PERFORMANCE AUDIT
DEPARTMENT OF REVENUE
AUDIT FUNCTION
Report to the Arizona Legislature
By the Auditor General
December 1995
Report # 95- 18
STATE OF ARIZONA
DOUGLAS R. NORTON, CPA OFFICE OF THE
AUDlTOR GENERaL
AUDITOR GENERAL
December 21, 1995
Members of the Arizona Legislature
The Honorable Fife Symington, Governor
DEBRA K. DAVENPORT, CPA
O~ P" T" A" C., TORGENER& L
Mr. Harold Scott, Director
Department of Revenue
Transmitted herewith is a report of the Auditor General, A Performance Audit of the
Department of Revenue audit function. This report is in response to a May 5, 1993,
resolution of the Joint Legislative Audit Committee. The performance audit was
conducted as part of the sunset review set forth in A. R. S. 5541- 2951 through 41- 2957.
This is the fourth in a series of reports to be issued on the Department of Revenue
( DOR). The report addresses DOR's need to develop an audit plan and tracking systems
to ensure productive use of audit resources. We also note that DOR needs to improve
its audit selection process. DOR needs to target more high- liability taxpayers for audit
and those industries or groups that typically err on their tax filings.
My staff and I will be pleased to discuss or clarify items in the report.
This report will be released to the public on December 22, 1995.
Sincerely,
~ bugl) asR . Norton
Auditor General
Enclosure
I I 2910 NORTH 44TH STREET . SUITE 410 m PHOENIX, ARIZONA 85018 . ( 602) 553- 0333 . FAX ( 6 0 2 ) 5 5 3 - 0 0 5 1
SUMMARY
The Office of the Auditor General has conducted a performance audit of the Arizona
Department of Revenue ( DOR) audit function pursuant to a May 5,1993, resolution of the
Joint Legislative Audit Committee. The audit was conducted as part of the sunset review
set forth in Arizona Revised Statutes ( A. R. S.) 5541- 2951 through 41- 2957. This is the fourth
and final audit of the Department conducted in response to the resolution.
DOR's audit function is responsible for ensuring taxpayer compliance with the State's tax
requirements and for ensuring that the State receives owed tax monies. A. R. S. 542- 117
authorizes DOR to conduct audits of taxpayers' returns and of those who failed to file a
return. In order to perform these audits, DOR has formed three audit sections: Individual
Income Tax Audit ( IIT), Corporate Income Tax Audit ( CIT), and Transaction Privilege,
Use, and Severance Tax Audit ( TPT).
During fiscal years 1990 and 1992, the Legislature provided DOR with an additional 297
audit and collections staff. These staff were provided for the purpose of collecting addi-tional
taxes owed the State. Initially these staff had a dramatic impact on dollars assessed,
from $ 69 million in fiscal year 1989 to $ 280.2 million in fiscal year 1992. However, since
fiscal year 1992, these assessments have dropped by 68 percent, to $ 89 million in fiscal
year 1995. The cause of the decline can be attributed to audit selection problems and
changes in tax laws.
Audit Plans and Tracking Systems
Could Help Management Oversight
( See pages 6 through 8)
An audit plan and better management information and tracking systems could assist DOR's
management in ensuring productive use of audit resources. Currently, DOR's audit sec-tions
do not have audit plans, and their tracking systems are limited. As a result, produc-tivity
has suffered. For example, individual income tax auditor productivity has dropped
from 1,807 audits per auditor per year in fiscal year 1993 to 851 in fiscal year 1995. The
reason for this is that DOR's primary source of individual income tax audits, a match- off
with federal data, has declined significantly because of changes to the States's tax laws
that conformed state tax calculations to more closely match federal tax calculations.
An audit plan and tracking system could help DOR prepare for and manage changes in
its workload, such as the decline in individual income tax audits, by determining alterna-tive
sources of audits and monitoring productivity. An audit plan would also enable
DOR to define productivity standards and allocate resources appropriately to achieve
productivity goals. In addition, a tracking system could help management monitor the
progress of audits through the audit process.
Further Improvements Needed
in Audit Selection Process
( See pages 10 through 14)
DOR needs to improve its audit selection processes. Three previous Auditor General re-ports
spanning a 14- year time period consistently found that DOR was not targeting a
sufficient number of high- liability taxpayers and that it had not developed audit selection
criteria and processes that ensured proper use of the criteria. To date, DOR has made
limited progress in these two areas.
DOR still needs to target more high- liability taxpayers for audit. High- liability taxpayers
comprise a significant portion of DOR's corporate income and transaction privilege tax
remittances. DOR, however, has only audited 98 of the top 239 corporate income taxpay-ers,
or 41 percent, for 1990. These audits identified an additional $ 11.8 million owed for
tax year 1990. Similarly, for 1990, DOR has audited only 59 of the top 100 companies
remitting transaction privilege tax and identified an additional $ 35.4 million in tax liabili-ties.
Other western states reported auditing a higher percentage of companies.
DOR also needs to improve its targeting of those segments of its audit populations that
are unlikely to comply with tax laws. These audits could be aimed at certain industries or
groups that typically err on their tax filings. Further, DOR needs to perform random
audits selected from the entire taxpayer population. Taxpayers are more likely to comply
with requirements if there is a potential that they may be selected for audit. Currently, no
random audits are performed because DOR views them as having a low dollar potential.
DOR needs to develop and implement systematic audit selection processes. DOR has
limited audit selection criteria and still relies primarily on auditor judgment for choosing
audits. Audit selection criteria and processes to ensure their proper utilization help en-sure
that limited audit resources are directed toward the most productive audits, and that
taxpayers are treated fairly.
Table of Contents
Paqe
Introduction and Background .......................................................... 1
Finding I: Audit Plans and
Tracking Systems Could
Help Management Oversight ....................................................... 6
Audit Planning ...................................................................................................... 6
Tracking Systems
Need Improvement... ............................................................................................ 7
Recommendations ................................................................................................. 8
Finding II: Further Improvements
Needed in Audit Selection Process ............................................ 10
Same Problems Found
in Earlier Audits .................................................................................................... 10
Better Targeting of High-
Liability Taxpayers Needed ................................................................................. 11
Noncompliant and Random
Audits Needed ...................................................................................................... 12
Systematic Audit Selection
Process Needed ..................................................................................................... 13
Recommendations ............................................................................................... 14
Sunset Factors .................................................................................. 16
Agency Response
Figure 1
Table 1
Figure and Table
Audit Function Net Assessments
Fiscal Years 1989 through 1995 .................................................. 3
Total Audits, Net Assessments, and
Average Net Assessment Per Auditor
Fiscal Years 1993 through 1995 .................................................. 5
INTRODUCTION AND BACKGROUND
The Ofice of the Auditor General has conducted a performance audit of the Arizona
Department of Revenue ( DOR) audit function, pursuant to a May 5,1993, resolution of
the Joint Legislative Audit Committee. The audit was conducted as part of the sunset
review set forth in Arizona Revised Statutes ( A. R. S.) 5541- 2951 through 41- 2957. This is
the fourth and final audit of the Department conducted in response to the resolution.
DOR's Audit Function
A. R. S. 542- 117 authorizes DOR to conduct audits of taxpayer returns and of those taxpay-ers
who failed to file a tax return. DOR's audit function helps ensure that taxpayers com-ply
with the State's tax requirements, and the State receives tax monies owed. To accom-plish
these goals, DOR has structured its audit function into three sections: Individual
Income Tax Audit ( IIT), Corporate Income Tax Audit ( CIT), and Transaction Privilege,
Use, and Severance Tax Audit ( TPT). I In fiscal year 1995, DOR's IIT section was autho-rized
138 FTEs, its CIT section 65 FTEs, and its TPT section 166 FTEs. Total expenditures
for these areas were approximately $ 11.3 million.
The process by which DOR audits taxpayers involves a number of steps. First, the audit is
identified through various audit selection programs, then the auditor obtains and ana-lyzes
necessary taxpayer information. IIT audits are primarily conducted in the office,
while both office audits and field visits are used to conduct audits of businesses and
corporations located both in Arizona and out- of- state. Finally, once the analysis is com-plete,
the auditor documents the audit results in one of three conclusions: 1) a dollar
assessment owed by the taxpayer, 2) a refund due to the taxpayer, or 3) no adjustment.
A transaction privilege tax is an excise tax on the privilege to engage in business in the State of
Arizona. This is commonly referred to as a sales tax.
A severance tax is imposed in lieu of a transaction privilege tax on the business of mining metallifer-ous
minerals and severing timber.
A use tax is imposed on the purchase price of tangible personal property from either in- state or out-of-
state vendors for which a transaction privilege tax ( equal to or greater than the Arizona rate) was
not paid.
Net Assessments Have
Declined Significantly
DOR's net audit assessments have declined sharply in the last four fiscal years. l From a
peak of $ 280.2 million in fiscal year 1992, DOR assessments dropped to $ 89.0 million in
fiscal year 1995, a 68 percent decline. Six years ago, the Legislature began a two- stage
audit and collections staffing increase at DOR for the purpose of collecting additional
taxes owed to the State. This effort, known as PIER I and PIER I1 ( Program for Increased
Enforcement Revenues), authorized the addition of 149 FTEs in fiscal year 1990 and an
additional 148 FTEs in fiscal year 1992, of which approximately 107 and 91 went to the
audit sections in respective years. Results from the PIER program increased the
Department's assessments from fiscal year 1990 to 1992. Since then, however, audit as-sessments
have declined. Figure 1 illustrates the rise and fall of DOR's audit assessments.
Figure 1
Audit Function
Net Assessment
Fiscal Years 1989 through 1995
Fiscal Years * gg ' 90 ' 9 1 ' 92 ' 93 ' 94
PIER I PIER II
shaded areas indicate the time periods PER I and PER 11 staffing increases we
Implemented
Source: Auditor General staff analysis of DOR's fiscal years 1989 through 1995 Corporate Income Tax
Weekly Reports, Transaction Privilege Tax Field Audit Summary by Fiscal Year Reports, and
Individual Income Tax Summary of Auditor Production Reports.
Net assessments include additional tax assessments charged to and refunds owed to taxpayers. These
amounts result from tax audit conclusions.
DOR's decline in audit assessments can be attributed to several reasons. The decline has
occurred in each of DOR's three audit sections. The Corporate Income Tax audit section's
( CIT) $ 40 million decline from fiscal year 1994 to 1995 resulted primarily from a 1994
corporate income tax law change, Senate Bill 1120. The decline in the Individual Income
Tax audit section's ( IIT) assessments over the past three fiscal years is due to changes in
the state tax law which conform state tax calculations more closely to the federal tax cal-culations,
thereby reducing potential taxpayer errors. The Transaction Privilege Tax au-dit
section's ( TPT) decline from fiscal year 1994 to 1995 is more difficult to determine.
DOR could not provide specific reasons, but suggested that problems with audit selec-tion,
tax law changes, and increased taxpayer compliance may explain the decline. Table
1 ( page 5) provides audit assessments and other productivity data from each of the three
audit sections from the last three fiscal years.
Audit Scope and Methodology
The purpose of this audit was to determine the effectiveness of the Department's audit
function. The audit report presents findings and recommendations in two major areas:
The need to develop audit plans and tracking systems.
4 The need to improve DOR's audit selection process.
In addition to these audit areas, the report contains a response to the 12 Sunset Factors
based on work performed during our four Department audits ( see pages 16 through 21).
A variety of techniques was used in performing tlus audit. Typical audit methods such as
interviews with Department management and staff, surveys of other states' audit func-tions,
review and analysis of agency information, and reviews of previous audit reports
were used. To determine the efficiency and effectiveness of the Department's audit func-tion,
the following steps were performed:
4 A review of all CIT field audits from fiscal years 1993 through 1995.
An analysis of CIT's 239 highest liability corporate taxpayers from 1990.
4 An analysis of TPT's 100 highest liability accounts from 1990.
Table 1
Total Audits, Net Assessments, and
Average Net Assessment Per Auditor
Fiscal Years 1993 throuqh 1995
Corporate Income Taxb
Number of Audits
Net Assessments
Average Net Assessment
Per Auditor
individual Income Tax
Number of Audits
Net Assessments
Average Net Assessment
Per Auditor
Transaction Privilege Tax
Number of Audits
Net Assessments
Average Net Assessment
Per Auditor
a Corporate income tax net assessments increased in fiscal year 1994 due to one assessment of $ 20,582,869.
CIT productivity numbers include field and office income tax audits.
~ Source: Table prepared by Auditor General staff using figures for fiscal years 1993 through 1995 obtained from Corporafe Income Tax Weekly Reports, Transaction Privilege Tax Field Audit Summary by Fiscal
Year Reports, and Individual Income Tax Summary of Auditor Production Rqorts.
I This audit was conducted in accordance with government auditing standards.
The Auditor General and staff express appreciation to the Director of the Department of
Revenue, and DOR staff, for their assistance during the course of our audit.
FINDING I
AUDIT PLANS AND TRACKING SYSTEMS
COULD HELP MANAGEMENT OVERSIGHT
DOR can strengthen management oversight and increase productivity through better plan-ning
and tracking of its work. Audit plans could help DOR manage its workload in the
face of changing conditions. In addition, better tracking systems would increase
management's ability to monitor performance and effect improvements.
Audit Planning
DOR needs to develop annual plans which could help the agency manage its resources in
a changing environment.
Currently, DOR does not have an annual audit plan in any of the three audit sections that
specifies such things as the projected number of staff hours available for audits, auditors'
expected productivity levels by types of audit, the number of audits to be performed,
and/ or the audit coverage desired. Formally establishing goals and objectives in a plan
can help DOR more proactively manage its resources.
DOR's changing environment requires careful management of its resources. DOR's work-load
and productivity can be affected by a number of factors including federal or state tax
law changes, changes in the economy, and information automation. Unless plans are de-veloped
to address these changes as a part of the ongoing workload, performance can
vary considerably. For example, the individual income tax law has been changed so that
state calculations conform more closely to federal tax calculations. This ultimately re-sulted
in a sharp reduction in the number of individual income tax audits perf0rmed. l
Auditor productivity dropped correspondingly. In fiscal year 1992, auditors each com-pleted
an average of 1,807 audits. In fiscal year 1995, auditors each completed an average
of 851 audits. In late fiscal year 1995, DOR did take action to address the decreased work-load
in individual income tax. DOR reassigned 10 positions to its collections function and
plans to reassign six more positions.
Staffing had been sigruficantly increased in fiscal years 1990 and 1992 through the PIER I and PIER I1
programs. However, there was an existing backlog of audits in the system. Therefore, the impact of
the increased number of auditors from the PIER programs and the reduction in new audits from the
tax law change did not appear until several years later.
6
However, because DOR has not specified its goals for numbers of audits, or auditor pro-ductivity,
it is difficult to know if the staffing adjustments it has already made are suffi-cient.
For example, over the past three years, DOR has audited from 3 to 5 percent of all
filed individual income tax returns. DOR management has also estimated that its audi-tors
can complete an average of 1,100 audits per year. Other states report auditing 2 per-cent
or fewer of their total individual income tax returns and the IRS audits less than 1
percent. Depending on the goals DOR sets for both number of audits and auditor produc-tivity,
it could still be overstaffed in the individual income tax audit section by three to
nine positions. Audit planning could address issues such as these.
Examples of audit plans can be found in other states such as Colorado, which prepares an
annual audit plan outlining management's resource allocation decisions. They use his-torical
data and current trends to project productivity levels based on available hours
and audit type priorities. In addition, they estimate hours that staff must spend on special
projects, training, and various administrative tasks. Each quarter, management evaluates
performance by comparing actual figures to projections. Based on the results, manage-ment
reallocates resources, if necessary, in order to accomplish productivity goals.
Tracking Systems
Need Improvement
BOR also needs to improve and make use of its tracking systems. Inadequate tracking
systems limit management's ability to monitor performance and use this information to
make improvements in their designated areas. Specifically, the cumbersome nature of the
corporate income tax audit section's ( CIT) system makes it difficult for management to
use collected information. CIT maintains partial information on several separate systems,
some computerized and some manual. To obtain status information, employees must
look at three distinct computer databases, four manual logs, and an index card system.
The following case example illustrates CIT's tracking problems:
One auditor failed to issue a $ 15,969 assessment because the period within which
DOR could issue a tax assessment expired pursuant to the statute of limitations. The
auditor needed taxpayer information to complete the audit. CIT received the informa-tion
in March of 1994; however, no one forwarded it to the auditor before the statute of
limitations expired in June of 1994 because, according to the supervisor, no one knew
about the audit in progress due to inadequate tracking information.
Although CIT needs to develop an integrated tracking system, the transaction privilege
tax ( TPT) and Individual Income Tax Audit ( IIT) sections could better utilize their exist-ing
tracking systems to improve management's planning and oversight. TPT has an auto-mated
system that provides audit status and outcome information. However, according
to management, although the system tracks audit statistics, it does not always perform
useful analyses of them. As a result, TPT must pull staff away from auditing activities to
7
analyze tracking information. In addition, IIT needs to ensure adequate use of its auto-mated
tracking system. Recently, the section added a feature to track hours spent per-forming
auditing and nonauditing activities. However, since the section has only recently
started to track this information, enough data is not yet available for management to
make resource allocation decisions based on it. DOR should ensure that when enough
data is compiled, it is utilized to make resource allocation decisions in IIT.
DOR needs to ensure that all tracking systems allow managers to efficiently and effec-tively
monitor auditor caseloads and that the systems capture statistical information nec-essary
to make resource allocation decisions. Colorado, California, and Utah utilize auto-mated
tracking systems to monitor their auditors' caseloads. Further, New Mexico annu-ally
captures and analyzes the amount of time auditors spend on work- related activities.
In addition, Texas and New Mexico analyze the number of audits conducted, the number
of auditors, total assessments, and audit hours for each tax type to make resource alloca-tion
decisions.
RECOMMENDATIONS
1. DOR should improve management oversight and tracking by formulating an audit
plan that details goals, objectives, and resource allocation decisions.
2. DOR should develop comprehensive management information and tracking systems
in all three audit sections.
FINDING II
FURTHER IMPROVEMENTS NEEDED
IN AUDIT SELECTION PROCESS
DOR needs to improve its audit selection process. Three previous Auditor General re-ports
over the last 14 years identified various problems with DOR's audit selection. DOR
needs to further improve targeting of high- liability and noncompliant taxpayers, perform
random audits, and implement systematic audit selection processes.
An effective audit selection system is vital to the success of an audit program. Because
audit resources are limited, it is important that those resources be utilized most efficiently
and effectively. According to two national experts on taxation, John F. Due and John L.
Mikesell,
"..., the methods for selection of accounts for audit are highly important, ifthe limited audit
program is to ofer maximunz efecfiveness, both in terms of generating revenue fiom the
legal tax base and of deterring noncompliance, that is, ensuring that the honest taxpayers
are not placed at a competitive disadvantage by cheats."'
Same Problems Found
in Earlier Audits
This audit found some of the same problems identified in 3 audit reports issued over the
last 14 years. For example, in 1981 ( Performance Audit Report 81- 2), DOR had not estab-lished
a systematic audit selection procedure for transaction privilege tax ( TPT) audits.
Similarly, in 1985, DOR still lacked a systematic audit selection system for TPT tax audits
and also lacked one for corporate audits ( Performance Audit Report 85- 5). In addition,
DOR was not sufficiently auditing taxpayers with the highest tax liabilities. Likewise, our
1987 follow- up report ( Performance Audit Report 87- 6) found continued problems with
DOR's audit selection system and also found that major taxpayers were not being au-dited.
' John F. Due and John L. Mikesell, Sales Taxation: State and Local Structure and Administration, Second
Edition ( 1 994).
Better Targeting of High-
Liability Taxpayers Needed
DOR needs to increase audits of high- liability taxpayers. These taxpayers, though few in
number, comprise much of the State's corporate and transaction privilege tax income and
should be targeted,
In tax year 1990, the largest 239 corporate taxpayers had a total tax liability of $ 134.8
million, or 68 percent of the total corporate tax liability. The top 100 transaction privilege
taxpayers remitted $ 670.4 million in TPT taxes in 1990, or 34 percent of the total TPT taxes
paid. Because of their significant tax liabilities and the impact of any noncompliance,
these companies should be targeted for audit.
Fewer high- liability taxpayers targeted - DOR appears to target fewer high- liability
taxpayers than other states. According to the Department's statute of limitations, a tax
return filed for a given tax year can be audited by DOR at any time within a four- year
timeframe following the tax year. The audit timeframe can be extended if the Department
has obtained a waiver. When an audit of a corporate taxpayer is conducted, DOR audi-tors
will typically consider auditing up to four years' worth of tax data.
An analysis of tax year 1990 found that DOR audited, or is in the process of auditing, 98 of
239, or 41 percent, of the largest corporate income taxpayers, These audits identified an
additional $ 11.8 million in taxes owed for tax year 1990 versus the reported 1990 tax
liability of $ 78.9 million for these 98 companiesel Further, when tax year 1990 was au-dited,
DOR also audited other tax years, resulting in millions of more dollars assessed
during the audit period. The 141 companies that DOR did not audit or has yet to audit
had a 1990 reported tax liability of $ 55.9 million.
DOR's audits of high- liability transaction privilege taxpayers also appear to be limited.
For tax year 1990, DOR has audited 59 of the largest 100 TPT taxpayers with total re-ported
tax liabilities of $ 288.4 million. These audits generated an additional $ 35.4 million
in tax liabilities for the audit period which encompassed tax year 1990. The other 41 com-panies
that were not audited reported $ 382 million in tax liability in 1990. In addition,
DOR has audited only 12 of the top 20 transaction privilege taxpayers in 1990.
' This additional liability figure does not include the assessment amounts for tax year 1990 for 24
companies that were still in the process of being audited by DOR at the time of this analysis.
Other states appear to audit more IzigIt- liabilifij taxpayers - Compared to DOR, other
western states appear to audit a higher percentage of high- liability taxpayers. Several
states, according to Due and Mikesell,
". . . establish regular audit cycles for the largest accounts, arguing that this approach
provides assurance that the tax base is protected; ... [ since] a large share of the sales tax base
is in the largest accounts. Furthermore, the productivity of the audit is likely to be high,
even in relation to the complexity of the audit."
Other states surrounding Arizona target more high- liability corporate and sales taxpay-ers.
For the largest corporate taxpayers, California audited 77 percent in 1995, and Texas
100 percent, and both Utah and Colorado target them but do not have enough resources
to audit all of them. Arizona, however, does not specifically target high- liability corpo-rate
taxpayers. For the largest sales taxpayers, California targets them on a three- year
cycle and selects those that are audit- worthy. Nevada audits all of them every other year,
and Utah and Colorado every three years. Arizona reviews all high- liability transaction
privilege taxpayers every four years.
Noncompliant and
Random Audits Needed
In addition to targeting high- liability taxpayers, DOR needs to use its remaining audit
resources to focus on noncompliant segments of the corporate and transaction privilege
tax base, as well as to perform random audits. Noncompliant taxpayers are those in a
particular industry or group that have tax compliance problems as identified through
audit research and experience. In addition, random audits should be performed to help
ensure compliance with tax laws by letting everyone know that they have the potential to
be audited.
Most ~ ro~ rcollrplinta~ xttp ayers not targeted - In addition to selecting more high- liability
taxpayer audits, DOR needs to better identify and target corporate and privilege tax trans-action
taxpayers most unlikely to comply with tax laws. For example, a review of other
states' sales taxation programs showed than many states identify noncompliant taxpay-ers
by 1) classifying sales taxpayers into industries based on problems and issues facing
those businesses and 2) assessing the productivity of past audits. In fact, according to a
recent tax publication, many states are moving toward this type of an approach to im-prove
their audit selection techniques.
Currently, DOR can improve methods to identify and target the taxpayers who are most
likely to err on their tax returns. Analysis of the Department's TPT section found it relies
heavily on auditor leads. From fiscal year 1990 through October 1994, the section selected
over 72 percent of its audits from auditor leads or tips. While this selection method is
used by most states, it is generally used in combination with other methods. In particular,
several states also give substantial attention to the results of past audits. This method
helps to identify those tax entities that often make errors on their tax returns. The TPT
section does have an audit program called " reaudit." However, this program reviews
only those taxpayers whose last audit generated an assessment above a certain amount.
In addition, we found that the CIT section does not systematically identify and examine
those taxpayers unlikely to comply with tax laws. Specifically, ClTs selection process is
performed manually and audit information is captured on index cards filed by company
name. The CIT section does not have or utilize any kind of automation to help select its
audits. Other states and the IRS are much more advanced, utilizing computerized sys-tems
to help select audits.
Rando~ lza udits not routinely peyfonlzed - Finally, DOR needs to ensure that all taxpay-ers
are potentially subject to audit. The foremost objective in audit selection is to maxi-mize
the recovery of tax revenue. An additional objective is to ensure overall compliance
with tax laws. A review of other states found that random audits are used to inform
taxpayers of the chance that they may be selected for audit. Currently, random audits are
not routinely performed because DOR does not regard them as having enough revenue
potential.
Systematic Audit
Selection Process Needed
DOR still needs to develop and implement systematic selection processes for all three
audit sections. DOR, in most cases, lacks audit selection criteria which are important for
two reasons. First, because audit resources are limited, it is vital that resources are di-rected
toward the most productive audits. Second, audit selection criteria and processes
help ensure that taxpayers are treated fairly. These types of guidelines and controls help
ensure that taxpayers are not unfairly targeted or do not go unaudited.
Audit selectio~ z& teria and processes treed i~~ zprovemen- t Previous Auditor General
reports criticized DOR for the lack of audit selection criteria. This review found that the
CIT section has some limited written audit selection criteria, but does not utilize them,
relying primarily on auditor judgment and geographic location to determine which au-dits
are performed. The TPT section has developed some audit selection procedures with
limited audit selection criteria; however, there is no tracking or documentation to deter-mine
whether the criteria are used. Finally, the IIT section relies primarily on auditor
judgment to select audits.
In addition to having audit selection criteria, DOR needs to put processes in place to
guide audit selection staff in the use of the criteria. In addition, these processes should
include supervisory review to help ensure that the audit selection criteria are being uti-lized
and applied appropriately, as well as to help ensure that all audit selection proce-dures
are being followed.
RECOMMENDATIONS
DOR can improve audit selection by:
1. Targeting a greater percentage of high- liability taxpayers for audit.
2. Increasing its targeting of noncompliant segments of its audit populations.
3. Performing random audits of taxpayers to encourage compliance with requirements.
4. Developing, documenting, and utilizing audit selection criteria.
5. Developing audit selection procedures that include supervisory monitoring and over-sight.
SUNSET FACTORS
In accordance with A. R. S. 541- 2954! the Legislature should consider the following 12 fac-tors
in determining whether the Arizona Department of Revenue ( DOR) should be con-tinued
or terminated.
1. Objective and purpose in establishing the Department.
Originally created in 1973, the Department of Revenue was established in part to
provide an integrated, coordinated, and uniform system of tax administration and
revenue collection for the State. In 1974, the Department assumed the powers and
duties of the Department of Property Valuation and the Estate Tax Commissioner,
as well as certain functions of the State Tax Commission.
The Department is responsible for the licensing, processing, collection, and en-forcement
of most taxes for the State of Arizona, including transaction privilege,
use and severance, individual and corporate income, withholding, luxury, prop-erty,
estate, fiduciary, bingo, flight property, private car, 911 excise, telecommuni-cations
devices, municipal water, and waste tire. In addition, the Department col-lects
taxes for over 73 local jurisdictions and various excise taxes for 14 counties.
Also, the Department is responsible for appraising the value of large, complex
properties in the State such as mines, utilities, railroads, and pipelines, and moni-tors
county performance in assessing other property.
2. The effectiveness with which the Department has met its objectives and pur-pose
and the efficiency with which the Department has operated.
Our reviews identify several areas where DOR needs to improve its effectiveness
and efficiency in fulfilling its statutory responsibilities. For example:
In our review of the Administrative Services and Taxpayer Support functions,
we noted that DOR could increase revenue and improve the efficiency and
effectiveness of some of its programs. For example, DOR could increase rev-enue
and improve tax processing efficiency by expanding the use of electronic
funds transfer ( EFT). In addition, DOR needs to improve telephone access to
the Taxpayer Information and Assistance Unit ( TIA). Also, the Department needs
to develop a more aggressive unclaimed property program to collect abandoned
property and, when possible, return it to rightful owners. Further, DOR is not
ensuring that criminal complaints related to bingo operations are investigated.
Our review of the compliance function found that the Department could in-crease
delinquent collections ( Performance Audit Report 95- 17). DOR could
collect additional millions of dollars owed to the State by following levy and
lien guidelines, using automatic withdrawal from taxpayer bank accounts for
installment agreement payments, and allowing taxpayers to pay their debt with
credit cards. In addition, DOR could collect an additional $ 6.6 million by add-ing
staff. Also, DOR can do much more to identify and collect delinquent taxes
from state employees and vendors. Further, our review indicates DOR could
more effectively utilize its collection resources by implementing a debtor pro-filing
program, which ranks the probability of collecting each account based on
past payment history.
In our review of the audit function, we noted that DOR could strengthen man-agement
oversight and improve the audit selection process ( see pages 6 through
14). DOR's audit assessments have dropped by more than $ 139 million from
fiscal years 1993 through 1995. DOR could improve productivity by develop-ing
and implementing an audit plan and a tracking system. In addition, DOR
needs to strengthen its audit selection process.
Our audit work in the property valuation and equalization area indicates that
DOR can improve administration of its property tax duties. ( Performance Au-dit
Report 95- 15). DOR needs to implement better methods to improve equity
and consistency of property values. In addition, DOR's centrally valued prop-erty
audit function needs to be more productive. Further, the Department has
inadequate controls in place to review and ensure the validity of property tax
data.
The extent to which DOR has operated within the public interest.
The Department of Revenue has operated within the public interest by administer-ing
and enforcing state tax laws, which generate revenue essential to the operation
of state government. Any programs ( education, entitlementprograrns, public health,
public safety) funded through the general fund are directly affected by the
Department's efforts. DOR's annual report shows $ 4.89 billion collected in fiscal
year 1994, $ 3.44 billion of which was deposited in the State General Fund. The
remainder was distributed to cities, counties, specific funds, or refunded.
However, the Department could do more to operate in the public interest. The
Department could do more to ensure equity and consistency of property values by
issuing equalization orders and requesting reappraisals when uniformity of prop-erty
values is poor. The Department could also do more to expand taxpayer access
to information and assistance by implementing various technological remedies. In
addition, DOR could be more aggressive in locating the owners of unclaimed prop-erty
by using the same techniques it uses to find individuals who owe money to
the State.
4. The extent to which rules and regulations promulgated by DOR are consis-tent
with the legislative mandate.
Based on our review, discussions with the Attorney General's Office, and informa-tion
provided by DOR, all rules promulgated are consistent with each division's
legislative mandate. With the exception of legislation enacted this year, all rules
currently mandated by statute have been promulgated.
5. The extent to which DOR has encouraged input from the public before pro-mulgating
its rules and regulations, and the extent to which it has informed
the public as to its actions and their expected impact on the public.
We found that the Department has not fully complied with open meeting law re-quirements.
The Department of Revenue supports three committees including the
Valuation Technical Advisory Committee, the Assessment Advisory Committee,
and the Education, Training, and Certification Committee. These committees hold
meetings to develop action plans for legislative implementation and recommenda-tions
for tax policy modification. Our review found that DOR failed to notify the
Secretary of State's Office regarding where public hearing notices for meetings of
these committees were posted, as required by A. R. S. § 38- 431.02. As a result, it is
unclear whether DOR posted public hearing notices at all required locations.
Despite these incidences, the Department uses several avenues to keep the public
informed of its actions. According to the Agency's Director, DOR publishes a
monthly newsletter and regularly mails inserts to taxpayers informing them of
changes that might affect them. The Department also organizes focus group meet-ings
with taxpayers and liaison groups with established industry representatives
such as the State Bar, Arizona Society of Certified Public Accountants, and the
Arizona Forum for Improved Taxation. Additionally, educational seminars are
put on by the Department for tax practitioners and the general public.
6. The extent to which DOR has been able to investigate and resolve complaints
that are within its jurisdiction.
DOR may receive complaints about an action taken by the Department, complaints
about DOR employees, complaints asserting citizens are not paying their taxes, or
complaints about DOR licensees. To improve complaint resolution, the Depart-ment
established an ombudsman's office in 1984 and formalized it into the Ari-zona
Taxpayer Assistance Office in 1986 in accordance with A. R. S. 542- 139.01.
Among several duties, the Office receives and evaluates complaints of improper,
abusive, or inefficient service by DOR employees, provides expeditious service to
taxpayers whose problems are not resolved through normal channels, negotiates
with Department personnel to resolve the most complex and sensitive taxpayer
problems, and takes action to stop or prohibit the Department from taking an ac-tion
against a taxpayer, who, in the judgment of the Problem Resolution Officer, is
suffering or will suffer a significant hardship. During fiscal year 1994- 95, the Prob-
lem Resolution Officer received 2,013 taxpayer contacts and filled 607 requests for
information. In addition, the Criminal Investigations Unit processed 1,019 crimi-nal
complaints during the same period.
The Department may also investigate instances of taxpayer illegal conduct and
seek criminal indictments and prosecution through the Arizona Attorney General's
Office.
The extent to which the Attorney General or any other applicable agency of
state government has the authority to prosecute actions under the enabling
legislation.
A. R. S. 542- 104. E authorizes the Attorney General's Office to prosecute actions nec-essary
to enforce titles 42 and 43, which enable the Department of Revenue to carry
out its prescribed duties. According to DOR, violations of agency statutes may be
referred to the Attorney General for enforcement.
8. The extent to which DOR has addressed deficiencies in its enabling statues
which prevent it from fulfilling its statutory mandate.
Although the Department has not sought legislation to amend its enabling stat-utes,
it has sought numerous technical and administrative changes and additions
to agency statutes over the years. Several bills were passed in the years 1991,1992,
1993,1994, and 1995. For example, in 1991, House Bill 2155 established a new pro-vision
which required taxpayers to include their tax identification number when
filing returns, statements, or other documents with DOR. In 1992, Senate Bill 1156
established DOR's authority to require certain taxpayers to remit their tax pay-ments
to DOR by electronic funds transfer. In 1993, Senate Bill 1083 amended pre-viously
existing legislation related to corporate estimated payment provisions and
provided a specific civil penalty for a taxpayer who fails to pay the full amount of
estimated income tax. In 1994, Senate Bill 1489 amended several existing provi-sions
related to, among other things, the modification and expansion of the
Taxpayer's Bill of Rights, establishing the Taxpayer Problem Resolution Officer's
position. Furthermore, in 1995, the Legislature passed Senate Bill 1374, which re-vised
the Department's administrative and enforcement procedures to allow con-tracts
with tribal governments and the exchange of confidential information with
these agencies.
9. The extent to which changes are necessary in the laws of the Department to
adequately comply with the factors listed in the sunset laws.
Based on our review and input from DOR, the Legislature should consider the
following statutory changes to improve the Department's efforts:
The Legislature should consider replacing the current property valuation stan-dard
with the International Association of Assessing Officers ( IAAO) recom-
mended valuation standard of full market value with a window of 10 percent
( Performance Audit Report 95- 15).
H The Legislature should consider providing DOR with additional authority to
enforce equalization orders. ( Performance Audit Report 95- 15).
H The Legislature should consider providing DOR with the authority to order
reappraisals based on inconsistent property values. ( Performance Audit Re-port
95- 15).
The Legislature should consider amending A. R. S. 5542- 179.01( D) and 42-
179.03( E) to provide the Department with statutory authority to assess back
taxes, penalties, and interest on centrally valued properties whether the tax-payer
intentionally or unintentionally underpaid ( Performance Audit Report
95- 15).
The Legislature should consider amending A. R. S. 542- 1451 to allow DOR to
charge an administrative fee for collection and administration of local taxes
( Performance Audit Report 95- 16).
H The Legislature should consider amending A. R. S. $ 544- 302 through 44- 316 to
reduce the holding periods on most types of property from five to three years
and even lower for some other types of property, such as utilities deposits and
refunds ( Performance Audit Report 95- 16).
The Legislature should consider amending A. R. S. 544- 318 to reduce paid news-paper
advertising of unclaimed property owners from four times a year to one
time per year, and use savings to fund additional locator services to return
property to rightful owners ( Performance Audit Report 95- 16).
H The Legislature should consider amending DOR's statutes enabling the De-partment
to implement a credit card program allowing taxpayers to use credit
cards to pay delinquent taxes. In addition, statutes would need to be revised to
allow DOR to absorb processing fees common to credit card transactions ( Per-formance
Audit Report 95- 17).
10. The extent to which the termination of the Department would significantly
harm the public health, safety or welfare.
Termination of the Department of Revenue could significantly harm the public
health, safety, and welfare if state funds were not available to ensure the continued
operation of essential state health- and safety- related services. In addition, the ter-mination
of the Department would eliminate enforcement and compliance pro-grams
that are needed to ensure that all taxpayers pay their fair share of the tax
burden.
11. The extent to which the level of regulation exercised by the Department is
appropriate and whether less or more stringent levels of regulation would be
appropriate.
The current level of regulation appears generally appropriate, especially given the
fact that the Department of Revenue is, generally speaking, not a regulatory agency.
Nevertheless, the Department has been given some regulatory authority in the
areas of licensing and registration, property tax administration, and bingo code
enforcement and regulation.
In addition, A. R. S. 542- 141( A)( 1) charges the Department with exercising ". . . gen-cral
supervision over county assessws in the administration of the state property tax laws
fm the purpose of ensuring that all property is uniformly valued fm state property tax
purposes." Our review, however, found that DOR may need to be more aggressive
in its efforts to ensure property values are equalized across the State. In addition,
our review found that the Legislature should consider providing DOR with addi-tional
authority to enforce equalization orders. Further, the Legislature should con-sider
providing DOR with the authority to order reappraisals based on inconsis-tent
property values. ( Performance Audit Report 95- 15). Finally, our review found
that the Department has failed to ensure that criminal complaints related to bingo
operations are enforced as required by A. R. S. $ 55- 401 through 5- 415 ( Performance
Audit Report 95- 16).
12. The extent to which the Department has used private contractors in the per-formance
of its duties and how effective use of private contractors could be
accomplished.
Private contractors are used extensively by the Department for activities such as
temporary services, computer programming, legal services, and printing. In addi-tion,
the Department also contracts for other services including news services, eco-nomic
forecasting information, some specialized training, electronic transfer pro-cessing,
credit reports, telephone access reports, and small dollar or out- of- state
collections. Given the nature of the Department's current functions, BOR's use of
private sector contractors appears appropriate.
Agency Response 11
FIFE SYMINGTON
GOVERNOR
ARIZONA DEPARTMENT OF REVENUE
1600 WEST MONROE - PHOENIX, ARIZONA 85007- 2650
HAROLD SCOTT
DIRECTOR
December 18, 1995
Mr. Douglas R. Norton, Auditor General
Ofice of the Auditor General
29 10 North 44th Street, Suite 4 10
Phoenix, AZ 850 18- 7243
Dear Mr. Norton:
We have received the final report of your performance audit of the Department of
Revenue @ OR) audit function. Following is our response to that audit report:
DOR has one of the most successful audit programs in the country
The report failed to place the Department's program in perspective. In July 1994, the
Multistate Tax Commission ( a national organization of taxing agencies created by
compact between member states) conducted a study of audit production for corporate
income and sales taxes. The study was aimed at audits of multistate taxpayers, although
some states responded to the study with data covering all taxpayers, not just multistate
taxpayers. Arizona does not segregate its data in that form and, therefore, the data for
Arizona applies to all corporate taxpayers. This will bias the data against Arizona.
Despite that bias, and despite the fact that Arizona is a small state with relatively small
taxpayer presence and relatively low tax rates, Arizona's corporate income and
transaction privilege tax audit programs are among the very most productive in the
country.
Using average Arizona data that the Auditor General collected demonstrates that the
corporate audit program is at least the sixth most productive such program in the country
trailing only relatively small states with small audit programs that permit the states to
select only the most productive audits ( e. g., Rhode Island with one auditor). Arizona
leads such states as California, Florida and Illinois which have large corporate presences
which naturally results in substantial audit changes ( a 10% change in the tax computation
OTHER LOCATIONS: Tucson Government Mall - 400 W. CONGRESS - TUCSON
East Valley - 144011460 E. SOUTHERN - TEMPE
Mr. Douglas R. Norton, Auditor General
December 18, 1995
Page 2
for a company which conducts 20% of its business in one of these large states will
produce a much larger dollar adjustment than a similar 10% change in Arizona where the
corporate presence may represent only 3% of the company's activities).
The Arizona transaction privilege and use tax audit program displays similar results.
Using averaged data and comparing Arizona to the other states in the Multistate Tax
Commission report, the Arizona audit program ranks fourth highest in the nation. Once
again, Arizona's program is more productive than the programs in such large states as
California ( 6% sales tax rate), Florida ( 6% rate) and Illinois ( 6.25% rate), all of which
have a higher tax rate.
While there is no similar study for Individual Income Tax Audit, it is noteworthy that the
average budget for the individual income tax audit section for the past five years is
approximately $ 3.5 million while the average assessment amount for the section is $ 69.4
million. This represents a return on investment of approximately 20 to 1.
The Department's Audit Selection program is an effective one
Modern management theory suggests that the best way to evaluate programs is through
their outputs rather than through some theoretical model. The ultimate output of an audit
selection program is the results of the audits. That is the approach taken by the only major
scholarly study performed on state tax audit selection -- An Evaluation of the Sales Tax
Audit Program of the New York State Department of Taxation and Finance, James T.
Barrese and Richard D. Pomp, Journal of State Taxation , Vol. 5, No. 1, Spring 1986.
Measured by its outputs, it is clear that the audit selection methodology of the Arizona
Department of Revenue is an effective one. The outstanding ranking of the Department's
audit programs when benchmarked against other states is clear evidence of this
conclusion.
The Auditor General looked at the audit plans, particularly with respect to the auditing of
high- liability taxpayers, of our neighboring Western states -- California, Utah and
Colorado for guidance in this area and compared Arizona to them unfavorably. Again,
based on actual audit results, this is an incorrect analysis.
The Department has contacted these states and secured their audit production per auditor
statistics for sales and use taxes for fiscal year 1994- 95. Accounting for assessments not
issued because of statutory changes in the 1994 legislative session, the assessment per
auditor comparison is as follows:
Mr. Douglas R. Norton, Auditor General
December 18, 1995
Page 3
STATE ASSESSMENT PER AUDITOR
ARIZONA $ 530,269
COLORADO $ 343,293
UTAH $ 388,85 1
CALIFORNIA $ 345,291
The Department believes that this comparison demonstrates the effectiveness of the
Department's audit selection criteria compared to the benchmark system proposed by the
Auditor General based on the activities of these states.
" Targeting" large taxpayers would be wrong and ineffective
The Department disagrees with the report's conclusion that it should " target" high
liability taxpayers for audit just because they have high sales volume.
DOR believes that large taxpayers should be audited when the appropriate audit indices
suggest that non compliance is likely to have occurred. Arizona reviews the tax returns of
all high liability taxpayers, their audit history and the experience of the Department in
auditing other, similar taxpayers. Based on these criteria, a decision is made whether to
subject those taxpayers to an audit.
This approach is consistent with the Due and Mikesell book cited by the Auditor General.
In fact, that book cites Arizona as one of the states that reviews large taxpayers. The
book does not state that size should be the exclusive determinant for audit selection.
Rather, it cites a number of audit selection criteria including audit leads, review of returns,
analysis of data on returns and previous audit experience which is consistent with the way
Arizona approaches audit selection.
Taxpayers who have established that they comply with the tax laws should not be audited
based on an arbitrary implementation of an audit selection policy. The Pomp Barrese
study determined that the " cell system" which ranked taxpayers for audit selection based
on 57 state industrial groups and 10 gross sales classes proved to be "... among the
poorest performers of the 19 different sources studied. "
Mr. Douglas R. Norton, Auditor General
December 18, 1995
Page 4
Both the scientific literature and the elemental principles of fairness dictate that the audit
selection criteria should not be based upon size alone. The Department results, using a
more dynamic method of analysis, demonstrates the validity of that position.
DOR's audit selection program does focus on non- compliant
taxpayers
Contrary to the assertion by the Auditor General, DOR's major audit selection focus is on
taxpayers likely to be non- compliant. The audit selection criteria utilized by DOR
selection staff is likelihood and dimension of non- compliance. The Department utilizes the
professional judgment of audit staff to reach this conclusion based upon a number of
factors, not the least of which is the knowledge base of the audit staff. This personal
experience was found by Pomp and Barrese to have "... performed relatively well, yielding
an average assessment per auditor day that was more than twice as large as the average
CCS [ cell system] audit. "
Random audits are not a fair and effective use of Department
resources
Audits are not without costs -- both to the taxpayer and the Department. Since a random
audit is not driven by a likely non- compliance, it can only be accomplished if it constitutes
a review of the major items on a taxpayer's return. Recent federal efforts to engage in
such auditing activities were met with public outcry of the unfairness of using taxpayers to
accomplish governmental goals having nothing to do with the expected accuracy of the
taxpayer's return.
The Department does not believe it is fair to subject taxpayers to audit to send a message
to other taxpayers. Whatever benefits might be realized from taxpayers who, fearing an
audit, might more accurately report their taxes will be more than offset because sending
that message is not without cost to the Department or taxpayers selected for random
audit.
Fiscal year 1994- 95 audit assessments were adversely impacted by
policy changes in tax law
Several changes were made to the tax laws during 1994. The largest, and most
significant, is referenced in the report: SB 1120. Senate Bill 1120 authorized corporate
groups of companies to elect to file a single return and treat all of their income as if it was
one entity apportioning to Arizona their total gains and losses irrespective of the
relationship, if any, of those gains and losses to Arizona activities. This law permitted the
taxpayers to elect this treatment retroactively to 1986.
Mr. Douglas R. Norton, Auditor General
December 18, 1995
Page 5
During this past fiscal year ( 1994- 95) the corporate audit section would have assessed
another $ 70 million but for this retroactive effect. Thus, the Section would have had its
most productive year yet. Instead, these audits had to be canceled and new audits
scheduled to review the amended returns of these taxpayers.
Retroactive changes were also made in the Transaction Privilege Tax audit area. These
changes included such exemptions as printing for out of state delivery, cars used by
educational institutions and their related entities, and leasing of trucks to motor carriers.
Accounting for audits not issued as a result of these laws generates an additional $ 22
million plus that would have been assessed in 1994- 95.
Finally, continued changes to the individual income tax laws have reduced audit
assessments. Greater conformity to the Internal Revenue Code has led to fewer errors by
taxpayers on their Arizona returns, and the significant reduction in individual income tax
rates accomplished by the Legislature and Governor during the past several years have
made the impact of any audit change significantly less. By Calendar year 1995, the
reduction in rates across the board is approximately 20% compared with tax year 1993.
Reductions in the commercial lease tax rate in the transaction privilege tax have had a
similar impact.
Nonetheless, the Sections have redesigned their audit programs and adjusting for these
changes. Individual income tax audits are up 50% after the first four months of this fiscal
year. Transaction Privilege and Use Tax audit assessments are up almost 100% during the
same period.
While the Department believes that its current, focused audit
selection systems are superior to those recommended by the
Auditor General, the Department will implement a test of the
recommendations
The Department has always taken seriously the recommendations of the Auditor General.
The Department is prepared to test in a controlled environment the recommendations of
the Auditor General and determine whether they will actually improve the audit selection
programs of the Department.
The Department has audit tracking systems but they can be
improved given adequate resources
The Department does have tracking systems for its audits, although, as noted in the report,
the ones for corporate income tax are the least thorough and sophisticated. On the other
hand, corporate income tax poses some unique problems in the development of tracking
systems given the movement of corporations in and out of various filing groupings.
Mr. Douglas R. Norton, Auditor General
December 18, 1995
Page 6
The Department will attempt to improve its tracking systems as it finds budgetary
resources to devote to the effort.
The Department does maintain audit plans, but not in the form
recommended by the Auditor General
Every year each audit section develops a plan for next succeeding five years that includes
the total number of audits that will be conducted and in some cases the total number of
audits by audit type. While these plans may not be packaged in the form utilized by
Colorado ( which one may recall has significantly lower productivity per auditor), that is in
part because the Department considers them to be dynamic plans that are subject to
constant adjustment as audit experience evolves. These plans are contained in overall
planning documents for each area, rather than in one all- inclusive audit plan document.
While this approach has been a productive one for the Department, the Department will
consider packaging these plans in a more formalized document.
Sincerely,
Harold Scott
Director
