2003 Fall Conference

1
2003 Fall Conference

• Advanced Strategies Session

Greg Judge, VP Senior Financial
Strategist (gjudge_at_fhlbdm.com)
1-800-544-3452, ext 1118 FHLB of Des Moines
2
Roadmap for Today

• Current interest rate environment in historical
  context
• Detailed Case Study (foundation)
• Two members with very different funding
  strategies
• Blended Funding Case Study (application)
• Members from case study competing for same loans.
• Conclusions and Bonus Case Study

3
When will things get back to normal? What is
normal?
4
Are We About To Repeat 1994?
Monthly Fed Funds Rates January 1993 -
September 2003
7.00
6.00
5.00
Whats Next ? And When?
4.00
281 bps rate increase during 1994
3.00
2.00
1.00
0.00
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
What Steps Can You Take Today to Lessen the
Impact of Rising Rates On Your Funding Costs?
5
Nearly Every Member is well below the district
median COF
What will happen to your COF as rates increase?
6
Classical A/L ApproachLow Interest Rate
Environment

• Lengthen liabilities and shorten assets
• Profit from better margins when rates increase
  later.
• Problem
• What if rates remain low for an extended period
  of time?
• Bigger Problem
• These actions are likely the opposite of what
  your customers want.
• Incentives may be required
• CD specials, teaser ARMS, etc.
• Result
• This approach may be expensive and
  counter-productive.

7
Following the Classical Funding Approach Doesnt
Always Work As Intended Ranking went from the
44th Percentile to the 90th Percentile
Extending Deposit Terms While Rates are Lower
Can these increased costs ever be recovered in
the future?
8
Domestic Deposit Composition and Growth Trends
1992-2003
All FDIC Insured Banks in USA - Aggregated
Information
Percent Change From 1992-2003
Transaction -9
Savings / MMDA 197
CDs 51
Trillions
4.50
4.00
3.50
3.00
2.50
2.00
50
1.50
38
31
1.00
0.50
0.00
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
Savings and MMDAs have been the fastest growing
deposit segments This has been a steady trend
over the last 10-11 years.
9
Rates on Interest Bearing Deposits 1999-2003
Deposit Categories and Median Rates
1,072 FHLB DM Members Under 1 billion
Transaction accts 2.04
Savings/MMDAs 3.18
CDs 5.25
FED Funds (QA) 4.53
8.00
7.00
6.15
6.00
5.34
5.00
3.79
4.00
3.15
3.09
3.00
2.47
2.00
2.12
1.25
1.00
0.71
0.00
Mar-99
Dec-99
Dec-00
Dec-01
Dec-02
Jun-03
Compare the relative cost behaviors which types
are lower?
10
Retail Deposit Rate History 1991 2002
Bankrate.com National Rate Survey
CDs
Non-maturity Deposits
Source Bankrate.com
Non-Maturity Deposits tend to be less
rate-sensitive than CDs.
11
Higher CD Concentrations Typically Result in
Higher Deposit COFs
Deposit Cost is a Function of Price and Structure
Member Banks in the FHLB Des Moines District
1999-2003
CDs Less Than 50 of Deposits
CDs Greater Than 50 of Deposits
100
90
CDs
39
38
80
42
45
46
57
56
58
58
60
70
MMDAs
60
50
40
Savings
30
20
Checking
10




0
Dec-99
Dec-00
Dec-01
Dec-02
Mar-03
Dec-99
Dec-00
Dec-01
Dec-02
Mar-03
4.16
4.87
3.41
2.28
2.09
4.46
5.21
4.01
2.84
2.59
-30 bps
-34 bps
-60 bps
-56 bps
-50 bps
-
-
-
-
-
Quarterly COF on Int Bearing Deposits (IBDs) and
Cost Difference
12
Case Study

• Two member banks in the FHLB DM District
• Both from the same state
• Similar sized
• Very different funding strategies and costs
• Subject 1 average deposit COF percentile 81st
• Always above district median
• Subject 2 average deposit COF percentile 10th
• Always below district median

13
Subject Member 1 High Cost of Funds
Consistently Above the Median Average
Percentile 81st
14
Subject Member 2 Low Cost of Funds
Consistently Below the Median Average
Percentile 10th
15
Peer Comparison Funding Mix
As a Percent of Assets
Low COF Member Has More Diversified Funding
Sources
Total deposits
Total borrowed money (inc FHLB adv)
FFP Sec sold under agree to repurch
High COF Member
Low COF Member
100





90
3
3
3
3
3
5
8
8
6
9
6
9
80
8
9
9
70
60
50
88
88
87
87
87
40
78
77
76
74
74
30
20
10



Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
3.39
3.14
3.04
2.84
2.75
2.14
2.07
1.81
1.56
1.38
125 bps
107 bps
123 bps
128 bps
137 bps
-
-
-
-
-
Quarterly COF on Int Bearing Liabs (IBL) and
Difference from Peer
16
Comparison of Deposit Structure Percent of
Deposits
Deposit Cost is a Function of Price and Structure
The mix of deposits is as important as the rates
on deposits
High COF Member
Low COF Member
100
90
26
26
27
27
27
CDs
80
59
60
70
61
62
61
MMDAs
60
50
40
Savings
30
20
Checking
10




0
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
3.36
3.10
2.99
2.79
2.68
1.90
1.79
1.57
1.35
1.37
146 bps
131 bps
142 bps
144 bps
131 bps
-
-
-
-
-
Quarterly COF on Int Bearing Deposits (IBDs) and
Difference From Peer
Low COF Member Has Less CDs in its Deposit Mix
17
How did these two members stack up nationally in
March 2003?
18
Low COF Member Has a Less Expensive CD Portfolio
High COF Member CD COF
Low COF Member CD COF
5.00
4.56
4.50
4.16
4.04
4.00
3.81
3.71
3.50
3.27
2.98
3.00
2.76
2.41
2.50
2.27
2.00
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
129 bps
118 bps
128 bps
154 bps
130 bps
The Structure of the CDs May Be Responsible for
this Difference
19
Detailed Comparison of Deposit Mix
Deposit Cost is a Function of Price and Structure
High COF Member
Low COF Member
100
5
6
6
6
6
90
CDs over 1 yr
13
14
14
13
13
29
30
30
31
31
80
7
7
8
7
7
CDs 3-12 mos
70
19
22
19
60
19
21
CDs 50
12
10
11
8
10
40
Savings/MMDA
30
Checking
20
10





0
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
Low COF Member Has Smaller Proportion of
Longer-Term CDs
20
Summary of Case Study So Far

• High COF Member Median Deposit COF 4.86
• Consistently above district median COF
• CDs represent about 60 of deposits
• Has longer-term and more costly CDs
• Uses little wholesale funding (advances)
• Low COF Member Median Deposit COF 3.70
• Consistently below the district median COF
• Non-maturity funding represents 70 of deposits
• Has shorter-term and less expensive CDs
• Uses wholesale funding to supplement deposits
• How Does This Affect Their Ability to Compete for
  Loans?

21
Implications of COF ManagementBlended Funding

• 1 million loan pool
• 15 year, fixed-rate mtgs
• 5.46 yield

High COF Member - Average IBD COF 4.49 Low COF
Member - Average IBD COF 3.31
Note The analysis shown in this section is for
informational purposes only and is not a
recommendation of a specific funding
combination. Funding strategies are specific to
individual members.
22
Blended Funding Steps

• Understand Characteristics of Deposits
• How might deposits act as rates change
• Understand Asset Characteristics
• Assuming 100 PSA on loan pool
• Normally would present several scenarios
• Blend in Appropriate Funding
• Will use same advance combination today for each
  member
• Normally would present several combinations

23
26-Quarter COF History and Current Advance Curve
Which Advances Fit With Your Deposit COF History?
Maximum Quarterly IBD COF Last 26 Quarters 5.26
High COF Member
Avg IBD COF 4.49
Advance Rates
6.00
5.57
1 Std Dev 5.31
5.19
5.12
4.95
5.00
Likely range of future deposit costs
4.78
4.57
4.30
3.89
4.00
3.39
-1 Std Dev 3.66
June 2003 IBD COF
2.86
3.00
2.20
1.90
2.00
1.49
1.32
1.24
1.19
1.00
0.00
1 week
1 month
6 month
1 year
1 1/2 year
2 year
3 year
4 year
5 year
6 year
7 year
8 year
9 year
10 year
15 year
MMA 15/15
The shaded advances will be used in this example
to fund the loan pool.
24
Average deposit COF for member
High COF Member Deposits plus 2-3-5-7 yr Advance
Ladder
25
Summary of Scenarios
Comparison of Projected Weighted Lifetime Spreads
How do combinations look under different
deposit/prepay assumptions?
Wtd Avg Lifetime Spread
Spread
High COF Member
2.00
1.74
1.80

• Future Deposit Costs are Unknown
• Three deposit COF scenarios are used to produce a
  range of spreads
• -1 standard deviation from average
• Average IBD COF
• 1 standard deviation from average
• Advances, loan rate, and prepay speed are held
  constant

1.60
1.33
1.40
1.20
0.94
1.00
0.80
0.60
0.40
0.20
0.00
1
2
3
3.66
4.49
5.31
100 PSA
100 PSA
100 PSA
2-3-5-7 Ladder
Run Number, Deposit COF (constant for entire
life), Asset Prepay Speed
26
26-Quarter COF History and Current Advance Curve
Which Advances Fit With Your Deposit COF History?
Maximum Quarterly IBD COF Last 26 Quarters 4.27
Low COF Member
Note The 1 standard deviation for this member
is BELOW the average IBD COF for the other member.
Advance Rates
Avg IBD COF 3.31
6.00
5.57
5.19
5.12
4.95
5.00
4.78
4.57
1 Std Dev 4.27
4.30
3.89
4.00
Likely range of future deposit costs
3.39
2.86
3.00
-1 Std Dev 2.36
2.20
June 2003 IBD COF
1.90
2.00
1.49
1.32
1.24
1.19
1.00
0.00
1 week
1 month
6 month
1 year
1 1/2 year
2 year
3 year
4 year
5 year
6 year
7 year
8 year
9 year
10 year
15 year
MMA 15/15
These same advances for the prior example will be
used with this member.
27
Exceeds highest IBD COF for member over last 26
qtrs.
Average deposit COF for member
Low COF Member Deposits plus 2-3-5-7 yr Advance
Ladder
28
Summary of Scenarios
Comparison of Projected Weighted Lifetime Spreads
How do combinations look under different
deposit/prepay assumptions?
Spread
High COF Member
Low COF Member
2.50
2.37
Projected Spread Difference Using Respective
Average COF on Deposits 58 bps
1.91
2.00
1.74
1.33
1.44
1.50
0.94
1.00
0.50
0.00
1
2
3
5
6
7
3.66
4.49
5.31
2.36
3.31
4.27
100 PSA
100 PSA
100 PSA
100 PSA
100 PSA
100 PSA
2-3-5-7 Ladder
2-3-5-7 Ladder
Run Number, Deposit COF (constant for entire
life), Asset Prepay Speed
Same Advances, Different Deposit Costs Look at
Effect on Spreads!
29
Funding Strategies Are Not One-Size-Fits-All
Difference in 26 qtr IBD COF Statistics
1 Std Dev 104 bps
Average 118 bps
- 1 Std Dev 130 bps
IBD COF
High COF Member
Low COF Member
6.00
5.31
5.00
4.49
4.27
4.00
3.66
3.31
3.00
2.36
2.00
1.00
Blended Funding Strategies are Specific to Each
Member
0.00
Different Costs Different Approaches - Does
the low COF Member necessarily need the same
advances?
30
Blended Funding Conclusions

• Higher deposit COF levels can make it more
  difficult to compete for loans.
• May require additional interest rate risk
  protection.
• Difficult to charge more than competition for
  loans (pass on costs).
• Typically results in lower spreads or lost
  business.
• Lower deposit COF levels can enhance ability to
  compete for loans.
• Can offer loans at the going market rate and have
  wider spreads.
• Can charge less on loans and have better spreads
  than competition.
• Can build in more interest rate risk protection
  and still have better spreads.

Lending Decisions are Dependent Upon Your Funding
Decisions
31
Best Practices for Lower COFs

• Use a benchmark to provide deposit pricing basis
  beyond local market area (i.e. FHLB Advance
  Curve, bankrate.com).
• Avoid pricing up existing deposits.
• CD specials, manipulating maturities with rates,
  etc.
• Be willing to use other funding sources to
  supplement the deposit base.
• Is it better to keep all balances at a higher
  cost or MOST of them at a lower cost?

32
Case Study

• Member has historically had a high COF on
  deposits.
• Typically priced deposits with competition.
• Strategy Change deposit pricing/structuring.
• Set CD rates below advance curve.
• Focus on non-maturity deposits.
• Use FHLB for growth needs beyond natural deposit
  growth.

33
This member historically used advances for
liquidity and match funding now also using us
to better manage the much larger core deposit
base.
Borrowed repos rather than paying market rates
for CDs
Became liquid paid back repos
Decided to extend CDs
New Plan !
34
Case Study Pricing Below the Advance Curve
Changed Deposit Pricing and Supplemented Balances
With Advances
Average Interest Bearing Liabilities
Cost of funds 3.12
Peer Cost of funds 2.88
-85 bps
-68 bps
450,000
4.00
3.47
3.33
400,000
3.50
3.12
350,000
2.75
3.00
2.62
300,000
2.50
250,000
2.00
200,000
1.50
150,000
1.00
100,000
0.50
50,000
0
0.00
Jun-02
Sep-02
Dec-02
Mar-03
Jun-03
27 bp
25 bp
24 bp
9 bp
10 bp
270
249
251
96
106
Quarterly Rate Difference From Peer and Cost
(Benefit) in 000's - Not Annualized
35
Final Conclusions

• Rates are at some historical lows and may be
  poised to increase.
• Following traditional approaches of extending
  deposit terms before rates increase may not
  produce the desired effect.
• Exercising restraint on rates, marketing to
  non-maturity depositors, and supplementing with
  advances may provide better results.
• Your ability to compete for loans and to maintain
  attractive spreads depends significantly upon how
  well you can control your deposit costs.

36
Action Steps

• Understand the Performance of Your Deposit Base.
• How does your 26Q deposit history or peer
  analysis look?
• Are you modeling your deposit base correctly?
• Take Another Look at Your Funding
  Practices/Policies.
• Do they help or hurt your ability to manage your
  COF?
• Adjust your policies or strategies if necessary.
• Contact your FHLB RVP or the Financial Strategies
  Group for assistance in getting started.

37
Be Very Cautious About Borrowing LT Advances That
Exceed Your Historical Deposit Costs
38
About half of the long-term fixed (LTF) advances
have been originated since December 2000. 86 of
the LTF portfolio had remaining maturities of
less than 7 years