Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Conclusions and Recommendations The first section below presents the committeeâs conclusions on experi- ence with stimulus spending in general, the role of transportation in a stimulus program, and planning and managing a transportation spend- ing stimulus program. The recommendations, in the second section, propose changes in established transportation programs that would make them more useful as instruments for counteracting economic downturns, provisions that would add to the usefulness of any future transportation stimulus program, and methods for evaluating transpor- tation projects in a stimulus spending program. CONCLUSIONS Effectiveness of Stimulus Spending Conclusions about transportation stimulus spending depend on assump- tions about the general effectiveness of fiscal policy. Estimates of the mag- nitude of the effects of stimulus spending vary over a wide range. However, empirical studies support the conclusion that federal stimulus spending, during a recession or period of high unemployment and when monetary policy is maintaining low interest rates, leads to an increase in gross domes- tic product and in employment, at least in the short run (within 1 or 2 years after the spending). The low and high estimates of output multipliers (the ratio of change in gross domestic product to the amount of the stimulus spending) that the Congressional Budget Office (CBO) uses in its reports on the effectiveness of the American Recovery and Reinvestment Act of 2009 (ARRA) (CBO 2012, 6) are indicative of the uncertainty in empiri- cal findings. The low and high CBO estimates are 0.4 and 2.2 for federal 108
Conclusions and Recommendations 109 government transfers to state and local governments for infrastructure, 0.5 and 2.5 for purchases of goods and services by the federal government, and 0.4 and 2.1 for transfer payments to individuals. Recent research based on experience since 2008 tends to support values of the multiplier above one as applicable when unemployment is severe and interest rates are being held near zero. The conclusions that follow concerning the effectiveness of transportation spending compared with other kinds of stimulus spend- ing and concerning the effects of the design of a stimulus program do not depend on the magnitude of aggregate multipliers, within this range. Value of Transportation Spending in a Fiscal Stimulus Program Credible grounds exist for including transportation capital expenditures as a component of a federal fiscal stimulus program, once the decision has been made to undertake such a program: â¢ If projects are selected with proper consideration of the value of the trans- portation services they will provide, the long-term benefits will offset the initial cost, so the expenditure may be justified even if the magnitude of the stimulus benefit is smaller than expected. Including transportation investment in a stimulus provides a hedge against the possibility that the government spending multiplier is smaller than expected. â¢ Stimulus that consists of accelerating planned expenditures (for example, planned road construction) adds less to public debt than expenditures that would not have been made in the absence of the need for stimulus. If the stimulus program changes only the timing of spending over the business cycle but not the total of spending, the long-term benefits and costs of the transportation system are little affected. The public gains the stimulus benefit of accelerated spending in a recession for a small cost (the cost of deviating from the schedule of expenditures that the transportation agency otherwise would have determined to be optimal). â¢ A diversified package with infrastructure as one component is a reason- able strategy because the relative sizes of the multipliers for different forms of stimulus are not well known. â¢ Transportation infrastructure improvement, by adding to the pro- ductive capacity of the economy, may raise consumersâ and investorsâ expectations for economic growth, providing an immediate stimulus
110 Transportation Investments in Response to Economic Downturns effect beyond that produced by equal expenditures for nonproductive purposes. â¢ Construction prices are likely to be lower during a recession, allowing transportation agencies to buy more with the funds available. Materi- als prices may have moderated, and there will be less competition for contractorsâ services and for use of construction equipment. â¢ In recessions in which the construction industry is strongly affected, as it was in 2007â2009, infrastructure spending may be well-targeted as stimulus. â¢ Although few estimates of multipliers for specific categories of gov- ernment spending are available, the research indicates that infra- structure spending can be at least as effective as other categories of stimulus spending in terms of short-term economywide employment and income impact. The objections raised historically to the use of transportation or infrastructure spending as stimulus must be balanced against the possi- ble advantages listed above. The main arguments in opposition are that, first, the time required to enact legislation and commence construction will delay spending beyond the time when the stimulus is needed; sec- ond, local governments will substitute federal funds for their own funds, reducing the net increase in spending; third, accelerating spending increases the risk of poor project selection decisions; and finally, public works projects have relatively high government budget cost per job (e.g., compared with public service employment programs). Another category of objections concerns the risk of disrupting existing transportation funding and governance arrangements if programs enacted as emer- gency measures become entrenched (i.e., the risk that increased reliance on federal funding and erosion of the established practice of paying for most transportation infrastructure with user-derived revenues will tend to discourage prudent investment). Each of these objections is examined below, for ARRA and for stimulus spending programs in general. Timeliness The timeliness of stimulus spending depends on the lag between the beginning of the economic downturn and the enactment of the measure
Conclusions and Recommendations 111 and on the lag in spending after enactment. ARRA was enacted in Feb- ruary 2009, 14 months after the onset of the 2007â2009 recession and 4 months before its end. Fifty-six percent of ARRA funds administered by the U.S. Department of Transportation (U.S. DOT) were expended through March 2011 and 73 percent through March 2012. Federal outlays for transportation lagged other categories of ARRA stimulus: the 73 percent paid out by March 2012 in U.S. DOTâadministered programs compares with 90 percent for all forms of stimulus in ARRA. Actual transportation construction spending occurred earlier than the federal expenditure is recorded in these data, because the federal gov- ernment reimbursed the states for expenditures they had already made, whereas in some ARRA programs (e.g., aid to individuals) most actual spending probably occurred after the federal outlay. Moreover, there is evidence that planned government spending has a stimulus effect as soon as it is announced and before government outlays occur. Unlike the previous postwar stimulus programs that included public works spending, ARRA was enacted before the recession that motivated them had ended, and nearly all of the funds ARRA provided, including most of the transportation funding, have been spent in a period when unemployment has been much higher than the prerecession level. The relative timeliness of the ARRA transportation spending is more the consequence of the duration of the recession and the slow rate of recov- ery than of speedier enactment and execution of the programs. The long duration of the downturn was foreseen in 2008 and conceivably influ- enced decisions on the amount of infrastructure spending in the act. The ARRA experience shows that transportation stimulus spending can be timely if the duration of the economic downturn can be predicted with some reliability and is expected to be protracted. Only 7 percent of ARRAâs $8 billion for grants to states for intercity passenger rail had been paid out by March 2012; obstacles to implement- ing this program are identified in the section below on management of the transportation stimulus programs. Thirty-two percent of highway funds remained unspent in March 2011 and 12 percent by March 2012; this rate of spending may represent a reasonable compromise between the goals of spending quickly and spending on worthwhile highway proj- ects to ensure long-term benefits.
112 Transportation Investments in Response to Economic Downturns Fiscal Substitution The effectiveness of federal transportation grants as stimulus depends on how the grants affect state and local transportation spending. A state or local agency that receives a federal transportation stimulus grant might reduce the amount of its own resources devoted to transporta- tion because other uses of the funds now have higher priority or because the agency lacks sufficient administrative capacity to spend all of the increased funding within the time limit of the grant. The substitution can occur over time. That is, a project funded with a stimulus grant may replace a project that a state had planned to con- struct in a future year with full or partial state funding; consequently, the state might maintain present-year spending at the level planned before ARRA but reduce its spending in the future year. When a transporta- tion capital expenditure is moved forward, the transportation benefit is earlier use of the constructed facility and the cost is the difference in the present value of the cost of construction today and the cost of construction later. If the future decline in spending occurs in a time of low unemployment, the stimulus impact (i.e., the short-term impact on employment and income) of the federal grant is not diminished. How- ever, if the acceleration of transportation capital spending under ARRA slowed spending while unemployment was still high in 2011 and 2012 (compared with the level of spending that would have occurred in the absence of ARRA), the net stimulus would be reduced. Maintenance-of-effort requirements, such as those imposed on ARRA transportation grants, have been difficult to implement, and their effect on state spending is uncertain. Substitution of federal ARRA funds for state funds would not necessarily indicate bad faith on the part of the grant recipients (who were required to certify maintenance of effort), because recipients have no certain means of determining what their spending priorities would have been in the absence of the grants. The Economic Development Administration evaluation of the 1976â 1977 federal stimulus programs concluded that substitution for local spending was substantial. There is also evidence that, historically, in the regular federal highway aid program, increases in the amount of federal highway aid distributed to the states have not led to equal increases in state highway spending. These earlier estimates of high rates of fiscal
Conclusions and Recommendations 113 substitution may not be applicable to the ARRA transportation grants. Most of the public works grants in the 1976â1977 stimulus program were extraordinary grants for nontransportation purposes, and no maintenance-of-effort requirement was imposed on the recipients. The observed substitution in the federal-aid highway program is a long-term effect occurring over the multiyear cycle of the program. In the short run, the structure of federal and state transportation funding poses obstacles to substitution in response to an extraordinary federal grant. Most state transportation spending is funded by revenue from dedicated tax sources. Some statesâ laws provide for emergency diversion of these funds, but it seems likely that overall, the trust fund arrangement tends to hinder fiscal substitution. The ARRA grants may have discouraged some states from enacting rate increases for the taxes that supply dedi- cated transportation revenue, but changes in these tax rates are infre- quent, so this form of substitution probably was not major. Fiscal substitution as a state response to ARRA grants would not nec- essarily have been harmful. States should know which spending and tax adjustments in response to the recession are best from the standpoint of the value of public services. If substitution is significant, the federal grants may nonetheless have a stimulus effect, but the magnitude of the effect will depend on how state and local governments adjust their spending and taxation in response to the aid rather than on the characteristics of the con- struction projects the recipients designate as funded by the aid. For exam- ple, if states responded to federal transportation grants by diverting some of their own funds from transportation to restoring their reserve funds, the stimulus effect of the federal aid would be diminished because the state had diverted funds from spending to saving. Federal restrictions on substitution (or incentives not to substitute) in a stimulus program are justifiable only if the federally designated uses of aid funds are believed to have greater value as economic stimulus than the uses that recipients would select. If Congress were to decide in a future stimulus program that the best use for a specified portion of federal stimulus spending would be trans- portation infrastructure, more effective measures to limit fiscal substi- tution than the maintenance-of-effort certification provision of ARRA might be available. First, the federal government could spend directly for transportation, but direct federal spending would be practical only for
114 Transportation Investments in Response to Economic Downturns the federally provided systemsâwaterways and air traffic control. Sec- ond, the federal government could motivate state and local spending on transportation infrastructure through challenge grants requiring state or local matching funds, but such a program would have difficult-to- predict effects on total state spending and taxation and consequently on the stimulus impact of the federal program. Improved understanding of state and local fiscal substitution in response to federal transportation grants would be useful not only in planning a future fiscal stimulus program but also in managing the regu- lar federal transportation aid programs. The recommendations below include a proposal for research to measure these responses. Risk of Poor Project Selection Decisions ARRAâs deadline for obligating transportation funds affected recipientsâ choices of the kind of spending to undertake with the grants. Highway and transit funds had to be obligated within 18 months of enactment (compared with up to 4 years allowed in the regular federal-aid pro- grams). Pavement resurfacing and bus purchases were important uses of funds, in part because they could be undertaken quickly. A large share of these expenditures had been planned in advance of ARRA and therefore met the investment criteria of the transportation agencies. Nevertheless, it is likely that, without the constraint of the deadline, the agencies would have been able to devote some of the funds to projects with greater long- term value. This conflict between the goals of long-term benefit from the services provided by the transportation improvement and short-term stimulus benefit is unavoidable. ARRAâs deadline may have been a rea- sonable compromise. The use of the federal transportation programs as instruments for stimulus spending may present a second kind of conflict with the qual- ity of transportation investment decisions in the long term. The ARRA transportation programs as well as other federal actions to sustain the Highway Trust Fund since 2008 have relied on a series of special contri- butions from the general fund rather than on revenue from user taxes and fees. In the same period, the capacity of these traditional revenue sources to sustain federal and state surface transportation programs deteriorated, and the federal transportation aid program was allowed
Conclusions and Recommendations 115 to lapse and then reauthorized in 2012 for an abbreviated term in sub- stantially altered form. The historically established structure of the fed- eral and state surface transportation programs is regarded by the state transportation agencies as having important advantages, including sta- bility of funding that allows long-term planning and public acceptability because user funding is perceived as fair. The structure also provides the discipline of a budget constraint that helps prevent total spending from greatly exceeding the value to users of the facilities provided. Frequent departure from the historical institutional and financial practices of the programs poses a risk of loss of these advantages. Budget Impact per Job The federal budget cost per job directly supported in U.S. DOTâ administered ARRA programs, according to grant recipient reports, was $318,000 per person year of employment, compared with $162,000 for all ARRA outlays for which recipient reporting of employment is required. ARRAâs recipient reporting requirements invited such comparisons; how- ever, the recipient-reported jobs data are poor measures of impact. They include employment by the recipients, their contractors, and subcontrac- tors, but they exclude employment by vendors (e.g., suppliers of con- struction materials for highway projects) and the effects on employment of increased consumption by newly employed workers. On the other hand, the net effect on employment could be lower than the recipientsâ reported jobs if recipients substitute a portion of federal funds for their own funds or if spending the federal funds crowds out other spending (e.g., by attracting workers with skills that are in short supply away from other projects). Therefore, the relatively high cost per direct job reported for the ARRA transportation programs does not show that these pro- grams were less cost-effective as employment stimulus than the others. CBOâs 2010 comparison of fiscal policy options, described in Chap- ter 2, ranked infrastructure spending against other stimulus options in terms of person years of employment per dollar of federal budgetary cost. Employment was the estimated total economywide impact of the spending, not just direct jobs as in the ARRA recipient reports. Infra- structure spending ranked higher than or comparably with nine of the 11 options considered. This ratio also is an imperfect measure of
116 Transportation Investments in Response to Economic Downturns the relative merit of alternative forms of stimulus because it takes no account of benefits in addition to stimulus that the alternative forms of stimulus spending would provide. To compare alternative expenditures that are intended to produce both benefits in the form of fiscal stimulus and benefits from transportation or other public services provided, a common unit of measure for the two kinds of benefits is needed. Evidence of Stimulus Effectiveness of Transportation Spending Only a few empirical studies report estimates of government expendi- ture multipliers applicable during recessions or periods of high unem- ployment and disaggregated by kind of government spending. Available estimates, summarized in Chapter 2, indicate that multipliers can vary substantially among spending categories and that infrastructure spend- ing can be at least as effective as other categories of stimulus spending. Transportation spending may have some special advantages that con- tribute to its stimulus impact. In recessions such as that of 2007â2009, in which the construction industry is strongly affected, infrastructure spending may be well-targeted as stimulus. (However, none of the research reviewed evaluated whether such targeted spending improves stimulus effectiveness.) Construction spending also may be more useful when a protracted recovery period can be reliably predicted, because the start-up lag in construction is not then a serious drawback. The relative effectiveness of different forms of stimulus will depend on economic circumstances, including the depth of the recession, the duration of the recovery, and possibly the sectors and regions most severely affected. Design of a Transportation Stimulus Program In summary, the conclusion of the preceding section is that, once a deci- sion to undertake a fiscal stimulus is made, transportation is appropriate as a component of a diversified program, especially if economic conditions indicate that the downturn is likely to be prolonged. Transportation spending would be less likely to be a useful stimulus in response to the kind of short-duration recession with rapid recovery of job losses that was typical before the 1970s because providing and spending the funds in time would be difficult. Transportation spending would not necessar- ily be less timely than other forms of government purchases in response
Conclusions and Recommendations 117 to a short recession. Moreover, in view of the time required historically for enactment of stimulus programs, whether any form of ad hoc fiscal stimulus program could be very effective in response to a short recession with quick recovery is open to question. Once a decision to include transportation is made, the planners of the program must determine the share of the total stimulus package to be devoted to transportation spending, the allocation of funding among categories of uses (e.g., in ARRA, the allocation among roads, transit, rail, aviation, and water and the exclusion of noncapital expenditures), and the details of grant eligibility rules and administration. Conclusions on the success of these design features in ARRA and lessons for any future stimulus programs are presented below. Transportation Share of the Overall Program The experience of ARRA indicates certain practical limits on the amount of funding for transportation that could be used effectively in any future stimulus program. Limits are imposed by state and local transportation agenciesâ capacities to manage an unanticipated surge in funding, by the capacity of the transportation construction industry, by federal admin- istrative agenciesâ capacities to manage and oversee new grant programs or expansions of existing programs, and by the opportunities available to invest the stimulus funds in worthwhile transportation projects. The optimum transportation share also depends on the relative stim- ulus effects of alternative kinds of government purchases and of other forms of fiscal stimulus. However, the practical capacity limits probably will be a more relevant guide than relative sizes of multipliers of different kinds of purchases. The funding increment that ARRA provided to highway and tran- sit programs was moderate compared with total spending in these pro- grams, and the basic administrative procedures of the regular federal-aid programs were already in place. For these reasons, the stimulus spend- ing does not appear to have been seriously constrained by the agenciesâ administrative capacities, although processing ARRA grants apparently slowed the statesâ processing of regular federal-aid highway grants. Similarly, the ARRA experience does not indicate that the value of transportation grants was seriously diminished by lack of worthwhile
118 Transportation Investments in Response to Economic Downturns transportation projects, according to the project selection criteria of the recipient transportation agencies. However, the ARRA obligation dead- line probably diverted some funds from uses that would have had greater transportation benefits. The slow rate of spending of the ARRA passenger rail grants and Trans- portation Investment Generating Economic Recovery (TIGER) multi- modal discretionary grants illustrates the importance of administrative capacity in the federal administrative agencies and among the grant recipi- ents. Both grant programs were new, and procedures had to be established for developing proposals, awarding grants, and monitoring performance. Total annual government spending on highways and transit in the United States is small in comparison with the total stimulus provided by ARRA. Therefore, even very large percentage increases in annual highway and transit spending could constitute only a minority of spending in a future comprehensive stimulus program comparable in size with ARRA. The committeeâs recommendations, presented in the final section of this chapter, include proposals for increasing the capacity of public agen- cies to manage transportation stimulus spending. Allocation of Funding Within the Transportation Program ARRA allocated the total amount of transportation funds provided among highways, transit, passenger rail, other modes, and the TIGER competitive grant programs. Passenger railâs share of ARRA transporta- tion funding was much greater than its historical share of federal trans- portation aid. Also, the ratio of transit funding to highway funding was slightly higher than historically. These allocations reflected Congressâs judgment of national investment priorities; considerations of the rela- tive stimulus effects of spending on the various modes do not appear to have been influential. The allocation has affected the stimulus impact of the program because spending of rail funds has been slow, but there is little basis for judging whether the stimulus impact of the funds that have been spent differed by mode. Similarly, whether the expenditure types that the obligation deadline favored (pavement resurfacing and bus purchases) have greater or lesser stimulus impact than the kinds of projects (e.g., bridge construction) that were difficult to undertake with ARRA funds is not known.
Conclusions and Recommendations 119 Administrative Rules Administrative rules governing grants to state and local government in the ARRA transportation programs concerning allocation of funds, maintenance of effort, timeliness of spending, and record keeping and reporting were intended to ensure the effectiveness of the spending as stimulus and to maintain accountability and public credibility. These are essential design objectives for a federal stimulus spending program. Fed- eral imposition of such requirements is appropriate because Congress is seeking to attain national objectives, which are not necessarily consistent with the objectives of the individual jurisdictions receiving grants. The ARRA rules affected the stimulus impact and long-term transpor- tation benefit of the spending. Some of the rules were objects of recipient complaints that they added costs or complexity to program administration and interfered with the recipientsâ ability to make best use of the funds. The form of the ARRA rules was influenced by the experience of dis- appointing results with earlier federal public works stimulus programs. Taking advantage of the ARRA experience in designing administrative rules for any future stimulus spending programs will be important. The committee was not able to evaluate the ARRA rules definitively. How- ever, it identified certain rules that require careful consideration before they are included in any future stimulus program: â¢ The recipient jobs reporting requirement, which produced data of lim- ited applicability for evaluating the success of ARRA and added to grant recipientsâ administrative burden. Recipients found requirements for reporting to multiple federal agencies especially burdensome. â¢ The requirement for giving priority to projects in distressed substate areas, which may not be an effective mechanism for targeting the unem- ployed and may have had only small impact on the location of projects. â¢ Exclusion of noncapital expenditures from eligibility. Following the rules of the established federal transportation aid programs, ARRA transportation grants could be used only for capital expenditures (with the exception of a small share of transit aid). Although evidence is not available on the relative stimulus impacts, in some circumstances, spending for operations and maintenance might be more stimulative than capital expenditures. (For example, maintenance spending might be more quickly carried out.)
120 Transportation Investments in Response to Economic Downturns The recommendations below include proposals for alternative rules con- cerning maintenance of effort, data collection and reporting, and match- ing requirements. RECOMMENDATIONS The committeeâs recommendations are in three areas: first, possible changes in the structure of the established federalâstate transporta- tion funding programs that would expand the industryâs capacity to absorb stimulus spending, if Congress decides to take such an action in the future, and that would help to stabilize government transportation spending over the business cycle; second, design and administration of any future transportation stimulus program; and third, evaluation of the effectiveness of transportation stimulus spending. 1.â Expand Transportation Agency and Construction Industry Capacity to Absorb Stimulus Spending Congress and the states should consider changes in finance, administra- tion, and planning practices that would increase the capacity of trans- portation programs to maintain spending in recessions and to absorb efficiently any future temporary federal assistance, such as the ARRA transportation programs, intended as economic stimulus. State and local government total capital spending and state and local noncapital transportation spending are procyclical; that is, they tend to accelerate when the economy is growing strongly and to deceler- ate during downturns. Transportation spending in the past may not have been as affected by the business cycle as other spending categories because of its reliance on dedicated revenues, the historical stability of the federal-aid program, and the multiyear schedules of many capital projects. Reinforcing the features of transportation spending that pro- vide stability over the business cycle or changing the program to facili- tate accelerated spending during recessions could have benefits for the economy overall and for the transportation system. For the economy, such changes could make a small contribution toward moderating the busi- ness cycle and mitigating recessions. Provisions to support transportation spending during recessions could benefit the transportation program
Conclusions and Recommendations 121 by avoiding disruption of construction schedules, taking advantage of lower construction prices during a recession, and avoiding the loss of skilled Â orkers in Â ransportation agencies and in the highway construc- w t tion industry. Fiscal policy is a federal responsibility. To the extent that avoiding cyclical swings in spending in transportation programs could improve the performance of these programs, the states also have an interest in stability. Modifications in the Primary Federal and State TransportationÂ Programs Modifications of established programs that Congress and the states should consider to enable transportation agencies more readily to main- tain or increase spending during economic downturns could include the following: â¢ Providing stability in the established federal transportation funding programs. The federal transit and highway aid programs operated on the authority of a series of short-term extensions from the expiration of the previous multiyear authorization in 2009 until 2012. The pro- gram similarly subsisted on short-term extensions from 2003 to 2005, between authorizations of multiyear programs. Failure to maintain revenue from user taxes or from other reliable sources has been the underlying source of instability. â¢ Maintaining a larger backlog of projects with completed designs and environmental reviews. Providing stability in aid programs and review of relevant federal-aid program rules are actions by the federal gov- ernment that could increase the willingness of state and local govern- ments to maintain project backlogs. Modification of the federal rule requiring a state to pay back federally reimbursed preliminary engi- neering costs of any highway project not started within 10 years of design may be required to avoid discouraging states from maintaining a larger backlog. â¢ Building a balance in the federal and state transportation trust funds during normal economic periods to be able to maintain or increase spending during recessions. â¢ Greater use of borrowing from future user-derived revenue (i.e., use of bonds backed by future tax revenue dedicated to transportation,
122 Transportation Investments in Response to Economic Downturns toll revenue, or anticipated federal grants) to maintain spending when current revenue slows. â¢ Implementing reforms to speed project delivery in the regular federal- aid surface transportation programs. (The 2012 federal surface trans- portation authorization legislation contains provisions for this purpose.) â¢ Legislative provision of standby authority to the executive to increase federal transportation aid to the states through the established pro- grams by specified amounts when specified economic conditions occur. The additional aid could be debited to the Highway Trust Fund, to be made up from future revenue or by drawdown of reserves. Each of these actions would require evaluation before a decision was made to implement it. Risks of such a policy, from the standpoint of the transportation program, include the possibility that some projects would be delayed, with loss of benefits, if funds were reserved for reces- sion spending; the possibility that unspent funds would be diverted to nontransportation programs; overindebtedness; and the possibility that states would reduce their transportation spending if they came to expect increased federal aid, drawn from the general fund, during recessions. Building a reserve fund would be challenging politically because it would require either an initial decrease in spending or an increase in revenue (e.g., through an increase in user tax rates). Some of these measures could have a budgetary impact on trans- portation programs. Increasing the balance in trust funds would require increasing revenue or reducing outlays for a period. Increasing the backlog of project designs would increase engineering expenditures for a period, until the backlog reached the desired level, although some of the extra expenditures eventually would have been made even if the backlog had not been increased. Furthermore, the added inven- tory would incur carrying costs, including the cost of obsolescence as plans age and the cost of having capital tied up in the plans. The increase in inventory also could have benefits, beyond the potential macroÂeconomic benefit of facilitating countercyclical spending. Trans- portation agencies would have more information about investment opportunities and would be able to make better use of unanticipated availability of funds (e.g., if a planned project were to fall through or if revenues unexpectedly increased).
Conclusions and Recommendations 123 Federal Competitive Assistance Program for Stabilizing Spending A competitive federal transportation grant and credit assistance program could provide a supplementary mechanism for accelerating spending dur- ing recessions. Congress should consider establishing a federal program that would coordinate the planning and funding of major infrastructure proj- ects that are economically justified and require federal participation. The program could award assistance at any time and would have the additional charge of increasing assistance quickly when defined criteria were satisfied to help sustain transportation construction during economic downturns. It would maintain a backlog of projects ready to be advanced when needed. It would operate through partnership with state and local government trans- portation agencies and could take the form of an extension of the charge of existing federal competitive grant and loan programs. The program would be in operation in advance of a future recession so that the administrative agency and eligible grant applicants would be experienced with its work- ings and a backlog of projects would have been identified. Distributing most ARRA transportation funds according to the appor- tionment formulas and procedures of the existing federal formula grant programs facilitated timely spending. Enactment was eased because Congress did not need to design and debate alternative allocations or administrative arrangements. Delay was minimized in the U.S. DOT agencies administering the grants because established administrative procedures could be followed for the most part. Finally, state and local agencies knew approximately how much funding to expect and faced no special application requirements to receive the bulk of the funds. However, distributing a portion of transportation funds in a stimulus program through competitive grants could help to achieve the objec- tives of the ARRA transportation grant rules. The arrangement would overcome the three main problems of transportation stimulus spending: â¢ Timeliness: Projects in the pipeline or under way would be immedi- ately available for accelerated funding. â¢ Risk of poor project selection: The administrative agency already would have evaluated project proposals and would maintain the back- log of eligible projects. â¢ Maintenance of effort: Because of their size or the need for federal participation, the projects funded would likely not be initiated by
124 Transportation Investments in Response to Economic Downturns grant recipients on their own during a recession (but would likely be carried out at some point in the future because of their merit). Several federal competitive transportation assistance programs are in operation or have been deployed in the past, including the TIGER grants in ARRA, the Federal Transit Administrationâs Transit New Starts program, the Transportation Infrastructure Finance and Inno- vation Act credit assistance program, and the Projects of National and Regional Significance competitive grant program in the 2012 surface transportation act. In the existing competitive grant programs, states and local governments submit proposals to the U.S. DOT for grants or credit assistance for capital expenditures for any kind of transportation project now eligible for federal aid. Proposals follow a standard format requiring credible estimates of costs and benefits and a financial plan that includes nonfederal contributions. The proposed new federal assis- tance program could take the form of an extension of the charge of these existing Âprograms that allowed them to coordinate their procedures and to Â ccelerate assistance during economic downturns. The scale of these a existing competitive programs (totaling a few billion dollars annually in grants and loans) would be appropriate for the scale of a counterÂ cyclical assistanceÂ Â rogram. Also, the administrative procedures of these p programs provide models for how a new assistance program could be organized. Any new federal involvement in selection of transportation infrastruc- ture investments, decisions that historically have been made at the state and local level, should be carried out cautiously. An earlier Transpor- tation Research Board policy study committee that proposed a federal competitive grant program for freight infrastructure identified potential drawbacks of such programs: the risk to transportation system efficiency if the federal program entails greater availability of subsidies; the risk that reduced involvement of state and local entities, which in most cases have the greatest stake in the project and will have responsibility for its operation and maintenance, will lead to poor project selection decisions; and limited capacity in federal agencies for evaluating project benefits and costs (TRB 2009, 162, 195, 207, 220). The earlier committee pro- posed guidelines with regard to funding, project eligibility, and evalua- tion to mitigate these problems (TRB 2009, 263â271, 277â280).
Conclusions and Recommendations 125 2.â Rationalize Design of Future Transportation Stimulus Spending Programs The five recommendations below concern first the need for early prepa- ration for a stimulus program and then each of the four main catego- ries of administrative rules in the ARRA transportation programs: grant eligibility and allocation, maintenance of effort, timeliness of spend- ing, and record keeping and reporting. As noted above, the rules were intended to ensure the effectiveness of the spending as stimulus and to maintain credibility with the public. Advance Rulemaking Rules needed for the U.S. DOT to administer the ARRA transporta- tion program were not in place at the time of enactment. Uncertainty about rules was not conducive to timely spending. To minimize delays in any future stimulus program, Congress should authorize the U.S. DOT to publish rules on maintenance of effort, project eligibility, and data reporting that would be available for future application. Allocation of Grants Any future transportation stimulus spending program should continue ARRAâs practice of allocating most of the funds according to the pro- cedures and formulas of the established federal transportation grant programs. This procedure was critical to the timeliness of the ARRA spending and avoided controversy. An allocation method that took into account interstate differences in economic conditions (as was the practice in some previous infrastruc- ture stimulus spending programs) might be able to increase the stimu- lus impact of the spending. However, the added complexity could lead to delay, and analysis methods for estimating the stimulus impact and effect on long-term transportation benefits of alternative allocations are not available. The ARRA requirement that each state favor economically distressed areas in selecting projects was poorly conceived and probably ineffective in amplifying the stimulus impact of the spending. Maintenance of Effort The maintenance-of-effort requirement in the ARRA transportation grant program was difficult to administer, and some state officials believe
126 Transportation Investments in Response to Economic Downturns the requirement was not applied uniformly. The effect of the require- ment on state spending is unknown. Any future transportation stimulus program should include an objective standard for maintenance of effort defined in terms of the grant recipientâs planned spending in specified categories. The definition should reflect the impact of declines in dedi- cated revenue on spending capacity of state and local transportation programs. Before more rigorous maintenance-of-effort requirements are imposed in a future transportation stimulus grant program, the effect of such requirements on the stimulus impact and transportation benefits of the program should be evaluated. Timeliness Requirements ARRAâs deadline for obligating federal grants for state and local highway and transit projects was not unreasonable in light of Congressâs concern with timeliness and the tardiness of spending under past public works stimulus programs. State officials reported to the committee their belief that they could have obtained greater long-term transportation benefits from the ARRA funds by undertaking projects that would have spent the funds over longer periods if deadlines had been longer. However, longer deadlines might have reduced the stimulus impact. In the design of any future transportation stimulus program, consid- eration should be given to alternative or additional means of ensuring timeliness: â¢ Providing multiple obligation deadlines, for example, a short dead- line applicable to a portion of the total funding to ensure immediate spending and deadlines equivalent to those in the regular federal-aid highway and transit programs for the remainder to allow states to undertake more complex projects with the stimulus funds; and â¢ Allowing temporary accelerated review and approval processes to speed construction for certain kinds of projects. The proposals in the previous section for changes in the regular federal- aid programs to make them more useful mechanisms for stimulus spending also would help ensure timely spending. Deciding how rigorous the timeliness requirements should be requires a method for objectively evaluating the trade-off between stim- ulus benefit and long-term transportation benefit. The risks of adverse
Conclusions and Recommendations 127 consequences from acceleration of environmental reviews or planning procedures would also need to be evaluated. Record-Keeping and Reporting Requirements Data collecting and reporting requirements in the ARRA transporta- tion grant programs that did not have clearly demonstrated useful- ness in managing the programs or evaluating their results should not be imposed in future programs. As part of its assessment of the ARRA programs, the U.S. DOT should prepare a plan for conducting neces- sary oversight of the performance of transportation grant recipients in any future transportation stimulus spending program to provide a basis for data collection and reporting requirements. The plan should specify methods that the U.S. DOT will use for monitoring timeliness of spend- ing and maintenance of effort and methods to be used by grant recipients for estimating transportation benefits of the spending. Requirements for duplicative reporting to multiple federal entities should be avoided. 3.â Measure the Effect of Federal-Aid Program Changes on Recipient Actions and Program Benefits The effect of the ARRA federal transportation grants on total state and local government transportation spending, total spending, and spend- ing priorities cannot be definitively assessed with the available evidence. The U.S. DOT should conduct research on how changes in the level of federal aid provided and in the rules of the federal-aid programs (e.g., with regard to matching shares and project eligibility) affect the spend- ing decisions of grant recipients. With congressional approval, part of this research could be experimental; that is, a federal program change could be enacted as a trial, with provision for evaluation of its effects. Reliable measurement of the direct impact of federal transportation stimulus grants on state and local government spending is a necessary first step in assessing economywide stimulus impacts. 4.â Define a Method for Balancing the Recovery and Reinvestment Goals of Transportation Stimulus Spending Most decisions in designing or managing a transportation stimulus spending program depend on balancing the stimulus benefit (immediate
128 Transportation Investments in Response to Economic Downturns job creation) against benefits of the future transportation services pro- vided by the facilities constructed. This balance must be taken into account in deciding whether to devote a portion of spending to trans- portation in a diversified stimulus program like ARRA and in the selec- tion of projects to be built with the funds allocated to transportation. For a given budget, the portfolio of projects that produces the greatest immediate employment is not likely to be the same as the portfolio that maximizes transportation benefits. No objective method for evaluating this trade-off was available to guide planning or management of ARRA. As part of its assessment of the ARRA transportation programs, the U.S. DOT should define a method for evaluating the combined transpor- tation and stimulus benefits of projects in a unified framework. Defining a single metric capturing all benefits in order to rank alternative programs or projects might ultimately be useful, although challenging. A necessary first step will be to develop quantitative estimates of each category of impact (including stimulus benefits and other benefits) of alternatives. As an illustration, the following are the steps for estimating each category of impact (stimulus benefits and other benefits) of program alternatives at the stage of planning the transportation component of a future stimulus package: 1. Define the alternative stimulus rules (for example, stringent timeli- ness constraints versus more permissive rules such as dividing the total of grant funds into short-term and longer-term categories). 2. Project (e.g., on the basis of experience with ARRA) how each of the alternative sets of rules would affect the mix of transportation proj- ects that the states would carry out during and after the period of stimulus grants. 3. Evaluate the transportation benefits of the mix of projects that would be carried out under each set of alternative rules. 4. Evaluate the stimulus benefits of each of the alternative project portÂ folios under assumptions about the duration and depth of the antici- pated recession. The largest difference might arise from the timing of spending. Policy makers could then be presented with a matrix of quantified impacts (short-term employment and income impact, long-term trans- portation benefit quantified in dollars, and possibly other short- and
Conclusions and Recommendations 129 long-term effects) for each of a set of stimulus program alternatives. Selection of the best alternative would rely on judgment but would be based on more systematic information about trade-offs than has been available in the past. Reducing the disparate benefits to a single metric would entail assign- ing an economic value to jobs created and incorporating this value into a benefitâcost analysis of the projects built in the alternative programs. The value of job creation would depend on the stage of the business cycle and would reflect the value that society places on helping individuals facing economic distress. Other categories of stimulus spending also yield benefits that are dis- tinct from their stimulus effectâthey may provide services or directly aid persons in distress. Decisions about the composition of a stimulus package will be made by elected officials, who must weigh the various impacts of the spending alternatives. Explicit analysis of both stimulus impact and other forms of benefit would give elected officials more balanced and useful information to guide decisions about allocating funds in a stimulus package. REFERENCES Abbreviations CBO Congressional Budget Office TRB Transportation Research Board CBO. 2012. Estimated Impact of the American Recovery and Reinvestment Act on Employ- ment and Economic Output from October 2011 Through December 2011. Feb. TRB. 2009. Special Report 297: Funding Options for Freight Transportation Projects. National Academies, Washington, D.C.