December 1, 1999
How
to compare the subsidy results
with
the previous release
The new release of the credit subsidy calculator will
produce subsidy estimates that differ from the previous release (r9). Though such differences are generally fairly
small, initial estimates of differences could be quite large if there are
uncorrected differences in the interest rate assumptions used for
discounting. This paper describes how
the interest rate assumptions used in the previous release should be adjusted
to make them consistent with the current methods.
Before the 1999 Budget, the interest rate assumptions used
in credit accounts were directly tied to the interest rates in the budget
assumptions for each year. As a result,
the rates for credit accounts often had a downward trend, which were not borne
out. A comparison of assumptions with actual data showed that the consistent
downward trend could be replaced by straight-lined estimates, which would be
simpler and at least as accurate as the assumptions they replaced. Beginning with the 1999 Budget, the interest
rate assumptions used for credit accounts were changed as follows:
The rates were those assumed to
prevail on the first day of the fiscal year.
(For example, the rates for the 1999 Budget would be those assumed to
prevail on October 1, 1998).
Previously, the rates for credit accounts were estimated annual
averages.
Rates would be straight-lined to all
future periods for the purpose of estimating subsidies for the budget. Previously, a downward trend was generally
assumed.
These rates would continue to be
used until replaced by actual rates.
Previously, the budget assumptions were replaced by revised estimates,
which were, at best, only a marginal improvement over the budget assumptions.
How to make a consistent comparison
If you wish to compare the results obtained with the CSC
with any previous budget assumptions, the following changes are needed:
Find the “similar maturity” rate
from the table below for the budget year and maturity category
Place that rate in all columns of
the “discount rate” line in your cash flow spreadsheet.
Adjust any other values in your
spreadsheet that depend on the interest rate assumptions. For example, if the borrower’s rate in a
direct loan program is tied to the Treasury rate, and adjustment in the cash
flows for interest payments and, perhaps for defaults and recoveries, would be
necessary.
Compute the subsidy using the
previous release and the revised spreadsheet.
Compute the subsidy using the CSC, the
revised spreadsheet, and with the rates for the budget year you selected in the
first step, above. (Rates are selected
using the Actions-Select Rates menu.)
These results will be comparable except for the changes in
discounting methods. Such differences
should be less than 1 percentage point.
If you do NOT make the changes above, you may find that
there are very large differences between the CSC and the previous release,
because of the difference in interest rate assumptions.
Rate assumptions
The following table presents the “similar maturity” rate
assumptions for the 1992-2000 Budgets, recalculated to be consistent with
current methods.
Budget year
|
1
year
|
1-5 years
|
5-10 years
|
10-20 years
|
20 years or more
|
2000…………...
|
4.55
|
4.77
|
4.91
|
5.01
|
5.06
|
1999…………...
|
5.38
|
5.66
|
5.87
|
6.03
|
6.11
|
1998…………...
|
5.42
|
5.82
|
6.10
|
6.23
|
6.28
|
1997…………...
|
5.21
|
5.42
|
5.57
|
5.67
|
5.71
|
1996…………...
|
6.56
|
7.34
|
7.84
|
8.07
|
8.13
|
1995…………...
|
4.07
|
5.09
|
5.69
|
5.93
|
5.98
|
1994…………...
|
3.94
|
5.54
|
6.47
|
6.82
|
6.87
|
1993…………...
|
5.14
|
6.24
|
6.91
|
7.18
|
7.23
|
1992…………...
|
7.06
|
7.38
|
7.54
|
7.63
|
7.68
|
The exact definitions of the intervals are:
1 year: 1 year or less
1-5 years: More than 1 year and less
than 5 years
5-10 years: 5 years or more, but less
than 10 years
10-20
years: 10 years or more,
but less than 20 years
20 years or
more: 20 years or more