Estimating the Impact of Liens on Taxpayer Compliance

Report
Estimating the Impact of
Liens on Taxpayer
Compliance Behavior and
Income
Taxpayer Advocate Service
Research & Analysis
June 2012
IRS uses Notice of Federal Tax Lien
(NFTL) to facilitate the collection of
unpaid tax debts.

When the IRS sends a tax assessment and a
taxpayer does not pay in full within 10 days, IRS
may file a NFTL.

IRS must file a NFTL in the appropriate location,
such as a county register of deeds.

NFTL establishes the priority of the government’s
interest in a taxpayer’s property.

NFTL places third parties on notice.
Lien Project 2011
TAS, Research and Analysis
2
The National Taxpayer Advocate is concerned
that lien filings, which are often harmful to
taxpayers, have increased without
corresponding increases in collections.
Inflation Adjusted Total Collection Yield vs. Liens Issued
40
1,200
35
1,000
30
Inflation
Adj Total Yield
(billions)
800
25
20
600
15
400
Number of
Liens Issued
(thousands)
10
5
200
0
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Inflation Adj. Total Yield
Lien Project 2011
Notices of FTLs Issued by IRS
TAS, Research and Analysis
3
Research objectives are to determine
the impact of IRS lien filing on:
•
Taxpayers’ payment behavior with
respect to the original liabilities incurred
in 2002;
•
Taxpayers’ payment compliance in
subsequent periods;
•
Taxpayers’ filing compliance in
subsequent periods; and
•
Taxpayers’ income in subsequent periods.
Lien Project 2011
TAS, Research and Analysis
4
Research Methods
Initial statistical concern: selection bias due to
nonrandom filing of tax liens. Addressed by:
• Matching approach selected: propensity score
matching
• Rosebaum and Rubin (1983)
• M. Caliendo and S. Kopeinig (2008)
• Matching algorithm, nearest available neighbor
method,
• “Greedy” Algorithm, Kosanke and Bergstralh, Mayo Clinic
• SAS program
Lien Project 2011
TAS, Research and Analysis
5
Research Methods
Taxpayers …
• Incurred unpaid tax liabilities in 2002 and
had no such liabilities at the beginning of
2002.
• Received NFTL between 2002 and 2004.
• Comparison Group:
Lien Project 2011
no NFTL filed.
TAS, Research and Analysis
6
Research Methods
Two Phase Analysis
• Phase 1
• Logistic regression estimated the propensity for
filing a tax lien.
• The propensity score variable was used for our
matching algorithm.
• All taxpayers have a propensity score value in
our analysis.
• Variables were derived from IRM sections on lien
filings.
Lien Project 2011
TAS, Research and Analysis
7
IRM NFTL Filing Criteria
(in model)

The aggregate unpaid balance of
assessment (UBA) is $5,000 or more.

Other indications of non-compliance.

IA that is not guaranteed, streamlined, or
in-business.

Currently not collectible.

Property exempt by bankruptcy.
Lien Project 2011
TAS, Research and Analysis
8
IRM NFTL Filing Criteria
(in model)

Case is closed as a hardship
(UBA>=$5,000).

NFTL filed and new liabilities
(UBA>=$2,000).

Consider lien filing in any situation where
taxpayer has broken a promise.
Lien Project 2011
TAS, Research and Analysis
9
Research Methods
Two Phase Analysis
• Phase 1:
• Performed match on propensity
scores.
• Developed comparable lien and nonlien groups.
• Mitigated selection bias issue by
establishing a matched set of
comparable lien and non-lien cases.
Lien Project 2011
TAS, Research and Analysis
10
Research Methods
Phase
2:
Compared lien group to non-lien group.
Developed logistic regression equations.
Used three categories of independent variables
• Individual taxpayer (TP) characteristics
• Income information
• IRS audit and collection activities
Lien Project 2011
TAS, Research and Analysis
11
Research Methods
Tax Compliance Models
•
•
Current Payment model investigated lien’s impact
on probability of taxpayer making sufficient
payments during study periods to reduce original
liability incurred in 2002.
Future Payment model investigated impact of lien
on the probability of the taxpayer staying
compliant with his payment of tax liabilities in all
periods subsequent to 2002.
Lien Project 2011
TAS, Research and Analysis
12
Research Methods
Tax Compliance Models
• Future Filing model investigated the tax lien’s
impact on the taxpayer’s timely filing behavior
during the study period.
• Future Income model investigated the impact of
liens on the taxpayer’s future income, measured as
change in the taxpayer’s total positive income
between the beginning and the end of the study
period.
Lien Project 2011
TAS, Research and Analysis
13
Phase 2 Tax Lien Results
Taxpayers with a NFTL were:
• Less likely than comparable taxpayers without liens
to reduce their 2002 liabilities.
• Less likely to timely file required returns after 2002.
• Less likely to generate greater total positive income
after 2002.
• More likely to pay subsequent liabilities.
Lien Project 2011
TAS, Research and Analysis
14
Phase 2 Tax Lien Results
•
Marginal effects (point estimate) show the impact of tax liens
on the probability of the outcome measures.
Models
A
A
2002-2005 2002-2006 2002-2007 2002-2008 2002-2009 2002-2010
Average
Current Payment
-6.36%
-6.00%
-5.99%
-5.21%
-4.78%
-4.54%
-5.48%
Future Payment
5.58%
4.69%
3.70%
2.77%
2.18%
1.23%
3.36%
Future Filing
-0.87%
-1.51%
-2.12%
-2.48%
-2.83%
-2.78%
-2.10%
Future Income
-7.89%
-7.61%
-6.70%
-6.38%
-5.78%
-5.16%
-6.59%
All models, except the future payment model, produced coefficients for the lien indicator that were negative
and significant. The lien coefficients for the future payment model were positive and significant.
Lien Project 2011
TAS, Research and Analysis
15
Research Limitations
• Due to data limitations, not all lien filing criteria in
propensity model.
• No lien or non-lien cases exist in the top ten
percent of propensity scores and few lien or nonlien cases are in the next five percent of propensity
scores.
• Two matches of lien cases against the sample of
non-lien cases were performed.
• Matched cases have some non-lien cases used twice and
have a weight of two.
Lien Project 2011
TAS, Research and Analysis
16
Future Research
 Consider
only events which occurred before
NFTL filing.
 Remove
CNC taxpayers from the analysis.
 Develop
models with economic indicators
(e.g., state unemployment rates).
 Investigate
when NFTLs are likely to be most
effective as a collection tool.
Lien Project 2011
TAS, Research and Analysis
17
Estimating the Impact of
Liens on Taxpayer
Compliance Behavior and
Income
Taxpayer Advocate Service
Research & Analysis
June 2012

similar documents