IRS' 1996 Tax Filing Season: Performance Goals Generally Met; Efforts to
Modernize Had Mixed Results (Letter Report, 12/18/96, GAO/GGD-97-25).

Pursuant to a congressional request, GAO reviewed the Internal Revenue
Service's (IRS) overall performance during the 1996 tax filing season,
focusing on: (1) changes in 1996 that relate to taxpayer services and
the processing of taxpayer refunds; and (2) some of IRS' efforts to
modernize its processing activities.

GAO found that: (1) IRS met or exceeded its timeliness and accuracy
goals for processing individual income tax returns and issuing taxpayer
refunds, answered more telephone calls from taxpayers seeking assistance
than it had planned to answer, and received more returns through
alternative filing methods than it had projected; (2) for the 1996 tax
filing season, IRS revised its procedures to limit the number of delayed
refunds to the volume of cases it could review, and focus on the cases
most in need of review, and as a result, IRS delayed many fewer refunds
in 1996 than it did in 1995 and avoided the kind of negative press it
received in 1995 as taxpayers and tax return preparers reacted to the
delays; (3) recognizing that much could be done to improve its systems
and procedures, IRS has initiated several modernization efforts, and
those efforts achieved mixed results in 1996; (4) IRS is developing a
strategy to increase the use of electronic filing and reassessing its
strategy for processing paper returns; and (5) IRS' decision to have
taxpayers send not only their payments but also their tax returns to a
lockbox and to have the banks sort those returns before sending them to
IRS has increased program costs, unnecessarily in GAO's opinion, by $4.7
million.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  GGD-97-25
     TITLE:  IRS' 1996 Tax Filing Season: Performance Goals Generally 
             Met; Efforts to Modernize Had Mixed Results
      DATE:  12/18/96
   SUBJECT:  Systems conversions
             Tax returns
             Tax refunds
             Tax administration systems
             Electronic forms
             Tax credit
             Customer service
             Personal income taxes
IDENTIFIER:  Earned Income Tax Credit
             IRS Earned Income Credit Compliance Initiative
             IRS TeleFile Program
             Internet
             World Wide Web
             IRS Volunteer Income Tax Assistance Program
             IRS Tax Counseling for the Elderly Program
             IRS TeleTax System
             IRS 1040PC Program
             IRS Distributed Input System
             IRS Service Center Recognition/Image Processing System
             IRS Document Processing System
             IRS Questionable Refund Program
             IRS FedState Electronic Filing Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Chairman, Subcommittee on Oversight, Committee on Ways
and Means, House of Representatives

December 1996

IRS' 1996 TAX FILING SEASON -
PERFORMANCE GOALS GENERALLY MET;
EFFORTS TO MODERNIZE HAD MIXED
RESULTS

GAO/GGD-97-25

IRS' 1996 Tax Filing Season

(268717)


Abbreviations
=============================================================== ABBREV

  DIS - Distributed Input System
  DPS - Document Processing System
  EIC - Earned Income Credit
  FMS - Financial Management Service
  IRS - Internal Revenue Service
  QRP - Questionable Refund Program
  SCRIPS - Service Center Recognition/Image Processing System
  SSN - Social Security Number
  TCE - Tax Counseling for the Elderly
  VITA - Volunteer Income Tax Assistance

Letter
=============================================================== LETTER


B-271107

December 18, 1996

The Honorable Nancy L.  Johnson
Chairman, Subcommittee on Oversight
Committee on Ways and Means
House of Representatives

Dear Chairman Johnson: 

This report responds to your request that we assess the Internal
Revenue Service's (IRS) performance during the 1996 tax filing
season.  Specifically, we discuss (1) IRS' overall performance during
the filing season, including information on changes in 1996 that
relate to taxpayer services and the processing of taxpayer refunds,
and (2) some of IRS' efforts to modernize its processing activities. 
In March 1996, we testified before the Oversight Subcommittee on the
interim results of our work.\1


--------------------
\1 Tax Administration:  IRS' Fiscal Year 1996 and 1997 Budget Issues
and the 1996 Filing Season (GAO/T-GGD-96-99, Mar.  28, 1996). 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

IRS uses various indicators to measure its performance during a
particular filing season.  On the basis of those indicators alone,
the 1996 filing season would be considered generally successful.  For
example, IRS met or exceeded its timeliness and accuracy goals for
processing individual income tax returns and issuing taxpayer
refunds, answered more telephone calls from taxpayers seeking
assistance than it had planned to answer, and received more returns
through alternative filing methods than it had projected. 
Contributing to the generally successful filing season were
improvements in two areas that were the sources of problems in 1995
(i.e., answering taxpayers' telephone calls for assistance and
issuing taxpayer refunds). 

As compared with the 1995 filing season, IRS received about 119
million fewer telephone calls for assistance and answered about 4
million more calls during the 1996 filing season.  That combination
of reduced demand and increased service enabled IRS to increase the
percentage of calls answered from 8 to 20 percent and to increase the
percentage of callers who were eventually able to get through to IRS
from 41 to 50 percent.  Although these results are much better than
last year, there is still considerable room for improvement. 

In 1995, IRS delayed about 7 million refunds to allow itself time to
verify Social Security Numbers (SSN), with an emphasis on returns
claiming the Earned Income Credit (EIC).\2

Although that verification effort had some positive results, there
were problems.  For example, IRS identified many more missing or
invalid\3 SSNs than it was able to pursue and eventually released the
refunds, after several weeks, without resolving the problems.  For
the 1996 filing season, IRS revised its procedures to (1) limit the
number of delayed refunds to the volume of cases it could review and
(2) focus on the cases most in need of review.  As a result, IRS
delayed many fewer refunds in 1996 than it did in 1995 and avoided
the kind of negative press it received in 1995 as taxpayers and tax
return preparers reacted to the delays. 

Although IRS met its timeliness and accuracy goals for processing
returns and refunds during the 1996 filing season, IRS set those
goals on the basis of what it expected that it could achieve with the
systems and procedures currently in place.  Recognizing that much
could be done to improve its systems and procedures, IRS has
initiated several modernization efforts.  Those efforts achieved
mixed results in 1996.  For example, although a new document scanning
and imaging system performed better in 1996 than it did in 1995, the
system still was not meeting performance expectations and may
eventually cost much more than originally estimated.  Also, although
the system that allows certain taxpayers to file their returns by
telephone worked very well and had a high rate of user satisfaction,
only about 10 to 14 percent of those eligible to use the system in
1996 did so. 

For all but one of the modernization projects discussed in this
report, IRS is taking steps to enhance its efforts.  For example, IRS
is developing a strategy to increase the use of electronic filing and
reassessing its strategy for processing paper returns.  The one
exception is IRS' use of lockboxes, which are postal rental boxes
serviced by commercial banks, to process Form 1040 tax payments. 
IRS' decision to have taxpayers send not only their payments but also
their tax returns to a lockbox and to have the banks sort those
returns before sending them to IRS has increased program costs,
unnecessarily in our opinion, by about $4.7 million. 


--------------------
\2 The EIC is a refundable tax credit available to low-income working
families with children and certain taxpayers without children. 

\3 An invalid SSN is one that does not match Social Security
Administration records. 


   OBJECTIVE, SCOPE, AND
   METHODOLOGY
------------------------------------------------------------ Letter :2

Our objective was to assess IRS' performance during the 1996 filing
season, including some of IRS' initiatives to modernize its
processing activities.  To achieve our objective, we

  -- interviewed IRS National Office officials and IRS officials in
     the Atlanta, Cincinnati, and Kansas City service centers who
     were responsible for the various activities we assessed;\4

  -- interviewed staff from the Department of the Treasury's
     Financial Management Service (FMS) about the use of lockboxes to
     process Form 1040 tax payments;

  -- analyzed filing season related data from various IRS sources,
     including its Management Information System for Top Level
     Executives;

  -- visited four walk-in assistance sites (two in Atlanta and one
     each in Kansas City, MO, and Mission, KS) to interview staff and
     taxpayers;

  -- visited two banks in Atlanta and St.  Louis that were being used
     by IRS as lockboxes to process tax remittances and analyzed
     cost/benefit data related to IRS' use of lockboxes;

  -- reviewed data on the results of and costs associated with IRS'
     decision to allow filers of paper returns to request direct
     deposits of their refunds;

  -- reviewed data on IRS efforts to identify and resolve
     questionable refund claims;

  -- reviewed computer system availability reports and periodically
     attended weekly operational meetings held by IRS' Network and
     Operations Command Center in February, March, and April 1996;

  -- analyzed IRS' toll-free telephone system accessibility data,
     telephone activity data for forms distribution centers, and
     accessibility reports for the IRS system (known as TeleFile)
     that enables some taxpayers to file their returns by telephone;

  -- reviewed data compiled by IRS, including the results of a user
     survey, on the performance of TeleFile; and

  -- reviewed relevant IRS internal audit reports. 

We did our work from January 1996 through September 1996 in
accordance with generally accepted government auditing standards.  We
requested comments on a draft of this report from the Commissioner of
Internal Revenue or her designated representative.  On November 6,
1996, several IRS officials, including the Assistant Commissioner for
Forms and Submission Processing, the National Director for Submission
Processing, and the National Director for Customer Service (Planning
and Systems), provided us with oral comments.  Their comments were
reiterated in a November 18, 1996, memorandum from the Acting Chief
of Taxpayer Service.  IRS' comments are summarized and evaluated on
pages 24 and 25.  IRS also provided some factual clarifications that
we have incorporated in the report where appropriate. 


--------------------
\4 We selected these three service centers because we had staff
available to do the work in those cities. 


   IRS GENERALLY MET ITS FILING
   SEASON GOALS FOR FISCAL YEAR
   1996
------------------------------------------------------------ Letter :3

Appendix I has data on 12 indicators that IRS uses to assess its
filing season performance.\5

These indicators relate to workload, such as the number of answered
telephone calls from taxpayers who are seeking assistance;
timeliness, such as the number of days needed to process returns or
issue refunds; and quality, such as the accuracy of IRS' answers to
taxpayer questions and the accuracy with which IRS processes
individual income tax returns and refunds.  As shown in appendix I,
IRS met or exceeded 11 of the 12 performance goals for the 1996
filing season and almost met the 12th goal (the number of
forms-ordering calls answered). 

Two specific aspects of IRS' filing season performance that are of
particular interest to taxpayers and that were the source of problems
in 1995 are (1) the level of taxpayer service being provided during
the filing season, especially the ability of taxpayers to reach IRS
by telephone, and (2) the timely issuance of refunds.  In 1995, as in
the past several years, taxpayers who sought answers to questions
about the tax law or their accounts had considerable difficulty
reaching IRS by telephone.  In 1996, IRS improved its telephone
accessibility while, at the same time, it reduced the availability of
face-to-face services at its walk-in sites.  Also, in 1995, millions
of persons had their refunds delayed as a result of new IRS
procedures for verifying the SSNs of dependents and EIC-qualifying
children.  The new procedures were designed to better ensure that
persons were entitled to the dependents and EICs they were claiming. 
In 1996, IRS implemented revised case selection criteria that
resulted in many fewer refund delays than in 1995.  Sufficient
information was not available when we completed our audit work to
assess the impact of IRS' revised procedures on the identification
and correction of questionable SSNs. 


--------------------
\5 Two other indicators (the number of individual income tax returns
received and the number filed through alternative methods) are not
included in the appendix but are discussed later in this report. 


      TELEPHONE ACCESSIBILITY
      IMPROVED WHILE THE
      AVAILABILITY OF WALK-IN
      ASSISTANCE DECLINED
---------------------------------------------------------- Letter :3.1

IRS officials have reaffirmed that service to taxpayers remains a
primary goal.  However, IRS took steps in 1996 to change the blend of
methods that it uses to deliver that service.  IRS placed more
emphasis on providing telephonic and computer-oriented service (such
as a new World Wide Web site on the Internet) while walk-in,
face-to-face assistance was deemphasized.  As a result, telephone
accessibility improved while many walk-in sites either closed or
offered a reduced level of service. 


         TELEPHONE ACCESSIBILITY
         IMPROVED
-------------------------------------------------------- Letter :3.1.1

An important indicator of filing season performance is how easily
taxpayers who have questions are able to contact an IRS assistor on
the telephone (i.e., telephone accessibility).  In reports on past
filing seasons, we discussed the difficulty taxpayers have had in
reaching IRS over its toll-free tax assistance telephone line.\6

Accessibility, as we define it, is the total number of calls answered
divided by the total number of calls received.  The total number of
calls received is the sum of the following:  (1) calls answered, (2)
busy signals, and (3) calls abandoned by the caller before an
assistor got on the line.  By our definition, accessibility of IRS'
toll-free telephone assistance improved in 1996, although it was
still low.  From January 1 to April 20, 1996, IRS reported receiving
about 114 million call attempts, of which about 23 million were
answered--an accessibility rate of 20 percent.  For the same period
in 1995, IRS reported receiving about 236.0 million call attempts, of
which 19.2 million (8 percent) were answered.  As the data for 1995
and 1996 indicate, a major reason for the improved accessibility in
1996 was the significant drop in call attempts.  IRS attributed that
drop to (1) fewer refund delay notices being issued, as discussed in
more detail later in this report, and (2) IRS' efforts to publicize
other information sources, such as its World Wide Web site on the
Internet. 

While acknowledging that telephone accessibility is low, IRS believes
that accessibility should be measured not only by the percentage of
calls answered but also by the percentage of callers who were able to
get through to IRS, which IRS refers to as the "taxpayer service
level of access." Accordingly, IRS summarized its toll-free telephone
assistance during the 1996 filing season as follows: 

     "For the period January 1, 1996, to April 20, 1996, IRS received
     calls made by 46 million callers.  IRS answered 23 million
     calls, or 50% of the callers.\7 Of the 114 million total call
     attempts received, 23 million or 20% received an answer.  The
     remaining 91 million attempts, often the result of redials,
     received a busy signal or were terminated by the callers because
     they did not want to wait in queue for an assistor.  The total
     number of callers mentioned earlier was determined by
     discounting for redials.  Therefore, the 114 million call
     attempts equates to 46 million callers.  This is an average of
     2.5 attempts per caller."

As IRS' data indicate, the accessibility of IRS' toll-free telephone
assistance during the 1996 filing season, whether measured as a
percentage of calls or callers, was still not good. 


--------------------
\6 Tax Administration:  Increased Fraud and Poor Taxpayer Access to
IRS Cloud the 1993 Filing Season (GAO/GGD-94-65, Dec.  22, 1993); Tax
Administration:  Continuing Problems Affect Otherwise Successful 1994
Filing Season (GAO/GGD-95-5, Oct.  7, 1994); and The 1995 Tax Filing
Season:  IRS Performance Indicators Provide Incomplete Information
About Some Problems (GAO/GGD-96-48; Dec.  29, 1995). 

\7 For the same period in 1995, IRS reported that 41 percent of the
callers were able to get through to an assistor. 


         AVAILABILITY OF WALK-IN
         ASSISTANCE REDUCED
-------------------------------------------------------- Letter :3.1.2

For the 1996 filing season, IRS closed 93 sites that had previously
provided walk-in assistance, reduced the operating hours of some of
the 442 sites that remained open, and eliminated free electronic
filing at many of the sites.  According to IRS, the closed sites were
selected on the basis of their historical volume of work and their
proximity to other walk-in sites.  As an indication of the effect of
these closures and cutbacks, IRS data showed that (1) walk-in sites
served about 2.8 million taxpayers from January 1 to April 20, 1996,
which was about 17-percent fewer taxpayers than were served during
the same period in 1995, and (2) about 59,000 electronic returns were
filed at walk-in sites in 1996, compared with about 104,000 in 1995. 

Concerned about the reduction in walk-in service, the House and
Senate conference agreement on the Treasury, Postal Service, and
General Government appropriation for fiscal year 1997 included a
provision that requires IRS to maintain the fiscal year 1995 level of
service, staffing, and funding for taxpayer services.  While noting
that this provision does not mean that IRS should be required to
rehire staff or reopen offices, the conference report said that "IRS
should be very sensitive to the needs of the taxpayers" who use
walk-in sites during the filing season. 

Walk-in sites provide various free services, including copies of more
commonly used forms and publications, help in preparing returns, and
answers to tax law questions.  We visited four walk-in sites and
asked taxpayers where they would go if the office were closed.  Many
taxpayers commented that they would go to another IRS office or a
professional tax preparer for assistance, and that they would call
the toll-free forms-ordering telephone number for forms or pick them
up at a library or post office. 

As indicated by the persons with whom we spoke, there are other ways
taxpayers can obtain the free services offered by walk-in sites,
although maybe not as easily.  For example, according to IRS, it
generally takes from 7 to 15 workdays to receive materials that are
ordered by telephone--longer if the materials are not in stock. 
Persons with access to a computer can download forms from the
Internet or the FedWorld computer bulletin board.  Free forms are
also available at libraries and post offices and through IRS' "fax on
demand" service. 

Taxpayers who need help in preparing their returns and do not want to
pay for that help may be able to take advantage of the tax
preparation services offered at sites around the country that are
part of the Volunteer Income Tax Assistance (VITA) and Tax Counseling
for the Elderly (TCE) programs.  According to IRS, these programs
help older, disabled, low-income, and non-English-speaking
individuals prepare their basic returns.  IRS data for the 1996
filing season indicate that there was an increased demand for
services at the VITA and TCE sites.  The data showed that although
the number of VITA and TCE sites around the country decreased by 513
compared with the 1995 filing season, about 71,000 additional
taxpayers took advantage of the service. 

Taxpayers who need answers to tax law questions can call IRS'
toll-free tax assistance number or IRS' TeleTax system, which has
prerecorded information on about 150 topics.  From January 1 to April
27, 1996, the number of tax law calls to TeleTax increased by about
11 percent over the same period in 1995 (i.e., 6.9 million in 1996
compared with 6.2 million in 1995). 

Still another option for free assistance is IRS' World Wide Web site
on the Internet.  Among other things, IRS' Web site includes copies
of forms, information similar to that on TeleTax, and some
interactive scenarios that taxpayers can use to help them answer some
commonly asked questions.  IRS reported that, as of May 1, 1996, its
Web site had been accessed more than 52 million times since January
8, 1996, when it first became available. 


      FEWER REFUNDS DELAYED IN
      1996
---------------------------------------------------------- Letter :3.2

In 1995, IRS took several steps in an attempt to better ensure that
persons were entitled to the dependents and EICs they were claiming. 
The most visible of those efforts involved the delay of about 7
million refunds to allow IRS time to verify SSNs, with an emphasis on
returns claiming the EIC.  The delays caused adverse reaction from
taxpayers and tax return preparers during the 1995 filing season. 

Although IRS' efforts in 1995 and the publicity surrounding those
efforts appeared to have had a significant deterrent effect (e.g.,
according to IRS, 1.5 million fewer dependents were claimed in 1995
than were claimed in 1994), the efforts were not without problems. 
For example, although IRS identified about 3.3 million returns with
missing or invalid SSNs and delayed any related refunds, it was able
to pursue only about 1 million of those returns.  For those cases it
was unable to pursue, IRS eventually released any refunds, after
holding them for several weeks, without resolving the problems. 
Also, IRS delayed about 4 million EIC-related refunds for taxpayers
whose returns had valid SSNs to check for fraudulent use of the same
SSN on more than one return.  IRS eventually released almost all of
those refunds, after several weeks, without doing the checks. 

For the 1996 filing season, IRS was more selective in deciding which
cases to review and which refunds to delay.  IRS tried to limit the
number of delayed refunds to the volume of cases it could review and
to focus its resources on the most egregious cases.  The most
significant change for the 1996 filing season was that IRS did not
delay EIC refunds on returns with valid SSNs.  IRS statistics on the
number of refund delay notices sent to taxpayers in 1996, concerning
dependent and EIC claims, indicated that IRS delayed far fewer
refunds in 1996.  As of September 6, 1996, IRS had mailed about
350,000 such notices compared with about 7 million in 1995.  Another
indicator that fewer refunds were delayed in 1996 is the decrease in
the number of "where is my refund" calls to IRS.  Taxpayers wanting
to know the status of their refunds can call TeleTax and get
information through the use of an interactive telephone menu.  During
the 1996 filing season, as of June 8, 1996, IRS reported receiving
48.2 million such calls, which was a decrease of about 15 percent
from the 56.6 million it reported receiving for the same period in
1995. 

In contrast to the negative reaction from taxpayers and practitioners
during the 1995 filing season, an executive of the largest tax
preparation firm told us that IRS generally did a better job in 1996. 
The executive said that the firm's clients received refunds quicker
and received fewer notices about problems, such as SSN mismatches. 
Likewise, in March 28, 1996, testimony before the Oversight
Subcommittee, a representative of the National Association of
Enrolled Agents said the following:  "Our members report they have
encountered far fewer problems this year compared to last year in the
area of refund processing .  .  .  ."


      IRS REVIEWED FEWER
      QUESTIONABLE SSNS AND
      IDENTIFIED FEWER FRAUDULENT
      REFUND CLAIMS IN 1996
---------------------------------------------------------- Letter :3.3

As part of IRS' increased emphasis on verifying SSNs in 1995, the
Examination function followed up on about 1 million returns that IRS'
computer, using certain criteria, had identified as having
questionable SSNs.\8 As of June 30, 1996, about 986,000 of those
cases had been closed--about 500,000 (51 percent) with no change in
tax liability and about 486,000 (49 percent) with changes totaling
about $808 million. 

In 1996, IRS (1) revised the criteria used to select cases in an
attempt to better focus its efforts and (2) identified about 700,000
returns for follow-up, which is about 300,000 fewer than in 1995. 
Because it takes time for IRS to complete its reviews, information on
results was not available at the time we completed our audit work. 
Thus, we do not know the impact of IRS' reduced level of effort in
1996.  However, a decrease in the number of cases reviewed does not
necessarily mean that IRS identified less noncompliance in 1996 than
in 1995 because only about one-half of the cases reviewed in 1995
were productive.  It is possible that IRS' revised criteria, despite
generating fewer cases, might have identified more productive cases
in 1996. 

The SSN verification/refund delay efforts previously discussed were
generally directed at identifying and correcting erroneous refunds
caused by honest mistakes or negligence.  Since the 1970s, IRS has
had a Questionable Refund Program (QRP) directed at identifying
fraudulent refund schemes. 

QRP results for January 1996 through September 1996 showed that IRS
had identified 20,521 fraudulent returns (involving claimed refunds
of about $55.4 million) during those 9 months.  These results are a
significant decline from the 59,241 returns and about $124.8 million
in refunds reported for the first 9 months of 1995.  QRP officials
attributed the decline to three things.  First, and most significant
in their opinion, was a staffing reduction that was part of IRS'
cost-cutting efforts in anticipation of reduced funding levels. 
According to the officials, the 10 IRS service centers were allocated
a total of about 379 full-time equivalent staff for the QRP in fiscal
year 1996 compared with 553 full-time equivalent staff in 1995, which
was a decrease of 31 percent.  The other two reasons cited by the QRP
officials were (1) the impact of enhanced upfront filters in the
electronic filing system that prevented bad returns from getting into
the system and (2) a decision to focus QRP efforts on certain kinds
of cases. 


--------------------
\8 The computers actually identified about 3.3 million problem cases
but, as previously noted, IRS was unable to pursue many of them. 


   IRS' EFFORTS TO MODERNIZE ITS
   PROCESSING ACTIVITIES HAD MIXED
   RESULTS IN 1996
------------------------------------------------------------ Letter :4

Although IRS was able to meet its processing goals (such as cycle
time,\9 processing accuracy, and refund timeliness) in 1996, those
goals were based on expectations as to what IRS could achieve with
the systems and procedures currently in place.  In that regard, there
is general agreement that much can be done to improve those systems
and procedures.  IRS has initiated several efforts toward that end,
including (1) providing alternatives to the filing of paper returns,
(2) using scanning and imaging technology to eliminate the manual
data transcription of paper returns, and (3) using lockboxes and
direct deposits to expedite the processing of tax payments and
refunds, respectively. 

Despite IRS' generally successful performance during the 1996 filing
season, there are still several concerns centering around IRS'
modernization efforts.  For example, although more returns were filed
using alternatives to the traditional paper form, the number of
returns filed through one of those alternatives (electronic filing)
fell short of IRS' projections.  Also, although a document scanning
and imaging system that was intended to streamline parts of IRS'
paper-processing operations performed better in 1996, the system
still is not meeting IRS' performance expectations and may eventually
cost much more than originally estimated.  Although data on the
results of IRS' use of lockboxes to process Form 1040 tax payments
indicate that the government is saving money, those savings are being
diminished significantly by the extra cost associated with having
taxpayers send not only their payments but also their returns to the
lockbox banks.  Finally, expansion of the direct-deposit option for
refunds to taxpayers who filed a paper return was not as widely
received by taxpayers as IRS had anticipated. 


--------------------
\9 Cycle time is the average number of days it takes service centers
to process returns. 


      NUMBER OF RETURNS FILED
      THROUGH ALTERNATIVE METHODS
      INCREASED BUT ELECTRONIC
      FILING FALLS SHORT OF
      PROJECTIONS
---------------------------------------------------------- Letter :4.1

As of October 18, 1996, IRS had received about 118.1 million
individual income tax returns, which was about 1.5 percent more than
the 116.4 million returns received as of the same period in 1995. 
While the increase in the overall number of returns filed was small,
the increase in the number filed through alternative methods was
substantially higher than in 1995 (about 50 percent).  IRS offers
three alternatives to the traditional filing of paper returns (i.e.,
electronic filing, TeleFile, and Form 1040PC).\10 As shown in table
1, most of the growth in alternative filings was due to TeleFile and
Form 1040PC.  Table 1 also shows that, of the three alternatives,
only electronic filing failed to meet IRS' projections. 



                                Table 1
                
                Number of Individual Income Tax Returns
                  Received Through Alternative Filing
                                Methods

                         Number of
                           returns
                               IRS
                         projected   Number of   Number of
                          for 1996     returns     returns  Percentage
                            filing    filed in    filed in      change
Filing method               season      1996\a      1995\a   from 1995
----------------------  ----------  ----------  ----------  ----------
Electronic              13,648,400  12,139,000  11,144,000        +8.9
TeleFile                 2,497,400   2,840,000     680,000      +317.6
Form 1040PC              4,617,600   7,031,000   2,912,000      +141.4
======================================================================
Total                   20,763,400  22,010,000  14,736,000       +49.4
----------------------------------------------------------------------
\a Data are as of October 18, 1996, and October 20, 1995. 

Source:  IRS data. 


--------------------
\10 Under the electronic filing method, return data in electronic
format are transmitted over communication lines through a third party
(such as a tax return preparer or an electronic return transmitter)
to an IRS service center, where the data are automatically edited and
processed.  Under TeleFile, certain taxpayers who are eligible to
file a Form 1040EZ are allowed to file using a toll-free number on
Touch-Tone telephones.  Under the Form 1040PC method, a taxpayer or
tax return preparer uses personal computer software that produces a
paper tax return in an answer-sheet format.  The Form 1040PC shows
the tax return line number and the data (dollar amount, name, etc.)
on that line.  Only lines on which the taxpayer has made an entry are
included on the Form 1040PC. 


         ELECTRONIC FILING
-------------------------------------------------------- Letter :4.1.1

Electronic filing has several benefits.  It enables taxpayers to
receive their refunds sooner than if they had filed on paper and
gives them greater assurance that IRS has received their returns and
that the returns are mathematically accurate.  The benefit for IRS is
that electronic filing reduces processing costs and facilitates more
accurate processing.  IRS began offering electronic filing in 1986. 
Since that time, 1995 was the first year that the number of
individual income tax returns received electronically decreased from
the number received the prior year.  IRS attributed that decline to
the secondary effects of measures it implemented to combat filing
fraud. 

IRS took several steps in an attempt to increase the use of
electronic filing in 1996.  For example, IRS (1) put increased
emphasis on the availability of On-Line Filing, a program that allows
taxpayers to file their returns, through a third party, via a
personal computer-modem link, and (2) extended the period during
which returns could be filed electronically by moving the closing
date from August 15 (the filing deadline for taxpayers who get one
extension to file) to October 15 (the filing deadline for taxpayers
who get a second extension).\11

Taxpayers' use of electronic filing recovered somewhat in
1996--increasing to about 12.1 million individual income tax returns
as of October 18 (about a 9-percent increase).  According to IRS, a
major contributor to this increase was growth in the Federal/State
electronic filing program.  Under that program, taxpayers can file
both their federal and state income tax returns through one
submission to IRS.  A taxpayer's federal and state data are combined
into one electronic record that is transmitted to IRS, which, in
turn, makes the state portion of the data available to the state. 
IRS reported that about 3.2 million returns were filed under the
Federal/State program in 1996 compared with about 1.6 million in
1995.  Some of the increase in electronic filing in 1996 was also due
to the steps discussed in the preceding paragraph.  According to IRS
data, 158,284 taxpayers had used the On-Line Filing option as of
October 18, and about 22,000 taxpayers had filed electronically
between August 9 and October 18, 1996.  Despite the increase in 1996,
electronic filings that year were still below the 13.5 million
individual returns filed electronically in 1994 and below IRS'
projection of about 13.6 million returns in 1996. 

A major impediment to the growth of electronic filing is that the
method is not completely paperless.  Taxpayers must send IRS their
W-2s and a signature document (Form 8453) after their return has been
electronically transmitted.  IRS must then manually input these data
and match them to the electronic return. 

In an attempt to eliminate the paper associated with electronic
returns, IRS tested the use of digitized signatures during the 1996
filing season.  The goal of that test was to gauge the willingness of
taxpayers and preparers to use an electronic signature pad in place
of signing a Form 8453.\12 The electronic signature was attached to
the electronic return and both were transmitted to IRS.  The test was
conducted at three locations (two VITA sites located on military
bases and a private, tax return preparation office).  According to
IRS officials, about 50 percent of the taxpayers who were offered the
chance to participate in the test agreed to do so.  Given the level
of participation in 1996 and positive preparer feedback, IRS plans to
expand the test in 1997, but details of that expansion will not be
finalized until just before the filing season begins. 

Besides eliminating the paper associated with electronic returns,
there are other steps IRS could take to increase the use of
electronic filing.  In October 1995, we reported that without some
dramatic changes in IRS' electronic filing program, many of the
benefits available from electronic filing could go unrealized.\13 We
recommended that IRS (1) identify those groups of taxpayers that
offer the greatest opportunity to reduce IRS' paper-processing
workload and operating costs if they filed electronically and (2)
develop strategies that focus on eliminating or alleviating
impediments that inhibit those groups from participating in the
program.  As of October 9, 1996, IRS was finalizing a new electronic
filing strategy. 


--------------------
\11 IRS also eliminated free electronic filing at many of its walk-in
sites.  However, because only a small number of returns had been
filed at walk-in sites in the past, the impact of that action on
electronic filing was minimal. 

\12 Test participants were also granted a waiver from the requirement
to submit W-2s.  According to an IRS official, tax return preparers
were instructed to review the paper W-2s and not prepare electronic
returns for taxpayers whose W-2s looked fraudulent. 

\13 Tax Administration:  Electronic Filing Falling Short of
Expectations (GAO/GGD-96-12, Oct.  31, 1995). 


         TELEFILE
-------------------------------------------------------- Letter :4.1.2

TeleFile generally provides the same benefits to taxpayers and IRS as
electronic filing.  However, TeleFile is more convenient and less
costly than electronic filing because the latter requires that
taxpayers go through a third party.  The increase in taxpayer use of
TeleFile in 1996 was due primarily to the program's expansion
nationwide.  As shown in table 1, IRS received about 2.8 million
TeleFile returns in 1996, when TeleFile was available to taxpayers in
50 states, compared with 680,000 in 1995, when Telefile was available
in only 10 states.  Although most of the increase was due to the
program's nationwide expansion in 1996, TeleFile use also showed a
significant rate of increase in the 10 states that were in the
program in 1995 (from 680,000 returns in 1995 to 804,732 in 1996--an
18-percent increase).  A major change that might have contributed to
the increase in TeleFile use was IRS' decision to make TeleFile
paperless in 1996.  Unlike past years, taxpayers did not have to mail
their W-2s or a signature document to IRS.  Instead of the signature
document, taxpayers used a personal identification number that was
provided by IRS.\14

IRS' Internal Audit Division reviewed the 1996 TeleFile Program and
concluded that management had "effectively prepared for and
successfully implemented" the nationwide expansion of TeleFile.\15
For example, Internal Audit noted that (1) its sample of returns
filed through TeleFile showed that all tax calculations were
correctly computed and that data had been posted accurately to IRS'
master file of taxpayer accounts and (2) taxpayer demand for TeleFile
during the 1996 filing season was generally met.  However, Internal
Audit also noted that IRS had not completed a system security
certification and accreditation and thus had no assurance that
taxpayer data were adequately secured.  According to Internal Audit,
certification is a comprehensive evaluation of a system's security
features; accreditation is a declaration that the system is approved
to operate.  As of November 21, 1996, according to the TeleFile
Project Manager, IRS was working to complete the certification and
accreditation. 

Internal Audit's evaluation and various statistics compiled by IRS,
including the results of an IRS survey of TeleFile users, indicate
that TeleFile worked very well in 1996.  For example, about 92
percent of the users surveyed by IRS said that they were very
satisfied with TeleFile.  However, it is important to note that only
about 10 to 14 percent of the more than 20 million 1040EZ filers who
IRS estimated would be eligible to use the system in 1996 actually
used it.\16 IRS did not survey the nonusers because, according to IRS
officials, past surveys showed that the most important reason
eligible users cited for not using TeleFile was their preference for
a paper version.  However, those past surveys did not probe into why
nonusers preferred paper. 

According to the TeleFile Project Manager, IRS plans several changes
to TeleFile for the 1997 filing season, which he estimates will
increase the participation rate to about 25 percent.  For example, he
said that eligibility to use TeleFile will be extended to married
persons filing jointly and TeleFile users will be able to take
advantage of the direct-deposit option that was available to other
taxpayers in 1996 (this option is discussed later in this report). 

The most significant change for 1997, in terms of its potential
impact on taxpayer participation, is IRS' decision to revise the tax
package sent to persons eligible to use TeleFile.  Instead of sending
eligible users a package that also contains a Form 1040EZ and related
instructions, in case they choose not to use TeleFile, IRS has
decided to send them a much smaller package that contains only the
TeleFile worksheet and instructions.  Although this action may
encourage more persons to use TeleFile and reduce IRS' overall
printing and mailing costs, it could be seen as imposing a burden on
persons who, for whatever reason, prefer not to use TeleFile and
would, in that case, need a Form 1040EZ.  It is unclear how taxpayers
will react to this change.  On the one hand, IRS summaries of three
1040EZ/TeleFile focus groups held in August and September 1996
indicated that focus group participants did not view the noninclusion
of Form 1040EZ as a burden because they could easily get a copy, if
needed, from their local library or post office.  On the other hand,
a mail survey that IRS sent to a random number of TeleFile users in
1996 showed that about 28 percent of the respondents thought it was
very important that the 1040EZ information be included in the
TeleFile package. 


--------------------
\14 Instead of mailing their W-2s to IRS, taxpayers were required to
enter certain data from the W-2s (such as the employer identification
number and the amount of taxable wages) into the TeleFile system. 

\15 Review of the 1996 TeleFile Program, IRS Internal Audit,
Reference No.  064401, June 14, 1996. 

\16 There was conflicting IRS information on the number of eligible
filers, which we were unable to resolve before finalizing the report. 
However, we are confident that the percentage of use falls somewhere
within the range cited. 


         FORM 1040PC
-------------------------------------------------------- Letter :4.1.3

The increase in the use of Form 1040PC during the 1996 filing season
resulted, in part, from the largest user's (a tax return preparation
firm) rejoining the program after dropping out in 1995.  For the 1995
filing season, IRS initially required that preparers provide
taxpayers with a specifically formatted legend explaining the Form
1040PC.\17 However, after the 1995 filing season began, IRS decided
not to require the specifically formatted legend but to allow
preparers to provide any type of descriptive printout that explained
each line on the taxpayer's Form 1040PC.  According to an executive
of the previously mentioned tax return preparation firm, (1) the firm
chose not to participate in the program in 1995 rather than comply
with the requirement for a specifically formatted legend and (2) IRS'
decision to change its requirement came too late for the firm to
change its plans.  The firm then rejoined the program for the 1996
filing season. 

The Form 1040PC was developed to reduce the number of pages that a
standard Form 1040 requires, which is a benefit to taxpayers and IRS,
and to streamline paper processing.  Although use of the Form 1040PC
reduces the amount of paper, IRS has not yet realized the full
processing efficiencies available from that form.  Because of
problems encountered with IRS' new document scanning and imaging
system, as discussed in the next section of this report, IRS
terminated plans to have Forms 1040PC scanned and, instead, is
manually keying data from the forms into its computers. 


--------------------
\17 The legend's purpose was to provide better supporting
documentation than was previously available to the taxpayers, and it
was to be used as an aid in doing things such as preparing state
returns and completing financial aid forms. 


      PROBLEMS HAVE HAMPERED
      EFFORTS TO MODERNIZE PAPER
      PROCESSING SYSTEMS
---------------------------------------------------------- Letter :4.2

The Distributed Input System (DIS), which is IRS' primary data entry
system for paper tax returns and other paper documents submitted by
taxpayers, has been in operation since 1984.  Although DIS generally
performed without major problems during the 1996 filing season, its
age is a source of concern within IRS.  IRS had planned to replace
DIS with two document scanning and imaging systems.  The first
replacement system, the Service Center Recognition/Image Processing
System (SCRIPS), was implemented nationwide in 1995 and is not yet
performing to IRS' expectations at that time.  On October 8, 1996,
IRS announced that the second planned system, the Document Processing
System (DPS), was being terminated. 


         SCRIPS IS NOT PERFORMING
         TO IRS EXPECTATIONS
-------------------------------------------------------- Letter :4.2.1

IRS experienced significant performance problems with SCRIPS in 1995,
which was the system's first year of nationwide operation.  Two major
problems were significant system downtime and slow processing rates. 

IRS made some hardware and software modifications that helped improve
the performance of SCRIPS during the 1996 filing season.  IRS
officials in all five SCRIPS service centers told us that SCRIPS
performed significantly better during the 1996 filing season than it
did in 1995.  Specifically, IRS data for April through June of 1995
and 1996 (the first 3 months for which IRS had comparable data)
indicate that system downtime decreased from 791 hours in 1995 to 43
hours in 1996. 

Despite the improved performance in 1996, SCRIPS (1) is still not
processing all of the forms that it was expected to process and (2)
may cost more than originally estimated.  In an October 1994 business
case for SCRIPS, IRS said that, by 1996, the system would be
processing all Federal Tax Deposit coupons and information returns,
all Forms 1040EZ, 50 percent of the Forms 1040PC, and 93 percent of
the Forms 941 (Employers Quarterly Federal Tax Return).  In fiscal
year 1996, SCRIPS processed all Federal Tax Deposit coupons and
information returns, as expected.  However, SCRIPS only processed
about 50 percent of the Forms 1040EZ and did not process any Forms
1040PC or Forms 941.  In addition, the cost estimate for SCRIPS has
increased from $133 million in October 1992 to a current estimate of
$288 million.  Part of the increase is due to the inclusion of
certain costs, such as for maintenance, that were not part of the
original estimate. 

We will be issuing a separate report that has more information on
SCRIPS' problems in 1995, its performance in 1996, and IRS' plans for
the system in the future. 


         DPS TERMINATED
-------------------------------------------------------- Letter :4.2.2

A second scanning system, DPS, was to replace SCRIPS and expand IRS'
imaging capability to more complex tax forms.  IRS expected DPS to
begin handling some of the DIS workload by the start of the 1998
filing season.  However, due to concerns about the future of DPS, IRS
reassessed its strategy for processing paper tax returns.  According
to IRS, part of the reassessment involved options, such as
outsourcing the processing of some returns and/or acquiring a new
manual data entry system to replace DIS.  As of September 26, 1996,
according to a cognizant IRS official, the reassessment was done but
a final decision had not yet been reached.  That reassessment took on
added importance when IRS announced, on October 8, 1996, that DPS was
being terminated.  IRS attributed that decision, at least in part, to
budgetary concerns and "the need to prioritize investments in systems
that have a direct and immediate benefit on improved customer
service, such as better telephone access."

The uncertainty of IRS' plans for processing paper returns means that
IRS may have to continue to rely on DIS longer than it had originally
expected.  In a February 1996 report, Internal Audit said that DIS
could be required to process forms until 2003.\18 Over the course of
the 1996 filing season, various service center officials expressed
concern about IRS' ability to adequately maintain and repair the
system.  Despite their concerns, DIS performed satisfactorily during
the filing season.  Officials also told us that, until this year, IRS
had not kept detailed maintenance records to capture DIS downtime. 
Thus, an accurate comparison of DIS downtime and system reliability
over the years is not possible.  We recently began a review of IRS'
ability to maintain current operating levels with its existing
systems. 


--------------------
\18 Review of the Distributed Input System Transition/Contingency
Plan, IRS Internal Audit, Reference No.  062208, February 23, 1996. 


      USE OF LOCKBOXES TO PROCESS
      FORM 1040 TAX PAYMENTS
      RAISES BURDEN, COST, AND
      SECURITY ISSUES
---------------------------------------------------------- Letter :4.3

IRS envisions that by 2001, most tax payments will be processed by
lockbox banks rather than by IRS service centers.  The banks process
the payments and transfer the funds to a federal government account. 
The payment and payer information are then recorded on a computer
tape and forwarded to IRS for use in updating taxpayer accounts.  One
reason for using lockboxes is the expectation that tax payments will
be deposited faster into the Treasury.  Faster deposits mean that the
government has to borrow less money to fund its activities and less
borrowing means lower interest costs (otherwise known as "interest
cost avoidance"). 


         IRS' TESTING OF LOCKBOXES
         TO PROCESS FORM 1040
         PAYMENTS
-------------------------------------------------------- Letter :4.3.1

Since 1989, IRS has used lockboxes to process payments sent in with
estimated tax returns (Forms 1040ES).  For the last several years,
IRS has been testing the use of lockboxes to process payments sent in
by individuals when they file their income tax returns (Forms 1040). 
For the 1996 test, IRS sent special Form 1040 packages to specific
taxpayers.  These packages included (1) mailing instructions and (2)
a payment voucher that could be scanned by optical character
recognition equipment. 

The test packages contained one return envelope with two different
tear-off address labels.  One label, which was addressed to a
lockbox, was to be used for a return with an accompanying tax
payment, and the other label, which was addressed to a service
center, was to be used for a return with no payment.  Taxpayers with
payments were instructed to put their returns, payments, and vouchers
in the envelope in their tax packages and to affix the label
addressed to the lockbox.  The bank that serviced the lockbox was to
separate the returns from the payments, deposit the payments, record
the payment information on a computer tape, sort the returns,\19 and
forward the returns and the computer tape to IRS for processing. 

IRS had tested another mailing method during the 1994 and 1995 filing
seasons.  This test involved the use of two envelopes.  One envelope
was addressed to a service center, and the other envelope was
addressed to a lockbox.  Taxpayers were instructed to put their tax
returns in the envelope addressed to the service center and to put
any payments and vouchers in the envelope addressed to the lockbox. 
The bank was to process the payments and vouchers as previously
described. 


--------------------
\19 In some cases, the payment sent in by a taxpayer does not fully
satisfy the tax liability reported on the return.  Because IRS gives
priority processing to those returns, so that it can speed up
issuance of a balance due notice to the taxpayer, the banks are to
segregate those returns from returns that do not have a balance due. 


         LOCKBOX BURDEN, COST, AND
         SECURITY ISSUES
-------------------------------------------------------- Letter :4.3.2

IRS has decided, for the 1997 filing season, to continue testing the
two-label method in certain tax packages.  According to an IRS
official responsible for the lockbox program, IRS will no longer use
the two-envelope approach due to the increased taxpayer burden IRS
anticipates the approach would cause.  She explained that IRS has
found, in its studies of taxpayer behavior, that, among other things,
taxpayers who participated in the test preferred to keep their
remittances and returns together.  Because of this, IRS believes that
asking taxpayers to split their tax payments from their returns is
burdensome. 

The studies referred to by IRS, all of which were done by a
contractor in 1993 and 1994, included mail and telephone surveys of
about 1,900 taxpayers, interviews with 46 individuals, and 5 taxpayer
focus groups.  We reviewed the contractor's reports and considered
the results to be inconclusive as they related to burden.  For
example, of the people surveyed by mail and telephone who said they
remembered what they did in the test, 45.9 percent said that they
felt uneasy about mailing their checks and returns in separate
envelopes while 41.2 percent said that they did not feel uneasy (the
other 12.9 percent did not know).  The results of the 46 interviews
showed a similar lack of consensus, in our opinion.  Several people
said that they preferred using one envelope because it was easier or
because they were worried about the payments and the tax returns not
getting linked if they were sent to two different places.\20 But,
several other people said that they preferred using two envelopes
because they were concerned about the confidentiality of their tax
returns or the increased risk of their returns getting lost.  Even
some of those who preferred one envelope expressed concern about the
banks' involvement in handling their returns. 

Burden is one issue to consider in deciding on the use of lockboxes;
cost is another.  Information we received from IRS and FMS indicates
that having taxpayers send their returns to the lockboxes along with
their payments has substantially increased the cost of the lockbox
service to the government.  During the first 8 months of the 1996
filing season, according to IRS, the lockbox banks had processed
about 7 million Form 1040 payments.  According to FMS, the government
paid the banks an average of $2.03 per payment in 1996--98 cents to
process each payment, 92 cents to sort each accompanying tax return,
and 13 cents to ship each return to a service center--and the same
fees will be in effect until April 1, 1997.\21 Fees after that date
are subject to negotiation between FMS and the banks.  Cognizant FMS
staff said that the banks have been charging such a high fee for
sorting returns to encourage IRS to stop having the returns sent to
the banks. 

Service centers process returns received from a lockbox bank in the
same manner as they process returns that come directly from
taxpayers, with one exception--the returns coming from the bank do
not have to be sorted by IRS.  According to IRS data, not having to
sort the returns saves IRS about 37 cents a return\22 --much less
than the 92 cents per return being charged by the banks.  Thus,
assuming a volume of 7 million returns, the government paid about
$6.4 million for a service (return sorting) that it could have done
itself for about $2.6 million, or about $3.8 million less.  Shipping
those returns cost the government another $910,000. 

According to FMS, the use of lockboxes to process Form 1040 tax
payments enabled the government to avoid interest costs of $15.7
million in fiscal year 1996.\23 This interest cost avoidance compares
with $1.6 million in fiscal year 1995.\24 Because these savings
result from faster processing of tax payments, having the banks sort
and ship the tax returns does not add to the savings and could, by
increasing the banks' workload, cause processing delays that would
reduce any savings. 

In an August 30, 1996, letter to Treasury's Fiscal Assistant
Secretary, IRS' Deputy Commissioner acknowledged the high costs
associated with having returns sent to lockboxes.  In a September 11,
1996, reply, the Assistant Secretary also expressed some concern
about the costs associated with the processing of Form 1040 tax
payments through lockboxes.  The Assistant Secretary said that "[t]he
most appealing option from a cost standpoint is the two-envelope
concept.  This option .  .  .  makes good business sense as tax
payments and tax returns are sent to the appropriate place best
prepared to handle them."

As a way to lower costs, the Assistant Secretary suggested that IRS
explore the possibility of not having the banks sort the returns and
have the sorting done by the service centers.  We discussed this
option with officials in IRS' National Office and at one service
center.  We were told that it would be difficult for service centers
to sort the returns once they had been separated from the payments
because the service center would not know if the taxpayer had fully
paid his or her tax liability.\25 According to the IRS officials,
that distinction is important because, as previously discussed,
returns involving less than full payment are given priority
processing to enable more timely issuance of the balance-due notice
to the taxpayer.  IRS had considered adding a checkbox on the return
for the taxpayer to indicate whether full payment was enclosed with
the return.  According to IRS, asking taxpayers to check such a box
would be another form of burden--although not a significant one. 

Security is a third issue that needs to be considered in deciding how
to use lockboxes.  As previously noted, several individuals who
participated in the focus groups and interviews about IRS' use of
lockboxes expressed concern that their returns would be lost or their
return data would be misused.  We did not do a thorough analysis of
security at the lockbox banks.  However, we reviewed security and
processing procedures at 2 of the 10 lockbox banks and found that
controls exist to minimize the risk of lost or misused tax data. 

IRS' lockbox procedures require that the tax returns be separated
from the payment as soon as the envelope is opened.  Only personnel
who open the envelopes and their supervisors are to have access to
the returns.  Security cameras are to monitor all of the lockbox
processing.  The returns are to be bundled and packed into boxes as
soon as they are separated from the payment.  Each day, the boxes of
returns are to be shipped by bonded courier to the service center. 
Background checks, such as a criminal record check, are to be done on
lockbox personnel hired by the bank.  These are the same checks that
are to be done on IRS service center personnel with the same duties. 
Bank personnel, like service center employees, are to sign statements
of understanding about the confidentiality of the information they
will process and the penalties for disclosing any of this
information.  IRS and FMS lockbox coordinators are to visit the banks
to ensure compliance with these procedures and are to submit
quarterly reports on the basis of those visits.  An FMS staff person
who was responsible for IRS' lockbox processing program told us there
have been no known incidents of disclosure of taxpayer information
from a lockbox bank. 

During our visits to the two banks, we observed the
security-surveillance cameras in operation and verified that badges
were being worn by all personnel and that access to the processing
area was controlled by a guard.  We also reviewed judgmental samples
of personnel files and, for each employee whose file we reviewed, we
(1) found that disclosure statements were maintained and (2) saw
evidence that background checks had been done. 


--------------------
\20 It is important to note that, if linkage were a concern, the use
of one envelope is not the answer.  Even though both the return and
payment originally go to the same location (the bank), they are
processed at two separate locations (the payment at the bank and the
return at the service center), just as they would be under the
two-envelope method. 

\21 FMS, not IRS, has been paying these fees and will continue to do
so in 1997. 

\22 According to IRS' data, which we did not verify, the 37 cents
includes direct labor, benefits, quality assurance, and overhead. 

\23 FMS, in computing interest cost avoidance, assumed that the
lockbox banks were able to deposit tax remittances 3 days faster, on
average, than the service centers.  Although we did not do sufficient
work to validate that assumption, information we obtained during this
review, along with related information obtained during reviews of
past filing seasons, indicates that the assumption is reasonable. 

\24 In our report on the 1995 filing season (GAO/GGD-96-48), we
referred to an interest cost avoidance of $44.3 million in 1995. 
That figure related to all tax remittances being processed through
lockboxes, including estimated tax payments.  The $1.6 million figure
relates just to remittances associated with Form 1040. 

\25 This would also be a problem under the two-envelope method
because the return and payment would be separated. 


      USE OF DIRECT-DEPOSIT OPTION
      WAS NOT AS WIDESPREAD AS IRS
      HAD EXPECTED
---------------------------------------------------------- Letter :4.4

Unlike past years, IRS allowed taxpayers who filed paper returns in
1996 to request that their refunds be deposited directly to their
bank account through an electronic fund transfer.  IRS included a
Form 8888 (Request for Direct Deposit of Refund) in almost all paper
tax packages.  IRS estimated that about 5 million taxpayers who filed
paper returns would request the direct-deposit option and, on
average, that the option would enable paper filers to get their
refunds 10 days faster than if they had waited for a paper check. 
IRS also estimated that it would cost about 25-percent less to
process a Form 8888 than it costs to mail a paper refund check (20
cents per form v.  27 cents per paper check). 

Only about 1.6 million taxpayers took advantage of the direct deposit
option.  An IRS official said that IRS will retain its goal of about
5 million direct-deposit refunds for the 1997 filing season.  IRS has
taken a couple of steps to enhance its chances of achieving that
goal.  Most significantly, it has eliminated the Form 8888.  Instead
of having a separate form, most of the individual income tax forms
will be revised to provide space for the taxpayer to request a direct
deposit and to provide the necessary bank account information.  Also,
as previously noted, TeleFile users will be able to request a direct
deposit in 1997. 


   CONCLUSIONS
------------------------------------------------------------ Letter :5

Was the 1996 filing season a success?  The answer depends on one's
perspective.  From IRS' standpoint, it was a success.  IRS met or
exceeded all but one of its performance goals and was very close to
meeting the other.  IRS was able to process individual income tax
returns and refunds without any apparent problem, with its aging
computer systems having made it through another filing season.  From
the taxpayer's perspective, the filing season was also successful in
many key respects.  For example, relatively few refunds were delayed
in 1996, unlike 1995 when millions of taxpayers were angered by IRS'
decision to delay their refunds while it checked dependent and EIC
claims; more taxpayers were given the opportunity to file by
telephone and to have their refunds directly deposited into their
bank accounts; and IRS' World Wide Web site on the Internet provided
a convenient source of information for taxpayers with access to a
computer.  However, there were some problems in 1996.  Although the
accessibility of IRS' toll-free telephone assistance improved,
taxpayers continued to have problems reaching IRS by telephone, and
some taxpayers may have been inconvenienced by the reduction in IRS'
walk-in services. 

IRS has several efforts under way to modernize the systems and
procedures it has used for many years to process returns,
remittances, and refunds.  These efforts are essential if IRS is to
successfully meet the demands of future filing seasons.  To date, the
results of those efforts have been mixed.  IRS has taken steps to
enhance its efforts.  For example, IRS is (1) expanding eligibility
for TeleFile and taking other steps in an effort to increase the use
of that filing alternative, (2) working to make electronic filing
paperless by broadening its test of digitized signatures, (3) making
it easier for taxpayers to request direct deposits of their refunds,
and (4) reassessing its strategy for processing paper tax returns. 

Even if IRS is successful in increasing the TeleFile participation
rate to 25 percent in 1997, that would still leave a large number of
eligible users who choose not to use TeleFile.  We believe that IRS'
efforts to expand the use of TeleFile could be enhanced if it had
more specific information on why eligible users prefer to file on
paper.  More specifics might help IRS identify barriers to TeleFile
use and develop mitigating strategies. 

We also question whether IRS' decision to have taxpayers send both
their tax returns and their tax payments to lockboxes and to have
banks sort those returns adequately considered both the costs to the
government and taxpayer burden.  Although it is important to minimize
taxpayer burden, the evidence we were given was not convincing
concerning the amount of burden associated with using two envelopes,
especially in light of the extra cost to the government associated
with using one envelope (about $4.7 million during the first 8 months
of the 1996 filing season).  It is understandable that persons
contacted by IRS' contractor, when asked to choose between one or two
envelopes, would pick one, because it is easier to put everything
into one envelope than to segregate things into two envelopes and pay
additional postage.  But, it is not clear that those persons
considered the use of two envelopes an unreasonable burden.  Nor is
it clear how those persons might have responded if they were told
that the use of one envelope causes the government to spend several
million dollars more than it would if taxpayers used two envelopes. 

The cost associated with using lockboxes to process Form 1040 tax
payments might become less of an issue if the government is able to
negotiate bank fees for sorting that are more comparable to the
service center costs for that activity.  Absent lower fees, an
alternative is to continue to have returns sent to the bank but to
have the banks ship the returns to the service centers unsorted. 
That would require IRS to add a checkbox to the return (which would
also be required if IRS decided to use two envelopes) but checking a
box would likely be perceived by taxpayers as less of a burden than
using two envelopes.  However, while a reduction in bank fees or a
decision to accept returns from the banks unsorted would make the
one-envelope method more advantageous, they would not relieve the
anxiety expressed by some taxpayers about their returns being lost or
misused by bank personnel. 


   RECOMMENDATIONS TO THE
   COMMISSIONER OF INTERNAL
   REVENUE
------------------------------------------------------------ Letter :6

If most eligible TeleFile users do not use the system during the 1997
filing season, as IRS is anticipating, we recommend that the
Commissioner of Internal Revenue conduct a survey to determine why,
including more specific information on why the nonusers prefer to
file on paper, and take steps to address any identified barriers to
increased user participation. 

If the government is unable to negotiate lockbox fees that are more
comparable to service center costs and in the absence of more
compelling data on taxpayer burden, we recommend that the
Commissioner, for filing seasons after 1997,\26 either discontinue
having returns sorted by the banks or reconsider the decision to have
taxpayers send their tax returns to the banks along with their tax
payments. 

We are not making any recommendations in this report to address
problems with telephone accessibility and electronic filing because
we have recently issued separate reports on these topics.\27 We will
also be issuing a separate report on SCRIPS. 


--------------------
\26 Our recommendation is directed at filing seasons after 1997
because it is too late for IRS to reconsider its lockbox plans for
the 1997 filing season. 

\27 Telephone Assistance:  Adopting Practices Used by Others Would
Help IRS Serve More Taxpayers (GAO/GGD-95-86, Apr.  12, 1995) and
GAO/GGD-96-12. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :7

We requested comments on a draft of this report from the Commissioner
of Internal Revenue or her designated representative.  Responsible
IRS officials, including the Assistant Commissioner for Forms and
Submission Processing, the National Director for Submission
Processing, and the National Director for Customer Service (Planning
and Systems), provided IRS' comments in a November 6, 1996, meeting. 
Those comments were reiterated in a November 18, 1996, memorandum
from the Acting Chief of Taxpayer Service.  IRS officials also
provided some factual clarifications that we incorporated in the
report where appropriate. 

IRS agreed with our recommendation that it determine why more
eligible taxpayers do not use TeleFile, including more specific
information as to why nonusers prefer to file on paper.  IRS
officials told us that by the end of fiscal year 1997, IRS would
conduct a focus group study of TeleFile nonusers to determine why
they prefer to file on paper and to identify any barriers. 

IRS officials said that steps have also been taken to address some
concerns identified by past nonuser surveys.  IRS believes that
taxpayers' preference for paper returns is linked to their
familiarity with the form.  The TeleFile worksheet that taxpayers had
been instructed to fill out and maintain as a record of their filing
did not have the same "official" appearance as a tax form.  For the
1997 filing season, according to IRS officials, TeleFile users will
be instructed to complete a TeleFile Tax Record instead of a
worksheet.  As described by the officials, the TeleFile Tax Record
will (1) include lines for the taxpayer's name and address, (2) look
more like the Form 1040EZ, and (3) be an official document.  IRS
hopes this change will provide potential TeleFile users with a higher
comfort level.  IRS officials also said that advertisements and other
publicity tools that were used in 1996 will be emphasized again in
1997 to educate the public on the simplicity of using TeleFile. 

In commenting on our second recommendation, IRS officials said that
IRS, in conjunction with FMS, has formed a task force to identify a
long-term solution for 1998 and beyond for directing Form 1040 tax
payments to lockboxes.  According to the officials, the group has
been tasked with (1) identifying options that complement Treasury's
goals of increasing the availability of funds and reducing the cost
of collecting federal funds, (2) reviewing what is required of
lockboxes by IRS to minimize operational and ancillary costs, and (3)
making recommendations to management.  The group is scheduled to
present their findings to management by March 1997.  This time frame
should provide IRS with information to make a decision on Form 1040
tax payment processing that could be implemented for the 1998 filing
season. 


---------------------------------------------------------- Letter :7.1

We are sending copies of this report to the Subcommittee's Ranking
Minority Member, the Chairmen and Ranking Minority Members of the
House Committee on Ways and Means and the Senate Committee on
Finance, various other congressional committees, the Secretary of the
Treasury, the Commissioner of Internal Revenue, the Director of the
Office of Management and Budget, and other interested parties. 

Major contributors to this report are listed in appendix II.  Please
contact me on (202) 512-9110 if you have any questions. 

Sincerely yours,

Lynda D.  Willis
Director, Tax Policy and
 Administration Issues


IRS' PERFORMANCE GOALS FOR THE
1995 AND 1996 FILING SEASONS AND
RELATED ACCOMPLISHMENTS
=========================================================== Appendix I

                 1995 filing season\a                1996 filing season\a
          ----------------------------------  ----------------------------------
Performa
nce
indicato
r         Goal              Accomplishment    Goal              Accomplishment
--------  ----------------  ----------------  ----------------  ----------------
Accuracy  Process 93.0%     94.1% were        Process 93.0%     94.0% were
of        accurately        processed         accurately        processed
returns                     accurately                          accurately
processe
d by
Code
and Edit
staff\b

Accuracy  Process 94.0%     94.6% were        Process 93.0%     94.2% were
of        accurately        processed         accurately        processed
returns                     accurately                          accurately
processe
d by
data
transcri
bers

Service   Process 9,800     10,820 returns    Process 10,000    12,174 returns
centers'  returns per       were processed    returns per       were processed
returns   staff year        per staff         staff year        per staff year
processi                    year
ng
producti
vity\c

Returns   11 days           Various types of  11 days           Various types of
processi                    1040s ranged                        1040s ranged
ng cycle                    between 5 and                       between 8 and 11
time\d                      9 days                              days

Accuracy  Process 97.0%     99.5% were        Process 98%       99.6% were
of        accurately        processed         accurately        processed
refunds                     accurately as of                    accurately as of
on paper                    May 1995                            May 1996
returns

Timeline  Issue within an   Issued within     Issue within an   Issued within an
ss of     average of 40     an                average of 40     average of 38
refunds   days              average of 36     days              days as of May
on paper                    days as of                          1996
returns\                    May 1995
e

Timeline  Payments          All payments      Payments          All payments
ss of     received 4/17/    received 4/17/    received 4/15/    received 4/15/
processi  95 through 5/2/   95 through 5/2/   96 through 5/1/   96 through 5/1/
ng tax    95 were to be     95 were           96 were to be     96 were
payments  deposited no      deposited by      deposited no      deposited by
submitte  later than 5/3/   5/3/95            later than 5/1/   5/1/96
d with    95                                  96
returns

Tax       Answer 16.5       19.2 million      Answer 19.2       22.9 million
assistan  million calls     calls were        million calls     calls were
ce                          answered (116%                      answered (119.5%
telephon                    of schedule)                        of schedule)
e calls
schedule
d to be
answered
\f

Taxpayer  Not applicable    Not applicable    Provide 41.4%     Provided 50%
service                                       level of access   level of access
level of
access\g

Accuracy  Answer 90%        90% were          Answer 90%        91% were
of tax    accurately        answered          accurately        answered
law                         accurately                          accurately
assistan
ce

Forms-    Answer 4.4        4.2 million       Answer 4 million  3.9 million
ordering  million calls     calls answered    calls             calls answered
telephon                    (95.7% of                           (98.3% of
e calls                     schedule)                           schedule)
schedule
d to be
answered
\f

Accuracy  Process 96.5%     97.2% processed   Process 96.5%     97.3% processed
of        accurately        accurately        accurately        accurately
processi
ng form
orders
--------------------------------------------------------------------------------
\a Data are as of April 1995 and April 1996, unless otherwise noted. 

\b Code and Edit staff prepare returns for computer entry by, among
other things, ensuring that all data are present and legible. 

\c The "returns processing productivity" indicator is based on the
number of weighted returns processed, which includes all returns
whether they were processed manually, through scanning equipment, or
electronically.  The different types of returns are weighted to
account for their differing processing impacts.  For example, a paper
Form 1040 has a higher weighting factor than a paper Form 1040EZ,
which in turn has a higher weighting factor than electronically
processed returns. 

\d Cycle time is the average number of days it takes service centers
to process returns. 

\e The "refund timeliness" indicator is based on a sample of paper
returns and is calculated starting from the signature date on the
return to the date the taxpayer should receive the refund, allowing 2
days after issuance for the refund to reach the taxpayer.  As
discussed in our report on the 1995 filing season (GAO/GGD-96-48),
the 36-day accomplishment cited for 1995 was slightly understated by
the exclusion of certain refunds that, according to IRS' standards,
should have been included.  That issue was not a problem in 1996. 

\f The "calls scheduled to be answered" indicator is the number of
telephone calls IRS believes its call sites will be able to answer
with available resources.  The indicator does not reflect the number
of calls IRS expects to receive. 

\g The "level of access" indicator is the number of calls answered
divided by the number of individual callers.  See pages 4 to 6 for
more information on this indicator. 

Source:  IRS data. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II

GENERAL GOVERNMENT DIVISION,
WASHINGTON, D.C. 

David J.  Attianese, Assistant Director, Tax Policy and
Administration Issues
William H.  Bricking, Senior Evaluator
Robert L.  Giusti, Senior Evaluator
Monika R.  Gomez, Evaluator
Christopher E.  Hess, Evaluator

KANSAS CITY FIELD OFFICE

Royce L.  Baker, Issue Area Manager
Doris J.  Hynes, Evaluator-in-Charge
H.  Yong Meador, Evaluator

ATLANTA FIELD OFFICE

Katherine P.  Chenault, Senior Evaluator
Jyoti Gupta, Evaluator

*** End of document. ***