Performance audit of the Arizona Department of Transportation's urban highways program for Maricopa County

              KPMG Peat Marwick
Management Consultants
Focused experience
for measurable results
Manufacturing and
Technology Industries
Financial Institutions
Health Care
Government Services
Insurance
Real Estate and Hospitality
Airports
Retail and Wholesale
Energy and Utilities
Corporate Transactions
Technology and Operations
Compensation and Benefits
State of Arizona
Office of the Auditor General
PERFORMANCE AUDIT OF THE
ARIZONA DEPARTMENT OF
TRANSPORTATION'S URBAN
HIGHWAYS PROGRAM FOR
MARICOPA COUNTY
Final Report
August 1991
DOUGLAS R. NORTOIU, CPA
AUDilOH GENERAL
STATE OF ARIZONA
OFFICE OF THE
AUDITOR GENERAL
September 9, 1991
Members of the Arizona Legislature
The Honorable Fife Symington, Governor
Mr. Charles Cowan, Director
Arizona Department of Transportation
Transmi tted herewi th is the report A Performance Audit of the Arizona
Department of Transportation's Urban Highwavs Prosram For Maricopa County.
This audit was conducted by the consulting firm KPMG Peat Marwick under
contract with the Auditor General and was in response to a May 13, 1991
resolution of the Joint Legislative Budget Committee.
A response to the audit from the Department of Transportation follows the
Executive Summary.
This report w i l l be released to the public on September 10 at 9: 00 a. m.
Sincerely,
DRN : I mn
Enc l osu re
2700 NORTH CENTRAL AVENUE " SUfTE 700 ' PHOENIX, ARIZONA 85004 ' ( 602) 255- 4385 FAX ( 602) 255- 1251
Peat Marwick
Certified Public Accountants
Suite 800
8150 Leesburg Pike
Vienna. VA 22182
Telephone 703 442 0030
August 30, 1991
Mr. Doug! as. Norton
Auditor General
Office of the Auditor General
2700 N. Central Avenue
Suite 700
Phoenix, Arizona 85004
Dear Mr. Norton:
KPMG Peat Marwick, in association with Dewberry & Davis and Linderlake Corporation, is
pleased to submit our final report for the PerjCormurlce Audit of the Arizona Department of
Transportation's Urbun Highways Program for Muricopu County. This report assesses
ADOT's management performance in carrying out the MAG Urban Highways Program
under original Program conditions, and recommends ways to improve ADOT's efficiency
and effectiveness in fulfilling its future overall Program responsibilities. The performance
audit responds to questions raised concerning ADOT's administration of the MAG Program
relative to the following five issue areas:
Original and current excise tax revenue forecasting processes
m Original and current program cost estimates
Priority programming process
Program management practices
m Right- of- way acquisition practices
This report presents the findings, conclusions, and recommendations resulting from the
analysis of questions comprising these issue areas.
K MPe at Marwick
Mr. Douglas Norton
Office of the Auditor General
August 30, 1991
2
We appreciate the opportunity to have assisted you and your staff in the conduct of this
performance audit. We also appreciate the cooperation we received from all those who
participated in the fact- finding portions of the performance audit effort.
Very truly yours,
KPMG Peat Marwick
PERFORMANCE AUDIT OF THE ARIZONA
DEPARTMENT OF TRANSPORTATION'S
URBAN HIGHWAYS PROGRAM FOR MARICOPA COUNTY
FINAL REPORT
CONTENTS
Executive summary
Introduction
Background
Objective and scope of the performance audit
Approach
Organization of report
Overview and status of MAG regional freewaylexpressway program
MAG plan and priorities
MAG program funding
MAG program costs
MAG program status
Review of excise tax revenue forecasting procedures
Review of program cost estimates
Review of priority programming process
Review of program management practices and procedures
Review of right- of- way acquisition laws and practices
Conclusions
Appendix A - Performance audit questions
Appendix B - List of interviewees
PERFORMANCE AUDIT OF THE ARIZONA
DEPARTMENT OF TRANSPORTATION'S
URBAN HIGHWAYS PROGRAM FOR MARICOPA COUNTY
FINAL REPORT
EXHIBITS
2- 1 MAG Regional Transportation Plan
2- 2 Freewaylexpressway priorities for excise tax and 15 percent
revenues adopted by MAG Regional Council January 29, 1986
2- 3 Priorities for MAG excise tax and 15 percent revenues adopted
by MAG Regional Council October 24, 1990
3- 1 Estimate of the annual amount of transportation excise tax
revenues to be raised
3- 2 Excise tax growth assumptions
3- 3 ADOT forecasting model - major economic and demographic variables
3- 4 Initial legislative staff estimates and initial ADOT forecasts of
excise tax revenues
3- 5 Total Maricopa County excise tax revenues - actual revenues and
forecasts
3- 6 Total Maricopa County personal income forecasts compared with
actual income
3- 7 Total Maricopa County dollar volume of building permit forecasts
compared with actual building permits
3- 8 Annual rate of excise tax revenue growth compared with forecasted
growth rate
3- 9 ADOT planning forecast of excise tax for FY 1991- 95 program
3- 10 Maricopa County RARF bonds issued to date
3- 11 Maricopa County RARF debt service and excise tax revenues
EXHIBITS
4- 1 Early estimates of regional transportation plan mileage and costs
4- 2 Regional Transportation Plan right- of- way cost estimates
4- 3 Pre- Proposition 300 transportation and population projections
4- 4 Original design features established for MAG freewaylexpressway
plan by MAGTPO in early 1985
4- 5 Original design features established for Proposition 300 plan
by ADOT Urban Highways Section in November 1985
4- 6 Post- Proposition 300 transportation and population projections
4- 7 MAG freewaylexpressway system population and VMT comparisons by
designlyear
4- 8 Forecasted average daily traffic volumes for MAG freewayl
expressway plan by corridor and design year
4- 9 Peak daily traffic volumes by type and size of highway
4- 10 Comparison of freewaylexpressway lanes planned versus lanes
required by corridor
4- 11 Per mile costs of urban highway right- of- way and construction
4- 12 Sample unit costs for urban freeway right- of- way and construction
4- 13 1991 design features for MAG freewaylexpressway program corridors
4- 14 Changes in design features for MAG freewaylexpressway plan
4- 15 Additional acreage for freeway- to- freeway interchanges
4- 16 Comparison of MAGTPO and ADOT cost estimate for MAG freewayl
expressway program
4- 17 Current cost estimates for MAG Program by corridor and function
EXHIBITS
4- 18 MAG freewaylexpressway program design status as of June 20, 1991
4- 19 MAG freewaylexpressway program construction contracts awarded
in fiscal year 1991
4- 20 Succeeding construction cost estimates for Agua Fria freeway
corridor
5- 1 Freewaylexpressway priorities for excise tax and 15 percent
revenues adopted by MAG regional council January 29, 1986
5- 2 Priorities for MAG excise tax and 15 percent revenues adopted
by MAG regional council October 24, 1990
5- 3 Comparison of MAG Program to statewide five- year highway
construction program budgets
5- 4 Original and current estimates of total MAG Program costs by
function
6- 1 Comparison of MAG Program to statewide five- year highway
construction program budgets
6- 2 ADOT urban highways section organization
7- 1 Summary of ADOT survey of state transportation agencies using
the " before and after" method of appraisal
7- 2 States permitting " before and after" method of appraisal by
statute or case law
7- 3 States prohibiting " before and after" method of appraisal by
statute
7- 4 MAG program right- of- way acquisition costs per acre by corridor
7- 5 Percentage of MAG program right- of- way acreage acquired by
corridor
EXHIBITS
7- 6 Remaining right- of- way costs by MAG program corridor
7- 7 Initial and current MAG program right- of- way cost estimates
7- 8 MAG program right- of- way settlements without trial
7- 9 MAG program right- of- way court awards based on trial reports
7- 10 MAG program right- of- way requests for authority to condemn
EXECUTIVE SUMMARY
This executive summary presents the objective, approach, and key conclusions and
recommendations for the Peflormance Audit of the Arizona Department of Transportation's
Urban Highways Program for Maricopa County.
Background
In 1985, the Arizona State Legislature enacted House Bill 2306 authorizing Maricopa
County voters to approve up to a ten percent increase in existing excise taxes designated as
a transportation excise tax ( commonly referred to as a one- half ( 112) cent sales tax). The
law enabled County residents to set aside excise tax revenue for the design and construction
of a metropolitan highway system. An urban highway plan for metropolitan Phoenix was
developed by the Maricopa Association of Governments ( MAG). The MAG Program called
for development of 233.5 miles of freeways and expressways to supplement the 86.5 miles
of completed or under- construction highways in the region over the period 1986 to 2006.
The additional excise tax to finance the MAG Program was overwhelmingly approved by
the voters in October 1985. The proceeds of the excise tax are to be used to design,
acquire right of way for, and construct freeways, expressways, and parkways in the MAG
plan. The 233.5- mile MAG Program is the largest urban freewaylexpressway development
program currently underway in the nation.
In December 1990, ADOT released a five- year progress report on the Maricopa County
transportation excise tax entitled, Maricopa County Transportation Excise Tax Review,
1986- 1990. This report noted that over $ 3 billion in additional revenue will be needed to
complete the MAG Program. This revenue shortfall coupled with other concerns about
Arizona Department of Transportation's ( ADOT's) management of the program resulted in
the Arizona State Legislature directing the State's Office of the Auditor General to procure
an independent performance audit of ADOT's Urban Highways Program.
Objective and scope of the performance audit
The purpose of the performance audit is to assess ADOT's management performance in
carrying out the MAG Urban Highways Program under original Program conditions, and to
recommend ways to improve ADOT's efficiency and effectiveness in fulfilling its future
overall Program responsibilities.
The Office of the Auditor General directed that the scope of the performance audit include
a review of ADOT's administration of the MAG Urban Highways Program relative to the
following five issue areas:
rn Original and current excise tax revenue forecasting processes
rn Original and current program cost estimates
rn Priority programming process
rn Program management practices
Right- of- way acquisition practices
A total of 46 specific questions were detailed in the five areas listed above by the Office of
the Auditor General for analysis in the performance audit. This executive summary
presents the major conclusions and applicable recommendations from the analysis of these
questions.
The scope of the performance audit did not include a financial audit of the MAG Program.
The scope also did not include a detailed preparation of long- range traffic projections or
development of Program revenue and cost projections, or underlying assumptions.
The performance audit used extensive data and estimates developed by ADOT. Such data
and estimates were reviewed for reasonableness, but not verified by the audit team.
Approach
Our approach included:
Numerous fact- finding interviews with MAG and ADOT managers directly
responsible for the MAG Program and its implementation
rn Interviews with current and former State and local elected officials, representatives of
the business community groups that have expressed concerns with MAG Program
management, and selected others knowledgeable of the overall MAG Program
rn Compilation and reviews of extensive ADOT and MAG reports, manuals, and
working papers
Contacts with selected other state and local agencies responsible for similar highway
and financing programs
The work plan was performed by a competitively selected project team composed of
performance audit, highway engineering, right- of- way acquisition, and financial analysis and
management specialists.
Conclusions and recommendations
Sections 3 through 7 of the report present detailed analyses, conclusions, and applicable
recommendations for each of the five areas of focus of the performance audit. This part of
the executive summary highlights the major conclusions and recommendations resulting
from the analysis of the 46 questions concerning these five areas.
Excise tax revenue forecasting procedures
Conclusions
Original excise tax revenue estimates for the MAG Program were developed at the planning
level of detail by an ad hoc committee, that included ADOT, MAG, legislative, and Phoenix
Chamber of Commerce Staffs. Inflation factors prepared for cost estimation purposes were
used to develop current dollar estimates of excise tax revenues over a twenty- year
timeframe. This caused the original revenue estimates to be high. Lower, more realistic
and conservative estimates of inflation should have been used for revenue estimation.
The original revenue estimation process was adequate for internal planning purposes, but
not for serving as the sole financial basis for a public tax referendum. Legislative staff
should have used a more rigorous approach to excise tax revenue forecasting, and should
have thoroughly qualified the estimates as to their level of diligence, assumptions, and
achievability within the voters' pamphlet.
The econometric model used by ADOT to forecast excise tax revenues appears reasonably
structured. However, the model's input variables lagged the local economic downturn in
the early years of the MAG Progam, resulting in consistently high forecasts compared with
actual collections. ADOT began to recognize in 1987 that its forecasts of MAG Program
excise tax revenues were optimistic. In 1989, the Department publicly issued an estimate of
excise tax revenues that was one- third less than its estimate of 1986. In 1990, the
Department began to reflect local economic conditions in its official excise tax revenue
forecasts to produce lower, even more realistic and conservative results, particularly in the
short term.
ADOT has adequately managed the bonding process for the existing half- cent excise tax,
given the high level of bonding directed by the MAG Regional Council. As a result, the
bonds issued by ADOT for the MAG Program enjoy high ratings from the major bond
rating agencies which enable the Department to receive very favorable interest rates on its
MAG Program bonds.
Adequate revenues are expected to be available to satisfy the covenants and debt service
requirements associated with currently outstanding Regional Area Road Fund ( RARF)
bonds.
Current planning estimates for total MAG Program revenues are expected to cover only
about half of the total costs of the Program as currently defined.
Recommendations
ADOT should continue to implement the improved short- term revenue forecasting process
and develop its official revenue forecasts, based on adjustments to independent data
variables which reflect local economic conditions.
ADOT should include a disclosure statement with all estimates, projections, and forecasts of
MAG Program revenues and costs, indicating the level of diligence, assumptions,
achievability, and intended use associated with the resulting figures.
ADOT should prepare an official annual revenue forecast issuance package for public
dissemination which includes the ADOT Director's approval of the updated revenue
forecasts for the MAG Program.
Program costs estimates
Conclusions
Initial MAG Program cost estimates were developed by MAGTPO staff, ADOT staff, and
outside consulting engineers. These initial cost estimates were based largely on areawide
planning studies, utilizing generalized per- mile costs for construction, design, and
right- of- way acquisition. Due to a lack of specificity regarding design features and
alignment, these initial MAG Program cost estimates did not represent reliable costs bases
for the MAG Program.
Significant expansion of the scope of the MAG Program contributed to the escalation of
Program costs, particularly during the early formative years when the corridor location and
design concept studies were being developed. The following changes in MAG Program
characteristics and features contributed to half of the cost increases associated with the
MAG Program, not including debt service costs. The other half of the increase in Program
costs was due to: rising costs for real estate, design work, and construction activities; costs
for mitigation measures ( such as noise walls, drainage features, asbestos abatement, and
hazardous waste site management); demolition and relocation; and additional access at
interchanges.
MAG Program Features
Right- of- way acres
Total lane- miles
Expressway lane- miles
Freeway lane- miles
Traffic interchanges
Fully directional interchanges
Miles of depressed freeways
The number of lanes per corridor appear reasonable based in traffic forecasts to the design
year of 2015 for all corridors on the MAG Program, except the Estrella Freeway, unless this
corridor remains a 2- lane road.
Initial per- mile cost estimates for MAG Program right- of- way acquisition and construction
were based on historical information from similar projects in Maricopa county. However,
these cost factors were understated since they were not inflated to the timeframe of the
MAG Program. In addition, certain difficult to estimate costs were not included in these
unit costs ( such as drainage, tunnels, interchange right- of- way, demolition, asbestos
abatement, and hazardous waste site management).
As the MAG Program evolved from the planning to the design stage of development, the
nature of corridor design features and location of alignment became more finnly
established. In most cases, the resulting design features followed standard design practices
and procedures. However, in several instances higher level design features ( such as the
depressed interchange between the Outer Loop and Superstition Freeways, various depressed
freeway sections, and modular signing along freeways in Tempe) were incorporated by
ADOT based on local community concerns regarding access and mitigation of visual, noise,
and drainage impacts of the new freeways.
Since 1985, MAG Program cost estimates have increased from $ 3 billion to over $ 7 billion,
in uninflated constant dollars. The Department's first major revision to the MAG Program
costs was produced in mid- 1988, as part of a report to the State Transportation Board, based
on the completion of location and design concept studies for many of the corridors making
up the Program. Current MAG Program cost estimates are much more realistic, based on
completion of all corridor location and design concept studies, completion of construction
plans for selected segments, and performance of project- specific right- of- way acquisition
and construction activities. However, further increases in MAG Program cost are expected
during the next fifteen years as additional corridors and sections are further developed and
as inflation drives up the costs of right- of- way acquisition and construction.
Since 1989, ADOT has began to institute a number of cost saving strategies to help control
the escalating costs of MAG Program projects. Those include:
Performing value engineering of project designs to identify lower cost design strategies
to address geometric, functional, and safety requirements
Staging construction activities so that partial projects which are adequate in the short-term
can be implemented within constrained budgets
Encouraging joint- funding arrangements with local communities and developers to
expedite projects schedules
Environmental impact mitigation requirements impacted the MAG Program ( particularly the
East Papago corridor, due to the presence of a hazardous waste site in the original
alignment). However, these impacts chd not represent major contributors to the cost
increases affecting the MAG Program.
While the Department's current reporting procedures and systems permit the timely
reporting of significant MAG Program cost changes, both within ADOT and MAG, the
Department and MAG continue to lack proactive, effective mechanisms to convey this
information to the general public. Prior releases of information by ADOT to the public and
the press on the status of Program cost estimates and their potential for achievement were
not sufficiently frequent or consistently conveyed by Department management.
Recommendations
ADOT should base its estimates of MAG Program cost on historical and current
information derived from actual projects and plans, that is adjusted to a consistent constant
or current dollar basis, depending on the intended use of the information.
ADOT should continually re- evaluate the MAG Program system needs based upon updated
estimates of regional growth. Design features should then be reassessed to determine their
adequacy, based on minimum design standards, updated traffic forecasts, local concerns, and
Program budget constraints.
ADOT and MAG should provide more timely, consistent, and comprehensive information to
the public regarding the MAG Program, including progress, status, and changes to revenues
and costs.
Any significant program changes which have major priority or fiscal implications need to be
resolved through the involvement of the MAG Regional Council.
Priority programming process
Conclusions
Both MAG and the Transportation Board have statutory authority to prioritize corridors for
construction in the MAG Freeway~ ExpresswayP lan. While the Transportation Board has
the ultimate statutory authority to approve roads for inclusion in the State Highway System,
to prioritize and program highway construction activities, and to administer the bonding
process relative to MAG Program funds, MAG has the final authority to define which roads
are included in the MAG Regional FreewayiExpressway Plan, and therefore, are eligible for
excise tax funding. MAG also has statutory authority to establish the priorities for MAG
Program corridors and to suggest construction schedules for MAG Program projects. This
provides MAG with significant leverage over the project programming of MAG Program
comdors and projects.
The current statutory roles of MAG, ADOT, and the Transportation Board regarding the
priority programming of the MAG Program appear reasonable and provide a useful check
and balance mechanism for ensuring consideration of local concerns and requirements
within the context of the State's overall highway program. Three fundamental limitations in
the current priority programming process as it is applied by ADOT and MAG to the MAG
Program are:
w Lack of long- term perspective to guide short- term priority programming decisionmaking
W Inadequate public involvement/ notification regarding yearly priority programming
deliberations by MAG
Lack of accountability of the MAG Regional Council for its authority and influence
over the MAG Program
Original comdor priorities were set by MAG for the MAG Program, based on documented
analyses and recommendations from MAGTPO, and incorporated into ADOT's five- year
highway construction programs by the Transportation Board. The MAG Regional Council,
based on technical support provided by MAGTPO, used generally representative and
reasonable criteria for prioritizing the design, right- of- way acquisition, and construction of
MAG Program comdors over the 20- year life of the excise tax funding mechanism. These
criteria included:
w Traffic conditions
Cost effectiveness
Project readiness
Outer Loop completion
System continuity
Geographic balance
There has been limited public involvement in or exposure to the priority setting process for
the MAG Program. The process is primarily controlled by MAGTPO, ADOT, and MAG,
with public involvement limited to public hearings on ADOT's five- year highway
construction program and open meetings of the MAG Management Committee, MAG
Regional Council, Transportation Board, and ADOT Priority Planning Committee.
In preparing five- year highway construction programs starting with fiscal years 1987
through 1991, the Department followed the original MAG program priorities, with only five
significant exceptions agreed to by MAG:
Agua Fria - Northern Avenue to 1- 10
W Sky Harbor Expressway
Grand Avenue
W Estrella Freeway ( interim road construction)
W Hohokam Extension
In most cases, changes to MAG Program priorities were adequately justified and
documented. However, in two of the five cases studied ( Agua Fria and Sky Harbor
Expressway), priority changes lack supporting documentation.
Programming actions to address the accumulating shortfall in MAG Program revenue
growth were not taken by ADOT until 1990. This delay was caused by a number of
factors, including: the front- loading of funds during the first five years of the Program
made possible by the high bonding of MAG Program revenues; continued optimistic
forecasts of excise tax revenues; and ADOT, MAG, and RPTA concerns about the impact
of MAG Program cutbacks on the 1989 public referendum concerning the VALTRANS
excise tax proposition ( which was subsequently defeated in February 1989). Once the total
Program revenue picture was better understood and the VALTRANS excise tax proposition
removed from the public agenda, the Department began to take decisive steps to curtail the
programming of MAG Program projects, as well as reducing the planned level of future
bonding. Projects were deferred in the schedule while others were staged to allow
completion of only those portions required during the short- term. To facilitate and guide
this adjustment process, the MAG Regional Council revised its MAG Program priorities,
whereby the sequencing of project functional activities are laid out without commitment to
fixed completion schedules.
The Department's priority programming process is oriented to a maximum five- year time
horizon. However, the MAG Program extends to the year 2005 in terms of excise tax
funding. By limiting its primary focus to the next five- year period, the Department and
consequently MAG may be making program adjustments which reflect primarily short- term
considerations instead of the full, long- term objectives of the MAG Program.
The allocation of MAG Program funds to preliminary engineering, right- of- way acquisition,
construction, and debt service has reflected the policy and priority programming guidance of
MAG, which called for expedited construction schedules, consistent with its stated corridor
priorities and an aggressive bonding program. To expedite construction schedules meant
performing preliminary engineering and right- of- way acquisition as early and rapidly as
possible, consistent with funding availability. As a result, relatively higher percentages of
MAG Program funds were spent on these functions in the first five years of the Program.
In the remaining years of the MAG Program, the past percentage allocation of funds for
construction is expected to more than double, while right- of- way acquisition is expected to
consume a significantly smaller percentage of Program resources. However, debt service
costs are expected to represent almost 20 percent of the remaining costs to complete the
MAG Program.
In the early years of the Program, preliminary engineering and right- of- way acquisition
costs for segments were often programmed together, which allowed such costs to be applied
to each activity as it evolved. In more recent years, design and right- of- way costs are
programmed separately in order to better control the use of MAG Program funds relative to
their intended allocation.
Recommendations
Both ADOT and MAG should consider the full MAG Program in terms of its revenues,
costs, scope, and schedule when developing priority programming decisions and adjustments
to be reflected in the annual updates to the Department's five- year highway construction
program.
ADOT should continue and expand its matching program for encouraging local government
and private sector funding participation in return for expediting project programming, in
order to leverage existing MAG Program revenues.
The Transportation Board and the MAG Regional Council should retain their dual statutory
authority over the priority programming of the MAG Program.
MAG Regional Council should be held accountable for its role in defining the MAG
Freeway/ Expressway Plan, developing section priorities, and programrning/ scoping projects.
Public accountability can be established through the conduct of public hearings and the
issuance of an annual fiscal status report on the MAG Program.
The MAG Regional Council should take a stronger role in coordinating the efforts of
member communities to influence the MAG Program in order to resolve conflicts, improve
consistency, and better control the number and extent of changeslenhancements being
requested.
ADOT should retain its statutory authority to cany out and administer the preliminary
engineering, right- of- way acquisition, and construction of the MAG Program.
MAG should formally recognize the priority- setting criteria: extent of local public and
private funding participation, in its priority- setting process for the MAG Program. This
criteria should continue to be used by ADOT to expedite projects which might not
otherwise receive timely programmed funding, provided they satisfy other important criteria
as well.
MAG should expand its priority- setting criteria for the MAG Program to include social and
community impacts on neighborhoods affected by the siting and subsequent construction of
new freeways. Accounting for these impacts during the priority programming process
would enable the MAG Regional Council and State Transportation Board to either
accelerate or defer projects with sensitive social and community implications in order to
better address the concerns of affected residents and business owners.
ADOT and MAG should maintain a more detailed and consistent set of documentation
regarding proposed and actual changes to MAG Program project/ section priorities, including
all correspondence, meeting notes, memoranda, and studies which describe the basis, source,
participants, deliberations, outcome, and rationale for the resulting decisions and actions.
Program management practices and procedures
Conclusions
ADOT lacks budgetary controls for the MAG Program at the program, corridor, and section
levels. The only budgetary controls for the MAG Program exist at the project level, based
on the Department's five- year highway construction program. In addition, the covenants
associated with the revenue bonds issued by ADOT for the MAG Program provide controls
over the level of MAG Program project costs programmed in the Department's five- year
programs.
The lack of budgetary controls at the program, corridor, and section levels indicates a
weakness in the adequacy of program- level oversight and management control. This
reflects on both ADOT, which is responsible for administering the design, right- of- way
acquisition, construction, and financing of the MAG Program, and MAG, which is
responsible for developing and updating the MAG Regional Transportation Plan and setting
the priorities to guide the scheduling of MAG Program activities. This weakness has
resulted in the Department and MAG ultimately committing to a program that cannot be
funded within the constraints imposed by the original authorizing legislation for Proposition
300. However, this has not impeded ADOT's ability to ensure adequate revenue coverage
to satisfy the covenants of outstanding MAG Program bonds.
During the early years of the MAG Program ( 1985- 1989), the Department lacked adequate
internal controls over changes in MAG Program costs. This contributed to the escalation of
Program costs in the first three years of the MAG Program. Since 1989, after the
Department recognized that the MAG Program costs would likely significantly exceed
available revenues, the Department began to institute a wide variety of procedures,
directives, forms, and reporting systems designed to control changes to the MAG Program
project budgets contained in the Department's five- year program. They also provide
effective controls to ensure the Department complies with the covenants associated with
outstanding MAG Program bonds. However, the absence of overall Program budgets at the
program, corridor, and section levels of detail preclude proper monitoring and control of
program, corridor, or project changes which relate to those portions of the MAG Program
beyond the latest five- year program.
The MAG Program is managed on a day- to- day basis by the Urban Highways Section
relative to project design activity, the Right- of- way Section and Urban Highways Section
relative to right- of- way acquisition activity, the Construction Section relative to construction
activity, and the Administrative Service Division relative to funding and bonding activities.
In addition, management oversight is provided by the Deputy State Engineers for
Development and Operations, the State Engineer, and the Director of the Transportation
Planning Division. The current allocation of responsibilities among ADOT units managing
the MAG Program appears reasonable.
Urban Highways Section staff coordinate and oversee the work of location and design
concept consultants, design consultants, and corridor management consultants, who in turn
manage individual section design consultants who prepare the actual construction plans.
Each group has well defined authority and responsibilities for their respective functions.
Recent changes in the responsibilities of the Urban Highways Section to help control the
programming of right- of- way acquisitions related to the MAG Program and to coordinate
ADOT efforts to better manage the parcel condemnation process represent effective
strategies for better controlling the cost of right- of- way acquisition for the MAG Program.
Significant pressures have been applied to ADOT throughout the last six years by local
officials, private interests, and local community groups requesting changes or enhancements
to the scope of MAG Program corridors and sections. Most have involved mitigation or
access issues. In the early years of the program, ADOT top management became quite
involved in resolving these issues, often agreeing to requests to maintain local consensus
and support for the Program. In recent years, the slowdown in the MAG Program and the
decline in estimated available Program revenues has reduced these requests and in some
cases encouraged local- match funding of projects to expedite their schedules. Earlier
commitments increased MAG Program costs, contributing to the current budget imbalance.
With most major location and design concept issues already resolved for the corridors
comprising the MAG Program, the Department has several important buffers in place to
effectively control outside pressures on the Program through the five- year program update
process. These buffers include the Urban Highways Section Engineer, Priority Planning
Committee, and the State Transportation Board. However, the effectiveness of these buffers
could be significantly improved if they had a realistic overall Promam budget to gauge the
fiscal impacts of requested changes proposed by outside influences. This is based on the
notion that the most effective buffer for the MAG Program to control outside influences is a
widely perceived, constrained budget.
Existing State statutes provide sufficient control over possible conflicts of interest situations
involving officers and staff of ADOT and MAG, as they relate to the MAG Program.
However, these statutes are not adequately reflected in administrative procedures manuals
for either ADOT or MAG.
Recommendations
Budgetary controls need to be established at the program, comdor, and section levels for
the overall MAG Program, including management forms, procedures, and reports which
track actual, programmed, and not- yet- programmed activity in terms of revenues by source,
costs by category, scope, and time frame. Project scope and costs should be managed
within the resulting revenue- constrained budget. The budget should be reviewed every six
months and updated annually.
ADOT should submit local requests for major MAG Program changes and enhancements
which would materially increase the cost of the Program to the full MAG Regional Council
via the MAG Program Fiscal Analysis Unit for review and internal MAG resolution prior to
final action by the Priority Planning Committee and State Transportation Board.
MAG should establish a one- to two- person MAG Program Fiscal Analysis Unit to monitor
and assess the fiscal status of the overall MAG Program, and to advise the MAG Regional
Council and MAG Management Committee regarding the consequences of major revenue,
financing, cost, scope, and schedule changes to the MAG Program.
The proposed MAG Program Fiscal Analysis Unit should prepare and issue for public
dissemination an annual report on the relative status of the MAG Program, in terms of
revenues, costs, scope, and schedule, indicating accomplishments, programmed activities,
and unprogrammed efforts needed to complete the Program. This will provide quicker,
more independent reporting to the public of the fiscal status and progress of the MAG
Program. This report should be followed up by MAG- sponsored public hearings, whose
results can serve as input to ADOT~ Transportation Board efforts to update the Department's
five- year highway construction program.
When describing the financidfiscal status of the MAG Program for internal and external
purposes, ADOT and MAG should be consistent in the use of either constant or current
( inflated) dollars for both revenues and costs, noting explicitly which basis is used.
ADOT and the MAG Regional Council should annually reassess the MAG Program budget,
scope, schedule, and financing strategies and adjust them to maintain the fiscal integrity of
the overall Program, consistent with the covenants of outstanding MAG Program bonds.
When appropriate, this should include reassessing prior location and design features to
determine if more cost- effective alternatives can be substituted. Such a process should
include consideration of:
w Corridor deletion or realignment
w Reduced number of lanes per corridor or segment consistent with realistic traffic
forecasts
Reduction of the frequency of RARF- funded traffic interchanges and crossroads ( to
every two to three miles)
Reduction in the miles of depressed freeways
Application of lower design standards which equal or exceed minimum urban design
guidelines
w Increased staging of MAG Program facilities to include freeway- to- freeway interchanges
Reduced crossroad lane widths to be consistent with adjacent local roads
w Local funding of all lighting beyond that required for minimum safety standards
Reduce local access to freeway- to- freeway interchanges
Local funding of right- of- way acquisitions
The MAG Regional Council and State Transportation Board should consider more moderate
bonding strategies which would enable the Department to develop and maintain a more
balanced program over the life of the available funding sources.
ADOT and MAG should provide their staffs with administrative procedures that more fully
explain what is allowed and what is prohibited under applicable conflict of interest laws.
Right- of- way acquisition laws and practices
Conclusions
ADOT's right- of- way acquisition policies and procedures are consistent with those of most
states and FHWA, with the major difference being the inability of Arizona agencies to
allow property enhancement in value to offset all or part of the purchase price for partial
takes. At least 24 states and all Federal agencies permit this practice, known as the " before
and after" method of appraising property.
Based upon the audit team's experience and research, we believe that current Arizona laws
strongly favor the interests of the property owner. A better balance should be provided by
allowing ADOT to apply the " before and after" appraisal method. Our experience and
research suggest that the " before and after" law fairly compensates property owners and
provides the needed protection to the State from excessive awards.
The ability of the Transportation Division staff in the Attorney General's Office to represent
ADOT in right- of- way condemnation cases has been limited by the defensibility of the
appraisals that are prepared for the properties. Past deficiencies in the ability of the
Attorney General's Office Transportation Division to represent ADOT resulted primarily
from insufficient coordination between ADOT and the Transportation Division and limited
staff resources within the Transportation Division.
During the past year, ADOT has attempted to improve coordination with the Transportation
Division by establishing a Condemnation Support Team to help coordinate technical support
for potential or pending condemnation cases; establishing a condemnation tracking report
system to help the Department monitor the status and progress of right- of- way
condemnation cases and workload; and having the Urban Highways Section serve as the
conduit for all condemnation cases being forwarded to the State Attorney General's Office.
The Right- of- way Section appears to have proper controls over all the necessary phases of
negotiation and appraisal, based on our review of ADOT policy manuals and our
investigation of interoffice memos and reports. While the Department has developed proper
controls over the right- of- way acquisition function, deficiencies such as those cited in the
full auQt report suggest that further diligence be applied to ensure that these controls are
properly and consistently applied. In recent years, the Highway Department Group has
instituted a variety of procedures aimed at improving the control of right- of- way activities.
These include:
Establishment of a condemnation support team in March 1990 to ensure proper internal
review and support of property acquisition cases which are likely or actual candidates
for condemnation proceedings
Provision of Urban Highways Section involvement in authorizing the Right- of- way
Section to proceed with property acquisition consistent with the five- year program
Institution of quarterly funding allocations to the Right- of- way Section by the
Administrative Services Division, in order to better control right- of- way expenditures
relative to the five- year program budget
D Implementation in early 1991 of a condemnation tracking system and monthly
condemnation status report
Development of the " red letter process" to facilitate communication and coordination
between ADOT and local zoning/ code enforcement agencies regarding prospective
development in MAG Program corridor alignments
ADOT has not reimbursed the RARF fund for the office rental savings resulting from the
Department's use of property acquired with RARF monies. This is not consistent with the
Department's treatment of private or commercial use of these kinds of properties. However,
the Department has stated it was informally advised by the Attorney General's Office that
there is no legal requirement to reimburse the RARF fund for its use of such properties.
Since the start of the MAG Program, estimates of the cost to acquire all right- of- way
required to complete the Program have grown by $ 1.1 billion, more than doubling the
original estimate. Design and scope changes that have occurred since the MAG Program
was started have driven up the acreage of right- of- way needed, increasing the overall cost
by an estimated $ 350 million. Condemnation awards and the general increases in property
values which had occurred between the time when the historical acquisitions were made
( upon which the preliminary MAG Program cost estimates were based) and the time when
ADOT began acquiring property for the MAG Program have resulted in an additional
estimated $ 750 million increase in the cost for MAG Program right- of- way.
Only about one- third of the required acreage for the MAG Program has been acquired to
date by ADOT. Unless significant additional funding sources are found or major portions
of the MAG Program are curtailed in scope or deleted, less than half of the total right- of-way
needed for the MAG Program will be able to acquired before the Program runs out of
available funds.
Recommendations
ADOT should continue to consider the full costs and risks of taking a condemnation case to
court versus accepting a negotiated settlement in determining the most cost- effective
strategy to handling right- of- way acquisition cases.
ADOT should avoid accelerated acquisition schedules, regardless of the size or funding
status of the MAG Program. ADOT should program its right- of- way acquisition efforts to
avoid " second takes," " double moves" for relocated businesses, and uncertainty on the part
of land owners by establishing a steady pace of right- of- way acquisition and thereby
controlling the time frame between funding authorization and actual acquisition.
ADOT should extend the Advanced Acquisition Program, consistent with budgetary and
priority programming constraints, and avoid advanced acquisition of properties which would
need to be condemned, except for demonstrated hardship cases.
ADOT should be joined by MAG, RSET, the local business community, and residents
interested in completing the MAG Program in the most fiscally prudent manner in
promoting legislative changes to pertnit the " before and after" appraisal method, while
avoiding cumbersome provisions such as those that would require State payment of
attorney's fees, loss of business damages, or proximity damages.
ADOT should reimburse the RARF fund for the Department's use of property acquired with
RARF monies.
ADOT should assess the staffmg requirement of the Transportation Division of the Attorney
General's Office, given the projected work load from the Department, and authorize/ fund
sufficient full- time or contract staff in a timely manner. Legal staff required by the MAG
Program should be funded out of the MAG Program funds, since they perform a direct
function in support of the acquisition of right- of- way for the MAG Program.
ADOT should continue its recent initiative to provide a condemnation support team to assist
the Office of the Attorney General in preparing for impending cases, and to seek alternative
alignments or design changes which could significantly mitigate the need to proceed with
condemnation prior to submitting cases to the Office of the Attorney General.
ADOT should continue to pursue land donations and third- party funding arrangements with
private developers, businesses, and local jurisdictions to lower the costs of right- of- way
acquisition and to increase its available funding. To facilitate this, ADOT should
significantly increase its local matching program for expediting projects in the Department's
five- year program.
ADOT's Right- of- way Section should consistently document all actions/ agreements
involving the appraisal, negotiation, settlement, and acquisition of property for the MAG
Program, and update its Right- of- way Parcel Status Report database in a more timely
manner to facilitate up- to- date management reporting.
ADOT Right- of- way Section staff and top management should continue to strive to comply
with the policies and procedures for guiding and controlling the Department's right- of- way
acquisition process. This will require periodiclrandom management reviews of
documentation to ensure compliance by staff.
Conclusions of the audit report
Problems and opportunities for improvement have been identified in each of the five areas
of focus of the performance audit. Many recommendations have been made to improve the
efficiency and effectiveness of ADOT's and MAG's management of the Program. Some of
these problems could have and should have been avoided by both ADOT and MAG while
some problems were very difficult to foresee, such as the significant economic downturn in
Maricopa County. In some cases, the problems were beyond the control of ADOT or
MAG, such as changing State laws concerning right of way acquisition, which is the
responsibility of the State Legislature.
In assessing the conclusions and recommendations in this performance audit, the following
should be kept in mind:
The MAG Program, which is intended to add about 230 miles of freeways and
expressways to the existing Maricopa County highway system, is the largest current
urban freeway and expressway program in the nation. This program is an enormous
undertaking by any standards.
The program is unique in terms of its being funded by local revenues and in terms of
ADOT's and MAG's institutional relationships and roles in the Program.
Virtually all urban highway and public transportation construction programs nationally
have encountered citizen concern and opposition because of real and perceived impacts
on neighborhoods , parks, businesses, and other land uses. Many of these programs have
also encountered schedule delays and, in certain cases, significant cost escalation
because of expanded design requirements; inflation in materials, labor, and right of way
costs; and schedule slippage.
Despite the many years of transportation planning for Maricopa County, the approval by
the Legislature and ultimately by the voters of the half- cent excise tax for the MAG
Program occurred over a relatively short period of time. An aggressive implementation
program was adopted by MAG and this required ADOT, which was heavily focused on
m a l highway construction programs, to quickly initiate a large- scale urban highway
design, right- of- way acquisition, and construction program in Maricopa County.
The original revenue and cost estimates for the MAG Program at the time of the
Proposition 300 vote were essentially planning level estimates that were not based on
detailed engineering studies or on sophisticated econometric analyses. These are some
of the reasons why revenue forecasts have been overly optimistic and costs have
increased beyond the original estimates. The dramatic and essentially unforseen
slowdown in the County's, the State's and the nation's economy also had a significant
impact on MAG Program revenues and costs.
This report contains many recommendations to correct known Program problems; to
improve existing ADOT and MAG policies, procedures, and practices; and to develop and
implement new procedures to enhance the efficiency and effectiveness of the MAG Program
in the future. The thrusts of the recommendations are to:
H Promote greater public accountability of both ADOT and MAG in terms of financial
management, revenue estirnation, cost control, and schedule adherence and to recognize
MAG's important plan development, priority setting, and financial policy roles in the
Program
w Promote and facilitate public involvement in and familiarity with the status and future
priorities of the Program through MAG's preparing an annual report for the Program
and holding a annual public hearing( s) on the status, future priorities, and costs and
revenue requirements of the Program, as well as by the Transportation Board and
ADOT improving the timeliness, clarity, and consistency of communications with the
public, elected officials, local governments, and other interests
Improve ADOT's revenue estimation, priority setting , and program management
practices to fully account for overall Program revenue and schedule constraints and
commitments, and not just focus on the next five- year highway program
Implement a budget- based monitoring, reporting, and control process for the overall
MAG Program so that short- term decisions regarding scope, financing, and scheduling
are not permitted to undermine the long- term viability of the Program.
Seek legislative approval of the " before" and " after" method of right- of- way acquisition
which is intended to control costs and more equitably balance property owner and State
interests in the right- of- way acquisition process
Control right- of- way costs by avoiding accelerated right- of- way acquisition and
advanced acquisition in cases of condemnation
Encourage MAG and the Transportation Board to consider more moderate bonding
strategies for the MAG Program which would enable the Department to develop and
maintain a more balanced program over the life of the available funding sources.
The specific recommendations presented in the report build upon the many improvements
ADOT has made in its policies, procedures, and practices, particularly in the last two to
three years. The recommendations, in our judgment, are feasible to implement with the
support of MAG and the Legislature. These recommendations should be implemented
regardless of whether additional revenues are approved to complete the balance of the MAG
Program. However, it is especially important that the audit's recommendations be
implemented if additional Program revenues are authorized. The implementation of and
adherence to the recommended program management procedures is a key to ADOT's and
MAG'S controlling local government and citizen requests for project enhancements if
additional funds become available to the MAG Program.
The MAG Program is scheduled to be complete in the year 2005 if adequate funding is
available. It must be recognized that forecasts of population, development, revenues, and
costs over such a long period are subject to many uncertainties. Well designed and
implemented financial management and program management procedures and systems will
help anticipate and respond to unforeseeable demographic and economic changes that will
inevitably occur over time.
1 ARIZONA DEPARTMENT OF TRANSPORTATION
I 206 South Seventeenth Avenue - Phoenix, Arizona 85007- 321 3
Phone: ( 602) 255- 7226 - FAX: ( 602) 255- 6941
1 FIFE SYMlNGTON
Governor
CHARLES E. COWAN
Director
September 6, 1991
Douglas R. Norton
Auditor General of Arizona
2700 North Central Avenue, Suite 700
Phoenix, Arizona 85004
Dear Mr. Norton:
Thank you for the opportunity to comment on the performance audit
conducted on the Arizona Department of Transportation's Urban Highway
Program by your consultant, Peat Mawick. First, I would like to compliment
your staff and those of Peat Mawick on the professional manner in which
they carried out this audit. Their willingness to openly discuss and present
all sides of issues in the final report is to be commended.
The Department accepts the findings as constructive in nature, and we are
already moving to implement recommendations. If you have followup
questions or comments in the upcoming weeks, I will be glad to meet with
you to discuss in more detail the implementation process.
'.- Charles E. Cowan
I HIGHWAYS . AERONAUTICS - MOTOR VEHICLE . PUBLIC TRANSIT . ADMINISTRATIVE SERVICES TRANSPORTATION PLANNING
1. INTRODUCTION
This section provides the background for and presents the objective, scope, and approach
for the Pelformance Audit of the Arizona Department of Transportation's Urban Highways
Program for Maricopa County.
Background
In 1985, the Arizona State Legislature enacted House Bill 2306 authorizing Maricopa
County voters to approve up to a ten percent increase in existing excise taxes designated as
a transportation excise tax ( commonly referred to as a one- half ( 112) cent sales tax). The
law enabled County residents to set aside excise tax revenue for the design and construction
of a metropolitan highway system. An urban highway plan for metropolitan Phoenix was
developed by the Maricopa Association of Governments ( MAG). The MAG Program called
for development of 233.5 miles of freeways and expressways to supplement the 86.5 miles
of completed or under- construction highway in the region over the period 1986 to 2006.
The additional excise tax to finance the MAG Program was overwhelmingly approved by
the voters in October 1985. The proceeds of the excise tax are to be used to design,
acquire right of way for, and construct freeways, expressways, and parkways in the MAG
plan. The 233.5- mile MAG Program is the largest urban freewaylexpressway development
program currently underway in the nation. The Program is also unique in that:
H It is locally funded by Maricopa taxpayers as opposed to state or federally funded like
most other urban highway programs
The 230- mile plan and associated implementation priorities are approved by MAG with
Arizona Department of Transportation ( ADOT) having responsibility for implementing
the Program
The primary funding source has a limited, fixed duration
In December 1990, ADOT released a five- year progress report on the Maricopa County
transportation excise tax titled, Maricopa County Transportation Excise Tax Review, 1986-
1990. This report noted that over $ 3 billion in additional revenue will be needed to
complete the MAG Program. This revenue shortfall coupled with other concerns about
ADOT's management of the program resulted in the Arizona State Legislature directing the
State's Office of the Auditor General to procure an independent performance audit of
ADOT's Urban Highways Program.
Objective and scope of the performance audit
The purpose of the performance audit is to assess ADOT's management performance in
carrying out the MAG Urban Highways Program under original Program conditions, and to
recommend ways to improve ADOT's efficiency and effectiveness in fulfilling its future
overall Program responsibilities.
The Office of the Auditor General directed that the scope of the performance audit include
a review of ADOT's administration of the MAG Urban Highways Program relative to the
following five issues areas:
Original and current excise tax revenue forecasting processes
Original and current program cost estimates
Priority programming process
m Program management practices
Right- of- way acquisition practices
A total of 46 specific questions were detailed in the five areas listed above by the Office of
the Auditor General for analysis in the performance audit. These questions are listed in
Appendix A. This report presents the findings and applicable recommendations from the
analysis of each of these questions.
The scope of the performance audit did not include a financial audit of the MAG Program.
The scope also did not include a detailed preparation of long- range traffic projections or
development of Program revenue and cost projections, or underlying assumptions.
The performance audit used extensive data and estimates developed by ADOT. Such data
and estimates were reviewed for reasonableness, but not verified by the audit team.
Approach
In order to meet the objectives of the performance audit, a work plan consisting of the
following six tasks was performed:
Task 1 - Review Revenue Estimates
Task 2 - Evaluate Estimates of Program Costs
Task 3 - Assess Priority Programming Practices
Task 4 - Evaluate Program Management Practices
Task 5 - Assess Right of Way Practices
Task 6 - Prepare Final Report and Present Findings
Our approach included:
Numerous fact- finding interviews with MAG and ADOT managers directly responsible
for the MAG Program and its implementation ( Appendix B contains a list of these
individuals)
Interviews with current and former State and local elected officials, representatives of
the business community groups that have expressed concerns with MAG Program
management, and selected others knowledgeable of the overall MAG Program
( Appendix B contains a list of these individuals)
Compilation and reviews of extensive ADOT and MAG reports, manuals, and working
papers
Contacts with selected other state and local agencies responsible for similar highway
and financing programs
The work plan was performed by a competitively selected project team composed of
performance audit, highway engineering, right- of- way acquisition, and financial analysis and
management specialists.
Organization of report
Following the introduction, Section 2 presents an overview and a brief status summary of
the MAG Program. This section provides useful background for readers that are not
familiar with the Program as well as the context for the detailed analyses documented in
other sections of the report.
Sections 3 through 7 document the analyses, conclusions, and applicable recommendations
for the five areas of focus of the performance audit. Each of the 46 questions in the
performance audit's scope is addressed in these sections. It is important to note that our
response is fully documented for each question with very limited cross- referencing to other
questions. This keeps the reader from having to make extensive cross- references to other
questions or sections of the report and also minimizes the potential for misinterpreting
conclusions and recommendations.
Section 8 summarizes the overall conclusions of the performance audit and identifies
important considerations for improving the efficiency and effectiveness of ADOT's and
MAG'S management of the MAG Urban Highways Program in the future.
2. OVERVIEW AND STATUS OF MAG REGIONAL FREEWAY/ EXPRESSWAY
PROGRAM
This section presents an overview of the MAG freeway/ expressway plan and priorities,
agency roles and responsibilities, the forecasted revenues and costs of the MAG Program,
and the Program's status as of July 1991. This overview and status report is intended to
provide readers with background information for evaluating the findings and
recommendations presented in Sections 3 through 8 of this report.
MAG plan and priorities
As required by federal law, an urban transportation planning program has existed in the
Phoenix urban area since the 1960s. The Maricopa Association of Governments ( MAG) is
the designated metropolitan planning organization ( MPO) for Maricopa County and is
responsible for directing the ongoing urban transportation planning program. Staff jointly
funded by MAG and ADOT perform the technical analyses for the urban transportation
planning program. The staff are in the MAG Transportation Planning Office ( MAGTPO)
and consist primarily of ADOT employees.
On July 25, 1985, MAG adopted its Regional Transportation Plan for Maricopa County.
This plan was approved by MAG elected officials and was based on transportation studies
of the East Valley, Central Phoenix, and the West Valley. The plan recommended
233.5 additional miles of freeway and expressway corridor to the 86.5 miles of completed
or under- construction freeways. Exhibit 2- 1 presents the MAG Regional Transportation
Plan.
The corridors included in the plan were broadly defined. Interchange spacing was
generalized and drainage, right of way, and selected other requirements could not be
detailed because of the generalized alignments and designs. The final alignment and design
of the facilities were the subject of engineering and environmental studies, public hearings,
and adoption by the State Transportation Board.
In January 1986, MAG adopted freeway/ expressway priorities for the MAG Program.
Exhibit 2- 2 summarizes these priorities by five- year period between 1986 and 2006.
ADOT's recent five- year progress report on the MAG Program noted that the original dates
for MAG priorities are no longer valid because of increased costs and revenue shortfalls.
EXHIBIT 2- 1
MAG REGIONAL TRANSPORTATION PLAN
ADOPTED JULY 24,1985
CAREFREE HIGHWAY
m Completed - Under Or Nearlng Construc!~~ n
[. IID Planned
Adopted Corridor Under Study
The precise. f~ narlo utes for all freeways
are subject to public hearings and
local input
W
EXHIBIT 2- 2
FREEWAYIEXPRESSWAY PRIORITIES FOR EXCISE TAX AND 15% REVENUES
ADOPTED BY MAG REGIONAL COUNCIL JANUARY 29,1986
SECTIONS
AGUA FRIA FREEWAY
Buckeye Rd.- Papago Freeway
Papago Freeway- Northern Ave.
Northern Ave.- Bell Rd
Bell Rd.- Black Canyon Freeway
ESTRELLA FREEWAY
S. R. 85- Grand Exprrssway
Grand Expressway- Black Canyon Freeway
GRAND EXPRESSWAY
Mchwell Rd.- Paradise Parkway
Paradise Parkway- Agua Fria Freeway
Agua Fna Freeway- Dysart Rd.
Dysart Rd.- Cotton Lane
HOHOKAM EXPRESSWAY
Mchwell Rd.- University Dr.
PAPAGO FREEWAY ( EAST)
I- 10- Hohokam Expressway
Hohokarn Expressway- Pima Freeway
PARADISE PARKWAY
Squaw Peak Parkway- Black Canyon Freeway
Black Canyon Fneway- 5lst Ave.
51st Ave.- Agua Fria Freeway
PIMA FREEWAY
Black Canyon Fneway- Squaw Peak Freeway
Squaw Peak Parkway- Swttsdale Rd
Scoasdale Rd.- Shca Blvd
Shca Blvd- Papago Freeway
Wpago Freeway- Superstition Freeway
PRICE PARKWAY
Superstition Fmway- Santan Freeway
RED MOUNTAIN FREEWAY
Pima Freeway- Country Club Rd.
Country Club Rd- Giibelt Rd.
Gilbert Rd.- Bush Highway
Bush Nghway- Elisworth Rd.
McKellips Rd- Superstition Freeway
SANTAN FREEWAY
I- lo- Price Parkway
Price Parkway- Gilbert Rd
Gilbert Rd.- Powcr Rd.
Power Rd.- Superstition Freeway
SKY HARBOR ACCESS FACZLITIES
Sky Harbor Exprssway ( Sky Harbor- ElO)
Sky Harbor Blvd ( 44th- 56th Streets)
SOUIH MOUNTAIN PARKWAY
Papago Freeway- Baseline Rd.
Baseline Rd- 7th St.
7th St.- Mariwpa Fneway
SQUAW PEAK PARKWAY ( EXTENSION)
Glmdalc Ave.- Thunderbird Rd.
Thundehid Rd.- Pima Fneway
E/ R Preliminary cnginaring and right- of- way purchase
AC Accelcntcd wnstruction started
C F~ nal wnstmction canpleted
S Staged wnstmction wmpletcd
* Full w ~ t m c t i o nb y 1995 if funds available
** Lowest wnstmction priority
Source: MAG
EXHIBIT 2- 3
PRIORITIES FOR MAG EXCISE TAX AND 15% REVENUES
ADOPTED BY MAG REGIONAL COUNCIL OCTOBER 24,1990
SECTION LEVEL I* LEVEL I1 LEVEL 111 LEVEL IV LEVEL V
AGUA FRIA FREEWAY
Buckeye Rd- Papago Freeway
Papago Fmway- Northern Ave.
Northem Ave.- Bell Rd
Bell Rd.- Black Canyon Freeway
EAST PAPAGO FREEWAY
Papago Freeway- Hohokam Expressway
Hohokam Expressway- Pima Freeway
ESTRELLA FREEWAY
Buckeye Road- Papago Freeway
Wpago Freeway- Grand Expressway
Grand Expressway- Black Canyon Fmway
GRAND EXPRESSWAY
McDowell Rd.- Paradise Parkway
Paradise Parkway- Agua Fria Freeway
Agua Fria Freeway- Dysart Road
Dysart Road- Cotton Lane
HOHOKAM EXPRESSWAYPARKWAY
University Drive- McDowell Road
McDowell Road- Thomas Road
PARADISE PARKWAY
Squaw Peak Parkway- Black Canyon Freeway
Black Canyon Freeway- 59th Avenue
59th Ave.- Agua Flia Freeway
PIMA FREEWAY
Black Canyon Freeway- Squaw Peak Fmway
Squaw Peak Parkway- Swttsdde Road
Swtlsdale Road- Shea Boulevard
Shea Boulevard- East Papago Freeway
East Papago Fmway- Superstition Freeway
PRICE PARKWAY
Superstition Freeway- Santan Freeway
RED MOUNTAIN FREEWAY
Pima Freeway- Country Club Road
Country Club RoadGilbert Road
Gilbcrt Road- Bush Highway
Bush Highway- Superstition Freeway
SANTAN FREEWAY
Mariwpa Freeway- Price Pakway
Price Parkway- Giben Road
Gilbcrt Road- Power Road
rower Road- Superstition Freeway
SKY HARBOR ACCESS FAC? LITIES
Sky Harbor Expressway ( Sky Harbor- ElO)
Sky Harbor Boulevard ( 44th- 56th Streets)
SOUIU MOCUTAIN PARKWAY
Papago Freeway- Baseline Road
Baseline Road- 7th Street
7th Street- Mariwpa Frecway
SQUAW PEAK PARKWAY ( EXTENSION)
Glcndale Avenue- Thunderbird Road
Thunderbid Road- Bell Road
Bell- Road- Pima Freeway
EIR
Em
Em
EIR
EIR
E/ R Preliminary engineering and right- of- way purchase
C Final wnst~ ctionw mpleted
S Staged construction wmpleted
* The portion of Price Parkway between the Superstition Freeway and Guadalupe Road. as well as the portion of the Red Mountain Freeway between the
Pima Freeway and Dokon Road, are included in Level I. Also, staged wnstluction of the Price Freeway between Pams Road and Galveston Road, as
well as staged construction of the Pirna F m a y between Bell Road and Smttsdale Road, am included ' in Level 1.
Source: MAG
MAG adopted revised priorities for the program in October 1990 and Exhibit 2- 3
summarizes the priorities. The definitions of the revised priority categories are as follows:
Level I: Completed segments and projects programmed in the ADOT FY 1991- 95
highway construction program, including those identified in the program for
construction in FY 1996 and FY 1997.
Level 11: Unprogrammed original 1990 and 1995 priorities.
Level III: Original 2000 priority projects.
Level N: Original 2005 priority projects.
Level V: Staged construction on the portions of the Estrella Freeway not programmed
to date.
Since the adoption of the MAG Plan and priorities in 1986, ADOT has conducted location
and design concept studies for all corridors and prepared detailed designs in many corridors.
These studies, which included extensive public involvement, generally resulted in expanded
design requirements ( e. g., type of facility ( freeway versus expressway), grade of facility
( depressed versus at- grade or above grade), number of lanes, number and types of
interchanges, right- of- way requirements, and environmental impact mitigation measures).
Factors contributing to this include:
Higher year 2005 population and employment forecasts for Maricopa County than were
used in original MAG planning studies
i Significantly higher year 2005 traffic projections for the MAG study area
Citizen and local government requests and requirements to develop corridor alignments
and designs that are compatible with adjacent development and mitigate environmental
and related impacts
Engineering requirements to meet ADOT's design standards and topographic,
environmental, and development conditions in each comdor
ADOT's report, Maricopa County Transportation Excise Tax Review, 1986- 1 990, presents
detailed assessments of how the geometric, operational, and right- of- way requirements in
each corridor changed based on the above factors during ADOT's ongoing design of the
Program.
MAG Program funding
The MAG Program is funded from two primary sources. The major funding source is the
transportation excise tax authorized pursuant to HB 2306 and approved by the Maricopa
County electorate on October 8, 1985. This tax equates to 10 percent of the state's
transaction privilege tax rate as of January 1986 on 16 separate classes of business activity
within Maricopa County. The revenues from this transportation excise tax are deposited in
the Maricopa County Regional Area Road Fund ( RARF) and can only be used for:
1. Payment of bond- related expenses and obligations.
2. The funding of reserve accounts for the repayment of bonds.
3. The design, right- of- way purchase, or construction of controlled- access highways which
are in the MAG plan and on the State Highway System.
4. Related grade separations of controlled- access highways which are included in the MAG
Plan.
In addition to RARF revenues, the MAG Propam is also financed with MAG 15 percent
revenues, which represent a 15 percent share of ADOT's 50 percent allocation of the
Arizona Highway User Revenue Fund ( HURF). Approximately 83 percent of the
15 percent revenues allocated to fund the MAG Program are statutorily dedicated. The
balance ( special 15 percent revenues) are allocated in accordance with ADOT policy. One
of the primary funding strategies recommended by the MAG Regional Council was to use
revenue bonds to be repaid from HURF and RARF funds to the extent possible. High
bonding levels in the early years of the MAG plan would accelerate construction of
facilities at the earliest possible date. The State Transportation Board has followed this
recommendation and has implemented an aggressive bonding program since 1986.
Other sources used to fund the MAG Plan include federal aid funds to pay for
improvements where MAG corridors intersect with federal aid- supported interstate
highways, third- party contributions from local communities and developers, and interest
income from unexpended bond moneys.
.4s discussed in Section 3 of this report, the actual excise tax collections have fallen short
of collections forecasted by ADOT each year since FY 1986. ADOT's latest 20- year
revenue forecasts are substantially below the legislative estimates originally developed in
1985. For example, ADOT's July 1991 MAG FreewaylExpressway System Status Report
notes that the original 20- year excise tax revenue forecast was $ 5.858 billion based on the
1985 legislative analysis, while its 1990 estimate is $ 3.834 billion. Considering 1991
revenues collections, ADOT's latest RARF trend revenue estimate is $ 3.410 billion.
Factors contributing to lower excise tax collections and forecasts are presented in Section 3
of this report. A major contributing factor has been the significant economic slowdown in
Maricopa County, the State, and the nation which began close to the time the Program was
initiated.
MAG Program costs
The projected costs to complete the MAG Program have increased significantly since 1985.
The latest ADOT total cost estimate to complete the Program is $ 7.1 billion ( in 1991
dollars) as compared to the $ 3.0 billion ( in 1985 dollars) estimate developed in 1985. The
cost increases have occurred for many reasons including:
H The completion of detailed corridor location studies in all corridors
H Expanded geometric, operational, and impact mitigation requirements
H Significantly greater right- of- way requirements and costs than originally estimated
H Inflation in material, labor, and related costs
Section 4 assesses the factors that have contributed to projected MAG Program cost
increases.
MAG Program status
Based on ADOT reports and estimates, the status of the MAG Program as of
September 1990 was:
H 14 miles of freeway/ expressways have been opened since 1985
29 miles of freeways/ expressways are currently under construction
Location studies have been completed in all MAG corridors
a Approximately 26 percent of design work has been completed
H Approximately 36 percent of required right of way acreage has been acquired
Approximately 8 percent of program construction has been completed
The overall program is estimated to be 21 percent complete
3. REVIEW OF EXCISE TAX REVENUE FORECASTING PROCEDURES
This portion of the performance audit addresses questions regarding the reasonableness of
forecasting and estimation procedures used for the excise tax increase enacted by Maricopa
County voters in 1985. For the purposes of this performance audit, we respond to inquiries
concerning:
Sources of original revenue estimates
Appropriateness of revenue estimation methods and assumptions
Adequacy of revenue forecast documentation
w Timeliness of revenue forecast updates
Reasonableness of current revenue forecasting processes
W Adequacy of ADOT's management of the RARF bonding process
w Sufficiency of excise tax revenues to meet debt services requirements of outstanding
RARF bonds
The following pages present the findings, conclusions, and, where appropriate,
recommendations resulting from the audit team's assessment of ADOT's excise tax revenue
forecasting procedures.
REVIEW OF EXCISE TAX REVENUE FORECASTING PROCEDURES
3.1 What entities were involved in developing original revenue forecasts? How were
original revenue estimates determined?
Background
In 1985, the Arizona State Legislature enacted House Bill 2306 authorizing an increase of
112 cent to Maricopa County's ( County) excise tax to support the design and construction of
a metropolitan highway system. This legislative authorization was followed in the same
year by a County- wide special election in which residents approved the levy of the
additional excise tax for the MAG Program. While the initiative approved by voters
authorized the collection of the increased excise tax and its expenditure on the MAG
Program, initial legislative planning also considered that a portion of the Program would be
funded from gasoline taxes.
Integral to the decision to utilize an excise tax increase as the primary funding mechanism
for the MAG Urban Highways Program, was estimation of the amount of funds that an
excise tax increase could potentially provide over the 20- year life of the Program. The
earliest public estimates of the additional excise tax that could be collected were printed in
the voters pamphlet of the special election. These estimates are shown in Exhibit 3- 1.
EXHIBIT 3- 1
ESTIMATE OF THE ANNUAL AMOUNT OF
TRANSPORTATION EXCISE TAX REVENUES TO BE RAISED
( Millions of dollars)
1986 $ 99 1991 $ 162 1996 $ 262 2001 $ 422
1987 $ 109 1992 $ 178 1997 $ 288 2002 $ 465
1988 $ 120 1993 $ 196 1998 $ 317 2003 $ 511
1989 $ 133 1994 $ 217 1999 $ 349 2004 $ 562
1990 $ 147 1995 $ 238 2000 $ 384 2005 $ 618
Source: Special Election Voters Pamphlet, Maricopa County
Board of Supervisors, October 8, 1985
The description of the estimates in the voters pamphlet stated that:
" Based upon population growth, inflation and other economic factors, legislative
staff estimates that the Transportation Excise Tax will produce $ 5.8 billion over
the twenty- year duration of the tax."
The voters pamphlet consistently used the term " estimate" to describe the future excise tax
revenues that may be collected. While frequently used interchangeably, an " estimate" has a
different definition than either a " forecast" or a " projection." It is widely recognized that a
" forecast" is the most definitive statement an economist will make about future information.
A forecast is typically characterized by assumptions which have a reasonable basis and have
been evaluated under a rigorous methodology. The term " projection" is typically used if
one or more of the underlying assumptions cannot be reasonably approximated, making a
projection somewhat hypothetical in nature. An " estimate" is typically supported by an
even less comprehensive, and correspondingly less precise, set of assumptions and process.
Due to their nature, many organizations restrict the use of estimates and projections to
specified or internal uses only.
Whether a forecast, projection, or estimate, it must be recognized that all future information
is subject to change as actual events occur. As the public does not typically distinguish
between the various levels of effort and confidence underlying future information, it is
generally necessary to clearly qualify future information with a statement regarding its
achievability.
No discussion of the achievability of the initial excise tax estimates, or the impact to the
Program if estimates were not reached, was provided to the public in the voters pamphlet.
The original excise tax revenue estimates were prepared by an ad hoc committee that was
composed of the following members:
W Terry Trost, Phoenix Chamber of Commerce
Dennis Smith. MAG Staff Coordinator
W Roger Herzog, MAG Transportation Planning Office ( MAGTP0) lADOT
Robert Lockwood, House Research Staff
W Charlie Miller, ADOT Director
Owen Ford, ADOT State Engineer
I Chuck Rider, ADOT
I Ron McReady, ADOT
I Suzanne Sale, ADOT Administrative Services Division Director
John Sernrnens, ADOT
Robert Mickelson, ADOT Deputy State Engineer
I Harry Reed, ADOT Transportation Planning Division Director
The majority of the committee was made up of staff with government finance, revenue
analysis, budgeting and legislative analysis experience. Some members also had experience
in economic forecasting and sales tax projections.
The committee was convened in November 1984 to develop a data base of information that
could be used to support planning for major transportation legislation that was expected to
be introduced in the 1985 Legislative Session. A part of the committee's charge was
reviewing appropriate funding alternatives for a major transportation program. This entailed
the review and compilation of estimates of future revenues from various funding sources,
including excise taxes. Developing official public forecasts for these potential funding
sources was not part of the committee's objectives.
The committee reviewed a number of alternatives for funding the Program, including
property taxes, gasoline taxes, and excise taxes. Due to the amount of funds that were
required based on very preliminary estimates of potential costs by ADOT, the Committee
determined that an excise tax increase provided the best alternative for collecting the
required funds.
The committee developed initial excise tax estimates for both a 114 cent and 112 cent tax
increase. Preliminary work on the revenue estimates indicated that a 114 cent excise tax
would not be enough to fund the Program, and further analysis of the 112 cent tax was
performed. The methodology used to develop the initial 112 cent excise tax estimates was
proposed by the Joint Legislative Budget Committee and reviewed and adopted by the
committee. ADOT representatives have stated that the Executive Budget Office and the
Department of Revenue also participated in the review of excise tax revenue estimates.
The methodology used by the committee to develop the initial estimates was based on
increasing the County's estimated excise tax collections for 1985- 86 by forecasted inflation
and the expected increase in real personal income over the life of the Program. Annual
inflation forecasts were obtained from DRINcGraw- Hill PRO, a well recognized,
econometric forecasting firm, and appear to have been developed based on the Producer
Price Index for Producer Finished Goods. The forecasts of the percentage increase in real
personal income were obtained from the Arizona Department of Economic Security ( DES).
These two factors were averaged for the fist five years of the Program, second five years,
and final ten years.
The factors used in the excise tax growth estimate are shown in Exhibit 3- 2. The actual
average annual growth rate for the period 1985- 1989 was 8.6 percent, which was comprised
of an annual average inflation rate of 3.8 percent and an average annual increase in real
personal income of 4.8 percent.
EXHIBIT 3- 2
GROWTH ASSUMPTIONS
Real Personal Excise Tax
Forecast' Income Forecast2 Growth Forecast
5.3% 10.5%
Second Five Years
FY 1990- 1991
through
N 1994- 1995 6.1 % 4.1% 10.2%
Final Ten Years
EY 1995- 1996
through
F'Y 2004- 2005 6.2% 3.8% 10.0%
Sources:
DRIMcGraw- Hill
Arizona Department of Economic Security
Based on discussions with ADOT, the original estimates of excise tax revenues were
developed by the committee for legislative planning purposes only. The inclusion of the
estimates in a public document like the voters pamphlet was not considered by the
committee as the estimates were being developed.
The voters pamphlet was drafted by MAG staff and Bob Lockwood of the House Research
Staff, reviewed by the Legislative Counsel's Office, and printed by the County. The
committee's estimates were converted to a calendar year basis and included in the voters
pamphlet as " legislative staff" estimates.
Criteria
The criteria used for evaluating the reasonableness of the initial estimation process
considered that the estimates were used for a public purpose and likely influenced voter
decisions to some extent. Therefore, a more comprehensive set of criteria was used to
review the estimates than would be necessary had the estimates been limited to internal or
legislative use only. Further, considering that these estimates were prepared by public
agency staff, assumptions should consider the risks of overstatement by being somewhat
conservative.
The criteria include the following:
The estimates should be developed through a rigorous process which considers various
alternatives to estimating future information
The estimates should be based on reasonable and sound econometric methods and
should consider the relative achievability of key assumptions
The estimates should be based on realistic but conservative long- term economic
assumptions
The estimates should be adequately qualified and assumptions disclosed to allow voters
to understand the nature and strength of the estimation process
In effect, since the estimates were used in a public document and impacted voter decisions,
they required the type of diligence of process used for economic forecasts.
Analysis
Using the process described earlier, the committee developed an excise tax revenue estimate
for the Program which grew at an average annual rate of 10.17 percent over the 20- year life
of the Program. This rate was comprised of equal parts inflation and real income growth in
the fist five years. In the second five years, the inflation component of the growth rate
was 6.0 percent and grew to 6.2 percent in the last ten years of the Program. This
illustrates that the initial excise tax estimates were highly dependant on actual inflation
occurring as forecasted.
DRI, like many other econometric forecasting firms, provides base case forecasts using its
best assessment of future economic activity. High and low forecasts are also produced by
adjusting the base case to general more optimistic and conservative forecasts. Importantly,
inflation is normally considered in the context of increasing costs and therefore high
inflation estimates are considered conservative. In the case of revenue estimates, however,
high inflation is not a conservative assumption.
In 1983 to 1985, actual inflation nationwide for consumer prices was between 3 and
4 percent per year, while producer price inflation was much lower. However, in late 1984
DRI was forecasting inflation for consumer prices to be 5 to 6.5 percent for the years 1986
to 1995. The committee used these high base case estimates of inflation for all years of the
estimates, even though:
Short- term trends pointed to lower inflation levels
High inflation is not a conservative assumption for a revenue forecast
Had a more conservative assumption for inflation been adopted, the initial revenue estimates
would have been closer to actual collections.
This conservatism may have been introduced to the committee's process had an external
review of the revenue estimates been performed by econometricians before they were issued
publicly. Alternatively, the public could have been made aware of the impact of a potential
overstatement of revenues through adequate disclosure in the voters pamphlet.
One argument that has been made related to the importance of the inflation assumption, is
that since it impacts both costs and revenues equally, its importance is minimized.
However, as described later in the report, revenues and costs were not analyzed together
and the total revenue figure, which included a high inflation estimate, became a benchmark
for the Program.
Finally, it is not clear how or if the estimates would have changed had the committee used
a more rigorous process and considered other economic indicators and variables. Trend
analysis based on inflation and real income growth is a relatively simple method of
estimating excise tax revenue and it is not clear that other, more comprehensive approaches
would have yielded substantially different results. Due to the importance and manner in
which this information was presented to the public, several approaches should have been
considered and the most conservative and supportable method adopted.
This more rigorous approach should have been considered before the legislative staff chose
to print the estimates in the text of the voters pamphlet.
Conclusions
Considering that the initial objective of the committee's efforts was to develop internal
planning estimates, the committee followed a generally reasonable estimation approach
based on trend analysis. The committee did not use a conservative assumption for inflation,
which accounted for much of the difference between the estimated revenues and actual
revenues to date. Considering the Committee's initial objective, other assumptions appear
to have been reasonable.
When the decision was made to include the information in a public document, a more
rigorous process should have been considered. Alternatively, the estimates should have
been clearly qualified as to their achievability in the voters pamphlet.
Recommendations
ADOT should develop a standard disclosure statement to accompany all estimates,
projections, or forecasts when the Department or its staff have been associated with
their development.
To ensure that internal estimates or projections are not used for public purposes in the
future without adequate disclosure, ADOT should adopt a policy of clearly labeling and
qualifying all estimates, projections, and forecasts of future information when they have
been associated with the development of the estimates, projections, or forecasts. This
qualification should include:
A statement regarding the level of diligence associated with each type of information
Specific description of the intended use of the information
A statement acknowledging that forecasts, projections, and estimates will likely vary
from actual results
By adding this standard statement to all future information using a stamp or a standard
form, ADOT will reduce the risk that other parties will place unwarranted reliance on
information ADOT has been associated with without appropriate understanding of the
information.
REVIEW OF EXCISE TAX REVENUE FORECASTING PROCEDURES
3.2 Were estimation methods and assumptions appropriate? How did assumptions
compare to those used in other forecast at the time?
Background
In 1985, House Bill 2306 amended the Arizona Revised Statutes Title 28 ( Transportation
Law) to allow Counties to authorize the collection of additional excise taxes to be used for
regional transportation purposes. By law, funds collected for these purposes are to be
deposited by the State Treasurer in a regional area road fund for the benefit of the County
from which the funds are collected. Transportation Law Article 3.1 Section 28- 1594.01
describes the use of excise taxes and specifies that "... the director ( of ADOT) shall
administer monies deposited in the regional area road fund."
After the passage of Maricopa County's 112 cent excise tax in October 1985, ADOT
determined that a more comprehensive and rigorous approach to revenue forecasting than
that used by the ad hoc committee was needed for planning and management of Program
funds. Revenue forecasts were considered critical to establishing the MAG Program portion
of the Department's five- year highway construction program and also necessary for
illustrating the viability of the Program's revenue stream for bond offerings.
To this end, ADOT issued a request for quotes ( RFQ) in December 1985 to retain a
consultant to develop forecasting techniques for the 112 cent excise tax increase. The RFQ
specified that the consultant must gather data, develop forecasting approaches, recommend
software, develop initial forecasts and provide a forecast model. The consultant was given
three months to complete the project and funding was limited to $ 10,000. The project was
awarded to Dennis Hoffman, Ph. D. and Don Schlagenhauf, Ph. D., economists associated
with Arizona State University ( ASU).
The model developed by the consultant team is a structural econometric model that
generates forecasts of taxable business activity over the twenty- year life of the program.
The model is organized to independently forecast 10 of the 16 taxable activities covered by
the tax increase legislation. Five of the activities-- retail excise, contracting, rental of real
property, utilities, and restaurants and bars-- historically account for over 90 percent of the
excise tax collections and therefore make up the most important components in the model.
Each of the components in the model is forecasted using a separate equation encompassing
specific relationships to the model's major economic and demographic variables. The
model calculates excise tax revenues by applying the appropriate tax rate to each forecast of
taxable activity. This structure will allow continued use of the model should tax rates
change in the future.
Consistent with structural econometric models, the ADOT model forecasts economic activity
based on time- series regression techniques. This approach determines the historical
relationship of the economic activity to be forecasted to other types of economic and
demographic information called variables. By combining the historical relationships with
forecasts of the economic and demographic variables, forecasts of the economic activity can
be developed. Similar to other econometric models, the ADOT model utilizes each
successive year of actual data to refine the relationships between variables and economic
activities.
The major economic and demographic variable forecasts on which the ADOT model relies
are:
Total personal income of Maricopa County residents
Total Maricopa County population
Number of passenger arrivals at Sky Harbor Airport
Dollar volume of building permits in the County
Exhibit 3- 3 on the following page describes the model's major economic activities that are
forecasted, the percentage of excise tax attributable to each activity, and the primary eco-nomic
or demographic variables which are used to forecast the activity.
The ADOT model was first used to develop official revenue forecasts for the Program when
the initial MAG Program Regional Area Road Fund revenue bond offering was developed
in July 1986. This initial forecast relied on the following values for the key economic and
demographic variables:
Personal income - Personal income was derived by adjusting upward the Eggert
Economic Enterprises, Inc. forecast of real personal income for Arizona by the inflation
factor used by DRINcGraw- Hill, Inc., combined with the historical growth difference
between Arizona personal income and County personal income. The forecasted per-sonal
income growth rate was 10.8 percent which was the sum of a 6.2 percent annual
growth rate in personal income and a 4.6 percent annual rate of inflation.
Population - The County population forecast was issued by the Arizona Department of
Economic Security in May, 1986. The forecasted growth was 3.5 percent per year.
Passenger Arrivals - Passenger arrivals at Sky Harbor Airport were estimated by
ADOT at 9.3 percent per year, which was within the range of City of Phoenix
estimates.
EXHIBIT 3- 3
ORECASTING MODEL
AND DEMOGRAPHIC VARIABLES
Percent of Major Determinants
Total Collections of Activity
Retail Sales 50.36% Population, Income
Contracting 11.99% Retail Sales, Building Permits
hotels and motels) 9.75% Income
9.64% Income
8.38% Income, Air Arrivals
- 9.88% Miscellaneous
- 100.00%
Source: Arizona Department of Revenue
Building Permits - Forecasted building permit activity was determined by ADOT based
on historical trends. A forecast of 13.8 percent annual growth in the dollar volume of
building permits was used in the initial forecasts.
The first official forecasts issued by ADOT in July 1986 are provided in Exhibit 3- 4. Also
shown are the original estimates used by legislative staff in support of HB2306. For this
comparison, legislative staff estimates were adjusted by ADOT to a fiscal year basis and are
therefore slightly different than the calendar year estimates provided in the voters pamphlet
and shown earlier in the response to Question 3.1 of this section.
Total revenues over the life of the Program were estimated at $ 5,858.4 million by
Legislative staff in 1985, while ADOT initially forecasted $ 6,006.3 million in 1986 using its
newly- developed excise tax revenue forecasting model.
EXHIBIT 3- 4
GISLATNE STAFF ESTIMATES AND INITIAL
FORECASTS OF EXCISE TAX REVENUES
( Millions of dollars)
1996 $ 250.8 $ 242.8
iscal Years 1986 and 2006.
Criteria
The criteria for evaluating the appropriateness of the excise tax revenue forecasting methods
and assumptions include:
Use of standard, econometric forecasting techniques
N Use of recognized data sources
Overall reasonableness of assumptions and data
Realistic but conservative approach in use of data
The use of standard, econometric forecasting techniques dictates that the model be able to
predict historical economic performance ( assuming historical relationships are valid), should
react to changing economic conditions, and should have long- term capabilities.
Analysis
This section describes the analysis procedures performed to evaluate whether estimation
methods and assumptions were appropriate. It is organized in the following subsections:
Use of forecasts
Forecast methodology
W Input data sources and assumptions
Procedures used by other agencies
Use of forecasts
Prior to analyzing the specific methodologies used to forecast excise tax revenues, it is
important to understand how the revenue forecasts relate to the Program.
The early revenue forecasts were used for two primary purposes: illustrating the strength of
the Program's revenue stream in the initial bond offering; and developing the five- year
program. The initial bond issuance covenants required that actual historical revenues be in
excess of expected debt service by 30 percent. Therefore, while rating agencies were likely
interested in the forecasts from a Program management perspective, they were likely much
more concerned about historical performance than the actual forecasts themselves.
The development of the five- year program was the other major reason for revenue forecasts.
Since the five- year program is a rolling plan that changes from year to year, early variances
in forecasts could be somewhat mitigated by changing priorities or revising bond issuance
schedules. Therefore, the five- year program is also not particularly impacted by forecast
inaccuracies.
The real risk to the Program of forecast variances is on the overall Program management
and the ability to complete the entire Program on time. Recent forecasts, which are
discussed in later questions, show that the relatively optimistic forecasts in the early years
combined with high growth rates in future years resulted in a significant overestimation of
Program revenues. As Program costs were not similarly reduced or constrained, a
significant shortfall for the overall Program is now anticipated.
Due to the compounding affect of growth on successive years of excise taxes, early
forecasts are critical to determining overall Program viability. Based on an ADOT analysis
prepared for this performance audit, the first five years of actual collections were lower than
forecasts by $ 71 million, or approximately 11 percent. While this may not seem particu-larly
significant when compared to a $ 5.8 billion Program, these initial years form the base
on which future growth is calculated.
Using the initial forecast's growth rates for the remaining 15 years, ADOT's analysis shows
that this early $ 7 1 million difference compounds to a difference of over $ 1.15 billion over
the 20- year life of the Program. This compounding is one of the major reasons for
currently projected Program shortfalls. Due to the compounding, reliable short- and long-term
forecasts are equally important to Program management.
Forecast methodology
The excise tax is actually a tax on 16 separate classes of business activity, and the ADOT
model forecasts the individual business activities, not the tax revenue. The classes of
taxable activity are very different, suggesting that the underlying determinants of economic
behavior and their effects for each taxable activity will vary. For example, economic
behavior that generates retail sales is likely to be different than the economic forces that
result in construction or contracting activity. Therefore the model was designed to
desegregate and separately estimate these classes for forecasting purposes.
Once the taxable activities are estimated, the model calculates the tax revenue for that
activity class by multiplying the total business activity by the appropriate tax rate. The total
tax revenue forecast is the sum of the individual activity tax forecasts. Assuming that the
appropriate indicators of each economic activity were selected, this methodology is reason-able
and consistent with the structure of the Program.
To evaluate whether the appropriate underlying variables were selected to predict each class
of activity, the regression statistics of the model were reviewed. Based on this review, it
appears that the relationships between input variables and the classes of activity are
appropriate.
It is important to remember that econometric models have limitations and cannot be
expected to predict structural changes or major events in an economy. The model is a tool
to be used by experienced economists who must incorporate their knowledge and experience
in both the application of input data and the interpretation of the model results.
This aspect of developing the forecasts is critical to ADOT because there have been several
major unanticipated economic events in the last five years and there have been signs of
some structural change in the regional economy. Since the original model was driven
primarily by historical performance, it did not react well to short- term changes. Consid-ering
the compounding impact of short- term variances discussed earlier, this presents a
significant risk to the forecast process and therefore careful management and interpretation
of results is very important.
Input data sources and assumptions
While the overall methodology for the model appears appropriate, the selection of data for
the underlying variables has been somewhat problematic since the initial forecasts. This
appears primarily due to the recession, Tax Reform Act of 1986, building slump, and
savings and loan failures which have significantly impacted the growth of the local
economy in the last five years. Most regional and national economists failed to predict the
magnitude of these downturns in the Arizona economy.
The four major input variables to the RARF revenue forecasting model included personal
income, population, air arrivals, and building permit value. The growth rates associated
with these variables, as used in the 1986 forecast of excise tax revenues and based on the
actual historical trends over the prior 25- year period ( 1961- 1985) are shown below.
Excise Tax Revenue Historical
Forecast Model Variable 1986 Forecast 1961- 1985 Average
Personal Income
Population
Air Arrivals
Building Perrnit Value
Even though the growth factors initially used for the input variables of the RARF revenue
forecasting rnodel were lower than the averages associated with the prior 25- year period,
1961- 1985 ( as shown above), these growth rates did not reflect an adequately conservative
outlook when considering the higher base in 1986, the 20- year forecast period, and the
lower inflation rates of the years immediately preceding the start of the MAG Program.
Of the four major input variables to the original forecast, each had a different level of
accuracy when compared to actual history and each contributed to a greater or lesser degree
to the difference between forecasted and actual results. Each variable is discussed below.
Personal income
Assumptions regarding personal income are perhaps the most important in the model.
Analysis performed by ADOT for this perf~ rmance audit shows that adjusting the initial
personal income growth forecast of 10.8 percent per year to the actual personal income
growth for the first five years of approximately 8.6 percent per year, accounts for 86
percent of the difference between the original forecasts and what the original forecasts
would have been if all variables were accurately forecasted for the first five years.
The compounded affect of this data difference is estimated by ADOT at $ 1.3 billion over
the 20- year life of the Program. Adding this difference to the $ 0.7 billion forecast decrease
caused by ADOT's current, more conservative growth expectation for excise taxes in the
next 15 years, explains the majority of the $ 2.2 billion difference between ADOT's original
Program forecast of $ 6 billion over 20 years and the latest official 1990 forecast of
$ 3.8 billion. Preliminary 1991 forecasts reflect continued conservative income growth, with
the 20- year total amounting to $ 3.4 billion.
The original forecast of Arizona personal income growth of 6.2 percent was taken from
Eggert Enterprise Inc., a firm providing consensus estimates for major national and regional
economic variables. These estimates are developed by combining estimates independently
developed by nationally recognized economists.
The original forecast of inflation of 4.6 percent was developed by using the DRI price
deflator ( inflation factor) and adjusting it for the historical relationship between inflation for
the State and the County. In the years immediately preceding the original forecast,
consumer price inflation had been in the 3 percent to 4 percent range. When used for
revenue estimation, high inflation is not a conservative assumption. Weighing recent
history to have a greater impact on short- term inflation rates would have been a more
conservative position.
Population
The population variable is also important to the forecasting process because the rate of
population growth in the County is a primary determinant of economic activity. The
forecasts of regional population growth were developed by the Department of Economic
Security ( DES).
DES's forecasts were quite accurate when the impact of the population variable on the
forecasts is evaluated. ADOT analysis suggests that if the original forecasts of population
growth in 1986 were adjusted to actual population statistics and forecasts made in 1990, the
original excise tax forecast would have changed by less than 114 of 1 percent.
Important to the data analysis is an understanding that population growth also influences
personal income growth. However, each of these variables is used in the model
independently. It is therefore important if at all possible that these variables be developed
in a consistent manner.
As discussed above, the population projections were obtained from DES, and the personal
income projections obtained from an independent source. It is quite possible therefore that
the two variables do not use similar approaches or assumptions. While the impact of this
difference may not be material at this time, it is not possible to determine if this will always
be the case. This is an area that warrants further review by ADOT in the future.
Air passenger arrivals
Air passenger arrivals are used as a proxy for the level of tourism in the County. This
variable is used in the forecast of the restaurant and bar taxable business activity. While
reasonably difficult to forecast in the long term, a review of the regression coefficients for
air passenger arrivals shows that this variable does not impact the overall forecasts to a
significant degree. ADOT has used official City of Phoenix and Federal Aviation
Administration ( FAA) forecasts for this variable.
Building permits
The dollar value of building permits has the most fluctuation over time of any of the
variables, making it the hardest to predict. Further, no source of forecasts for this data are
available, requiring ADOT to develop forecasts for this variable internally.
The original forecasts incorporated a 13.8 percent annual growth rate, which was 32 percent
below the 25- year historical average of 20.2 percent. Building permit activity actually
declined over the last five years making this assumption account for an additional 28
percent difference between the original forecast and what the forecasts would have been if
accurate data had been used in the initial five years. ( Note that the 86 percent decrease
caused by personal income plus the 28 percent decrease caused by building permits are
offset by a 14 percent increase because of tax changes in the law that actually increased
collections over the forecast assumptions.)
A 13.8 percent growth rate may reflect somewhat the dramatic growth in building that
occurred in the County in the early 1980s. However, sustaining that level of growth over
20 years was not a realistic and conservative assumption. A 13.8 percent compounded
growth over 20 years would make annual building permit activity grow from $ 3.8 billion in
1985 to $ 57.5 billion in 2006. However, seven times in the previous 20 years, building
permit activity has actually declined in the County. Given the 1986 Tax Reform Act and
other unforeseen events, the results of the last six years confirmed the optimism of the
initial growth rates assumed for building activity in Maricopa County.
Procedures used by other agencies
To review forecasting procedures used by other agencies, we interviewed representatives
from:
Phoenix Regional Public Transportation Authority
San Bernardino Association of Governments
San Diego Association of Governments
Santa Clara County Traffic Authority
Each of these is discussed separately below.
Phoenix Regional Public Transportation Authority ( RPTA)
The RPTA developed excise tax revenue projections in conjunction with the preparation of
the regional public transportation plan that was defeated by Maricopa County voters in
February 1989. The RPTA projections were developed at a less detailed level than the
ADOT projections. The principal underlying the RPTA methodology is that there are three
variables that significantly affect future revenues:
Population growth
Real per capita tax base growth
Inflation
Population growth uses the official MAG population forecast, for both resident and non-resident
components of population. The per capita tax base is represented by real dollar per
capita taxable transactions and is grown by an annual rate of 0.85 percent, for both resident
and non- resident contributions to the tax base. The real dollar total taxable transactions
were then calculated and a weighted average of the individual tax rates was applied to
derive the revenue projection in real dollars.
Inflation is applied to translate the real dollar projection into a current dollar projection. In
this way, RPTA can evaluate the projection under alternative inflation assumptions.
In effect, the RPTA methodology is a much simplified form of the ADOT model, since
both take into the account the same economic and demographic forces that affect the
revenue stream. Both use population, inflation, and taxable expenditures to estimate
revenues.
San Bernardino Association of Governments ( SANBAG)
In 1989, San Bernardino County voters passed a one- half cent sales tax rate increase to
finance public transportation and transit system improvements in the County. The County
expects to soon issue bond anticipation notes in anticipation of a major bond offering in the
Spring of 1992.
The County has determined that they will rely on outside parties to develop forecasts.
SANBAG contracted with a local university economist to develop forecasts for the voter
information literature for the 1989 referendum. SANBAG has tasked their financial advisor
for the bond offering with the responsibility for overseeing the development of the
forecasts, which will likely be performed by an economic consultant.
San Diego Association of Governments ( SANDAG)
SANDAG has an economic input- output model that simulates the San Diego County
economy as an interrelated component of the national economy. This model incorporates
significantly more detail in the variables used to generate sales tax revenue forecasts than
do most other agencies with similar responsibilities. In the SANDAG model, employment
is the primary determinant of economic activity. The employment data used by the model
is at a detailed industry sector level which contributes to a greater degree of accuracy in the
forecasts. Using this comprehensive approach, SANDAG's model has been slightly under-estimating
actual retail sales over the past five years.
Input- output models of this nature and complexity are very expensive to develop and
maintain. The original SANDAG model was developed under a government grant at a cost
in excess of $ 500,000. The model is updated every five years and requires almost a full
person- year for update and analysis.
Santa Clara County Traffic Authority
Santa Clara County voters passed a one- half cent sales tax rate increase to finance the
construction or upgrade of three County freeways. The County created the Santa Clara
County Traffic Authority to oversee all aspects of this ten- year program. In 1987, the
Authority issued $ 200 million in bonds which were then refunded by a $ 274 million bond
issued in 1990. The forecasts used in the preparation of the bonds were prepared in
conjunction with the County's own economic forecast and based on an arithmetic increase
over the prior year's taxable business activity. These forecasts are tracked and updated
monthly based on actual monthly revenues. Forecasts have been within 5.5 percent of
actual results.
Conclusions
Conclusions based on the analysis are as follows:
The model's early development, while constrained by funding and time frame
commitments, is reasonable and consistent with econometric techniques.
The major variable in the model is personal income and in retrospect the initial estimate
of personal income growth provided by a well- recognized independent source did not
accurately anticipate continued low inflation.
The population variable was accurately forecasted by a well- recognized source.
The air arrival variable is forecast by the City of Phoenix and the FAA but has little
impact on the revenue forecast.
The building permit variable, while lower than historical trends, was not conservatively
estimated in retrospect when considering the compounding affect of a 13.8 percent
growth rate and the large amount of building activity that compounding eventually
suggests.
H While other more expensive approaches have been used successfully by other agencies,
the modeling level for ADOT appears appropriate.
Overall, the use of external data sources, which did not fully recognize regional factors
such as recent decreases in inflation or anticipate future decreases, contributed to a
general lack of conservatism in the forecasts in the early years of the MAG Program.
Importantly, ADOT staff recognized many of these factors in the early years of the Program
and, as discussed in Question 3.4, updated later forecasts to reflect many of these points.
Recommendations
ADOT should continue the refinement of its forecasting procedures and continue to
apply subjective judgment in a rigorous, orderly manner based on knowledge of local
economic conditions in reviewing and using forecasted independent data variables to
ensure that conservatism is present in forecasts of excise taxes.
The primary area where the original economic forecasting process could have been
improved was more conservative selection of values for personal income and building
permit activity growth. These areas require that the economist responsible for preparing the
forecast carefully evaluate data sources and variable forecasts and, recognizing the
sensitivity of the forecasts to each assumption and the Program's needs for conservative
revenue forecasts, adjust as necessary each assumption, as is now being done by ADOT.
Use of external data forecasts developed by outside parties for non- specific purposes should
be documented as it is performed and carefully disclosed with the forecasts whenever they
are published.
REVIEW OF EXCISE TAX REVENUE FORECASTING PROCEDURES
3.3 Was the forecasting process adequately documented and subject to appropriate
re view and approval?
Background
ADOT's process for preparing the initial revenue forecasts was described in the response to
Question 3.2 of this section. The documentation supporting this process includes:
Request for Quotes for obtaining a consultant to develop a forecasting model
Proposal responses from potential consultants
Consultants final report on development of forecasting model with model documentation
Historical data series and sources of data used to develop initial variable relationships
Input data values, data sources, and description of other assumptions
Complete model equations and resulting coefficients for each forecast series and annual
updates
Resulting forecasts
Comparison of variable and excise tax forecasts to actual experience for the first five
years of the Program
Criteria
The criteria for evaluating the adequacy of the review and approval process include:
Senior management review, approval, and sign- off prior to the issuance of forecast
Formal issuance of forecasts complete with assumptions and necessary qualifications
Analysis
ADOT's forecast development process calls for Administrative Services staff to compile
latest historical data, develop assumptions, and prepare forecasts during the summer of each
year. Documentation of these activities is maintained. Final forecasts are presented to the
Director of the Administrative Services Division for review and approval. Once approved,
forecasts are presented to the Department Director. Formal sign- off for approval of
forecasts is not performed.
Conclusions
ADOT's procedure for updating revenue forecasts is well documented from a forecast
development standpoint. However, the review, approval, and issuance process is not
formalized in written procedures.
In the past, revenue forecasts have been approved by ADOT and issued as part of each
succeeding revenue bond official statement. A more formal issuance, which incorporates an
analysis of changes to major assumptions for review by the public, would provide more
visibility to the forecasting process. This would also allow ADOT an opportunity to clearly
qualify and explain forecasts to the public and others.
Formal sign- off by the ADOT Director should be part of that process.
Recommendations
ADOT should prepare an official revenue forecast issuance package which includes the
ADOT Director's approval.
The official revenue forecast should include:
A cover letter signed by the Department Director specifically describing the intended
use of the revenue forecasts
ADOT's standard qualification statement regarding the achievability of the revenue
forecasts ( see response to Question 3.1)
A description of all key assumptions
Analysis of changes in major assumptions
Sensitivity analysis of the impacts on the revenue forecast of under and over- forecasting
key independent variables
This approach will allow the Department's approval to be more formally documented, and
provide a package with complete documentation of the forecasting process for review by the
public and others.
REVIEW OF EXCISE TAX REVENUE FORECASTING PROCEDURES
3.4 Were the forecasts updated in an appropriate and timely manner? Were
appropriate and timely adjustments made in the program in response to the updated
forecasts?
Background
Excise tax revenue forecast updates impact the Program in two important areas: issuance of
bonds and Program planning and management. Forecast updates are generated by ADOT
Administrative Services Division staff in July and August of each year. These forecasts are
prepared soon after final sales tax revenue data are made available from the Arizona
Department of Revenue for the prior year. The forecast update process takes several weeks
to complete.
Annual forecast updates were prepared for 1987, 1988, 1989, and 1990. The forecast
update process for 1991 was underway at the time of our review.
Throughout the four years of updates, ADOT staff closely monitored and compared actual
experience with the forecasts of excise taxes and the key model variables. Exhibit 3- 5 on
the following page shows a comparison of the revenue forecasts from 1986 to 1990, and the
actual revenues collected by the Program.
Criteria
The criteria for evaluating the appropriateness and timeliness of excise tax revenue forecast
updates include:
Use of standard, econometric forecasting techniques
Use of recognized data sources
Overall reasonableness of assumptions, data, and adjustments
H Realism and conservatism in overall approach and use of data
Regular updates which correspond to Program needs
EXHIBIT 3- 5
( Millions of Dollars)
107.2 $ 106.3 -
1990 110.8 129.5 126.4 121.0 114.9 $ 1 10.8
199 1 113.3 143.3 140.4 132.9 123.1 117.0
1992 - 159.0 155.2 144.8 132.7 125.0
1993 - 176.7 171.5 157.9 143.1 137.0
1994 - 196.4 189.8 172.1 154.4 150.0
1995 - 218.2 210.5 188.0 166.5 162.0
1996 - 242.8 234.3 206.3 180.0 174.0
1997 270.5 261.6 226.9 195.0 188.0
1998 - 301.4 29 1.3 249.0 210.7 202.0
1999 - 336.3 322.8 275.7 228.3 218.0
2000 - 375.6 358.5 305.9 247.3 234.0
2001 - 419.9 399.8 341.1 267.8 252.0
2002 - 469.7 447.2 381.4 290.0 27 1 . O
2003 - 526.4 498.5 425.1 3 14.2 292.0
2004 - 589.7 553.8 472.2 340.6 3 14.0
2005 660.5 617.5 526.5 369.3 338.0
2006 - 43 1.5 402.7 193.3 233.8 212.3
Source: Arizona Department of Transportation
Analysis
Beginning with the initial forecasts in 1986, ADOT staff identified that forecasts were
generally high when compared to the actual excise tax collections. This was primarily
thought to be the result of economic downturns that affected the personal income and
building permit variables. Analysis showed that population and air passenger arrival
forecasted inputs were having little affect on the growing shortfall in forecasted excise tax
revenues.
In the period from 1986 to 1990, ADOT continued to use the Blue Chip Panel ( previously
Eggert Enterprises, Inc.) economists consensus measure for real personal income and adjust
it for inflation forecasted by DRI. Inflation was also adjusted in relationship to historical
differences between national and regional inflation levels. However, this measure of
personal income continued to produce results that were somewhat higher than actual for the
years 1986- 1990.
Exhibit 3- 6 on the following page shows the comparison over time.
Analysis performed by ADOT staff suggest that these data differences were the primary
reason for differences between actual and forecasted excise tax collections.
The second data concern involved the internal forecasts that ADOT prepared for building
permits. Again, due to the building slump, Tax Reform Act of 1986, savings and loan
failures, and general economic downturn, forecasts of building permit activity were
forecasted too high. Exhibit 3- 7 shows the forecasts over the period 1986- 1990.
In response to revenues and key forecast variables not meeting expectations, ADOT
determined in 1988 that more fundamental and lasting changes in the economy may be
occurring which necessitated review of the underlying data relationships and overall
modeling approach. The model was still very capable of forecasting history, however, it
was showing a consistent weakness in quickly responding to the major short- term changes
that the Arizona economy was experiencing. Considering the initial long- term focus of the
model, this was to be expected.
The first effort at improving the modeling process was a re- analysis of the current
forecasting model. This was conducted in the Fall of 1989 by Dennis Hoffman, Ph. D. of
ASU, one of the authors of the original model. This analysis developed procedures to
correct for serial correlation of the error terms in the forecasting process. It was thought
that minimizing the error terms would produce more accurate forecasts. However, this
adjusted mode1 actually produced higher revenue forecasts than the original model, and was
therefore not used by ADOT.
In 1990 ADOT again initiated efforts to improve forecasting capabilities by seeking a
method which would produce more accurate short- term forecasts while incorporating its
long- term forecasts. In this effort, ADOT contracted with The Hickling Corporation to
conduct a study of these effects. This study resulted in the report: " Risk Analysis of
Expected Revenues from the Transportation Excise Tax, Maricopa County: 1990- 2006."
EXHIBIT 3- 6
TOTAL MARICOPA COUNTY PERSONAL INCOME
FORECASTS COMPARED WITH ACTUAL INCOME
( Millions of Dollars)
ua1 July 86 July 87 July 88 July 89 July 90
7,805 $ 26,404
0,312 28,754 $ 28,210
32,676 3 1,7 15 30,495 $ 3 1,798 - -
1989 35,198 35,109 33,97 1 34,513 $ 34,873 -
1990 37,693 38,901 38,08 1 38,500 37,598 $ 37,274
* Actual income is adjusted to Convert from calendar year to fiscal year data.
I Source: Arizona Deparbnent of Transportation
The risk analysis process refines the forecasting process by incorporating independent
ranges of value for the input variables, by measuring the probability or " odds" that an
outcome will actually occur. This is accomplished by attaching ranges