Role of Paid Preparers in Tax Compliance: New Evidence

The Leviner-Richison study contributes some fascinating new evidence on tax compliance by taxpayers who use different types of preparers. Some of their results present quite a contrast to the previous literature in the subject, which I discussed in my last post.

Leviner and Richison's new work analyzes a random cross-sections of 1999 tax returns claiming EITC that had been selected to study noncompliance with the Earned Income Tax Credit (EITC) rules. They initially divided those returns into nine different categories based on the type of preparer: (1) self-prepared, (2) CPA preparer, (3) attorney preparer, (4) Enrolled Agent, (5) HR Block/Jackson Hewitt (the two big national chains), (6) Other Professional Tax Preparer, (7) Friend/Relative, (8) IRS/VITA/TCE (volunteer free tax prep and IRS taxpayer assistance), and (9) Other.

The big chains (#5) and Other professional preparer (#6) accounted for 56% of the returns in the sample. Very few returns in the study sample were prepared by CPAs and attorneys that the authors decided to combine those categories for their analysis. That's not too surprising, since EITC recipients are low-income taxpayers and very few attorneys and CPAs specialize in serving that clientele, aside from those who volunteer in VITA.

Here's how they summarize their results:

As illustrated in Table 1 below, our analysis reveals that CPA/Attorney, HR Block/Jackson Hewitt, and IRS/VITA/TCE staff have the lowest percentage of returns with change (either positive or negative) to EITC when the original amount claimed on the returns is compared to that concluded by the IRS after audits and reviews....

"Change" is IRS-speak for finding an error rate in an audit. So a high change rate means a high error rate. They found both overclaim and underclaim errors, but, not too surprisingly, the overclaim error rate was about an order of magnitude greater than the underclaim error rate. That is, when the IRS found an error, it was far more likely to be that the taxpayer's return had claimed too much tax refund rather than too little refund.

An important note to keep in mind: the rules for claiming "qualifying children" on a tax return were quite different in 1999 than they are today. The rules are still complicated, confusing, and subject to abuse today, but this was arguably even more true of the rules that applied in 1999.

That said, this is a REALLY discouraging error rate. Bear in mind that these returns were a RANDOM representative set of returns filed in 1999.

The authors have additional cautionary notes to bear in mind in interpreting this data:

An examination of the Adjusted Gross Income line (AGI, Table 2) reveals a nearly 50 percent rate of returns with change for CPAs/Attorneys, and over 60 percent for Enrolled Agents. It is possible that the financial circumstances EITC claimants have are complex enough to confuse even the most trained of preparers. Paid preparers usage is believed to be more common among taxpayers with complicated returns which may go some way toward explaining errors on returns filed by preparers (as opposed to taxpayers filing for themselves) generally. This might be particularly the case with regards to taxpayers engaging the most trained and experienced preparers such as those who are CPAs and Attorneys. Even so, errors made on paid prepared returns do not necessarily mean that these errors are the result of the preparer’s, as opposed to taxpayer’s, misconduct.

In other words, it's important to bear in mind the correlation vs. causality problem here. It's entirely possible that the higher error rate of Enrolled Agents vs. unenrolled preparers may reflect the greater complexity of the returns they prepare rather than a lesser degree of competence or conscientious adherence to the law.

They also looked at patterns of error in AGI reported by different types of preparers, summarized here:

The types of preparers to exhibit the highest rate of change in claimed AGI are: Other Professional Tax Preparer, Other Preparer, and CPA/Attorney, (in that order). HR Block/Jackson Hewitt and IRS/VITA/TCE have almost half that rate of change and are the most accurate compared with other preparer types.

Again, we need to keep in mind the previous cautionary note about correlation vs. causality, because the types of taxpayers who patronize different types of preparers are not necessarily the same.

This study was intended as exploratory rather than conclusive, but it certainly provides some interesting data for consideration.

Correlation vs. causality: tax pros and audit probability and LOVE

Tax Practitioner Credentials and the Incidence of IRS Audit Adjustments

2003, Accounting Horizons

John Hasseldine, Peggy A Hite

Abstract

A random selection of Internal Revenue Service office audits from October 1997 to July 1998, the type of audit that concerns most taxpayers, is analyzed. Taxpayers engage paid preparers in order to avoid this type of audit and to avoid any resulting tax adjustments. Whether there are more audit adjustments and penalty assessments on tax returns with paid-preparer assistance than on tax returns without paid-preparer assistance is examined. By comparing the frequency of adjustments on IRS office audits, it is concluded that there are significantly fewer tax adjustments on paid-preparer returns than on self-prepared returns. Moreover, CPA-prepared returns resulted in fewer audit adjustments than non CPA-prepared returns. The study included 2,253 audit cases; 71% of the audited taxpayers prepared their own tax return, 19% hired a CPA, and 10% hired a non-CPA. Most adjustments come from deduction errors, and CPA-assisted returns have significantly lower likelihood of having a deduction adjustment.

Given that most paid preparers are not CPAs, a naive interpretation of the data above could lead to an erroneous conclusion, to wit: hiring a CPA to prepare your tax return "causes" a higher likelihood of IRS audit than hiring a non-CPA to prepare it.

An alternative explanation could be that the subset of taxpayers whose finances are sufficiently complicated as to make them likely to attract the interest of the IRS are also more likely to hire a CPA than they are to hire a non-CPA.

Since the study also found that IRS audits in the sample turned up fewer errors on CPA-prepared returns than on non-CPA-prepared returns, choosing a CPA might well have been a very good decision on their part.

In a study such as this one, there are always many unobserved variables which could be a source of what econometricians refer to as LOVE ("Left Out Variable Error.")

The sample of taxpayers selected for office audit is clearly not a random representative cross-section of taxpayers. Taxpayers who prepare their own tax returns are a minority of taxpayers, but they constituted 71% of those audited in the study sample. It's an interesting question as to whether preparing your own tax return raises the probability of audit, after statistically controlling for the complexity and other attributes of the return.

Time series, cross section, and quasi-experiment analysis simple example

Three ways to use the data to predict changes in labor effort in response to benefit guarantee change

Approach #1Time series analysis:
compare Arkansas in 1996 vs. Arkansas in 1998
Use just two data points (same place, different times)
Benefit guarantee went down by $1,000 (5K to 4K) and labor went up 200 hours (across that time period within Arkansas.

Time series analysis prediction:
$1,000 benefit guarantee cut ==> 200 hour increase in labor

Problem with this? Lots of other things changed over time, maybe the increase is due to something else besides the benefit guarantee that changed over that time period.

Approach #2Cross section analysis:
compare Louisiana in 1998 vs. Arkansas in 1998
Use just two data points (two places, same time)
Benefit guarantee is $1000 less in Arkansas than in Lousisiana (5K vs. 4K) and labor is 100 hours greater in Arkansas than in LA that year

Cross-section prediction:
$1,000 benefit guarantee cut ==> 100 hour increase in labor

Problem with this? Lots of other differences between Louisiana and Arkansas besides the benefit differnce, maybe the increase is due to something else different between the two states.

Approach #3Quasi-experiment analysis also known as "difference-in-difference" analysis

Use all four data points across time AND space.

Use Louisiana as "control group" for study because benefit guarantee stayed the same there.

Use Arkansas as "treatment group" for study because benefit guarantee changed there.

Compare CHANGE over time in labor in Arkansas vs. CHANGE over time in Louisiana labor

change over time in Arkansas (treatment group)
= 200 more hours of work = 1,200 - 1,000

change in Louisiana (control group)
= 50 more hours of work = 1,100 - 1,050

Now take difference between 200 and 50 = 150 hours predicted effect of $1,000 increase in benefits

Quasi-experiment prediction is
$1,000 benefit guarantee cut ==> 150 hour increase in labor

This approach isn't perfect, but it's better than the other two, because it "controls" for some of the differences between the states (i.e., those that stayed the same over time.)

Example above is very simple. In practice might use more time periods, more variables, more states, etc.

The price elasticity of soda consumption

Today's NYT has an article on the soda tax prominently featuring price elasticity, which we'll talk about in class today.

Here's an excerpt.

The scientific paper found that a beverage tax might not only raise revenue but have significant health effects, lowering consumption of soda and other sweet drinks enough to lead to a small weight loss and reduced health risks among many Americans.

The study cited research on price elasticity for soft drinks that has shown that for every 10 percent rise in price, consumption declines 8 to 10 percent.

John Sicher, the publisher of Beverage Digest, a trade publication, said that a two-liter bottle of soda sells for about $1.35. At 67.6 ounces, if the full tax was passed on to consumers, that would add 50 percent to the price. A 12-can case, which sells today for about $3.20, could rise by $1.44, a 45 percent increase.

“A one cent per ounce tax would create serious problems and potentially adversely impact sales for the American beverage industry,” Mr. Sicher said.

The New England Journal of Medicine also released an article on the topic this week. The graph from their article shows dramatic increases in soda consumption during a time period in which we know that obesity increased. There is clearly correlation, but causality is much harder to establish. We need to look beyond the simple time series data presented here.