A Review of DOT Compliance with GASB 34 Requirements (2004)

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.

16 CHAPTER 4 CASE STUDY OVERVIEW PROCEDURES Case study interviews were conducted with six state DOTs—three (i.e., Michigan, Tennessee, and Washington) were modified approach states, two (i.e., South Carolina and Vermont) selected the depreciation approach, and one (i.e., Texas) used both approaches for different major asset classes. Four of the interviews (i.e., South Carolina, Tennessee, Texas, and Washington) were conducted on site during all- day sessions. In the interest of economy, two interviews (i.e., Michigan and Vermont) were conducted via extended con- ference calls. To facilitate comparison of the interview results, a structured question-and-answer format, tied to survey top- ics, was developed by the research team. A generic version of this interview structure was previously submitted to the NCHRP panel for review and comment. The panel approved the structure with some comments and suggestions, which were incorporated into the document. Individualized inter- view structures were then developed for each of the six states, tied to each state’s specific responses to the electronic survey. The individualized interview questions were then trans- mitted to each of the states at least 2 weeks before the inter- view date to permit them to conduct any necessary research and to identify the appropriate personnel to take part. For the on-site interviews, the research team conducted a series of sessions with each of the Department offices involved in GASB 34 implementation and met with a representative from the State Comptroller’s Office or other member from the cen- tral state government. These sessions were generally con- ducted without monitoring from the Department finance office in order to encourage a candid expression of views. Other office representatives were asked only those questions that affected their areas of responsibility. Although the interviews benefited from the structure of prepared questions, the research team did not rigidly follow this format. If a topic was of particular interest to a state rep- resentative or that participant’s state had employed an espe- cially noteworthy approach, the researchers tried to follow that discussion wherever it might lead. It was thought that the particulars thus obtained were more important than ensuring that each particular question was rigidly addressed by each state. All interviews were conducted by at least two, and sometimes three, members of the research team. On the DOT side, participants ranged in number from 2 to 10. Following the interviews, the research team prepared a summary of the remarks and provided these to each state for review and comment. All comments suggested by the states were accepted and incorporated into the reports, as reflected in Appendix F to this report (which is available in NCHRP Web Document 63). OVERVIEW The selection of approach—modified or depreciation— was influenced by many factors, but the most important appears to be the maturity of the DOT’s asset management informa- tion system. Those DOTs with information systems in place that generated the information necessary to support the mod- ified approach tended to select that approach; absent such sys- tems, the depreciation approach was preferred. The detailed questionnaires that follow this section address a wide range of GASB 34 implementation issues, while this overview focuses on seven issues judged to be of particular interest: • Implementation procedure/organization—each of the interviewed DOTs had some involvement with central state finance entities (e.g., comptroller’s office, division of finance and administration, or state auditor), but in all cases the DOT’s infrastructure assets represented the overwhelming majority of the state total. The DOTs pri- marily relied on internal committees to conduct the work with little or no contribution from new hires or consul- tants. An early start was generally deemed to be critical to success, with 18 months prior to the close of FY02 the typical beginning date. • Determination of condition targets—for modified approach states, the calculation of meaningful condition targets was a key step in the process, and various meth- ods were employed. Different condition targets were adopted for the two principal asset classes, roads and bridges; targets were not established for other asset classes. Fiscal constraint was an important aspect in determining these targets in order to ensure that they were realistic. • Estimated cost to achieve targets—linking targeted conditions to required expenditures is problematic for the modified approach states. The principal difficulties

17 are that (1) the management information systems, as deployed at the time, were not mature enough to gener- ate reliable estimates and (2) the cost definitions con- tained in GASB 34 were not compatible with the DOTs’ budget practices and management systems. For several reasons, the DOTs are working to improve their perfor- mance on the first issue; there is less interest in address- ing the second. Several DOTs made the point that they view the comparison between planned and actual expenditures as much less significant than the compari- son between targeted and actual conditions. • Categorization of costs—a related source of difficulty is the categorization of costs among capital, preserva- tion, and maintenance. The GASB 34 guidelines use a functional approach to these categories—maintenance costs achieve the original design life; preservation costs extend that design life, but do not increase capacity or service; and capital costs increase capacity or service. However, the traditional DOT definitions relate more to type of construction—a full reconstruction project is viewed as capital whether or not lanes are added; a resur- facing project is viewed as preservation whether or not there are ancillary safety benefits. A potential solution is to allocate costs within a project to the three cate- gories, but this is strongly resisted as impractical by the DOTs, which typically must account for hundreds, if not thousands, of projects each year. The conflict between traditional and GASB 34 categories is particularly pro- nounced for the division between capital and preserva- tion. Accordingly, this discrepancy is less significant for depreciation approach states because those states capi- talize both categories, but it is important for modified approach states where preservation costs are supposed to be expensed. Some DOTs have questioned the degree to which the difference between the structures is mate- rial in the overall financial statements. • Estimated historical costs—in order to reduce the bur- den associated with estimating historical costs, GASB 34 deferred the effective date of this requirement for 4 years and further provided that it was necessary to report costs incurred only since July 1, 1980. However, each of the interviewed states elected to report histori- cal costs in the initial year of GASB 34 implementation, and four of the six went back to well before 1980. Esti- mates were prepared using a combination of AASHO: The First 50 Years, financial statements, project records, and written down replacement costs. • Additions and retirements—the reporting of additions to and retirements of infrastructure assets in the financial statements was troublesome for many DOTs. Although DOTs typically track such changes in physical assets through management programs and other inventory rec- ords, a link to costs recorded in the accounting system typically did not exist. Many states initially found it nec- essary to achieve this link manually and with other ad hoc methods. Most have now developed automated links. • Usefulness of and interest in information—there was general agreement among the interviewed DOTs that the information generated in the GASB 34 exercise was potentially useful in budgeting and resource allo- cation, particularly the information included in Required Supplemental Information for the modified approach states. However, all reported that in FY02 that benefit remained a potential and was not realized—the new information generated virtually no interest or inquiries outside the agencies. One DOT observed that this is just the first year of implementation. It expects that general interest will increase as the accuracy of the infor- mation improves and as a time series of validated data becomes available and trends can be observed. MICHIGAN Michigan DOT (MiDOT) decided on the modified approach on the basis of its long tradition (25 years) of asset manage- ment and its commitment to the asset management philosophy. The following items from the interviews were deemed to be particularly noteworthy: • MiDOT received guidance from the state comptroller and state auditor, but the bulk of the work was completed within the Department and was accomplished without benefit of a formal committee structure. • A conservative approach to selecting the condition targets was used, with the targets well below current conditions. • The Department had difficulty in computing the estimated expenditure levels necessary to achieve these targets, in part because the GASB structure and definitions for the Required Supplementary Information were not compati- ble with MiDOT budget and management procedures. • MiDOT categorized a full reconstruction as a capital cost, even if the number of lanes remained the same, on the theory that full reconstruction inevitably improves service levels based on current design standards. The Department did not differentiate between preservation and maintenance expenditures. Such differentiation was not deemed necessary because both categories are expensed in the financial statements. • For historical costs, MiDOT indexed all expenditures back to the average year of construction. • Unlike many states, Michigan did not have difficulty in accounting for additions to and retirements from the infra- structure network. The Department used 250 work type codes to classify a project as either capital or preservation/ maintenance (see case study for listing). Costs are ini- tially recorded as “construction in progress” and then transferred to the appropriate accounts on project com- pletion. The Department did not attempt to allocate costs within a single project to capital and preservation/ maintenance components, judging that to be adminis- tratively prohibitive.

• The Department reports that, thus far, GASB 34 has not had a measurable effect on resource allocation and bud- geting decisions, nor have Department staff received any expressions of interest in the new information from the legislature or the general public. SOUTH CAROLINA In selecting the depreciation approach, South Carolina DOT’s (SCDOT’s) primary concern was to comply with GASB 34 in the most efficient and effective manner. SCDOT concluded that its management systems were not adequate to support the modified approach, with the key weakness being data for secondary roads. The road valuation system for sec- ondary roads is being updated, which may lead to selection of the modified approach in the future. Another concern, however, was the potential effect of not meeting condition targets on funding levels. The following items from the inter- views were deemed to be particularly noteworthy: • GASB 34 implementation began in 2000 and was over- seen by a statewide committee with consultant support. However, 95% of the assets were SCDOT’s and only SCDOT assets were classified as infrastructure. • As a depreciation approach state, SCDOT did not target condition levels or estimate the costs to achieve such targets. • As a depreciation approach state, SCDOT grouped expen- ditures for preservation and capital projects together. It did not experience difficulty in differentiating these proj- ects from maintenance expenditures. • For historical costs, SCDOT used AASHO: The First 50 Years for 1914–1964 expenditures (with current replace- ment value used to allocate costs among roads, bridges, and right-of-way), internal financial statements for 1964 to the early 1990s and current expenditure records from the early 1990s to 2002. • To account for additions to and retirements of infra- structure assets, SCDOT modified its procedures so that, for GASB purposes, project closing occurs when expen- ditures are complete (rather than when open to traffic). Additions and retirements are recorded once per year by journal entry. • SCDOT acknowledged that the depreciation approach provides less meaningful information because it is incon- sistent with the Department’s preservation program and the nature of infrastructure assets. However, it sees a danger in modified approach data being used to make inappropriate comparisons with other DOTs or budgets. Accordingly, SCDOT recommends that more detailed standards be developed for condition assessments and disclosures to minimize the possibility of unfair dis- closures caused by inconsistencies. 18 TENNESSEE Tennessee DOT (TDOT) assumed from the outset that it would use the modified approach and did not seriously con- sider the alternative. Existing management systems—in par- ticular, the Tennessee Road Information Management Sys- tem (TRIMS), the bridge management system (PONTIS), and the Maintenance Management System (MMS)—gener- ate the necessary data, and the asset management approach is consistent with the Department’s philosophy. The follow- ing items from the interviews were deemed to be particularly noteworthy: • Tennessee began implementing the modified approach earlier than most—in mid-1999, just before the formal publication of GASB 34. It was essentially a TDOT effort, but the Department met early with the Depart- ment of Finance and Administration and the state audi- tor. An internal TDOT committee initially met quarterly, but more frequently toward the end. • Condition targets for bridges were based on 75% of the deck area being neither structurally deficient nor functionally obsolete (the FHWA National Bridge Index goal). Condition targeting for roads was a new ven- ture and was based on the Maintenance Rating Index (MRI) produced by the MMS. The MRI is determined for 1/10 mile road segments based on a sample size of 7% annually. • The link between targeted conditions and required expen- ditures is weak. Neither MMS nor PONTIS, as currently deployed, is mature enough to generate reliable cost esti- mates. For FY02 and FY03, estimated costs are based on projections derived from historical funding patterns. • Projects are categorized as either capital or preservation/ maintenance, and all costs within the project are in one category or the other. Allocation of costs within a proj- ect is not considered practical with over 1,500 new proj- ects per year. A full reconstruction, even without addi- tional lanes, is treated as capital based on the theory that current design standards will always generate significant benefits. On the other hand, a resurfacing project that might have safety benefits is classified as preservation/ maintenance. • TDOT historical costs were derived in three tiers: AASHO: The First 50 Years for 1914–1964 with con- struction costs allocated among roads, bridges, and right- of-way based on current replacement cost; high-level appropriation codes for 1964–2001; and project-level information beginning in 2001. • In order to account for retirements from infrastructure assets, it was necessary to modify TRIMS. This was done manually for FY02 and FY03. In the future, TRIMS will generate a report on lane miles removed. • TDOT believes that the information generated by the GASB 34 exercise is potentially useful, but, thus far, has

received no expressions of interest from elected offi- cials or the general public. However, the Department notes that this is just the first year of implementation. It expects that general interest will increase as the accu- racy of the information improves and as a time series of validated data becomes available and trends can be observed. TEXAS Texas DOT (TxDOT) used the modified approach for most of its assets, but selected the depreciation approach for bridges and some minor asset classes. This combination approach was used because the bridge management system (developed in house) includes a good inventory from which to make depreciation calculations, but does not have asset management functions. Another factor was that TxDOT views bridges as having a more definable lifecycle than roadways, so bridges are more appropriate to the depreciation calculation. If TxDOT develops a bridge management system with the necessary functionality, it might consider a shift to the modified approach for bridges. The following items from the interviews were deemed to be particularly noteworthy: • The Comptroller’s office led the state’s overall GASB 34 implementation committee, including a Capital Asset team in which TxDOT participated. The committees met for about 11/2 years—an early start was a key to suc- cess. A Capital Asset Guide was prepared and is avail- able on the internet. In addition, an internal TxDOT committee met throughout the period. • TxDOT developed condition targets for roadways based on a condition assessment system that sampled condi- tions on 5% of the network (10% for Interstates). The targets are fiscally constrained and are approved for- mally by the Transportation Commission. • The correlation between targeted conditions and esti- mated required expenditures is weak and will require further development over the next few years. However, TxDOT believes that the comparison between targeted and actual condition levels, as opposed to the compari- son between planned and actual expenditures, is the more meaningful aspect of the exercise. • Full reconstruction of a roadway, even without addition of lanes, was categorized as a capital expense, as was a dualization project that included both new road con- struction and overlay of an existing roadway. There was no attempt to allocate the costs of such projects between preservation and capital because the effort would be extensive and cost prohibitive. • Historical costs for highways were estimated through a combination of AASHO: The First 50 Years and TxDOT financial reports. For bridges, the current year replace- ment cost for each bridge category, measured by square 19 feet of bridge deck, was indexed back to the year of construction. • Unlike many states, TxDOT did not report difficulty in accounting for additions to and retirements of infrastruc- ture assets. There is now an annual entry to the account- ing system for construction-in-progress and fixed asset classes. The threshold for an addition is when 85% of anticipated project expenditures have occurred. • TxDOT believes that the information generated by the GASB 34 exercise is potentially useful for resource allocation decision-making, but, to date, there has been virtually no interest in this information by elected offi- cials or the general public. The Department is disap- pointed that its significant effort in seriously addressing GASB 34 has produced little benefit other than com- plying with the requirement. VERMONT The Vermont Agency of Transportation (VTrans) selected the depreciation approach, primarily because its asset man- agement system does not meet the specific requirements of the modified approach under GASB 34. Further, VTrans views finance and accounting as inherently separate from asset man- agement issues and does not perceive an advantage in includ- ing reports on asset condition and related matters in the finan- cial statements. The following items from the interviews were deemed to be particularly noteworthy: • During the implementation process, VTrans worked with the Director of Statewide Financial Reporting and the State Auditor, but there was not a statewide committee. Within VTrans, a steering committee composed of the Director of Administration and the Director of Program Development/Chief Engineer made the decisions and directed activities, which were carried out by a working committee. • As a depreciation approach state, VTrans did not target condition levels nor estimate the costs to achieve such targets. • As a depreciation approach state, VTrans grouped expen- ditures for preservation and capital projects together. It did not experience difficulty in differentiating these proj- ects from maintenance expenditures and followed the GASB 34 guidelines without exception. • 1980–1993 historical cost estimates were derived from project ledger data with some adjustments to exclude maintenance costs and to allocate among asset classes. From 1994 to the present, the project cost system includes detailed object codes that simplify the conver- sion to asset classes. • Additions to and retirements of infrastructure assets by asset class represented new information that had to be developed by the preparation of forms for each project.

These forms identified, by asset class, which project costs were for new or replacement of infrastructure. • The State of Vermont has not yet issued an FY02 Com- prehensive Annual Financial Report (CAFR) because of a change in its accounting system. Accordingly, the GASB 34 disclosures are not yet publicly available. VTrans believes that the information generated in the modified approach would be useful in preparing budgets and making the case for funding infrastructure preser- vation. Some of this same type of information might be developed under the depreciation approach. WASHINGTON Washington DOT (WSDOT) used the modified approach based on a good asset management system already in place. It did not want to use a different methodology for GASB 34 purposes. However, absent the existing pavement and bridge management systems, it may have been encouraged to adopt the depreciation approach. The following items from the interviews were deemed to be particularly noteworthy: • WSDOT employed a relatively structured implementa- tion procedure organizationally, using both an executive committee and a working committee that held quarterly meetings with the State Office of Financial Management. • Condition targets were derived from budget allocations that were then translated into expected conditions. This was a fiscally constrained approach following a method- ology developed years ago. • The Required Supplementary Information (RSI) compar- ison between planned and actual expenditures was diffi- cult because the asset management system is organized 20 by program, rather than by cost category. The reported amounts were derived from budgetary information, less amounts capitalized in the preservation program. • Regarding cost categorization, the GASB definition of preservation presented difficulties. For asset management purposes, WSDOT considers the complete reconstruc- tion of a roadway as a new asset that should be capital- ized. Reconstruction inevitably results in improvements reflecting current design standards. Allocation of proj- ect expenditures between capital and preservation was considered impractical, with hundreds of projects per year. WSDOT suggests that GASB evaluate a more sophisticated approach for categorizing capital, preser- vation, and maintenance costs and that FHWA defini- tions be considered. • For historical costs, WSDOT went back to 1980 and used actual capital outlay costs in construction programs. Cost data were accumulated from different accounting and asset management data sources and required ad hoc reporting. • Before GASB 34 there was no need to link asset man- agement data with financial statements. Thus, it was dif- ficult in FY02 to account for additions to and retire- ments of infrastructure assets. WSDOT developed ad hoc reporting to accomplish this accounting; in the future it is anticipated that the link will be automated. • WSDOT hopes that in the future the GASB 34 informa- tion will be useful in making the case for funding infra- structure preservation (although this did not occur in the initial year of implementation). In this regard, WSDOT questions the wisdom of requiring a shift to the depreci- ation method if condition targets are not being met—a failure to meet condition targets should be publicly dis- closed, and this would be lost under depreciation.