An official website of the U.S. Govt
On March 13, 2018, the IRS announced that on September 28, 2018, it would end the offshore voluntary disclosure program (OVDP), as it had only attracted 600 applicants in 2017. So now is a good time to take a step back to review the program in the broader context of the research on tax amnesties.
Settlement programs and other voluntary disclosure or correction programs generally offer some form of amnesty. Offering broad tax amnesties on a regular basis, as many states do, can erode voluntary compliance. However, settlement programs are important tools for tax administrators and governments around the world. The puzzle is when settlement programs should be offered and how they should be structured. In Volume 2 of my 2017 Annual Report to Congress TAS reviewed the tax amnesty literature in an effort to shed light on these questions.
In summary, TAS found that narrowly-tailored programs aimed at specific objectives are less likely to have negative effects on compliance than broad amnesties aimed at accelerating short-term revenue. Research (e.g., study 1, study 2, and study 3) suggests that any negative effects can be mitigated if the program is coupled with other measures to improve compliance. Some have suggested that when the government increases enforcement, it can even accelerate compliance gains by offering a limited amnesty at the same time. If so, then compliance-enhancing measures such as increases in penalties, automated information exchanges, and third-party reporting could also provide opportunities for tax agencies to use settlement programs to accelerate the transition to compliance norms while reducing disputes. Indeed, the Organization for Economic Cooperation and Development (OECD) recommends that tax agencies consider offshore voluntary disclosure programs (OVDPs) as they receive more information about citizens’ and residents’ assets in other countries, and at least 47 countries have adopted them, including the U.S.
Our research finds that when noncompliance has become the norm, the government’s failure to address it in a way that is viewed as reasonable and proportionate – for example by putting taxpayers on notice and offering an amnesty or amnesty alternative – could pose an altogether different kind of risk to voluntary compliance.
Research suggests that broad and frequent tax amnesty programs generally do not generate revenue that exceeds their true long-term costs. Nonetheless, it can still make sense for tax agencies to offer limited amnesties or amnesty alternatives to:
Even broad amnesties can be useful. Tax agencies use broad amnesties to promote future compliance when taxpayers have acted reasonably or the government contributed to the noncompliance. For example, the U.S. Supreme Court has called into question the legality of the requirement for out-of-state sellers to collect state sales and use taxes. Some states have responded by offering sellers broad amnesty for past noncompliance if they agree to collect these taxes in the future.
Although even broad amnesties can promote future compliance, those that merely reduce or eliminate penalties are so different from broad amnesties – posing such smaller risks to voluntary compliance – that we refer to them as amnesty alternatives. They generally attract participants who made inadvertent errors and who are the least committed to noncompliance. Economists have suggested that allowing people to pay their back taxes with interest before their noncompliance is detected should have little effect on economic deterrence (i.e., the expected economic costs and benefits of noncompliance). Accordingly, they should not attract people who cheat for economic reasons.
A review of state tax amnesties bears this out, suggesting that even broad amnesties generally attract people who owe relatively small amounts (e.g., inadvertent errors), rather than those with large chronic delinquencies. According to another study, amnesties generally attract those who want to comply. Even among those whose original noncompliance was intentional, a survey of participants found that their noncompliance was often due to their inability to pay. Moreover, an analysis of OVDP participation around the world suggests that one major category of holdouts includes those who remain “unwilling to pay the tax due.”
Providing penalty relief to those with small delinquencies whose misreporting was inadvertent and who correct it before being contacted by the tax agency, does not raise the same concerns as a broad amnesty. Participants are voluntarily paying taxes with interest, albeit late. These taxpayers may not even be viewed as getting special treatment because all taxpayers can avoid accuracy-related penalties, even after detection, if they can establish the misreporting was due to “reasonable cause” under IRC § 6664(c). When viewed in this way, amnesty alternatives could be viewed as fairer than asserting penalties for every violation. If so, they might increase trust for the IRS. A 2012 study by TAS found that trust for the IRS is correlated with voluntary compliance. In addition, one simulation found that amnesties regarded as fair increased voluntary compliance. Another found that they had significantly greater positive effects on voluntary compliance when citizens voted on them.
Consistent with these fairness considerations, the IRS has a longstanding amnesty program that promotes voluntary compliance by allowing taxpayers to avoid the accuracy-related penalty (but not the failure to pay penalty) by filing “qualified amended returns” (QARs) any time before being contacted by the IRS, as provided by Treas. Reg. § 1.6664-2. If they come forward before being detected, they can also generally avoid being recommended for criminal prosecution under the IRS’s longstanding voluntary disclosure practice (VDP), provided they follow through by cooperating and paying the liability, as described in IRM 9.5.11.9.
Moreover, as tax agencies receive more financial information from third parties, such as foreign tax agencies and financial institutions, settlement programs that include an element of amnesty are becoming the norm. An OECD survey of VDPs around the world found that close to half of the countries (19 out of 47) that responded waived all monetary penalties for taxpayers who make voluntary disclosures. Thus, when a tax agency increases its ability to detect underreporting, its failure to offer at least an amnesty alternative may be viewed as outside the norm.
Adopting tax administration strategies that are responsive to the taxpayer’s motivational posture – applying amnesty to those who made honest mistakes, and reserving enforcement-oriented treatments for those committed to noncompliance – is consistent with the so-called “responsive regulation” tax compliance model endorsed by the OECD Forum on Tax Administration Compliance Sub-group, and a number of tax agencies throughout the world. This model suggests that when an agency assumes all violations are intentional (e.g., by always asserting penalties) it can reduce the perceived legitimacy of the agency and the tax system.
When the government does not enforce a law, it encourages otherwise-honorable people to mirror the government’s priorities and view compliance as unimportant. It allows noncompliance to become the norm. In such cases, some have suggested that imposing sudden and severe sanctions against the few people that the government has the resources to audit may seem unfair and disproportionate, undermining the agency’s legitimacy. Along the same lines, in the context of speed limits, others have observed that:
By warning drivers of the presence of enforcement, the aim is to provide the driver with every opportunity to change behavior. If the enforcement is transparent and warnings have been presented, then it is hard for the offending driver to claim that the procedure is unfair.
Perhaps for this reason, the IRS routinely delays implementation of penalties for the failure to comply with new requirements, presumably to give taxpayers a reasonable transition period to learn about and comply with them. Similarly, increasing enforcement or increasing the penalties for previously unenforced or lightly-penalized conduct is likely to seem fairer if preceded by an amnesty or amnesty alternative.
As an example, the Australian Taxation Office (ATO) reportedly damaged its reputation when addressing a mass-marketed tax scheme. Rather than proactively responding to questions about whether these schemes were legitimate, it was silent while tax professionals advised taxpayers that they were. After taxpayers invested, it adjusted their accounts and suggested they were tax cheats. The ATO subsequently adopted a settlement program that waived interest and penalties, but the damage had already been done. Years later a survey found that most participants still thought the settlement was unfair, held more negative views toward the agency, and reported about the same level of efforts to minimize taxes as before the ATO adjusted their accounts.
ATO’s experience is consistent with experiments, which suggests that people reciprocate by punishing unfair behavior, even if it is not in their economic self-interest. Conversely, an amnesty alternative could improve voluntary compliance if it seems fair, puts people on notice that noncompliance will be punished in the future, and fosters trust that the tax agency will address noncompliance in a reasonable and proportional manner that takes the taxpayer’s facts and circumstances into account (e.g., the taxpayer’s motivational posture). At the margin, there is no reason to think that these factors are less important than economic deterrence. Thus, the failure to offer an amnesty alternative before ramping up enforcement or increasing penalties could pose risks to voluntary compliance – potently greater than the relatively small risk of offering an amnesty alternative.
Next week we will apply what we’ve learned about amnesties to the IRS’s OVDP.
The views expressed in this blog are solely those of the National Taxpayer Advocate. The National Taxpayer Advocate presents an independent taxpayer perspective that does not necessarily reflect the position of the IRS, the Treasury Department, or the Office of Management and Budget.
