survey
Value creation with big data analytics for enterprises: a surveyValue creation with big data analytics for enterprises: a survey
Value creation with big data analytics for enterprises: a survey
TELKOMNIKA JOURNAL
 
3 Strategies to drive more data driven outcomes in financial services
3 Strategies to drive more data driven outcomes in financial services3 Strategies to drive more data driven outcomes in financial services
3 Strategies to drive more data driven outcomes in financial services
TamrMarketing
 
Analytic Transformation | 2013 Loras College Business Analytics Symposium
Analytic Transformation | 2013 Loras College Business Analytics SymposiumAnalytic Transformation | 2013 Loras College Business Analytics Symposium
Analytic Transformation | 2013 Loras College Business Analytics Symposium
Cartegraph
 
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET Journal
 
FPA-Trends-Survey-2022-SAP.pdf
FPA-Trends-Survey-2022-SAP.pdfFPA-Trends-Survey-2022-SAP.pdf
FPA-Trends-Survey-2022-SAP.pdf
James Myers
 
How To Biuld Internal Rating System For Basel Ii
How To Biuld Internal Rating System For Basel IiHow To Biuld Internal Rating System For Basel Ii
How To Biuld Internal Rating System For Basel Ii
FNian
 
The Value of Predictive Analytics and Decision Modeling
The Value of Predictive Analytics and Decision ModelingThe Value of Predictive Analytics and Decision Modeling
The Value of Predictive Analytics and Decision Modeling
Decision Management Solutions
 
Loan Default Prediction Using Machine Learning Techniques
Loan Default Prediction Using Machine Learning TechniquesLoan Default Prediction Using Machine Learning Techniques
Loan Default Prediction Using Machine Learning Techniques
IRJET Journal
 
What Are the Challenges and Opportunities in Big Data Analytics.pdf
What Are the Challenges and Opportunities in Big Data Analytics.pdfWhat Are the Challenges and Opportunities in Big Data Analytics.pdf
What Are the Challenges and Opportunities in Big Data Analytics.pdf
Mr. Business Magazine
 
Supervised and unsupervised data mining approaches in loan default prediction
Supervised and unsupervised data mining approaches in loan default prediction Supervised and unsupervised data mining approaches in loan default prediction
Supervised and unsupervised data mining approaches in loan default prediction
IJECEIAES
 
Data mining on Financial Data
Data mining on Financial DataData mining on Financial Data
Data mining on Financial Data
AmarnathVenkataraman
 

Similar to McKinsey_Data_and_analytics_innovation_inCPM.pdf (20)

Foundry Data & Analytics Study 2021
Foundry Data & Analytics Study 2021Foundry Data & Analytics Study 2021
Foundry Data & Analytics Study 2021
 
Views From The C-Suite: Who's Big on Big Data
Views From The C-Suite: Who's Big on Big DataViews From The C-Suite: Who's Big on Big Data
Views From The C-Suite: Who's Big on Big Data
 
Who's big on big data?
Who's big on big data?Who's big on big data?
Who's big on big data?
 
Data & Analytics Sample Slides_NEW.pdf
Data & Analytics Sample Slides_NEW.pdfData & Analytics Sample Slides_NEW.pdf
Data & Analytics Sample Slides_NEW.pdf
 
Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...
Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...
Projecting Impact of Non-Traditional Data and Advanced Analytics on Delivery ...
 
Infrastructure that can stand the test of time | Accenture
Infrastructure that can stand the test of time | AccentureInfrastructure that can stand the test of time | Accenture
Infrastructure that can stand the test of time | Accenture
 
The Relevance of Data Analytics in External Audit.pdf
The Relevance of Data Analytics in External Audit.pdfThe Relevance of Data Analytics in External Audit.pdf
The Relevance of Data Analytics in External Audit.pdf
 
big data analysis concepts by dr. lahe
big data analysis  concepts  by dr. lahebig data analysis  concepts  by dr. lahe
big data analysis concepts by dr. lahe
 
CIO Tech Priorities 2023_Sample Slides.pdf
CIO Tech Priorities 2023_Sample Slides.pdfCIO Tech Priorities 2023_Sample Slides.pdf
CIO Tech Priorities 2023_Sample Slides.pdf
 
Value creation with big data analytics for enterprises: a survey
Value creation with big data analytics for enterprises: a surveyValue creation with big data analytics for enterprises: a survey
Value creation with big data analytics for enterprises: a survey
 
3 Strategies to drive more data driven outcomes in financial services
3 Strategies to drive more data driven outcomes in financial services3 Strategies to drive more data driven outcomes in financial services
3 Strategies to drive more data driven outcomes in financial services
 
Analytic Transformation | 2013 Loras College Business Analytics Symposium
Analytic Transformation | 2013 Loras College Business Analytics SymposiumAnalytic Transformation | 2013 Loras College Business Analytics Symposium
Analytic Transformation | 2013 Loras College Business Analytics Symposium
 
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data Mining
 
FPA-Trends-Survey-2022-SAP.pdf
FPA-Trends-Survey-2022-SAP.pdfFPA-Trends-Survey-2022-SAP.pdf
FPA-Trends-Survey-2022-SAP.pdf
 
How To Biuld Internal Rating System For Basel Ii
How To Biuld Internal Rating System For Basel IiHow To Biuld Internal Rating System For Basel Ii
How To Biuld Internal Rating System For Basel Ii
 
The Value of Predictive Analytics and Decision Modeling
The Value of Predictive Analytics and Decision ModelingThe Value of Predictive Analytics and Decision Modeling
The Value of Predictive Analytics and Decision Modeling
 
Loan Default Prediction Using Machine Learning Techniques
Loan Default Prediction Using Machine Learning TechniquesLoan Default Prediction Using Machine Learning Techniques
Loan Default Prediction Using Machine Learning Techniques
 
What Are the Challenges and Opportunities in Big Data Analytics.pdf
What Are the Challenges and Opportunities in Big Data Analytics.pdfWhat Are the Challenges and Opportunities in Big Data Analytics.pdf
What Are the Challenges and Opportunities in Big Data Analytics.pdf
 
Supervised and unsupervised data mining approaches in loan default prediction
Supervised and unsupervised data mining approaches in loan default prediction Supervised and unsupervised data mining approaches in loan default prediction
Supervised and unsupervised data mining approaches in loan default prediction
 
Data mining on Financial Data
Data mining on Financial DataData mining on Financial Data
Data mining on Financial Data
 

Recently uploaded

Satta matka guessing Kalyan result sattamatka
Satta matka guessing Kalyan result sattamatkaSatta matka guessing Kalyan result sattamatka
Satta matka guessing Kalyan result sattamatka
➑➌➋➑➒➎➑➑➊➍
 
Best Practices in Negotiation for Business Management Students
Best Practices in Negotiation for Business Management StudentsBest Practices in Negotiation for Business Management Students
Best Practices in Negotiation for Business Management Students
zoyaws1
 
Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...
Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...
Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...
Lynch Creek Farm
 
A Complete Guide of Dubai Freelance Visa and Permit in 2024
A Complete Guide of Dubai Freelance Visa and Permit in 2024A Complete Guide of Dubai Freelance Visa and Permit in 2024
A Complete Guide of Dubai Freelance Visa and Permit in 2024
Dubiz
 
Transforming the Future of Limo Services.pptx
Transforming the Future of Limo Services.pptxTransforming the Future of Limo Services.pptx
Transforming the Future of Limo Services.pptx
limocaptaincom
 
PETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAA
PETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAAPETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAA
PETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAA
lawrenceads01
 
Growth Buyouts - The Dawn of the GBO (Slow Ventures)
Growth Buyouts - The  Dawn of the GBO (Slow Ventures)Growth Buyouts - The  Dawn of the GBO (Slow Ventures)
Growth Buyouts - The Dawn of the GBO (Slow Ventures)
Razin Mustafiz
 
Case study on Indian Ecommerce logistics
Case study on Indian Ecommerce logisticsCase study on Indian Ecommerce logistics
Case study on Indian Ecommerce logistics
UnheardShayari
 
Corporate Governance for South African Mining Companies
Corporate Governance for South African Mining CompaniesCorporate Governance for South African Mining Companies
Corporate Governance for South African Mining Companies
James AH Campbell
 
foodgasm restaurant and Bar pune road.docx
foodgasm restaurant and Bar pune road.docxfoodgasm restaurant and Bar pune road.docx
foodgasm restaurant and Bar pune road.docx
PraghyaBhandari
 
Mandated reporting powerpoint to help with understanding your role
Mandated reporting powerpoint to help with understanding your roleMandated reporting powerpoint to help with understanding your role
Mandated reporting powerpoint to help with understanding your role
khidalgo2
 
A STUDY OF MUTUAL FUND OF BANK OF INDIA .pdf
A STUDY OF MUTUAL FUND  OF BANK OF INDIA .pdfA STUDY OF MUTUAL FUND  OF BANK OF INDIA .pdf
A STUDY OF MUTUAL FUND OF BANK OF INDIA .pdf
rsonics22
 
NewBase 05 July 2024 Energy News issue - 1736 by Khaled Al Awadi_compresse...
NewBase   05 July 2024  Energy News issue - 1736 by Khaled Al Awadi_compresse...NewBase   05 July 2024  Energy News issue - 1736 by Khaled Al Awadi_compresse...
NewBase 05 July 2024 Energy News issue - 1736 by Khaled Al Awadi_compresse...
Khaled Al Awadi
 
High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...
High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...
High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...
dimplekumaridk322
 
brojjeddah Home Services Company in Saudi Arabia
brojjeddah Home Services Company in Saudi Arabiabrojjeddah Home Services Company in Saudi Arabia
brojjeddah Home Services Company in Saudi Arabia
brojjeddah
 
TALENT ACQUISITION AND MANAGEMENT LECTURE 2
TALENT ACQUISITION AND MANAGEMENT LECTURE 2TALENT ACQUISITION AND MANAGEMENT LECTURE 2
TALENT ACQUISITION AND MANAGEMENT LECTURE 2
projectseasy
 
BBA Final SML 501 INTERNATIONAL BUSINESS .pdf
BBA Final SML 501 INTERNATIONAL BUSINESS .pdfBBA Final SML 501 INTERNATIONAL BUSINESS .pdf
BBA Final SML 501 INTERNATIONAL BUSINESS .pdf
mcdopex6
 
AI at Work​ The demystification of AI and real-world stories on how to apply ...
AI at Work​ The demystification of AI and real-world stories on how to apply ...AI at Work​ The demystification of AI and real-world stories on how to apply ...
AI at Work​ The demystification of AI and real-world stories on how to apply ...
Auxis Consulting & Outsourcing
 
Cheslyn Jacobs- TymeBank: Building Consumer Trust in Digital Banking
Cheslyn Jacobs- TymeBank: Building Consumer Trust in Digital  BankingCheslyn Jacobs- TymeBank: Building Consumer Trust in Digital  Banking
Cheslyn Jacobs- TymeBank: Building Consumer Trust in Digital Banking
itnewsafrica
 
Pricing sophistication - auto insurance telematics
Pricing sophistication - auto insurance telematicsPricing sophistication - auto insurance telematics
Pricing sophistication - auto insurance telematics
Matteo Carbone
 

Recently uploaded (20)

Satta matka guessing Kalyan result sattamatka
Satta matka guessing Kalyan result sattamatkaSatta matka guessing Kalyan result sattamatka
Satta matka guessing Kalyan result sattamatka
 
Best Practices in Negotiation for Business Management Students
Best Practices in Negotiation for Business Management StudentsBest Practices in Negotiation for Business Management Students
Best Practices in Negotiation for Business Management Students
 
Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...
Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...
Christmas Decorations_ A Guide to Small Christmas Trees, Candle Centerpieces,...
 
A Complete Guide of Dubai Freelance Visa and Permit in 2024
A Complete Guide of Dubai Freelance Visa and Permit in 2024A Complete Guide of Dubai Freelance Visa and Permit in 2024
A Complete Guide of Dubai Freelance Visa and Permit in 2024
 
Transforming the Future of Limo Services.pptx
Transforming the Future of Limo Services.pptxTransforming the Future of Limo Services.pptx
Transforming the Future of Limo Services.pptx
 
PETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAA
PETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAAPETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAA
PETAVIT SIP-05.pdfAAAAAAAAAAAAAAAAAAAAAAAAA
 
Growth Buyouts - The Dawn of the GBO (Slow Ventures)
Growth Buyouts - The  Dawn of the GBO (Slow Ventures)Growth Buyouts - The  Dawn of the GBO (Slow Ventures)
Growth Buyouts - The Dawn of the GBO (Slow Ventures)
 
Case study on Indian Ecommerce logistics
Case study on Indian Ecommerce logisticsCase study on Indian Ecommerce logistics
Case study on Indian Ecommerce logistics
 
Corporate Governance for South African Mining Companies
Corporate Governance for South African Mining CompaniesCorporate Governance for South African Mining Companies
Corporate Governance for South African Mining Companies
 
foodgasm restaurant and Bar pune road.docx
foodgasm restaurant and Bar pune road.docxfoodgasm restaurant and Bar pune road.docx
foodgasm restaurant and Bar pune road.docx
 
Mandated reporting powerpoint to help with understanding your role
Mandated reporting powerpoint to help with understanding your roleMandated reporting powerpoint to help with understanding your role
Mandated reporting powerpoint to help with understanding your role
 
A STUDY OF MUTUAL FUND OF BANK OF INDIA .pdf
A STUDY OF MUTUAL FUND  OF BANK OF INDIA .pdfA STUDY OF MUTUAL FUND  OF BANK OF INDIA .pdf
A STUDY OF MUTUAL FUND OF BANK OF INDIA .pdf
 
NewBase 05 July 2024 Energy News issue - 1736 by Khaled Al Awadi_compresse...
NewBase   05 July 2024  Energy News issue - 1736 by Khaled Al Awadi_compresse...NewBase   05 July 2024  Energy News issue - 1736 by Khaled Al Awadi_compresse...
NewBase 05 July 2024 Energy News issue - 1736 by Khaled Al Awadi_compresse...
 
High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...
High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...
High Profile Girls Call Bhubaneswar 🎈🔥000XX00000 🔥💋🎈 Provide Best And Top Gir...
 
brojjeddah Home Services Company in Saudi Arabia
brojjeddah Home Services Company in Saudi Arabiabrojjeddah Home Services Company in Saudi Arabia
brojjeddah Home Services Company in Saudi Arabia
 
TALENT ACQUISITION AND MANAGEMENT LECTURE 2
TALENT ACQUISITION AND MANAGEMENT LECTURE 2TALENT ACQUISITION AND MANAGEMENT LECTURE 2
TALENT ACQUISITION AND MANAGEMENT LECTURE 2
 
BBA Final SML 501 INTERNATIONAL BUSINESS .pdf
BBA Final SML 501 INTERNATIONAL BUSINESS .pdfBBA Final SML 501 INTERNATIONAL BUSINESS .pdf
BBA Final SML 501 INTERNATIONAL BUSINESS .pdf
 
AI at Work​ The demystification of AI and real-world stories on how to apply ...
AI at Work​ The demystification of AI and real-world stories on how to apply ...AI at Work​ The demystification of AI and real-world stories on how to apply ...
AI at Work​ The demystification of AI and real-world stories on how to apply ...
 
Cheslyn Jacobs- TymeBank: Building Consumer Trust in Digital Banking
Cheslyn Jacobs- TymeBank: Building Consumer Trust in Digital  BankingCheslyn Jacobs- TymeBank: Building Consumer Trust in Digital  Banking
Cheslyn Jacobs- TymeBank: Building Consumer Trust in Digital Banking
 
Pricing sophistication - auto insurance telematics
Pricing sophistication - auto insurance telematicsPricing sophistication - auto insurance telematics
Pricing sophistication - auto insurance telematics
 

McKinsey_Data_and_analytics_innovation_inCPM.pdf

  • 1. CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited March, 2022 Discussion Document Data and analytics innovations in Credit Portfolio Management
  • 2. 2 McKinsey and IACPM together completed a survey on new developments in data and analytics for credit portfolio management The survey had 3 main objectives Survey range of industry practice on: Ÿ Emergence of alternative data sources for credit risk identification, assessment, and monitoring Ÿ Related evolution of analytics tools Understand degree to which different data types and analytical approaches are in use/under consideration across institutions, geographies, subsectors, and CRE/C&I portfolio segments Develop insights on current state and path forward for participants to incorporate next generation data and analytics CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 3. McKinsey & Company 3 44 financial institutions participated in the survey from across Americas, APAC, and EMEA Asia Pacific Americas Europe Middle East and Africa Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 4. McKinsey & Company 4 Content Summary of survey results 20 mins Perspectives on selected topics 30 mins Questions 10 mins CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 5. McKinsey & Company Summary of results D Investment and Strategic Goals Top investment areas have been data tech and data acquisition. Over the next 2 years, participants expect a greater role of innovative data and advanced analytics in improving Credit Strategy and Client Experience A Trends Over the next 2 years, larger number of participants expect significant increase in the use of internally developed advanced techniques and new types of external data F Impact Use of innovative data and/or advanced analytics improves model accuracy , turn-around-time, automated decisioning and time spend on analysis, with higher benefit observed in the SME segment E Use cases Machine learning models are primarily gaining traction for Risk Scoring of SMEs and Early Warning across the board Innovative external data sources are more used for Corporate segment while SME segment uses more innovative internal data sources B Challenges Data quality assessment and talent management are the top challenges for use of both advanced analytics and innovative data solution C Climate Majority of the participants believe that Climate and ESG is the next biggest challenge for credit assessment Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 6. McKinsey & Company 6 A: Over the next 2 years, a vast majority of participants expect significant increase in the use of new types of external data and internally developed advanced techniques Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 Expectation for the next 2 years In the past 2 years How has your firm’s data/analytics for credit decisions changed in the past 2 years and how do you expect it to change in the coming 2 years? In the past 2 years, over 60% of the participants have seen an increase in the: Ÿ Use of new types of internal and external data Ÿ Use of internally developed advanced techniques Ÿ Size of data and analytics team Over the next 2 years, even larger % of participants expect this trend to continue % participants see increase in trends 72% 74% Increased use of new types of internal data Increased use of internally developed machine learning and other advanced analytics techniques Increased use of vendor-developed machine learning and other advanced analytics techniques Increased use of new types external data Change in the size of the data & analytics team Change in share of automated credit approvals 66% 64% 49% 32% 91% 81% 59% 79% 81% 66% CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 7. McKinsey & Company 7 B: Data quality assessment and talent management are the top challenges for use of both advanced analytics and innovative data solution 42 49 14 26 26 40 9 42 7 2 5 63 42 30 28 28 19 14 14 12 2 2 Validation difficulty Cost of data Data Quality Ability to explain Regulatory Resources Skepticism/Probability Unmet expectations Risk concerns Other Fragmentation Currently, what are the major challenges faced by your firm that constrain the use of innovative data or advanced analytics (e.g., machine learning and AI)? Top 3 challenges for use of innovative data solutions Top 3 challenges for use of advanced analytics solutions Percentage Major challenges for use of advanced analytics solutions are: Ÿ Attract, retain and develop resources Ÿ Ability to explain Ÿ Data quality Ÿ Validation While for using innovative data, key challenge in data quality assessment and talent Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 8. McKinsey & Company 8 C: 86% of the participants believe that Climate and ESG is the next biggest challenge for credit assessment 86 58 51 42 9 7 Post COVID-19 model adjustments and uncertainties Capital, Provisioning, or regulatory stress testing model requirements Climate and ESG Incorporating Machine Learning models within regulatory and risk constraints Effectively competing with attackers Other – please specify 27: What are the biggest challenges facing credit risk and credit portfolio management analytics in the next 2- 3 years? Percentage 25: For incorporating the impact of climate risk ,are you using existing loss models with climate shock applied to input variables? Or are you developing new loss models to assess it? Percentage 86% of the participants believe climate and ESG is the next biggest challenge for credit assessment, followed by COVID-19 adjustments, capital and regulatory exercise, and using machine learning models One third of the participants plan to use existing credit models to translate the climate impact to credit risk and another one third of the participants plan to develop new loss models for climate assessment 35 33 33 Using existing models Developing new loss models Not yet exploring Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 9. McKinsey & Company 9 12% Yes - Obligor climate risk scorecard Yes - Adjustments to obligor rating methodology No Yes - Climate stress loss scenarios 36% Yes - Other (1) 39% 21% 18% Percentage of participants SME Mid-Market Corporate 34% 21% 59% 21% 28% 54% 17% 39% 20% 22% 27% 21% 52% 39% 15% CRE 1: E.g., Adjustments to obligor rating methodology and climate stress loss scenarios, but beyond 12 months. Bucketing of risks (geography, industry, property type segments). N=33 N=29 N=41 N=35 Have you implemented or are planning to implement in the next 12 months any changes to the credit assessment/ adjudication and monitoring models to capture the impact of climate change? (transition and physical risks) C: > 50% of participants have implemented/or are planning to implement climate stress loss analyses in mid-market, corporate, or CRE space Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 10. McKinsey & Company 10 D: In the past 2 years, top investment areas have been data tech and data acquisition – this trend is expected to continue over the next 2 years 13 5 8 2 6 6 10 4 9 5 6 7 10 9 1 0 Data scientists and engineers for development and validation Data acquisitions Data tech (e.g. cloud, visualization) 11 Data scientists and engineers for data quality processes 3rd Party/Vendor models 0 0 Other, please specify 1 25 22 22 20 3rd largest Largest 2nd largest 15 7 4 2 5 1 9 6 6 8 5 5 7 11 10 1 1 0 20 29 21 0 20 11 Where have you made the most investments in the past 2 years and where do you expect to invest the most in the coming 2 years? In the past 2 years, the top investment areas for participants were data tech and data acquisition And this trend in expected to continue over the next 2 years with higher expected investment Other top investment areas include talent for both development/ validation and data processing Number of votes Expectation for the next 2 years In the past 2 years Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 11. McKinsey & Company 11 E: Machine learning models are primarily gaining traction for Risk Scoring of SMEs and Early Warning across the board Early warning indicators Credit pricing 39% Risk score/ Rating/ PD model Stress Testing IFRS 9/ CECL 44% 44% 39% 32% 37% 68% 24% 56% 46% 7% 10% 5% 29% 5% 12% 2% 0% 0% 0% 0% 0% 5% 0% 0% Simulation-based models (e.g., monte-carlo approaches to economic capital) Statistical approaches (e.g., linear or logistic regression or CHAID) Machine learning tree algorithms (e.g. random forest, gradient boosting) Expert based qualitative models Other machine learning models (e.g. support vector, NLP, neural network) 44% Stress Testing Risk score/ Rating/ PD model Early warning indicators 41% Credit pricing 47% IFRS 9/ CECL 44% 38% 66% 72% 56% 53% 38% 6% 9% 3% 6% 28% 0% 13% 16% 6% 3% 0% 0% 0% 3% 3% Corpo- rate SME What methodologies are being “used in production”, “validated” or “in pilot” for each of the listed use cases for the Corporate and SME portfolio Expert based and statistical models are most widely applied approaches across the spectrum of use cases Simulation based models are more widely used for stress testing Machine learning models are getting traction for Risk Scoring, Early Warning, and Pricing Percentage Innovative data sources Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 12. McKinsey & Company 12 E: Innovative external data sources are more used for Corporate segment while SME segment uses more innovative internal data sources Which of the following categories of data are being used in production, under pilot or under consideration for credit risk management use cases within the Corporate portfolio? External data sources: For Corporates, over 50% of the participants are using, piloting or considering New media or social media and 3rd Party account data, a higher proportion than for SMEs Both segments use E-commerce data at similar rates Internal data sources: For Corporates, over 70% of the participants are using, piloting or considering Automated client/ issuer financials and internal credit behavior data, while for the SME segment in addition to above 2, internal cross product data also has large share External data sources Internal data sources 97% 88% 76% 59% 53% 33% 32% 88% 76% 62% 56% 44% Economic and market forecasts Credit bureau and rating agency data Financial market data on issuers News media or social media Other high frequency data 3rd Party account data Automated client / issuer financials E-commerce data Internal credit behavior data Internal cross product behavior Internal client interaction data Internal consumer-wholesale crossover data 91% 63% 35% 32% 30% 29% 19% 83% 91% 76% 45% 38% Corporate SME Percentage Innovative data sources Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 13. McKinsey & Company 13 F. Automated decisions are largely a feature of SME portfolios SME 18% 18% 16% 34% 19% 11% 72% 8% 4% Mid-market 87% 13% Corporate 89% 11% CRE None and Not implemented Less than 10% 11-30% 31-50% Above 50% Percentage of participants, where applicable N= 27 25 31 27 Fully automated decisions for a material portion of the portfolio is almost exclusively a feature of SME portfolios However, there are pockets of portfolio with full automation, even for mid-market and others Key insights In the past 3-5 years, what was the typical percentage of automated decisions based on models in your portfolio? Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 14. McKinsey & Company 14 F: Where implemented, use of innovative data and/or advanced analytics typically achieved increased automation 80 20 60 0 10 40 70 50 30 0 14 70 4 7 9 5 37 22 11 23 0 Mid-market 7 0 7 SME CRE 9 Corporate 5 26 27 36 In terms of increasing automation, what benefit have you seen in the past 3-5 years from the use of innovative data and/or advanced analytics? 63% participants reported an increase in automation for the SME segment, followed by Mid- Market (27%), Corporate (19%) and CRE segment (14%) Where implemented, use of innovative data and/or advanced analytics typically improved automation by up to 10% Higher improvement (11-50%) in automation is typically observed in SME segment, which involves the highest level of direct automated decisioning Increased by 11-50% Implemented but no uplift in automation Increased by up to 10% Increased by more than 50% 70% 36% 26% 27% XX% % of participants implemented Percentage of participants that implemented Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 15. McKinsey & Company 15 F: Where implemented, use of innovative data and/or advanced analytics significantly improved turn-around-time “TAT” for SMEs 10 40 60 50 70 20 30 0 7 9 12 0 41 3 7 19 30 Mid-market SME 20 28 30 12 0 Corporate 10 0 13 CRE 63 52 44 56% participants reported decrease in TAT for the SME segment, followed by Mid- Market (43%), CRE (32%), and Corporate (31%) Where implemented, use of innovative data and/or advanced analytics typically improved turn- around-time “TAT” by up to 10% Higher improvement (11-50%) in TAT is typically observed in SME segment but also to some extent in Mid-Market and CRE segment In terms of accelerating TAT to decision, what benefit have you seen in the past 3-5 years from the use of innovative data and/or advanced analytics? Decreased by 11-50% Implemented but no reduction in approval time Decreased by upto 10% Decreased by more than 50% 63% 52% 41% 44% Percentage of participants that implemented XX% % of participants implemented Source: McKinsey/IACPM Survey on data and analytics innovations in Credit Portfolio Management – October 2021 CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 16. McKinsey & Company 16 Agenda Summary of survey results 20 mins Perspectives on selected topics 30 mins Questions 10 mins
  • 17. McKinsey & Company 17 Perspectives on specific topics : Climate Risk, Next generation data and analytics, and Scenario planning and forecasting Climate Risk Next generation data and analytics Scenario Planning and Forecasting Using cross-product data with help from AI/ML can drive both revenue growth and automated credit decision For larger obligors, availability of analytics and accessible data is key for turnaround time reduction, however, doing so requires treating data quality as more than a ‘regulator-required’ effort Setting up a modular, portfolio specific methodology for scenario analyses is critical, prioritization will depend on exposure to high-risk sectors Climate risk impact on portfolio requires inter-disciplinary skills and mobilization across credit, front-line, and model risk management Building a flexible infrastructure to forecast and optimize portfolio is more critical than ever Rather than waiting for a full-scale solution, banks would want to establish analytics and organizational capabilities that enable rapid ‘what-if’ analyses CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 18. McKinsey & Company 18 Key learnings on the materiality of Climate Risk on Credit, based on our extensive work with banks The majority of the risk is in knock-on impacts that most banks do not model Direct damages are immaterial on credit; Knock-on effects can dwarf direct impacts, e.g., 4.5x the impact of direct 1st order impact for a Muni flood example Material risk drivers include community deterioration, geographic transition risk, and broader ecosystem impacts (e.g., insurance cost and availability) Real estate losses are driven by asset pricing (property values and cap rates), not physical damage Risk is concentrated in “pockets” across the portfolio; banks need to take a targeted approach Both physical and transition risk lie in very targeted areas of the portfolio: for example, for a large global bank we identified that approximately ~15% of their loan book was materially exposed to climate risk Banks need to perform heatmapping to focus their efforts on the high risk portfolios and risks: even within a CRE portfolio for a large US bank, we found that majority of the credit impact came from 10% of the portfolio ​ C urrent SVA (estimated) ​ O il and Gas ​ C ommercial Real Estate ​ A griculture ​ M anufacturing ​ E xpected 2030 SVA ​ M ining ​~700 ​ T ransportation ​ U tilities ​ R esidential ​~1100 ​-35% ​Impact on Economic profit1, CAD mn 1. Calculated as the return on invested equity above the return on equity (estimated at 10.2% based on McKinsey CPAnalytics) times invested equity (allocation based on exposure). Assumes that percentage NII decline = percentage net income decline Climate is not a “capital” problem; however it can have real impact on returns / economic profit A ‘CCAR mindset’ of focusing on capital risks will underestimate the business value risk and miss the opportunity to steer the business For a North American bank, we identified that 35% of economic profits could erode by 2030 without taking action on key pockets of climate risk exposures The “average” impact is moderate in the near-term, but there is high degree of counterparty-level variability We found that even for high risk industries, the average impact is moderate: For example, in a portfolio of upstream O&G companies, the impact by 2025 under below 2C scenario was ~7% median impact on EBITDA However, the difference between winners and losers is stark: in the upstream O&G example, we saw several counterparties with up to ~40% impact on EBITDA, while there were other companies that saw a positive EBITDA impact Climate Risk CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 19. McKinsey & Company 19 Scenario analyses for corporate and mid-market obligors needs to be sector-specific: oil and gas example Module Specific Data Carbon cost Upstream Oil & Gas Acute impact Competition module Stage 2 profit revenue, cost Integration module Physical risk impact Transition risk impact Climate-stress obligor financials Supply Chain Oil & Gas Cleantech Scenario selection (e.g. NGFS Scenarios) Scenario expansion and country downscaling (e.g., damage curve, transition pathways) i ii iii iv v vi vii Obligor Financials Oil and Gas Asset- Level Data Clean Tech Market Shares Obligor Emissions Abatement Cost Curves Methodology isolates individual aspects of scenario analysis through modules that calculate Oil and Gas obligor impacts from: Ÿ Oil and Gas demand changes across Upstream and Downstream Operations Ÿ Clean Technology demand changes Ÿ Carbon Costs Ÿ Acute Physical Hazard damage costs Climate Risk CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 20. McKinsey & Company 20 Physical risk impact Revenue impact Carbon cost impact Sanitized example : Demand destruction, direct carbon cost, and market impacts for a specific obligor Driven by current business composition, breakeven cost, and oil and gas price Driven by carbon emissions which drives carbon costs, abatement and ability to pass cost to customers/ market impacts Demand destruction Market impacts Current valuation 7% Adaptation Physical impact 3% 40% 0% Demand creation 50% Direct carbon costs 5% Abatement 30% Full impact 100% 41% Delayed transition scenario – 2050 Valuation impact waterfall – average impacts weighted by exposure Takeaway • Demand destruction, direct carbon cost and market impacts are the top 3 drivers of the climate impact to the O&G portfolio • Demand destruction is related to the business composition and breakeven cost • Direct carbon cost is related to the carbon price and the amount of output that can be sustained • Market impact is related to the ability to pass cost to customers and gain market share from other players ASSUMING NO MANAGEMENT/BANK ACTION Climate Risk CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 21. McKinsey & Company 21 Sanitized example: Portfolio level impact for a sample of public oil and gas obligors 200 1,300 40 300 65 1,700 800 35 25 70 15 30 1,100 400 20 600 45 700 500 50 55 900 1,000 60 1,600 1,200 0 100 1,500 1,400 Cimarex Energy Co. Exxon Mobil Corporation Total SE Petrobras Scope 1 CO2eq per $M total revenue Breakeven Oil Price (USD/bbl) Seven Generations Energy Ltd. Woodside Petroleum Ltd BP p.l.c. Aker BP ASA Antero Resources Corporation APA Corporation Equinor ASA Conocophillips Marathon Oil Corporation Santos Limited Suncor Energy Inc. Canadian Natural Resources Limited Chevron Corporation Cnx Resources Corporation Eog Resources, Inc. OMV Ag Pdc Energy, Inc. Pioneer Natural Resources Company Royal Dutch Shell Plc 500 25 300 600 20 15 70 40 30 800 35 50 900 1,000 400 1,100 45 1,200 1,300 55 1,400 60 65 1,500 0 700 100 200 1,600 1,700 Total SE Scope 1 CO2eq per $M total revenue Exxon Mobil Corporation Antero Resources Corporation BP p.l.c. Marathon Oil Corporation Equinor ASA Aker BP ASA Woodside Petroleum Ltd Suncor Energy Inc. Cnx Resources Corporation Canadian Natural Resources Limited Royal Dutch Shell Plc Chevron Corporation Pioneer Natural Resources Company APA Corporation Santos Limited Eog Resources, Inc. Cimarex Energy Co. OMV Ag Pdc Energy, Inc. Petrobras Seven Generations Energy Ltd. Breakeven Oil Price (USD/bbl) Conocophillips Impact of each driver on earnings for upstream companies Bubble size: Market cap 100% of current market cap 2030: Immediate transition scenario 2050: Delayed transition scenario A B Note: Scenario analysis assumes no mitigation action by bank or borrowers; Source: McKinsey Energy Insights, Rystad Earnings increase Earnings decrease 51-100% Earnings decrease <=50% Climate Risk CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 22. McKinsey & Company 22 Case study: Using a phased framework to develop bespoke scenario impact assessment tool for CRE portfolios (1/2) Impact to Net Operating Income Mechanism to translate financials to expected loss Identify risks and specify scenarios Assess impact on obligor’s financials Quantify portfolio impact Identify hazards Ÿ Flood, hurricane and wildfire most important, given the bank’s footprint Ÿ Transition risk leading to GVA impact in counties with high fossil-fuel dependency Downscale scenarios Ÿ Hurricane wind severity in every mile across areas where storms have been observed Ÿ Probability of flood and inundation levels Additional drivers of credit worthiness Ÿ Community and infrastructure resilience or vulnerability Ÿ Additional sponsor risk Ÿ Risk mitigants like levees Impact to Property Value Ÿ Decline in property value due to change in expected impact of climate events Ÿ Uninsured damage from physical risk event Ÿ Short term revenue decrease due to business interruption Ÿ Increase in insurance cost due to increase in frequency and severity of physical risk Ÿ NOI and property valuation impact Loan- to-value and DSCR of properties, translated into PD, LGD and EL Ÿ Detailed sensitivity testing, e.g.: — Change in cap rate — Insurance unavailability for specific hazard like fire, or in a specific state Climate Risk CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 23. McKinsey & Company 23 Case study: Using a phased framework to develop bespoke scenario impact assessment tool for CRE portfolios (2/2) Sanitized example from US based CRE portfolio of a large bank #1 100 140 2040 9 2030 Current 23 8 2050 Total relative EL by 2050 Climate impact to EL arranged by deciles 70% 40% 10% 0 20% 60% 0.2 50% 30% 80% 90% 100% Ø 4.6 0 43.0 1.6 0.6 0.3 0.1 0.1 0 Average indexed increase in EL 100 334 125 72 Total relative EL by 2050 37 Current 2030 2040 2050 3 2 3 10 5 4 2 Florida 0 1 Ohio 5 New York 1 14 9 California 5 0 1 Uninsured damage Increase in insurance cost Uninsured business interruption Reduced demand for properties exposed to climate risk Transition risk & Heat stress Net Cash Flow decline by transmission channel With insurance Without insurance Evolution of climate shocked Expected Loss Scenario : Delayed Transition Evolution of climate shocked Expected Loss Scenario : Delayed Transition EL increase by year (starting EL indexed to 100) Climate Risk CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 24. McKinsey & Company 24 Deposits data can be used to create an up-to-the-minute estimate of SMB’s financials Strategic implications Risk signals extracted from transactions Estimated revenues and profits Business seasonality Sectoral dependencies SME financial position Multi-banking clients Loans with other lenders Interest rate changes by other lenders Fintech relationships (e.g. Paypal) Competitive information Deposits transactions can be analyzed using a transaction classifier to derive an estimate of Ÿ SMB’s financials (e.g. revenues, revenue growth, and profits) Using a reinforcement learning and natural language processing, deposits transactions are analyzed to provide structure to unstructured data In a client application, this method yield 95% transaction classification accuracy and over 300 risk signals and competitive indicators, resulting in substantial improvement in predictive power of credit risk models Competition to become the primary bank will intensify In turn, banks that host the deposit account where the payroll is deposited have a competitive advantage Deposit accounts are already increasingly being bundled with other financial products Fees for deposit accounts are expected to reduce further Context Next generation data analytics CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 25. McKinsey & Company 25 Case example: Leveraging transaction data to transform credit decisioning Up to 40% Accuracy of financial inflows and outflows from classifier 95% Predictive risk signals – linked to transaction classifications >300 Traditional lending process More sophisticated understanding of risks More customer insights More predictive power Reimagined credit decisioning additional predictive power in some subsegments Illustrative; Client example Customer provides bank account and provides simple historic financial statements ETB: Transactional data from own system, used to build synthetic financial statements Customer completes credit application form Customer completes online application (pre-populated, leveraging API enabled data sources) NTB: Customer provides permission1 to access bank transactions data to build synthetic financial statements RM manually reviews data and performs credit assessment (scoring model, uses credit bureau, and historic/ narrow financial data) Algorithm integrates data and performs credit assessment (e.g., leveraging transaction classier to build financial statements, link to risk signals) Manual decision communicated to customer Decision communicated to customer (instant time yes) Next generation data analytics CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 26. McKinsey & Company 26 AI4DQ An AI-driven asset deployment driven approach can accelerate data quality improvement significantly 1 AI toolkit for data 2 Deployment accelerators 3 Training modules Detection: Ready to implement package to assess data quality Correction: Relationship discovery and anomaly detection to find errors Repair: AI driven correction through an open-loop process Deployable as pipelines that can be stand-alone for immediate results and integrated into Data Platforms to continuously monitor and improve data quality (e.g., Apache Airflow integration with Collibra), platform agnostic deployment A new way of working, including roles, talent, an a fast-paced Agile operating model Co-development of solutions through build-operate-transfer to sustain the impact When implemented together these 3 components significantly accelerate data quality capabilities CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited AI driven data quality
  • 27. McKinsey & Company 27 Data Quality can be addressed through modularized tools in three core areas Detection Correction Repair How machine learning helps Automatic generation of reports on the profile of data, inferred relationships between tables and anomalies, root cause identification to prioritize upstream interventions Rule-mining and clustering algorithms recommend corrections to data quality errors, quantify confidence, and help estimate business impact Validate corrections with experts (open-loop) confirming only lower confidence recommendations manually and automated approval of validated changes Illustration Detect top 1th percentile interest rate, unusually late maturity date as a potential error Correct loan type with attribution, country of origination, and address with 95% confidence Flag and automatically fix reporting dates that were corrected by experts repeated Objectives Measure the quality of each record on a scale of Low, Medium, High data quality to help prioritize remediation Recommend corrections to data quality errors prioritized by business impact and measure the confidence (e.g., 95%) Validate recommendations with experts and incorporate changes into underlying data systems feeding reports Ready-to- deploy tools Root cause identification Comparison to 3rd party data Data relationship discovery Automated DQ rule generation Collaborative workflows Anomality detection Entity disconnect identification Free-form text corrections Programmatic data validation Corrections that integrate attribute, record, and database signals CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited AI driven data quality
  • 28. McKinsey & Company 28 High impact examples AI being used to improve data quality Automated identification of errors in CRE loan data Root cause detection and correction to errors in 5M customer accounts Accelerating an enterprise data transformation with AI Corrected 5M free-form occupations to accelerate AML Customer Risk Rating using a neural language model and traditional fuzzy matching Automated data quality detection and correction with AI (human-in- the-loop) to reduce data transformation timeline by 30-40 percent Automated detection of data quality errors in over 100,000 loans using time series anomaly detection to detect issues in real time and prioritize remediation CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited AI driven data quality
  • 29. McKinsey & Company 29 Scenario planning forecasting approach needs to be tailored to bank’s footprint flexible to incorporate different driving factors Full estimates from business drivers and optimize across BUs Interest Income Credit Loss … Canada UK US B2 B1 B3 A3 A1 B4 A4 A2 B5 Scenarios from underlying reasons … Canada UK US Illustrative Develop inflation scenarios based on set of underlying reasons (e.g., supply chain bottleneck leading to ‘cost-push’ inflation) and consider incorporation of important macro- linkages (e.g., currency fluctuation risk) Business drivers from scenarios Identify drivers that are likely to be inflation- sensitive for business portfolios / geographies, and direction of impact – e.g., inflation may drive up transaction volume in the short term; however, inflation may also reduce demand in the long term Develop analytics to project the underlying drivers and the business portfolio financials Based on projection results, synthesize implications for strategic decision-making Scenario Planning and Forecasting CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 30. McKinsey & Company 30 Uncertainty in economic path require a forecasting approach that is nimble, flexible, and responsive to emerging risks like inflation Ÿ Model and analytics developed for relevant drivers and separately for relevant geography Ÿ Models capture time trend, macroeconomic factors, seasonality and latent portfolio factors in a transparent manner Ÿ Designed to capture, test and visualize business actions under different macro- conditions Ÿ Structured to generate report on portfolio and sub-segment level with different ‘what-if’ Ÿ Flexibility to expand current internal scenarios with new variables Ÿ Ability to add macro and event driven overlays Ÿ Ability to refresh scenarios weekly vs. monthly Versatility Coherence Flexibility Scenario Generation and Refinement Customized scenarios for portfolio/segments Ÿ Can be macro-economic or event driven Ÿ Recent macro-trends (e.g., EPOP ratio) core to design Business driver forecast through models Volume, Revenue, and Expense Forecast Models tied to business drivers (e.g., line utilization) and scenario conditioned Implementation Engine One-click solution for aggregation, reporting and visualization Scenario Planning and Forecasting CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 31. McKinsey & Company 31 In addition to standard capital or liquidity constraints, exploring emissions related constraints can help in portfolio alignment Capital consumption Carbon emission Preliminary list of constraints incorporated in the approach (to be refined based on observations during design phase) Product Industry Carbon limit Capital limit Approach to incorporate emissions related constraints BU Example output by industry at a loan level Carbon consumption Economic return Energy Materials Telecom Capital Risk Weighted Assets Expected Loss Origination/balance growth Liquidity coverage ratio Emissions Ÿ Include carbon limit based on benchmark scenario at North America with relevant downscaling to Canada/Alberta Ÿ Add constraint for net zero target of total portfolio emissions Ÿ Develop functionality to add sector specific targets and connect with potential sector-specific carbon intensity metric Loans out Loans in Scenario Planning and Forecasting CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 32. McKinsey & Company 32 Microsegment based forecast models are coded in engine with a Balance Sheet Optimizer added for ongoing scenario planning Node 2 Node n Node 1 Node 1 O/P Node 2 O/P Scenario File Input internal data Config File Overlay File . . . . . . Node n O/P Input Pre-processing (e.g., Python) Run pre-processing code in Python e.g., overlays, variable transformation Prepare input files and data Data aggregation and Optimizer Aggregation Aggregate predictions, conduct sensitivity Cashflow Engine Cashflow calculator (e.g., Spark) Run microsegment level calculations Strategic planning FP & A Add-on modules Feed output to additional modules per specific use case Client mgmt. Architecture tailored based on level of manual intervention desired, and size of input data Example forecasting engine architecture for volume and revenue estimation Scenario Planning and Forecasting CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 33. McKinsey & Company 33 Content Summary of survey results 20 mins Perspectives on selected topics 30 mins Questions 10 mins CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited
  • 34. McKinsey & Company 34 Questions CONFIDENTIAL AND PROPRIETARY Any use of this material without specific permission of McKinsey & Company is strictly prohibited