report of the president's commission to study capital ... · iii the president’s commission...

WASHINGTON, D.C. FEBRUARY 1999

REPORTof thePRESIDENT’SCOMMISSIONto STUDYCAPITAL BUDGETING

Page

i

TABLE OF CONTENTS

Letter of Transmittal ................................................................................................................ iiiAcknowledgements .................................................................................................................... vPreface ........................................................................................................................................ 1Executive Summary .................................................................................................................. 3What is ‘‘Capital’’? ..................................................................................................................... 9Trends in Capital Spending ..................................................................................................... 13Budgeting Capital ..................................................................................................................... 17Do Current Budget Conventions Distort Decisions About Federal Capital Spending? ....... 25Recommendations ...................................................................................................................... 31Pros and Cons of a ‘‘Cap’’ on Capital Spending ...................................................................... 39A Capital Budget and Depreciation ......................................................................................... 43Other Versions of a Capital Budget ........................................................................................ 45Conclusion .................................................................................................................................. 45Endnotes .................................................................................................................................... 47Selected References and Bibliography ..................................................................................... 51Appendix A: Executive Order 13037: Commission to Study Capital Budgeting, and the

Amendments .......................................................................................................................... 53Appendix B: Commission Membership .................................................................................... 57Appendix C: List of Witnesses and Written Statements ....................................................... 59Appendix D: Outline of Material on the Website ................................................................... 61

GENERAL NOTES

This publication may be purchased from the U.S. Government Printing Office andits bookstores. An order form is included in this publication. The report is alsoavailable, with testimony and other supporting materials, on the Internet at:http://www.whitehouse.gov/pcscb.

Numerical detail in this document may not add to the totals because of rounding.

iii

The President’s Commissionto Study Capital Budgeting

Co-Chairs

Kathleen Brown

Jon S. Corzine

Members

Willard W. Brittain

Stanley E. Collender

Orin S. Kramer

Richard C. Leone

David A. Levy

James T. Lynn

Cynthia A. Metzler

Luis Nogales

Carol O’Cleireacain

Rudolph G. Penner

Steven L. Rattner

Robert M. Rubin

Herbert Stein

February 1, 1999

Honorable William J. ClintonThe PresidentThe White House1600 Pennsylvania Avenue, N.W.Washington, D.C. 20500-0001

Dear Mr. President:

We are hereby submitting the final report of the Commission to Study Capital Budgeting.

As you requested, we have concentrated on capital spending by the federal government. However, we haveconcluded that capital spending by all levels of government, as well as by the private sector, provides thenation with important long-term benefits.

Our research shows that the current budget process does not permit decision-makers in the executivebranch and Congress to pay sufficient attention to the long-run consequences of their decisions. Thisresults in inefficient allocation among capital expenditures and shortchanges the maintenance of existingassets.

In this report, we propose a series of recommendations that we believe would improve each of thecomponent parts of the budget process: setting priorities currently and for the long run, making budgetdecisions in the current year, reporting on those decisions, and subsequently evaluating them in order tomake improvements in future years. We do not propose, however, the current adoption of a formal capitalbudget, as defined and discussed in the report.

To implement the proposed recommendations, the executive branch and Congress must ensure that theappropriate information is made available to decision-makers and the public throughout the budgetprocess. As a result, policy makers will be both properly informed when deciding how to spend taxpayers’money, and held accountable by the public for those decisions.

This report reflects the views of commissioners from many different backgrounds. We reached ourconclusions after conducting nine hearings, at which more than thirty experts from the private and publicsectors presented their views. While the members of the commission endorse the recommendationspresented herein, individual members do not necessarily agree with all of the analysis or with each andevery word of the report.

The commission worked diligently to carry out your directions. We hope that our recommendations willhelp the Administration, future presidents, and the Congress in improving the budget process, especiallyas it relates to decisions about capital spending.

Respectfully,

Kathleen Brown, Co-Chair Jon S. Corzine, Co-Chair

v

ACKNOWLEDGEMENTS

The commission operated under the auspices of the U.S. Department of the Treasury. Assistingthe commission were staff members of the Office of Management and Budget and the Depart-ment of the Treasury, as well as individuals from organizations with which some of the commis-sioners are affiliated. The commission is grateful for all this support, and wants to acknowledgethe following people who made contributions to the overall effort and to this report:

Dick Emery, Executive Director to the commission (OMB)Robert E. Litan, writer of the report (The Brookings Institution)E. William Dinkelacker, Designated Federal Official (Department of the Treasury)

Barry Anderson (former Executive Director and formerly of OMB)Barbara Basser-Bigio (Goldman, Sachs & Co.)Bethann J. Beck (PricewaterhouseCoopers)Jesse Brackenbury (Levy Institute)Thomas J. Craren (PricewaterhouseCoopers)Philip R. Dame (OMB)Joshua Gotbaum (OMB)Lawrence W. Hush (OMB)Robert W. Kilpatrick (OMB)Sarah Lee (OMB)Patrick Locke (OMB)Rusty Moran (OMB)Jason Orlando (OMB)Justine Rodriguez (OMB)Arthur Stigile (OMB)

1

1 The full text of the initial order and subsequent amendmentsare shown in Appendix A. Other materials the commission exam-ined in carrying out its duties, including summaries and full ver-sions of the testimony the commission heard from a variety of ex-perts and interested parties, are posted on the website of the com-mission at: www.whitehouse.gov/pcscb.

2 The staff from the various organizations who provided assist-ance to the commission are listed in the Acknowledgements.

PREFACEBy Executive Order 13037, issued on March

3, 1997, the President of the United Statesestablished this Commission to Study CapitalBudgeting. The order directed the commissionto prepare a report discussing various aspectsof capital budgeting, including the budgetingof capital in other countries, state and localgovernments, and the private sector; theappropriate definition of capital; the roleof depreciation in capital budgeting; andthe effect of a capital budget on budgetarychoices, macroeconomic stability, and budg-etary discipline. 1

Since its formation, the commission hashad nine meetings, has heard testimony andreceived written submissions from many indi-viduals from the public and private sectors,

and has reviewed the relevant and voluminousprofessional literature. It has carried outits work on its own. The Administrationdid not provide any instructions concerningparticular results or suggestions that it wantedthe commission to explore or recommend.

This report is the product of the commis-sion’s hearings and of deliberations amongits members and associated staff. 2 The mem-bers of the commission endorse the rec-ommendations presented in the report, al-though individual commissioners may notagree with all of the analysis or with eachand every word. In some cases, the separateviews of certain commissioners on selectedsubjects are provided in footnotes to thereport (which are signified by alphabeticalletters; all other numbered notes after thispreface are found at the end of the report).

3

EXECUTIVE SUMMARYThe subject of capital budgeting—or indeed

public budgeting for any purpose—may appearto be of interest to only a special audience:government professionals ‘‘inside the Beltway’’and perhaps some analysts in the investmentcommunity. Nothing could be further fromthe truth.

The budget of any organization, privateor public, is a statement of both the resourcesto be made available to the organizationand the priorities of those who manageit. The budget that the President submitsto the Congress, which in fiscal year 1999covered expenditures of nearly $2 trillion,tells the American people how the administra-tion proposes to spend their taxes and, untilrecently, the proceeds of federal debt issuedto finance the shortfall between total expendi-tures and revenues. The budget is thusinherently a political document, but in thebest sense of the term. This is becauseit reflects the collective judgment of theindividuals in a democracy about how muchpublic funds are to be raised and howthey are to be used.

This commission has devoted its attentionto one particular kind of expenditure inthe federal budget: spending on ‘‘capital.’’Although this term has been defined invarious ways for different purposes, a commonelement among all of the definitions is thatcapital spending—whether undertaken by theprivate or public sector—is intended to gen-erate benefits over the long run.

In this report, we have concentrated oncapital spending by the federal governmentbecause it is our charge. But we cannotemphasize too strongly that capital spendingat all levels of government, as well asby the private sector, provides importantbenefits to the nation as a whole in significantpart because those benefits are deliveredover the long run. It is easy in the day-to-day battles over budget policy to forgetthat such spending helps determine the kindof society that we and our children willlive in—not just this year but many years

from now as well. We therefore encouragethis president and future presidents to helpeducate American citizens about the impor-tance of devoting current resources towardfuture needs—in the form of spending oncapital by both the private and public sectors.

Most firms in the private sector, as wellas many state and local governments, recog-nize the importance of capital expendituresby making decisions about them separatelyfrom decisions about how much to spendon annual operating expenses. By contrast,the federal government has never done this.

This commission has been directed to exam-ine whether this practice ought to bechanged—that is, whether the federal govern-ment should adopt a ‘‘capital budget’’—and,if not, what other steps, if any, shouldbe taken to improve the federal decision-making process as it relates to spendingon capital or ‘‘investment’’ expenditures.

Capital budgeting is a process that takesexplicit account of capital spending levels.In this report, we primarily examine versionsof a capital budget in which either: (1)the size of the deficit or surplus is madeto depend, in part or in whole, on theamount of expenditures defined as ‘‘capital,’’or (2) a single decision is made about howmuch to spend on ‘‘capital,’’ under somedefinition. A variation of the first definitionis what we label the ‘‘simplistic’’ versionof the capital budget, one in which capitalspending may be financed, in part or intotal, by borrowing. We treat the seconddefinition as the equivalent of imposing aseparate ‘‘cap’’ on expenditures defined tobe capital, or in the alternative, a processwhereby the depreciation of capital is explicitlytaken into account in the budget process.We briefly note in a concluding section thatthere are other, perhaps less formal, variationsof a capital budget that we do not extensivelyanalyze here.

The commission had its origins duringthe Congressional debate about whether toamend the Constitution to require the federal

4 REPORT OF THE PRESIDENT’S COMMISSION TO STUDY CAPITAL BUDGETING

a Comment of Commissioners Corzine, Kramer, Leone, Levy,O’Cleireacain, Rattner, and Rubin: We wish to register our strongopposition to any amendment to the Constitution that would man-date balanced federal budgets. The macroeconomic straightjacketimplied by such a change in the Constitution would cost the nationdearly in lost growth, unnecessary unemployment, and slow recov-ery from recessions. Indeed, were such an amendment to pass, itwould be essential that many spending items be exempted rou-tinely, while others be exempted under clearly defined cir-cumstances. Rather than simplify the budget process, it would thenbecome more confused and opaque. In addition, democratic govern-ance would suffer since the ability of Congress and the presidentto respond to public priorities would be unduly constrained.

Specifically, in a recession tax receipts fall and spending for suchitems as unemployment insurance rises. This imbalance offsets re-cessionary forces, thus speeding recovery. It is one of the reasonseconomic downturns have been less severe since World War II thanbefore. Indeed, the insistence on trying to balance the budget inthe early 1930s is generally considered to have deepened the GreatDepression. The counter-cyclical advantages of the current systemare not trivial. Giving them up may lead to real costs, particularlyamong working men and women: income lost when governmentcannot fight a recession is lost forever.

b Comment of Commissioners Lynn, Penner, and Stein: We donot favor adopting at this time a capital budget of any kind, wheth-er of the kind here labeled ‘‘simplistic’’ or any other known to us.We endorse the qualification ‘‘at this time’’ to allow for the possibil-ity that future developments in information, sophistication, anddiscipline in the budgetary process might recommend a differentcourse.

c Comment of Commissioners Corzine and Levy: These weak-nesses in the budget process may have macro as well as micro con-sequences. One of the aggregate effects of sub-optimal choices may,at times, be either an inadequate or an excessive level of capitalspending.

d Comment of Commissioners Corzine, Leone, and O’Cleireacain:We believe it is both possible and desirable to move towardclassifying the federal budget in two parts: as ‘‘capital,’’ in thesense of investment with long-term effects: and as ‘‘operating,’’such as consumption expenditures and transfer payments for thecurrent year. This approach, which is consistent with private sectororganizations’ practices, would enable the U.S. government to bet-ter understand, manage, and finance its commitments.

As is the custom at the state and local levels of government, acapital budget classification does not mean that the governmentwould lose its flexibility to manage during periods of fiscal con-straint/plenty. Nor does it mean that all capital expenditures mustbe financed from borrowed funds. Moreover, the definition of cap-ital, like other aspects of the current budget structure, could be re-fined and updated over time.

government to have a balanced budget everyyear. Nothing in this report should be con-strued as support for the balanced budgetamendment considered by the Senate in 1996. a

Nor does the commission endorse the adoptionof the simplistic version of the capital budget.Furthermore, a majority of the membersof the commission does not support, at thistime, adopting a budget procedure that wouldimpose a separate cap on capital spending. b

The reasons for reaching these conclusionsare spelled out in the body of the report.

At the same time, we have concludedfrom our study of existing practices andafter gathering evidence from a wide rangeof experts, that the existing federal budgetprocess—as it affects decision-making aboutcapital expenditures as well as other typesof spending—has significant weaknesses. In-sufficient attention is paid to the long-runconsequences of budget decisions. Capitalspending in particular is inefficiently allocatedamong projects. Moreover, the current processshortchanges the maintenance of existing as-sets. c

Accordingly, the commission urges the Con-gress and the executive branch to undertakea thorough examination of how the budgetprocess may be improved beyond addressingcapital-related needs. Toward this end, itmay be productive for both branches tocreate a new Commission on Budget Conceptsto aid them with this task.d

In the meantime, we believe there area series of constructive responses to theshortcomings we have identified, though theydo not include adopting any particular formof a capital budget as we have just definedthe term. These responses are aimed atimproving each of the component parts ofthe budget process: setting priorities currentlyand for the long run, making budget decisionsin the current year, reporting on those deci-sions, and subsequently evaluating them inorder to make improvements in future years.Key to achieving these improvements is ensur-ing that the appropriate information is madeavailable to decision-makers and the publicthroughout the process so that policy makers(1) are properly informed when deciding howto spend taxpayers’ money and (2) can beheld accountable by the public for thosedecisions.

The recommendations we summarize belowtake account of two important features offederal budgeting.

First, many government efforts have objec-tives, such as the management of foreignaffairs or the defense of the nation, thatcannot be readily measured in monetaryterms. In stark contrast, it is relativelyeasy to keep score in the private sector,where firms are often judged by a singlemetric, such as current profitability, return

5EXECUTIVE SUMMARY

e Comment of Commissioners Kramer, Leone, and O’Cleireacain:We believe the text over-emphasizes ‘‘theoretical’’ market disciplinewhen it comes to borrowing for capital by the states. Most states,as a simple matter of ‘‘capacity to pay,’’ could borrow much morethan they do. In fact, almost always, in the real world the actualconstraints are political (including referendum requirements) andpractical—demands for current revenues limit the amounts avail-able for debt service.

on equity, or the dollar value of their share-holders’ equity.

Second, borrowing is subject to less dis-cipline at the federal level than it is atlower levels of government. States and local-ities cannot ‘‘print money’’ to cover the debtsthey issue, whereas one arm of the federalgovernment—the Federal Reserve—has theability to ‘‘monetize’’ debt issued by theTreasury. A related difference is that federaldebt is viewed by the marketplace as prac-tically free of default risk, whereas statesand localities have a strong interest in main-taining high credit ratings, which constrainsborrowing at the state and local level. e

These considerations necessarily imply thatfederal budgeting rules should not simplyreplicate rules that may be used in theprivate sector or at the state and locallevels of government. But at the same time,because the existing federal budget processhas the weaknesses we have noted, certainimprovements are appropriate. We have con-centrated on suggestions for the executivebranch; however, as will become evident below,certain of these require the cooperation ofand concurrence by the Congress.

We also recognize the essential role ofthe American people as monitors, advocates,and parties whose interests ultimately areat stake during the budget process. Forthis reason it is important to increase thetransparency of that process—not only toenhance the quality of inputs to the Congressfrom the private sector and other levelsof government, but also to increase the federalgovernment’s accountability to the Americanpeople.

To facilitate the setting of prioritiesamong all programs, not just those involv-ing capital expenditures, the commissionrecommends:

Recommendation 1: Five-Year StrategicPlans.—Although federal agencies are now re-

quired (under the Government Performanceand Results Act) to prepare strategic plansevery three years and performance plans an-nually, this process should be improved in sev-eral respects:

• The strategic plans should (1) be preparedannually, (2) be integrated with the an-nual performance plans and the agencies’five-year budget projections that are nowsubmitted to the Office of Managementand Budget (OMB), and (3) be includedas an integral part of the budget justifica-tions sent to the Congress.

• The strategic plans of the agencies andtheir annual budgets should be tied to thelife-cycles of their capital assets.

• OMB should standardize the formats ofthese plans, in consultation with GAO andCBO, to make them more useful to policymakers.

• OMB should expand its efforts to evaluatethe plans and facilitate the Administra-tion’s use of them for government-wideplanning.

• Congress should take such plans into ac-count in deciding on annual agency appro-priations. It should also consider how itmight improve its own procedures so thatit can pay more attention both to thelonger-run implications of its current yeardecisions and to issues with longer-runconsequences. In undertaking this task,Congress might find it useful to take ad-vantage of the wide range of institutionalexpertise available to it, including re-sources within the Congressional BudgetOffice, the General Accounting Office, andthe Congressional Research Service.

Recommendation 2: Benefit-Cost Assess-ments.—There should be an ongoing effortwithin the federal government to analyze thebenefits and costs of all major government pro-grams (whether or not related to capital spend-ing), so that they can be adjusted, refashioned,or eliminated, as appropriate. OMB, the agen-cies, and the Congress (through GAO and CBOin particular) should be given the resourcesto carry out this important function.


f Comment of Commissioner Levy: I urge the Congress to ad-dress the lease-purchase problem as part of a special or com-prehensive amendment to the current budget process. I discuss thisissue in greater detail in a subsequent footnote (l).

To improve the process by which an-nual budget decisions are made, thecommission recommends:

Recommendation 3: Capital AcquisitionFunds.—To promote better planning andbudgeting of capital expenditures for federallyowned facilities, Congress and the executivebranch should experiment by adopting for oneor more agencies separate appropriations for‘‘capital acquisition funds’’ (CAFs). Budget au-thority would be lodged in the CAFs for feder-ally owned capital assets. The CAFs would‘‘rent out’’ their facilities to the various pro-grams within each agency, charging them theequivalent of debt service.

• CAFs would help ensure that individualprograms are assessed the cost of usingcapital assets.

• By spreading capital costs across entireagencies, CAFs would help smooth out thelumpiness in appropriations sometimes as-sociated with large capital projects.

• If the CAF experiment proves successful,the CAF approach should be adoptedthroughout the government.

Recommendation 4: Full Funding forCapital Projects.—All capital projects, or us-able segments thereof, should be fully fundedbefore the work begins. In this way, Congresscan fully evaluate their likely costs and bene-fits before appropriating funds for them.

Recommendation 5: Adhering to theScoring Rules for Leasing.—Existing rulesthat govern the scoring of leases should bestrictly followed by both agencies and the Con-gress. This will discourage the signing of short-term leases when it is cheaper over the longrun to construct or purchase a facility. f

Recommendation 6: Trust Fund Re-forms.—Although trust funds for highways,airports, and other uses insulate certain typesof spending from the balancing process thatis inherent in the rest of the budget, they canbe useful if the funds going into them trulyrepresent charges or fees for the use of thegovernment services they support. But this

purpose is fulfilled only if the monies raisedby earmarked taxes or fees to support infra-structure or other types of capital—averagedover some reasonable period, such as threeyears—are actually spent on the dedicateduses.

• To ensure that this is done, the President’sbudget should disclose the earmarkedtaxes or fees and spending of these variouscapital-related trust funds. This will allowpolicy makers to make informed decisionsabout whether to increase spending on theauthorized activities or reduce the chargesnow being assessed purportedly to financethose activities.

• State and local governments that are re-cipients of capital-related grants from thefederal government should be required tomaintain their capital—such as high-ways—as a condition to receiving any ad-ditional federal aid (unless those govern-ments can demonstrate that there is nolonger a need for the assets the federalgovernment initially supported).

Recommendation 7: Incentives for AssetManagement.—The executive branch and theCongress should experiment with incentives toencourage agencies to manage their assets effi-ciently. One possibility might be to allow, onan experimental basis, one or more agenciesto keep a limited portion of the revenues theyraise from selling or renting out existing as-sets.

Steps must be taken to improve themethods that are used to give the resultsof those decisions (and the programsthey support) to the public and policymakers. In particular:

Recommendation 8: Clarification of theFederal Budget Presentation.—The Presi-dent’s annual budget should contain a break-down of proposed current and projected federalspending over the budget year and the subse-quent four years among the following cat-egories: investment, operating expenditures,transfers to individuals, and interest. Such abreakdown would make available to policymakers and the wider public the President’slong-run vision for federal spending. This in-formation might also encourage Congress to

7EXECUTIVE SUMMARY

g Comment of Commissioners Lynn, Penner, and Stein: We donot believe that this four-way classification of expenditures wouldbe helpful in making good budgetary decisions.

find ways of taking a longer-run view in itsannual budget deliberations. g

Recommendation 9: Financial StatementReporting.—Reporting on financial activitiesand asset positions of the federal governmentshould be enhanced in a number of ways tobetter inform the Congress and the publicabout the ways in which the federal govern-ment’s assets are being used and maintained:

• Federal agencies should be required toissue to policy makers and the public moredetailed information (both in print formand on their websites) about the composi-tion and condition of the federally ownedor managed capital assets under their con-trol. OMB should consolidate these re-ports, which should continue to be basedon independently developed accountingstandards, and report on them in sum-mary fashion in the annual budget.

• There should be enough information in theconsolidated reports to provide Congressand the public with accurate benchmarksfor making appropriate comparisons bothin the current year and over time.

• The calculation of depreciation in variousgovernment reports should be standard-ized.

With more comprehensive, objective informa-tion on how the federal government as awhole, as well as individual agencies andprograms, have used resources, increased ordepleted assets, and undertaken new invest-ments, debates over critical national policieswould be better informed. Private corporationsreport audited financial results and assetand liability positions to investors. By thesame token, the federal government shouldmake available to the American people auditedfinancial statements and underlying detailthat go well beyond the information shownannually in the unified budget. Just ascorporate decision-makers have accurate ac-counting data to help them assess past per-formance and make decisions about the future,Congress and the public should also haveaccurate accounting on federal assets andinvestments.

Recommendation 10: Condition of Exist-ing Assets.—Work is planned at the federallevel for agencies to begin developing stand-ardized methods for estimating deferred main-tenance. The commission strongly supportsthese efforts and encourages OMB to workwith the agencies to complete this taskpromptly and to implement its results. In addi-tion, the federal government, working withstates and localities, should endeavor to reporton the condition of assets owned at these lowerlevels of government, or at least those thathave received federal support. In combinationwith the rest of the information provided inthe audited financial statements, data on de-ferred maintenance will enable policy makersto develop sound plans for maintaining exist-ing assets and spending on new ones wherethat is advisable.

Finally, steps should be taken to im-prove the process used in evaluatingthe impact of past budgetary decisions,so that policy makers can be in a positionto make improvements, if warranted.

Recommendation 11: Federal ‘‘ReportCard.’’—Under OMB guidance, agenciesshould assess the extent to which major invest-ment projects have produced returns in excessof some benchmark cost of capital, such asthe prevailing interest rate on long-term fed-eral debt, the average cost of capital expectedby private market investors, or some otherthreshold that OMB believes the public wouldfind useful. This federal ‘‘Report Card’’ couldbe included in the President’s annual budget.The commission recognizes that the projectsfor which it might be feasible to provide amonetary analysis may account for a relativelysmall fraction of total spending; nonetheless,it believes that over time advances in estimat-ing techniques may permit a larger fractionof total spending to be evaluated in this man-ner. Where benefits and costs cannot be ex-pressed in monetary terms, the evaluationsshould identify project objectives and assessoutcomes qualitatively.

The foregoing recommendations are summa-rized in the table on the following page.The columns in the table refer to threedifferent classes of capital, which are discussedin the body of the report: the federal govern-ment’s own assets (such as buildings in


which federal agencies are located), the federalgovernment’s investment in assets owned bystate and local governments (such as high-ways), and the federal government’s invest-ment in what we have labeled intangiblenational assets that are financed but notowned by the government (such as benefitsaccruing from federal expenditures on researchand development and or on education). Ourrecommendations are then classified both bythe stage of the budget process at whichthey are directed and by the types of capitalthat they are likely to affect. Because anumber of our recommendations are designedto improve decision-making with respect toone or more categories of capital, they arelisted in multiple columns.

While the primary responsibility for initiat-ing most of the foregoing recommendationsrests with the executive branch, in certaincases Congress also has an important role.

Indeed, virtually all of the recommendationsrequire active Congressional cooperation ifthey are to have a positive effect on thebudget process and budget decisions.

Although the commission as a whole doesnot endorse setting a separate cap on capitalspending, it nonetheless discussed the tech-nical details of such a change in budgetprocedure. The concluding section of thisreport contains our findings on these issues,outlines the key pros and cons of subjectingcapital spending to its own limit, analyzesproposals to reflect depreciation of capitalassets in the budget process, and brieflydescribes some alternative versions of a capitalbudget.

In sum, the federal budget process canbe and should be improved. The commissionbelieves the recommendations outlined in thisreport would help accomplish this objective.

Summary of Recommendations by Stage of the Budget Processand Type of Capital Affected

Federal Investment In:

Federal Assets State/Local Assets Intangible Assets

Strategy and Planning .......... Five-Year Plans Five-Year Plans Five-Year PlansBenefit-Cost Analysis Benefit-Cost Analysis Benefit-Cost Analysis

Decision-Making .................... CAFsFull FundingProper Lease ScoringTrust Fund Reforms Trust Fund ReformsInvestment Life-Cycle Plan-

ningStates/Localities Maintain

AssetsIncentives for Better Asset

Management

Reporting ................................ Improved Financial Report-ing

Improved Financial Report-ing

Improved Financial Report-ing

Audited Financials Audited Financials Audited FinancialsAsset Inventory Asset Inventory

Evaluation .............................. Report Card Report Card Report Card

9

h Comment of Commissioner Levy: A distinction must be madebetween practical and theoretical definitions. Defining investmentbased on its benefits (such as ‘‘increasing social welfare’’ or ‘‘in-creasing long-term growth’’) is useful in theoretical discussions, butno accounting is possible since we can never be sure which outlaysqualify. At the same time, practical definitions—such as those em-bodied in Generally Accepted Accounting Principles (GAAP)—al-ways have shortcomings, but still can be very useful. If we are toconsider using investment or capital in federal accounting andbudgeting, then we must resign ourselves to the use of practicaldefinitions. The definitions of the Federal Accounting StandardsAdvisory Board (FASAB) are a functioning example.

WHAT IS ‘‘CAPITAL’’?This commission has been charged with

examining capital budgeting in other coun-tries, states and local governments, and theprivate sector, and, in the process, withaddressing a number of questions about capitalbudgeting. It is only appropriate, therefore,to begin with the threshold issue: whatis ‘‘capital’’ (or its annualized counterpart,‘‘investment’’)?

The commission has not settled on, nordoes it endorse, a single definition of capital. 1

Instead, a series of distinctions between dif-ferent types of capital or ‘‘investment’’ spend-ing, both by governments and by firms inthe private sector, seem warranted for dif-ferent purposes (and different commissionersplace varying amounts of emphasis on alter-native definitions of capital).

One distinction relates to the functionsof capital. At its broadest level, any spendingthat yields benefits beyond the typical report-ing period (such as a year) should be consid-ered to be investment, and ‘‘capital’’ refersto the assets created by this spending. Sucha definition would encompass spending notonly on physical or fixed assets, such asstructures and equipment, but also on humanand a variety of intangible assets. ‘‘Humancapital’’ consists of the skills imparted toindividuals through training and educationthat enable them to increase their earningsnot just in a single year, but potentiallythroughout their lives. Intangible assets cancover a very broad class of items. In privatesector financial accounting, for example, intan-gibles are often measured by the expendituresrequired to gain patents, copyrights, trade-marks, or other intellectual property protec-tion. Certain types of public spending—includ-ing research and development (R&D), defense,nutrition, disease prevention, police protection,and drug treatment and prevention pro-grams—may also produce intangible assetsthat deliver, or are at least designed todeliver, benefits over years, if not lifetimes.

Broad definitions of investment or capitalcould be useful for several purposes. For

example, to the extent citizens and policymakers are interested in enhancing economicgrowth, the definition should count both pri-vate and public sector spending on buildings,equipment, research and development (includ-ing some defense-related R&D), and educationand training. An even broader definitionwould be justified if the goal were to measurecapital aimed at improving social welfare—one that included expenditures on nationaldefense and police to enhance security aswell as spending on childhood immunization,maternal health, nutrition, and substanceabuse, to improve the health and well-beingof citizens over many years. h

The accounting standards used in the pri-vate sector do not take such an expansiveapproach to the definition of capital. Generallyspeaking, they limit capital to physical andcertain intangible assets (such as investmentsin intellectual property). Similarly, the Na-tional Income and Product Accounts (NIPA)—the federal government’s statistical systemfor collecting and reporting data on overalleconomic activity—define capital to be spend-ing only on physical assets. 2 It is importantto keep in mind, however, that while theseaccounting standards may be conservative,they do not necessarily constrain the waymanagers think about spending that provideslonger-run benefits. For example, althoughprivate sector accounting standards defineemployee training expenditures as an expense,this spending typically generates longer-termbenefits to the firm (and to the employees).The fact that these expenditures are writtenoff during the course of a year does notstop managers or investors from considering


them as investments in the future well-being of the firm.

A second distinction relates to who ownscapital: specifically, whether it is owned pri-vately or publicly (and if publicly, by federal,state, or local governments). Individuals andfirms reap most of the benefits from thespending on capital they undertake; however,the public benefits when government is mak-ing the expenditures. For example, governmentspending to educate each generation of citizensbenefits the entire public by ensuring thatthe population continues to be literate, cog-nizant of the benefits of our system ofgovernment, and able to work in an ever-changing economic environment. Similarly,when the government spends money on thenation’s defense or finances basic scientificresearch, the benefits accrue to all citizens.Appropriately enough, economists call invest-ments that confer benefits on a wide classof parties ‘‘public goods’’ because no privateperson or firm can capture all of theirbenefits. Identifying and funding those pro-grams that produce returns to society wellabove the cost of capital is especially importantfor enhancing economic growth.

These points highlight the different criteriathat are used to decide whether to addto private and public capital. In the privatesector, capital spending decisions are madebased primarily on how they affect sharehold-ers, and are evaluated predominantly in mone-tary terms. In the public sector, decisionsabout capital take into account the impacton the public at large and rest on bothmonetary and non-monetary considerations.

A third distinction is between federal govern-ment capital and national capital. Federalgovernment capital, as we use the term,refers only to those assets the governmentowns, such as federal buildings or federalmilitary hardware. National capital is a broad-er term, including all government spendingaimed at delivering long-term benefits toany portion of the nation, whether or notit is owned by the federal government. So,for example, using the broad functional defini-tion of capital discussed above, national capitalwould include spending at all levels of govern-ment on roads and other physical assets,research and development, and education and

training, among other items. At the federallevel, what OMB labels as ‘‘federal investmentoutlays,’’ illustrated in Table 1, representsfederally financed national capital regardlessof who owns it. 3 As the table shows, nearlyhalf of the federal government’s investmentoutlays in fiscal year 1997 were devotedto physical capital, about one-third to researchand development, and the balance to educationand training—roughly the same proportionsthat were prevalent during the earlier partof the decade. 4

Federal government capital, in contrast,can be defined as including only assets ownedby the federal government, so it can beaccounted for in a fashion similar to theway capital is measured in the private sector.For example, OMB’s Capital ProgrammingGuide, which provides guidance to federalagencies on capital planning, procurement,and management, defines ‘‘federal capital’’to include land, structures, equipment, andintellectual property (including software) be-longing to the federal government that hasan estimated useful life of at least twoyears. Consistent with this definition, Table2 illustrates how the federal governmentprovided almost $66 billion of budget authorityfor fiscal year 1997 on ‘‘major capital acquisi-tions’’: government buildings, information tech-nology, and ‘‘other items’’ (weapons systemsin the case of the Department of Defense,and facilities and equipment for other agen-cies). The table shows that the major partof the federally owned investment was fordefense-related purposes.

This distinction between ‘‘national’’ and ‘‘gov-ernment’’ capital is of more than academicinterest. As discussed below, the governmentof New Zealand has adopted a separatecapital budget but only for government capital.In contrast, the General Accounting Officehas suggested defining a budget target thatis a variation of national capital: publicinvestments that promise ‘‘to raise the privatesector’s long-run productivity,’’ which wouldinclude spending on infrastructure, non-de-fense R&D, education and training, and somedefense activities, but would specifically ex-clude what GAO calls ‘‘federal capital,’’ suchas government-owned buildings, weapon sys-tems, and land [GAO, 1993].

11WHAT IS ‘‘CAPITAL’’?

Table 1. FEDERAL INVESTMENT OUTLAYS,FISCAL YEAR 1997

(billions of dollars)

Outlays Percentof total

Physical capital:Direct federal defense ..................................................................... $52.4 23%Direct federal nondefense ............................................................... 19.7 9%Grants to state and local governments ......................................... 41.5 18%

Subtotal, physical capital ........................................................... 113.6 50%

Research and development:Defense ............................................................................................ 40.2 18%Nondefense ...................................................................................... 30.9 14%

Subtotal, research and development .......................................... 71.1 31%

Education and training:Grants to state and local governments ......................................... 25.0 11%Direct federal ................................................................................... 19.0 8%

Subtotal, education and training ................................................... 44.0 19%

Total, federal investment outlays ........................................ 228.8 100%

Source: OMB, Analytical Perspectives, Fiscal Year 1999, p. 125.

Table 2. MAJOR FEDERAL CAPITALACQUISITIONS, FISCAL YEAR 1997

(budget authority, in billions of dollars)

Construction and rehabilitation:Defense military construction and family housing 4.2Corps of Engineers ................................................... 1.6General Services Administration ............................ 1.4Department of Energy .............................................. 1.2Department of the Interior ...................................... 1.0Other agencies .......................................................... 5.6

Subtotal, construction and rehabilitation ........... 15.1

Major equipment:Department of Defense ............................................ 42.8Department of Transportation ................................ 2.2NASA ......................................................................... 0.6Department of the Treasury .................................... 0.3Other agencies .......................................................... 4.4

Subtotal, major equipment ................................... 50.3

Purchases of land and structures ............................... 0.3

Total, major acquisitions ........................................ 65.7

Source: OMB, Analytical Perspectives, Fiscal Year 1999, p. 135.


A fourth definitional distinction is betweencapital created by (1) direct government spend-ing and (2) public and private capital spendinginduced by government policies. The advantageof confining any definition to direct spendingis that measurement is relatively easy. None-theless, if the objective is to measure theimpact of overall government policy on na-tional capital (narrowly or broadly defined),then a definition based only on the govern-ment’s direct expenditures is too limited.A full accounting would also require inclusionof capital spending at the state and locallevels and by the private sector that maybe brought about by such policies as federaldeficit reduction (through lower interest rates),and targeted tax incentives, as well as regu-latory mandates such as those requiring orinducing expenditures on pollution controlor occupational safety. 5 Granted, such inducedspending may be very important; however,the operational problem with adding inducedexpenditures is that they cannot be directlymeasured, but instead must be estimated,using economic models or survey responses.

The different definitions underscore theproposition that ‘‘capital’’ is not a single,uniform concept, but one that varies accordingto why the term is being used. Indeed,this is one reason that most members ofthe commission are opposed to recommendingthat a separate capital budget using onesingle definition of capital be adopted fordecision-making purposes. Nonetheless, defini-tional issues should not stand in the wayof illuminating the consequences of choosingamong different government programs, wheth-er or not they are labeled as capital. Norshould debate over definitions distract atten-tion from (1) the need to improve planningand evaluation for whatever expenditures pol-icy makers may choose to label as capital,or, (2) in the case of federal capital inparticular, the need to identify the assetsthe government has and report them ina coherent way.

Finally, one important characteristic ofmuch (but not all) capital spending is thatits value declines over time. Buildings and

machines wear out. Patents and copyrightshave limited lives. Even the value of basiceducation and training may decline in aworld of continuing technological change,which requires many workers to upgradetheir skills constantly to maintain their earn-ings.

Accounting standards in the private sector,as well as the concepts reflected in theNational Income and Product Accounts, takeaccount of the declining value of capitalitems by requiring property and plant andequipment (but not land) to be ‘‘depreciated’’or ‘‘amortized’’ over their ‘‘useful lives.’’ Theannual amounts of depreciation or amortiza-tion represent expenses that, along with sala-ries, supplies, rent, taxes, and other expenseitems, are deducted from annual revenueto determine profits each year. 6 A numberof different methods for depreciation andamortization are in use, ranging from the‘‘straight-line’’ method (that computes the an-nual deduction simply by dividing the originalcapital investment by the years of usefullife) to various forms of ‘‘accelerated deprecia-tion’’ (that deduct more in the early yearsof an asset’s useful life and less in lateryears). Businesses may also use depreciationmethods for financial accounting purposesthat are different from those they use tocompute their income tax liability.

Some state and local governments accountfor the declining value of their debt-financedcapital assets by including in their annualbudgets the annual debt service on the bondsthey issued to finance the investments. Debtservice includes interest and the annualamount of the principal of the bond thatis paid off (similar to amortization of principalon a mortgage that individuals may takeout to finance their homes) or put intoa ‘‘sinking fund’’ that is eventually usedto pay off the bonds when they mature.The amortization component of the debt serv-ice charge is analogous to depreciation, butwith a time profile that is the oppositeof accelerated depreciation—much larger de-ductions in the later years than in theearlier years.

13

1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995

-10

-5

0

5

10

15

20

NET CAPITAL SPENDING ON PHYSICAL ASSETS

FIGURE 1

Federal Government

Private

State and Local

As a Percentage of GDP

TRENDS IN CAPITAL SPENDINGTwo important distinctions are useful to

keep in mind when considering trends incapital spending:

• The amount of capital spending in anygiven year is defined as ‘‘investment.’’ Thisis different from the total amount of theexisting capital ‘‘stock,’’ which is the cumu-lative total of all previous investmentminus cumulative depreciation.

• Investment, in turn, is often measured intwo ways: as either the ‘‘gross’’ amountof spending on capital items or the ‘‘net’’investment, which is the gross figureminus annual depreciation. Many analystsconcerned with the contribution of capitalto economic output pay more attention tothe net figures than the gross figures.

Figure 1 depicts trends in net spendingon physical assets alone, as a share ofGDP, by the private sector, state and localgovernments, and the federal government. 7

Since World War II, the shares of suchnet spending in GDP have been reasonablystable—more so in the public sector thanthe private sector—with private investmentsubstantially exceeding public investment.Meanwhile, within the government sector,since 1950 net investment at the state andlocal level has consistently outpaced federalspending. It is important to note, however,that about one-quarter of state and localinfrastructure spending is financed by federalgrants, and much of the rest has beensubsidized by the federal tax exemption onmunicipal and state debt.


-4

-2

0

2

4

6

8

RATE OF NONFARM PRODUCTIVITY GROWTH

FIGURE 2

Annual Percent Change

1948 1953 1958 1963 1968 1973 1978 1983 1988 19931997

Average(1948-1973)

Average(1974-1997)

One reason that public investment hasbeen of special interest to economists, busi-ness, and policy makers is its impact oneconomic output. In particular, some econo-mists have argued that the decline in thepublic investment-to-GDP ratio shown in Fig-ure 1 has contributed significantly to theslowdown in long-term growth from the firsthalf of the post-World War II era to thesecond. 8 As shown in Figure 2, althoughit has picked up in recent years, over thepast 25 years the annual rate of productivitygrowth in the United States, which determinesthe growth in average living standards, hasbeen substantially below that of the 1948–73period, which some have characterized asa ‘‘golden age.’’ Figure 3 suggests that theslowdowns in public spending and productivitygrowth have occurred more or less aroundthe same time.

The claim that the first slowdown (ininfrastructure spending) ‘‘caused’’ the otherone (in productivity growth) has proven tobe highly controversial, however. Among other

things, various economists have claimed thatthe causation runs the other way: that is,(1) public capital spending has slowed becauseeconomic growth has slowed; (2) the publiccapital buildup in the 1950s and 1960s waslargely associated with once-in-a-generationevents—the construction of the interstate high-way system and the construction of schoolsfor the baby boom generation—that couldnot have been expected to be repeated afterthey were completed; and (3) the statisticalestimates used to prove that the slowergrowth in capital spending caused the slow-down in productivity growth are highly sen-sitive to the time period examined. 9 Moreover,as already shown in Figure 1, public sectorinvestment in physical assets is considerablysmaller in magnitude than private sectorinvestment, which has also decreased relativeto GDP during the same period in whichpublic investment has declined. Both of thesefacts raise the question of whether andwhy public capital spending in particularshould be singled out as being primarily

15TRENDS IN CAPITAL SPENDING

1948 1952 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996

-2

0

2

4

6

8

10

Per

cent

age

of G

DP

-3

-2

-1

0

1

2

3

4

5

6

7

8

Ann

ual P

erce

nt C

hang

e

NONFARM PRODUCTIVITY GROWTH AND PUBLIC INVESTMENT

Nonfarm Productivity GrowthAnnual Percent Change

Net Public Capital Spending on Physical Assets as a Percentage of GDP

FIGURE 3

responsible for the trends in productivitygrowth.

There’s no need to resolve the debateover what has caused the slowdown in meas-ured productivity growth over the past 25years to conclude that all types of capital(fixed assets, human capital, and intangibles),whether owned by the private or publicsectors, remain important to economic growth.Economic theory has long pointed to thatconclusion. The challenge for decision-makersin both the private and public sectors isto undertake those investments that realisti-cally promise returns that exceed the costof financing the investments; otherwise, scarceresources will be wasted.

Some observers have attempted to drawpolicy implications for the United Statesby comparing the intensity of investmentactivity here (both public and private), aswell as rates of return on investment, withsimilar figures for other industrialized coun-tries. Such comparisons, however, do notprovide a standard for judging the appropriate-

ness of the amount of total investment inthis country, and still less for judging theamount of investment spending by the federalgovernment.

In any event, several points should bemade about those comparisons. First, thoughby conventional measurements the UnitedStates invests a smaller share of its GDPthan other advanced countries do, that isnot true (1) if investment is defined morebroadly to include expenditures on education,research and development, consumer durablesand defense capital, and (2) if the relativeprice of investment goods and other outputis correctly calculated [Kirova and Lipsey,1998]. Second, comparing investment/GDP ra-tios may not be as illuminating as comparingrates of return on capital. When this isdone, the United States typically comes outon top of other countries. Third, significantdifferences in definitions and demographicconditions make comparisons of public invest-ment particularly complicated across coun-tries. 10

17

BUDGETING CAPITALThe executive order directs the commission

to report specifically on capital budgetingpractices used in the private sector, by stateand local governments and in other countries,and then to explain the relevance of thosepractices for budget decisions made by thefederal government.

By definition, a budget is a constraintbecause it implies the existence of a finiteamount of resources that can be allocatedamong alternative uses. But what is it thatlimits the amount of available money? Thevastly different answers to this question forprivate firms, state and local governments,and the federal government help shed lighton the extent to which capital budgetingpractices followed elsewhere are suitable forthe federal budget.

Capital Budgeting in the Private Sector

The American economy is populated byover twenty million businesses, large andsmall, which surely have different ways ofbudgeting capital expenditures. Nonetheless,certain conventions have become standardizedthrough custom and repetition, as well asthrough formal professional practice. As aresult, it is possible to describe a stylizedprocess that many firms, typically largerpublicly held corporations, use to analyzetheir capital spending options, to choose amongthem, and then to account for those choices.To help understand these conventions, itis useful to refer to three basic financialstatements that are found in the annualreports of publicly held companies: the balancesheet, the income statement, and the state-ment of cash flows.

The balance sheet provides a financial snap-shot at a single point in time, usuallyat the end of a reporting year, of thefirm’s assets (on one side) and liabilitiesand net worth (on the other). The twosides add to the same total. Assets are‘‘financed,’’ as it were, by borrowing (liabilities)and shareholders’ contributions (paid-in capitaland retained earnings). Broadly speaking,three categories of assets are reported on

the balance sheet: short-term assets (suchas cash, marketable securities, receivables,and inventories), fixed assets (structures andequipment) minus any cumulative deprecia-tion, and intangible assets minus any cumu-lative amortization. Using the nomenclatureof this report, capital for private firms consistsof fixed assets and, under some definitions,intangible assets as well. 11 It is worth notingthat private sector accounting has been stand-ardized in Generally Accepted AccountingPrinciples (GAAP), which are used to preparefinancial statements. 12 The Financial Account-ing Standards Board, an independent bodyof experts, is responsible for seeing thatthe principles embodied in GAAP are main-tained, updated, and applied in a fair andreasonable manner. 13

The income statement is an accountingof revenues and expenses over a certaintime frame, typically a year, with the dif-ference representing the firm’s profit or loss.Because businesses exist to generate profits,spending decisions by private companies—including whether and how much to investin capital projects—are judged predominantlyby their likely impact on profitability. Invest-ments in capital projects by definition aredesigned to deliver benefits over the longrun, so capital spending does not appearon the income statement. Instead, the depre-ciation or amortization of existing capitalrecorded on the balance sheet shows upon the income statement as an expensethat reduces reported profits.

Where, then, might spending on capitalshow up? The typical place is on the statementof cash flows. This statement combines infor-mation on where a firm gets its moneyand where it spends it during the courseof a year: on operating activities, intereston any outstanding debt, and the full costof capital projects.

How do firms decide how much capitalspending to undertake, and of their manypossible options, which projects to pursue?


Here, again, practices surely vary. But certainfacts and conventions are widely understood.

First, most firms cannot spend withoutlimits: they are constrained by their cashon hand, revenue likely to be realized inthe short run, and how much additionalcash they might be able to raise by sellingexisting assets, borrowing, or selling newequity. 14 In turn, creditors and investorsdecide whether to provide funds, if theyare requested, and on what terms basedon the firm’s ability to repay its debts(in the case of borrowings) and generateprofits (in the case of equity sales). Inshort, firms in the private sector are subjectto market discipline.

Second, it is standard practice in privateindustry for firms to assess their capitalprojects by estimating their ‘‘net presentvalue.’’ Net present value (NPV) is calculatedby projecting the future cash flows the invest-ment is likely to generate (such as rentalsfrom a building or cost savings from investingin new equipment or machinery), ‘‘discounting’’the future cash flows by the ‘‘time valueof money,’’ taking appropriate account ofthe risk of investment, and then subtractingthe initial cost of the endeavor. Future cashflows are discounted because a dollar todayis worth more than a dollar to be receivedin two, three, or several years hence (sincethe dollar today can be invested in a financialinstrument and earn a rate of interest).

According to standard practice, it makeseconomic sense to undertake a capital projectonly if its NPV is positive (the discountedreturns are greater than the project’s cost),and even then a firm may decide not toproceed. 15 For example, if the discount rateis 10 percent, a project costing $1 millionbut projected to generate net revenues of$200,000 annually for ten years, would havea NPV of $229,000. But if annual net revenuesare projected to be only $100,000 over thesame time period, the project should notbe pursued because its NPV is a negative$386,000 (which doesn’t even cover theproject’s cost).

Passing the NPV test, however, does notmean that a project will be authorized. Afirm may have many potential projects thatlook promising when judged by their NPVs;

however, it might not pursue all of thembecause it may have strategic objectives thatcannot be readily quantified which limit therange of investments it can undertake. Thefirm may also be reluctant for other reasonsto seek outside financing (preferring to under-take only those projects that can be financedwith cash on hand), or to limit its borrowingor sale of equity.

Third, regardless of which of these ap-proaches (or others) private firms may employto decide how much capital investment toundertake and which projects to pursue,all of them ultimately measure the probablesuccess of the projects by a single metric—the likely effect on future financial perform-ance. Moreover, the process of evaluatingthese undertakings is different from thatof deciding whether to make certain expendi-tures for operating purposes (the expensesnecessary to keep the business running ona day-to-day basis). These decisions do notrequire long-run projections of impacts ordiscounting into the future, although tech-niques such as calculating NPVs are oftenused to decide whether to terminate existinglines of activity. Accordingly, operating budg-ets are often prepared and overseen in theprivate sector through a process that isseparate from the capital budget (althoughboth processes are often linked by an overallmanagement plan). 16

Finally, a firm’s decision to undertake oneor more capital projects is not necessarilylinked with a decision about how to financethose projects. Some firms, averse or unableto take on additional debt, may financeall, most, or part of their capital projectswith cash on hand; others may borrow;and still others may sell equity. But justbecause capital spending may require a sepa-rate decision and budget, it need not befinanced to any degree with additional debt.

Capital Budgeting by State and LocalGovernments

Just as there is no single capital budgetingpractice prevalent in the private sector, theapproach to capital budgets also varies amongstate and local governments. Nonetheless,some general tendencies are worth noting. 17

19BUDGETING CAPITAL

First, most state governments maintaina capital budget separate from the operatingbudget. However, states differ substantiallyin how they define capital, the degree towhich capital is separate in the governor’sproposed budget and in the legislature’s budg-et, and the means by which they financecapital expenditures. 18

Second, whether or not states budget capitalspending separately from other expenditures,most states have long-range capital plans,ranging from three to ten years, with fiveyears being the most frequent planning hori-zon. The spending figures in these planstend not to be as detailed as the figuresincluded in the annual budgets.

Third, available survey evidence indicatesthat the states most satisfied with theircapital budgeting process use some methodof keeping their legislatures regularly in-formed about capital needs. Some state legisla-tures also have a separate committee chargedwith overseeing all or most capital projectsand their financing.

Fourth, unlike the private sector, wheredifferent capital projects can be judged bythe common standard of impact on profit-ability, governments are responsible for avariety of functions, including police protec-tion, health care, and education, whose bene-fits generally cannot be reduced to dollarsand cents. This is a common situation sharedby all levels of government. Nonetheless,governments must set priorities in decidinghow to spend tax revenues and any borrowedfunds.

How do state governments set prioritiesin deciding on their capital expenditures?Although some do it project-by-project, orcase-by-case, most states have formal mecha-nisms, either in statute or by practice, forsetting priorities. Many states that take thisapproach set priorities on a functional basis,allocating expenditures for higher education,transportation, aiding local governments, orprotecting natural resources. Others have stat-utes that give priorities to certain activities,such as health and safety.

Fifth, contrary to popular belief, state gov-ernments do not always finance their capitalprojects by borrowing. To the contrary, states

often dip into general revenues to pay forcapital items, although the extent to whichthey are allowed or choose to do so varies.Other major sources of revenue for statecapital spending include excise taxes (suchas taxes on gasoline) or grants from thefederal government. In addition, while debtservice—interest and repayment of principal—typically shows up in state operating budgets,no state budget includes charges for deprecia-tion. 19 Many states impose user fees onintended beneficiaries of capital projects inorder to help service the debt issued tofinance them.

Finally, most states have either constitu-tional or statutory limits (often with referen-dum requirements) on the amount of debtthey may issue. State borrowing is alsodisciplined by the market. Rating agenciesdetermine the ratings they give to a state’sbonds, which strongly influence the interestrate at which those bonds can be marketed.These ratings are set in significant partby measuring the amount of state debt out-standing against the economic output gen-erated in the state. Higher interest ratesdue to adverse ratings can force states tolimit their borrowing.

As a broad generalization, local governmentsfollow procedures and conventions similarto those outlined for state governments.

Current Budgeting by the FederalGovernment

It may be surprising to some that through-out much of American history, the federalgovernment had no central budget. Untilthe Budget and Accounting Act of 1921,which created the Bureau of the Budget,each individual agency submitted a budgetto Congress. Since 1921, the Bureau of theBudget (now OMB) has coordinated the prepa-ration and submission of a Presidential budgetfor the entire executive branch. The Presidentis required to submit the budget for thecoming fiscal year by the first Monday inFebruary. This gives Congress eight monthsto enact the legislation that will continuethe operation of most government operationsand programs. If the necessary appropriationslaws have not been enacted by October 1,temporary ‘‘continuing resolutions’’ usually


provide funds until full-year appropriationsare enacted.

Although the Congress considers the Presi-dent’s budget proposals, it usually does notactually pass a law setting forth a budget(although, as discussed below, the ‘‘budgetresolution’’ passed by Congress establishesa framework for later Congressional consider-ation of different pieces of the budget). Instead,it enacts thirteen separate appropriations billsfor the approximately one-third of all federalspending that is deemed to be ‘‘discretionary.’’The thirteen appropriations bills are developedfor full Congressional consideration by thesame number of subcommittees of the Appro-priations Committees of each chamber.

The other two-thirds of the budget coversso-called ‘‘mandatory spending,’’ which ismainly for entitlement programs such asSocial Security, Medicare, Medicaid, and un-employment insurance. Mandatory spendingcontinues at levels regulated by standinglaws unless Congress enacts legislation tochange them (for example, by changing abenefit formula). The same is true of taxreceipts. Congress assigns responsibility forlegislation governing mandatory spending andreceipts to the authorizing (rather than appro-priations) committees.

Until the Congressional Budget Act of 1974,Congress had no procedures for coordinatinglegislation governing appropriations, manda-tory spending, and revenues into an overallfiscal policy. Instead, a fiscal policy simplyemerged as the sum of all of the enactedbills. The 1974 Act aimed at bringing moreorder to the budget process by creatingseparate budget committees in both the Houseand the Senate, and the Congressional BudgetOffice (the congressional counterpart to OMB),which provides information to Congress aboutthe costs and effects of legislation. In addition,the Act requires Congress first to decidewhat the projected budget surplus or deficitshould be and then to be guided by thatdecision in enacting spending and revenuebills.

More specifically, the 1974 Act calls forCongress to adopt each year a ‘‘budget resolu-tion’’ that sets a ceiling on total outlaysand a floor on total receipts. The resolution,which is not presented to the President

because it is technically not a law, alsoallocates ‘‘budget authority’’ and ‘‘outlays,’’by functional categories, to the appropriationscommittees (for discretionary spending) andthe authorizing committees (for mandatoryspending). The appropriations committees, inturn, further allocate budget authority amongtheir thirteen subcommittees, which mustreport bills back to the full committee consist-ent with those allocations. The resolutionmay also direct authorizing committees toachieve a specified amount of savings byreducing mandatory spending or increasingreceipts. Finally, the 1974 Act establishedparliamentary rules (‘‘super-majority’’ votingrequirements in the Senate) to stop billsthat violate the budget resolution.

The distinction between ‘‘budget authority’’and ‘‘outlays’’ is fundamental to understandingthe way budget decisions are actually made.Congress grants budget authority (BA), ena-bling agencies to incur obligations. Thoseobligations, in turn, require outlays (actualcash payments). Capital expenditures andoperating expenses typically have very dif-ferent ‘‘outlay rates.’’ Capital projects areoften completed over several years, so theoutlays for them are spread out over someperiod of time. In contrast, the outlays forsuch things as salaries of government workers,repairs, and maintenance, along with pay-ments under the various entitlement pro-grams, typically coincide with the amountof BA for the same year.

The Budget Enforcement Act of 1990 addedfurther requirements to the budget processfor fiscal years 1991–95. The BEA has beenextended twice so that its requirements nowapply (with amendments) through fiscal year2002:

• The BEA divided all discretionary spend-ing, of which capital spending is a part,into categories and imposed statutory lim-its or ‘‘caps’’ on each category (on bothBA and outlays). The categories changefrom year to year, but currently consistof defense, non-defense, violent crime re-duction, highways, and mass transit. Theseparate caps for defense and non-defenseare replaced after fiscal year 1999 by a‘‘discretionary spending’’ category, whilethe other categories remain intact. The

21BUDGETING CAPITAL

violent crime reduction category expiresafter fiscal year 2000, leaving the discre-tionary, highways, and mass transit cat-egories. Increases in taxes do not increasespending allowed by the caps (althoughthe BEA rules allow discretionary spend-ing to be offset by fees charged for goodsand services when the fees are authorizedin appropriations acts). The caps were in-tended to restrain the growth of spending,whether or not additional tax revenues formore spending could be found.

• The BEA contained ‘‘pay-as-you-go’’(PAYGO) provisions to ensure that the cu-mulative impact of changes in legislationaffecting mandatory spending or receiptsdo not increase the deficit. In other words,any increases in mandatory benefits mustbe financed either by cuts in other manda-tory spending or by increased revenue. Be-cause most capital expenditures are dis-cretionary, the PAYGO rules seldom applyto capital spending.

The federal budget contains several typesof funds. The ‘‘general fund’’ is the broadestand includes income and some excise taxreceipts. It also includes proceeds of generalborrowing, on the revenue side of the budget;on the expense side, it includes nationaldefense, interest on the federal debt, operatingexpenses of most federal agencies, and somecapital expenditures (broadly defined) on R&D,education, and infrastructure and other phys-ical capital spending. ‘‘Special funds’’ areearmarked for specific purposes; while theyare not designated by law as ‘‘trust funds,’’they do not differ from them in substance. 20

Most special funds are financed by userfees. ‘‘Trust funds’’ also have dedicated uses,and are financed by user fees or taxes;when their surpluses are borrowed, the fundsreceive interest. A few of the best-knowntrust funds are those for Social Security,Medicare, and highways (although there areabout 150 such trust funds in total). 21

Although each of the trust funds is tech-nically distinct, they are reported on a com-bined basis in a ‘‘unified budget,’’ a conceptadopted in January 1968 (for the FY 1969Budget). The unified budget provides thebottom-line impact of all federal spendingand taxing on the economy by indicating—

through the cash deficit or surplus—the im-pact on credit markets.

The unified budget also consolidates bothoperating and capital expenditures, whichmeans that the federal government does nothave a separate budget for capital expendi-tures. The receipts and outlays shown inthe unified budget are similar to a cashflow statement in the private sector, whichalso provides a comprehensive accounting ofincome and spending.

There have been several efforts since WorldWar II to address the question of whetherbudget procedures should be changed to pro-vide for separate consideration of capitaland operating expenditures. 22 For example,a capital budget was incorporated in theTaft-Radcliffe amendment to the EmploymentAct of 1945, which was passed by the Senatebut rejected in the House. The 1949 HooverCommission did not recommend a separatecapital budget, but it did suggest that thegovernment publish budget estimates for cur-rent operating expenditures and capital out-lays separately under each major functionor activity in the budget.

There were periodic attempts in Congressduring the subsequent two decades to adopta capital budget, but these were often opposedby the executive branch and never resultedin legislation. The capital budget was firmlyrejected in 1967 by the President’s Commissionon Budget Concepts, as it was in previousstudies by the American Institute of CertifiedPublic Accountants and the U.S. Chamberof Commerce. Interest in the idea returnedin the 1980s with the apparent approvalof Comptroller General Charles Bowsher andthe suggestion by President Reagan in 1986that the idea be studied. Interest in capitalbudgeting surfaced again during Congressionaldeliberations in 1995–96 over the proposedBalanced Budget Amendment (BBA) to theConstitution. Some of the proponents of theBBA wanted the amendment applied onlyto operating expenses of the federal govern-ment, excluding some defined capital thatcould be financed by government debt.

The federal budget process today continuesto budget operating and capital expenditurestogether. 23 During the course of its delibera-tions, the commission heard several expla-


nations of why this is the case (althoughnot all commissioners agree with each ofthem).

First, for reasons already discussed, federalpolicy makers have not been able to agreeon a single definition of capital or investmentin the public sector. While a technical analysisthat accompanies the budget (today it isknown as Analytical Perspectives) has useda stable definition of investment for manyyears, the use of the term investment inthe budget to describe policy proposals haschanged with the political priorities of dif-ferent administrations. 24 Given the changingpriorities of the Congress and different admin-istrations through time, it is not surprisingthat no single definition of public capitalhas emerged.

Second, capital is one of a number ofinputs (along with materials and labor) thatthe federal government uses to deliver itsservices (directly or through state and locallevels of government) to the public. Thepublic, in turn, judges the government notby the inputs it uses, but by the amountand perceived quality of the output it delivers.On this view, budget decisions should focuson the goals to be achieved (such as providingeducation or securing the national defense),and not on the mix between capital andother inputs judged necessary to achievethem.

Third, although there is no necessary con-nection between capital spending and itsfinancing—indeed, many states, localities, andother authorities have clearly defined capitalbudgets without financing all capital throughborrowing—there have been fears that a‘‘capital budget’’ would allow what is calledcapital to be debt-financed (in large partor in the entirety). Those who believe theseconcerns are justified also fear that adoptionof a capital budget could create a strongtemptation for policy makers to classify awide range of expenditures as capital orinvestment (1) to avoid having to pay forthem out of tax receipts or (2) to avoidhaving them subject to caps on discretionaryspending. This is especially true for highvisibility projects for which there are clear,short-term political benefits to elected officials

in both branches of government who advocatethem.

The fears about excessive spending areof special concern: while it is true thatthe federal government cannot borrow withoutlimit, federal borrowing is far less constrainedby financial markets than is the case forborrowing by private firms and state andlocal governments. Investors understand thatpeople and capital can easily move to otherlocales if state or local taxes are consideredto be too high. This limits the ability ofstates and localities to borrow. Simply put,the added taxes that are required to servicetheir debts could cause individuals or compa-nies to move to other areas if they believethat the additional services are not worththe higher taxes. 25 By contrast, individualsand corporations in this country are farless likely to move to other countries inresponse to changes in taxes here. Further-more, investors also understand that thereis a buyer of last resort for federal debt—the Federal Reserve, which regularly addsto the money supply by buying Treasurysecurities.

Capital Budgeting in Other Countries

The national governments of very few otherindustrialized countries currently have a cap-ital budget. At one time, Sweden, Denmark,and the Netherlands engaged in the practice,but all have since abandoned it. However,New Zealand and more recently the UnitedKingdom have adopted different versions ofa capital budget for decision-making purposes.

In 1988, New Zealand’s national governmentintroduced a capital budget for government-owned fixed assets. Spending on these itemsis separately budgeted and not shown onthe government’s operating budget, whichis compiled under the accrual method ofaccounting. Depreciation of government capitalis reflected on the operating statement, analo-gous to the way it would be accountedfor in a private business in the incomestatement. Nonetheless, the full cost of capitalassets must be appropriated in advance. 26

In June 1998, the United Kingdom an-nounced an even bolder capital budgetinginitiative. Under this approach, the Britishgovernment has established for a three-year

23BUDGETING CAPITAL

period a budget for all physical investmentand grants in support of capital spending.A two-part financing rule has been announcedto accompany the budget: (1) the ‘‘goldenrule’’ under which the government will borrowonly to invest (and not to support currentspending), averaged over the economic cycle;and (2) a limitation on borrowing to ensurethat the public debt-to-national income ratiois stable over the economic cycle. The newsystem was adopted with the explicit intentionof encouraging more spending on public cap-ital, raising net public investment as a shareof GDP from 0.75 percent to 1.5 percent[Brown, 1998, p. 6].

It is too early to judge the results fromeither of these initiatives. Still, at least

three features of the governmental systemsin both countries are noteworthy. First, neithergovernment counts expenditures on educationand R&D—part of what we have labeled‘‘national capital’’—as capital for budgetingpurposes. Second, the governments in bothNew Zealand and the United Kingdom operatewithin a parliamentary system under whichthe party controlling the executive branchalso controls the majority in the Parliament.Accordingly, the proposed budget of the execu-tive branch is expected to be adopted intolaw, unlike in this country. Third, agencyheads in both New Zealand and the UnitedKingdom have greater authority to managetheir operations, with incentive-based pay,than do their counterparts in the UnitedStates.

25

DO CURRENT BUDGET CONVENTIONSDISTORT DECISIONS ABOUT FEDERAL

CAPITAL SPENDING?A central question the commission has

addressed is to what extent, if any, doesthe current federal budget process lead toless-than-ideal decision-making about capitalspending? We answer this question in twoparts: whether and to what extent the currentprocess leads to a bias one way or anotherin (1) capital spending in the aggregate,and thus relative to other types of spending(the possible ‘‘macro’’ bias), and (2) the alloca-tion of capital spending among differentprojects and activities, including maintenanceof existing capital assets (possible ‘‘micro’’biases).

Is There a Macro Bias?

The commission reviewed evidence andheard testimony suggesting that the currentbudget system has important biases in bothdirections with respect to capital spending—no matter how the term is defined. It isimpossible to know which biases predominate,however, without first having an objectivestandard of what level of aggregate spendingis optimal.

It is difficult enough for a private firmto calculate its ideal level of capital spending,taking account of expected future profitabilityand the riskiness of the investments. Butcalculating an ideal level of capital spendingfor the government is far more complicated.Since the government is not a private firm,its activities cannot be judged by the profit-ability standard often used in the privatesector. Instead, government has many differentobjectives that are not easily compared, suchas influencing the distribution of resourcesamong different geographic regions and incomegroups, ensuring national security, protectingthe environment, and facilitating economicgrowth. In principle, it might be possibleto calculate and even budget an ideal amountof capital spending for one of these purposes;but the commission has found nothing thatprovides a supportable and objective way

of specifying an ideal level of all capitalspending under any definition. For this reason,the commission does not believe that anyonecan say authoritatively whether the existingbudget process has a ‘‘macro’’ bias towardtoo much or too little total spending oncapital.

Even so, it may be interesting to knowwhether recent changes in budget conventionshave caused capital spending totals to moveeither up or down without specifying whethersuch changes may be desirable. For example,what effect, if any, have the caps on discre-tionary spending that have been in placesince fiscal year 1991 had on capital spending?In particular, have caps crowded out capitalprojects?

To investigate this question, the commissionexamined multi-year averages for spendingof different types as a share of GDP, bothbefore and after 1990. Table 3 presentsthe results.

The table shows essentially no differencein spending-to-GDP ratios in each of thefour categories displayed, including overalldiscretionary spending, between the five yearspreceding the introduction of the caps andthe succeeding years. It is true that thespending ratios for both periods are substan-tially below the levels in years before 1985,especially the 1970s; but with the exceptionof direct physical capital (whose spendingas a share of GDP dropped in the 1970s),the declines in the spending ratios occurredin the 1980s during the Reagan Administra-tion, before the caps were enacted.

It is impossible to know what capitalspending (or, for that matter, overall discre-tionary spending) would have been in theabsence of the caps, so we cannot statewith certainty that the caps had no constrain-ing impact on capital spending. But Table3 demonstrates that if the caps have sup-


Table 3. FEDERAL NONDEFENSE INVESTMENT ANDDISCRETIONARY OUTLAYS AS A PERCENTAGE OF GDP

Nondefense InvestmentNon-

defenseDiscre-tionary

Physical Capital

Direct Grants R&D Edu-cation

1962–69 .................................................... 0.39 0.65 0.84 0.46 3.861970–79 .................................................... 0.29 0.75 0.60 0.93 4.491980–84 .................................................... 0.27 0.68 0.48 0.77 4.531985–90 .................................................... 0.28 0.53 0.38 0.56 3.621991–97 .................................................... 0.28 0.51 0.41 0.59 3.69

Source: OMB

i Comment of Commissioner Stein: I believe the critical issuehere is whether the outlay rate is slower than the benefit rate—to the decision-maker or to the country.

j Comment of Commissioner Levy: A multi-year outlay period forcapital can at best lessen the bias against capital spending, but Icannot see how it could exert a bias in favor of capital spending,as seems to be implied by the text. When the outlay caps are thebinding constraint, Congress may indeed ‘‘spend more than it oth-erwise would’’ if budget authority were binding. However, that doesnot imply that Congress would spend as much as it would with aclear, long-term perspective. As long as the life of the purchasedcapital is longer than the period over which its purchase outlaysare scored, then the scoring system is biased against the purchaseof such an asset. I have great difficulty imagining many examplesof government capital for which the length of the outlay period isas long as—not to mention longer than—the life of that capital.

pressed capital spending they probably havedone so to no greater extent than theyhave for discretionary spending in the aggre-gate.

One feature of the current federal budgetprocess—the general practice of having thefull cost of all capital acquisitions appropriatedby Congress before any portion of the acquisi-tion can be made or the project started—has been alleged to act as a bias againstpublic capital investment, specifically govern-ment-owned capital. 27 The commission be-lieves, however, that full funding is importantbecause it ensures that policy makers considerthe total costs of an initiative before authoriz-ing and appropriating the funds for it. Other-wise, policy makers would be tempted tofund only a portion of a capital projectin the initial years, which means it wouldbe too far along to stop later. We discussbelow how failure to fully fund projectsin the past has produced substantial waste.

Nonetheless, it is possible that decision-makers defer some necessary, but large, cap-ital projects because funding them requiresauthorized spending to ‘‘spike’’ in a givenyear. To the extent this occurs, aggregatepublic investment may fall short of someideal figure.

How serious a problem this actually turnsout to be, however, depends to a significantdegree on whether spending is more con-strained in any year by the caps on budgetauthority or on outlays. As it turns out,the caps on budget authority (BA) seldomhave constrained spending. Instead, in mostyears since the BEA was enacted, the outlaycaps have been reached first. As already

noted, capital projects also tend to havelow outlay rates—that is, they spend outtheir budget authority over several years.When the outlay caps under the BEA arethe binding constraint, the slower outlayrates for capital projects could induce Congressto spend more than it otherwise would onpublic capital. This is because operating ex-penses, including maintenance, tend to spendout quickly, and thus get scored as outlaysin the forthcoming budget year. i Of course,there are projects so large that even ifthe outlays are spread over several years,the annual outlay is still a ‘‘spike’’ andspending could be constrained if the outlaycaps are binding. j

Efforts to get around budget spikes, mean-while, produce distortions of their own. Asjust noted, agencies can be tempted to use‘‘camel’s nose under the tent’’ budget tacticsthat have led to inefficient outcomes. Another,potentially wasteful budget maneuver foravoiding spikes is for agencies (sometimeswith Congressional blessing) to enter intoshort-term leases rather than to constructor purchase property at the outset—even

27DO CURRENT BUDGET CONVENTIONS DISTORT DECISIONS ABOUT FEDERAL CAPITAL SPENDING?

when the life-cycle cost of the purchasewould be lower than the cost of stringingtogether a series of short-term leases. Bothof these ‘‘tricks’’ demonstrate that seeminglyarcane scoring rules can have a real impacton budget decisions.

Are There Micro Biases?

Although it may not be possible to determinewhether current budgeting procedures havecaused a sub-optimal amount of total capitalspending, there is much greater reason tobelieve that the current system generatesbiases at the micro level: that is, capitalspending is allocated among capital projectsand initiatives, including the maintenanceof existing capital assets, in a less-than-ideal fashion.

The Congressional Budget Office has re-viewed the available studies of the measuredeconomic returns from different activities,finding a very large variation—from programsthat have produced estimated social returnswell in excess of the cost of capital, tothose that are producing almost no positivereturns. 28 Significantly, the CBO cites evi-dence indicating that maintenance can paysocial dividends well in excess of the returnsrealized on some large new projects [CBO].

The commission recognizes that budgetingis not a mechanistic exercise solely in searchof initiatives with the highest economic re-turns. 29 But in deciding how much attentionto pay to efficiency and how much to distribu-tional objectives, policy makers must workwithin a structured framework that (1) con-fronts them with the implications of therelevant tradeoffs and (2) provides maximumincentives for producing cost-effective deci-sions. Of particular interest to the commissionis the need for federal decision-makers totake adequate account of the interests ofAmerican society over the long run. Thecommission has concluded, however, that inseveral respects, the current budget processimpedes the ability of decision-makers toachieve these important objectives.

To understand the basis for this conclusion,we first briefly review the key phases ofthe current federal budget cycle, and thendiscuss its shortcomings.

Phases of the Current Budget Cycle

The ‘‘budget process’’ of any organizationis usefully understood as the combinationof four important, separate functions: planningand analysis, which leads to budget rec-ommendations; the making of budget decisions;accounting and reporting of the results; andevaluation of the outcomes of budget decisionsand subsequent readjustment in programs,where appropriate. We have already describedthe legal process by which budget decisionsare made. At the risk of some over-simplifica-tion, here are some key features that explainhow the federal government carries out theother three functions.

The process begins generally 18 monthsin advance of each fiscal year at the agencylevel, when individual departments and agen-cies develop internally the budget requeststhey will make to the President (initiallythrough OMB) for that fiscal year. Untilrelatively recently, with few exceptions, agen-cies focused their budget plans only on asingle year and generally paid little attentionto their long-run plans. This changed tosome extent with the enactment of the Govern-ment Performance and Results Act of 1993(GPRA), which requires agencies to submitfive-year strategic plans to OMB every threeyears. The first such plan was submittedin 1997, the next one is due in 2000.

For the most part, the strategic plansare descriptive in nature and do not containout-year spending/revenue projections. None-theless, the agencies separately provide toOMB their spending and revenue projectionsfive years out under presidential policy. OMBuses these projections to present in thePresident’s annual budget five-year projectionsof revenue, by major source, and outlaysin aggregated form and at the function andprogram level (OMB’s data base includesprojections at the ‘‘account’’ level beyondthe budget year, but these are not shownin the budget).

The GPRA requires agencies to submitperformance plans to OMB and the Congresseach year. The Act also requires OMB toprepare a government-wide plan. These plans,the first of which was submitted with thePresident’s budget for FY 1999, are supposedto lay out the agencies’ goals in objective,


quantifiable terms (such as the airplane acci-dent rate for the Federal Aviation Administra-tion) for that budget year.

With respect to capital projects in particular,OMB’s Capital Programming Guide requiresagencies to analyze their life-cycle costs andbenefits as part of any request for fundingof planned projects. Once budget decisionsare made, the results are reflected in annualreports issued by both OMB and CBO display-ing the agencies’ current and historical spend-ing patterns.

Agencies also prepare balance sheets thatreport their assets and liabilities. The ChiefFinancial Officers Act of 1990 required allcabinet departments, major independent agen-cies and the government as a whole tohave audited financial statements. These fi-nancial statements are prepared in accordancewith federal accounting standards developedby the Federal Accounting Standards AdvisoryBoard (FASAB). 30 Of particular interest tothis commission, these standards require thefinancial statements to disclose in footnoteform estimates of deferred maintenance, effec-tive with the statements for fiscal 1998.In his fiscal year 1999 budget, the Presidentset a goal of having an unqualified opinionon the consolidated (government-wide) finan-cial statements for that year. Furthermore,twenty of the twenty-four agencies underthis Act are committed to obtaining unquali-fied opinions on their own statements inthe same time frame [OMB and CFO Council,1998].

Various mechanisms are in place for evalu-ating the outcomes and ongoing progressof federal programs. The agencies typicallyhave evaluation efforts under way. Congressperiodically asks the General Accounting Of-fice to prepare independent evaluations. None-theless, no ongoing systematic, government-wide evaluation process is in place, whetherfor capital spending (however defined) orother types of spending.

Shortcomings of the Current Process

As reflected in the foregoing summary,a number of significant improvements havebeen made in recent years in certain stagesof the federal budget process. Even so, thecommission has concluded that the existing

process, at each of its various stages, stillcontains a number of important shortcomings.A broad theme that ties the various flawstogether is that the federal government—both the executive and legislative branchesconsidered together—is so heavily focusedon each current budget year that too littleattention is paid to longer-run matters. Fur-thermore, policy makers are not held suffi-ciently accountable for the longer-run implica-tions of their current decisions. This showsup in part in wasteful spending on somecapital projects, a shortchanging of mainte-nance of existing assets, and perhaps somemissed opportunities (which are inherentlydifficult to measure, but nonetheless real).

• While the strategic plans and performancegoals required of agencies under the GPRAare a major step forward, a number of im-portant defects remain at the planningstage of the process:

—Updated long-term strategic plans arenot required of the agencies annually, norare they integrated with the five-yearbudget plans submitted by the agencies.Furthermore, because there is no system-atic format for the strategic plans, theymake it too easy for agencies simply tojustify their missions rather than providetrue forward-looking plans for achievinglonger-term results-oriented objectives ina cost-effective manner.

—There is uneven progress among theagencies in stating goals and missions inperformance plans, as required under theGPRA.

—Insufficient attention is paid to benefit-cost analyses of capital spending initia-tives in particular, both before and afterthey are proposed. The analytical require-ments set forth in OMB’s Capital Pro-gramming Guide apply only to govern-ment-owned assets, and not to the broadertypes of assets that belong in any defini-tion of national capital (at a minimum,infrastructure spending, R&D, and edu-cation and training). Many agencies andOMB lack the resources to design and con-duct benefit-cost analyses, while Congresspays what the commission believes to beinsufficient attention to such analyses inits oversight, authorizing, appropriations,

29DO CURRENT BUDGET CONVENTIONS DISTORT DECISIONS ABOUT FEDERAL CAPITAL SPENDING?

and budget resolution activities. Further-more, GAO and CBO do not have enoughanalytical resources to review the work ofthe agencies, and thus to assist Congressin assessing the merits of capital spendingproposals.

• Because budget decision-making is inher-ently a political process, it is likely thata bias exists favoring projects with highlocal visibility and a concentrated impacton employment (such as roads, buildings,and waterways) and against those that areless visible and have a more diffuse impacton employment (such as computers for theInternal Revenue Service or the Social Se-curity Administration). Although thisproblem can never be fully overcome, itcan and should be mitigated by a commit-ment by both the executive and legislativebranches (1) to sound analysis before ap-proving new projects and (2) to supportingevaluations of the impact of those initia-tives after they have been undertaken (seebelow).

• As discussed above, failure to fully fundcapital projects in advance can lead towasteful spending. For example, the Gen-eral Accounting Office has found that in-cremental funding practices have led tosubstantial cost overruns, schedule slip-pages, and terminations in the Depart-ment of Energy’s major acquisitions [GAO,1996]. The associated waste in expendi-tures has been substantial: the four can-celed projects since 1983—the Super-conducting Supercollider being the primeexample—cost $8 billion before they wereterminated.

• Several aspects of the trust funds estab-lished to support certain types of capitalspending—such as the construction ofhighways, airports, and water projects—are problematic. As a threshold matter,the existence of the capital-related trustfunds themselves insulates the programsthey support from the annual balancingof priorities across the government. At thesame time, the commission recognizesthat, in principle, the trust fund devicemay be justified (1) where the revenuesgoing into them represent charges or feeson users of the services they support and

(2) the earmarked fees and taxes are spenton the purposes for which the funds werecreated.

The tendency toward surplus in some trustfunds has become a problem under currentscoring rules. Specifically, these rules treatrevenues going into the trust funds on themandatory side of the budget, but classifythe spending out of the trust funds asdiscretionary spending and thus subject tocaps. Congress and the administration tooka major step toward rectifying the imbalancein the highway trust fund generated bythis difference in scoring with the enactmentof the Transportation Equity Act for the21st Century in 1998. This legislation createsseparate BEA caps for highway and masstransit spending, and it sets the caps equalto the receipts from motor fuels taxes collectedthe previous year. 31 The commission doesnot endorse the specific spending formulain this act as a model for other trustfunds; however, it does believe that theprinciple of tying spending out of the capital-related trust funds to the tax and fee revenuethat flows into them, averaged over somereasonable time period, is a good one tofollow.

The current budget decision-making processalso exerts biases against both routine andmajor maintenance, such as rehabilitationand remodeling (which represents a differenttype of capital expenditure). As already noted,the presence of the outlay caps feeds sucha bias because the budget authority forboth types of maintenance has associatedwith it a more rapid outlay rate than budgetauthority for new construction. In addition,there currently is no mechanism assuringthat state and local governments receivingfederal support for new capital projects ade-quately maintain those assets, once theyhave been constructed or acquired (nor dorating agencies generally allow maintenanceto be bonded). This can defer maintenance,in turn leading to excessive funding fornew assets when it may be more cost-effective to maintain existing assets.

The shortchanging of maintenance is aggra-vated by the lack of accurate and timelyinformation on the condition of federal andfederally funded assets. Granted, recently


adopted federal financial accounting standardsrequire the audited financial statements ofthe agencies to be accompanied by footnotesdisclosing the extent of deferred maintenance;yet footnote disclosure is not a substitutefor a more complete and detailed reporton the actual condition of federally ownedassets. In addition, the federal government’sfinancial statements do not contain informa-tion on the condition of assets at the stateand local levels, some of which the federalgovernment has funded. 32 Information aboutthe current condition and even obsolescenceof assets is critical if policy makers areto design effective maintenance and capitalspending programs.

The commission cannot stress too stronglythe importance of having reliable estimatesof deferred maintenance. Currently, thereis no generally accepted method for agenciesto use in estimating deferred maintenance.This is a significant shortcoming since soundpolicy making requires having accurate infor-mation of deferred maintenance in settingspending priorities and in deciding whetherto purchase new assets or fix existing ones.This shortcoming has led the FASAB topropose an amendment to its current stand-ards that would relax the audit requirementfor the information reported on deferred main-

tenance. In conjunction with this change,OMB is planning to organize a task forceto develop methods for making consistent,government-wide estimates of deferred mainte-nance, which should enable these estimatesto be fully audited. Still, until better andmore-consistent information about the condi-tion of federally owned and financed assetsis routinely made available, policy makerswill be unable to make fully informed decisionsabout whether to fund new projects or putmore money toward maintaining existing as-sets.

Though efforts have been made to evaluatethe effectiveness of government programs,we believe there is still little systematicretrospective analysis within either branchof the federal government to determine wheth-er capital projects generated the benefitsand came within the cost projections thatwere originally promised.

In sum, we recognize that it is difficultto determine whether the existing budgetprocess produces insufficient or excessiveamounts of capital spending in the aggregate;however, there are several reasons for believ-ing that aspects of the process contributeto a sub-optimal allocation of capital spendingamong various projects while shortchangingmaintenance.

31

RECOMMENDATIONSThe commission considered a range of pro-

posals to address the problems that havejust been identified. We believe the appro-priate response is to make improvementsin each of the component parts of the budgetprocess. Many of the recommendations weoutline below relate to improvements in infor-mation, but others also entail changes inthe ways that budget decisions are actuallyconsidered and made.

Better Planning and Analysis

Long-range planning for all kinds of expend-itures and operations of the federal govern-ment is essential (1) to ensure that servicesare delivered to the public in the most-effective manner and (2) to allow policymakers to judge how much and what kindsof capital are needed to provide public serv-ices. 33 Given the difficulty of terminatingprograms and initiatives once begun, thepreparation and publication of long-run planscan help ensure that resources are wiselycommitted to new programs before they arelaunched, while facilitating ongoing readjust-ment in priorities when appropriate. Thecommission advances the following rec-ommendations to help improve this process.

Recommendation 1: Five-Year Strategic Plans

Although the GPRA made major stridesin requiring agencies to prepare five-yearplans, we have pointed to a number ofgaps in the existing planning process thatshould be filled.

First, the five-year plans should be preparedannually (not just every three years) andshould be integrated with the annual perform-ance plans. Furthermore, the plans shouldbe an integral part of the budget justificationssent to Congress.

Second, the plans should be reconciledwith the longer-run budget projections thatthe agencies already submit to OMB. Inparticular, the plans need to state results-oriented objectives—not just for the currentbudget year under current budget policy,

but ideally with respect to future projectedchanges in policy.

Third, the plans and annual budgets shouldbe tied to the life-cycles of the agencies’capital assets. The following elements of cap-ital planning are common in the privatesector and among state and local governments,and should be standard practice for thefederal government: a needs assessment forsuch additional capital assets; a realisticmaintenance schedule, funded appropriately;and recognized replacement cycles.

Fourth, OMB should develop standardizedformats for the plans (in consultation withGAO and CBO) so that policy makers inboth the executive and legislative branchescan more easily compare the plans of oneagency to another. Among other things, theplans should be less voluminous than manycurrently are, should record past successesin achieving defined results-oriented objec-tives, should identify shortcomings that needto be addressed, and should spot challengesthat remain to be tackled. The plans shouldalso identify major future outlays for physicalassets (segregated in a separate ‘‘capital acqui-sition fund,’’ as discussed below) in a levelof detail that OMB should specify.

Fifth, OMB should expand its efforts toevaluate the plans (together with benefit-cost analyses of major projects, as discussedbelow) and to consider them in connectionwith government-wide planning. Among otherthings, the plans should help identify pro-grams and efforts that are no longer needed,programs that might be better carried outby other federal agencies or other levelsof government, and new programs that maybe truly necessary. The results of this exerciseshould be considered in the preparation ofthe President’s annual budget.

Sixth, in considering agency appropriationrequests, the Congress should take accountof the agencies’ five-year plans and of OMB’sannual evaluations of those plans, as reflectedin the President’s budget. Congressional au-thorization, appropriations, budget resolution,


and oversight hearings should focus on theseplans and evaluations. Congress should alsostudy ways in which it might improve itsown procedures to give more weight to thelonger-run implications of its current yeardecisions and to issues with longer-run con-sequences. In undertaking this task, Congressmight find it useful to take advantage ofthe wide range of institutional expertise avail-able to it, including resources within theCongressional Budget Office, the General Ac-counting Office, and the Congressional Re-search Service.

Recommendation 2: Benefit-Cost Assessments

The benefits and costs (both expressedin monetary terms to the extent practical)of alternative options should be consideredbefore decisions are made. This principlehas been part of executive branch regulatoryrulemaking (for ‘‘major’’ rules) for over twodecades. It has recently been required offederal capital projects as well through OMB’sCapital Programming Guide.

The commission believes that several exten-sions beyond existing practice are warranted.First, the benefit-cost requirement should beextended beyond federally owned capital assetsto the broader array of undertakings associ-ated with a definition of national capital.To some extent, this is already done, althoughnot in a systematic fashion. Most agenciesfund evaluations of their programs. We aresuggesting that the evaluation process becomemore systematic and institutionalized. Policymakers should not wait for sporadic economicstudies of individual programs prepared byacademic scholars to appear in the professionalliterature. Instead, there should be an ongoingeffort within the government to analyze thebenefits and costs of all major programs—whether or not related to capital expendi-tures—so that they can be adjusted, refash-ioned, or eliminated, as appropriate. As apractical matter, it may be useful to beginby requiring benefit-cost analyses only for‘‘major’’ initiatives, such as those over acertain dollar threshold; later on, smallercapital projects and government programscould be analyzed in the same fashion.

Second, more resources within the agencies,OMB, CBO, and GAO, should be devoted

to carrying out this mission. Those resourcesshould also support OMB in its effort tobecome a clearinghouse for ‘‘best practices’’in evaluation techniques that the agenciescan and should draw upon in preparingtheir own analyses. Given the many billionsof dollars at stake each year, it wouldbe penny-wise and pound-foolish not to spendmillions of dollars for analysis to help producebetter information for decision-makers in bothbranches of government and for the public.(A related need is for the government toprovide a stronger commitment to improvingits base of statistical data on the entireeconomy. Some of this information is impor-tant in preparing benefit-cost and other analy-ses of various existing and proposed govern-ment programs.)

Third, working with the agencies, OMBshould periodically review the evaluation tech-niques they use and, where appropriate, pro-vide guidance to improve them.

Improving the Decision-Making Process

The commission believes that several meas-ures short of adopting a separate ‘‘capital’’budget could improve the quality of budgetpolicy decisions. These recommendations areset forth below.

Recommendation 3: Capital Acquisition Funds

As an experiment, the commission believesit would be useful for Congress and theexecutive branch to have one or more agencieswith capital-intensive operations establish aseparate ‘‘capital acquisition fund’’ (CAF) with-in their budgets that would receive appropria-tions for the construction and acquisitionof large capital projects. The CAFs woulduse that authority to borrow from the Treas-ury’s general fund and then charge operatingunits within the agency rents equal to thedebt service (interest and amortization) onthose projects. 34 In addition, the CAFs wouldacquire all existing capital assets of theagency so that all the costs of all suchcapital could be allocated within the agency.

To ensure uniform implementation of theproposal, OMB should issue guidance aboutwhat capital items belong in the CAFs,such as federal buildings and other largecapital purchases by the agencies. 35

33RECOMMENDATIONS

The main advantage of CAFs is that theyshould improve the process of planning andbudgeting within agencies. If units or divisionswithin agencies are charged the true costsof their space and other large capital items,they are likely to make more efficient useof those assets. CAFs could also help addressthe spike problem by smoothing out thebudget authority required for any large capitalprojects proposed by units within agencies.In principle, Congress could take this smooth-ing function one level higher by either formallyor informally budgeting CAFs across all ofthe agencies within the jurisdictions of eachof the thirteen appropriations subcommittees.However, there is still merit in having CAFsmanaged at the agency level to promoteaccountability. 36

If the CAF experiments realize the foregoingbenefits, the commission would urge thatCAFs be used for all agencies.

Clearly, the CAFs would not replace theGeneral Services Administration, which man-ages the Federal Buildings Fund (FBF), agovernment-wide revolving fund establishedin 1972. The FBF acquires office buildingsand rents space in them to federal agencies.The GSA can and does delegate its authorityto agencies to acquire their own office spaceunder some circumstances. In such cases,an agency would acquire its office spacethrough its CAF. Greater use of this delegationauthority would be appropriate if agenciescould demonstrate that the CAFs led themto improve their capital asset managementpractices. In addition, GSA would negotiatethe acquisition of space for multiple agenciesthat seek to co-locate in a single facility.

Recommendation 4: Full Funding for CapitalProjects

Full funding of capital projects encouragesdecision-makers to consider the life-cycle costsand benefits of projects before they are under-taken and to compare the funding requiredwith other governmental priorities. This prac-tice should be continued.

Nonetheless, large projects in particularmay produce funding spikes that may causethe postponement of such initiatives in favorof smaller, less cost-effective projects, or eventheir cancellation. This problem can be ad-

dressed, without sacrificing the principle offull funding, by providing advance appropria-tions for all useful and programmaticallyseparate segments of particular projects. Auseful segment is one in which the benefitsexceed the costs even if no further fundingis appropriated. For example, if the fullproject envisions acquisition of multiple air-craft, a useful segment would be the numberof aircraft for which benefits exceed costseven if no additional aircraft are ever author-ized.

The preparation of five-year plans by theagencies should also help remedy the spikeproblem by alerting OMB and the Congressto potential future funding needs for largeprojects. If policy makers become aware ofthese requirements, they might be able tobetter adjust their annual appropriations ac-cordingly.

Recommendation 5: Adhering to the ScoringRules for Leasing

In principle, the scoring rules in the BEAare designed to eliminate any bias thatpolicy makers might have in deciding whetherto acquire or lease capital assets used inthe delivery of government services. Theydo this by requiring the present value ofso-called capital leases—those that are thefunctional equivalent of a purchase—to bescored up front, as if they were purchases;in this way, policy makers can make accuratecomparisons between the two options anddecide which is the least expensive. Underthe current BEA rules, which are modeledafter private sector accounting standards, acapital lease is one in which (1) the leasetransfers ownership of the property by theend of the lease term, (2) the net presentvalue of the lease payments is at least90 percent of the fair value of the property,or (3) the term of the lease is at least75 percent of the expected life of the asset.

The current rules give the agencies andCongress an incentive to be creative. Specifi-cally, they can enter into a succession ofshorter-term leases that do not meet thequantitative criteria for defining a capitallease precisely, which means the full costof the lease does not need to be scoredup front. Although this is legal under the


k Comment of Commissioner Penner: I believe that a rule requir-ing the capitalization of all short-term lease payments should beadopted and that the estimation problems associated with such arule are no more severe than those encountered in estimating thecost of many credit programs.

l Comment of Commissioner Levy: The scoring of leases versuspurchases of capital assets should be addressed by Congress, eitherin isolation or as part of a comprehensive overhaul of the budgetprocess. For example, Congress might require any short-term leaseor building to be certified as the superior choice in the long run.I agree with Commissioner Penner that short-term leases shouldbe capitalized for purposes of comparing them with the cost of pur-chasing a capital asset, but I would like to emphasize that a cap-italized five-year lease cannot be compared with the price of abuilding that will last at least 30 years. Analysts should considerthe cost of leasing over 30 years, or else compare the options overfive years with the estimated market value of the purchased build-ing added back at the end of five years.

current rules, it can result in wasteful spend-ing when, computed appropriately on a presentvalue basis, the less expensive alternativeis to buy the asset.

In principle, this problem could be remediedby a rule that required the capitalizationof all short-term lease payments expectedin the future. To be effective, however, thisrule would require strict scrutiny of estimatesof future lease payments—something thatmay be difficult and expensive to do inits own right. k In addition, there is a riskthat any rule requiring the capitalizationof all leases could discourage the use ofshort-term leases that are highly cost-effective,such as when agencies are downsizing orbetween moves to different locations.

Though the commission believes that thebest course for now is to retain the existingBEA rule, both the agencies and the Congressshould strictly adhere to it. This shouldbe easier to do when agencies are preparingstrategic plans every year. These plans couldexpose the intentions of the agencies withrespect to capital assets in particular. Inturn, OMB and Congress would be ableto identify programs where purchase is moresuitable than leasing, as well as becomealert to possible spending spikes that couldbe smoothed by the other recommendationsalready outlined (CAFs and advance fundingfor useful, separate project segments). l

Recommendation 6: Trust Fund Reforms

Various trust funds—for highways, airports,the air traffic control system, water projects,and certain other purposes—have been createdwith the ostensible purpose of assuring the

funding of capital projects. The funds havebeen financed with fees or taxes assessedon those who use the facilities (such asthe gasoline tax to help support highwayconstruction and the airline passenger tickettax to help fund airport equipment andconstruction).

The commission believes two important re-forms of current trust funds are necessaryto make them more cost-effective.

First, averaged over some reasonable periodsuch as three years, the revenues from taxesand fees dedicated to the trust funds support-ing infrastructure or capital spending shouldbe spent for designated purposes: capitalspending and maintenance. OMB should high-light in either the budget or accompanyingdocuments the extent to which trust fundmonies are being spent for such purposes.If spending on the earmarked uses is notsufficient to exhaust the revenues over somereasonable period, then Congress should lowerthe specific taxes or fees so that the revenuethey raise is more in line with the spendingthey are intended to finance.

Second, state and local governments thatreceive federal support for capital items (suchas highways)—whether or not such supportis provided through a trust fund—shouldbe required to maintain assets financed bythe federal government as a condition ofreceiving any additional federal support. Theone possible exception to this general ruleis where state and local governments candemonstrate that the assets the federal gov-ernment initially funded are no longer needed(as could be the case with roads in ruralareas where the population has dwindled).Otherwise, the federal government risks fi-nancing new infrastructure that may be unnec-essary. States and localities seeking federalaid for capital projects should be requiredto certify that they have met the maintenancerequirement, and the relevant federal agenciesshould check these certifications. To the extentthat this maintenance requirement representsan ‘‘unfunded mandate,’’ the commission be-lieves it is one that could readily be justifiedas a mechanism to help ensure the efficiencyof spending at all levels of government onfederally supported capital projects. 37

35RECOMMENDATIONS

m Comment of Commissioners Lynn, Penner, and Stein: We donot believe that this four-way classification of expenditures wouldbe helpful in making good budgetary decisions.

Recommendation 7: Incentives for AssetManagement

In addition to improving the informationavailable to decision-makers and changingthe scoring rules, it is important that agencieshave financial incentives to manage theirassets efficiently. In the private sector, firmsclearly have such incentives; the better theymanage, the more money they are likelyto make.

Federal agencies operate under much tighterconstraints in managing their assets thanis the case in the private sector. Withfew exceptions, agencies cannot sell, exchange,or lease assets on their own. Instead, ifthey no longer have a use for certain property,they must report it as ‘‘excess’’ to the GeneralServices Administration. In turn, the GSAmust first offer it to other federal agencies;if no agency claims it, the property canthen be offered to state and local governmentsand various non-profit organizations.

The commission encourages the administra-tion and the Congress to expand the freedomof agencies to manage their assets and toconsider ways to give the agencies incentivesto do so efficiently. One possibility wouldbe to allow, on an experimental basis, oneor more agencies to keep a limited portionof the revenues they raise from selling orrenting out existing assets.

Better Information

The third stage in the budget processis the reporting of the results. The commissionrecommends two key improvements in thisarea.

Recommendation 8: Clarification of the FederalBudget Presentation

Policy makers must be cognizant of thecumulative impact of their many micro budgetdecisions when planning how much to spendon individual government programs or decid-ing to alter the tax code. In short, theyshouldn’t lose sight of the forest when plantingindividual trees. The forest should be plainlyvisible for the American people to see, inuser-friendly form.

One set of forest level figures, of course,includes the aggregate totals of spending

and revenue, and the resulting projecteddeficit or surplus in the unified budget.In recent years, the goal of a balancedbudget has been the guiding principle fordecision-making about the budget. In addition,given the strictures of the caps, policy makersnecessarily pay attention to the broad spend-ing breakdowns defined by the Budget En-forcement Act—namely the distinction betweenmandatory and discretionary spending andwithin the latter, the distinction betweendefense and non-defense spending.

The commission believes policy makers alsoshould pay attention to another set of broadcategories of spending: operating expenditures(defense and non-defense), investment spend-ing, transfer payments made to individuals,and interest on the federal debt. Apart fromthe fact that federal policy makers do notbudget depreciation, the separation of operat-ing and investment spending would be analo-gous to a similar division used in the privatesector and in most state and local govern-ments. The breakout of transfer paymentsto individuals is useful because of the federalgovernment’s deep involvement in this area,protecting individuals against financial lossesdue to unemployment, retirement, disability,and illness. Interest on the federal debtshould be reported separately because it isa financing expense rather than an operatingexpenditure.

Table 4 below is illustrative of the typeof information that should be highlightedin future budget presentations, with the notesbelow explaining how the figures displayedwere calculated. m The definition of investment,in particular, is a broad one, as it includesnot only spending on federal assets, butalso federal spending on education and R&D,as well as federal capital grants to statesand localities. Since the definitions of invest-ment spending in particular may vary fromadministration to administration, it wouldbe useful if something like Table 4 wereconstructed using alternative definitions ofinvestment. Also of use would be a chartshowing historical trends in spending in thedifferent categories, especially as a percentageof GDP, as well as projected future spending


Table 4. BREAKDOWN OF FEDERAL SPENDING BY BROADFUNCTIONAL CATEGORIES, FISCAL YEAR 1997

(outlays in billions of dollars)

OutlaysPercentof TotalOutlays

Percentof GDP

Investment ........................................................................................... 229 14 3Operating expenditures ....................................................................... 243 15 3Transfer payments to individuals ...................................................... 935 58 12Net interest .......................................................................................... 244 15 3Undistributed offsetting receipts ........................................................ –50 –3 –1

Total ..................................................................................................... 1,601 100 20

Notes: For purposes of this table, ‘‘Investment’’ includes all investments in physical capital (wheth-er or not owned by the federal government), in research and development, and in education andtraining. ‘‘Operating expenditures’’ are calculated by subtracting from total outlays all of the othercategories shown in Table 4. ‘‘Transfer payments to individuals’’ includes spending on such items asunemployment insurance and food stamps, Social Security (retirement and disability payments),Medicaid, and Medicare, but does not count student aid payments (which are included in Invest-ment). ‘‘Net interest’’ primarily includes interest paid on federal debt held by the public. ‘‘Undistrib-uted offsetting receipts’’ counts receipts from other parts of the budget for the employer share of em-ployee retirement, rents and royalties on the Outer Continental Shelf, and the sale of major assets.

Source: OMB, from Historical Tables, Fiscal Year 1999 Budget.

in the various categories. This should besupplemented with charts or tables showingexpected changes in the net capital stockfor major categories of physical assets.

For policy makers, the kind of informationjust described would highlight to what extentthe President proposes to invest for thefuture, to operate the federal government’svarious functions (excluding depreciation,which is not counted as spending undercurrent budget accounting concepts), and toarrange for transfers to qualifying individuals.It would also explain how spending for allthese activities may be constrained by theobligation to pay interest on the cumulativeamount of federal debt.

Recommendation 9: Financial StatementReporting

No sensible private firm would decidewhether to undertake a new investment,such as a new building or plant, withoutdetailed knowledge of the composition, condi-tion, and value of its existing facilities. Yetfor decades the federal government operatedthis way, without having an updated andaccurate inventory and report of the conditionof its own assets—let alone those of the

other levels of government to which it rou-tinely makes grants. Moreover, public policydebates about national priorities have notbeen as well informed as they should havebeen. Specifically there has been no easyway for the public, the media, or evenexpert analysts to evaluate such questionsas whether there is an ‘‘infrastructure deficit,’’or whether budget cuts to reduce the unifiedfederal budget deficit were achieved throughsensible economies or by neglecting improve-ments or additions to the preexisting publiccapital stock.

As discussed earlier, the CFO Act of 1990makes major strides in rectifying this situationby requiring individual federal agencies andthe government as a whole to issue auditedfinancial statements. Furthermore, work isplanned for developing standardized methodsfor estimating deferred maintenance. The com-mission strongly supports these efforts andencourages OMB to work with the agenciesto complete this task promptly.

One important consequence of the CFOAct is that the federal government nowpublishes consolidated financial statements.These share two important principles withprivate financial accounting practices that

37RECOMMENDATIONS

are essential to objective, consistent, andtrusted reporting: (1) the use of definitionsbased on independently determined accountingstandards (determined by FASAB), which aredesigned to be insulated from the politicalprocess; and (2) the independent auditingof the financial data, which helps assurethe public that the information is not manipu-lated to achieve political ends.

Still, more should be done.

First, the calculation of depreciation invarious government reports should be stand-ardized. Currently, depreciation of capitalitems reported in the Analytical Perspectivesvolume of the President’s budget is computedwith reference to the replacement cost ofthe assets, whereas depreciation reported onfinancial statements is based on historicalcosts of assets. This kind of inconsistencyshould be eliminated so that depreciationis reported consistently in all governmentfinancial reports.

Second, the agencies should make theiraudited financial statements, together withdetailed breakdowns of assets and their condi-tion, widely available in printed form andthrough publication on their websites. Thefinancial statements should continue to beprepared on the basis of independently devel-oped accounting standards.

Third, this information should also be con-solidated at the government-wide level, eitherby OMB or GAO. The resulting aggregatereport, with appropriate detailed breakdownsby agency and type of investment, shouldalso be audited and published in writtenand electronic form.

The annual audited statements, togetherwith the detailed breakdowns on the conditionof federally owned assets, will be valuabletools for the agencies in preparing theirlonger-term strategies, for preparation of thePresident’s annual budget, and for Congressin both assessing the agencies’ out-year plansand deciding on current year appropriations.Policy makers and analysts would also beable to use the consolidated report, in conjunc-tion with the information on the conditionof federally owned assets, to judge the settingof priorities across the government and toassess whether the government has unmet

needs that are likely to show up in futurebudgets. Furthermore, the report would en-hance the public’s ability to understand howand to what extent their tax dollars arebeing spent on current activities or usedto increase the public capital stock. It wouldalso reveal, for example, whether the capitalstock was growing at an unreasonably rapidrate, or at the other extreme, contracting.

Fourth, the consolidated reports should pro-vide information based on multiple conceptsof investment, including the current FASABdefinition of government investment as wellas alternative concepts the public and theCongress might find useful. Toward this end,FASAB should examine the feasibility ofdeveloping alternative definitions—especiallythose that take account of investments inhuman capital and other intangible assets.Multiple views of investment would promotebetter understanding of the federal govern-ment’s past use of resources and its currentneeds.

Recommendation 10: Condition of ExistingAssets

The commission believes there should bebetter information on the condition of existingassets. As previously noted, work is plannedat the federal level for agencies to begindeveloping standardized methods for estimat-ing deferred maintenance. The commissionstrongly supports these efforts and encouragesOMB to work with the agencies to completethis task promptly and implement its results.In combination with the rest of the informationprovided in the audited financial statements,data on deferred maintenance will enablepolicy makers to develop sound plans formaintaining existing assets and spending onnew ones, where that is advisable.

OMB should also work with the agenciesto compile an annual report on the conditionof state and local infrastructure, or at leaston that portion that has been federally as-sisted. The commission recognizes that thisis a major, long-term undertaking and requiresthe cooperation of state and local governmentsto help identify what data are availableand additional information that needs tobe collected. This endeavor may also callfor federal legislation requiring the states


and localities to report information abouttheir assets to the federal government. Butin this ‘‘information age,’’ there is no reasonfor citizens and policy makers throughoutthe country—and especially those at the fed-eral level—to remain unaware of the conditionof assets that have been financed or supportedwith federal tax dollars.

Improving the Evaluation of BudgetDecisions

Recommendation 11: Federal ‘‘Report Card’’

Finally, it is critical that the federal govern-ment have mechanisms in place for constantlyevaluating the outcomes of budget decisions.Many agencies already do this (althoughwith varying degrees of success). Still, thereis room for improvement.

In particular, a natural companion to therecommendation that the benefits and costsof major capital projects be assessed beforethey are undertaken, is that the agencies,under OMB’s guidance and review, should(1) regularly conduct benefit-cost analysesof existing major capital spending initiativesand (2) report the results in a manneruseful for decision-makers and the public.Such a ‘‘Report Card,’’ which could be includedin the annual Analytical Perspectives thataccompanies the budget, could identify whichinvestment projects have produced returnsto society in excess of some benchmark ‘‘costof capital’’—such as the prevailing interest

rate on long-term federal debt, the averagecost of capital expected by private investors,or other thresholds that OMB determinesuseful to the public. Furthermore, it is impor-tant that the agencies and OMB use suchexisting mechanisms as the Government Per-formance and Results Act (GPRA), the FederalAcquisition Streamlining Act (FASA), andthe Clinger-Cohen Act to evaluate publicinvestment programs. In this way, policymakers can make mid-course alterations, iffeasible, and learn from the successes andweaknesses of past efforts to help producewise spending decisions in the future.

To be sure, not all programs have benefitsthat can be easily quantified, let alone ex-pressed in monetary terms. Indeed, the com-mission recognizes that the projects in whichit may be feasible to provide a monetaryanalysis may account for a relatively smallfraction of total spending; nonetheless, overtime, advances in estimating techniques maypermit a larger fraction of total spendingto be evaluated in this manner. Furthermore,as in the regulatory sphere, OMB and theagencies should do the best they can withthe available data. Where the benefits ofprojects cannot be measured in monetaryterms, the evaluations should identify theobjectives of the projects and assess theirbenefits qualitatively. Meanwhile, OMB shouldtake the lead in identifying ways to improveboth the collection of information useful tosuch analyses and analytic techniques.

39

PROS AND CONS OF A ‘‘CAP’’ON CAPITAL SPENDING

There is no inherent reason that a capitalbudgeting process used for decision-makingmust be linked with any particular financingrule. In principle, capital spending couldbe subject to an appropriations process sepa-rate from the one used for operating expendi-tures. To ensure spending restraint, a separatecap on capital spending could also be imposed.

Most members of the commission do notsupport a capital spending cap. Several com-missioners, however, believe that moving inthis direction might be appropriate, but withthe understanding that it would require achange in the way both the executive andlegislative branches do business (which mightbe facilitated by a study involving representa-tives of both branches).

In this section, we discuss issues thatwould have to be resolved if Congress andthe administration were to agree on includinga separate cap on capital spending (undersome definition) as part of the budget process,as well as some implications of taking sucha step. In the process, we outline argumentsin favor of and against making this changein budget procedures. This discussion assumesthe continuation of the spending caps thatare now part of the Budget EnforcementAct.

Implementation Issues

A cap on capital spending could not beimplemented for decision-making until at leastthe following three issues are resolved:

First, how would capital be defined? Akey argument against adopting a capitalcap is that there is currently no consensuson what definition would be most useful—or even whether any form of capital spendingshould be broken out and treated differentlyfor budget purposes. Furthermore, if on theone hand, capital spending eventually is fi-nanced at least in part by borrowing, thetemptation to expand the definition will grow.On the other hand, if a separate cap is

imposed without a financing rule, it couldimpart a bias against any investment expendi-tures left out of the definition of capital.

Those favoring a separate capital cap arguethat as long as policy makers identify aspecific objective, a definition of capital wouldfollow more easily. Indeed, if Congress wereinclined to adopt the idea of a cap, itcould ask FASAB, CBO, or some other bodyto provide it with recommended definitionsof capital and then decide to use one ofthem.

Second, on what basis should any capitalspending cap or target be set? Critics ofthe idea argue that there is no objectivemethod for answering this question, especiallyif capital is broadly defined to include, say,all national assets. Indeed, as mentionedearlier in this report, there is no way toknow whether or not the current budgetprocess has a macro bias precisely becauseit is impossible to make an objective statementabout the optimal level of broadly definedpublic capital.

Supporters of a capital budget cap haveseveral responses to this. One is that policymakers already routinely make tradeoffs ofprograms with diverse objectives in the currentbudget process, and it would be no differentif they were asked to do so specificallyfor all capital spending; indeed, having anational discussion on that issue would behelpful. Another response is that settinga limit on capital spending would becomeconceptually more manageable if capital weremore narrowly defined to be consistent witha single objective.

Third, How exactly would Congress imple-ment a capital cap? One answer is thatCongress could simply set a non-bindingtarget for capital spending in the annualbudget resolution. A more ambitious stepwould be to impose a statutory cap thatwould actually constrain total appropriationsfor capital spending. Presumably, the appro-


n Comment of Commissioner Levy: Although there are reasonsto limit the size of federal debt and deficits, I cannot agree thatdeficits ‘‘divert national saving away’’ from other uses. I and othereconomists argue that investment generally determines saving, notthe other way around. Certainly ‘‘saving equals investment’’ is afact, an accounting identity. However, the notion that governmentactions to increase or decrease public saving will similarly increaseor decrease investment is a theoretical proposition that is neitheruniversally accepted nor empirically proven. Notably, it ignores theoffsetting impact of changes in fiscal policy on business saving(profits).

priations committees would divide up thecapital total among each of the thirteenappropriations subcommittees, as they do nowwith the so-called section ‘‘302(b)’’ discre-tionary spending allocations. 38

Critics of the idea would argue that thereis no way to guarantee that spending withinany capital allocation is truly for capitalrather than just labeled as such. Supporterswould respond that as long as Congressagreed upon a definition of capital, an inde-pendent scorekeeper like CBO would ensurefaithful implementation of the cap.

Implications

We turn next to the implications of settinga cap on capital expenditures (under somedefinition) that both proponents and opponentsof the idea have claimed.

Impact on Budgetary Choices

Advocates of such a cap argue that itwould have at least two salutary effects:it would focus greater attention on the totalamount of resources devoted to achievinglonger-run objectives, and it would improvethe allocation of limited resources towardthe most cost-effective initiatives.

Opponents of a separate cap on capitalspending have several responses, apart fromthose already outlined in the rest of thisreport. Arguably, the claimed macro andmicro benefits of a cap could be attainedthrough the improved reporting requirementsand longer-term agency spending plans werecommend, without running the risks ofseveral potential adverse consequences. Also,as already discussed, any definition of capitalcould create a bias in favor of those itemsincluded within the definition whiledisadvantaging any capital or other itemsthat might fall outside it. This problemmight be mitigated, of course, to the extentthat policy makers defined capital broadly—if not initially, then in later years. Buta more expansive definition might weakenbudget discipline, which could lead to excessivepublic borrowing.

Impact on Budget Discipline

In principle, budget discipline would notbe weakened if a capital budget were adopted

without any rule that capital—gross or net—be financed by borrowing. Indeed, advocatesof a capital budget might argue that aseparate cap on public capital spending wouldpromote budget discipline at the micro level,where limited resources are allocated amongalternative uses. If policy makers were explic-itly required to trade off different typesof capital spending, they might be morecareful about which capital projects theyauthorize.

Critics of a separate cap on capital spendingargue that it would tempt policy makersto adopt a borrowing-for-investment rule pre-cisely because capital is identified with thelong run. Future generations, after all, willreap the benefits of such spending, so whynot have them incur the cost of financingit as well? To the extent public investmentbecomes debt-financed as a matter of course,policy makers would then have incentivesto move expenditures within the definitionof capital so that they could be debt-financed.This could lead to excessive government bor-rowing, which would lower economic growthby diverting national saving away from poten-tially more productive uses in the privatesector. n In addition, future generations mightnot appreciate the benefits of programs orprojects authorized many years before, normight the programs be suitable for the in-tended beneficiaries.

More broadly, with present expenditureprograms and taxes, the federal governmentwill apparently run surpluses in the unifiedbudget, under current budget conventions,for some years to come—although these projec-tions (which have often proved to be incorrectin the past) could miss the mark in thefuture. The country needs, and does nothave, policies and procedures for decidinghow big those surpluses should be, assumingthe projections of surplus prove to be reason-ably accurate.

41PROS AND CONS OF A ‘‘CAP’’ ON CAPITAL SPENDING

Deciding how much of a surplus (in theunified budget) to achieve is difficult. Federalsurpluses add to the national saving, thesource from which private investment canbe financed, and thus contribute to economicgrowth. Logically, the proper size of thesurplus should depend on the rate of privatesaving, on expected technological advance,and on expected change in the size andcomposition of the population. It should alsorepresent a social choice between the consump-tion of the present generation and the con-sumption of future generations. To recommendhow these and probably other relevant vari-ables should be taken into account in decidingon the proper size of the budget surplusis beyond the charge, as well as the com-petence, of this commission. We do, however,want to recognize that in such a process,some weight might be given to the amountof federal investment as a factor influencingthe proper size of the surplus. In rejectingboth the Balanced Budget Amendment aswell as a simplistic capital budget that wouldfinance all capital with debt, we do not

mean to reject consideration of the totalamount of federal capital (however it isdefined) in developing a more sophisticatedfiscal policy in the future.

Impact on Macroeconomic Stability

It is difficult to reach firm conclusions,in the abstract, concerning the impact acap on capital spending would have on fiscalpolicy and hence on macroeconomic stability.The effect of federal fiscal policy on therest of the economy in any given yearis typically measured by the change in thestructural budget balance (the surplus ordeficit assuming some given level of economicactivity, typically full employment). Each year,in the course of agreeing on a budget,Congress and the administration togetherdecide how large or small that change inthe structural fiscal balance should be. Thecommission cannot say with any degree ofcertainty whether the adoption of a separatecapital cap would systematically move fiscalpolicy in the direction of stimulus or contrac-tion.

43

Table 5. BORROWING TO FINANCENET INVESTMENT, FISCAL YEAR 1999

(billions of dollars)

National investment ........................................... $130Less: Depreciation .............................................. –72*

Net Investment ................................................... 58

Less: Trust fund revenue earmarked for cap-ital projects ....................................................... –46

Total investment purportedly requiring debt-financing ........................................................... 12

Projected FY 1999 surplus in the unifiedbudget ............................................................... 10

* Depreciation is not included for capital projects fi-nanced by earmarked trust fund receipts. OMB estimateddepreciation on these assets in FY 1999 to be $11 billion.

Source: OMB, Analytical Perspectives, Fiscal Year 1999,p. 156.

o Comment of Commissioner Levy: Under some types of a capitalbudget, there might be more gross and net investment, whichunder a borrowing-for-investment rule, would justify more borrow-ing.

A CAPITAL BUDGET AND DEPRECIATIONNo government currently budgets deprecia-

tion of infrastructure [GAO, 1995]. Nonethe-less, one variation of a capital budget thatwe consider here would attempt to mimicthe income statement of a private firm byhaving depreciation recognized as an expensein the annual operating budget.

In principle, Congress could provide a per-manent appropriation for depreciation; how-ever, such a step would not be a meaningfulexercise of control because depreciation rep-resents an expense that in effect has beenobligated in the past—the portion of theoriginal cost of an asset that is being usedor consumed in a given year. 39

Nevertheless, recognizing depreciationshould be part of any decision rule linkingthe amount of capital spending to the amountof federal borrowing. If this were done, how-ever, it would be important to recognizethe difference between net and gross invest-ment. If the government borrows to financegross investment, without taking account ofdepreciation of existing capital, it will beplacing an undue burden on future genera-tions. Only net additions to capital—grossinvestment minus depreciation—generate fu-ture benefits and thus could appropriatelyjustify additional borrowing.

For this reason, any borrowing-for-invest-ment rule would have to include depreciationof all capital assets. But if this becomesthe rule—borrowing for net investment—itturns out that the level of additional borrowingthat might be justified is considerably belowwhat some proponents of a borrowing-for-investment rule might anticipate (assumingcurrent levels of gross investment by thefederal government). o This is illustrated inTable 5, which shows that in fiscal year1999, the Clinton administration intendedto spend $130 billion on national capital,defined by both OMB and GAO to include

all federal spending that contributes directlyto economic growth (most federally ownedassets, federally financed infrastructure, R&Daimed at enhancing growth, and federal ex-penditures on education and training). Inthe same year, OMB estimated that theexisting stock of these assets would depreciateby $72 billion, leaving net investment of$58 billion. 40 In addition, OMB also estimatedreceipts of $46 billion in FY 1999 fromspecial taxes earmarked for trust funds (suchas the Highway Trust Fund) dedicated tosupport capital spending. Subtracting thesereceipts from the net investment figure wouldyield borrowing of just $12 billion.

An issue related to depreciation is thefollowing proposal offered by certain individ-uals who testified before the commission:that the budgeting for capital be switchedfrom the current convention, under whichthe full cost of capital projects is appropriatedup front, to a system of accrual accounting,in which the costs of such projects (andtherefore their appropriations) would bespread out over their useful lives. Sucha change in scoring would have the following


objective: to remove the alleged bias underthe current system against capital spendingthat arises because large capital expenditurescan cause spending to bump up against,and even exceed, the caps on discretionaryspending.

Earlier, however, we suggested that therewas no clear evidence as to whether ornot this was occurring. But even if it were,proper accrual accounting requires deprecia-tion of existing as well as new capital.Table 5 suggests that investment net ofdepreciation is substantially below the levelof gross investment spending (although bothcould be very different under a system ofaccrual budgeting).

Realizing this, some have argued that thescoring of capital items should be changedonly for future projects. Depreciation on exist-ing assets would be ignored. But this would

mean that new capital projects would nothave to compete for resources with previouslyapproved projects. The commission stronglyrejects this approach, which clearly wouldbe inconsistent with standard accrual account-ing practices. Moreover, if the federal govern-ment were to adopt accrual-based budgeting,it would be inappropriate to confine it tothe scoring of capital. Other programs, includ-ing federal insurance and pensions, woulddeserve accrual budgeting as well. In fact,these programs, which now appear to bewell financed when scored on a cash basis,also have large liabilities; consequently, whenscored on an accrual basis, they would implya much larger level of total federal spendingthan the amount now being reported. Decision-makers could then decide to curtail ratherthan expand capital spending (which is notthe objective of some of those who haveurged the adoption of accrual budgeting).

45

OTHER VERSIONS OF A CAPITAL BUDGETSo far, we have considered alternative

versions of a capital budget: one linkedto a financing rule, a capital spending cap,and incorporation of depreciation into theconsideration of capital (including a switchto accrual-based accounting). These are notthe only ways to define or implement a‘‘capital budget.’’

It may be possible to think of a capitalbudget process that is much more informalthan any of the versions just outlined. Forexample, federal policy makers might findways to use or publicize multiple financial

statements—not just the unified budget, butseparate ‘‘operating’’ and ‘‘capital’’ budgetsor statements. Discipline might come froma specific constraint on the growth of debt—much like that recently adopted in the UnitedKingdom—rather than from any balancedbudget rule. There surely are other variations,many of which would require a major overhaulof the current budget process (which constrainsspending through spending caps and PAYGOrules relating to taxes and mandatory spend-ing). However, the commission has not consid-ered all of these alternatives in depth.

CONCLUSIONNo matter how it is defined, capital spending

by the federal government is important be-cause it delivers long-run benefits to thenation. The challenge for federal policy makersis to ensure that an appropriate amountof resources is devoted to such spendingin the aggregate as well as among thevarious projects and initiatives that appearto fit within the definition of public capitalor investment.

Clearly, there is no objective way to assesswhether the current federal budget process

leads to a bias, one way or the other,in the amount of total capital spendingby the federal government. The commissionhas concluded, however, that the processhas a number of flaws. These shortcomingshave led and continue to lead to a less-than-ideal allocation of capital spending amongindividual programs, as well as between newinvestments and maintenance of existing cap-ital assets. In our view, the combinationof recommendations outlined here would helpremedy these flaws.

47

ENDNOTES

1. This should not be surprising. One thoughtful economist writing in 1965 noted that ‘‘thenumber of different definitions of ‘‘capital’’ employed in the writings of economists defy enumera-tion’’ [Dewey, p. 4].

2. The Bureau of Economic Analysis, which is responsible for the National Income andProduct Accounts (NIPA), has developed an experimental account for research and developmentcapital, however.

3. The outlays shown in the table include the subsidy component of federal credit programsaimed at supporting or stimulating capital spending.

4. It is important in reading Table 1 to bear in mind that the different categories of capitalspending may have very different economic (and non-economic) effects. For example, it is highlylikely that all, or close to all, federal expenditures on defense capital and R&D create capitalthat would otherwise not exist. Some federal spending on non-defense capital—such as highwaysand other capital grants to the states—may displace spending that would otherwise occur at thestate and local levels. Similarly, some portion of the subsidies on student loans probably getstranslated into higher tuition rather than more education. At the same time, it is also possiblethat federal matching grants for infrastructure may encourage states and localities to investmore than they otherwise would. In addition, it may be fairer or more efficient for the federalgovernment to finance certain infrastructure than for local residents to bear all of the cost. Thekey point is that different types of federal capital spending have different impacts on the na-tion’s overall stock of capital (as do federal surpluses, deficits, and taxes).

5. The federal tax code contains a variety of incentives designed to enhance various types ofcapital spending, including (but not limited to): tax-exemption of interest on state and localbonds used to finance infrastructure and other physical investment; tax incentives for private re-search and development expenditures; and various tax incentives that support investment ineducation.

6. In the National Income and Product Accounts, depreciation is also deducted to determinethe federal government’s ‘‘current surplus or deficit.’’

7. The GDP data have been adjusted for inflation using the chain-weighted GDP deflator,while investment expenditures have been deflated using a chain-weighted investment deflator.

8. Aschauer, 1989; see also Munnell, 1992. The Boskin Commission report recently arguedthat the inflation data are overstated for various reasons, which if true, would also mean thatreal output and productivity are understated. This report has been the subject of considerablecontroversy, however, among economists.

9. For a summary of such studies, see Gramlich, 1994 and CBO, 1998.


10. There are differences in ownership of certain key sectors of the economy where invest-ment in physical assets is especially important. For example, transportation and utility servicesthat are publicly provided in other countries are not provided by governments here (utility serv-ices being a prime example, with the exception of some federal hydroelectric projects and munici-pally owned power companies). In addition, in countries where the government provides hospitalcare services (such as the United Kingdom), investments in hospitals show up as governmentcapital spending, whereas in the United States most health care is delivered privately (with theexception of military and veterans’ hospitals and some municipally owned hospitals). Similarly,in the United States, much higher education is provided privately, whereas in many countrieshigher education is more likely to be provided publicly. While these differences in ownership pat-terns between countries do not affect comparisons of total national investment, they do distortcomparisons of capital spending by governments.

11. The use of the word ‘‘capital’’ in the financial accounting context can be confusing, sincethe term is often interpreted as the shareholder’s contribution to the company, and not a cat-egory of assets, which is the way the term is often defined by economists and government policymakers.

12. Under GAAP, capital assets are recorded, with some exceptions, at their original costs(minus any cumulative depreciation in the case of fixed assets), and not at their current marketvalues.

13. There is a GAAP for state/local governments, and the body responsible for its principlesis the Governmental Accounting Standards Board (GASB).

14. For mature firms with access to credit, equity is typically the last means of financing(other than through stock options to employees) because new equity dilutes the ownership per-centages of existing shareholders. For new or young firms without a track record of profitability,equity may be the only means of financing, whether by selling new shares or granting stock op-tions or shares to employees in lieu of cash.

15. An alternative way of evaluating capital projects that is sometimes used is to computetheir internal rates of return, or IRR, and to compare the result with the discount rate. The IRRis that discount rate that theoretically equates the discounted future cash flows to the cost of theproject or that produces a zero NPV. If the IRR exceeds the discount rate, then proceeding withthe project is justified. In practice, however, the IRR can be difficult to compute and yields dif-ferent results from NPV when cash flows are very uneven.

16. This need not always be the case, however. Firms that manage their spending throughsomething analogous to the ‘‘statement of cash flow’’ in effect combine their budgeting for operat-ing expenses and capital items.

17. The material in this subsection is drawn from National Association of State Budget Offi-cers, 1997; OMB, 1998, p. 154; Hush and Peroff; and GAO, 1986.

18. Typically, states include in their definitions of capital expenditures major maintenance,although dollar thresholds for defining what maintenance is ‘‘major’’ also vary across states.

19. States do record depreciation expense in their proprietary (or commercial-type) fundsand in trust funds where net income, expense, or capital maintenance is measured.

20. Examples of special funds include the Land and Water Conservation Fund and the Na-tional Wildlife Refuge Fund.

21. Two other types of government funds are ‘‘public enterprise funds’’ (revolving funds thatconduct business-type operations with the public) and ‘‘intragovernmental funds’’ (that do thesame within and between government agencies).

22. The historical material summarized in this and the subsequent two paragraphs drawson Nuzzo.

49ENDNOTES

23. It should be noted that although the budget does not distinguish between capital andoperating expenditures, the Analytical Perspectives volume of the budget contains informationthat makes that distinction at an aggregate level and for major programs.

24. The Reagan administration defined investment primarily to cover defense expenditures.The Bush administration broadened the term to include federal expenditures on R&D, infra-structure, child immunization, drugs, the environment and energy, and programs aimed at pre-serving America’s heritage (such as those for the arts, humanities, and museums). The Clintonadministration has used a similar definition, but has concentrated on transportation, environ-ment, rural development, energy, community development and defense conversion, housing, edu-cation, justice, health care, and investments in information technology to improve the delivery ofgovernment services.

25. For an elaboration of this point, see Eichengreen, p. 84. Indeed, there is empirical evi-dence indicating that state governments have been effectively rationed out of the market whenthe ratio of their outstanding debt rises above 9 percent of state economic output [Bayoumi,Goldstein, and Woglom].

26. New Zealand also imposes a ‘‘capital charge’’ on each agency, which is paid to the Treas-ury twice a year. Although the capital charges of the various agencies are washed out on theoverall government’s budget, they were adopted as a means of encouraging departments to man-age their capital assets wisely [Troup Testimony]. Below, we suggest that the federal govern-ment experiment with a similar procedure, the establishment of ‘‘capital acquisition funds.’’

27. The Adequacy of Appropriations Act and the Antideficiency Act require all agencies tohave budget authority for all obligations, including capital acquisitions.

28. The ‘‘social’’ rate of return of a project measures the benefit of the project to the nationas a whole, taking into account both economic and non-economic considerations (such as equityand freedom). Social returns exceed the returns earned by the private sector alone where theprojects generate benefits beyond those reaped only by those who undertake them. For example,it is widely acknowledged that much basic R&D undertaken by the government generates bene-fits for many firms and industries, as well as society as a whole. The same is true for education,which confers benefits not just on the individuals who receive it, but also on the entire society tothe extent that a more educated work force is likely to come up with new ideas that make busi-nesses more productive.

29. As former CBO Director Robert Reischauer pointed out in his testimony, the federal gov-ernment also pays attention to distributional concerns: ‘‘On the basis of economic considerationsalone, the federal government would allocate far less to roads and bridges and public buildingsin North Dakota than it now does. But there is agreement that all areas of the country shouldenjoy the advantages of a modern highway system, even where the economic payoff is minimal.’’[Reischauer, p. 3].

30. The FASAB consists of nine members: one representative each from OMB, Treasury,GAO, and CBO; two representatives from other executive branch agencies; and three representa-tives from the private sector or state and local government. FASAB has developed two state-ments of accounting concepts and ten statements of standards applicable to accounting by thefederal government.

31. The Act significantly increased total funding for highways to $217 billion for FY1998–2003, a substantial increase over the $155 billion authorized for the preceding five years.The commission as a whole takes no position on the merits of this funding level, but notes onlythat the linkage between future spending and revenue dedicated to the trust fund addresses theproblem that, in prior years, motor fuels tax revenues were not being fully used for their in-tended purpose.


32. Nor do the aggregate investment and capital stock data currently reported in the Ana-lytical Perspectives and in the National Income and Product Accounts reveal the physical condi-tion of those assets (which are reported at current cost minus an adjustment for accumulated de-preciation).

33. As Paul Posner from the GAO told the commission: ‘‘Prudent capital planning can helpagencies to make the most of limited resources, while failure to make timely and effective capitalacquisitions can result in increased long-term costs’’ [Posner at 14]. As an example, Posner point-ed to planning failures that have led to cost overruns, schedule delays, and performance short-falls in the Federal Aviation Administration’s modernization program. Similar problems appearto have plagued the computer modernization program at the Internal Revenue Service. In its re-cent Capital Programming Guide, OMB encourages agencies to develop long-term capital plansas part of their planning process and to use these plans to develop their annual budget justifica-tions.

34. Debt service is an appropriate rental charge whether or not the federal governmentmust borrow to finance a certain project. In particular, even if the government is running anoverall surplus, there is an opportunity cost associated with the acquisition of a capital item—measured by the cost of borrowing—associated with not having an even larger surplus.

35. It would not be appropriate or useful to include in the CAFs grants to states or localitiesfor what, in other contexts, may be deemed to be capital expenditures, such as those for high-ways. The grant itself is the program; highways and other federally assisted capital assets arenot being used to provide federal services, so there are no federal programs to which the cost ofusing this capital should be allocated for budget decision-making. Moreover, spending ‘‘spikes’’tend to be associated with the construction or acquisition of federally owned facilities; spendingon highways and other ‘‘capital’’ items tends to be relatively smooth from year to year.

36. Some agencies have portions of their budgets considered by more than one appropria-tions subcommittee. For example, while most of the budget of the Department of the Interior isconsidered by the Interior subcommittee, the Energy and Water Development subcommittee hasjurisdiction specifically over the budget of the Bureau of Reclamation (an agency within Inte-rior). Similarly, the Labor/HHS subcommittee oversees most of the budget of the Department ofHealth and Human Services, but the Agriculture subcommittee has jurisdiction over the budgetof the Food and Drug Administration. In these cases, it may be necessary to establish multipleCAFs that fit jurisdictional boundaries of the appropriations subcommittees.

37. Moreover, a federal mandate linking federal funding to state and local support of main-tenance might encourage rating agencies to allow bonding for maintenance.

38. The number refers to the section of the BEA that provides for allocating spending totalswithin the cap among the appropriations subcommittees.

39. Note that one virtue of a CAF is that the rental rate that would be charged implicitly onthe use of capital assets would include a charge for depreciation.

40. The depreciation total reported by OMB and shown in the table includes depreciation ofeducation and R&D expenditures.

51

Selected References and Bibliography

Aschauer, David Alan. ‘‘Is Public Expenditure Productive?’’, Journal of Monetary Economics,1989, Vol. 23, pp. 177–200.

Bayoumi, Tamim; Morris Goldstein; and Geoffrey Woglom. ‘‘Do Credit Markets Discipline Sov-ereign Borrowers? Evidence from U.S. States’’ (International Monetary Fund, Washington,D.C., 1994).

Brown, Gordon. ‘‘Statement By The Chancellor of the Exchequor on the Economic and FiscalStrategy Report,’’ House of Commons, June 11, 1998.

Congressional Budget Office. CBO Papers: The Economic Effects of Federal Spending on Infra-structure and Other Investments, June 1998.

Dewey, Donald. Modern Capital Theory (Columbia University Press, New York, 1965).

Eichengreen, Barry. International Monetary Arrangements for the 21st Century (The BrookingsInstitution, Washington, D.C., 1994).

General Accounting Office. Department of Energy: Opportunity to Improve Management of MajorSystem Acquisitions (GAO/RCED–97–17, November 26, 1996).

————————. The Role of Depreciation in Budgeting for Certain Federal Investments (GAO/AIMB–95–34, February 1995).

————————. Budget Issues: Incorporating an Investment Component in the Budget. (GAO/AIMD–94–40, November 1993).

————————. Budget Issues: Capital Budgeting Practices in the States. GAO/AFMD–86–63FS, July 1986).

Gramlich, Edward M. ‘‘Infrastructure Investment: A Review Essay,’’ Journal of Economic Lit-erature. September 1994, Vol. 32, No. 3, pp. 1176–1196.

Hush, Lawrence W., and Kathleen Peroff. ‘‘The Variety of State Capital Budgets: A Survey,’’Public Budgeting and Finance. Vol. 8, Summer 1988.

Kirova, Milka S., and Robert E. Lipsey. ‘‘Measuring Real Investment: Trends in the UnitedStates and International Comparison,’’ Federal Reserve Bank of St. Louis Review. January/February 1998, pp. 3–18.

Munnell, Alicia H. ‘‘Infrastructure Investment and Investment Growth,’’ Journal of EconomicPerspectives, Fall 1992, Vol. 6, No. 4, pp. 189–98.

National Association of State Budget Officers. Capital Budgeting in the States (September 1997).

Nuzzo, James L.J. ‘‘Whither a Federal Capital Budget.’’ Unpublished manuscript (May 1994).

Office of Management and Budget. Analytical Perspectives, FY 1999 Budget (February 1998).

———————— and Chief Financial Officers Council. Federal Financial Management StatusReport & Five Year Plan (June 1998).

Posner, Paul L. ‘‘Budgeting for Capital,’’ Testimony before the Commission. March 6, 1998.

Reischauer, Robert. Testimony before the Commission. April 24, 1998.

Schick, Allen. The Federal Budget Process: Politics, Policy, Process. (The Brookings Institution,Washington, D.C., 1995).

Tillet, Ronald L. ‘‘Capital Planning and Budgeting in the Commonwealth of Virginia.’’ Testimonybefore the Commission. May 8, 1998.

Troup, George. ‘‘Capital Budgeting in the New Zealand Government.’’ Testimony before the Com-mission. May 8, 1998.

53

APPENDIX AExecutive Order 13037: Commission to Study Capital Budgeting, and

the Amendments

Federal Register

Presidential Documents

Vol. 62, No. 44 Thursday, March 6, 1997

Executive Order 13037 of March 3, 1997

Commission To Study Capital Budgeting

By the authority vested in me as Presidentby the Constitution and the laws of theUnited States of America, including the Fed-eral Advisory Committee Act, as amended(5 U.S.C. App.), it is hereby ordered asfollows:

Section 1. Establishment. There is estab-lished the Commission to Study Capital Budg-eting (‘‘Commission’’). The Commission shallbe bipartisan and shall be composed of 11members appointed by the President. Themembers of the Commission shall be chosenfrom among individuals with expertise inpublic and private finance, government offi-cials, and leaders in the labor and businesscommunities. The President shall designatetwo co-chairs from among the members ofthe Commission.

Sec. 2. Functions. The Commission shallreport on the following:

(a) Capital budgeting practices in othercountries, in State and local governmentsin this country, and in the private sector;the differences and similarities in their capitalbudgeting concepts and processes; and thepertinence of their capital budgeting practicesfor budget decisionmaking and accountingfor actual budget outcomes by the FederalGovernment;

(b) The appropriate definition of capitalfor Federal budgeting, including: use of capitalfor the Federal Government itself or the

economy at large; ownership by the FederalGovernment or some other entity; defenseand nondefense capital; physical capital andintangible or human capital; distinctionsamong investments in and for current, future,and retired workers; distinctions between cap-ital to increase productivity and capital toenhance the quality of life; and existingdefinitions of capital for budgeting;

(c) The role of depreciation in capitalbudgeting, and the concept and measurementof depreciation for purposes of a Federalcapital budget; and

(d) The effect of a Federal capital budgeton budgetary choices between capital andnoncapital means of achieving public objec-tives; implications for macroeconomic stability;and potential mechanisms for budgetary dis-cipline.

Sec. 3. Report. The Commission shall adoptits report through majority vote of its fullmembership. The Commission shall reportto the National Economic Council by March15, 1998, or within 1 year from its firstmeeting.

Sec. 4. Administration. (a) Members ofthe Commission shall serve without compensa-tion for their work on the Commission. Whileengaged in the work of the Commission,members appointed from among private citi-zens of the United States may be allowedtravel expenses, including per diem in lieu


of subsistence, as authorized by law forpersons serving intermittently in the Govern-ment service (5 U.S.C. 5701–5707).

(b) The Department of the Treasury shallprovide the Commission with funding andadministrative support. The Commission mayhave a paid staff, including detailees fromFederal agencies. The Secretary of the Treas-ury shall perform the functions of the Presi-

dent under the Federal Advisory CommitteeAct, as amended (5 U.S.C. App.), exceptthat of reporting to the Congress, in accord-ance with the guidelines and procedures estab-lished by the Administrator of General Serv-ices.

Sec. 5. General Provisions. The Commissionshall terminate 30 days after submittingits report.

WILLIAM J. CLINTON

THE WHITE HOUSE,

March 3, 1997

Federal Register


Vol. 62, No. 211 Friday, October 31, 1997

Executive Order 13066 of October 29, 1997

Amendment to Executive Order 13037, Commission To Study CapitalBudgeting

By the authority vested in me as Presidentby the Constitution and the laws of theUnited States of America, and in order toincrease the membership of the Commissionto Study Capital Budgeting, it is herebyordered that the second sentence of section

1 of Executive Order 13037 is amendedby deleting ‘‘11’’ and inserting ‘‘no morethan 20’’ in lieu thereof. It is further orderedthat section 3 of Executive Order 13037is amended by deleting the words ‘‘by March15, 1998, or’’.

WILLIAM J. CLINTON

THE WHITE HOUSE

October 29, 1997

55APPENDIX A

Federal Register


Vol. 63, No. 240 Tuesday, December 15, 1998

Executive Order 13108 of December 11,1998

Further Amendment to Executive Order 13037, Commission To StudyCapital Budgeting

By the authority vested in me as Presidentby the Constitution and the laws of theUnited States of America, and in order toextend the reporting deadline for, and theexpiration date of, the Commission to StudyCapital Budgeting, it is hereby ordered thatExecutive Order 13037, as amended, is further

amended by deleting in section 3 of thatorder ‘‘within 1 year from its first meeting’’and inserting in lieu thereof ‘‘by February1, 1999’’ and by deleting in section 5 ofthat order ‘‘30 days after submitting itsreport’’ and inserting in lieu thereof ‘‘onSeptember 30, 1999’’.

WILLIAM J. CLINTON

THE WHITE HOUSE

December 11, 1998

57

APPENDIX B

Commission Membership

Co-Chairs

Kathleen Brown is President of Bank of America’s Private Bank West. She has been withBank of America since 1994. From 1991 until 1994, she served as California’s 28th Treasurer,responsible for managing the state’s investment portfolio, and administering bond sales to fi-nance schools, parks, prisons, housing, health facilities, and environmental programs. Ms. Brownwas the Democratic nominee for Governor of California in 1994.

Jon S. Corzine is Co-Chairman and Senior Partner of the investment banking firm Goldman,Sachs & Co. Since joining the firm in 1975, he has held a variety of positions including partner-in-charge of government, mortgage and money markets trading, co-head of the Fixed Income Di-vision and of the firm’s treasury and finance functions, and Chief Executive Officer.

Members

Willard W. Brittain, of New York, New York, is Global Managing Partner ofPricewaterhouseCoopers.

Stanley E. Collender, of Washington, D.C., is a Senior Vice President and Managing Directorof the Federal Budget Consulting Group at Fleishman-Hillard.

Orin S. Kramer, of Englewood, New Jersey, is a general partner of Kramer Spellman, L.P.,which manages investment vehicles focusing on the financial services industry.

Richard C. Leone, of Princeton, New Jersey, is the President of The Century Foundation, for-merly the Twentieth Century Fund, Inc., a public policy research institution in New York. Hewas New Jersey State Treasurer and Chief Financial and Budget Officer for 1973–1977.

David Levy, of Pound Ridge, New York, is the Vice Chairman of the Jerome Levy EconomicsInstitute of Bard College and Director of the Levy Institute Forecasting Center.

James T. Lynn, of Bethesda, Maryland, is retired Chairman and Chief Executive Officer ofAetna Life & Casualty. During the Ford Administration, he served as Director of the Office ofManagement and Budget.

Cynthia A. Metzler, of Washington DC, is a Partner with the law firm of Pepper HamiltonLLP. She was formerly Acting Secretary of Labor during the Clinton Administration.

Luis Nogales, of Beverly Hills, California, is President of Nogales Partners, and was Chairmanand CEO of Embarcadero Media and United Press International and President of Univision.

Carol O’Cleireacain, of New York, New York, is a Senior Fellow at The Brookings Institutionand former Finance Commissioner and Budget Director of New York City.

Rudolph G. Penner, of Washington, D.C., holds the Arjay and Frances Miller Chair in PublicPolicy at the Urban Institute. He is a former Director of the Congressional Budget Office.

Steven L. Rattner, of New York, New York, is Deputy Chief Executive of the investment bank-ing firm Lazard Freres & Co. LLC.

Robert M. Rubin, of Southampton, New York, is Executive Vice President and Director of AIGTrading Group, an international currency and commodity dealer.

Herbert Stein, of Washington, D.C., is a senior fellow of the American Enterprise Institute. Heserved as Chairman of the Council of Economic Advisers under Presidents Nixon and Ford.

59

APPENDIX CList of Witnesses and Written Statements

The following persons have appeared before the commission to testify, have presented a writtenstatement, or both. The individuals are identified below in alphabetical order. The commission’swebsite (at http://www.whitehouse.gov/pcscb) has a one-page staff summary for the statement ofeach individual and the full written testimony if written testimony was submitted. An asterisk(*) beside the name indicates that a written statement was submitted to the commission.

• James J. Abel, Executive Vice Presidentand CFO, Lamson & Sessions, on behalfof Financial Executives Institute (appear-ance on May 8, 1998)*

• William R. Buechner, Director of Econom-ics and Research, American Road andTransportation Builders Association (ap-pearance on January 30, 1998)*

• Letitia Chambers, Chambers AssociatesIncorporated (appearance on September16, 1998)*

• G. Edward DeSeve, Acting Deputy Direc-tor for Management, Office of Manage-ment and Budget (appearance on June 26,1998)

• Senator Pete V. Domenici (NM) (January30, 1998)*

• Robert Eisner, William R. Kenan ProfessorEmeritus of Economics, Northwestern Uni-versity (appearance on April 24, 1998)*

• Senator Michael B. Enzi (WY) (appearanceon January 30, 1998)*

• Financial Executives Institute: Proposalon Federal Capital Asset Budgeting. Thisproposal was mentioned in the testimonyof James J. Abel on May 8 (see entryabove for James J. Abel). The proposalwas presented to the House GovernmentReform and Oversight Committee in May1995*

• Gary Gensler, Assistant Secretary for Fi-nancial Markets, Department of the Treas-ury, with Roger Anderson, Deputy Assist-

ant Secretary for Federal Finance (appear-ance on April 24, 1998)*

• Edward M. Gramlich, Member, Board ofGovernors, Federal Reserve System (ap-pearance on March 6, 1998)*

• Representative Amo Houghton (NY) (Jan-uary 23, 1998)*

• Martin Ives, formerly a member of theGovernmental Accounting StandardsBoard (GASB), which establishes account-ing standards for State and local govern-ments, and the Federal Accounting Stand-ards Advisory Board (FASAB), which rec-ommends accounting standards for theFederal Government (appearance on May8, 1998)*

• Raymond G. Kammer, Director, NationalInstitute of Standards and Technology,Department of Commerce (appearance onMarch 6, 1998)

• Representative Dennis J. Kucinich (OH)(February 4, 1998)*

• Senator Frank R. Lautenberg (NJ) (ap-pearance on January 30, 1998)*

• Frank E. Lewis, Price Waterhouse (ap-pearance on May 8, 1998)*

• David Mosso, Chairman, Federal Account-ing Standards Advisory Board (appearanceon May 8, 1998)*

• Delegate Eleanor Holmes Norton (DC) (ap-pearance on January 30, 1998)*


• Representative James L. Oberstar (MN)(appearance on January 30, 1998)*

• June E. O’Neill, Director, CongressionalBudget Office, with James Blum, DeputyDirector (appearance on April 24, 1998)*

• Ralph R. Peterson, Chairman, Construc-tion Industry Round Table (December 3,1998)*

• David Plavin, President, Airport CouncilInternational (appearance on March 6,1998)

• Paul Posner, Director, Budget Issues, Gen-eral Accounting Office (appearance onMarch 6, 1998)*

• Franklin D. Raines, Director, Office ofManagement and Budget (appearance onApril 24, 1998)

• Martin A. Regalia, Ph.D., U.S. Chamberof Commerce, (June 5, 1998) *

• Robert D. Reischauer, Senior Fellow, Eco-nomic Studies, The Brookings Institution(appearance on April 24, 1998)*

• Edward G. Rendell, mayor of Philadelphiaand Chairman, Rebuild America Coalition,with James Lebenthal, Vice Chairman,Rebuild America Coalition, and WilliamBertera, Executive Director, RebuildAmerica Coalition (appearance on January30, 1998)

• Robert E. Rubin, Secretary of the Treasury(appearance on March 6, 1998)

• Charles L. Schultze, Senior Fellow Emeri-tus, The Brookings Institution (appearanceon April 24, 1998)*

• Lawrence F. Skibbie, President, NationalDefense Industrial Association (June 22,1998)*

• Representative John M. Spratt, Jr. (SC)and Richard Kogan (January 30, 1998)*

• Rick Swanson, Executive Director, Insti-tute of Management Accountants (appear-ance on May 8, 1998)*

• Ronald L. Tillett, Secretary of Finance,Commonwealth of Virginia (appearance onMay 8, 1998). Mr. Tillett also distributedthe ‘‘Six-Year Capital Outlay Plan:1998–2004’’ for the Commonwealth of Vir-ginia.*

• Senator Robert G. Torricelli (NJ) (appear-ance on January 30, 1998)

• George Troup, Deputy Chief of Mission,Embassy of New Zealand (appearance onMay 8, 1998).* In addition to his initialwritten statement, Mr. Troup later pro-vided written answers (June 18) to certainquestions asked during the testimony. Atthe testimony he made available five sup-plementary documents:

Graham C. Scott, Government Reform inNew Zealand, International MonetaryFund, Washington DC, 1996;

New Zealand Treasury, Financial State-ments of the Government of New Zealand,For the Nine Months Ended 31 March1998 (1 May 1998);

New Zealand Treasury, A Guide to theManagement of Departmental Fixed Assets,1991;

New Zealand Treasury, A Guide to theManagement of Departmental Purchasing,1991; and

New Zealand Treasury, A Guide to theManagement of Departmental WorkingCapital, 1991.

• Carol Cox Wait, President, Committee fora Responsible Federal Budget, June 4,1998*

• Representative Robert E. Wise, Jr. (WV)(appearance on January 30, 1998)*

61

APPENDIX DOUTLINE OF MATERIAL ON THE WEBSITE

http://www.whitehouse.gov/pcscb

Report of The Commission

Letter of TransmittalAcknowledgementsPrefaceExecutive SummaryWhat is ‘‘Capital’’?Trends in Capital SpendingBudgeting CapitalDo Current Budget Conventions Distort Decisions About Federal Capital Spending?RecommendationsPros and Cons of a ‘‘Cap’’ on Capital SpendingA Capital Budget and DepreciationOther Versions of a Capital BudgetConclusionEndnotesSelected References and BibliographyAppendix A: Executive Order 13037: Commission to Study Capital Budgeting, and the

AmendmentsAppendix B: Commission MembershipAppendix C: List of Witnesses and Written StatementsAppendix D: Outline of Material on the Website

Supplementary Materials

I. White House Press ReleasesII. Activities of the Commission

A. Summary of the Activities and Process of the CommissionB. Meetings

1. Schedule2. Agenda3. Minutes

III. Testimony And StatementsA. List of Witnesses and Written StatementsB. One-page Staff SummariesC. Written Testimony and Statements

IV. Bibliography of Selected Published Materials Reviewed by the CommissionV. Commission Working Papers

The papers in this section were prepared by staff and submitted to the commission for itsinformation and consideration. The commission has not necessarily endorsed the viewsexpressed in the papers.A. A Matrix of Capital Budgeting Problems and OptionsB. Options Presented for Discussion

Planning1. A Longer Term Perspective for Capital Needs


Supplementary Materials—Continued

2. Information on Federally Owned Capital Assets3. Using Benefit/Cost AnalysisBudgeting4. Dedicated Revenues for Selected Capital5. Investment Targets6. Full Funding for Capital Acquisitions7. Capital Acquisition Funds8. Cost of Labor and Support Services9. Budget Treatment of Leases and PurchasesAcquisition10. AcquisitionManagement in Use11. Capital Is Often Poorly Maintained12. Disposition of Capital Assets

C. Controversies Surrounding Capital Budgeting1. Defining Capital Spending2. Assessing Possible Biases Affecting Investment3. Analyzing Macroeconomic Outcomes

D. Synopses of Selected Programs or ProjectsTransportation1. Capital Management in the Coast Guard2. Historical Surface Transportation Program Funding and Impact of Recent Re-

authorization Bill on Funding DecisionsDefense3. Department of Defense Acquisition ProcessesScience4. National Institute of Standards and Technology Campus5. Space6. Synopsis of Budgeting for Investments in New Facilities for the Department of

Energy7. National Science Foundation ResearchInformation Technology8. Internal Revenue Service (IRS) ModernizationHuman Investment9. WIC (Special Supplemental Nutrition Program for Women, Infants, and Chil-

dren)Water Resources10. Water Resources Development Projects

E. How States Budget for CapitalF. Synopses of Accounting Information

1. Basic Information about Federal Financial Standards and Statements2. Federal Accounting Standards for Capital Investment3. Consolidated Financial Statements of the United States Government for FY

19974. Financial Management Status Report and Five-Year Plan

G. Other Reports1. Synopsis of the Capital Programming Guide2. Recommendations of the Grace Commission on Capital Budgeting and Planning3. Depreciation Methods

63APPENDIX D


4. Budget Impact of a Capital Budget Framework5. Generational Accounting6. Revised Asset Scoring Rule for BEA Scoring

VI. Bibliography of Recent Efforts to Improve Federal Budgeting for CapitalA. Recent Laws Regarding Budgeting

1. The Congressional Budget and Impoundment Control Act of 19742. Federal Capital Investment Program Information Act of 19843. Balanced Budget and Emergency Deficit Control Act of 19854. Budget Enforcement Act of 19905. Federal Credit Reform Act of 19906. Chief Financial Officers Act of 19907. Government Performance and Results Act of 19938. Omnibus Budget Reconciliation Act of 19939. Federal Acquisition Streamlining Act of 199410. Government Management Reform Act of 199411. Federal Acquisition Reform Act of 199612. Information Technology Management Reform Act of 1996 (Clinger-Cohen Act)13. Balanced Budget Act of 1997

B. Governmental Studies Regarding Capital Budgeting1. Report of the President’s Commission on Budget Concepts, Washington, DC,

1967.2. President’s Private Sector Survey on Cost Control (Grace Commission), Report

on Federal Management Systems, ‘‘FMS 5: Capital Budgeting,’’ September 1983,pp. 96–111.

3. General Accounting Office reports, 1989–1998. GAO issued many reports in thisperiod that discussed budgeting for capital, including the following:

Budget Policy: Prompt Action Necessary to Avert Long-Term Damage to theEconomy (GAO/OCG–92–2) (June 1992).Budget Issues: Incorporating an Investment Component in the Federal Budget(GAO/AIMD–94–40) (November 1993).Budget Issues: The Role of Depreciation in Budgeting for Certain Federal In-vestments (GAO/AIMD–95–34) (February 1995).Budget Issues: Budgeting for Federal Capital (GAO/AIMD–97–5) (November1996).Executive Guide: Leading Practices in Capital Decision-Making. (GAO/AIMD–99–32) (December 1998).

4. Report of the National Performance Review, Creating a Government that WorksBetter & Costs Less, Washington, September 1993, p.111. See also the accom-panying report, Improving Financial Management, ‘‘FM12: Manage Fixed AssetInvestments for the Long Term,’’ September 1993, pp. 67–72.

C. Progress in Implementing Modifications in Budgeting for Capital1. OMB Circular A–94: ‘‘Guidelines and Discount Rates for Benefit-Cost Analysis

of Federal Programs,’’ (October 29, 1992).2. Executive Order 12893: Principles for Federal Infrastructure Investments (Janu-

ary 26, 1994).3. OMB. Statement of Federal Financial Accounting Standards (SFFAS) No. 6, Ac-

counting for Property, Plant, and Equipment (November 1995); and SFFAS No.8, Supplementary Stewardship Reporting (June 1996).

4. OMB Circular A–130: ‘‘Management of Federal Information Resources,’’ Trans-mittal Memorandum No. 3, February 8, 1996.



5. OMB Memorandum 97–02, ‘‘Funding Information Systems Investments,’’ Octo-ber 25, 1996. This memorandum is also known as ‘‘Raines Rules,’’ because OMBDirector Franklin D. Raines issued the memorandum.

6. OMB. Capital Programming Guide (Supplement to Part 3 of OMB Circular A–11), July 1997.

7. OMB. Budget of the United States Government, Fiscal Year 1999 (February1998).a) ‘‘Principles of Budgeting for Capital Asset Acquisitions,’’ in the AnalyticalPerspectives volume of the Budget, p. 140.b) Chapter 6: ‘‘Federal Investment Spending and Capital Budgeting,’’ of theAnalytical Perspectives volume of the Budget. This chapter includes projectionsof both budget authority and outlays to four years beyond the budget year, asection on major Federal capital proposals in the budget, the ‘‘Principles ofBudgeting for Capital Asset Acquisitions,’’ estimates of capital stocks and de-preciation, and a section on capital budgeting.c) Proposal of Budgeting for Results, Budget volume, page 44.

8. Department of the Treasury. Consolidated Financial Statements of the UnitedStates Government (most recently, March 31, 1998).

9. OMB Circular A–11 (Part 3): ‘‘Planning, Budgeting, and Acquisition of CapitalAssets,’’ (July 1, 1998).

10. Agency Strategic Plans and Annual Performance Plans. These agency plans areprepared pursuant to the Government Performance and Results Act of 1993 andguidance in OMB Circular A–11, Part 2: ‘‘Preparation and Submission of Strate-gic Plans and Annual Performance Plans,’’ July 1, 1998.