|
KAL_II Loan Servicing Valuation Manual
|
This chapter of the KAL_II Manual deals with the individual
portfolio segments. A segment is a subdivision of the total portfolio.
The loans in each segment will exhibit similar characteristics. When we
evaluate the loans as we treat them as if they were uniform in terms of
prepayments, servicing costs, foreclosure requirements, etc.
Every loan is certainly unique. This does not mean that we should evaluate
every loan as a separate segment. There is some number of segments that
makes sense for each portfolio. Too many segments will mean an unnecessary
amount of work. Too few segments will not provide the level of accuracy
we seek. The segmentation decision is up to you. Start out with as few
segments as possible. As you become proficient with the Model you can further
refine your segments.
This is a list of some of the commonly used segmentation
characteristics. If you are bidding on purchased servicing you may already
have the data broken into these categories.
1. Loan type
2. Investor remittance requirement
3. Interest Rate
4. Geographic Location
5. Delinquency ratio
A Loan Status Category is a period of time during which the servicing costs for a segment of loans remain relatively constant. The Model uses these statuses, which you define, to determine the servicing cost for each segment for each month of the loan life. A complete definition of a loan status includes the following information:
Description of the loan status
Beginning time period of the status
Length of time in months of the status
Beginning balance of the status in loans or percents
Payoff probability at the end of the status
Cost to process while in the status
The KAL_II Model is unique in its ability to simulate
a loan servicing operation. The simulation uses many variables that other
models do not and it may appear a formidable task to collect the data necessary
for a complete ealuation. Actually, most of the data required by the model
is readily available form the servicing and accounting departments.
Before you start collecting the data read through this chapter of the manual.
The concept of a loan segment is not new. You already have much of the
segment data. When new portfolios are being valued for a potential purchase,
most of the data is provided by the seller.
Data Input Forms
We have provided data input forms to assist you in the data collection
phase. Use these forms as a guideline and as a quick reference of what
is needed to run the Model.
Fill in Data Fields
We recommend that you fill in all the data fields, even
if the data variable is set to zero. This prevents incorrect data being
inadvertently used by the program. Step through each screen in the order
that they are presented. This will insure that no data fields are missed.
The KAL_II Model provides many powerful functions to consolidate
different loan segments. The Model provides the same investment criteria
for both the portfolio valuation and the group consolidation. All the financial
reports are available for both simulations.
Before you consolidate any portfolios, make certain that you have evaluated
and verified the individual components. The group consolidation is time
consuming process. The program evaluates each segment separately and then
adds the results to a group total.
Although it is possible to use a different loan status
configuration for each segment, it is easier to evaluate the simulation
results if all segments have similar statuses.
The KAL_II Model provides powerful budgeting functions
through the ability to start the servicing of a portfolio any time in the
next five years. This function is intended solely for budgeting purposes.
When you want to calculate a present value for a portfolio make sure you
set this field to zero.
GNMA GNMA I Example
Loan Portfolios
Change Data
How Categories Defined *
Amortization Factors
Beginning Status *
Cashiering Patterns *
Debt Financing *
Extra Income Sources *
Foreclosure *
Growth & Other Tables *
Impound
Late Fee *
Miscellaneous
Payoffs
Remittance Process
Service Fees
Transition and Cost *
Quit Menu
* - These fields can be similar for most portfolios being
evaluated by the same company.
A Loan Status Category is defined to be a period of time
during which the loan servicing costs are likely to remain constant for
each month in the loan status category. These categories are determined
by the delinquency structure and marginal cost structure of the individual
firm. During the remainder of the manual we will refer to loan status and
loan states. This does not mean the geographical state but rather the delinquency
state.
The categories can be longer than one month. A condensed status always
lasts for two or more months. Once in a condensed status, loans stay in
the condensed status for the full number of months. There can be a total
of six (6) loan status categories. The GOOD_LOAN status and the OFF_BOOKS
status are automatically assigned by the program.
At the completion of the loan status category there are three situations that can occur for a loan:
the loan becomes current
the loan pays off
the loan moves to the next status.
When the last status is complete (FORECLOSE category),
the loans are either removed from the books (OFF_BOOKS category), reinstated
or paid off. Always define the last status as OFF_BOOKS. This tells the
program that no more categories are required.
In the example, which is used throughout the manual, it is assumed that the mortgagor's payment is due on the first day of the month. A loan is one month delinquent on the 32nd day after the payment was due. We call the loans that are late during the first month the "thirty day delinquents".
There are two types of loan status categories. The first is one month long and it is called a regular category. The second type can be longer than one month and it is called a condensed category. The condensed categories are different for the following reasons:
1. A loan stays come current or payoff from the within
the condensed category.
3. The probability ratios, loan processing costs and payoff ratios which
apply to this status take all the months of the status into consideration.
For example, the probability of a loan in the GT_90_Day status in the condensed
category until the status time period completely elapses.
2. Loans can not going from three months delinquent to four months delinquent
is 100%. It is only as the loans leave the status that the payoff ratios
and probability of moving to the next status apply.
We have provided some suggested loan status categories that are easy to use and that most mortgagees feel comfortable with. Remember, a loan cannot come current from the fourth month of delinquency if it is in the GT_90_DAY status; it can only go to the fifth month of delinquency and then to the sixth month. At the point the loan leaves the GT_90_DAY status the probability of paying off, the probability of coming current and the probability of moving on to Next status will all apply.
It is possible to use different categories for different portfolio segments. The program will consolidate the portfolios correctly with very little loss of detail information.
We have provided a sample portfolio file that is composed of GNMA I loan servicing. This file can be used as a sample file while reviewing this manual.
The program automatically assigns the first status as GOOD_LOAN. This status includes all payments that are made in the month that the payment is due. The mortgagor may pay anytime from the first day to the last day of the month. If the mortgagor's payment was not received on the first day of the month it was due, the following may occur:
Day 10 - the mortgagor is sent a letter reminding him/her
that their payment has not been received.
Day 15 - A collector calls the mortgagor requesting payment before the
Late Charges are posted.
Day 17 - A late notice is sent explaining to the Mortgagor that they are
now responsible for a late charge.
Day 23 - a possible default notice is sent to the Mortgagor.
Day 31 - A Demand Letter is sent notifying the mortgagor that only two
payments, the full delinquent amount, will now be accepted.
This one month period would represent a "loan status category". The status would have a time span of one month and a related cost associated with this month. All costs, including customer service, collections (for the first month), escrow and related administrative costs are included in the cost to process this status (GOOD_LOAN). The costs for this status can be assigned under the ten day collection cost.
Assume the firm spent $32 to process the loan during this month (GOOD_LOAN). This cost does not include costs which the program handles specifically such as pay-off cost, set-up cost and late charge notice cost.
During the second month (which we will always refer to as the First Month of Delinquency) if the Mortgagor has not made his payment, the following events may take place:
The mortgagors are contacted and the seriousness of the
delinquency is explained to them.
A property inspector is sent to the home to determine the condition of
the property.
Demand letters may be prepared and sent.
There is increased contact with the mortgagor especially if there is a
sincere intent on the part of the mortgagor to bring the loan current.
We define this period as a second loan status category, DEL_1_MON. The marginal costs during this period are estimated to be $4.00. This includes the cost of the property inspection, collection calls, etc.. The servicing cost vary significantly from the Good_Loan status and it is important to establish the first month of delinquency as a separate loan status.
This is the second month of delinquency. Additional costs, collection notices, are usually absorbed during the second month in addition to the costs that were absorbed in the first month.
During the third month of delinquency the servicer is usually preparing the loan to be transferred to the foreclosure department. Additional contact with the mortgagor may be attempted. There is usually additional expense in monitoring the property through external agencies or using the servicer's own personnel.
Sometime during the delinquency period the costs of servicing
the delinquency from month to month may become uniform. During the fourth
to the sixth month of delinquency, while the foreclosure process is taking
place (after the collector has determined that the mortgagor is not able
to make any more payments), the costs may be uniform and consistent from
month to month. It would be logical to combine these three months into
one loan status category and determine what the total marginal cost to
process the loan during this period (3 months) will be.
If there are additional costs that need to be accounted for only in the
fourth month of delinquency then a 120 Day loan status category could be
created and the relevant inputs developed.
Loans which enter the foreclosure loan status stay in this condensed status for six months. The probability of payoff for the first five months of the status is zero.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Definition of Loan Status Categories
Status Length in Months
0 Good Loan 1
1 Del 1 Mon 1
2 Del 2 Mon 1
3 Del 3 Mon 1
4 GT 90 Day 3
5 Foreclose 6
6 Off Books 0
7 0
Define a title for each loan status. Make the title name meaningful since it is used in several places in the program. Loan status categories should cover periods which require similar administration costs.
Define the number of MONTHS that a loan stays in this
status before moving to the next status. The length of time in a particular
status will directly affect the cost to service that loan. By shortening
the time a loan spends in foreclosure you directly effect the total cost
to foreclose that loan.
It is possible to completely redefine the categories that are presented
in this example. These categories represent an acceptable level of detail
and seem to match most delinquency reporting categories. The program will
automatically adjust to whatever categories are used.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Amortization Factors
Title of Portfolio: Standard Model Comparison
Number of Loans: 10000
How Enter Balance: 1 (1=Average, 2=Total)
Loan Balance 50000 (Prior to This Month's Payment)
Original Amort Pd: 360 (Months)
Original Maturity: 360 (Months)
Average Loan Age: 12 (Months = Maturity-Remaining Life)
Portfolio Category: P (P=Purchase, E=Exist, N=New Prod)
When Become Active: 0 (Number of months in future,0=Now)
Type of Loans: 1 (1=Fixed, 2=GPM, 3=ARM)
Fixed Interest Rate: .11 Annual Percent
This title may contain any characters. It will appear at the top of the screen when you are working with this file and on any reports generated using this file. Make the title descriptive of the portfolio. This portfolio is composed of GNMA I type loans. If the portfolio segment is composed of adjustable rate mortgages and it is the second segment to be evaluated then an appropriate name might be ARM Segment 2.
The number of individual loans in the portfolio is used in many of the calculations. The entry above is for a segment of 10,000 loans.
This field allows you to enter the portfolio loan balances as either an average loan amount or as the total for the portfolio.
(1) Loan - Enter one "1" if you want to enter
the loan balances as an Average Loan Balance.
(2) Total - Enter two "2" if you want to enter the balances as
the Total Dollar Balance of the portfolio.
The Loan Balance is the amount of loan principal outstanding on the day the simulation begins, before receiving any loan payment due the first month. This is the amount prior to this month's payment. Since a "1" was enter in the prior input field this is the Average Loan Balance, $50,000.
The original amortization period is used to compute loan payments, interest and principal. This provides a means to provide for only partial amortization. The model will allow amortization periods up to 360 months (30 Years).
The original maturity period is used in connection with the amortization period to determine loan payments, interest, etc.
The average loan age determines the number of months that have passed since the first payment was made. It must be less than the original loan maturity. The average loan age is equal to the original maturity less the remaining life. Use the full Loan Life even if the loans are liable to balloon (become fully due) before the end of the loan maturity. This example shows that 12 months (one year) has passed since the mortgagor began making payments.
This determines where the loans in the portfolio segment came from. Enter:
(P) for loan servicing that is purchased from an external
company.
(E) for loans in an existing portfolio.
(N) for loan servicing that is added through new production.
Enter the month of the simulation when the portfolio becomes
active. This is the number of months from today. Enter "0" if
the loans are being added this month or are in the existing portfolio.
If the portfolio is to be purchased and servicing to begin six months from
today then enter "6" in the field.
Present Value - Whenever the present value of the portfolio is being determined
it is important to enter "0" in this field. The portfolio valuation
will not take place if there is anything but zero in this field.
This determines the type of loans that are in this portfolio segment. If the portfolio contains more than one type of loan then the portfolio should be segmented and each component analyzed separately. The group simulation should then be performed. The group simulation will add each segment together regardless of loan type.
(1) FRM - Enter a one "1" if the loans are
Fixed Rate Mortgages (FRM's).
(2) GPM - Enter a two "2" if the loans are
Graduated Payment Mortgages (GPM's).
(3) ARM - Enter a three "3" if the loans are
Adjustable Rate Mortgages (ARM's).
This is the average interest rate for the portfolio. If several different interest rates are being used, it is best to segment the loans by similar rates and use the group simulation described in Chapter 7. Enter the rate as a percentage; i.e. 9.5% is entered as 0.095000.
To use this section of the program it is necessary to take a representative GPM loan and use that specific loan as a basis for the evaluation. There can obviously be many different types of GPM loans. By varying the Inputs you should be able to determine how sensitive the Present Value is to changes in those inputs.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Title of Portfolio: GNMA I Example
Number of Loans: 100
How Enter Balance: 1 (1=Average, 2=Total)
Loan Balance 50000 (Prior to This Month's Payment)
Original Amort Pd: 360 (Months)
Original Maturity: 360 (Months)
Average Loan Age: 12 (Months = Maturity - Remaining Life)
Portfolio Category: P (P=Purchase, E=Exist, N=New Prod)
When Become Active: 0 (Number of months in future, 0=Now)
For Graduated Payment (GPM) Loans
Loan Interest Rate 0.000000
Current Payment 0.00
Annual Pmt Increase 0.000000
Number of Increases : 0
Month Next Increase : 0 (After Start of Portfolio)
This is the average interest rate of this portfolio segment. If the interest rates vary significantly it might be best to segment the loans into more than one GPM portfolio segment. If the average current loan rate was 9.75% you would enter this number as 0.097500 into the input field.
This is the average current Principal and Interest (P&I) payment for the loans in the portfolio segment. In this example the average payment per loan in the segment is $575.00.
This is the average annual payment increase calculated
as a percent of the monthly P&I payment for the loans in this segment.
The portfolio can be analyzed and an average of all loans in the portfolio
segment can be used. If the average for the segment is 15%, then enter
this number as 0.15000.
This is the number of increases that remain according to the GPM loan program. In this example there is a total of 6 increases allowed over the life of the loan.
This is the first month after the loans become active, if purchased, that the payments are allowed to increase.
There are many different types of ARM products available
to mortgagors. As a result of a lack of standardization in the mortgage
industry the evaluation of an ARM portfolio will be substantially more
time consuming than that of a Fixed Rate portfolio Analysis. You are encouraged
to develop as much detail as possible for this analysis. A typical large
Mortgage loan servicer may have as many as 60 different ARM products. As
you become familiar with the Simulation you will be able to determine how
these ARM products can best be segmented and evaluated.
For purposes of evaluation it is important to understand how the inputs
for the ARM's affect the portfolio Value.
Cost to Process - These types of loans usually require a significant amount of additional effort in relation to the normal Escrow and Customer Service work. In addition, the ARM Indexes must be maintained and checked on a regular basis. Companies often have a separate group of employees who devote their time specifically to ARM Servicing. These additional employees would be taken into consideration in the cost to process Section of Inputs.
Cash Flow - Since the ARM interest rates may change every year there will be a corresponding change in the available P&I float over the same period. An increase in the ARM Index will cause an increase in the P&I Float and therefore earnings on P&I float.
Delinquency - Should the ARM index rise too high there could be an increase in the level of delinquency and foreclosure due to unexpected high monthly payments. Payoff and collections costs would then increase significantly.
Payoffs - During periods of decreasing interest rates the mortgagors may take advantage of the lower rates and refinance to a Fixed Rate Mortgage. This would significantly increase the payoff ratio.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Title of Portfolio: GNMA I Example
Number of Loans: 100
How Enter Balance: 1 (1=Average, 2=Total)
Loan Balance 50000 (Prior to This Month's Payment)
Original Amort Pd: 360 (Months)
Original Maturity: 360 (Months)
Average Loan Age: 12 (Months = Matur - Remain Life)
Portfolio Category: P (P=Purchase, E=Exist, N=New Prod)
When Become Active: 0 (Number months in future, 0=Now)
ARM (Adjustable Rate Mortgages
Key to Index Used : 1 1=Market, 2=Table)
Month of Next Change: 0
Months Between : 0
Current Loan Rate 0.000000
Max Ann Rate Change 0.000000
Max Possible Rate 0.000000
Min Possible Rate 0.000000
Margin from Index 0.000000
The Index to the change in the ARM interest rate can be tied to either the market rate or to a separate table that follows this screen.
(1) Market Index - Enter a "1" if you wish the
program to use the Market Index (entered under the economic input screens)
as the index for the ARM annual Index.
(2) Table - Enter a "2" if you wish to enter your own ARM index
Table. If this option is selected the program will bring up the alternate
table for the User to input a set of ARM indexes. The program will adjust
the ARM interest rates to this ARM index table.
This is the month following the START of the portfolio that the index will first be changed. If the portfolio is not going to be started for six months the month of next change will need to be adjusted accordingly. For example: If today is January 1 and the portfolio will be started on June 1 and the payment is due to change on September 1, then the month of next change is 2 (Two). In our example we assume the portfolio is started in the first month (0) And the month of next change is 12 (one year).
This is the number of months between the change of the loan rates based on the ARM index. A one year ARM would have a value of 12 months or one year.
This is the current average loan rate on the ARM portfolio segment. The number "0.10000" represents a current loan rate of 10%.
This is the Maximum Annual Rate of change by which the loan principal and interest payment can be adjusted. An ARM with a 10% rate of change would have an input value of 0.10000 or 10%. This would mean that the ARM P&I payment could be adjusted annually up or down by no more than 10% of the current P&I payment amount.
This is the highest rate possible that can be charged for the ARM. A maximum possible rate of 14% would mean that the loan can not be adjusted to an annual interest rate above 14%.
This is the minimum rate that the mortgagor can expect to pay on this ARM loan. The loan will not be adjusted to an annual rate below this number. If the minimum possible rate is 8% then the loan will never be adjusted to a rate less than 8% regardless of the type of index used.
This is the percentage over or under the index that will be used to compute the actual annual rate charged on the ARM loan. If the margin is +0.02000 (2%) and the index is 10% then the ARM rate used to calculate the amortization schedule for this loan will be 12%. The index used is determined by the index key as entered above.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Index Table for ARM Loans
Thru Month Annual Rate
12 0.0800
24 0.1000
60 0.1200
360 0.1000
Starting from Month 1 input a schedule of ARM indexes to be used in calculating the ARM amortization schedules. If there was to be no change in the index then the thru month is "360" and the annual rate would be 8%. This would leave the loan rate as 10% (8% plus margin from index of 2%) which is the current loan rate.
If loan interest rates are expected to change at the end of year 1 to 12% then the first two entries in the table would be
(12, 0.08000) and
(24, 0.10000).
In this section you define the beginning loan status categories.
The program will distribute the loans into the status categories defined
in the previous section.
We have provided two separate means to accomplish this. First, you can
enter the beginning distribution as delinquency percentages. These percentages
are usually available from monthly delinquency reports. Second, you can
enter the actual loan counts for each status. It may be necessary to use
this method if the portfolio has been segmented and the delinquency ratios
are not available for each segment. The program will then convert the status
counts into status ratios.
As you enter the percentage or number of loans in each status, the program
will tabulate the amount remaining to be distributed. You will not be able
to exit this screen until all loan status categories have been given values,
the total of all percentages is equal to 100% or the all the loans have
been allocated.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Beginning Loan Distribution
Initial Loan Status Entered As -->> 1
Number of Loans: 10000.00
Status Percentage
Good Loan 1.00000
Del 1 Mon 0.03000
Del 2 Mon 0.02500
Del 3 Mon 0.02000
GT 90 Day 0.01500
Foreclose 0.01000
(Undefined) ______
Entry-->> <<
(Pct Left to Allocate 0.000000)
This is the percentage or number of loans in each loan status. The loans can be entered using either method. The entry method depends on the data available for the simulation.
(1) Percent - Enter a "1" to use Percent of
loans in each status category. In this instance you would enter a set of
delinquency ratios such as:
(2) Number - Enter a "2" to use the Number of loans in each status
category. If you want to enter the total loan counts in each delinquency
status then enter the number of loans into the input field.
The total number of loans in the portfolio is established in the amortization section of the Model. If the percentage method is chosen the program will expect you to allocate 100% of the loans to one or a combination of the categories
The selection of a "2" in the Initial loan status
field will cause the program to show the number of loans that need to be
allocated. You do not enter this field. The program calculates the amount
remaining to allocate as a convenience to assist in allocating all loans
that are in the portfolio.
Cashiering patterns describe the timing of the monthly collections from the mortgagors. There are four patterns that the Model explicitly examines. The first, and most important, is the collections of the current loans. This information can be approximated by using the monthly delinquency report to determine how many loans are current on each day of the month. This should give a reasonable estimate of the mortgagors payment habits. We would like to make decisions that result in the mortgagors paying earlier in the month. The Model will give you a financial estimate of the value of these decisions. Earlier payment of the mortgagor's funds should increase the value of the servicing.
<$&s4140[v]>
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Cashiering Patterns
-->> Collections on Current Loans
Delinquencies Coming Good
Foreclosures
Payoffs
This is the Cashiering Patterns Change Data screen.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Cashiering Patterns Loans Staying Good
Day Fraction Day Fraction Day Fraction
1 11 21
2 12 22
3 13 23
4 14 24
5 15 25
6 16 26
7 17 27
8 18 28
9 19 29
10 0.50 20 0.75 30 1.00
Entry-->> _____________ (0.0-1.0) %
Indicate when during the month funds in this category
are deposited in the bank by entering the fraction of the total amount
that has been deposited by the end of the day. Enter only the days for
which you have data. It is necessary to move the cursor to the day you
want to enter and then enter the percentage on the line that is titled
"Entry". The program will interpolate between your entries. To
blank out an entry, enter minus one (-1) fractions must be increasing.
Press {ESC} when finished.
An entry of 0.50 on Day 10 would mean that 50% of the money is received
by the 10th of the month. A second entry of 0.750 on the 20th of the month
would mean that 75% of the funds were received by the 20th. The program
will automatically adjust to 100% as of the end of the month. If you want
all the funds received as of the end of the month, put 0.00 on the 29th
of the month and 1.00 in the 30th of the month.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Cashiering Pattern - Delinquents Coming Good
Day Fraction Day Fraction Day Fraction
1 11 21
2 12 22
3 13 23
4 14 24
5 15 25
6 16 26
7 17 27
8 18 28
9 19 29 0.00
10 20 30 1.00
Entry-->> _____________ (0.0-1.0) %
Indicate when during the month funds in this category
are deposited in the bank by entering the fraction of the total amount
that has been deposited by the end of the day. Enter only the days for
which you have data. The program will interpolate between your entries.
To blank out an entry, enter minus one (-1) fractions must be increasing.
Press {ESC} when finished.
In this example all the funds are shown being deposited on the last day
of the month. It may be that the collections for this loan status category
are focused on month end due to the intensive late month collection efforts.
To see how your entries are translated by the program you should check
the Rates and Factors (R&F) screens in the portfolio simulation section
starting with Screen {S6-110}. These screens outline every day of the month
and the Total Percentage received Month-To-Date.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Cashiering Pattern - Foreclosures
Day Fraction Day Fraction Day Fraction
1 11 21
2 12 22
3 13 23
4 14 24
5 15 25
6 16 26
7 17 27
8 18 28
9 19 29
10 20 30 1.00
Entry-->> _____________ (0.0-1.0) %
Indicate when during the month funds in this category
are deposited in the bank by entering the fraction of the total amount
that has been deposited by the end of the day. Enter only the days for
which you have data. The program will interpolate between your entries.
To blank out an entry, enter minus one (-1) fractions must be increasing.
Press {ESC} when finished.
In this example the funds are assumed to be received in an even distribution
throughout the month. The average cash available from foreclosures would
then be one-half of the total amount received. The Program uses this information
to calculate the average bank balances and ultimately the earnings on P&I
float. An improvement in a cashiering pattern (money which is collected
sooner in the month) can be shown to produce additional bank balances and
therefore additional earnings.
Vary these collections patterns to determine the value of additional, incremental
collections efforts.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Cashiering Pattern - Payoffs
Day Fraction Day Fraction Day Fraction
1 11 21
2 12 22
3 13 23
4 14 24
5 15 25
6 16 26
7 17 27
8 18 28
9 19 29
10 20 0.50 30 1.00
Entry-->> _____________ (0.0-1.0) %
Indicate when during the month funds in this category
are deposited in the bank by entering the fraction of the total amount
that has been deposited by the end of the day. Enter only the days for
which you have data. The program will interpolate between your entries.
To blank out an entry, enter minus one (-1) fractions must be increasing.
Press {ESC} when finished.
In this example we assume that 50% of the funds are received by the 20th
of the Month. Even though there is no additional Input, the program will
interpolate between the 20th and the last day of the month. The second
50% of the funds will be received in ten equal daily amounts as calculated
by the Program.
Many models perform some degree of analysis concerning
the effectiveness of different financing alternatives. The KAL_II Model
performs an after debt service analysis that includes the determination
of the present value, internal rate of return and break-even price. A financial
screen is available for you to examine the monthly effects of reinvesting
the cash generated from the loan servicing.
The factors in this section describe the financing structure used to purchase
servicing portfolios. It is assumed that the firm has two sources of capital,
debt financing and equity capital. The program allows a wide variety in
the timing of the repayment of the debt as well as the interest charged.
It is also possible to directly allocate the purchase price to debt financing.
This usually results in a lower financing cost since debt is almost always
less expensive than equity. These inputs are used only in after-debt, or
budget analysis.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Debt Factors
Fraction Financed with Debt : 0.7500 (0.0 to 1.0 Purchase Price)
Key to Debt Interest Rate : 1 (1=actual,2=spread market)
Months Between Principal Pmts: 12
Key to Repayment Method : 2
(1 = Use up available funds )
(2 = Equal Principal Payment)
(3 = Equal to Amortization )
For Equal Payments:
Number of Years of Payments : 12
Enter the fraction of the purchase cost which will be financed with debt. The fraction is expressed as a decimal with a value between 0.0 and 1.0 (e.g. 50% as .50). In our example, 75% of the purchase price will be financed with debt.
The key determines whether the debt table is a specific table or a spread from the market index.
(1) Annual Debt Table - Enter a "1" to input
the cost of debt using the Annual debt Rate Table. This would be the instance
where the Firm was able to arrange Long Term financing at predetermined
Interest Rates.
(2) Market Index - Enter a "2" for the debt table to be a spread
from the Market Interest Index.
Enter the months between principal payments on the debt. An entry of 12 months wuld result in a single payment once a year.
There are three methods available to retire the debt:
(1) Available Funds - Enter a "1" if debt repayment
will be from available funds. All the available funds, after taxes are
paid, are used to retire the debt.
(2) Equal Principal Payments - Enter a "2" if equal principal
payments will be made. In our example we have defined a fixed rate loan
with predetermined interest rates and equal, annual principal payments.
(3) Amortization Schedule - Enter a "3" if the repayment schedule
is going to exactly match the amortization of the servicing investment.
This will cause the debt to be repaid monthly and for the same number of
years as the loan amortization.
Enter the number of years over which the debt will be repaid. This only applies to repayment method 2.
In our example the 12 year repayment schedule might indicate an attempt to match the often used FHA 12 year life Assumption.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Annual Debt Rate
Thru Annual
Month Rate
360 0.10000
Up to ten (10) separate periods can be entered in the rate tables. Put the number of the month that the rate changes and the new borrowing rate. Enter the new annual as a percentage (0.10 is 10%) in the annual rate field.
The bank in this example has agreed to a 10% fixed interest rate for twelve years.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Extra Income Sources
Income that applies to all loans--------------------------
Extra Income : 3.00 ($/Loan/Year)
Key to Growth : 2 (1=Inf, 2=Table)
Customer Base Cross Selling (Servicing Premiums)----------
Monthly Premium : 15.00 (per Loan )
Commission : 0.370000 (Fraction of Premium)
Key Premium Growth: 2 (1=Inf, 2=Table )
Current Balance : 25000
These fields define the income that applies to all loans. The amount you enter in these fields will be applied to all the loans in the portfolio. When you complete this section the premium penetration table, ancillary income growth table and the premium growth tables follow this screen.
Enter the average dollars per year that will be earned
on all loans still on the books even if they are delinquent. This may be
fees from selling mailing lists, promotions or from marketing services
that you do not provide. It may also include moneys earned for loan payoff
fees or document processing fees.
If there are 10,000 current loans in this portfolio segment, then each
loan will contribute an additional $3.00 to income; i.e. $30,000 for the
year. This amount will be included in the earnings screen as additional
income. Remember, the amount you enter in this field will be automatically
applied to EVERY LOAN in the segment.
There are two ways to determine the growth in these earnings.
(1) Inflation Rate Table - Enter a "1" and the program will use the general inflation rate. The premiums in our example will grow at a rate equal to the inflation rate. If you are investigating the effects of inflation, it is best to tie all variable inflation rates to the general inflation rate. Changing one table, the general inflation table, will automatically change all other tables.
(2) Ancillary Income Growth Table - Enter a "2" and the program will ask you to input a table of growth factors for this category of earnings.
These factors affect the insurance premiums sold to a PORTION of the mortgagors.
Generally, insurance premiums are paid to the loan administration area from the insurance company or agent. If the premiums are received by some other department of your company you still may want to include the benefit in the evaluation of the portfolio.
Enter the average monthly premium you collect from the mortgagor as of the start of the simulation. These are payments that you collect, and remit to the insurer at periodic intervals.
Enter the fraction of the monthly premium that you take as commission for collection the premiums for the insurer.
This field will determine how the premiums grow over the life of the loans.
(1) Inflation Rate Table - Enter a "1" and the
program will use the general inflation rate.
(2) Premium Growth Table - Enter a "2" and the program will ask
you to enter a table of growth rates for the amount of monthly premium.
This is the current dollar balance in the premium impound account. The amount is entered in dollars. This will be the base amount for interest earnings until the entire account is remitted to the insurer using the remittance schedule for insurance premiums.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Premium Penetration Table
Thru Annual
Month Rate
12 0.10000
24 0.20000
360 0.25000
This table will appear after you enter the data in the
extra income section. Use {Page Down} key if the tables do not appear.
This is the percentage of the portfolio that will generate insurance premium
income. In the following example only 10% of the loans will pay premiums
the first year (12 months). Due to an anticipated marketing program the
penetration is expected to move to 20% the second Year and finally to 25%
in the fourth through the 30th year.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Ancillary Income Growth
Thru Annual
Month Rate
360 0.04000
This table will determine how the ancillary income premiums
will grow over time. If you do not want the premiums to grow according
to the general inflation rate then use this table to set up the expected
growth rate of the premiums.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Premium Growth
Thru Annual
Month Rate
360 0.05000
This table describes how the dollar amount of the monthly
premiums from the optional insurance program will grow over the life of
the loan. It is possible to have this growth tied to the general inflation
rate.
The program uses several different methods to represent
foreclosure costs. The first is to simply record all costs and losses as
a single number; hard costs. If the hard cost is the amount of the actual
foreclosure loss then no hard cost would be recovered.
Next, it is possible to estimate the P&I and T&I advance that is
not recovered when the loan is foreclosed. This requires a greater understanding
of the foreclosure costs. The interest payments lost and principal balance
lost fields can be used together with, or instead of, the P&I and T&I
advance fields. Define whichever scenario that has the most meaning to
your particular situation.
The Model supplies different methods to analyze foreclosure costs because
these costs are large and they often have a serious effect on the portfolio
valuation. Use the Model to measure the effects of different foreclosure
agreements you might use in the sale or purchase of loan servicing.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Foreclosure Factors
Hard Out-of-Pocket Cost : 350
Hard Costs Recovered : 0.650000
Service Fees Recovered : 0.000000
P&I Advances Recovered : 0.900000
Tax & Ins Advances Recov : 0.900000
Interest Payments Lost : 2
Principal Balance Lost : 0.010000
This is the real dollar amount paid to people and companies
outside the firm to foreclose a loan. This would include legal fees, interest
lost, REO costs incurred, property management not recovered, etc.
This is the fraction of the hard dollar amount which you
will get back after foreclosure. Enter it as a decimal (e.g. 10% as .10)
Enter the fraction of the service fees which were not
received when the loan was delinquent which you will get at the end of
foreclosure. Enter it as a decimal (e.g. 10% as .10)
Enter the fraction of the P&I which was advanced while
the loan was delinquent which you will recover at the end of foreclosure.
Enter it as a decimal (e.g. 10% as .10)
Enter the percentage of Tax and Insurance payments which
were advanced while the loan was delinquent and which you expect to get
back when the loan completes foreclosure. Enter it as a decimal (e.g. 10%
as .10)
Enter the number of interest payments which are "lost"
as a result of foreclosure. This is a function of government loan guarantee
procedures. Enter "0" if no interest payments are lost.
Enter the fraction of the loan principal which will be
taken as a loss at the end of foreclosure. This is a good place to show
the effect of miscellaneous costs which do not really fit in other areas.
Enter a fraction as a decimal (e.g. 1% as .01)
This is a summary of the many tables that used to calculate
the different growth factors. They are entered in this section ad a convenience
to the User. Each table can be found in the appropriate section of the
model. The tables can also be examined in detail in the rates and factors
section of the portfolio simulation section. These are tables that are
used in the program to calculate growths rates and interest rates. They
can be entered here or in the appropriate section.
This is the only place that the Servicing Cost table can
be entered. Check the factors and rates section to be certain that the
table has been entered as intended.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Growth & Other Tables
Ancillary Income
Debt Interest Rate
Extra Premium Growth
Insurance Impound Growth
Premium Penetration
-->> Servicing Cost Growth
Tax Impound Growth
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Growth Service Costs
Thru Annual
Month Rate
360 0.050000
Enter the month and the annual percentage growth that
is expected in servicing costs. The rate you enter is actually applied
on a monthly basis at 1/12 of the annual rate. In this example the servicing
costs would grow at .417% per month (5% / 12 months)
Impounds
GNMAEX GNMA I Example
Loan Portfolios
Impound Inputs
-->> Balances and Indexes
How Paid Out During Year
Insurance Growth Rate (if not inflation)
Property Tax Growth Rate (if not inflation)
Rate Paid to Borrowers on Impounds
Quit Menu
We have several objectives to accomplish in this section
of the program:
We must determine what the initial impound balances should
be. If you know the exact amount then the task is easy. Because of the
annual remittance cycle for impounds it is usually difficult to find the
average annual balance. The KAL_II program will assist you in determining
what the annual balances should be under the simulated conditions.
It is important to consider the value of the impounds in the present value
analysis. The KAL_II program deals with these impounds in a very detailed
manner. We often need to make estimates of these balances when analyzing
a portfolio for purchase. In this case it is best to take rates and averages
based on your own experience with the different types of servicing under
investigation.
We want to determine what the tax and insurance balances
will be when we take the delinquency information into consideration. Most
programs will not allow any variation in the impound accounts as a result
of changes in the delinquency ratios. If the delinquent ratios rise by
2% impounds in the accounts will drop.
It is always in the best interest of the firm to collect all the escrows
that are due. An escrow analysis is usually performed on an annual basis
to make certain that sufficient funds are being collected to pay the individual
mortgagor's current tax and insurance bills.
In conjunction with our economic scenario we want to estimate
how the impounds grow and do they grow on an annual basis or a monthly
basis. The tax impounds may not grow as quickly as the hazard insurance
impounds. These rate scan be set to different growth rates.
The remittance schedule for each type of impound account
need to be determined. We use three separate schedules to remit the impounds:
a. Tax impounds are usually remitted once a year. IN the
last several years it has become popular with different axing authorities
to begin biannual or tri-annual remittances. Some counties will accept
monthly remittances. By using different schedules you can determine the
effects on earnings from these decisions by the taxing authorities.
b. Hazard insurance premiums are usually remitted on an
annual basis. The remittance month coincides with when the home was purchased.
It often happens that the heavy remittance months for hazard insurance
occur in the summer months. If you track your escrow accounts on a monthly
basis you will have enough information to determine what percent of the
insurance is remitted in what period.
c Impound interest will be paid according to the remittance schedule you
define in this section.
d Special insurance premiums are remitted on a regular monthly basis as
they are collected.
Every state has different regulations concerning interest
payments to the mortgagor on the impounds accounts. You can set the impound
interest paid in several ways. If you expect to pay interest on all impounds
all the time then set the rate in the Firm section of the program. If you
are going to pay interest only in certain states then set the interest
rate in the portfolio section.
Cashiering Patterns
Standard Model Comparison
Loan Portfolios
Change Data
Impound Information
Monthly Property Tax : 50.00 (Average Per Loan )
Key to Growth Rate : 2 (1=Inflation, 2=Growth Table)
How Tax Impound Grows: 2 (1=Monthly, 2=Annually )
Monthly Insurance Pmt: 30.00 (Average Per Loan )
Key to Growth Rate : 2 (1=Inflation, 2=Growth Table)
How Ins Impound Grows: 1 (1=Monthly, 2=Annually )
Initial Impound Balance: 1 (1=Calculate, 2=Input)
Extra Months Tax : 2
Extra Months Ins : 2
Distribute Total Impound : Y (Y=Yes, N=No)
If Yes, Actual Impound Balance 4,800,000
This is the average monthly tax payment that is deposited
into the escrow account. This money is available to be used to offset or
draw interest.
Either the inflation table or a specific can be used to
determine the growth rate in the tax impound account. This assume that
property taxes will grow over the next years.
(1) Inflation Rate - Enter "1" to use the inflation
rate table.
(2) Specific Table - Enter "2" to use the separate property tax
growth table.
The impounds can be set up to grow on a monthly or on
an annual basis.
(1) Monthly - Enter 1 "1" to have the tax impounds
grow on a monthly basis using the annual rate table.
(2) Annual - Enter a "2" to have the tax impounds grow on an
annual basis.
This is the average monthly hazard insurance payment in
dollars per loan.
Key to Insurance Growth Rate
Either the inflation table or a specific growth table
can be used to determine the growth in the insurance impounds.
(1) Inflation Rate - Enter "1" to use the inflation
rate table.
(2) Specific Growth Table - Enter "2" to use the separate Hazard
insurance growth rate table.
The Impounds can be set up to grow monthly or annually:
(1) Monthly - Enter "1" to have the insurance
impounds grow on a monthly basis using the annual rate table.
(2) Annual - Enter a "2" to have the insurance impounds grow
on an annual basis.
Calculate (1)
The program will calculate what the initial balance in
the impound account should be based on the impound payment table and the
average payment size.
This is the number of extra months of property tax that
are collected in the monthly payments.
This is the number of extra months of Hazard insurance
that is collected during the year.
Distribute Impound Balance (Yes)
This is the combined initial impound balance. The program
will distribute this amount between the tax and hazard insurance accounts
based on the monthly payments and extra months of impounds held.
Distribute Impound Balance (No)
The program will calculate what the initial impound balances
should be for both the tax and insurance impound accounts.
If you know the beginning impound balances, enter them
directly in this section. Enter the tax and insurance impounds separately.
The previous amount entered for impound balances was for the combined total
of both accounts.
This is the actual tax impound balance in the bank account.
This is the actual hazard insurance impound balance in
the bank account.
Impound Payments
Number of Required Payments to be Paid Out in Each Future Month
Property Tax Acct Insurance Acct Impound Interest
Month Payments Month Payments Month Payments
1 - 0 1 - 1 1 - 0
2 - 0 2 - 0 2 - 0
3 - 0 3 - 1 3 - 0
4 - 6 4 - 0 4 - 0
5 - 0 5 - 0 5 - 0
6 - 0 6 - 0 6 - 0
7 - 0 7 - 8 7 - 0
8 - 0 8 - 1 8 - 0
9 - 0 9 - 1 9 - 0
10 - 0 10 - 0 10 - 0
11 - 6 11 - 0 11 - 0
12 - 0 12 - 0 12 - 12
This is the number of remittance payments that will be
paid in each future month. There is a maximum of twelve (12) months to
assign in the tables. This represents 100% of the payments received during
the year.
1. For equal payments during the year, enter 1 in each
month.
2. For semi-annual payments, enter 6 in two of the future months.
3. For annual payments, enter 12 in only one future month.
This is the number of insurance payments (1-12) that will
be made in the months following the start of the portfolio. The months
do not represent January through December. The months represent the months
from the start of the portfolio.
This is the number of interest on impounds payments that
must be paid to the mortgagors. Enter the pattern that will repeat in the
future. The example assumes that all interest on impounds will be paid
12 months after the start of the portfolio.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Insurance Impound Inflation
Thru Annual
Month Rate
360 0.060000
Enter the month and the insurance impound growth rate
in the table. Up to ten sets of values can be entered. The program will
interpolate within each adjustment period.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Property Tax Inflation
Thru Annual
Month Rate
360 0.020000
Enter the month and the property tax impound growth rate
in the table. Up to ten sets of values can be entered. The program will
interpolate within each adjustment period.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Pay on Impounds
Key to Rate Paid on Impounds: 1 (0=Firm,1=actual,2=spread)
Thru Annual
Month Rate
360 0.000000
Every state has its own laws regarding the payment of
interest on impounds. We have allowed the user a wide range of choices
in this area. If you desire to pay interest on all impounds then select
"0" in the key field. All portfolios that being are analyzed
using the current firm file can be set to pay different rates for the interest
on impounds. It is more likely that only certain segments of the portfolio
will be required to pay interest on impounds. In this case you would select
"1" or "2" depending on whether you want to use the
spread feature or wish to enter the paying rate directly.
(0) Firm Rate - If the firm rate is chosen then all portfolios
analyzed with the active firm data file will pay interest on impounds.
If only some of the portfolios will pay on impounds then select 1 or 2
for this field. Then, only those portfolios will pay interest.
(1) Actual - This will cause the program to pay interest according to the
actual rate entered in the pay on impounds table found in chapter 3.
(2) Spread - This will cause the interest rate on impounds to be a spread
from the market index. The spread can be either positive or negative.
Late Fees are usually received from two sources:
a. Loans which have not made their payment by the Late
Fee Trigger Date.
b. Loans which are more than 30 days delinquent.
The KAL_II Model allows late fees to be collected on loans
that have not made their payment by the late fee trigger date. Other models
only allow collection of late fees on loans that have been delinquent thirty
days or more.
Late fees are an important source of ancillary income. The determination
of the late fees can be a critical issue. Many models actually show an
increase in value when the delinquency ratios rise. This occurs for several
reasons. First, no allowance is made for the increase in servicing cost
as a result of the delinquencies. Next, other models collect late fees
on all loans including those in foreclosure. Finally, other models make
no allowance is made for the timing of the receipt of the late fees.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Late Fees
Late Fees Triggered-------------------
On Any Payment Received After: 16 (Day of Month)
Late Fee on P&I : 0.040000
Late Fee on T&I : 0.000000
Fraction Late Fees Collected : 0.650000
For Previously Good Loans Only--------
Late Notice Triggered : 17 (Day of Month)
Cost of Late Notice : 10.00
This is the day of the month that the late fees are triggered.
The late Fees are due and payable after this date.
This is the late Fee that is due on the Principle and
interest payment. It is expressed as a percent and is entered as a decimal
between 0.0 and 1.0.
This is the late Fee expressed as a percent that is due
on the tax and insurance impound portion of the monthly loan payment.
This is the percentage of the late Fees that are collected.
It is represented as a percentage of the total late fees charged. It is
entered as a decimal between 0.0 and 1.0.
The following information applies only to loans that were
current and have now become one month delinquent. Payments that come in
AFTER the trigger day incur the late fee or the cost of the late notice.
This is the day the late notice is sent. The date is used
to determine the additional cost of loans not paying before this date.
Payments which come in after this date trigger the late notice.
This is the cost of the late notice in dollars per late
notice. It may also include the first month's collection costs.
The miscellaneous section covers three areas:
. Purchase Price - In order to determine the present value
of the portfolio you must decide what price you are willing to pay for
the portfolio. If the price you enter is less than the break-even price
you will have a positive net present value. If the price paid is greater
than the break-even the net present value will be negative.
. Conversion Costs - This topic is discussed in detail in the firm section
under conversion costs.
. Growth in Servicing Costs - The table for the servicing cot inflation
is found under growth and other tables.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Miscellaneous
Key to Purchase Price : 1 (1=Pct,2=Dollars)
Price as Fraction of Balance : 0.020000 (Enter 1% as .01 )
Price in Dollars : 0.00
Key to Conversion Costs : 1 (1=Per Loan,2=Total)
Conversion Cost Amount : 20.00
Key to Growth Service Costs : 1 (1=Inf, 2=Table)
(1) Percentage - Enter a "1" if the Purchase
Price is entered as a Percentage of the loan Balances.
(2) Dollars - Enter a "2" if the Purchase Price is entered as
a total amount in dollars.
The price is expressed as a fraction of the total portfolio
loan balance. If the portfolio price was 2.0% of the principal balance
the number entered would be 0.20000.
The price is expressed as a total dollar amount.
There are two methods available to express conversion costs:
(1) Per Loan - Enter a "1" if the conversion
costs are expressed in a per loan basis.
(2) Total Amount - Enter a "2" if the following conversion costs
are expressed as a Total amount for the conversion.
This is either a per loan amount or a total Dollar amount
depending on what was entered in Key to conversion costs field.
Key to Growth in Service Costs
The servicing cost growth table is entered in the growth table section.
(1) Spread - Enter "1" if the growth in servicing
costs is tied to the inflation table.
(2) Specific Table - Enter a "2" if the separate servicing cost
growth table is used.
The KAL_II program uses four methods to determine the
prepay rate of a portfolio segment:
1. Experience Based Forecast
2. F.H.A. Multiplier (FHA)
3. Constant Payoff Ratio (CPR)
4. Public Services Administration (PSA)
We understand there is some uncertainty in using any one
particular method. The effects on present value can be substantial if different
methods are used. In years prior to 1984 it was not unusual to experience
Runoff's in the range of 4%-6% per year. In recent years, runoff rates
of 15%-20% have been common. What is the correct rate to use? It is really
a function of the loans that you are originating or purchasing. Obviously,
loans with higher interest rates will have higher prepayment rates. As
current rates rise to meet the higher mortgage rates the runoff of those
loans may slow down.
Rather than attempt to exactly predict runoff it is better to look at several
different scenarios. What we really are trying to do is to estimate Risk.
The easiest method to use is the Experience-Based Forecast. You can enter
exactly what you feel is a reasonable estimate of Runoff. In addition,
you can use this method to enter the actual runoff for the year and approximate
the change in value of your segments or of your total portfolio.
Another factor to consider is that different types of loans may experience
different runoff rates. ARM loans may not payoff as quickly as the fixed
rate loans. If any loans have high prepayment penalties this may also slow
down the runoff.
In order to determine how the program has calculated the runoff, examine
the backup screens in Chapter 7. This will give you the payoff probabilities
for each month of the loan life.
There are two fields labeled "Interest Due on Payoffs".
One field is the interest due from the mortgagor. The second field is the
interest that is paid to the investor. Interest differential is the difference
in interest received from payoffs and the interest remitted on the payoffs.
It is very possible that the interest collected does not equal the interest
remitted. If this occurs there may be an additional payoff cost.
It is also possible there could be a gain from this differential. If you
find there is a negative cost in the payoff cost field, check the two interest
owed fields and make certain you have set the fields correctly.
The probability of payoff you enter in the transition
table will not be reflected in the payoff table given in Section 7. Only
the prepayment pattern selected in this section is shown in that table.
If you service loans that require prepayment penalties,
show these costs as a reduction in the payoff cost.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Payoffs
General Factors---
->> Cost to Process Payoff 50.00
Interest Owed on Payoff 2
Forecasting Parameters---
1. Estimates of Remaining Loans
2. F.H.A. Experience
3. Constant Percentage
4. P.S.A.
Choice of Method---
Select Method 1
Quit this menu
This is the total dollar cost to process a single payoff.
It may vary with different firms. The amount should include the full cost,
including facilities and data processing expense, of the payoff for a single
loan.
This is the interest that the mortgagor owes the mortgagee.
If there is a difference between what is received by the mortgagee and
what is paid to the investor then the difference is shown in the payoff
costs and can be seen in Chapter 8 as differential interest.
(0) No Interest - No interest is owed by the mortgagor
when the loan is paid off.
(1) Actual Days - Enter a "1" if the interest owed by the mortgagor
is the actual days until payoff (VA loans).
(2) Full Month - Enter a "2" if a full month's interest is owed
by the mortgagor (FHA loans).
Forecasting Parameters - Payoff Method
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Experienced Based Probabilities
Thru Annual
Month Rate
12 0.920000
60 0.750000
120 0.500000
360 0.200000
This is an estimate of what percent of the remaining loan
balances will be after each period expires. It is an estimate you make
of what the runoff is expected to be for the entire life of the loan. Base
the estimate on the loan interest rate, loan type, loan age, location,
etc. When using this method you have the advantage of specifying exactly
what you want to happen. In our example, 92% of the loans payoff in the
first year, 75% have paid off by the fifth year, 50% payoff by the tenth
year and 20% payoff in the last year.
This method uses the available F.H.A. tables to calculate
the payoff probability. The tables are updated on an annual basis. The
FHA multiplier tells the program the factor by which to multiply the standard
F.H.A. prepayment rate.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Constant Percentage Probability
Thru Annual
Month Rate
0 0.000000
360 0.200000
This is the actual rate at which the loans will run-off.
Enter the period through which the rate stays constant. The loans paying
off will be determined by multiplying the portfolio balance by the CPR.
Public Services Administration - If the P.S.A. method
is used then this is the multiplier that will be applied to the standard
annual PSA prepayment rate.
Enter Number of Selection
The program will use whichever method of the four available.
Enter the number of the method you want to use.
One of the primary benefits of collecting funds from the
mortgagors is the right to hold these funds for negotiated length of time.
These funds are often used by mortgagees to offset fixed rate loans. The
mortgagee is able to borrow from the bank at a rate that is substantially
less then the current interest rate.
The KAL_II model provides a thorough examination of these funds. You can
quickly and effectively determine how changes in the investor requirements
will affect the different sources and uses of funds. In addition, the Model
provides standard valuation variables, such as P&I constants and average
P&I balances.
The following is a list of topics that are discussed in the manual. Not
all these topics are presented in this section. Check the index to find
the topic that you are interested in.
1. Principal & Interest Balances
Initial balance
P&I constant
Length of time that the balances are held
Remittance method
Remittance schedule
Average monthly balance
2. P&I Advance Account
Initial advance required
Monthly advance requirement
Recovery of the advances
3. Payoffs Funds
Length of time that payoff funds are held
Interest owed to the investor on payoffs
Remittance method
Remittance schedule
We have provided for the most common types of remittance
processing. The category "Other" can be used to create your own
particular investor remittance pattern. In addition, it is possible to
alter the first five patterns to your own particular needs. Each of the
first five patterns have predefined variables to make it easier to do a
quick analysis. It is interesting to change these patterns and examine
how the changes affect the servicing valuation or the cash flows.
a - GNMA 1 GNMA I Type Servicing
b - GNMA 2 GNMA II Type Servicing
c - FNMA MBS Federal National Mortgage Association
Mortgage Backed Securities
d - FNMA AES Federal National Mortgage Association
AES Servicing
e - FHLMC PC Federal Home Loan Mortgage Corp
Participation Certificates
f - Other Investor Owned Servicing
We recognize that new remittance requirements may be established
with each new security that is developed. There should be sufficient flexibility
in the model to provide for any of these new changes.
Each segment of a portfolio can be a different loan servicing
type. When the segments are consolidated the program will analyze each
segment separately and then combine the results. Segments that share P&I
accounts can be defined in the group section. When segments share P&I
accounts it is possible for an advance from one segment can be paid by
the positive cash balance available in another segment. This is particularly
helpful when evaluating GNMA loan servicing.
This table defines how you expect the remittance checks
to clear your bank account. The program will begin the pattern on the day
of the month that you remit to the investors. In our GNMA I example that
day is the 15th of the month. We allow no checks to clear the first day
after remitting. On the second day (after remitting) 60% of the checks
clear. By the 20th of the month 70% of the checks will clear. Remember
that this is a cumulative percentage. All checks must clear by the end
of the month. The program interpolates between each fraction. If you expect
all funds to clear the same you remit day you put 100% on the 15th of the
month.
This pattern is very important to the daily cash flow calculation. Most
accounting departments have a good idea of what this pattern looks like.
It is helpful to monitor your clearings on a monthly basis in order to
determine any changes in that pattern. A late mailing to the investors
will immediately show up in your remittance clearings.
The clearance pattern screen will not appear unless the remittances begin
some time in the middle of the month. You set the table to begin the clearings
after the day of the month that you remit.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Remittance Processing
Key to Type of Servicing : a
( a. GNMA 1 )
( b. GNMA 2 )
( c. FNMA MBS )
( d. FNMA AES )
( e. FHLMC PC )
( f. Other )
This input field describes the remittance requirements
for the investor who owns the loans that are being evaluated. The categories
listed above can be altered to suit your particular requirements. They
are offered only as a starting point for the simulation.
Once you change remittance types the changes you made to the prior remittance
type will be lost. Keep a record of types that you define yourself. It
is easy to forget exactly what you used for a definition. The program will
print a copy of the data entered here as part of the reports for this section.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Remittance Processing
Description of Servicing Method: GNMA I Example
For Surviving Loans---------------
Delay in Remitting Payments : 0 (Months)
Day of Month to Remit : 15 (0 = Pass thru)
Must Deposit All Funds : Y (Y=Yes, N=No )
Advance Required : 2 (0, 1=Int, 2=P&I )
Maximum Period of Advances : 360 (Months )
Remittance Clears Same Day : N (Y=Yes, N=No )
Calculate Initial P&I Receivables : Y (Y=Yes, N=No )
For Loans Paying Off--------------
Interest Owed on Payoffs : 2 (0,1=Days,2=Full)
Must Pass Payoffs As Received : N (Y=Yes, N=No )
(if Yes, Days to Wait Until Pass) 0 (Days )
(if No, Delay in Passing Payoff) 1 (Months 1)
This refers to the type of loan servicing that is being
evaluated. If you add your own input definitions, then "NAME"
the servicing method for future reference.
This is the number of months that pass before remitting
the funds to the investors.
This is the day of the month that the funds are remitted.
If "0" is entered then the Remittance is considered as a "Pass-Through".
The clearance pattern screen will begin check clearances as of the date
that you enter here.
This determines whether all funds must be deposited into the account.
(Y) Deposit - All funds are deposited into the bank accounts.
(N) No Deposit - Funds are not deposited into the bank accounts.
This field defines whether you must advance payments to
the investor whether they are received or not.
(0) No Advance - If "0" is entered then no advances
are required to this investor.
(1) Interest - Indicates that interest must be advanced on the remittance
date to the investor.
(2) P&I - Indicates that interest and Principle must be advanced to
the investor for all loans due for the prior month.
This is the number of months that you are required to
continue making the advances. If the loans are delinquent you must continue
to advance through this month.
This tells the program whether the remittance to the investor
clears the same day.
(Y) Same Day - Remittance clears the same day.
(N) Later Date - Remittance clears at a later date.
This applies to a purchase where there is an initial principle
and interest advance that needs to be determined.
(Y) Advance - The program will calculate the P&I advance
based on the initial delinquency ratios.
(N) No Advance - Do not calculate an initial P&I advance.
This is the interest owed by the mortgagee to the investor.
If the mortgagee cannot collect a similar amount from the mortgagor then
an interest differential will be included in the calculated payoff cost.
(0) None Owed - If "0" is entered then no interest
is owed on payoffs.
(1) Days - Indicates that the interest calculate on a per diem is owed
to the investor.
(2) Full Month - Indicates that the full months interest is owed regardless
of the payoff date.
Indicates whether the payoff amount must be passed through
as it is received. Either the payoff is held for a number of days or the
payoff is held for a number of months.
Pass Payoffs - "Yes"
Enter the number of days that is allowed before the payoff
funds must be sent to the investor.
Passing Payoff - "No"
Number of months you are allowed to wait until the funds
are passed to the investor.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Clearance Pattern of Remittance to Investors
Day Fraction Day Fraction Day Fraction
21
22 0.8000
23
24 0.9000
15 0.0000 25
16 0.6000 26
17 27
18 28
19 29
20 0.7000 30 1.0000
Entry-->>____________<<
Indicate when during the month funds in this category
are deposited in the bank by entering the fraction of the total amount
that has been deposited by the end of the day. Enter only the days for
which you have data. The program will interpolate between your entries.
To blank out an entry, enter minus one (-1) Fractions must be increasing.
Press {ESC} when finished
Servicing Fees are usually charged in two different ways.
First, there is the conventional fee that is expressed as a percentage
of the loan balance. With this type of fee the annual servicing fee revenue
will reduce in proportion to the loan balances. loan balances are reduced
through normal amortization, payoffs and foreclosures. The second method
charges a flat fee to the owner of the loan. As an example, this may be
$75.00 per loan. Although the revenue received under this method may not
be as high initially, the servicer usually does not have responsibility
for additional charges such as foreclosure losses. In addition, as time
passes the revenue per loan will stay constant. Under the first method
the fees received for low balance loans may not totally offset the servicing
costs.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Servicing Fees
Type of Fees : 1 (1=Pct of Balance, 2=Dollars/Loan)
Thru Annual
Month Rate
360 0.004400
Servicing fees can be expressed as either a percentage
of the remaining loan balances or as a dollar per loan amount.
(1) Percent - Enter a "1" to show annual rate
as a percentage (%) of the loan Portfolio balance.
(2) Dollar/Loan - Enter a "2" to show annual rate as a fixed
dollar amount ($$$) per loan.
On each line, enter the month and the annual rate at which
servicing fee is earned on the declining loan balance. Months must be increasing.
GNMA_I servicing has a standard servicing fee of 44 basis
points.
Transition Tables are used by the KAL_II Model to show
how the portfolio changes over time in respect to:
1. Probability of delinquency patterns
2. Payoffs within each delinquency category
3. Costs to service good loans
4. Costs to service each delinquency category
A transition table is a SET of probabilities, payoff ratios
and processing costs that remain constant over a period of time. This period
can be any length from one month to 360 months. Transition tables are established
at the your discretion. The transitions tables define the loan administration
costs and, therefore, have a significant impact on the valuation of the
portfolio. These tables are used to simulate expected market and operational
conditions in sufficient detail to show the cost effects of changing operational
conditions.
A maximum of ten transition tables can be defined for each loan portfolio
segment. A complete set of probabilities and costs can be used for each
of these transition tables.
There are many reasons why you would want to use several transition tables.
You may feel that we are in a economic period that has an unusual amount
of Foreclosures. At some point in the future, economic conditions will
change and we will again experience lower foreclosure ratios.
Being that the cash flows are heaviest in the early years it might be important
to determine delinquency pattern for the next ten years using one pattern
for each year. The remaining years will use the last pattern entered. The
program will always extrapolate the final patterns or rates entered to
the end of the portfolio life.
A transition table defines the probabilities that a loan
will do one of the following at the end of each loan status:
a. move on to the next status,
b. become a current loan,
c. pay off.
The sum of these probabilities must always equal one.
In this section of the program you define what will be the probability
of each of these states occurring.
These probabilities can be determined in several ways.
First , they can be calculated directly from the firm's monthly delinquency
ratios provided by the delinquency reporting system. Second, we have provided
tables that show certain delinquency patterns and the probabilities associated
with each pattern. The tables use average delinquency ratios to give a
starting point when determining the delinquency probabilities. Finally,
there is a spreadsheet program that will calculate the probabilities for
a specific delinquency pattern.
Delinquencies and Transition Probabilities
In our GNMA I Example file we have defined a 3.3% probability
that a Good_Loan will become one month delinquent. Therefore, the number
of thirty day delinquency loans will be:
9000 (Good_Loans) x 3.3% (Prob) = 297.0 (Del_1_Mon Loans)
The new thirty day delinquent ratio will be approximately:
297.0 (Del_1_Mon) / 9980 (Remaining Loans) = 3% Delinquency
In this brief example we have not included the loans that
have come current from each loan status category, the loans that have been
foreclosed and the loans that have paid off. The example at the end of
this section works through the first month of the portfolio life.
In the GNMA I example we have supplied with the KAL_II model we use three
transition tables. The delinquency ratio patterns are as follows:
GNMA I Example - Transition Table Delinquencies
Period Month 30 60 90 +120 F/C Total Table One 1-24 3.00% 2.50% 2.00% 1.50% 1.00% 10.00% Table Two 25-60 2.50% 2.00% 1.00% .75% 1.00% 7.25% Table Three 31-360 2.50% 1.50% 1.00% .75% 1.00% 6.75%
There is an extended example at the end of this section
of the manual.
It is possible to change the computed payoff probability
and set the prepayment specifically for each delinquent category. This
recognizes the fact that delinquent loans may prepay differently than current
loans.
Probability of Paying Off
From Probability of
Status Paying Off
Good Loan - Del 1 Mon Computed
Del 1 Mon - Del 2 Mon Computed
Del 2 Mon - Del 3 Mon Computed
Del 3 Mon - GT 90 Day .200
GT 90 Day - Foreclose .200
Foreclose - Off Books .400
In this example we have allowed the program to determine
what the probability of payoff is for the first three loan categories.
We then entered a 20% payoff probability for DEL_3_MON and GT_90_DAY. The
probability of payoff for FORECLOSE is set to be 40%.
An important function of the model is to allow an analysis
of the servicing value based on different marginal servicing costs for
each loan status category. The cost to process loans in each status directly
effects the overall servicing costs and, therefore, the value of the portfolio.
The operational servicing cost data should be sufficiently understood to
be able to determine what is the marginal cost to process loans in each
loan status category. The total servicing costs for each period are shown
on the earnings screen in the portfolio section. By using this screen as
a comparison to the firm's actual servicing cost input cost data used in
the model can be verified to the actual servicing results. Keep in mind
that there are nine possible transition tables. Each table can have a separate
set of processing cost data. This allows the cost structure of the servicing
operation to be varied many times over future years.
The marginal costs are adjusted by the inflation rate
on servicing costs (see Miscellaneous Screen). The cost to process the
Good_Loan status is given in annual amounts on a per loan basis. The other
categories are given as the additional or marginal cost to process the
loan status category on a monthly basis. Individual statuses can be selected
using the arrow keys. Enter the value and press {Enter}.
The program uses an ANNUAL cost per loan for the cost
to service a Good Loan. This is the cost to service a loan before any Delinquency,
payoff, or late notice cost is added. It is assumed that most users will
have a good idea of their annual processing costs.
The program uses MONTHLY cost per loan for all other categories.
All costs in these tables should be stated in today's
dollars. The program will adjust for inflation according to the factors
input in the relevant cost and inflation sections.
Developing the servicing cost for each status can be a
time consuming process. It is often best to start with a few categories
and learn to use them effectively. As the simulation is developed over
time it may be helpful to expand the cost accounting System to track the
desired categories and be able to generate the data for direct input to
the program.
Costs to process a status are always given as the monthly
marginal cost for the entire loan status time period.
From Cost to Process
Status in this Status
Good Loan - 32.00 Annual
Del 1 Mon - 4.00 Monthly
Del 2 Mon - 6.00 Monthly
Del 3 Mon - 8.00 Monthly
GT 90 Day - 12.00 Monthly
Foreclose - 15.00 Monthly
In this example the cost to process a GOOD_LOAN is $32.00
per year or $2.67 per month. The Additional cost to process a loan that
is one month delinquent (DEL_1_MON) is $4.00 making the total cost for
the month equal to
$2.67 Cost to process GOOD_LOAN ($32.00/12 months)
+4.00 Marginal Monthly cost to process DEL_1_MON status
---------
$6.67 Total Monthly Servicing costs
The Marginal cost to process a DEL_2_MON would then be
$2.67 plus $6.00 or $8.67 per Month.
Finally, if a loan is in Foreclosure the total cost to process a Foreclosure
is:
$ 2.67 GOOD_LOAN
+15.00 Marginal Cost to Process FORECLOSURE
------------
$17.67 Total Monthly Processing Cost
This is the cost to process Foreclosures, not the total
Loss on the Foreclosed loan. Review the foreclosure section of the Manual
for the inputs for loans in Foreclosure.
We have provided the user with this cost mechanism because we recognize
the important effect that delinquent loans can have on the cost to service
a loan. It is beneficial to the user to see how a better collection program
can affect the cost to service the segment under consideration. A decrease
in delinquency ratios should produce a significant decrease in servicing
costs. This type of analysis is helpful in justifying the additional dollar
outlays that often accompany increases in servicing productivity.
GNMAEX GNMA I Example
Loan Portfolios
Change Data
Transition Table: 1
Which Applies from Month: 1 thru Month: 24
From Prob Move on Prob Pay Cost to Process When
Status To This Status Fully In This Status
Good Loan - Del_1_Mon 0.0330 Computed Good Loan 32.00 Annually
Del_1_Mon - Del_2_Mon 0.8330 Computed Del_1_Mon 4.00 Per Month
Del_2_Mon - Del_3_Mon 0.8000 Computed Del_2_Mon 6.00 Per Month
Del_3_Mon - GT_90_Day 0.2500 Computed Del_3_Mon 8.00 Per Month
GT_90_Day - Foreclose 0.3330 Computed GT_90_Day 12.00 Per Month
Foreclose - Off Books 0.7500 Computed Foreclose 15.00 Per Month
Entry--
Each transition table is given a number beginning with
1 (one).
This is the beginning month and ending month which this
transition table covers. In this example transition table #1 covers the
period from 1 month to 24 months. If a second table is not entered the
program will extend this table to the last month of the Simulation.
In these fields you define the probability that the loan
will go to the next status. These probabilities are used to determine what
the delinquency pattern will be in any month during the life of the portfolio.
The payoff ratio is used to calculate the probability
of the loan paying Off at the time the loan LEAVES the loan status category.
There are nine possible transition tables so there can be ten sets of payoff
probabilities.
If it is possible to determine the probability of payoff for each loan
status then enter these payoff probabilities into the data fields. The
program will then use these probabilities instead of those calculated using
the prepayment method entered in the payoff section. If you are uncertain
what the actual prepayment probabilities are, then use the computed feature.
The program will compute the probability of payoff for
each status by entering "C" in the {Prob_ Pay_Fully] field. The
data entered in the payoff Section is used to determine what the payoff
probability will be (Experience, FHA, CPR or PSA).
Only one payoff method can be selected for each portfolio segment. The
program uses the selected method for all transition tables in this portfolio
segment analysis. If you chose a value other than "Computed"
as the probability, the program will override the payoff method selected
in Section 4.K.
Enter the cost to service the loans in this portfolio
segment. The Good_Loan cost is in annual dollars per loan and the delinquency
costs are in additional (marginal) monthly costs per loan.
This example traces the loan status calculations for the
first month delinquency. It is important to follow through a single month
in order to determine how the probability ratios should be calculated.
This example uses a different portfolio description in order to provide
a simpler example.
Loans Loan Status Probability Payoff Probability
930 Good Loan - Del 1 Mon 0.0500 1.0 %
40 Del 1 Mon - Del 2 Mon 0.2000 10.0 %
20 Del 2 Mon - Del 3 Mon 0.3000 20.0 %
10 Del 3 Mon - GT 90 Day 0.4000 30.0 %
0 GT 90 Day - Foreclose 0.5000 Computed
0 Foreclose - Off Books 0.7500 Computed
GOOD-LOAN MONTH ONE
BEGIN DEL-1 PAYOFF CURRENT TOTAL
Probability 5% 1% 94% 100%
GOOD-LOAN 930 46.5 9.3 874.2 930
This means that there is a beginning balance of 930 loans in the GOOD-LOAN status category. Of this balance 46.5 loans move on to DEL-1-MON, 9.3 loans payoff and the remainder, 874.2, stay current in the GOOD-LOAN status.
DEL-1-MON MONTH ONE
BEGIN DEL-2 PAYOFF CURRENT TOTAL
Probability 20% 10% 70% 100%
DEL-1-MON 40 8 4 28 40
This means that eight loans go on to the DEL-2-MON status,
4 loans payoff and the remainder, 40, become current.
DEL-2-MON MONTH ONE
BEGIN DEL-3 PAYOFF CURRENT TOTAL Probability 30% 20% 50% 100% DEL-2-MON 20 6 4 10 20
This means that six loans go on to the DEL-3-MON status,
4 loans payoff and the remainder, 10, become current.
DEL-3-MON MONTH ONE
BEGIN GT-90 PAYOFF CURRENT TOTAL Probability 40% 30% 30% 100% DEL-3-MON 10 4 3 3 20
This means that four loans go on to the GT-90-DAY status,
3 loans payoff and the remainder, 3, become current.
We take the sum of the individual calculations as follows:
BEG GOOD 1-MON 2-MON 3-MON GT-90 OFF END
GOOD-LOAN 930 874.2 46.5 9.3 915.2
DEL-1-MON 40 28.0 8.0 4.0 46.5
DEL-2-MON 20 10.0 6.0 4.0 8.0
DEL-3-MON 10 3.0 4.0 3.0 6.0
GT-90-DAY 0 4.0
OFF BOOKS 20.3
Totals 1000 915.2 46.5 8.0 6.0 4.0 20.3 1000.0
After Month One there are :
915.2 Loans in Good_Loan status,
46.5 Loans in Del_1_Mon,
8.0 Loans in Del_2_Mon,
6.0 Loans in Del_3_Mon,
4.0 Loans in GT_90_Day, and
20.3 Loans Paid-off.
Since no loans completed the GT-90-DAY status there can be no loans in foreclosure.