PAGENO="0001"
INTEREST RATE GUIDELINES FOR FEDERAL
DECISIONMAKING
HEARING
BEFORE THE
SUBCOMMITTEE ON ECONOMY IN GOVERNMENT
OF THE
JOINT ECONOMIC COMMITTEE
CONGRESS OF THE UNITED STATES
NINETIETH CONGRESS
SECOND SESSION
JANUARY 29, 1908
Printed for the use of the Joint Economic Committee
~ I ~
PROPFP 1
-.
COLLLL u JI~ E~( L ~
CAMDEN1 N. J. 08102
MAR 2 0196
U.S. GOVERNMENT PRINTING OFFICE
~`~85-654O WASHINGTON : 1968
For sale by the Superintendent of Documents, ILS. Government Printing Office
Washington, D.C. 20402 - Price 25 cents
PAGENO="0002"
SENATE
JOHN SPARKMAN, Alabama
J. W. FULBRIGHT, Arkansas
HERMAN E. TALMADGE, Georgia
STUART `SYMINGTON, Missouri
ABRAHAM RIBICOFF, Connecticut
JACOB K. JAVITS, New York
JACK MILLER, Iowa
LEN B. JORDAN, Idaho
CHARLES H. PERCY, Illinois
HOUSE OF REPRESENTATIVES
RICHARD BOLLING, Missouri
HALE BOGGS, Louisiana
HENRY S. REUSS, Wisconsin
MARTHA W. GRIFFITHS, Michigan
WILLIAM S. MOORHEAD, Pennsylvania
THOMAS B. CURTIS, Missouri
WILLIAM B. WIDNALL, New Jersey
DONALD RUMSFELD, Illinois
W. E. BROCK 3D, Tennessee
ECONOMISTS
WILLIAM H. 1~IOoRE JOHN B. HENDERSON
DONALD A. WEBSTER (Minority)
SUBCOMMITTEE ON ECONOMY IN GOVERNMENT
WILLIAM PROXMIRE, Wisconsin, Chairman
JOINT ECONOMIC COMMITTEE
(Created pursuant to sec. 5(a) of Public Law 304, 79th Cong.)
WILLIAM PROXMIRE, Wisconsin, Chairman
WRIGHT PATMAN, Texas, Vice Chairman
JOHN R. STARK, Eccecutive Director
JAMES W. KNOWLES, Director of Rc~search
GEORGE R. IDEN
SENATE
JOHN SPARKMAN, Alabama
STUART SYMINGTON, Missouri
LEN B. JORDAN, Idaho
CHARLES H. PERCY, Illinois
HOUSE OF REPRESENTATIVES
* WRIGHT PATMAN, Texas
MARTHA W. GRIFFITHS, Michigan
WILLIAM S. MOORHEAD, Pennsylvania
THOMAS B. CURTIS, Missouri
DONALD RUMSFELD, Illinois
(II)
PAGENO="0003"
CONTENTS
STATEMENTS AND SUBMISSIONS
Proxmire, Senator William, chairman of the Subcommittee on Economy
in Government: Page
Opening remarks
Announcement of hearing 2
Staats, Hon. Elmer B., Comptroller General of the United States, accom-
panied by Ellsworth H. Morse, Jr., Director, Office of Policy and Special
Studies; Keith E. Marvin, Deputy Director for Systems Analysis; Daniel
Rathbun, former Deputy Director for Systems Analysis; and Ted M.
Rabun, staff member 3
"Survey of Use by Federal Agencies of the Discounting Technique in
Evaluating Future Programs," report to the Joint Economic Com-
mittee by the Comptroller General, January 29, 1968 33
(III)
PAGENO="0004"
PAGENO="0005"
INTEREST RATE GUIDELINES FOR FEDERAL
DECISIONMAKING
MONDAY, JANUARY 29, 1968
CONGRESS OF THE UNITED STATES,
SUBCOMMITrEE ON EOONOMY IN GOVERNMENT
OF THE JOINT EcoNoMIc COMMITrEE,
Wa8hington, D.C.
The Subcommittee on Economy in Government met, pursuant to
notice, at 10 a.m., in room 1202, New Senate Office Building, Hon. Wil-
liam Proxmire (chairman of the subcommittee), presiding.
Present: Senators Proxmire, Symington, and Jordan.
Also present: John R. Stark, executive director.
Chairman PROXMIRE. The subcommittee will come to order.
This morning the Subcommittee on Economy in Government of the
Joint Economic Committee continues its study of Federal expendi-
`tures, particularly methods of improving efficiency and reducing waste
in `the $175 billion of Federal expendi:tures.
During this subcommittee's hearings last September on planning-
programing-budgeting (PPB) `system's, we heard three top economists
testify that current Government policies favoring l'ow discounting
rates in cost-benefit `analysis lead to gross overinvestment in certain
sectors, and contributes to inflationary pressures and slower economic
growth. To the extent that this is true, it involves waste of our
resources.
At my request, agencies recalculated, at various higher discount
rates, proposed public works projects; in many cases the higher dis-
counting showed that project costs overwhelmed benefits, indicating
that such planned investments were actually a drain from taxpayers'
pockets.
An initial survey-the subject of today's hearing-was carried out
by the General Accounting Office. The survey `attempted to discover
the extent and `type `of discounting practices now utilized by Federal
agencies. Virtually every Government agency was covered in the
study. The survey shows clearly that there are inconsistencies in cur-
rent budgetary practices and points out the need for reform in dis-
counting procedures.
We will include the complete `survey report in the record at the close
of today's proceedings.
(See p. 33.)
Senator PR0XMIIm. However, before we hear from our witnesses,
an important point should be made. While there is virtual agreement
that Government discount rates are too low as compared to discounts
used in the private sector, we are not asking that the Government
institute one single unique and unchangeable rate to be applied to
(1)
PAGENO="0006"
2
every Government agency. Instead, we are more interested in the
range of rates now utilized; our point is that the entire discounting
structure should be adjusted upward to some parity with private
rates, and that some degree of Government standardization be
achieved.
I might say that it is encouraging to note that in his forthcoming
budget, President Johnson called for increasing the interest rate used
in evaluating water resources projects. The Budget says, and I quote:
The Water Resources Council is developing a more appropriate rate to be
applied in formulating and evaluating water projects. The revised rate will
be related to the average estimated current cost to the Treasury of long-term
borrowing. It will be higher than the rate now in use for project evaluation.
The new rate will ~e applied to future projects in order to assure the most
effective use of Federal funds in the development of the Nation's water resources.
Hopefully the higher rate should reflect the results of the GAO
study, which puts the cost of long-term borrowing around 71/2 percent
rather than just the current Treasury rate which is around 4.875
percent. It might also be advisable to apply the higher rate to cur-
rent or just-started public works projects. If many of these projects
showed negative present value, there would be a strong case for
discontinuing them.
Of course, my own feeling is that opportunity costs are more
relevant and the Stockflsch report which was, as you know, a part
of the hearings which we had last September, I think indicated that
private industry was better than 10 percent on the average, manu-
facturing being around 15 percent of return on invested capital before
taxes.
With this in mind, we are very happy to have as our witness the
Honorable Elmer B. Staats, Comptroller General of the United States,
and members of his staff.
At this point in the record we will include the announcement of
this hearing
(Announcement follows:)
JANUARY 20, 1968.
CONGRESS OF THE UNITED STATES JOINT ECONOMIC COMMITTEE SUBCOMMITTEE ON
ECONOMY IN GOVERNMENT
Senator William Proxmire (D-Wis.), Chairman of the Joint Economic Com-
mittee, said Sunday the results of a General Accounting Office probe into serious
inconsistencies in the Government's system for determining spending priorities
will be disclosed at a hearing, Monday, January 29th.
The Wisconsin Senator was joined in announcing the hearings, to be conducted
by the Subcommittee on Economy in Government, by Representative Thomas B.
Curtis (R-Mo.), ranking minority member. Senator Proxmire is the subcommittee
Chairman.
Senator Proxmire said: "Hearings conducted last September by the subcom-
mittee showed virtual unanimity among economists in the view that discounting
procedures used by the Federal Government in evaluating the benefits of proposed
projects were economically irrational.
"No attempt is made in present practice to consider the rate of return that
public investment might earn if employed elsewhere in either the public or private
sectors and, worse still, there is not even any correspondence between rates used
by different agencies, even for similar programs."
The Comptroller General of the United States, Mr. Elmer Stoats, and his staff
will testify at the Japuary 29th hearing, which is scheduled for 10:00 a.m., in
room 1202 of the New Senate Office Building.
PAGENO="0007"
3
Mr. Staats, before you proceed with your statement I will ask
Senator Jordan if he has a statement?
Senator JORDAN. No; I have no statement.
STATEMENT OF HON. ELMER B. STAATS, COMPTROLLER GENERAL
OP THE UNITED STATES; ACCOMPANIED BY ELLSWORTB K.
MORSE, JR., DIRECTOR, OFFICE OP POLICY AND SPECIAL STUDIES;
KEITH E. MARVIN, DEPUTY DIRECTOR FOR SYSTEMS ANALYSIS;
DANIEL RATHBITM, FORMER DEPUTY DIRECTOR FOR SYSTEMS
ANALYSIS, AI~fl) TED M. RABUN, STAFF MEMBER
Mr. STAATS. Thank you very much, Mr. Chairman. We are pleased
to appear before your Subcoramittee on Economy in Government to
discuss the results of our recent survey of the use by Federal depart-
ments and agencies of the discounting technique in making evaluations
of future programs. Our report on this survey is being submitted today
to your committee in accordance with your request of December 19,
1967.
We have noted with interest the work your committee is doing to
establish a more effective basis for evaluation of the economics of
Government programs. We welcomed the hearings you conducted in
September 1967 on the subject of the planning-programing-budgeting
system and we have read with great interest your statement on dis-
counting in the report. on those hearings. I would like to say, Mr.
Chairman, I thought this was an excellent report. While our review
of this subject; goes back for quite some time, we particularly welcomed
the kind of statement included in that report, and we feel that with the
committee's report and, hopefully with our report, we will make
progress in this area.
The General Accounting Office `has also `been interested in the subject
of planning, programing, and budgeting and specifically the discount-
ing technique. We recently conducted a survey of 23 selected Federal
agencies to obtain information on the use made of the discounting
technique in making evaluations of their future programs. I share
your views that discounting is a matter of increasing importance as
the use of PPB causes us to look further into the future at alternative
programs. This is so because by the t&dmique of discounting the
amounts of estimated future costs and benefits are stated in terms that
are comparable to present costs and benefits. The discounting tech-
nique does this by the application of a compounded rate reflecting
the costs of money.
Federal agency programs generally involve a series of aunual costs
and a flow of benefits over time. Calculation of the present values of
costs and benefits through discounting makes possible a comparison of
alternative programs in terms of a ratio of benefits to costs, which
gives consideration to the time periods in which benefits will be real-
ized and costs incurred and the time value of money.
The discounting technique can be helpful to the decisionmaker even
in those cases in which the benefits associated with programs cannot
be measured in dollars. Here, the present values of the costs of the
programs can be presented to the decisionmaker for his decision as
to whether the perhaps dimly perceived benefits are worth their costs.
Before commenting on the specific aspects of our survey, I would
PAGENO="0008"
4
like to discuss our conclusi2ns at this time so you may consider them
in the light of what our survey ha;s shown. Many Federal agencies
have made good use of the discounting technique :~fl evaluating inth-
vidual projects. By applying the discounting technique to alternatives
within a single program, the agencies have been in a better position to
select the most economical alternative.
In our opinion, however, there is even a greater need for a; consistent
discounting policy when decisionmakers must choose from many com-
peting projects, either within an agency or among agencies. This calls
for a common standard, with justifications for variations only in spe-
cial circumstances. Although some agencies indicated that the discount
rate is viewed as an aid in choosing between pro~rams within an
`igency, there `tppears to be little recognition in practice that this kind
of evaluation of Federal programs calls for a common yardstick for
use by all agencies.
Our study brings tout the significant impact on benefit-cost ratios
of discount rate variations. For example, a proposed program showing
a benefit-cost ratio of 2:0 without discounting might have a benefit-cost
ratio of only 1: 1 if costs and benefits were discounted at a 10-percent
rnte. Our survey has disclosed variation in the discounts ranging
from about 3 to 12 percent. This sort of reduction in benefit-cost ratios
as a result of discounting would be characteristic of many programs
because the high initial investment costs are not reduced and, there-
fore, become greater in relation to the future annual costs and annual
benefits.
in our opinion, the general acceptance of the technique of discount-
ing by Federal `tgencies should be supplemented with impi ovements
necessary to bring about consistency in and among agencies in the
discounting rates used, and in the techniques and underlying concepts
employed. We believe such improvements are needed if this aid is to be
of most effective use to the agencies, the Bureau of the Budget, and
to the Congress in its evaluation of executive agency programs sub-
nutted foi consideration
We believe that improvement in the direction suggested will still
require `i subst'tnti'il `tmount of `iddition'iJ study We h'tve received
statements from several Federal agencies in which they express recog-
nition of the need for standardization and offer to work with us on
further studies.
With these conclusions in mind, I would now like to describe the
highlights of our survey.
Our survey of 23 Federal agencies disclosed that there is a variety
of policies and practices for the use of the discounting technique.
Ten of the 23 `~gencies used the discounting technique in ev'duiting
their fiscal year 1969 programs.
Eight of the 23 `igencies did not use the technique for ev'du'tting
1969 progr'tms but pl'rnto do so in the future
Five of the `igencies do not use discounting `md did not st'mte th~t
they plan to do so in the future
The 10 `igencies that use the discounting technique cite `m v'mriety
of rationales as support for the discount rates they use, which vary
fiom `tbout 3 to 12 percent As `mu ex'mmple, the Office of Economic
Opportunity has used rates of both 3 `mud 5 percent to ev'mlu'vte its
Job Corps `mud IJpw'mrd Bound progr'mms `mnd h'ms used `mr'mte of 5 per
PAGENO="0009"
5
cent for the family planning program. The stated rationale for
selecting these rates was that they were safely on the conservative side
for estimates of this type and gave consideration to the secular growth
in the price `of quality-constant labor.
The General Services Administration used a rate of 4.5 percent in
its analyses leading to `the decision to request~ funds `to `buy sites for
additional buildings for its fiscal year 1969 facilities program. The
4.5 percent was selected as an estimate of the long-term productivity
of capital.
We were advised by the Department of Transportation that for fis-
cal year 1969 programs discounting was used only in the consideration
of three investment programs of the Federal Aviation Administration:
facilities and basic systems, radar components, and en route automa-
tion. A rate of 4.2 percent was chosen because this was approximately
the discount rate (4.25 percent) of the Federal Reserve Board at the
time the studies were undertaken. The Department acknowledged that
some of its other programs may appropriately be suitable for dis-
counting. The Department stated that complete analytical studies are
made on a selective basis and discounting was considered relevant only
for the investment programs mentioned above.
The Agency for International Development uses discount rates
which vary depending on the type of project, the opportunity cost in
the particular country, foreign exchange scarcity, and other factors.
In a recent evaluation of a powerplant project, a discount rate of 8
percent was used as representative of the opportunity cost of money
in the country concerned. In evaluating highway projects in other
countries, discount rates of 8,10, and 12 percent were used.
The Department of the Interior uses several different discount rates
in its evaluation of programs. The interest rate specified by Senate
Document 97, which is the one to which you referred, Mr. Chairman,
in your opening remarks as being modified in the President's budget
message, which was 31/8 percent, was used to evaluate long-term Fed-
eral investment programs in water and land resources. In utility-type
programs 6 percent was used as representative of such programs where
the risk is considered to be relatively low. A 12-percent rate has been
used by the Department of the Interior in its evaluation of certain re-
search and development programs, such as energy and mineral re-
sources, where exploitation, production, and processing is considered
to be a private rather than a public function.
The Atomic Energy Commission reported that it used several rates.
For its analysis of fiscal year 1969 production of special nuclear ma-
terials activities, the discount rate used was 5 percent; however, ana-
lyses were also made using rates of 7.5 and 10 percent to test the sensi-
tivity of the analyses to the discount rate. The 5-percent rate was se-
lected because it was a conservative estimate of the cost of long-term
borrowing by the Department of the Treasury. In the Commission's
reactor development studies, discount rates of 5, 7, and 9 percent were
used since rates of 6 to 7 percent are typical of those used by investor-
owned utilities.
Mr. Chairman, I believe these examples bring out rather clearly the
variety of discoun'ting rates and rationales used by individual Federal
agencies in evaluating their programs. Our report summarizes more
complete the information that we obtained from the 23 departments
and agencies.
89-654 0-68-2
PAGENO="0010"
6
On the basis of our survey, it is evident that there is little agreement
among the agencies as to the rationale that should be used to determine
an appropriate discount rate. There has been no central grndance to
the agencies on this matter and, except for those programs which con-
cern water and related land resources projects, the agencies have been
free to choose whatever discount rate or rationale they considered ap-
propriate. The rationales described to us and the variety of rates used
clearly do not evidence a common understanding by Federal agencies
of the applicability of the technique to Federal programs.
In those agencies that did not use discounting rn their analysis of
fiscal year 1969 programs, there is also a lack of agreement. At one
extreme is the view taken by the Department of Labor that its man-
power development assistance program could be evaluated in terms
of a 1-year horizon even though program benefits are expected to con-
tinue for 5 to 20 years, depending on occupations for which training
is carried on. The implication here is that a very high discount rate
is applicable since benefits beyond the first year are ignored.
At the other extreme is the practice of making evaluations on the
basis of total undiscounted costs and benefits over the life of a pro-
gram. This procedure implies a zero discount rate since the dollar costs
and benefits estimated for future years are given the same importance
as current costs and benefits. The National Aeronautics and Space Ad-
ministration, the Department of Housing and Urban Development,
and the Department of Commerce are among the important agencies
that did not use discounting in their evaluations of fiscal year 1969
programs. As noted in our report, 13 of the 23 agencies we surveyed did
not use discounting.
The fact that 18 of the agencies included in our survey either use or
plan to use discounting is, we believe, an indication that the technique
is receiving increasing acceptance in Federal agencies as an important
aSI)ect of the decisionmaking process. Several of the agencies advised
us that one reason they have not used discounting in the past was that
their analyses were not developed sufficiently to permit discounting.
Presumably, they will use discounting in the future.
There are several schools of thought followed by the various agencies
in determining their particular discount rate. Two of these schools of
thought appear to be predominant although there are various inter-
pretations in actual practice.
One school of thought holds that the rate should be determined by
and be equal to the rate paid by the Treasury in borrowing money. A
second school of thought holds that the rate should be determined
by what is foregone; namely, the return that could have been earned
in the private sector of the economy when the decision is made to
commit resources to the public sector.
Neither school of thought provides clear guidance on the specific
discount rate which should be used. Cost to the Treasury, for example,
will vary depending upon the definition applied, from 3 to 8 percent
or more. The average rate of return in the private sector also varies
depending upon historical periods selected and upon the weighting of
the various segments of the private sector which are used in computing
an average.
A discount rate of slightly over 3 percent is the cost to the Treasury,
if based upon the average rate payable on outstanding U.S. securities
having maturity of 15 years or more, as prescribed by Senate. Docu-
PAGENO="0011"
7
ment 97 which relates to land and water resources. The rate determined
by the procedure prescribed in Senate Document 97 is ait the low end
of the range of rates in use by the agencies and therefore may be an
overly conservative estimate of interest costs on Government bor-
rowing.
Long-term rates show no sign of returning to the level of 10 or 15
years ago. Furthermore, the legal restriction on long-term interest
rates has forced a substantial amount of refinancing of the public debt
through the sale of higher yield short-term securities. It, therefore,
appears to us that the current average yield rate reported in the
Treasury Bulletin is a better basis than Senate Document 97 for deter-
mining interest costs. In this connection, we have noted with interest
that your committee has requested the views of the Water Resources
Council on the propriety of the discount rate determined under Senate
Document 97. This is a little bit out of date because, as you noted in
your opening remarks, the budget being submitted to the Congress
today calls for an increase in the discount rate.
A variation, which we believe has considerable merit, on the pure
interest school of thought is to include the effect of forgone Federal
taxes which would be collected from the private sector of the same
funds were invested there.
As brought out in our report, if the full costs of borrowing, includ-
ing an estimate of forgone taxes from the private sector, are considered,
the difference between the various schools of thought is narrowed sub-
stantially. If this concept is accepted, it would appear that there is a
good possibility of a satisfactory reconciliation of varying points of
view regarding the rate to be used.
To conclude, Mr. Chairman, we believe that the results of our
survey of Federal agency practices suggest that the case for discount-
ing is being accepted, but that there are si~nifioant differences of
opinion in the agencies over the appropriate discount rates to be used.
Because of the wide variation in discount rates and techniques being
used by the executive `agencies to evaluate and justify their programs
and because there is strong impetus to use of the discounting technique
provided by Federal agency adoption of planning-programing-budget-
ing systems, we believe that centralized guidance is needed. The Con-
gress itself may wish to provide guidance to the executive agencies
onthis important topic.
This concludes my statement, Mr. Chairman.
I have with me here this morning, at my left., Dr. Rathbun, who has
been with us and `did a great deal of the work involved in the report,
and who has now left us and is at the University of Pittsburgh. He
has returned today for this reading.
And also, to my left, is Mr. Morse, who is the head of our Office of
Policy and Special Studies which had the responsibility for this re-
port, and Keith Marvin, who has replaced Mr. Rathbun as head of
the systems analysis group in Mr. Morse's office. Mr. Ted Rabun is on
his staff.
I would like to also add we received on Friday a short notice from
the Bureau of the Budget indicating that they were considering the
possibility of central guidance in this area. This, of course, we would
welcome and we believe it is in line with the general recommendations
we are making.
PAGENO="0012"
8
Chairman PROXMIRE. Thank you very much.
I have had a chance to go over-very quickly, of course, because we
just received it this morning-your excellent report to this committee
on the survey of the use by Federal agencies of the discounting tech-
nique in evaluating future programs; and at the suggestion of~ John
Stark, our staff director, I think it is a good suggestion, we will send
this to key Members of Congress, at least those Members of Congress
who seem to be interested in this kind of proposal, and I think might
take some action on the basis of it. I think it is a most significant and
convincing presentation of the importance of achieving some con-
sistency, and of extending as much as we can the use of the benefit/cost
system.
Let me ask you, Mr. Staats, without the benefit/cost system an-
alyses, without discounting, without using a discount rate, how can
an agency possibly evaluate with any. objectivity its investment pro-
gram?
Mr. STAATS. Well, I think the answer is that there has to be some
evaluation of benefits and costs in any judgment that is involved in a
budgetary process. The question is whether or not it is a good judg-
ment or a bad judgment. The question of whether or not you are
making decisions erroneously in terms of priorities depends heavily
upon the use of some analysis of this kind.
As suggested in your committee report, the danger may well be that
we are spending money in some areas which will yield far less in terms
of return for our tax dollar than we would if we invested in some
other areas, and this, of course, is the great danger.
Chairman Pnoxa\IIRE. Isn't it true also that, in terms of economic
growth, if we take from the private economy funds that will yield a
return of 10 percent and invest those funds in a Federal project at
31/8 percent it represents a misallocation of resources?
Mr. STAATS. That is correct.
I don't see how you can avoid that conclusion.
Chairman PR0xMIRE. And it slows down the growth of the economy?
Mr. STAATS. There has been much good work done, Mr. Chairman,
in this area. It is a highly controversial area, as you know, but without
- some statement of policy in this area, cost effectiveness studies which
are an inherent part of the planning-programing-budgeting system
will not be effective. You can get not only misallocation in an agency
but misallocation in other agencies, particularly when they work in
similar fields. The water resources and land area is a good illustration.
Research projects which involve several agencies or manpower train-
ing are other examples. There has to be some consistency in approach
used. If there are variations there ought to be some good rationale or
good basis for making a change.
Chairman PR0xMIRE. Recognizing the limitations which you prop-
erly recognize in your report-since all of us recognize the discount-
ing process is no substitute for intuition, job evaluation, and all that
goes into making these decisions-it would seem to me if the Congress
and the administration are going to have any effective priority sys-
tem, discolmting would be most helpful to us. I remember late last
year when we were so desperately anxious to reduce spending, some
of us were in a very, very difficult position because tite President-I
am sure, with great sincerity-said it was awfully hard for, him to
PAGENO="0013"
9
cut spending. Members of the Appropriations Committee-and I am
a member-were very much aware of this, concerned as to ~where we
could cut properly. It was obvious that an across-the-board cut of 2
percent or 5 percent or 10 percent was most unsatisfactory. If we had
had a much greater knowledge of the comparisons such as we could
get from using the discount ratio in developing benefit-cost studies, I
think we would have been in a far better position to approve those
programs that had a higher yield and, perhaps, to postpone some of
those programs that had a lower yield. Not that that would have been
the only consideration, but it would have been a very helpful one.
It would have been an objective criteria that everybody could recog-
nize as objective and we could have proceeded with much greater order
and t'here would have been a more productive Government investment
tool.
Mr. STAATS. I agree, Mr. Chairman.
I wonder if it would be helpful to ask Mr. Rathbun to comment a
].ittle further.
Chairman PROXMIRE. Yes, I wish Mr. Rat.hburn would comment.
Mr. STAATS. In that connection, I believe it is of particular interest
that appendix I of this report brings out perhaps even more sharply
than my testimony, the variations and the differences in the practices
followed by the different agencies.
Chairman PR0xMIRE. I was struck by page 10, too, of the survey,
and I want to ask about this in a few minutes, showing the agencies
that did not use any discounting technique for the 1969 budget
and five of which, as you say, don't plan to use discounting in the
future.
Mr. STAATS. That is correct.
Chairman PROXMIRE. Mr. Rathbun, do you want to comment?
Mr. RATHBUN. Yes, Mr. Chairman.
I will make several observations.
First, it seems to me that there is a clear and present danger that
the mixture of practices which we have today may give us poorer
results than we would have in the absence of discounting. It is possible,
of course, for agencies to evaluate undiscounted costs and benefits.
They could simply sum the total costs and benefits, and it is not clear
to me that the present mixture of practices will give us a better result
than we would get under conditions in which no one discounted.
Chairman PROXMIRE. May I interrupt at that point and say how-
ever, within the broad category, for instance on the reclamation pro-
grams, isn't it of some benefit to the Congress and budget and to all
of us to have 31/s percent apply, although I think it is a much too low
rate, nevertheless it is better to have that than nothing. If we didn't
have anything here it would seem to me you would have almost a pure-
well, not completely, but a very large political effort on the part of
powerful Members of the House and powerful Members of the Senate,
those situated as chairmen of the appropriate subcommittees, the Ap-
propriations ~ommittees would be in a very strong position to push
prolects that had, even with a. 3i/8-percent discount, a negative cost
ratio. Whereas if you had a program that had a 4 to 1, 5 to 1, or 6 to 1
benefit-cost ratio it is a pretty good argument for it even if you don't
have any political push behind it., whereas if its benefit-cost ratio is
unity itis a little easier to oppose it.
PAGENO="0014"
10
Mr. RATHBtTN. Well, I~ agree that use of the 31/8 percent in these
important water resources areas probably is helpful but I am dis-
turbed by the thought that in other areas, an agency using 12 per-
cent, for example, may be rejecting projects which are really much
more attractive-
Chairman PROXMIRE. I think it is an excellent point. The same
agency-Interior, as you pointed out-uses 12 percent on those proj -
ects which are competitive or comparable to private industry, and
31/8 on some of their other projects.
Mr. RATHBUN. A second observation, it seems to me, is that the stage
is set for substantialprogress.
Chairman PROXMIRE. The what?
Mr. RATIIBUN. The stage is set for substantial progress. In our dis-
cussions with the agencies in connection with their response to the
questionnaire on this subject, we detected a rather widespread acknowl-
edgement that the differences between the. opportunity costs school
and the cost to the Treasury or borrowing school are not terribly
important, are not as important as they were thought to be a few years
ago, because when you approach the cost to the Treasury of borrowing
in a realistic way and include taxes foregone-
Chairman PRox~rIRE~ I think it is a very interesting contribution,
very, very good. That comes to about 7½, 8 percent as compared to
maybe 10 percent for a strict opportunity cost application.
Mr. Stockfisch was going a little higher than 10 percent, he wanted
10 to 20, but I think this is an excellent conception because, as you
say, it reconciles differences of opinion between the cost of the Treas-
ury, the realistic cost, and the foregone opportunity cost.
Mr. STAATS. It doesn't close the gap, but it approaches the oppor-
tunity cost figure of 10 percent as a recognized figure. You are within
a 2- or 2½-percent difference.
Chairman PROXMIRE. Especially the way interest rates are going.
Mr. STAATS. Yes.
Chairman PRoxMnu~. It won't be long before they will be what we
think are consistents.
Mr. RATHBUN. `When you couple this recognition fact that the rates
advocated by the two schools are drawing together with the wide-
spread acceptance of the notion of discounting, I think you have the
stage set for rather substantial progress.
A third observation is that a good deal of additional work remains
to be done even after the agencies move up to a higher discount level
and agreement is secured .on the use of this level. I think of some work
going on in the General Accounting Office now on the evaluation of
benefits. For years and years the benefits associated with reclamation
projects have been based-the calculation has been carried out-on the
assumption that all of the power produced by these projects could be
sold at going rates. To an increasing extent this is not true and there
is going to have to be a lot of additional work on the evaluation of
these benefits.
Similarly in the case of Senate Document 97, where you do have
the discount rate spelled out rather clearly, the various agencies use
quite different practices in, for example, treating the benefits asso-
ciated with the enhancement of the value of shorelands.
PAGENO="0015"
11
So my point is, agreements are going to have to be secured not only
on a higher rate, but on the procedures to be followed in evaluating
the benefits and measuring the costs.
Chairman PROXMIRE. Thank you very much.
I would like to ask Mr. Staats, you said in your statement that some
of these agencies that plan during the future to use discounting tech-
niques, have not done so yet-he said their analysis has not developed
sufficiently to warrant the use of discounting techniques. This just
seems to be incomprehensible to me. Does this mean these agencies don't
know what the benefits of the investments are, don't know what their
costs are? It would just appear offhand that if an agency comes into
the Bureau of the Budget with an investment program and says "We
are not sure what our costs are or we know our costs, but we don't
know what we will get by making the investment," it would be very
easy for the Budget Bureau to say, "No; find out what your benefits
are or we are not going to spend this money," especially now with the
pressure all of us feel to keep spending down.
Mr. STAATS. Of course, a part of it has been, I think, obvious lack of
central guidance in this area. But in part, I think, it reflects the diffi-
culty of measurement, particularly in some of the programs in the
social welfare area-the Great Society type programs. Mr. Rathburi
has had conversations with some of these agencies on this point.
Chairman PROxi~rIIn. Well~ Health, Education, and Welfare is using
a discounting technique for the 1969 program. I don't know how
extensive is their use of it, `but they are using it. The space program
doesn't use it at all. The Post Office Department didn't use it at all;
the Department of Labor, the Peace `Corps, the National Science
Foundation, the Federal Power Commission, some of these agencies
have very small investment programs and are primarily operating
agencies, but to the extent they have any investment programs it seems
to me, and the Post Office Department has a big investment program,
they ought to know what their benefits are going to be, what they are
going to get for their investment and also know what their costs are.
Mr. STAATS. I would like to read a paragraph from the reply we
had from the National Aeronautics and Space Administration which
has a very major capital investment program.
Chairman PROXMJRE. Yes, indeed.
Mr. `STAATS. `They reply along these lines; they say:
The evaluation of `alternatives and `priorities within `NASA is accomplished
principally through the agency's programing and `budget process and within the
framework of the Bureau of the Budget PPB system. Although this is the first
year we have submitted our budget request under the requirements of PPBS,
NASA has since its inception used comparable techniques in development and
evaluation of budget submissions. `Among other things, our programing system
permits U's to consider the total `cost implications in future years of on-going
program's and o'f proposed new programs, and is directly tied in to t'he prepara-
tion of our annual `bu'dget estimates and to our annual operating plan. We use a
standard practice, `advanced analytical methods in the consideration of alterna-
tives. Trade-off type studies relating to performance, costs, schedule and other
pertinent factors are used in project planning and management, and in the
planning and conduct of individual missions. We have not, however, applied
specific discout rates. That is to say we have not attempted to express in terms
of percentage `discounts, the effect of future costs and `benefits, although these
economic factors are implicit in our consideration of alternative programs and
assessment of priorities. Frankly; we find it extremely difficult to quantify such
elusive economic considerations as they affect research and development efforts
in the space environment.
PAGENO="0016"
12
I cite this as one kind of situation where they are not using discount
rates. They say itis implicit. This may or may not be true.
Unless they have some more systematic approach to it it seems to me
it would be very difficult to say in a case of this type that it is implicit in
all of these decisions
Chairman PROXMIRE. My time is up. But I woud just say that if it
is implicit, let's make it explicit. After all if they can say they don't
know what the benefits of space exploration are then maybe we had
better forget about it. They should be able to make an attempt at quan-
tification. If we don't have some kind of objective measurement in these
areas, it just means we are pushing money out in the hope that maybe
it will do some good. We don't know what good it will do and we can't
even make an estimate.
If the Defense Department, with all of the great problems involved
there, and the enormous amount of money, in determining cost effec-
tiveness and so forth can do it as widely as they do it, it seems to me
that space and other agencies that make investment programs should
do it, that they should certainly make their implicit findings explicit so
that we know what we are doing.
Senator ,Jordan ~
Senator JORDAN. Thank you, Mr. Chairman.
Mr. St'aats, you have made a very fine presentation of a complicated
subject here this morning and, as you pointed out, and rightly so, there
are a variety of systems in use in implementing a discount rate by some
agencies, and other agencies use none whatever.
In your report you mentioned explicit and implicit discount rates.
Can you point out the difference between the two? You have touched on
it here in your exchange with the chairman but I think we need to
understand that a little more fully. Will you e1aborate on that?
Mr. STAATS. If I may, Senator Jordan, turn to Mr. Rathbun on this
question. He is far more familiar with this, and I am sure he can give
you a better answer.
Mr. RATHBUN. With respect to the implicit rate, there are a couple of
points that are completely clear. One is that if an agency does not dis-
count in any sense but simply adds up the total costs and total benefits,
the implicit rate is zero. They are not discounting in any way, shape, or
form.
At the other extreme, if an agency makes a decision on the basis of
costs in the first year, makes decisions on the basis of costs and benefits
in the first year, then the implicit discount rate is 100 percent, because
this agency is ignoring all future costs and benefits.
In between these two extremes we have, of course, a host of im-
plicit rates, and the implicit rate in this middle ground is the rate
that makes the present value of costs and benefits over the full life
of the program equal to the costs and benefits over the period used by
the agency in making a decision. In other words, if an agency makes
decisions on a weapons system, for example, that has an actual life
of 20 years, but the agency makes the decision `md ev'tluates the weap
ons system on the basis of 5 ye'tr costs, th'mt is, on the basis of what
happens during the first 5 years, then the implicit discount rate is
the rate that would make the present value of costs and benefits over
the 20 years equal to the costs `md benefits ovei the 5 ye'mr pei iod used
by the agency in evaluating the progr'mm
PAGENO="0017"
13
As you may have noted, there is a table presented in the report which
shows the approximate implicit rates under various conditions.
Senator JORDAN. I can appreciate the problem that is created here
because frequently the benefits are quite difficult to define. I can under-
stand with respect to a water resources project that costs allocable, we
will say, to power or to flood control or irrigation can be determined
quite accurately and definitely by people experienced in the field. But
frequently we find allocation of nonreimbursable costs on hard to
define intangibles such as recreation or enhancement of fish and wild-
life. We find if a benefit/cost ratio is not as favorable as it should be
there is an inclination to transfer some of the costs from the reim-
bursable, compensable area over into an area more nebulous for the
public good, one hard to define and intangible.
Mr. STAATS. What you are saying is quite true and, of course, there
have been efforts to quantify and put in dollar terms many of these
benefits which are virtually impossible to quantify.
Yet, on the other hand, a judgment has to be made and the plea we
are making is that those judgments should, to the extent possible, be
consistent both within the agency and among the agencies if the
budget is to reflect the priorities which are based on a total assessment
of all benefits and all costs. This is as true in other fields, more true,
perhaps, than it is in the water resources field, where we have been
employing benefit/cost ratios to determine these projects that go into
the budget for a great many years.
It goes back well beyond, as you know, the issuance of Senate Docu-
ment 97.
Senator JORDAN. Yes.
Mr. STAATS. It is about the only area where, until the PPB system
was initiated, we have had this kind of test to evaluate proposed
programs.
But even the Corps of Engineers follows, as you know, somewhat
different practices in assessment of benefits than the Bureau of Recla-
mation.
Senator JORDAN. Yes.
Mr. STAATS. Sometimes in the same river valley. This is the kind of
inconsistency that we would like to see removed.
To go back to your question on implicit versus explicit rates, if you
have a copy of our report, on the bottom of page 14 is brought out
fairly graphically an illustration showing how this affects the Vet-
erans' Administration as against the Department of Commerce
program. (.See p. 51, this volume.)
I would like to ask Mr. Rathbun if he would elaborate on this. Here
is a case where they are both talking about structures and the capita]
outlays and they are using different practices, so that you obviously
are going to come out with different benefits/cost ratios.
Mr. RATHEUN. Well, as pointed out there, in the case of the VA, the
discount rate is 20 percent, whereas the Department of Commerce
practice involves the use of an implicit rate of only 8 percent. Ob-
viously this is most unsatisfactory. If the two programs are of equal
merit, the one in the VA would be rejected out of hand `because of this
use of a very short period of time in evaluating the program, and the
associated implicit rate of 20 percent. Whereas the same stream of cost
and benefits would yield a much higher benefit/cost ratio in the De-
89-654 O-68-3
PAGENO="0018"
14
partment of Commerce because they h'ive choseii to use `t form of
analysis that involves `i much lowei implicit r ite
Senatoi JORDAN This is `i good example `md it does pomt out th'mt
developments of equal merit, if used under the application of a differ-
ent type of discount rate or process of establishing a different kind of
discount rate, might find themselves discriminated against or coming
out one against the other unfavor'ibly
Mi RATHBTJN Mr Jordan, I ~iould like `ilso to sty, `mmd this is
pointed out in the report th'mt there `ire undoubtedly c'mses where the
benefits c'in't be quantified, but even here it w ould seeni helpful to use
`m reasonable discount rate `md `ipply it to the costs md present to the
decisionmaker the present v'mlue of the costs ot these two progr'mrns
`mmd let him use his judgment in pl'icing `m ~ mine on these h'mrd to quan
tify benefits
Senator JORDiN Would it be youi iecommend9tiom then th'mt `ill
programs should have some kind of discount rate applied to them,
even if they are intangible and hard to define?
Mr. STAATS.YeS; I think that would be our view.
The point I w ould like to emph'msize, md w hich Mi R'mthbun has
already made, is that w e are not suggesting a h'mrd `mmd f'mst formul'm to
be applied in every case
Sen'ttor JORDAN Yes
Mr STAATS What we are suggesting is `need `it le'tst on the cost
side to apply a common policy in developing one of the essential pieces
of mform'mtion th'mt the Congress and the executive branch ought to
have in arriving at their judgment. It is the kind of information that
any prudent businessman would rely on for any new investments lie is
m'mking w hether he is t `ilkmg `ibout h'ii dw `ii e, i ese'ii cli `mmd de\ elop
ment, or welf'mre programs for his employees We `ire saying `is `t
minimum it ought to be the cost of money plus `i t ictom foi t'ixes
foregone There m'm~ be ~ `irntions from th'mt, dependin~ upon the
individual program, but at least it would give you a benchmark of
policy which could be `mpphed consistently thioughout the executi~ e
br'mnch
Senator JORDAN. What you are saying is that as these agencies com-
pete for the taxpayer's dollar for development we ought to use a
common yardstick insofar as possible?
Mr. STAATS. That is right.
Sen'mtor JORDAN In ev'mlu'itmg theii merits1
Mr STAATS Otherwise w e m'iy be speiidmg more th'mn w e ire gomg
to get b'ick We will be m'iking in mi estment in m p001 i enture
Th'mt is the d'ingei, `mmd that is the ie'ison prii `ite indiistrm m'tkes
these kinds of `in'mlyses
Senator JORDAN. Do you think the skill in evaluating programs is
improving `is more and more peo}~le become conscious of the need foi
est'iblishm~ some kind of priority in spending
Mr. STAATS. Yes;~ I think so. In my own experience going back ni
the budget field for `i good many ye'irs it seems to me w e `ire m'mking
progress in how to do this It is nei ci gomg to be pei fect There `ire
`mlii `iy~ gomg to be dis'igreemeiit'~ `imong people ii ho ii e foi 01
`mg'iinst `m gn en progr'im But the f ict th'it w e h'ii e now developed
centr'il st ifs in the Bure'mu of the Budget `mmd in the other `igencies
in the executive dep'mrtments, who `ire m'ikmg this `i speci'il concern
PAGENO="0019"
15
is a useful step forward, and the fact that the Congress more and:
more has been asking for these kinds of justification when agencies
present their authorization requests and their budget requests is also
a hopeful sign. In general terms, I think there has been progress but
we have a long way to go.
Senator JORDAN. Thank you. My time is up.
Chairman PROXMIRE. I notice on page 10 of your report you have.
a list of the agencies with which you made your inquiries, and we find
here that., as I said before, in a number of them, they did not use any
discounting technique in 1969, and five indicate they have no plans to
use discounting techniques in the future, to wit, the Interstate Com-
merce Commission, the Export-Import Bank, the Veterans' Adminis-
tration, the Department of the Treasury, the Department of
Commerce.
Now, especially when you come to the latter three, Veterans' Admin-
i~tration, Department of the Treasury, and Department of Commerce,
they all have substantial investments. How can they justify the asser-
tion that they have no intention of using discounting techniques. Cer-
tainly the Treasury, where they originated the formulas that we have
used, certainly ought to have an acute notion of the future compared
with the present value of the dollar. The Treasury should recognize
the importance of discounting. What kind of responses did these agen-
cies give to you when you asked them?
Mr. STAATS. Of course, the table on page 10 does not fully state the
views of the agencies. There is a more detailed statement `in appendix
I but let me, if I may, Mr. Chairman, read a more recent letter that
came from the Department of Commerce, which states in part:
While we do not now use discount techniques in our programs universally and
do not have definite plans for the ~ of these techniques in the future, we are
considering the use of discount analysis for the evaluation of some programs.
such as capital outlays. We agree, therefore, that a standardized concept should
be adopted for Government-wide use. While we have not reviewed the matter in
depth we believe that the second alternative appears to be more practical;
namely, that of recognizing that the cost of Treasury borrowing as a cost of
investment in the public sector.
They do not make an unqualified statement that they would not. do it.
They are first seeking central guidance and second they are-
Chairman PROXMIRE. If the Department of Commerce recognizes
the cost of Treasury borrowing as the, what do they say, opportunity
costs?
Mr. STAATS. No; Treasury borrowing.
Chairman PR0xMIRE. The' cost of Treasury borrowing on this real-
istic basis of foregone taxes and makes it 7.1/2 and 8 percent, what does
it really mean unless they say they have a way of calculating what
their benefits are and then what their costs are and use the discount
system to come up with a. basis for a decision? Simply t.he awareness
that the costs foregone are 71/2 or 8 percent., it seems to me, isn't a very
good guide unless they tell us what they do with this.
Mr. STAATS. That is, of course, true for most of these agencies that
were referred to, which have a sizable capital outlay.
Chairman PROXMIRE. The Veterans' Administration and Depart-
ment `of Commerce especially, but, of course, the Post Office Depart-
ment does, too. They indicate that they plan to use it sometime in the
future. How about these other agencies that did not use it in 1969 but
PAGENO="0020"
16
plan in the future to use it, when-I don't mean to ask you in detail
on each one, but wh'~t w `~s the gener'tl feeling ~ rj~~ they w ould be
able to move in 1970, 1971, 1972, or soon
Mr RATHBTJN First, let me refer b'tck to the Dep'trtrnent of Corn
merce and the Veter'tns' Administr'ition, the two ~gencies that st'ited
they did not use discount r'ttes It should be iecognized th'tt in the
c'~se of the Veterans' Adirnmstr'~tion they do ev'tlu'tte progi `trns in
terms of cost md benefits over `t ~ ye'ir period
Chairm'tn PROXMIRI Yes, you gtve an example of th~t
Mr.RATHBJN. So they are using a high implicit rate.
Furthermore,. there are elements within the Department of Com-
merce that do essentially the same thing. In one case they have used
what h'tppens over the first 7 ye'trs for progr'irns h'tving an estim'~ted
life of 11 years.
Ch'urmqn PROXMIRE So this st'~tement on page 10 is not-I don't
ine'in it is misleading, but unless you re'~d it in connection with what
Mr R'tthbun just told us, you get `t misconception It says discounting
techniques not used md no intention of using it fins isn't true They
use it but they use it in a way which is not stand'trdized `rnd not
rehted to wh'tt othem depariments are doing and it goes right b'tck
to your initial skepticism, Mr. Rathbun, this inconsistency is almost
worse than not having a system at alL
Mr RATHBTJN It m~y very well be
Churman PR0xMIRE It may well be in some cases
Mr. RATHBUN. Now, with respect to your question about when these
`igencles will get underway, I c'tn't recall `~ single case in which some
time w'is specified In t'dking to the people some of them seem to feel
they `tre going to be in `tposition to do this right `tn `ty, in the next year
or so Others `tre somewhat more undecided, `tnd I suspect it will be
2 or 3 years
Chairman PROXMIRE Well now, that brings me to what I think is
the most encouraging statement we have he'trd this morning and that
w'ts your st'ttement that you think ~ e may be on the verge of `tbre'tk
through, you said something like that, you s'tid we m'ty be on the verge
of making this much more univers'illy `ipplicable `ind much more effec
tive. What can we do as Members of the congress? Do you recommend
any legisl'ition that we c'tn adopt, Mr St'i'tts or Mr 1R'mthbun, that
would be helpful here or is this something where our inquiry and youi
inquiry and publication of it `tnd the welcome position the President
has taken this morning in his budget on the w'mter resources projects
indicates this is coming along ~ eli `ind indicates we have to pursue oui
inquiry and let other Members of Congress kno\\, but it doesn't need
legislation ~
Mr ST&ATS I h'iven't pursued the in'itter to the point of coming to i
conclusion-
Chairman PROXMIRE. We would like to hurry it up.
Mr STAATS Whether legisl'ition `is such would be necess'uy, I ~ ould
say that unless we can show more visible progress th'tn we h'ive hid,
some further `tction on the p'irt of the legisl'ttive bi `inch w ould cer
tainly be in order
It had been our hope th'it this iepoit, together with the report of the
.Joint Economic Committee in December might result-
Chairman PRoxMI~ This repoi t ought to shock Congress I think it~
indicates, `is you s'ty, `iterrific disp'irity, `number of `tgencies, `ibout
PAGENO="0021"
17
half of them, that don't use it, didn't use it in the budget; and without
discounting techniques there is no objective basis on which you can
evaluate investment programs and establish any kind of objective
criteria.
Mr. STAATS. Plus the fact that there is continuing through most of
the replies that we received from the agencies, an indication of recog-
nition of the need for it.
Chairman PROXMIRE. So we know they are misallocating resources
within the Government and misahlocating resources that should be
available in the private sector.
Mr. STAATS. If I may make a suggestion, Mr. Chairman, possibly
this committee miglht consider referring this report to some of the
agencies as well as to the interested committees of the Congress, with
a view to possible later hearings after they have had a chance to review
the report. Our report was completed, as you know, only in the last few
days.
Chairman PROXMIRE. I think it is an excellent suggestion.
Mr. STAATS. While we have given the agencies the benefit of the
replies we have had, they have not had access to our completed report.
Chairman PROXMIRE. I know it. is awfully hard to give us, on the
basis of this inquiry especially, but. could you or Mr. Rathbun or Mr.
Marvin, any of you gentlemen, give us an estimate of the percenta~e of
Government investment programs that now use the discount technique,
and the percentage that don't?
Mr. STAATS. I don't believe we made any-
Chairman PROXMIRE. You can't. even make a rough approximation,
whether there are half of them? We have to do this, of course, within
the definition that Mr. Rathbun has suggested because there is an im-
plicit discounting, but it would be helpful, I think, to know how much
of this enormous amount we are investing, billions and billions of
dollars a year, is based on a discounting technique that enables you to
apply objective criteria and how much is not based on it as well as
to know the very important discrepancies between the systems used.
Mr. STAATS. It might be useful and meaningful.
Chairman PRoxmRE. It would help if it were not only in terms of
percentage, but in terms of dollars.
Mr. STAATS. Yes.
Mr. RATHimN. I think it is clear at this time it is probably well
above half.
Chairman PRox~rIRE. Probably what
Mr. RATHBUN. Well above half.
Chairman PRoxmRE. Well above half that use the discountiiig?
Mr. RATIIBUN. Right, I say this because partly the Defense Depart-
inent with its very, very large expenditures does use the discount rate,
but I would agree with Mr Staats that a detailed-
Chairman PRoxl%rIRE. How about the consistency within the De-
fense Department?
Mr. RATITBUN. It. is consistent. It is it) percent.
Chairman PRoxMIRE. Ten percent, is that right?
Mr. R.ATHBUN. I think the Defense 1)epartrnent has gone through
to a very considerable extent what some of these agencies are going
to go through. Three or foui years ago in the Defense Department
there was a grea.t controversy about the propriety of discounting these
differences of opunon were eliminated and then there was controversy
PAGENO="0022"
18
on over just what the rate should be At the present time there are no
questions raised `~bout the propriety of discounting `~nd 10 percent is
used on `tn `ieross the board basis
Ch'urman PROXMIRE I iimtgine the fact thit the Defense Dep'~rt
ment now uses a 10-percent discount, uses it on an across-the-board
b'tsis on the enormous `tmount of investment they m'~ke, m'~de `t gre'tt
difference in their expeditures If they h'td not been using that, but
using inste'td `t 3 percent or 31-/s w hich h'is been used by Interior 01
Corps of Engineers the expenditures ~ ould h'~ve been `~ great de'tl
more You can `trgue this has s'ived the t'txp'tyer billions of doll'trs,
w ithout any exagger'ttion, for ye'u s
Mr STAATS In a very recent letter which we received from the
Defense Department they first s'Ly they agree on the desirability of `t
gre~tei degree of st'tnd'~rd~z'~tion of Gm ernment prtctice Then they
`Liso say that
Our discount rates correspond closely to the opportunity cost concept, but in
arriving at our rates we did not use exactly the opportunity cost concept as you
describe it.
Incidentally, the Department's regulations, are fairly recent, about
2 years old The Dep'irtment's letter goes on
Since our reasoning led to what I believe to be the same end result I doubt
that the differences are worth noting
My staff \\ould be glad to discuss this mattei in moie detail uith ~oi.ii staff
if you wish I doubt that the cost of Treasury borrowing is the rele\ ant rate
foi the t~ pe of probiems ~ ith w hich w e are concerned The data w e ga~ e on
the projects where discounting was used and the analysis considerably undel
btates the use of these practices The 73 projects mentioned are those where
the discount calculation actually appeared in the budget submission. There must
be hundreds of other. investments decisions in which discounting techniques are
used,. but which do not show up in the data because they involve amounts that
are not large enough to require individual consideration in the budget or because
the discounting techniques showed the project was not sufficiently attractive to
warrant submission Moreover discounting concepts are used on lease or bu~
decisions and we did not include these in the list For example the recent
~ir Force decision regardmg its Phase II computers `.~ ith ~ hich I think you
are familiar was based on data that included an analysm that calculated the
implicit interest cost of leasing for a year
Chairman PROXMIRE I think that is a most helpful response, and it
`Suggests we ought to give more consideration to opportunity cost ~s
~ dl `ts Treasury costs of borrowing ~Jthough they `~re coming to
gether Mr Rathbun estimated much more than half of investments
by the Government is subject to discounting techniques because of the
Defense Dep'~rtment, excluding the Defense Department, I take it on
the b'~sis 0± these very big agencies that have big investments pro
grams but don't use discounting techniques th'Lt it is very possible th'tt
subst'inti'tll~ less th'tn h'ilf of the nondefense sectoi of our budget is
subject to discount
Mr STAATS That would be my guess If you exclude the Defense
Dep'irtment it `v~ ould be subst'intnily less th'tn h'tlf
Chairman PnoxMIIui I think you `inswered another question I h'id
in mind I was going to ask you what sort of discounting method the
comptroller sees as most fe'isible Is there `i difference bet~w een st'tnd
`trdization in form and so on ~ I think you de'dt with that unless y ou
wanted to comment further
I would like to ask, have you made any estimate of the misallocation
costs of the Government, of one agency using explicit rates, one using
PAGENO="0023"
19
implicit rates, or one using discounting and one not applying dis-
counting?
Mr. STAATS. I don't think we have done anything beyond pointing
out the possibility or likelihood that this has happened. We have not
actually gone beyond that. point, that is to say we have not actually
examined the Appropriations Acts from the standpoint of those items
which were excluded or included on the basis of higher or lower use
of discount rates. I suppose this would be possible, but our main pur-
pose in this report was to bring out sharply the differences in practices
among the different agencies and, therefore, the need for a more com-
mon policy concerned.
Chairman PROXMIRE. Then you can properly conclude on the basis
of this study that the Federal Government. has been engaged in prac-
tices which do inevitably result in ~t substantial amount of misalloca~
tion of resources. We know that. You are not giving us an estimate,
but my own estimate is that it would be in the billions of dollars a
year as compared with the kind of investments we would make if we
had a standard system which was used throughout Government-and
not rigidly, still permitting value judgment, intuition and going ahead
with low yield programs if there were other reasons for doing so-
if we had this criteria, a basis for the Congress and President making
a decision.
Mr. STAATS. This certainly could be a proper conclusion. I think un-
less a decisionmaker has this kind of information before him, certainly
the opportunity is there for errors of judgment, and I think it would
be almost certain that this does occur in the absence of this kind of-
Chairman PRox~1InE. You showed a dramatic example of comparing
the Department of Commerce and the Veterans' Administration with
respect to a building, one 5 years and one over a period of, a greater
period of years, 8-percent discount compared to 20-percent implicit
discount.
Mr. STAATS. Of course, as you know-
Chairman PRoxi~m~. Of course, where you have the Defense De-
partment investing on a 10-percent basis, and by and large the water
projects, and so forth, on a 3-percent, three and a fraction percent
basis, we know there is a misallocation there.
Mr. STAATS. We used this particular illustration because they are
both in the nature of construction type programs, I think at. the same
time though we would have to recognize they are in noncompeting
areas. This 1)1'Oblem that we are talking about comes up most. sharply
where you have agencies who are in a seiise in competing programs.
The land and water resources field is a good case in point, where you
have Agriculture, Interior, TVA, and the Public Health Service, all
concerned with investments in water supply or water improvement.
It also comes up dramatically in any other areas where you have a
miumber of agencies that are concerned with carrying out common ob-
jectives, conimon programs on a Government-wide basis, and simi-
larly as in the example we brought out. here in the Interior Depart-
ment, even within a department. If you put the Department of In-
terior, budget together, all of these programs in some sense or other
are related, so that you need a common policy to assure even within a
given department that they are following a common policy.
Chairman PRoxl%IIRE. I want to come back, but my time is up.
PAGENO="0024"
20
Senator Jordan ~
Senator JORDAN. Mr. Staats, you note that agencies do not evidence
a clear understanding of the discount rate technique?
Would y~o~i expand on that a little bit ? Is it because they lack the
st'tff capability to implement it or `tre they unwilling to f~ce up to the
rigors of the discipline required9 How do you expl'im it~
Mr STAATS I would like to he'tr from my colle'tgues here on this
point also But nay own person'tl feeling would be th'it it is not so much
opposition as that there hasn't been adequate staff who understood
and knew how to employ these techniques. I think it has been more
in that category than in the resistance to the idea. Would you care to
comment, Mr R'tthbun ~
Mr RATHBTJN I would simply idd th'it in in'my c'ises they have
never been asked to do this. The Bureau of the Budget has, of course,
`isked them to think of discounting in the c'ise of some projects, but
many, m'tny agencies hqve nevei been iequested to give `my serious
thought to discounting, and `is `t result they h'i\ e not given the requisite
i ime `ind attention to this
Senator JORDAN Were you going to s iy something9
Mr STA &TS I ~i `is going to `tdd th'it it is only f'urly recently, within
the last 3 ye'irs, that the executive hi `inch h'is formally extended its
cost effectiveness concept on `i Government wide b'tsis, `ind th'it h'ts
brought this problem more sharply into focus. I think that without
cost-effectiveness studies being made on these major programs there
h'isn't been too much re'ison for employing `t discount rate But in the
water resources field, where we h'ive come to underst'tnd `ind use this
techmque o~ ci m'tnv ye'trs, you do h't\ e pretty highly refined tech
mques w ithm those `tgencies now bec'tuse they `tie requned to use it `is
`t b'isis for `my project before it is `tuthorized `tnd ig'tin befoie `tppl.o
pri'ttions `ire rn'tde The `tnsw ei ~ `tries `i gie'tt de'il depending upon
w hether or not there h'is been `t foi m'tl progi `tm which requires the
use of these types of cost-effectiveness techniques.
Senator JORDAN I would be interested, Mr St'iats, in what the
Post Office Department has to say `ibout it They `iie listed here in youi
table on page 10 as being in the category of "Discounting not used but
plans to use discounting in the future" Do you h'ive any paiticular
comment w ith respect to the Post Office Dep'trtinent2 Wh'tt h'ts been
their `ittitude mci how effecti\ e w ouki you think discounting tech
mques w ould be in the Post Office Dep'titment9
Mr. STAATS. I would think, Senator Jordan, so far as their capital
investments progr'tms particul'irly `ire concerned, such `is f'tcilities 01
buildings, this type of technique would be very relevant. It would be
perhaps easier to employ `there than almost anywhere.
Senator JORDAN. The third line from the botton on page 10 is where
they `ire listed in the table
Mr. STAATS. In their formal letter to us the Department stated that,
"The Post Office did not use discount i `ites in in'ilyzing fisc'tl ye'u
1969 progr'tms However, in cert'un inst'tnces discounts `ire `tpphed to
individu'tl projects For ex'tmple, in e~ `idu'tting w hethei to `ib indon't
Feder'tl building for a new le'tsed f'icihity 01 to extend `tnd modernize
it by using plant `md equipment funds, the i cut sti e im foi `t40 ye'u
period is estumited The ~1 esent w orth of the i cut sti e'tm is then
determined by `ipplying i. 4 percent i'tte fictoi Ihe 4 percent h'ts been
PAGENO="0025"
21
applied for several years and does not take into account the current
high interest rates. The examination of the two choices leads to the pre-
ferred alternative. The Post Office Department also used discount rates
in certain reports required by the Congress. For example, Public Law
90-15 requires reports on those leased buildings which exceed 20,000
square feet of gross interior space. A comparison is made between the
annual cost of leasing and the annual cost of the proposed facility if
it were constructed for Federal ownership. The total cost of the
project is developed and then converted to an annual cost by using a
41/2-percent rate for the period of time of proposed usage by the Gov-
ernment. Depending upon the project this period of time may be from
15 to 40 years at 4l/2-percent~ interest. This annual cost of con-
struction for Federal ownership is then compared against the annual
cost of leasing as a basis for choosing the preferred alternative."
Then they outline further studies which they are making with the
purpose and intent of extending this concept to other parts of the
Post Office Department.
I think here again it comes back to the point we were discussing a
minute ago and that is the fact that this is still in the process of being
extended and employed in different programs with the agencies. But
again they are doing it without central policy or guidance.
Senator JORDAN. That part of any agency or Department's budget
that relates exclusively to services wouldn't be adapted to the applica-
tion of this kind of technique, would it?
Mr. STAATS. Well, now, they do indicate that they are taking further
steps to try to improve the studies that they have been making. They
say, for example, the Department is generating standardized guide-
lines which will stress the use of discount rates in comparative invest-
ment decisions. They further indicate that they are developing a
methodology for major facility projects including an illustrative ex-
ample of alternative life cycle costs and state further that the dis-
counting adjustment is applied to the current dollar estimates to con-
vert the expenditures to a present value basis. This correction enables
a uniform comparison of investment outlays even though they are
made in different years.
This is indicative of the further studies which they are making with
the intent of applying this on a more extensive basis in the Department.
Senator JORDAN. If I may turn now to the Department of the In-
terior and the various rates that are used there in discounting, will
you tell us, Mr. Staat.s, what is the average maturity rate of Federal
long-term obligations now, do you know?
Mr. STAATS. Those held by the Treasury?
Senator JORDAN. Yes.
Mr. STAATS. Already issued?
Senator JORDAN. Yes.
Mr. STAATS. The average has been coming down because of the recent
experience that I referred to in my statement. More and more of issues
are on a shorter-term basis. We can supply more specific information
for the record.
(The following material was subsequently supplied:)
Current information concerning the average tinie until maturity of Federal
long-term obligations can be found in the December 1967 Treasury Bulletin.
The following table is taken directly, or is computed from data on page 31 of
that issue of the Treasury Bulletin:
89-654 0 - 68 - 4
PAGENO="0026"
22.
MATURITY DISTRIBUTION AND AVERAGE LENGTH OF MARKETABLE INTEREST BEARING PUBLIC DEBT
Total amount Percentage in maturity classes Average
End of fiscal year outstanding - length
or month (millions) Within 1-5 years 5-10 years 10-20 years 20 years (years)
1 year and over
1958 166,675 40.7 25. 5 12. 9 16. 6 4. 3 53l~
1959 178,027 41.0 32.8 9.6 12.1 4.5 4~%s
1963 203, 508 41. 9 28. 5 18. 4 4. 1 7. 1 5342
1967-June 210,672 42.6 33.9 11.5 4.0 8.0 47%~
1967-November_~-_ 226, 081 45. 2 34. 2 9. 5 3. 7 7. 4 434~
The average length of both short `md long term obligations is shown in this
table It is clear that the average hac declined mainly because of the decrease
in 10-20 year maturities and the increase in under 5-year maturities, relative to
the total In fact most of the $60 billion inclease in the total has been in
maturities under 5 years
Sen'~tor JORDAN This is the point I w'inted to lead up to, the fact
that the maturity time of outstanding so-called long-term Federal
borrowings is growing shorter and shorter because more of the debt,
as it mat.urès,is being financed on a short-term basis.
Mr STAATS Con ect
Senator JORDAN. And, therefore, a rate of 31/4 percent which has been
calculated and used by some agencies, some agencies of the Department
of the Interior, is an unrealistic rate in the context of the present
interest costs. Is that a fair statement?
Mr. STAATS. That is correct.
Senator JORDAN. And it is not at all likely to be an effective or
realistic rate in the very near future.
Mr. STAATS. I would agree.
Senator JORDAN. There was a time when it was set up, no doubt, that
it did represent a meaningful cost~ of money because they used an
`tverage r'tte of long term borrow ings But since, in the hst 10 ye'u s,
interest rates have been escalating at a rather rapid rate, any rate based
based on eaiher prognostic'ttions of wh'i~t the tot'tl m'rturity of `dl
investments is would be outmoded in the present rate structure of in-
terest in the cost of money
Mr STAATS Yes And I think the st'~tement from the budget message
that the chairman read reflects agreement on this point now.
Senator JORDAN. Before we get through with this, do you expect
that we will be able to draft legislation that would give us some guide-
lines that would help establish those uniformities?
Mr. STAATS. I would not want to rule this out of the picture. It would
be my reaction, however, that it may not be necessary or, perhaps even
desirable right itt this moment because the subject needs more debate,
it seems to me There needs to be certun explor'ition of the possibility
0± i econcihng the opportunity cost concept which p1 `tctically `dl econ
omists would `trgue foi, to the ide'~ th'tt v~ e `tie suggesting in our
report-of taking in `tddition to the yield r'tte on the long term issues
`t taxes foregone kctor w hich ~i ould bring you w ithin tw o to hi o `tnd `t
half percent of the figure \vhi~h most of the opportunity cost ad-
vocates would support, which is around 10 percent.
Senator JORDAN. Most of the agencies that are using this device now,
that are using this kind of technique, are doing it because of adminis-
trative direction r'tther th'tn by legislation
PAGENO="0027"
23
Mr. STAATS. That is correct; yes. So far as I know there is no legis-
lation on the statute books which governs the discount rate on any
program.
Senator J0IWAN. Execpt for the statutes having to do with water
resources development.
Mr. STAATS. Of course, that is not a statute. That was an executive
branch action, and it was incoporated in a Senate report or Senate-
issued document.
I think you could correctly interpret that as having the agreement
of the Senate Interior Committee. But it was not a legislative action
on the part of Congress as a whole.
Senator JORDAN. I think it is important to bring out that such im-
plementation of the technique of discounting as is presently done is
wholly by administrative order and without being based on legislation.
Mr. STAATS. To the best of my knowledge there is no legislation on
the subject.
Senator JORDAN. In your opinion, the wisest course to follow is to
see if we can get some voluntary compliance before we even start
talking about legislation.
Mr. STAATS. I think the danger would be to try to enact into law
a specific fixed rate. I think there would have to be room for flexibility.
Senator JoRDAN. Yes.
Mr. STAATS. You could not, it seems to me, specify a fixed rate.
If the concept could be written into a law which would still rec-
ognize the need for variations in individual situations, that should
be done.
Senator JORDAN. Which would still permit the flexibility that the
system would require in its application by the several departments.
Mr~ STAATS. Yes. I think this would be the area where you would
have difficulty at this point in time, that is in being very specific about
what flexibility under what circumstances, you would permit in given
programs. That would be the problem.
Senator JORDAN. Thank you, sir.
Chairman PROx~rIRE. I would like to ask you, Mr. Staats, something
that I think can give us part of the answer to this difficulty we have
in the Federal Government with its great divergency in applying the
discount factor.
W~hy, of all of the subagencies of the Defense Department is the
Corps of Engineers the only one that has this 31/4 percent or 31/8
discount rate, meaning a return of 3% percent on the investment in
that area, where all the rest of the Defense Department has 10 percent,
virtually all the rest?
Mr. STAATS. Well, I think-
Chairman PROXMTRE. After all, the Defense Department knows
what it is doing; it has pioneered in this area and has a clear rationale
for applying opportunity costs.
They made a fine response, and yet in the Corps of Engineers they
have an exception.
Mr. STAATS. The Corps of Engineers follows a different rate because
it was agreed on for all the water resource development programs in
1962.
The purpose was to get the consistency of the discount rate to be em-
ployed by the Interior Department, the TVA, the Corps of Engi-
PAGENO="0028"
24
neers-'tll of the `tgeneies concerned with w itei i esoui ces develop
ment
Chairman PROXMIRE. In other words, what you are saying is that
the President, with the acquiescence of the Congress, a little more than
acquiescence in view of Senate Document 97, and the Government
decided th'it they `ire going to gi~ c preference `md `i ~ ery decisive
preference to w'tter projects o~ er othei projects of the Defense De
partment, `tnd they `ire going to do it by h'i\ ing `t discount f'tctor
which tells people th~t `t project will yield `i benefit higher th'in its
cost in'i w'iy th'it is self deceptive
If you `ipphed the 10 percent discount f'ictoi inste'id of the 3 pci cent
discount factor, I think most of the water projects never would have
been funded.
Mr. STAATS. It certainly would have ruled out a. great many, no
question `thout it
Ch'mrn'in PRox~rIrr A gre'it m'tny And on the b'isis of leports
th'it I h'is e h'id wheie I h'i~ e h'id w heie I li'i~ e inquiied `ibout them,
I find th'tt most of them would h'ive `i less th'in umty benefit cost r'ttio
It is a policy on the part of the Congress to favor the water projects.
This m'iy be `t good policy, but I think Cong,iess should do it moie
with its eyes open, `ind w e w ould bc better off if w e h'id `i umfoim
system `ind s'iid, "All right, we `tie going to fund the w `itei pio~ects
if the benefit is half the cost."
Mi ST~&TS But to st'tte the othei side of the issue `i bit, there `ire
those also who would argue that they have consistently uncleresti-
m'ited the benefits
Ch'iirm'tn PPoXMIPE That is possible It is ~ ei ~ possible But I think
it is important, if that is the case, that. the benefits should be more
ie'tlistic'tlly `ippi `used i `ither th'in foolmg `ii ound with the discount
f'tctoi, `ind h'tvmg this cle'u disci imination
Mr STAATS I believe from pci son'tl experience-
Chairman PROXMIRE. This is why I differ a little bit in your reply
to Senator Jordan when you said there is a danger in a fixed rate. I
am not so sure that we should rule that out, because I think a fixed rate
might have its benefits, or `it le'ist if not `t fixed r'ite, `in `issei tion
of the determm'ttion on the p'irt of the Congress, detei min'ttion on the
p'irt of the Government, to h'ive `t st'ind'trdized system th~ oughout
You might w'tnt to v'iiy the rate to everybody foi ie'isons, `is in
terest rates go up or for various other reasons.
Mr STA~TS I thmk my response would be `i little different if we
were thinking in terms of `t inimmum or floor, `ind then go higher foi
such f'ictors `is obsolescence, the difficulties of estim'tting the benefits
and the costs, and risk factors. Even a. fixed rate--
Ohairman Piiox~nRE. But, you see, my argument is that this should
not be a substitute for decision. This is a guide to decision, a help to
decision. In other words, all this would do would give you what the
discount factor develops in terms of the m'tthematic'il formul't Then
you have to use your own v'tlue judgment
Then you have to step in `ind m'ike youi own decision, `ind it
would seem to me if you h'ive a unifoi iii system then you `ire in `trench
better position to make `t clear cut decision b'ised on these other
factors You know what you are working with
Senator SYMINGTON Would the chairman yield ~
Chairman Pnox~nnii Yes
PAGENO="0029"
25
Senator SYMINGTON. It would appear we get more benefit from
water development than from some other aspects of the Federal
budget. Correct me if I am wrong.
For example, in my State we had a close question on a certain
project when it came to cost-benefit ratio. As a result, we finally built .a
dam; and whereas nearly every rural county in Missouri has lost pop-
ulation and lost income, this particular group of counties, as a result
of this dam, has had a tremendously increased economic gain, to the
point where bank deposits are now over 300 percent more than they
were a few years ago.
This, of course, creates a much-needed additional tax base, not only
for the State but for the Federal Government.
It is my experience that when you develop water on a proper eco-
nomic basis it is a big help to any economy; which I would not think
true of some other investments made on other Federal programs.
Chairman PR0xMIIIE. If the Senator would yield on that point-
Senator SYMINGTON. Perhaps I misunderstood.
Chairman PRox~IIRE. No; I think it is a very excellent point. It
comes right to the crux of the problem.
I think the difficulties, however, are that the benefits may very well
have been underestimated for water projects, in which case I think we
should do a better job of estimating the indirect as well as the direct
benefits. Then you are in a position to make your evaluation.
The Senator from Missouri has far greater experience in private
enterprise than virtually any other Member of the Senate. I think
you would agree that any businessman, in making an investment,
would want to know his rate of return, and would want to have it
standardized so that regardless of whether he was investing in some-
thing that was close to his heart or somebody else's heart, he would
have the facts, the figures, uniformly applied, and then he makes his
decision.
Senator SYMINGTON. Well, I say to my chairman that the penalty
for being late is probably expressed by my observations.
Chairman PRoxi~rIRE. No, they are very good.
Senator SYMINGTON. I am sorry I could not have been here all
the time.
Mr. STAATS. If I might comment on what you and Senator Syming-
ton have been saying here, I think we would all agree that an effort
to relate benefits and costs on any investment-type program is a desir-
able thing. This is like being for home and motherhood. I think this
is an invaluable kind of technique.
The real problem is in terms of how good are our estimates of
what the benefits and costs are going to be for the future, and we are
projecting ahead a good many years. In water resource projects, for
example, we have had the reclamation law on the statute books since
1902, and we had Corps of Engineers programs long, long before that..
I think the Congress and the Executive would both be in difficulty
if we do not have some kind of a cost-and-benefit analysis in connection
with these water projects, because it not only affects the decision of
whether you are going to build at all or not, but in terms of what
combination of benefits. Are you going to put recreation in? Are you
going to put power in? Or are you just going to have flood control? Are
you going to have reclamation as well?
PAGENO="0030"
26
~ Now, this same principle carries over into other programs as well.
It is just that we have not. developed these techniques as far in some
of these other programs `~s w e h t~ e 1n the w `ttei i esoui ce field
Chairman PROXMIRE A lot of people feel discounting in i'~ `ipply
to w `itei piojects `ind `tlso to Defense Dep'iitment in~ estrnent, but it
does not `ipply to hurn'in resoui ces We h'id ~ ery excellent testimony
from the Assistant Secretary of HEW, Mr. Gorham, who did a fine
job, I thOught, of showing where they applied discounting and cost-
benefit study to their studies in the National Institutes of Health, and
found that they can have an excellent guide to their decision on the
basis of carefully figuring the benefits and the costs and finding out
that some programs have a far quicker and better payoff in saving
ln es, `ind so forth, th'in others
So that I think we are underestimating the value of this across the
board in a very great number of areas where it is just beginning.
There h'is been some pioneering in thc Defense Dep'trtment and
elsew hei e, but this is the kind of technique th~t c'ui be most helpful
to the Congress, it seems to me.
Mr STAATS It is, of course, more difficult in these types of piograms
administered by HEW. We are faced in the General Accounting
Office with a very tough problem at the moment growing out of the
recently en'tcted povei ty `iuthoriz'ition, the `iuthoriz'ttion for the Office
of Economic Opportunity, where w e h'tve been given the t'tsk of
evaluating the effectiveness of oui programs administered undei that
act
This is extremely difficult in such progr'tms `is He'td St'irt, the
Neighborhood Youth Corps, `ind the community `iction progr'ims
Chairm'tn PRox~iniE But the ver~ f'tct you `ii e ±orced to think h'n d
about the benefits of the program, you are forced to determine where
you can get the most for your dollar, even in an antipoverty program,
it seems to me, is a good, wholesome requirement. It means that you are
going to invest that money more carefully.' It means you are going to
find ways of economizing, it means you can close down the less efficient
operations th'it h'ive'i lower payoff
Recently the President announced, I guess just ovei the w eekend, the
closing down of some of the Job Corps c'tmps One of them w'is in my
St'ite, `ind I regretted it very, very much
But I must say they `ipplied criteri'i w hich h'id some b'isis of objec-
tivity, `ind they had `i good method foi determining this I m'iy have
disagreed with his decision, but it w `is done on `ib'isis, `is `innounced,
and without political pressure, and this is most. helpful in the economy
to'a truer sense, that is geting more for your dollar.
Mi STA~TS As the budget becomes tighter `ind the people becomt
more and more concerned about the grow th of Government expendi
tures, it seems to me it only m'ikes good, commonsense to `ipply criteri'i
where we `ire making an effort to relate the p tyoff to the investment
that we `ire making
Chairman PROXMIRE May I `isk ~ ou `ind Mr R'ithbun `tnd `ilso Mr
Marvin, if you would like to comment, if you do not feel th'tt this
technique h'is the potentiality `ind c'tn be `ipplied more ~ idely, much
more w idely, in the future than in the p'ist, `iuid by doing so w ill gi~ e
the Congress and the administration a far better notioii of how to in-
vest these tax dollars and spend them more wisely?
Mi STAATS I w ould like to h'ive Mr R'ithbuii `insu ci
PAGENO="0031"
27
Mr. RATHBUN. I believe it is a mistake to view discounting as some-
thing applicable to only an investment project. I think it is applicable
to all projects, whether they involve investments or flow of services,
because they do involve a flow of costs and benefits over time.
Discounting is fully as useful in, say, the evaluation of some pro-
gram involving services as in and out-and-out investment program.
Chairman PROXMIRE. Thank you very much.
Do you want to comment?
Mr. STAATS. I think it would be of interest to hear what the De-
partment of Health, Education, and Welfare has to say on the-
Chairman PROXMIRE. Mr. Gardner, I notice, had some different
views and Mr. Gorham modified-
Mr. SmATS. The letter indicates that:
"You have expressed concern"-that is the GAO-"that wide dis-
parity among agency practices and rates may produce an inferior pat-
tern of resource allocation. Standard discount rates are important
when we are choosing among alternative programs with the same or
similar objectives." This is the point we made a while ago.
In these cases, proper estimation of costs and benefits could lead to a use-
ful ranking of projects to maximize the returns from available dollars. But
where the benefits of public programs are diverse or intangible, frequently af-
fecting different age groups and regions, a common discount rate, whatever the
level, may not be helpful in determining the best allocation of Government
resources.
In the present state of the art, no amount of analysis is going to reveal
whether the Nation benefits more from sending a slum child to preschool, pro-
viding medical care to an old man, or constructing a large water resources
project. Currently the decisions of how much health, education, and so forth,
and which groups in the population shall benefit, are largely based on value
judgments and politics. With respect to the basis of discount rates, we believe
they should reflect the opportunity costs of the funds utilized by the project.
It seems equally clear that there is a wide difference of professional opinion--
reflected in the results of your survey-as to the appropriate rate.
This concludes the substance of the letter.
Chairman PRox~1IRE. I thought it was a very intelligent observa-
tion, and I think, of course, he is right. Value judgments have to be the
main basis for these things, and I think the example he gives is a good
one. Once again, however, I do not understand why this is not a help-
ful guide under all circumstances.
I understand that the use of the discount technique in evaluating
education programs shows education to be a fine investment, excellent
investment, and that it is something that you can use quite well.
Mr. STAATS. I think the caution needs to be-
Chairman PRox1~rTRE. There is, I think, a feeling of hesitance on the
l)ar~ of people because they think it is a little rough and unthinking
and unfeeling and mechanical, and that your human resources pro-
grams are going to suffer. But I think they could not be more wrong.
1 think that these are the programs where there is a real payoff and
where you are going to get encouragement if they are good programs.
Once again it means you are going to evaluate these programs with
a hard head, too, mid save money where there is waste, and if you can
expose waste and you can keep constantly asking, insisting, that the
programs justify itself, and show that there is a payoff.
This is true even in the programs like Head Start that we all feel in
0tH~ heart, everybody wants to help every little child, but even some
of these programs may be wasteful. To the extent that they are, we
ought to correct them and improve them.
PAGENO="0032"
28
I am sorry I have gone over my time, Senitor Symmgton I ~ou1d
like to yield to you
Senator SYMINGTON I would ask one question, Mr Chairm'rn When
you take a water project where you have flood control, reclamation, and
power-I think those would normally be the three main interests-
from the standpoint of what you might call good business manage-
ment under sound accounting principles, you would be willing to have
a. closer rate normally for flood control than for power creation, would
you not9
Mr STAATS This is one of the kinds of cases that I had in imnd when
I referred to the need for some flexibility
Sen'ttor SYMINGTON If you have `t flood, theie is dis~ster, `tud gen
erally there is no w `ty that c'tn be prevented by private enterprise
investment.
Leaving reclamation and its intriguing appeal from the standpoint
of obtaining jobs where companies want to move and have happy em-
ployees, going from flood control over to power, you often have dis-
agreement with private power interests as to whether it is to the bene-
fit of the community and the country to put in any public power,
do you not9
Mr STAAIS That is true
I w ould still feel th'tt an analysis in connection w ith flood control,
though, is `L highly useful `inalysis in terms of both location of the
project `nid the extent to which you c'tn de'd w ith the problem through
upstre'Lm me'isures `is `tg'unst the l'trge d'ims w hich w e h't\ e gotten so
used to
Sen'ttoi SYMINGTON I `igree I only ine'int y on w ould be w illing to
put `t lesser cost r'itio `ind percenhge estnn'ite on `iflood contiol proj
ect, in the development of the over'ill project, th'in you would re the
cost of power development in the p'irticul'u thrn in question
I v~ `is not disputing, in `iny w'iy, the necessit~ for flood control I w'is
in the executive br'inch w hen w e tried to get `idequ'ite flood p1 otectioii
for Kans'is City, w hich piotection w ould h'tve cost `ibout $200 million
In the 1950-51 flood `iround K'ins'ts City w e lost $1 billion So we lost
five times moie in 1 ye'tr, in that he'tvy flood, th'in the tot'tl cost of
flood protection w ould h'tve been
So I w `is trying to figure if it w is not logic'ii to hi~ e t closer return
stim'tte cost ratio-lower would be bettei th'in closer-on the flood
control `ispect of `iw `itei development th'tn on `t power de\ eloprnent
included `is `t component p'ut of the s'tme w `iter de\ elopment
Mr ST~ ~rs I would `insw ei th'it in tw o w `tys One is I think th'tt
you might h'tve `t b'isis foi diffeienh'itmg on the use of discount foi
`ipow er project `is `ig'unst `t nonpow ei pio~ect where you do h'n e op
portumties for priv'tte investment is w eli `is public m~ estment
But I would think `iso th'it while we `tie t'ilking `thout discount
r'ites `tnd cost effectiveness studies, it is impoi t mt th'it w e `iso m'ike
`mpropei estiin'tte of benefits
My own peison'tl experience w ith the K'mns'is City flood situ'ition
was th'mt there h'id not been tdequ'tte in'iiysis Qfl en to the benefits,
01 th'tt project would h'mve been built e'uiiei
There `ire m'tny `ti guments in~ oh ed, including thc clisloc ttion of
busmesses `mci f'ums, `ind so foi th, in the ii e'm, md w h'ttnot But it
w'ts not `in `idequ'ite `issessment of the benefits in tim it p'titicui'ii c'tse,
in my opinion
PAGENO="0033"
2~
Senator SYMINGTON. At that time I was handling the relief in ques-
tion in the executive branch.
Mr. STAATS. I would like to say one thing further, Mr. Chairman,
mostly by way of emphasis. I think cost-effectiveness studies, including
the discount rate, are most useful when we are talking about a com-
mon type program. It is not too useful in comparing, say, guided
missiles with job training. They are most useful when we are talking
about ways in which you can accomplish a given object in the least
cost terms, and while we have talked a great deal here this morning
about flood control and water development, one of the things which
has not been given adequate attention in the area, in my opinion, has
been the possibility of flood-plain zoning as one of the ways in which
you can achieve the same objectives at much lower costs, and this is
because we have not-
Senator SYMINGTON. What was that?
Mr. STAATS. Zoning in the flood plain where you have an established,
known flood area, which you can zone to use for different purpOses
than otherwise it might be used.
Maybe it can be used for a park or maybe it can be used for one
purpose or another that does not suffer severe damage from floods.
Now, this is a subject which is getting a~ lot more attention now as
a result of the executive branch report put out about a year ago.
In many cases it would be much cheaper to control the situation be-
fore it develops, in effect, rather than having to build expensive struc-
tures to protect them after they are once built up.
Now, this is better for the future situation, and it does not really
solve the problem if you already have that development in place.
Chairman PR0XMIRE. Senator Jordan?
Senator PJORDAN. Just one more question in defense of flexibility.
The Department of Defense has many varied activities. We men-
tioned a good deal about flood control and navigation, even improve-
inent of water quality, a proper function of some water project, and
these are benefits and costs to which you apply one set of criteria. But
how do we evaluate those prioritywise as against an expenditure of the
same Department of Defense which it will make for a missile or a silo
which we hope we will never have to use
Mr. STAATS. Well, I do not believe that any of the techniques of cost-
effectiveness is going to give you an answer to that question, in my
opinion.
I do believe that it is important if you are considering the anti-
ballistic missle program that you have a careful analysis made of any
other way in which you could accomplish the same objective. I think
that is what we are really saying here.
Senator JORDAN. I only bring this out in defense of your recom-
inendation of flexibility, with which I fully agree. I do not think you
could establish one overall rate that would be applicable even in one
department of Government, the T)efense T)epartrnent.
Mr. STAATS. I think there can be some statement of principles, and
I think there can be far greater consistency than we now have.
Senator JORDAN. Yes. Thank you.
Chairman PROXMIRE. Thank you.
I have just one other question, and that is whether or not you have
any indication of how many programs are calculated at more than just
one discount rate?
89-654 0 - 68 - 5
PAGENO="0034"
30
Mr. STAATS. Some of the agencies indicated to us that they use more
than one discount rate in a given program, to test the effect of different
rates of discounts.
Chairman PROXMIRE. When they make a presentation to Congress,
however, they come down with one, or do they take the one that helps
them the most ~
Mr. STAATS. It might be useful; but we do not have this information.
Chairman PROXMIRE Well, it is obvious that there is a tremendous
need for development and improvement in the executive branch with
respect to evaluation of programs, and also `i great need to develop
them so that Congress can h'ive `t better understanding of these tech
niques and uses so we can evaluate the proposed spending programs
better than we do.
It is my intention to pursue this matter during this yeai with agency
heads for the purpose of i~chieving a greater effectiveness in public
expenditures, and I think, as you suggested or implied, it would be `i
good ide'L for us to htve he'~rings and have these agency he'~ds come
up at `i later time
We might be tied up in this committee for several months now with
a very heavy schedule but later on in the year we hope to pursue this
further because I think that this is `~ most constructive report which
gives promise, I think of saving enormous "tmounts of money in the
future, plus providing `t much better, sharper kind of investment
program
Senator SYMINGTON Mr Chairman, may I ask one question ~
Chairman PROXMIRE Yes, indeed I am happy to yield to Senatoi
Symington
Sen'ttor SYMINGTON Mr Comptroller Gener'tl, I noticed in youi
ieport th'tt the average `tnnuql return rate in the private sector of com
panies regulated by Government-State or Feder'il-h~s heavy varia
tion, from around 4 percent up to 15 percent
The other day `t public utility executive ~ i ote me protesting `tbout
Government regulation, or lack of regulation, of the television indus
try He said television was `dso `~ Government utility, that the chan
nels-wavelengths-were owned by the people through the Govern-
ment, and licenses were renewed for each station periodically by the
Government. But he said there was no supervision of profit as against
investment in the television industry, `md that there was growing re
sentment `tmong othei regulated utilities `mbout the condition in ques
tion
Wts his comment accurate about i egulation ~
Mr STAATS I do not believe I could answer that question
Senator SYMINOT0N. I would like to answer my constituent.
Mr. STAATS. I have not personally heard the question framed in
those terms
Senator SYMINGTON. There is an aspect to Comsat, private and pub-
lic, a semi Government agency I presume `dl c'mble income returns `tre
regulated
I did not have any answers, and wondered if you did
Mr STAATS I want to look it up
PAGENO="0035"
31
Chairman PRoxMnu~. I want to thank you very, very much, Mr.
Staats, for your usual extremely competent testimony that you and the
members of your staff have given. It was a most constructive morning.
The committee will stand adjourned, subject to subsequent hearings
we will have during the year.
Mr. STAATS. Thank you very much.
(Thereupon, at 12noon, the subcommittee adjourned.)
[The General Accounting Office report to the Joint Economic
Committee, on which this hearing was based, follows:]
PAGENO="0036"
PAGENO="0037"
Survey Of Use By Federal
Agencies Of The Discounting
Technique In Evaluating
F utu re Progra m 5
JAN.29, 1968
2
REPORT TO
JOINT ECONOMIC COMMITTEE
CONGRESS OF THE UNITED STATES
BY THE COMPTROLLER GENERAL
UNITED STA TES
OF THE
(33)
PAGENO="0038"
34
COMPTROLLER GENERAL OF THE UNITED STATES
WASHINGTON DC 20548
B- 162719
Dear Mr. Chairman:
In accordance with your request of December 19, 1967, we are
submitting our report on the results of our survey of use by Federal
agencies of the discounting technique inevaluating. future programs.
Discounting is atechnique which reduces thedollar value of future
program costs and benefits by a compounded rate reflecting the cost of
money. The discounting of future costs and benefits makes them com-
parable to present costs and benefits, i.e., comparable in terms of their
present value.
In October 1967 we sent a list of questions concerning discounting
practices to the heads of 23 Federal agencies. The agencies' replies to
these questions are summarized in appendixes of this report. On the
basis of the responses of the agencies, the following points are clear:
Although some Federal agencies do not use and do not plan
*to use discounting, the great majority of agencies view dis-
counting as an increasingly important aid to decisionmaking.
Whether based on Treasury borrowing costs or other con-
siderations, discount rates used in evaluating programs
vary over an extremely wide range (3 to 12 percent).
It is possible that the wide disparity in agency practices and dis-
count rates may produce a more serious misallocation of resources
than would exist in the absence of discounting In our opinion the gen-
eral acceptance of the technique of discounting by Federal agencies
should be supplemented with improvements necessary to bring about
consistency in and among agencies in discounting rates, techniques, and
underlying concepts.
PAGENO="0039"
35
B-162719
With respect to the discount rate, one school of thought holds that
the rate should be determined by and be equal to the rate paid by the
Department of the Treasury for borrowed money. Another school of
thought holds that the rate should be determined by what is foregone,
namely, the return that could have been earned in the private sector of
the economy when the decision is made to commit resources to the pub-
lic sector. Our survey revealed that some Federal agencies rely on
their own interpretation of these schools of thought while others employ
different criteria, including the Federal Reserve Board rediscount rate
and agency borrowing cost.
The results of the survey, as shown in appendixes I and II,
have been sent to the agencies for their consideration, A substantial~
amount of further study of specific problems and discussion of the prob-
lems with the agencies will be necessary if general guidelines estab-
lishing a standardized basis for discounting are to be promulgated.
This report also points out that, if the full costs of borrowing, in-
cluding foregone taxes from the private sector, are considered, the dif~
ference between the various schools of thought is narrowed substantially.
It therefore appears that there is a possibility of a satisfactory recon-
ciliation of varying points of view regarding the rate or rates to be used,
The report emphasizes that the important matter disclosed by our study
is the Inconsistency of present practices.
Because of the extreme variation in discount rates and techniques
being used by the executive agencies for evaluating and justifying their
programs and because there is strong impetus toward the use of the
discounting technique by Federal agency adoption of planning-
programming-budgeting systems, the Congress may wish to provide
guidance to the executive agencies on this important topic.
We have prepared this report for the use of the Congress because
of our belief that some measure of standardization regarding the use of
-2-
PAGENO="0040"
36
B.- 162719
the discounting technique is needed to enhance the quality and consiS-
tency of information concerning Federal programs which is presented
to the Congress for its consideration.
Sincerely yours,
Comptroller General
of the United States
The Honorable William Proxmlre, Chairman
Joint EcOnomic. Committee
Congress of the United States
- 3-
PAGENO="0041"
37
C 0 n t e n t~
INTRODUCTION 1
BACKGROUND 2
Explicit and implicit discount rates 3
Effect of changes in discount rates on benef it-
cost ratios 5
Determination of discount rate 7
Discount rate for planning water and related land
resources projects 8
RESULTS OF SURVEY 10
Summary of agency practices 10
Divergent agency views on discount rates 11
Significance of differing practices 13
Conclusions 15
Matter for consideration by the Congress 16
Appendix
APPENDI~S
Discount rates used by Federal agencies
in the analysis of individual pro-
grams in fiscal year 1969 I 19
Federal agencies not using discounting
in the analysis of individual pro-
grams in fiscal year 1969 II 24
Illustration of calculation of total
cost to the Government of borrowed
funds III 25
Illustrative copy of a request for in-
formation concerning discounting sent
to selected agencies IV 29
* Letter dated December 19, 1967, to the
Comptroller General of the United
States from the Chairman of the Joint
Economic Committee V 32
PAGENO="0042"
38
INTRODUCTION
The General Accounting Office has made a survey of the
us~by 23 selected Federal agencies of the technique of dis-
counting in making evaluations of future programs. The
nature of the discounting technique is described in the
background section of this report.
Our survey was made pursuant to the Budget and Account-
ing Act, 1921 (31 U.S.C. 53), and the Accounting and Audit-
ing Act of 1950 (31 U.S.C. 67). The survey was performed
in Washington, D.C., and was completed in December1967.
We undertook this survey because of our belief that the
rapid growth in program expenditures by the Government and
the increasing complexity of Federal programs point up the
need for responsible officials in~ the Government to make
effectiye use of objective aids to decisionmaking. This
report discusses one such aid--the technique of discount-
ing--and describes its use by the Federal agencies included
in our survey.
Our survey was directed primarily toward determining
the extent to which the discounting technique is presently
employed by Federal agencies and the extent to which those
agencies not employing this technique plan to do so and to-
ward identifying the discount rates currently in useby the
agencies. We did not evaluate the appropriateness of the
agencies' stated policies and practices as they were re-
vealed to us, neither did we examine into the actual dis-
counting techniques and practices followed by the agencies
or into other policies and practices used by them when
evaluating Federal programs.
An illustrative copy of a request for information con-
cerning discounting sent to each of the selected agencies
is attached as appendix IV. Our request for information
and our survey did not cover information concerning water
and related land resources programs for which recognition
of uniform agency practice is contained in Senate Docu-
ment 97 (see discussion on p. 8) or programs that come
within the purview of Bureau of the Budget Circular A-76
entitled "Policies for Acquiring Commercial or Industrial
Products and Services for Government Use."
1
PAGENO="0043"
39
BACKGROUND
The theory underlying the discounting technique is
that benefits from Federal programs to be realized in the
near future are valued more highly than benefits to be
realized in the more distant future and that costs which
must be incurred in the near future loom larger than costs
that will be incurred in the more distant future. The dis-
counting of future benefits and costs makes them comparable
to present benefits and costs, i.e., to the present values
of benefits and costs. The numerical standard used in mak-
ing these intertemporal comparisons is called the discount
rate.
Federal agency programs, like other programs, gener-
ally involve a series of annual costs and a flow of bene-
fits over time. Calculation of the present values of costs
and benefits through discounting makes possible a compar-*
ison of costs and benefits, usually expressed in terms of a
ratio of benefits to costs, which gives consideration to
the time periods in which benefits will be realized and
costs incurred. Such comparisons (or ratios) are useful in
evaluating programs and in choosing between alternative pro-
grams. Furthermore, the discounting technique can be help-
ful to the decisionmaker in those cases in which the bene-
fits associated with programs cannot be measured in dollars.
Here, the present values of the costs of the programs can
be presented to the decisionmaker for his decision as to
whether the perhaps dimly perceived benefits are worth their
costs. The technique can also be helpful in making compar-
isons of the costs of programs that have equal benefits.
The present values of program benefits and costs will
not usually be the only information decisionmakers need con-
cerning benefits and costs. For example, there may often
be value in tables which show the time-phased undiscounted
costs and benefits in a manner that enables the decision-
makers to see the full impact of benefits and costs in each
fiscal year.
2
PAGENO="0044"
40
EXPLICIT AND IMPLICIT DISCOUNT RATES
Discount rates are often explicit but in some cases
th'ey may be implicit. An explicit discount rate is a rate
which is identified and used to calculate the present val-
ues of future benefits and costs. However, if in an analy-
sis the assumed life of the~ program is different from the
more probable life expected on the basis of experience or
studies, then an implicit discount rate is actually being
used.
The meaning of such an implicit discount rate can be
illustrated by the following example in which a 10-year
program life is used in an evaluation study but the most
probable program life is 20 years. In order for the evalu-
ation study based on 10 years to lead to the same conclu-
sion as one based on 20 years, the present values must be
equal in both cases. V
In this example the net annual program benefits will V
continue beyond the 10-year program life assumed in the V
study. Since the net annual benefits in the eleventh year
through the twentieth year are not recognized in the 10-
year study art implicit discount rate is being used... In
column A of the following table the total undiscounted val-
ues of program costs and benefits are shown for the assumed
program life of 10 years. In column B the program costs
and benefits for the most probable program life of 20 years
are shown discounted at 8 percent--the discount rate that V
*is required to equate the benefit-cost ratios in columns A
and B. The implicit discount rate mV this analysis which V
uses an assumed program life of 10 years (column A) is thus
V 8 percent.
Present~~a1~AgS
(~) V
(A) Most probable
Undiscounted Assumed V program
amounts 10-year Iif~ life--20 VyVear~
(millions) V
Implicit discount rate 8.0~
Initial investment $10 $10.0 $10.0
Annual operating costs 1 10.0 9.82
Annual benefits 3 30.0 29.45
Ratio of present value of benefits to
present value of total costs (benef it-
costrstio), rounded 1.5 1.5
3
PAGENO="0045"
41
Shown in the table below are the approximate discount
rates, for a range of the most probable program lives, that
would be implied by analyses in which explicit discounting
is not used but in which the assumed program lives are
shorter than the most probable program lives. The implicit
discount rates in an actual case may be somewhat different,
depending on theform of the cost or benefit streams over
time.
Implicit Discount Rates in Percent
Most probable
program life -
(years)
3 ~-
4 2.
Benefit-cost ratios may be very sensitive to the ex-
plicit discount rate used in calculating the present value
of costs and benefits. As shown in the following example,
which involves a program with a probable life of 25 years
that requires an initial investment of $50 million, changes
in the rate can turn what would appear to be an attractive
program into an unattractive program and vice versa.
Assumed Dro2ram~ life
1
5
10
15
20
25
(years)
15 20
100
100
100
100
100
25
5 10
15 -
18 6
19 8
20 9
Ratios of benefits to cost
Initial Annual
invest- operating Annual Undis-
merit costs benefits counted
(millions)
$50 $12 $16 1.14
Dis-
counted
at
3 percent
*Dis-
counted
at
7 percent
0.98
1.08
PAGENO="0046"
42
EFFECT OF CHANGES IN DISCOUNT
RATES ON BENEFIT-COST RATIOS
The chart on page 6 shows the inverse relationship be-
tween discount rates (shown on the horizontal axis) and
benefit-cost ratios (shown on the vertical axis) for three
different programs which have constant annual benefits and
constant annual operating costs. The three cUrves illus-
trate the effect of discounting on the benefit-cost ratios
of the three hypothetical programs with differing lives,
each of which, if undiscounted, would reflect a benef it-
cost ratio of 2; that is, total undiscounted benefits for
each program would be double the total undiscounted costs.
With discounting, the favorable benefit-cost ratios deteri-
orate as higher discount rates are applied. For example,
Program A has a benefit-cost ratio of about 1.7 at 3 per-
cent, but only about 1.1 at 10 percent.
The curves drawn for these three hypothetical programs
are probably typical of such curves for many programs; how-
ever, the curve for any particular program may be different
because of the behavior of the program's benefits and
costs. Such behavior would depend in part upon the assumed
life of the* program and the amount of initial investment
cost relative to annual benefits and annual costs
The chart on page 6 also illustrates that changes in
the higher discount rates have less relative effect. on
benefit-cost ratios than do changes in the lower discount
rates. Therefore, a greater absolute error or wider range
of uncertainty or variation may be tolerated in the higher
discount rates than in the lower discount rates. For exam-
ple, an increase in the discount rate from 3 to 6 percent
reduces the program B benefit-cost ratio by about 23 per-
cent; whereas, an increase from 10 to 13 percent reduces the
benefit-cost ratio by about 17 percent.
5.
PAGENO="0047"
0
I
43
11 1~ ~3
DISCOUP4T RATE
PAGENO="0048"
44
DETERMINATION OF DISCOUNT RATE
With respect to determination of the rate, one school
of thought holds that the rate should be determined by and
be equal to the rate paid by the Treasury in borrowing
money. Another school of thought holds that the rate
should be determined by what is foregone, namely, the re-
turn that could have been earned in the private sector of
the economy when the decision is made to commit resources
to the public sector. Proponents of both views agree that
the use of different discount rates has an effect On finan-
cial judgments which is similar to charging different
prices for, say, identical labor in different programs.
Neither school of thought provides clear guidance on
the specific discount rate that should be used. Cost to
the Treasury, for example, will vary, depending upon the
definition applied, from 3 to 8 percent or more. The aver-
age rate of return in the private sector also varies de-
pending upon historical periods selected and upon the
weighting of the various segments of the private sector
which are used in computing an average.
A discount rate of about 3.2 percent is the cost to
the Treasury, if based upon the average rate payable on
outstanding United States securities having a maturity of
15 years or more, as prescribed by Senate Document 97 (dis-
:cussed below).
The rate of return that can be earned in the private
sector is estimated by most researchers to be significantly
higher than the cost of Government borrowing based on Sen-
ate Document 97. According to one researcher, the average
rate of return in the private sector amounted to 15.4 per-
cent for manufacturing comp~nies and 4.1 percent for rail-
roads in the years 1961-65. These and other estimates of
the same researcher are shown in the following table.
1Prepared statement of Jacob A. Stockfisch, Hearings before
the Senate Subcommittee on Economy in Government of the
Joint Economic Committee, Ninetieth Congress, on the
Planning-Programming-Budgeting System: Progress and
Potentials.
7
PAGENO="0049"
45
Annual
average rate
of return
1961-65
Segment (percent)
Manufacturing 15.4
Electric utilities 9.3
Gas pipelines 8.6
Telephone 11.9
Railroads 4.1
Motor carriers (note a) 14.7
Oil pipelines 15.6
Airlines 8.2
aAverage for 1961-64 only.
Therefore, if Senate Document 97 is used as the basis
for the cost of Government borrowing, the difference be-
tween the two schools of thought is quite significant. If,
on the other hand, Treasury borrowing costs are calculated
on the basis of total costs to the Government, including
corporate and individual income taxes foregone as a result
of borrowing by the Government to finance programs, an es-
timate of between 7 and 8 percent results. (See illustra-
tive calculation in app. III.) Thus, if Government costs
are calculated on this basis, the practical importance of
the difference between the two schools of thought is
greatly reduced.
DISCOUNT RATE FOR PLANNING WATER
AND RELATED LAND RESOURCES PROJECTS
Senate Document 97 entitled "Policies, Standards, and
Procedures in the Formulation, Evaluation, and Review of
Plans for Use and Development of Water and Related Land Re-
sources" (87th Cong., 2d sess.), which provides congres-
sional guidance on discount rates under certain circum-
stances, states that the discount rates "shall be based
upon the average rate of interest payable by the Treasury
on interest-bearing marketable securities of the United.'
States outstanding at the end of the fiscal year preceding
such computation, which upon original issue, had terms to
maturity of 15 years or more."
8
PAGENO="0050"
46
The document also states that "This procedure shall be
subject to adjustment when and if this is found desirable
as a result of continuing analysis of all factors pertinent
to selection of a discount rate for these purposes."
Senate Document 97 was developed by the Secretary of
the Army; the Secretary of Agriculture; the Secretary of
Health, Education, and Welfare; and the Secretary of the
Interior, and it was approved by President Kennedy on
May 15, 1962. As indicated in the title, the document re-
* lates only to water and related land resources.
9
PAGENO="0051"
47
RESULTS OF SURVEY
SUMMARY OF AGENCY PRACTICES
In our survey of the use of the discounting technique,
selected agencies were asked to respond to a questionnaire
concerning the discount rates used, rationale for the rates
chosen, and plans for future use by agencies which were not
evaluating fiscal year 1969 programs on the basis of dis-
counting.
The table below shows the agencies' plans for using the
discounting technique.
/~gency Plans for Using the Discountinm secn,sq~
Discounting used in Discounting not used
analysis of fiscal but plans are to use
year 1969 programs discounting in future
Yes No Yes No
Tennessee Valley Authority X
General Services Administration X
Department of Agriculture X
Department of Defense X
Office of Economic Opportunity X
Department of Transportation (Federal Aviation
Administration) X
Department of Health, Education, and Welfare X
Atomic Energy Commission X
Agency for International Development X
Department of the Interior X
Federal Communications Commission X X
Department of Housing and Urban Development X X
Interstate Commerce Commission X X
National Science Foundation X X
Federal Power Commission X X
Export-Import Bank of Washington X X
Department of Labor X X
Peace Corps X x
Veterans Administration X X
Department of the Treasury X X
Post Office Department X X
National Aeronautics and Space Administradon X X
Department of Commerce X X
10
PAGENO="0052"
48
As shown in the table, 10 of the 23 agencies queried re-
port that discounting is used in making decisions. An ad-
difional eight agencies do not now use discounting but re-
port that they plan to do so in the future. The remaining
five agencies do not use discounting at the present time
and do not state that they plan to do so in the future.
However, two--Department of Commerce and Veterans Admini s-
tration-- of these five agencies reported that some~pro-
grams are evaluated in terms of periods of time which are
shorter than the probable actual life of the programs--a
procedure which involves implicit discounting. (See dis-
cussion of implicit discounting which begins on p. 3.)
Details of rates used by the agencies which use the
discounting technique appear in appendix I.
Explanations of the plans of agencies which do not use
the discounting technique appear in appendix II.
DIVERGENT AGENCY VIEWS ON DISCOUNT RATES
A divergence of opinion on discounting is reflected in
agency practices. Some agencies use the Treasury cost of
borrowing money as the discount rate while others use a
rate based on the return on investment in the private sec-
tor of the economy. Still others.employ different criteria
to determine the agency's discount rate including the Fed-
eral Reserve rediscount rate and agency borrowing cost. The
agencies included in our survey use discount rates which
vary over an extremely wide range--from about 3 percent to
12- percent.
Within each school of thought there are important dif-
ferences of opinion. Of those agencies which tie the dis-
count rate to Treasury borrowing costs, one uses theesti-
mated cost of new money to the Treasury, another uses the
average cost of money to the Treasury, others use the cost
prescribed by Senate Document 97 (about 3.2 percent at the
time of our review).
One of those agencies which would tie the discount rate
to the rate of return in the private sector uses a rate of
11
PAGENO="0053"
49
return on a safe investment and a slightly higher rate
(3 percent and 5 percent) while another uses for some pro-
grams the rate representative of average capital returns
in the private sector (presumed to be 12 percent) and for
some programs permits the analyst to determine the rate on
the basis of his judgment as to the nature of the program
and the kind of analysis considered most meaningful.
Views are equally disparate in those agencies which
report that discounting was not used in their analyses of
individual programs for support of their fiscal year 1969
budget decisions. At one extreme is the view that deci-
sions on programs should be made on the basis of first year
costs and benefits--a procedure which implies a discount
rate of 100 percent, since the future is ignored. At the
other extreme is the view that decisions should be based on
total undiscounted costs and benefits--a procedure w~uch
implies a discount rate of zero, since costs and benefits
applicable to, say, the 2Ot-i year are treated as be~'g as
important as current costs and benefits.
PAGENO="0054"
50
SIGNIFICANCE OF DIFFERING PRACTICER
The fact that some agencies use discounting and some
do not will tend to affect adversely the quality of deci-
sions. This is illustrated in the following (hypothetical)
cases.
Case A involves a program in an agency that ~ use
discounting. Estimated costs and benefits of this program
over its 25-year life are:
Initial investment costs
Annual operating costs
Annual benefits
Present value of total costs
discounted at 4.5 percent
Present value of total bene-
fits discounted at 4.5 per-
cent
Benef it-cost ratio
Initial investment costs
Annual operating costs
Annual benefits
Total costs
Tothl benefits
Benefit-cost ratio
On the basis of the benefit-cost ratio, the net value
of program A is marginal. To the extent that benefit-cost
ratios are considered significant by the decisionmakers,
there would be a tendency to prefer program B rather than
program A because total undiscounted benefits exceed total
costs by a relatively wide margin, even though if program B
were subjected to the rate of discount (4.5 percent) used
for program A it would be much less attractive than program A
(benefit-cost ratio of 0.97 compared with 1.07). Thus, the
benefit-cost analyses for programs A and B would, if only
the program A analysis used discounting, tend to bring about
* $12,000,000
2,000,000
3,000,000
41,656,000
44,484,000
1.07
Case B involves a program in a different agency which
does not use discounting. Estimated costs and benefits of
this program over its 25-year life are:
$16,000,000
1,500,000
2,500,000
53,500,000
62,500,000
1.17
13
PAGENO="0055"
51
a misallocation of resources since the results of the anal-
yses would favor program B.
In the above example, the tendency to misallocate
will, of course, increase with higher discount rates. The
following table illustrates this tendency as the discount
rate increases from zero to 10 percent.
Pro- Pro- Differ-
Benef it-cost ratio gram A gram B ~
Undiscounted 1.21 1.17 .04
Discounted at 4-1/2 per-
cent 1.07 .97 .10
Discounted at 10 percent .90 .77 .13
Differences in discounting practices (see app. I) in
and among the agencies which do use discounting will tend
to produce the same kind of misallocation. For example, a
power-production program in the Tennessee Valley Authority
(TVA) (life of 50 years) may have a benefit-cost ratio of
1.14 (if discounted at 4.5 percent), while in the Depart-
ment of the Interior the same flow of costs and benefits
would yield a ratio of 0.96 (if discounted at 6 percent) or
only 0.58 if the project were felt to be risky (and for
this reason discounted at 12 percent). Thus, even though
both agencies use discounting, the fact that different
rates are used tends to favor programs of the agency using
the lower discount rate and thus a misallocation of re-
sources could be the result.
Similarly, interagency differences in implicit dis-
count rates in~ agencies that do not use explicit discount-
ing (see app. II) tend to produce~ misallocation. As shown
in the table on page 4, the implicit discount rate in-
creases as the difference between the most probable
(longer) actual life of a program and the period of time
considered in evaluating the program increases. The Veter-
ans Administration (VA), for example, evaluates programs on
the basis of a 5-year period, while the most probable life
of, say, a hospital is about 25 years. Thus, the implicit
discount rate is about 20 percent. The Department of Com-
merce, on the other hand, evaluates some programs having a
1L4
PAGENO="0056"
52
most probable life of at least 11 years in terms of a 7-year
period--a procedure which implies a discount rate of about
8 percent. Under these conditions, the benefit-cost ratio
for a given stream of benefits and costs will be higher in
the Department of Commerce than in the VA. To the extent
decisions turn on the relationship between costs and bene-
fits, the VA will be inclined to reject projects which
would be promoted under the standards employed in the De-
partment of Commerce. Hence, resources will tend to flow
from VA projects to Department of Commerce projects.
CONCLUSIONS
Federal agencies that use the explicit discounting
technique may be making good use of this tool in evaluating
individual projects. The analyst who examines a given
project in detail, develops the discount rate which he con-
siders appropriate, and calculates the present value of
benefits and costs is in a better position to make defen-
sible recommendations than if his analysis ignored the time
periods over which benefits will be realized and costs in-
curred.
In our opinion, however, there is a greater value for
the discounting technique when the decisionmaker must
choose between many competing projects and this calls for a
common standard with justification for variations in the
discount rate which may be appropriate in special circum-
stances. Although some agencies indicate that the discount
rate is viewed as an aid in choosing between programs
within an agency, there appears to be little recognition
that the evaluation of Federal programs calls for a common
yardstick for use by all agencies.
In our opinion, the general acceptance of the technique
of discounting by Federal agencies should be supplemented
with improvements necessary to bring about consistency in
and among agencies in discounting rates, techniques, and
underlying concepts. We believe such improvements are
needed if this aid is to be of most effective use to the
agencies, the Bureau of the Budget, and the Congress in its
evaluation of agency programs submitted for consideration.
15
PAGENO="0057"
53
Obstacles to reform in this important area are many
and varied. For example, the specific nature of decisions
that must be made needs to be examined into in order that
criteria for evaluating discounting techniques, discount
rates, and underlying concepts can meet the needs of the
various decisionrnakers. It is clear to us that some mea-
sure of standardization is needed to enhance the quality
and consistency of information presented to the various de-
cisionmakers for their consideration.
MATTER FOR CONSIDERATION BY TUE CONGRESS
Both the case for discounting and the choice of dis-
count rates have been s~ibj ects of dispute for years. How-
ever, the results of our survey of Federal agency practices
suggest that the case for discounting is being accepted but
that there is a significant difference of opinion among the
agencies over discount rates. Because of the extreme vari-
ation in discount rates and techniques being used by the
executive agencies for evaluating and justifying their pro-
grams and because there is strong impetus toward the use of
the discounting technique provided by Federal agency adop-
tion of planning-programming-budgeting systems, the Congress
may wish to provide guidance to the executive agencies on
this important topic.
16
PAGENO="0058"
54
APPENDIXES
APPENDIX I
Page 1
DISCOUNT DATES USED DY FEDERAL AGENCIES SN TUE
ANALYSIS OF INDIVIDUAL FRODRAIIS IN FISCAL YEAR 1969
Rationale
Treaaury
borrowing
cost Other
Fertili net noun itinno and
donelopoect
Poaur supply nod use
Fsoiiitivc progrea.
4.5 te 5.5 Coat of nanny to TVA (cots 1)
Estimated productivity of
4.5 capital (note 2)
I (note 3)
Tennessee Valley Authority
G~aiaral Services Adoinistre-
tion
Department of Agriculture
Office of Economic Opportunity
Tranopoetaiioo
(FadeioI AviatIon Adeiniorra-
tlco)
Atomic Energy Coneniawion
}ioucltg loans 4.875
Water and sewer loans nod
grants 4.8/5
Rural elnotrificotion louao 4.875
Rural telephone loans 4.875
Rural reoetal district loans 4.875
Rural conservation and dcccl-
aptcot loans 4.8/5
Farn operating macs 4.875
Farm real aotate looms 4.875
Production efficienny 5.0
Jobcorpo 3.0 end 5,0
3.0 " 5.0
5.0
x
x
S
S
x
x
Upward brood
Faoily planning program
Facilitleo and Ionic
Radar nonpoonnta
Enrirhed uraniun pcaduction
plaooicg
Radiouttlon waste nacngnnnnt
planning
Valuw of opecial oucleor
natnoiulc
Rudiat tots pustwurizntintt
Reactor developsestt
41 shipyard projects
14 sIc stations
58 other ctaticnc
I'
x
"Rote or return on a safe in-
tentmsast and a slightly
higher tate (note 4)
Do.
Federal Reserve rediscount
Do.
Use of 7.0 on/ 10.0 cot en-
plalned
Rate sued In industry
6.0 and 7.0 typical of rota
uned by private utIlities;
9.0 uoad to tcrrodueonr lob
factor (note 5)
Tire preference for rattan
ye. futurn mosey uncrlFicra
in private eeorur (note 6)
4.2
4.2
4.2
5.0
5.0
5.0, 7.5.
and 18.0
Gavetrrsent
coors--S.D
Industry
heoefito--i5.O
5.0 to 9.0
10.0
`p.o
10.0
Departmnest af Defense
19
PAGENO="0059"
55
APPENDIX I
Page 2
DISCOUNT RATES USED BY FEDENAL AGENCIES IN THE
ANALYSIS OF INDIVIDUAL PROGRA1IS IN FISCAL YEAR 1969 (continued)
Rationale
~UYT!~
Frog~pcs
Rote
(ESYTANt)
borrooing
coot
Other
Agency Ens- Intornoticuol Do-
telopoent
Paces- piunt in Afghotintac
llighstoy is Bolivia
6.0 (cost only)
12.0
Opportunity s-ott of rnney
(tote 7)
Development on foreign to-
thongeucarc ity. oppocto.
nity costs, ond other foctur
(note H)
R6odt in Guyana
Rondo in British Guiana
10.0
8.0
Do.
Department of the interior
Utility program:
Low rink
Average
6.0
12.0
Representative returns on in-
vestreot (note 9)
Average return in private
Department of Health, Educa-
tion, and Welfare
Energy nod ninerol develop-
cent prograna io which ex-
ploitation in a private
function
Aquatic living reaourcco
Indian resercati ooreoources
development
Button inveotnent progronu
(adult education, voca-
tional rehabilitation, stork
experience)
12.0
3.1 and 6.0
3.1
0-6.0
X
Saxe (note 9)
No special explanation
(note SO)
Related to voter or land re-
sournen, no S.D. 97 applies
(note Il)
(note 12)
Individual di neanee (tubes-cu-
ionic, canter, nyphilin,
arthritta, rotor vehicle in-
jury)
Costofillnouo
Cancer control
0-10.0
4Uand6.U
4.0 ` 6.0
The noten to appendix I
appear on page 21.
(note 12)
(note 12)
(nate 12)
PAGENO="0060"
56
APPENDIX I
Page 3
NOTES TO APPENDIX I
1. The Tennessee Valley Authority (TVA) stated that its
Power Supply and Use Program rates "are based on the
expected costs of money which TVA must pay over the
period of the evaluation. Since the power program is
currently financed from earnings from the sale of power
and from the sale of revenue bonds, the future cost of
money varies with the proportion acquired from the dif-
ferent sources utilized as well as from changes in in-
terest rates."
2. The General Services Administration (GSA) stated that
in its facilities program the costs of alternatives
were discounted to present value at 4.5 percent for 50
years, the estimated life of the buildings. GSA stated
that the 4.5 percent rate was selected "as an estimate
of the long term productivity value of capital" and was
applied on the assumption that "the relationship be-
tween the costs of alternatives would hold over the
life of the project under examination."
3. The Department of Agriculture stated that its analysts
"often prefer to calculate internal rates of return for
comparing investment-type programs, rather than use a
benefit/cost ratio analysis which depends upon dis-
counting." The "internal rate of return" is analogous
to the "yields" of an investment. The procedure is to
find the "internal rate" (the rate of discount) that
equates the present value of the proceeds from an in-
vestment with the present value of the outlays on the
investment.
4. The Office of Economic Opportunity advised us: that the
rates of 3 and 5 percent "were safely on the conserva-
tive side for estimates of this type," and that they
represented a "rate of return on a safe investment and
a slightly higher rate." The agency also advised us
that these rates gave consideration to the secular
growth in the price of quality-constant labor.
2~
PAGENO="0061"
57
APPENDIX I
Page 4
5. The Atomic Energy Commission stated that in its "reac-
tor development studies, discount rates of 6 to 7 per-.
cent are illustrative of those used by investor-owned
utilities. (Companion studies employ appropriate dis-
count rates based on data from the Federal Power Corn-
missionapplicable to public power systems.)" The
5 and 9 percent rates were used to determine the sen-
sitivity of the reference value of 7 percent.
6. The Department of Defense stated that all the programs
to which it applied the-10 percent discount rate were
related to its military construction program. The dis-
count rate was selected "to reflect the amount of time
preference for current versus future money sacrifices
that the public exhibits in nongovernmental transac-
tions. The 10 percent rate is considered to be the
most representative point within a range of plausible
rates obtained from considering this public time pref-
erence."
7. The Agency for International Development stated that
the cost of its Afghanistan power plant project was
discounted at 8 percent, assumed to be the opportunity
cost of money in that country. Benefits were not dis-
counted since "it is assumed that the expected demand
for power must be met." The computations were for a
useful life of 39 years.
8. The Agency for International Development (AID) stated
that the three Latin American roads projects were dis-
counted for 20 years at the rates shown in the rate
column of appendix I. Latin American roads projects,
in general, are discounted to present value by AID at
the opportunity cost of long-term capital "usually
placed upwards of 8 percent and possibly ranging to
16 percent in some Latin American countries." AID
analysts also compute the internal rate of return of
such projects. The internal rate of return is de-
scribed in note 3.
9. The Department of the Interior stated that no uniform
or single discount rate had been applied in its major
analyses outside the water and land resource area. The
22
PAGENO="0062"
58
APPENDIX I
Page 5
selection of the discount rate generally had been left
to the judgment of the analyst. But 6 percent, after
taxes, is "considered to be representative of utility-
type program in which risk is relatively low; 12 per-
cent presumed to be representative of average capital
returns in the private sector, etc." The time frame
for analysis is varied from a long-term useful life
concept to shorter periods of 20 years or less for
programs oriented to the private sector or to short-
term objectives. In the mineral resource area, the
Department's general approach is to identify and com-
pare internal rates of return as a means for ranking
or establishing program priorities. The internal rate
of return is described in note 3.
10. The Department of the Interior stated that in its
aquatic resources programs "Alternative program levels
and mixes were subjected to benefit-cost analysis us-
ing discount rates of 3-1/8 and 6 percent against
benefit flows over 5, 10, and 15-year periods."
11. The Department of the Interior stated that, in its
Indian-reservation resources development programs,
comparative analyses of alternative programs "(irri-
gation, range development for livestock, dry-farming,
timber production and industrial development) were
based on an interest rate of 3-1/8 percent; however,
the periods of analysis necessarily varied because of
distinct program characteristics." This discount rate
was selected in accordance with Senate Document 97
since the programs were related to water and land re-
sources.
12. The Department of Health, Education, and Welfare
stated that "We feel that discounting a future stream
of dollars to present value is helpful, but we are un-
certain what rate to set. We (use) several to see
whether the difference is critical, for the specific
purpose of the study. If it does not seriously dis-
turb relative rankings we note this. If it does have
a significant effect, we wish to inform the reader of
the study of this."
23
PAGENO="0063"
APPENDIX II
59
FEDERAL AGENCIES NOT USING DISCOUNTING
IN ThE ANALYSIS OF INDIVIDUAL PROGRM~ IN FISCAL YEAR 1969
Ag~pcy comments
National Aeronautics and Space Administra-
tion
Federal Power Commission
Federal Communications Commission
Peace Corps
National Science Foundation
Department of Labor
~. ~gencies thato].an to~ise_dLacounting in future
Department of Housing and Urban Development Discounting used in appraisal of fiscal year 1968 programs,
with the rate determined by then-current Treasury borrowing
costS. Agency has a strong interest in the development of
policies for measuring the costs and benefits applicable to
Fedeisl programs.
Analyses of Federal Power Commission programs, other then
those involving water resources, are still in a preliminary
stage; decisions on appropriate discount rate must await
completion.
Detailed cost-benefit analyses will be made in the future
when additional staff capability is developed.
The evolution of planning-programming-budgeting systems in
the Peace Corps has pot reached the stage where discount
rates are applied to costs and benefite.
Analyses hawe not been so sophisticated as to require die-
counting.
Agency has not attempted to exprese in terms of percentage
discounts, the effect of future (total) costs aiad benefits
although these economic factors are implicit in considera-
tion of alternati~re programs and the assessment of priori-
ties,
Agency han considered discounting at length. Would tend to
one a projected rate of growth in Gross National Product as
repre3enting the opportunity coat of financing socio-economic
programi, with sensitivity analysis to indicate the effects
of different rateS and different time horizons. Manpower
development assistance program evaluated in terms of coot in
the first year, while the program will continue for 5 to
20 years.
Agency has used discouncing in past in its facility modern-
ization program and in its lease-purchase reports to the
Congress. Currently studying the incorporation of discount-
ing concepts aud practices on a wider scale.
Interstate Commerce Commission has no programs which lend
themselves to the use of discount rates in measuring costs
and benefits.
Unefs,l life span of programs cannot be estimated in terms of
duration based on a known or foreseeable termination point.
Mont programs, therefore, are evaluated on a 5-year projec'.
tion basis.
Programs are, in the main, continuing ones, and problems of
efficiency relate primarily to the cost side; usually alter-
native costs for the same objective. This kind of compari-
son does not require use of a discount rate to evaluate fu-
ture benefit streams. In those cases where procurement of
capital equipment is involved, the benefits are generally
large enough so that a simple payout period is all that is
necessary.
Na comment.
Agency procedures involve appraisals of programs in terms of
periods s,hich are shorter than the estimated useful lives of
the programs. The Economic Development Administration cur-
rently evaluates programs with 11 years' minimum lives in
terms of the fiscal year 1967-74 period. The Envtronmental
Science Setvices Administration evaluates 10- to 50-year
programs in terms of 5 years. The Bureau of Standards eval-
uates indefinite (continuing) programs in terms of 5 years.
Post Office Department
B. Agencies that havg np stated plans to use
discounting or had no comments
Interstate Commerce Commission
Veterans Administration
Department of the Treasury
Export-Import Bank of Washington
Department of COnsserce
214
PAGENO="0064"
60
APPENDIX III
Page 1
ILLUSTRATION OF CALCULATION OF TOTAL COST
TO THE GOVERNMENT OF BORROWED FUNDS
METHOD I
The current interest cost of borrowing long-.
term money is approximately 5 percent. The moving
average rate specified by Senate Document 97 is cur-
rently about 3.2 percent. Therefore, a rate of in-
terest approximately halfway between 5.0 and 3.2
could be used for initial consideration as the Gov-
ernment cost of borrowed money. 4.O7~
Add to this cost:
1. Corporate taxes foregone by the Government
if the average corporate return on invest-
* ment is 12 percent before taxes1, if the
fraction. of dollars borrowed by the Govern-
ment which would have gone into corporate
investment is 65 percent2, and if the mar-
ginal corporate tax rate is 40 percent.
(.12) (.65) (.4) 3.l7~
2. Personal lfaxes foregone by the Government
if the average return on proprietorship,
personal income-producing investments, etc.,
is such that the remaining 35 percent of
money borrowed by the Government would have
earned a 10-percent return for the persons
taxed, and if such return would be taxed at
a composite marginal rate of 30 percent.3
(.1) (.35) (.3) 1.07~
3. (a) Taxes foregone by the Government on
dividends that would have been received by
individuals from corporations if the corn-
posite marginal tax rate applicable to
individuals is 30 percent, if the taxable
dividends payout is 40 percent of corporate
25
PAGENO="0065"
61
APPENDIX III
Page 2
profits after taxes, and if the assump-
tions as to corporate earnings and the
marginal tax. rate shown above urn.er
(1) are applicable. The marginal corpo-
rate tax rate is assumed to be 40 percent
therefore 60 percent of corporate earn-
ings is assumed available to the corpo-
ration for payment of dividends.
(.3) (.4) (.12) (.65) (.6) = .67~
(b) Personal taxes foregone by the Govern-
ment if the corporate investment is
financed by bonds rather than by corpo-
rate earnings, if corporate bonds carry
an interest rate of 5 percent, if the
fraction of dollars borrowed by the Gov-
ernment which would have gone into cor-
porate investment is 65 percent, and if
the composite marginal tax rate applic-
able to individuals is 30 percent
( 05) ( 65) ( 3) = 1 07
(c) Actual overall financing arrangements
by corporations will generate tax reve-
nues under both (a) and (b), therefore
the cost to the Government may be as-
sumed to be somewhere between .6 per-
cent and 1.0 percent, say about .87~
Subtract from this cost:
1. Income taxes collected on Government in-
terest payments, if investment in bonds
(see rate above of 4 percent) are divided
between corporations and individuals in such
a way that the tax rate is 35 percent.
(.04) (.35) ~7
Cost to Government 7 57
26
PAGENO="0066"
62
APPENDIX III.
Page 3
METHOD II
On an aggregate basis, a similar result may be
computed assuming a composite corporate and personal
marginal tax rate of 50 percent4 and a taxable re-
turn of 10 percent on any money not borrowed by the
Government.
(.5) (.1) 5.07.
Cost of Government borrowing (see explanation
under Method I) 4.07~
Less taxes on Government bond interest (.04)
(.35) (see explanation under Method I)
Cost to Government 7.67~
1Various economists have examined rates of return before
taxes in the private sector. Stockfisch (see footnote on
page 7), arrives at an average of 13.5 percent. Stiegler,
National Bureau of Standards, determines a rate of 14 per-
cent. Variations in this estimate result from considera-
tion of differing time periods, weighting, etc. Our
estimate of 12 percent used for this appendix is somewhat
conservative in comparison with the recent experience
noted by these economists.
2See Raymond Goldsmith's "Flow of Capital Funds in the Post-
war Economy," National Bureau of Economic Research, 1965,
where a table of gross capital consumption by major seg-
ments of the economy is shown. We are interested here in
capital consumption for purposes of productive investment.
Most household borrowing can be excluded as investment in
consumption which would also result from payments by the
Government to labor involved in Government programs.
State and local capital consumption can be left out of
this consideration. The corporate share of the remainder
is approximately 65 percent.
3A table of marginal tax rates for various income levels is
contained in a study done by the Institute for Defense
Analyses for the Office of Economic Opportunity, as
27
PAGENO="0067"
63
APPENDIX III
Page 4
summarized in "R-1l6, Federal Poverty Program, Assessment
and Recommendations," January 1966. This document shows
that the average marginal rate is approximately 30 per-
Cent for the higher income levels, from which personal
income-producing investments tend to originate.
4mis is a rough composite marginal rate for corporate and
personal taxpayers that provides approximately for the
separate estimates shown in 1, 2, and 3 for Nethod L
28
PAGENO="0068"
64
COPY APPEND1 X IV
Page 1
Dear Mr. Secretary:
The General Accounting Office is making a study of the
practices followed in the major Federal agencies in measur-
ing the costs and benefits applicable to proposed Federal
programs.
The examination of alternative means of achieving an
agency's objectives is, of course, of central importance in
the allocation of resources. Typically, the alternatives
considered will differ with respect to the phasing of both
costs and benefits over time. Determination of the relative
merits of the alternatives, therefore, necessarily requires
that costs and benefits be discounted.
It is clear that the choice of the preferred alterna-
tive turns to some extent on the explicit or implicit rate
at which costs and benefits are discounted. It is equally
clear that in the absence of general guidance on this sub-
ject a variety of quite different discounting practices have
developed in the Federal agencies.
While different rates in different program areas may be
appropriate, the absence of agreement on basic concepts,
clearly reflected in the continuing disputes over basic pol-
icies, establishes a presumption that existing practices
merit review.
It is with this thought in mind that we have prepared
the enclosed questionnaire, which I ask you to complete and.
return by November 15, 1967. Cases involving water and re-
lated land resources are excluded, since these are handled
under the rules set forth in Senate Document No. 97; simi-
larly, cases involving make-or-buy decisions are excluded,
since these are covered by Bureau of the Budget Circular
No. A-76.
Any questions you may have should be addressed to
Daniel B. Rathbun, Deputy Director for Systems Analysis, in
our Office of Policy and Special Studies. (129-5309)
29
PAGENO="0069"
65
APPENDIX IV
Page 2
My intent at this point in time is simply to develop
co~nprehensive and accurate data on existing discounting
practices. The results will be tabulated and copies sent to
you. The results will also be discussed with the leaders in
the executive and legislative branches who have shown in-
creasing interest in this subject and who have asked if a
greater degree of standardization in evaluative practices
would be beneficial.
Your cooperation will be appreciated.
Sincerely yours,
1sf Elmer B. Staats
Comptroller General
of the United States
The Honorable
The Secretary of Housing and
Urban Development
30
PAGENO="0070"
66
APPENDIX IV
Page 3
THE USE OF DISCOUNT RATES TN
EVALUATING FEDERAL PROGRAMS
Instructions:
A. Exclude water and related land resource programs and programs involv-
ing the acquisition of commercial or industrial products and services.
B. If discounting practices of organizational components differ, please
complete separate questionnaires for each component.
C. Please return the completed questionnaire(s) to the Comptroller Gen-
eral by November 15, 1967.
1. Did you use a discount rate or rates in the analysis of fiscal year 1969
programs?
YES NO
2. If "yes" is checked, please identify (1)the fiscal year 1969 programs for
which discount rates were used and (2) the rate or rates used in each case.
FISCAL YEAR 1969 PROGRAMS DISCOUNT RATE(S) USED
3. If "yes" is checked, please describe briefly the rationale employed in se-
lecting rates.
4. If "no" in question No. 1 is checked, did your program evaluation proce-
dures involve appraisals of alternative programs or systems in terms of
periods which were shorter than the estimated useful lives of the programs
or systems (e.g., a comparison of systems with estimated useful lives of
20 years on the basic of 5-year costs)?
~ LII NO
5. If"yes" is checked, please identify the programs or systems, the time pe-
nod used in evaluating alternatives, and the estimated useful lives of the
programs or systems.
PROGRAM DURATION ESTIMATED USEFUL LIFE
(SYSTEM) OF PERIOD OF PROGRAM (SYSTEM)
PAGENO="0071"
67
APPENDIX V
ogre~ of tfje ~iuteb ~`tate~ ~
JOINT ECONOMIC COMMITTEE w.~e,ocx .aT.'~N.
(~n~1tD punsw,~~Totrc.I(.) C~ 1CLAWIO~,7?TH~G~ES$) ~ME5 ~. K~oWLEI.
December 19, 1967
The Honorable Elmer B. Staats
Comptrofler General of the United States
General Accounting Office
Washington, D. C.
Dear Elmer:
It has been called to my attention that
the General Accounting Office has undertaken an
examination of the use that Federal agencies make
of discount rates and an evaluation of such programs.
As you know, this is a subject in which we have a
continuing interest; the issue was examined in the
recent hearings before the Subcommittee on Economy
in Government of the Joint Economic Committee,
90th Congress, 1st Session.
Since this work deals with actual practices
in the Federal agencies, it would complement the
material presented in these hearings. I would, there..
fore, appreciate it if you would make the results of
your work available to the Joint Economic Committee
as soon as possible.
Sincerely,
)
/*~ /`//
I ~/ /
/ -
William Proxmire
Chairman
32
0
PAGENO="0072"