![NinaOlson2019](https://www.taxconnections.com/taxblog/wp-content/uploads/NinaOlson2019.jpg?resize=90%2C90&ssl=1)
As part of the TAS Report to Congress the study addresses two research questions regarding taxpayers attitudes. The first question: “How do attitudes towards paying taxes vary among different types of taxpayers?” For this question, the focus is on a comparison of self-employed taxpayers (audited and unaudited) and wage earners (who did or did not experience either IRS ID theft processing procedures or who experienced the IRS questioning the legitimacy of their refund return).
The second research question is: Do attitudes among audited self-employed taxpayers vary in accordance with the type of audit and the outcome of the examination?” As wage income is usually subject to third-party reporting, wage earners tend to have relatively few opportunities for tax noncompliance in comparison with self-employed taxpayers (Kleven et al., 2011). However, it is unlikely that opportunity alone drives tax compliance behavior. Personal beliefs, social norms, and past experiences with the IRS shape taxpayer attitudes. Personal experiences might be particularly relevant for taxpayers who have been victims of tax fraud involving ID theft or who have been suspected of tax fraud. Solving these cases frequently delays legitimate refund claims substantially, which imposes financial hardship on taxpayers, potentially erodes trust in the IRS, and might adversely impact voluntary compliance. Against this background, analyzing the attitudes of different occupational groups will contribute to an understanding of the determinants of tax noncompliance.
Attitudes of self-employed taxpayers who recently experienced an audit might explain differential behavioral responses to examinations (Gemmel & Ratto, 2012; Beer et al., 2015). For instance, audited taxpayers might misperceive the probability of future audits (Mittone et al., 2017). Alternatively, audits might crowd out the intrinsic motivation to comply among honest individuals (Lederman, 2017) or further decrease the willingness to pay of taxpayers whose cheating has gone undetected. Behavioral responses will likely reflect the type and outcome of the examination. Field audits, for instance, might have a more powerful impact on attitudes than correspondence audits. Similarly, taxpayers may have a more positive outlook following an audit that concludes with a tax refund than one that results in an additional tax assessment. However, the attitudinal implications of tax audits that result in no tax change are less clear. In such cases the subjective appraisal of the audit experience might have a strong effect on future reporting behavior.
SURVEY INSTRUMENT AND SAMPLE DESIGN (SECTION 4) The Slippery Slope Framework was employed as a guide in designing the survey instrument. We have constructed scales from the survey responses to relevant sets of related questions that reflect the various determinants of compliance behavior under this framework. For instance, the perceived audit probability scale is based on a combination of the individual rankings (on a scale from 1 to 9) for two questions on the perceived risk of audit (questions DF1 and DF2). 6 Figure 5.2 provides an overview of the survey scales. A copy of the survey instrument is provided in Appendix A. The individual survey items, the correlations between these items, and the construction of scales based on these items are presented in Appendix B.
Sample Selection
Separate survey samples were drawn for different taxpayer groups. Two samples of wage earners were drawn of approximately 400 respondents each. The first was a random sample of wage earners who appeared to have been subject to an ID theft investigation or a suspected illegitimate refund claim. Ultimately, some of the respondents were dropped from the sample after discovering that they likely did not actually experience such an investigation. After their removal, we were left with 357 respondents. The second wage earner sample was drawn randomly from the overall wage-earner population. Separate samples of audited and unaudited self-employed taxpayers were selected. To be eligible for selection, an audited taxpayer had to meet the following criteria:
■ Had an operational audit between tax years 2010 and 2015.
■ Filed a Schedule C return for at least the three years covering the first tax year preceding the audit through the first tax year following the audit.
■ All income tax returns were filed chronologically.
■ The audit for tax year T was initiated prior to the filing of the tax year T+2 return.
■ No examinations were initiated or ongoing two years prior to the audit under consideration.
■ The taxpayer was not a resident of Puerto Rico.
To explore differences in taxpayer attitudes according to the type of audit (office, field, and correspondence) and the audit outcome (positive tax adjustment, no-change, and tax refund), we drew separate samples of audited self-employed taxpayers for each of the nine (3×3) possible audit type and outcome combinations. For each type of audit, a target of 100 respondents was set for taxpayers receiving a tax refund as a result of the audit, 150 respondents for those experiencing no tax change, and 200 for those experiencing an additional tax assessment. To help insure that a reasonably comparable “control group” of unaudited Schedule C filers could be sampled, we performed a propensity score analysis of the likelihood of an audit. Where possible, we under sampled taxpayers with very high or very low propensity scores and oversampled those with more moderate scores. In certain cases, these targets could not be met, so additional respondents were drawn from a different category in such cases.
Once the surveys of the audit samples were completed, we selected samples of unaudited Schedule C filers meant to be broadly similar to the audited Schedule C respondents. A generalized propensity scoring procedure was employed that matched unaudited self-employed taxpayers to the audit respondents on the basis of the predicted likelihood associated with each of the three types of audit. 8 This matching process was carried out separately for each audit tax year (2010 through 2015) to account for the fact that the audit selection process tends to vary from year to year. Separate “control samples” were drawn for each audit tax year. Within each year, separate subsamples were randomly drawn for each of six different propensity score classes. The sample sizes were set sufficiently high to ensure that the target number of respondents could be achieved within each subsample.
Sample Characteristics
Ultimately, the sampling process resulted in a combined sample of 3,486 useable responses. 9 Overall, 61 percent of the respondents are male, and the mean age in the sample is 55 (SD=13.4). A majority of respondents holds at least a high-school degree and works full-time. Our sample includes almost 900 individuals who state that they are currently not working; most of them (622) are retired. The final sample includes 2,729 self-employed taxpayers (1,363 audited and 1,366 non-audited) and 757 wage earners (357 who experienced an ID theft investigation and 400 from the general population.) Some socio-economic and demographic statistics for our sample are presented in Figures 5.3 and 5.4.
Under our sampling design, almost 50 percent (1,363 out of 2,729) of the Schedule C taxpayers who were surveyed had experienced fairly recent audits. Table 2 presents some socio-economic and demographic statistics of the audited respondents. We distinguish between different audit types and audit outcomes. Overall, 454 taxpayers received a field audit, 453 an office audit, and 456 a correspondence audit. Further, 603 individuals received an additional assessment, 503 no additional assessment, and 257 taxpayers a tax refund. The mean age of audited Schedule C taxpayers is slightly higher than the sample average (57.41 years, SD=12.44) and 62 percent are male. The majority of audited self-employed taxpayers have at least a college diploma and work full-time. Among those who responded to the questions regarding employment type, most are exclusively self-employed, although some report also working for another employer.
PRELIMINARY FINDINGS (SECTION 5)
This section presents the findings of our survey of self-employed taxpayers and wage earners. First, we analyze differences between audited Schedule C taxpayers, non-audited schedule C taxpayers, and wage earners. Second, we investigate the relationship between different audit types (field audit, office audit, or correspondence audit) and outcomes (positive adjustment, no adjustment, refund) and taxpayer attitudes. Third, we analyze how attitudes of taxpayers who are aware that their ID may have been stolen to unlawfully claim a tax refund or who the IRS suspects of an illegitimate refund claim differ from those who are not in this situation. Finally, we examine how the results relate to the assumptions of the Slippery Slope Framework. Attitudes of Audited and Non-audited Schedule C Taxpayers and Wage Earners To investigate how attitudes vary among taxpayers, we have performed a multivariate analysis of covariance (MANCOVA) using taxpayer type (Schedule C audited, Schedule C non-audited, and wage earners) as the explanatory variable and survey scales (as presented in Figure 5.5) as the dependent variables. 12 After controlling for gender and age, we find significant effects of taxpayer type (F(42, 5986) = 6.87, p < .001, eta2 = .046); gender (F(21,2992) = 5.22; p < .001, eta2 = .035), and age (F(21,2992) = 10.37; p < .001, eta2 = .068). Figure 5.5 shows the adjusted means and standard errors of the survey scales by taxpayer type. These survey scales represent the mean of the responses to the relevant survey items.
The results indicate that self-employed taxpayers are more likely to consider cheating, are more likely to view taxpaying as a coercive process, and perceive higher audit probabilities and fines for noncompliance than wage earners. Audited self-employed taxpayers perceive an especially high risk of audit. Consequently, audited Schedule C filers exhibit lower trust in the IRS and are relatively more likely to agree strongly with the statement: “Paying taxes feels like something is taken away from me rather than a contribution to society.”
Interestingly, audited self-employed taxpayers perceive higher degrees of procedural justice, information justice, interpersonal justice, and distributive justice than the unaudited control group. This indicates that those who recently experienced IRS audit procedures view them as more transparent, respectful, and appropriate in their outcomes than those who have not. A deeper investigation of these differences in justice perceptions is undertaken below, where we break down the results by audit type and examination outcome.
The survey responses reveal that experiencing an audit induces strong negative emotions. Audited taxpayers report higher levels of fear, anger, threat, and caution when thinking about the IRS. Moreover, audited taxpayers also feel less protected by the IRS. We do not observe differences in self-reported tax knowledge, attitudes towards paying taxes, and social norms between different taxpayer types.
Attitudes of Audited Schedule C Filers by Audit Type and Outcome
We first investigate whether audited taxpayers are aware that they were audited. Figure 5.6 shows that only 61.3 percent of audited self-employed taxpayers indicate they have been audited. However, we observe substantial differences between audit types. Audit awareness is particularly low in case of correspondence audits, where only 38.8 percent indicate they had been audited. For field audits, more than two-thirds (67.0 percent) recall the audit, while the rate of recall is highest for the office audit group (73.7 percent). Audit awareness is generally higher in the case of positive tax adjustments than when then examination results in either no adjustment or a tax refund. Surprisingly, only 55.3 percent of audited taxpayers who received a refund as a result of the examination recall the audit.
Recent Comments