Sie sind auf Seite 1von 29

CUT – OFF & STRATEGIES

PRESENTER: IOANNA ROUSSOU

Bucharest, 20/10/2010

Main Topics

 Cut – Off Table

 Cut – Off Objectives

 Profitability

 Profitability Considerations

 Practical Implications

1
Main Topics

 Cut – Off Table

 Cut – Off Objectives

 Profitability

 Profitability Considerations

 Practical Implications

Initial Step
Estimated Good Bad Odds
What is the result of a Score Development? at this score – Number of
Goods for each Bad
Expected TTD Number of
Applications at A Score Distribution – A Cut-Off Table Applications
this score expected at this
Total TTD -
Applications Expected Expected Interval G:B Approval Bads score and above
Portfolio
Score Booked Goods Bads Bad Rate Odds TTD Above Rate Above Bad Rate
Percent of
169 1,161 1,062 99 8.5% 10.8 52,536 67.5% 2,356 4.5% Applications
170 1,180 1,083 97 8.2% 11.2 51,375 66.0% 2,257 4.4% expected at this
Expected Bads171 at 1,198 1,103 95 7.9% 11.6 50,195 64.5% 2,160 4.3%score and above
this score 172 1,214 1,120 94 7.7% 12.0 48,997 62.9% 2,065 4.2%
173 1,228 1,136 92 7.5% 12.4 47,783 61.3% 1,971 4.1%
174 1,241 1,151 90 7.3% 12.8 46,555 59.7% 1,879 4.0%
175 1,251 1,163 88 7.0% 13.2 45,314 58.1% 1,789 3.9%
176 1,259 1,173 86 6.8% 13.7 44,063 56.5% 1,701 3.9% Number of Bads
177 1,266 1,183 83 6.6% 14.2 42,804 54.9% 1,615 3.8% expected at this
178 1,270 1,189 81 6.4% 14.7 41,538 53.3% 1,532 3.7%score and above
Estimated Bad
179 1,272 1,194 78 6.1% 15.2 40,268 51.6% 1,451 3.6%
Rate at this score
180 1,272 1,197 75 5.9% 16.0 38,996 50.0% 1,373 3.5%
Cumulative Bad
Rate at this
score and above

2
Key Points
Interval Bad Rate
It is the bad rate at that specific score-only if you accept all applications at
score 175
Portfolio Bad Rate
It is the bad rate of the whole portfolio-if you accept all applications from
score 175 and above
Total TTD -
Expected Expected Interval G:B Approval Bads Portfolio
Score Applications TTD Above
Goods Bads Bad Rate Odds Rate Above Bad Rate
Booked

173 1,228 1,136 92 7.5% 12.4 47,783 61.3% 1,971 4.1%


174 1,241 1,151 90 7.3% 12.8 46,555 59.7% 1,879 4.0%
175 1,251 1,163 88 7.0% 13.2 45,314 58.1% 1,789 3.9%
176 1,259 1,173 86 6.8% 13.7 44,063 56.5% 1,701 3.9%
177 1,266 1,183 83 6.6% 14.2 42,804 54.9% 1,615 3.8%

Interval vs Marginal Bad Rate


Cumulative vs Portfolio Bad Rate

Cut – Off …table types


Various versions in the industry:
Score groups in ranges At Amount Level

Incoming Bad Rate per TTD Loss From Revenue


Score Band Score
Applications Band Applications Bads From Goods

170 - 179 10% 15% 169 1,161 128,130 107,474


180 - 189 10% 7% 170 1,180 125,541 109,600
190 - 199 10% 4% 171 1,198 122,953 111,624
200 – 209 10% 3% 172 1,214 121,659 113,344
173 1,228 119,070 114,963
174 1,241 116,482 116,481
Observed Bad Rate vs. Predicted
Interval Bad Interval Bad
Score TTD Applications Rate Rate
(Predicted) (Observed)

169 1,161 8.50% 10.20%


170 1,180 8.20% 9.02%
171 1,198 7.90% 7.11%
172 1,214 7.70% 6.93%
173 1,228 7.50% 8.25%
174 1,241 7.30% 8.76%

3
Key Question
Bad Definition
60+ with 18 average outcome period or
90+ with 24 average outcome period

What is my Loss Rate?

X Bads Y Losses
Is it included in the table?
Relationship between Bad and Loss definition

6% Bad Rate = 2% Loss Rate

Cut-Off Objectives

4
Main Topics

 Cut – Off Table

 Cut – Off Objectives

 Profitability

 Profitability Considerations

 Practical Implications

Objectives
Book Better Customers
What is a better customer?
Reduce Losses
Are less credit losses always good for the business?
Increase Market Share
Increase the approval rate – what is optimal?
Maximise Number of Automated Decisions
Where is manual verification and assessment worthwhile?
Increase Profits
How do I measure profitability?

5
Objectives
What is your aim in setting a Cut-Off?

Cut-Off Table
Total TTD -
Applications Expected Expected Interval Approval Bads Portfolio
Score Booked Goods Bads Bad Rate G:B Odds TTD Above Rate Above Bad Rate

169 1,161 1,062 99 8.5% 10.8 52,536 67.5% 2,356 4.5%


170 1,180 1,083 97 8.2% 11.2 51,375 66.0% 2,257 4.4%
171 1,198 1,103 95 7.9% 11.6 50,195 64.5% 2,160 4.3%
172 1,214 1,120 94 7.7% 12.0 48,997 62.9% 2,065 4.2%
173 1,228 1,136 92 7.5% 12.4 47,783 61.3% 1,971 4.1%
174 1,241 1,151 90 7.3% 12.8 46,555 59.7% 1,879 4.0%
175 1,251 1,163 88 7.0% 13.2 45,314 58.1% 1,789 3.9%
176 1,259 1,173 86 6.8% 13.7 44,063 56.5% 1,701 3.9%
177 1,266 1,183 83 6.6% 14.2 42,804 54.9% 1,615 3.8%
178 1,270 1,189 81 6.4% 14.7 41,538 53.3% 1,532 3.7%
179 1,272 1,194 78 6.1% 15.2 40,268 51.6% 1,451 3.6%
180 1,272 1,197 75 5.9% 16.0 38,996 50.0% 1,373 3.5%

Strategy Curve – Current Situation


Strategy Curve
100.0%

90.0%

80.0%

70.0%

60.0%
Approval Rate

50.0%

Current:
40.0%
Approval Rate: 56%
30.0%
Bad Rate: 5.2%

20.0%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

6
Strategy Curve – Current Situation
Strategy Curve
100.0%

90.0%

80.0% Strategy Curve


New Score
70.0%

60.0%
Approval Rate

50.0%

Current:
40.0%
Approval Rate: 56%
30.0%
Bad Rate: 5.2%

20.0%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

Setting Cut-Off-Application Scores


Can set Cut-Off based on a single criteria - a target
 Approval Rate
OR
 Portfolio Bad Rate

A solution which is a compromise between the two


I want to keep the Approval Rate above 55%
But I don’t want the Portfolio Bad Rate to be more than 5%

Use Cut-Off Tables


 Simple
 Measurable

7
Strategy Curve–Maintain Bad Rate
Strategy Curve
100.0%

90.0%

80.0%

70.0%

60.0%
Approval Rate

50.0%

40.0%

30.0% Current:
Approval Rate: 56%
20.0% Bad Rate: 5.2%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

Strategy Curve–Maintain Bad Rate


Strategy Curve
100.0%

90.0%

80.0%

70.0%

60.0%
New:
Approval Rate

Approval Rate: 78.2%


50.0% Bad Rate: 5.2%

40.0%

30.0% Current:
Approval Rate: 56%
20.0% Bad Rate: 5.2%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

8
Objectives–Maintain Bad Rate
What is your aim in setting a Cut-Off?

Cut-Off Table
Total TTD -
Applications Expected Expected Interval Approval Bads Portfolio
Score Booked Bads Goods Bad Rate G:B Odds TTD Above Rate Above Bad Rate

155 773 101 672 13.1% 6.7 66,057 84.7% 3,786 5.7%
156 804 102 702 12.6% 6.9 65,284 83.7% 3,685 5.6%
157 835 101 735 12.0% 7.3 64,480 82.7% 3,583 5.6%
158 866 102 764 11.8% 7.5 63,645 81.6% 3,483 5.5%
159 897 103 794 11.5% 7.7 62,779 80.5% 3,381 5.4%
160 927 103 824 11.1% 8.0 61,882 79.3% 3,277 5.3%
161 957 104 853 10.9% 8.2 60,954 78.2% 3,174 5.2%
162 986 104 883 10.5% 8.5 59,997 76.9% 3,070 5.1%
163 1,015 104 911 10.2% 8.8 59,010 75.7% 2,966 5.0%
164 1,042 103 939 9.9% 9.1 57,996 74.4% 2,863 4.9%
165 1,068 103 966 9.6% 9.4 56,954 73.0% 2,759 4.8%
166 1,093 102 992 9.3% 9.7 55,886 71.7% 2,657 4.8%

Strategy Curve–Maintain Approval Rate


Strategy Curve
100.0%

90.0%

80.0%

70.0%

60.0%
Approval Rate

50.0%

40.0%

30.0% Current:
Approval Rate: 56%
20.0% Bad Rate: 5.2%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

9
Strategy Curve – Maintain Approval Rate
Strategy Curve
100.0%

90.0%

80.0% New:
Approval Rate: 56%
70.0% Bad Rate: 3.9%

60.0%
Approval Rate

50.0%

40.0% Current:
Approval Rate: 56%
30.0% Bad Rate: 5.2%
20.0%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

Objectives - Maintain Approval Rate


What is your aim in setting a Cut-Off?

Cut-Off Table
Total TTD -
Applications Expected Expected Interval Approval Bads Portfolio
Score Booked Bads Goods Bad Rate G:B Odds TTD Above Rate Above Bad Rate

170 1,180 97 1,083 8.2% 11.2 51,375 65.9% 2,256 4.4%


171 1,198 95 1,103 8.0% 11.6 50,194 64.4% 2,159 4.3%
172 1,214 94 1,120 7.7% 12.0 48,996 62.8% 2,064 4.2%
173 1,228 92 1,137 7.5% 12.4 47,782 61.3% 1,970 4.1%
174 1,241 90 1,151 7.2% 12.8 46,554 59.7% 1,879 4.0%
175 1,251 88 1,163 7.0% 13.2 45,314 58.1% 1,789 3.9%
176 1,259 86 1,174 6.8% 13.7 44,063 56.5% 1,701 3.9%
177 1,266 83 1,182 6.6% 14.2 42,803 54.9% 1,615 3.8%
178 1,270 81 1,189 6.4% 14.7 41,538 53.3% 1,532 3.7%
179 1,272 78 1,194 6.2% 15.2 40,268 51.6% 1,451 3.6%
180 1,272 75 1,197 5.9% 16.0 38,996 50.0% 1,373 3.5%
181 1,270 73 1,196 5.8% 16.3 37,724 48.4% 1,298 3.4%

10
Strategy Curve – Maintain Approval Rate
Strategy Curve
100.0%

90.0%

80.0% New:
Approval Rate: 56%
70.0% Bad Rate: 3.9%

60.0% New:
Approval Rate

Approval Rate: 78.2%


50.0% Bad Rate: 5.2%
40.0%

30.0%
Current:
Approval Rate: 56%
20.0% Bad Rate: 5.2%

10.0%

0.0%
0.5% 0.7% 0.8% 1.1% 1.5% 2.0% 2.6% 3.4% 4.2% 5.1% 6.0% 6.7% 7.3% 7.6% 7.8%

Portfolio Bad Rate

Setting the Cut-Off


As a result of the new scorecard we have many options, amongst them are

Strategy Criteria Approval Rate Bad Rate


Current 56.0% 5.2%
Strategy 1 Same Approval Rate 56.0% 3.9%
Strategy 2 Same Bad Rate 78.2% 5.2%

But which is the best option?


• Booking 22% more of the incoming applications but having the same bad rate
• Reducing the bad rate by 1.3 percentage points but still booking 56% of applications
Or Is it something else?

11
Profitability
This brings us back to the first question

What is the target for the cut-off and the overall aim of business?

In most cases the business target is to maximise profitability

By balancing Risk and Reward the goal is:

Reward (revenue) and Risk (losses) BREAK EVEN POINT

Maximum Profit

Setting Cut-Off-Application Scores


Setting the Cut-Off based on maintaining approval or portfolio bad rate is
straightforward but difficult to optimise:
Still reject profitable business
or
Accept unprofitable business

The answer is to correctly balance Risk and Reward!


But what does this mean?

Optimise profitability by use of:


 Profitability Analysis
 Modeling

12
Profitability

Main Topics

 Cut – Off Table

 Cut – Off Objectives

 Profitability

 Profitability Considerations

 Practical Implications

13
Profitability
The principles are straightforward:
In correctly developed scores we have two categories of accounts:
 Goods
 Bads
Each individual score has an associated probability that an applicant, if
booked, will become Bad

Therefore if we know
 The Reward from a Good
AND
 The Risk from a Bad
We can use this information to decide on the cut-off

Setting The Cut-Off – Profitability


Steps:
Use Bad Definition to create Profit & Loss Accounts for the two sub-
portfolios
Portfolio of Good Accounts
Portfolio of Bad Accounts

Find the Profit and Loss for each at the account level

Use resulting Good/Bad ratio to set score cut-off

14
Setting The Cut-Off – Profitability
Required Information
Either:
All variables in Profit & Loss accounts calculated on an account by account
basis

Or:
Portfolio Information for all variables
and
Flow Rates and Delinquent Balances

Profit & Loss Statement


Good Bad

Gross Interest Revenue 27,205,678 490,000


Cost of Funds 7,773,051 140,000

Net Interest Revenue 19,432,627 350,000

Fees 2,880,000 90,000


Interchange Income 4,760,000 190,000

Total Card Income 27,072,627 630,000

Marketing Costs 4,800,000 200,000


Acquisition Cost 2,400,000 100,000
Operating Expense
Premises 1,350,000 18,000
Telecomms 220,000 12,000
Systems etc 420,000 10,000
Customer Service Costs 5,538,000 95,345
Credit Collection Expense 300,500 276,000

Total Expense 15,028,500 711,345

Credit Losses 4,990,370


Fraud Losses 2,332,800 97,200

Total Profits before Tax 9,711,327 -5,168,915

15
Profit & Loss Statement
Good Bad

Gross Interest Revenue 27,205,678 490,000


Accounts Cost of Funds 7,773,051 140,000
designated as
Good in Score Net Interest Revenue 19,432,627 350,000
development
Fees 2,880,000 90,000
Interchange Income 4,760,000 190,000

Total Card Income 27,072,627 630,000

Marketing Costs 4,800,000 200,000


Acquisition Cost 2,400,000 100,000
Operating Expense
Premises 1,350,000 18,000
Telecomms 220,000 12,000
Systems etc 420,000 10,000
Customer Service Costs 5,538,000 95,345
Credit Collection Expense 300,500 276,000

Total Expense 15,028,500 711,345

Credit Losses 4,990,370


Fraud Losses 2,332,800 97,200

Total Profits before Tax 9,711,327 -5,168,915

Profit & Loss Statement


Good Bad
Accounts
Gross Interest Revenue 27,205,678 490,000 designated as
Cost of Funds 7,773,051 140,000
Bad in Score
development
Net Interest Revenue 19,432,627 350,000

Fees 2,880,000 90,000


Interchange Income 4,760,000 190,000

Total Card Income 27,072,627 630,000

Marketing Costs 4,800,000 200,000


Acquisition Cost 2,400,000 100,000
Operating Expense
Premises 1,350,000 18,000
Telecomms 220,000 12,000
Systems etc 420,000 10,000
Customer Service Costs 5,538,000 95,345
Credit Collection Expense 300,500 276,000

Total Expense 15,028,500 711,345

Credit Losses 4,990,370


Fraud Losses 2,332,800 97,200

Total Profits before Tax 9,711,327 -5,168,915

16
Profit & Loss Statement – Account Level
Good Bad

Gross Interest Revenue 283.4 122.5


Cost of Funds 81.0 35.0

Net Interest Revenue 202.4 87.5

Fees 30.0 22.5


Interchange Income 49.6 47.5

Total Card Income 282.0 157.5

Marketing Costs 50.0 50.0


Acquisition Cost 25.0 25.0
Operating Expense
Premises 14.1 4.5
Telecomms 2.3 3.0
Cut-Off Set atSystems
ScoreetcYielding 4.4 2.5
Customer Service Costs 57.7 23.8
12.8 Goods:Bad Odds
Credit Collection Expense 3.1 69.0
1,292.2/101.2 = 12.8
Total Expense 156.5 177.8

Credit Losses 0.0 1247.6


Fraud Losses 24.3 24.3

Total Profits before Tax 101.2 -1,292.20

At Score Range Level


At the score equivalent to Good:Bad Odds of 12.8:1 it is breakeven
At Odds below 12.8:1 bookings are unprofitable
Above Odds of 12.8:1 bookings are profitable
Estimated Revenue
Loss From Net Profit
Score From
Total TTD Goods Bads Bad Rate G:B Odds Bads At Score
Goods

169 1,161 1,062 99 8.5% 10.8 107,474 128,130 -20,655


170 1,180 1,083 97 8.2% 11.2 109,600 125,541 -15,942
171 1,198 1,103 95 7.9% 11.6 111,624 122,953 -11,329
172 1,214 1,120 94 7.7% 12 113,344 121,659 -8,315
173 1,228 1,136 92 7.5% 12.4 114,963 119,070 -4,107
174 1,241 1,151 90 7.3% 12.8 116,481 116,482 0
175 1,251 1,163 88 7.0% 13.2 117,696 113,893 3,802
176 1,259 1,173 86 6.8% 13.7 118,708 111,305 7,403
177 1,266 1,183 83 6.6% 14.2 119,720 107,422 12,298
178 1,270 1,189 81 6.4% 14.7 120,327 104,833 15,493
179 1,272 1,194 78 6.1% 15.2 120,833 100,951 19,882
180 1,272 1,197 75 5.9% 16 121,136 97,068 24,068

17
At Portfolio Level
Approving only above Good:Bad Odds of 12.8:1 yields the Maximum Total Profit
At Odds below 12.8:1 the portfolio is less profitable
Above Odds of 12.8:1 the portfolio is less profitable
Revenue
Approval Goods Bads Portfolio Loss From
Score TTD Above From Total Profit
Rate Above Above Bad Rate Bads
Goods

169 52,536 67.5% 50,180 2,356 4.5% 5,078,216 3,049,229 2,028,987


170 51,375 66.0% 49,118 2,257 4.4% 4,970,742 2,921,100 2,049,642
171 50,195 64.5% 48,035 2,160 4.3% 4,861,142 2,795,558 2,065,584
172 48,997 62.9% 46,932 2,065 4.2% 4,749,518 2,672,606 2,076,913
173 47,783 61.3% 45,812 1,971 4.1% 4,636,174 2,550,947 2,085,227
174 46,555 59.7% 44,676 1,879 4.0% 4,521,212 2,431,877 2,089,335
175 45,314 58.1% 43,525 1,789 3.9% 4,404,730 2,315,396 2,089,334
176 44,063 56.5% 42,362 1,701 3.9% 4,287,034 2,201,502 2,085,532
177 42,804 54.9% 41,189 1,615 3.8% 4,168,327 2,090,198 2,078,129
178 41,538 53.3% 40,006 1,532 3.7% 4,048,607 1,982,776 2,065,832
179 40,268 51.6% 38,817 1,451 3.6% 3,928,280 1,877,942 2,050,338
180 38,996 50.0% 37,623 1,373 3.5% 3,807,448 1,776,992 2,030,456

At Portfolio Level
Portfolio Profit
2500000
Score 174
2000000

1500000
Total Profit

1000000

500000

0
107 117 127 137 147 157 167 177 187 197 207 217 227 237 247

-500000

-1000000
Score

So is it possible to set the best cut-off without considering profitability?

18
Scenarios
Market Share
Increase Cut-Off Table

Approval Portfolio
Score TTD Above G:B Odds Bads Above Total Profit
Rate Bad Rate

161 60,954 78.2% 8.2 3,174 5.2% 1,739,263

174 46,554 59.7% 12.8 1,879 4.0% 2,089,335

176 44,063 56.5% 13.7 1,701 3.9% 2,085,532

Reduce Losses

Profitability Considerations

19
Main Topics

 Cut – Off Table

 Cut – Off Objectives

 Profitability

 Profitability Considerations

 Practical Implications

Profitability – Is it Easy?
Risk is easy...
Ever 90+ Bad Rate
25.0%

20.0%

15.0%

10.0%

5.0%

0.0%
Up to 139 140 - 149 150 - 159 160 - 169 170 - 179 180 - 189 183 - 189 190 - 199 200 - 209 210+
Application Score

20
Profitability – Is it Easy?
Bank Policies
Approval Rate Product Type
Delinquency Level
Collections
Terms & Existing
Frauds Marketing
Conditions Customers

New
Cross Sell Limits
Clients

Attrition Usage

Account Maintenance

Profitability – Is it Easy?
Profitability is not - reality!
Average Total Revenue by Application Score
Credit Cards
800
735
689
700

600
531 530
491
500
427
389
400 366
342
309
300

200

100

0
Up to 139 140 - 149 150 - 159 160 - 169 170 - 179 180 - 189 183 - 189 190 - 199 200 - 209 210+
Application Score

21
Profitability – Is it Easy?
Profitability is not - reality!
Inactivity Rate and Average Balance
Credit Cards
1000 100.0%
Average Balance
900 % Inactive 90.0%

800 80.0%

700 70.0%
Average Balance

600 60.0%

% Inactive
500 50.0%

400 40.0%

300 30.0%

200 20.0%

100 10.0%

0 0.0%
Up to 139 140 - 149 150 - 159 160 - 169 170 - 179 180 - 189 183 - 189 190 - 199 200 - 209 210+
Score

Profitability – Is it Easy?
Profitability is not - reality!
Instalment Interest Income By Tenor
Loans Portfolio
12,000
Gross Interest Income
Net Interest Income
10,000 CoF

8,000
7,164

6,000

4,000 3,340

2,000 1,632
830

0
96 48 24 12
Tenor

22
Key Questions
Over What Period do you consider Costs and Revenues?

• For Tenor Loans, actual Loan Durations are easy to calculate…


For Revolving Loans how long should be considered?

• Profitability over 3 or 5 years will be very different to 20 years!!


Many Credit Cards will have active life of 20+ years…

Key Questions
What is the Real Value of a Customer?

• Should we consider Renewals?

• How should we consider the propensity of a customer to desire further


products in the future?

Handling of these questions greatly affects the cut-off decision


• Profitability – renewals, multiple products and greater time period
• Type of Applicants – we want to approve
Young applicants have greatest potential for growth

23
Cut-Off Strategy

Main Topics

 Cut – Off Table

 Cut – Off Objectives

 Profitability

 Profitability Considerations

 Practical Implications

24
Single Score Cut-Off Strategy
Score Distribution
1.8%

1.6%

1.4%
Reject Accept
Percent of Population

1.2%

1.0%

0.8%

0.6%

0.4%

0.2%

0.0%
107 117 127 137 147 157 167 177 187 197 207 217 227 237 247

Application Score

Multiple Score Cut-Off Strategy


Score Distribution
1.8%

1.6%

1.4% Reject Grey Area Accept


Percent of Population

1.2%

1.0%

0.8%

0.6%

0.4%

0.2%

0.0%
107 117 127 137 147 157 167 177 187 197 207 217 227 237 247

Application Score

25
Multiple Score Cut-Off Strategy
In the Grey Area customers are referred for additional underwriting
• It generally shows a Lack of Confidence in Score
• It can make Sceptics more comfortable

Can be useful as a temporary measure when the applicant distribution or


performance is not known for instance:
• Generic scorecard
- New Products
- New Sources

Multiple Score Cut-Off Strategy


Considerations to be taken into account with this strategy are:

• What additional criteria are being considered?

• Are they already in the scorecard?

• Is it judgmental or can these be converted in to rules?

• Can underwriters make a better decision than the score?

• Can it be appropriately tracked and monitored?

• Are there comprehensive reject and approval codes?

26
Implications
Probability of Default
25.0%

Cut – Off
20.0%

15.0%
Bad Rate

10.0%

5.0%

0.0%
Up to 139 140 - 149 150 - 159 160 - 169 170 - 179 180 - 189 183 - 189 190 - 199 200 - 209 210+
Score Ranges

Implications
Probability of Default
25.0%

Cut – Off
20.0%

15.0%
Bad Rate

Low Sides Performance


Grey Area objectives
10.0%

5.0%

0.0%
Up to 139 140 - 149 150 - 159 160 - 169 170 - 179 180 - 189 183 - 189 190 - 199 200 - 209 210+
Score Ranges

27
Implications
Significant Profitability Implications:
Your cut-off strategy effects your populations of customers

Depending on the score type


• Attrition
• Risk
• Usage, etc.
The score cut-off will impact
• Balances
• Delinquency
• Risk

Ultimately, each score cut-off will impact your Profitability!!

Additional Aspects
1. Complicated Structures of 2 or 3+ Scores
2. Risk Based Pricing
3. Limits
4. Macro Environment
5. Basel II Requirements – PDE, LGD, EAD

28
Thank you!!

29

Das könnte Ihnen auch gefallen