2012 Spring Report of the Auditor General of Canada

 

Performance audit reports

This report presents the results of a performance audit conducted by the Office of the Auditor General of Canada under the authority of the Auditor General Act.

A performance audit is an independent, objective, and systematic assessment of how well government is managing its activities, responsibilities, and resources. Audit topics are selected based on their significance. While the Office may comment on policy implementation in a performance audit, it does not comment on the merits of a policy.

Performance audits are planned, performed, and reported in accordance with professional auditing standards and Office policies. They are conducted by qualified auditors who

  • establish audit objectives and criteria for the assessment of performance;
  • gather the evidence necessary to assess performance against the criteria;
  • report both positive and negative findings;
  • conclude against the established audit objectives; and
  • make recommendations for improvement when there are significant differences between criteria and assessed performance.

Performance audits contribute to a public service that is ethical and effective and a government that is accountable to Parliament and Canadians.

Main Points

What we examined

The Canada Revenue Agency works to ensure that Canadians pay their required share of taxes and that the revenue base is protected. It is responsible for administering the Income Tax Act, which specifies when taxpayers are required to file a return. A non-filer is an individual, a corporation, or a trust who fails to file a tax return as required by legislation.

Under the Excise Tax Act, businesses that meet certain criteria are required to register for the goods and services tax/harmonized sales tax (GST/HST). Non-registrants are businesses that fail to comply with this requirement.

The Agency, through its Non-Filer/Non-Registrant (NF/NR) program, works to encourage individuals, corporations, and trusts to comply with the filing requirement and, in the case of businesses, with the GST/HST registration requirements. In the 2010–11 fiscal year, the NF/NR program’s salary budget was $39 million of the Agency’s total budget of $4.5 billion, and it employed 700 of the Agency’s approximately 39,000 employees. This relatively small program area generated $2.8 billion of additional taxes, interest, and penalties assessed in each of the 2009–10 and 2010–11 fiscal years.

We looked at what the Agency has done to address non-compliance with filing and registration requirements. The audit focused on how the Agency identifies non-filers and non-registrants and how it plans, monitors, and reports on its actions to improve compliance.

Audit work for this chapter was substantially completed on 5 January 2012. Further details on the conduct of the audit are in About the Audit at the end of the chapter.

Why it’s important

Income tax is the single most important source of government revenue. For the income tax system to operate fairly and effectively, it is important that all individuals and corporations file returns when they are required by law to do so, and that they pay the taxes they owe. Not doing so reduces the amount of money available for important government programs such as health care, education, and the environment. Similarly, businesses that are required to register for GST or HST, but do not, may have a negative impact on government revenues. The Agency has identified non-filers and non-registrants as high-priority issues to address, as part of its efforts to combat non-compliance, including the underground economy.

What we found

  • Because the Agency does not have the resources to pursue all non-filers, it has developed a risk-scoring model to identify those it will pursue. Agency analysis indicated that the files it chooses to pursue result in returns filed and taxes assessed. However, it has not tested its screening to determine whether the files it chooses not to pursue from the initial 2.5 to 3 million files identified by the matching process should in fact be pursued. In addition, two thirds of the files initially selected for pursuit in the field are later determined to have low potential, and work on them is discontinued. In other words, the Agency does not know if its risk-scoring process is as effective as it could be.
  • The Agency uses identification projects to find taxpayers who may be participating in the underground economy. The majority of projects completed were meeting or exceeding their targets for number of tax returns filed and amounts assessed.
  • The Agency’s planning process tends to repeat actions from one year to the next to identify non-filers and non-registrants. While it considers the cost of its processes to pursue these files when planning work, the Agency is not taking advantage of its research findings in its work plans in order to improve its results in achieving compliance.
  • Current performance indicators focus on routine program activities—for example, tax dollars and the number of taxpayers assessed—rather than on the longer-term impact of the program, in particular the Agency’s success at getting non-compliant taxpayers to change their behaviour and file their returns. Various audits and program evaluations over the years have recommended action to develop additional performance measures and risk management of the Non-Filer/Non-Registrant (NF/NR) program and to improve its use of internal and external information. The Agency has made limited progress in implementing many of these recommendations.

The Agency has responded. The Agency agrees with all of the recommendations. Its detailed responses follow the recommendations throughout the chapter.

Introduction

4.1 The mission of the Canada Revenue Agency is to administer tax, benefits, and related programs and to ensure taxpayer compliance, on behalf of governments across Canada. As Canada’s tax administrator, in addition to protecting Canada’s revenue base, the Agency’s primary goal is compliance—ensuring that taxpayers meet their obligations. Under the Canada Revenue Agency Act, the Agency is responsible for enforcing the Income Tax Act and the Excise Tax Act.

4.2 As part of fulfilling that responsibility, the Agency has created the Non-Filer/Non-Registrant (NF/NR) program. The program’s mandate is to encourage individuals, corporations, and trusts to comply with their requirements under the Income Tax Act to file tax returns as well as to encourage businesses to register for the goods and services tax/harmonized sales tax (GST/HST). The NF/NR program’s stated objective is

to develop fair, responsible and effective strategies to achieve high levels of filing and registration compliance commensurate with the level of funding while generating a high return on investment (fiscal impact and production [returns obtained]).

4.3 The Income Tax Act identifies situations that require individuals to file returns, which they must do to receive income tax refunds. Since many individuals are employees, and income tax is deducted at source from their pay, many receive refunds once they have filed their returns. Therefore, filing is in their best interest.

4.4 Filing an income tax return is the first step in the compliance continuum. The Agency assesses taxes once a taxpayer has filed a return; if necessary, it may also assess taxes even if a taxpayer has not filed a return. Once the return is filed, if taxes are owed, a different compliance risk may arise—that of whether a taxpayer can or will pay the amount owed. Taxpayers may not file their returns for a variety of reasons. Some forget, some procrastinate, and others do not file their returns or do not report their income in an effort to avoid tax. The Agency provides tools to taxpayers to help them comply and, when necessary, intervenes with a variety of actions.

4.5 An individual taxpayer is required to file a return when one of the following criteria is met:

  • Tax is payable.
  • The Agency requests or demands a return.
  • The taxpayer has elected to split pension income with a spouse or common-law partner.
  • The taxpayer wishes to receive the Working Income Tax Benefit.
  • The taxpayer has disposed of capital property or realized a taxable capital gain.
  • The federal government has overpaid Old Age Security or Employment Insurance benefits.
  • The taxpayer owes money to a Registered Retirement Savings Plan (RRSP) because of the Home Buyers’ Plan or the Lifelong Learning Plan.
  • The taxpayer is required to contribute to the Canada Pension Plan.

Individuals, corporations, or trusts who do not file a tax return as required are called “non-filers.”

4.6 The Agency administers benefits on behalf of other government departments as well as other levels of government. Often, these benefits are based on information in the income tax return. Unless individuals file a tax return, they cannot receive the benefits for which they may be eligible. The Agency aims to facilitate access to benefits, such as child benefits and the GST/HST credit, through its service delivery programs.

4.7 In addition, under the Excise Tax Act, businesses must register for the GST or HST if they provide goods or services equivalent to $30,000 or more each year. Charities must register after earning $250,000 in revenue; for public service bodies, such as schools, universities, and hospitals, the threshold is $50,000. Taxi drivers and certain non-resident businesses must register regardless of revenue. Businesses that should register but do not are called “non-registrants.”

4.8 Approximately 700 of the Agency’s 39,000 employees work in the NF/NR program. In the 2010–11 fiscal year, the program’s budget was $39 million of the Agency’s total budget of $4.5 billion. Compared to the other ways the Agency deals with compliance issues, such as audits and income tax collections, the program is relatively small. Despite its small size, the NF/NR program generated $2.8 billion of additional taxes, interest, and penalties in each of the 2009–10 and 2010–11 fiscal years.

4.9 The Agency has identified the pursuit of non-filers and non-registrants as a high priority in its strategy to combat non-compliance, including the underground economy. The Agency’s strategy incorporates

4.10 The Agency recently reviewed the delivery model for its NF/NR program. The review looked at operational processes, workload management, and resource use to identify improvements that would result in better returns on investments for the program. In March 2011, the Agency summarized its findings, and it is currently reviewing the results.

Focus of the audit

4.11 The objective of the audit was to determine whether the Canada Revenue Agency adequately addressed filing non-compliance (non-filers) and registration non-compliance (non-registrants).

4.12 We looked at how the Agency planned suitable methods to address both types of non-compliance, whether these plans incorporated past experience and research, and whether they took into account the cost of using these methods. We also looked at how the Agency validated its selection decisions and how well its projects addressed non-filing risks. Finally, we analyzed the way the Agency monitored and reported its actions toward non-filers and non-registrants. In all of these areas, we considered past audit and evaluation results, both internal and external, and recommendations the Agency had previously accepted.

4.13 We did not focus on the issue of whether taxpayers are paying the taxes that had been assessed.

4.14 More details about the audit objectives, scope, approach, and criteria are in About the Audit at the end of this chapter.

Observations and Recommendations

Identifying non-filers and non-registrants

4.15 Most taxpayers in Canada file their tax returns on time. Others file late before Agency intervention. The Agency will assess interest and penalties for late-filed returns. If a tax return has not been filed, the Agency will start the non-filer process.

4.16 There are two types of non-filers: known and unknown. Known non-filers are those that the Agency can identify. It does this by matching information slips (for example, T4 slips reporting employment income and T5 slips reporting investment income) filed by employers or other entities to tax returns filed by taxpayers.

4.17 Unknown non-filers are those the Agency cannot identify through information slips or other documents. They are often self-employed or operate within the underground economy. Because there is little or no automated information about unknown non-filers and they are harder to find, they are at higher risk of going undetected than known non-filers.

4.18 To start the identification process, Agency computer systems match information slips to personal tax returns to determine if a taxpayer has filed a return and if the amount reported is correct. Information slips that do not have matching tax returns identify potential non-filers. Each year, this process identifies 2.5 to 3 million individuals who are potential non-filers. Many are not determined to be non-filers, since they may not be obliged to file a return, or they may file their returns late. Another automated process identifies corporate non-filers by matching their corporate business numbers to corporate income tax returns.

4.19 Once this initial matching process has identified potential non-filers, the Agency applies a risk-scoring process to select which taxpayers to pursue. The initial process for pursuing them is automated (Exhibit 4.1).

  • The Agency sends letters, reminding them to file their outstanding returns.
  • If the Agency does not receive a return within 40 days, it may send a second “request to file” letter or it may send a demand to file.
  • Finally, if it has not received a return, the Agency may transfer the file to a tax centre or a tax services office where a staff member will begin a manual follow-up process. In some cases, the process ends with the prosecution of a taxpayer, resulting in a fine or jail term or both.

Exhibit 4.1—Process for pursuing individual non-filers, using the 2009 tax year as an example

Flow chart showing the process for pursuing individual non-filers in 2009

[text version]

*Total taxes assessed does not match the annual report, because this is the T1 process for individuals and does not include corporations, trusts, or GST/HST.

The Agency has validated its process for selecting files but not for rejecting files

4.20 We looked at whether the Agency validates (tests the validity of) its screening process for identifying non-filers, and whether it knows if the process was appropriate in identifying files to pursue. It is important to validate this process, because it could include (“screen in”) individuals who owe little or no tax or who are due a refund (low potential), and exclude (“screen out”) individuals who owe high amounts of taxes (high potential). For 2010 and 2009, we examined whether the Agency validated its screening process to confirm that it was excluding low potential files and including high potential files.

4.21 According to guidance from the Organisation for Economic Co-operation and Development (OECD) on how to evaluate the effectiveness of an organization’s strategies to encourage compliance, it is important to validate the process that the organization uses to select cases. This process is described in Exhibit 4.2. A flawed selection process could result in high costs and poor results, whereas sound selection processes result in low costs and positive results.

Exhibit 4.2—The Compliance Risk Management Process

Flow chart showing the Compliance Risk Management Process

[text version]

Source: Organisation for Economic Co-operation and Development (OECD), Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies

4.22 The individual (T1) risk-scoring process is designed to estimate the amount of tax that may be owed, on a case-by-case basis, using information available to the T1 matching process. The accuracy of that estimate can be validated only when the taxpayer files a tax return, and the actual amount of tax assessed can be compared to the estimated amount. One of the Agency’s challenges is that it does not know how much tax, if any, is owed until it receives a return.

4.23 We found that the Agency’s analysis of its T1 file-selection process indicated it has some ability to predict the potential of the files from non-filers who were assessed after having been identified in the matching process. Risk scoring allows the Agency to prioritize returns with high potential. For the majority of cases, particularly when the file-selection process did not select a case for enforcement action, the Agency does not know how well its risk-scoring process works.

4.24 The Agency’s research does not provide a reliable estimate of the tax potential of the files that the Agency decided not to pursue. To get a reliable estimate of the potential of those files, the Agency could randomly select a small number of accounts from the 2.5 to 3 million individuals that are identified annually as potential non-filers. It could then develop baseline measures to determine the effectiveness of its risk-scoring process. The Agency has developed a new risk-scoring model for personal tax returns and has run pilot projects to test its effectiveness. This new model also does not address the potential in the files the Agency decided not to pursue. (Our recommendation is at paragraph 4.29.)

The Agency needs to validate its process for registering potential non-registrants

4.25 The Agency’s process for identifying potential GST/HST non-registrants involves matches, leads, projects, and previously identified non-filers. The Agency focuses on businesses that it believes ought to be registered but are not.

4.26 Its work involving non-registrants uses approximately five percent of the NF/NR program resources. Agency guidance indicates that when an individual is being reviewed as a non-filer, staff must consider and act on GST/HST registration issues at the same time. We were unable to determine if this was being done consistently.

4.27 We looked at businesses that were identified as potential non-registrants between 1 April 2009 and 31 March 2011, to determine whether the Agency validated its methods for pursuing non-registrants. We found two areas of concern:

  • The Agency cannot manage the volume of potential non-registrants that its matching process identifies. Each year, this automated system identifies around 185,000 potential non-registrants and sends their files for review. The Agency has the capacity to review about half of these files. This means that a large number of potential non-registrants may not be reviewed on a timely basis.
  • We also noted a file-selection issue. Given the volume of potential non-registrants identified, it is important for the Agency to select files according to their tax potential. While the current method includes elements of risk assessment, the Agency has not completed analysis that would allow it to determine the effectiveness of its file-selection method.

4.28 In summary, the Agency does not know if its selection methods to pursue potential GST/HST non-registrants are effective. Given the Agency’s limited resources, it is important to prioritize files to pursue.

4.29 Recommendation. The Agency should determine the effectiveness of both its file selection and rejection processes for the total population of potential non-filers and non-registrants.

The Agency’s response. Agreed. The Agency will develop a methodology for determining the effectiveness of the file selection and rejection processes. It is expected that this methodology will be applied to the non-filer/non-registrant population in 2013–14; and, in 2014–15, the file selection and rejection processes will be adjusted, as required, based on the evaluation of the testing results from previous years.

File selection for manual pursuit needs improvement

4.30 Once taxpayer files have gone through the automated process, and taxpayers have still not filed a return, the files are transferred to tax centres and tax services offices (collectively called “the field”) for resolution. Agency field staff will attempt to find current addresses for taxpayers, using the Internet and government databases. They may use a variety of methods—including phone calls, personalized letters, personal visits, and taxpayer lifestyle analyses—to achieve filing compliance.

4.31 We noted that the Agency has been conducting research to find the optimal enforcement strategy for individual non-filers. Using data about filing history, the Agency has been trying to identify the non-filers that are highest potential and, as such, should be sent directly to the field, skipping the step of sending an automated letter. The Agency has piloted a new model based on this research, and the initial results are encouraging.

4.32 When field staff pursue a file manually, the potential results are

  • the taxpayer files a return;
  • the Agency assesses the tax payable for the outstanding tax year, under subsection 152(7) of the Income Tax Act; or
  • the Agency discontinues pursuit of a return, because it determines that the file lacks revenue potential or that no return is required.

In some cases, the Agency may apply penalties or prosecute taxpayers. When it has successfully prosecuted a taxpayer, it issues a press release to inform the public of the consequences of not filing a return or not reporting income.

4.33 Guidance from the OECD states that tax authorities should ensure that they choose the correct files to pursue, and that they consider the cost of getting a non-filer to comply. We looked at

  • how the Agency selects files to pursue manually,
  • whether it evaluated the effectiveness of manual interventions, and
  • how it addressed any shortcomings.

4.34 Tax centres and tax services offices do not have the capacity to pursue every potential file transferred from the automated process. The Agency has identified an excessive inventory of non-filer cases at tax services offices as a significant challenge, so choosing which files to pursue is important.

4.35 Field staff can use the amount of tax potential that the Agency estimates to select the potential non-filers that it pursues. They also have a number of other tools to help choose files. However, an Agency analysis, conducted as part of an NF/NR program renewal exercise, determined that the tax potential calculated by the Agency does not meet the field officers’ needs. In addition, the analysis found the procedures used to select files at the tax services offices varied, as did the quality of the file selection. About two thirds of the files selected for manual pursuit ended up being discontinued, partly because the field staff determined low tax potential after they started to work on the file. When a file is discontinued, a decision is made to stop work on it, even if the taxpayer has not filed a return.

4.36 A stronger risk assessment system would allow staff to prioritize and focus on files with high potential. It would also provide the Agency with the assurance that it can eliminate more low-potential files from its inventory and allow it to use its resources more efficiently and effectively.

4.37 Recommendation. The Canada Revenue Agency should implement a robust risk assessment system to prioritize its selection of files.

The Agency’s response. Agreed. As noted in this report, the Agency’s current effort towards improving file selection through data analysis is encouraging. The Agency is committed to developing and applying new risk assessment methodologies and business intelligence solutions to improve the selection of both low- and high-complexity files. The Agency will continue this work in 2012–13, so that a more robust risk assessment process can be implemented by 2015–16.

The Agency does not pursue all non-filing corporations

4.38 The Income Tax Act states that, unlike individuals, all corporations are required to file a tax return, whether or not they owe taxes. The Agency matches corporation business numbers to corporate tax returns. If there is no corresponding return, the corporation is a non-filer.

4.39 We looked at whether the Agency ranks corporations for pursuit and excludes corporations based on predetermined criteria. We found that corporations are assigned a risk rating based on the estimated tax owing. However, the rating is for information purposes only; the Agency does not use that rating to select files to pursue.

4.40 The Agency’s administrative position is that it will not pursue certain types of corporations—such as incorporated municipalities and non-profit organizations—because they are exempt from income tax. Without tax returns for certain types of corporations, it is difficult for the Agency to determine if they still meet the conditions for tax exempt status.

Identification projects find non-filers for whom the Agency lacks information

4.41 The Agency has identified the underground economy as one of the biggest risks it faces. The combination of the declining economy and a growing number of self-employed individuals increases the risk of revenues going unreported and under-reported and of a reduced tax base.

4.42 According to the Agency, identification projects are one of the ways that it identifies and addresses filing and registration non-compliance that is missed, or is not prioritized, by the automated process.

4.43 These projects may be based on various factors, such as local knowledge and experience, regional concerns, and national issues. The focus of identification projects tends to be on self-employed individuals who carry on a business, because the Agency’s automated systems address potential non-filers who receive information slips. For example, the Agency’s automated matching process would not identify a contractor who offers a low “cash-only” price for renovations.

4.44 Often, the research to identify which businesses to look at begins at a tax services office, where staff review a large number of businesses for filing compliance. As sources of information, they use the Internet and directory and database searches, as well as projects conducted at other field offices. They also seek information from other levels of government.

4.45 The Agency’s 2009 Risk Action Plan indicated the need for a more horizontal approach to enable a coordinated corporate strategy. Furthermore, it indicated that more research, testing, and analysis would be critical to help the Agency focus on high-risk industries and cases. The Agency’s analysis of completed projects was to be finished by early 2011, but it remains outstanding. In addition, the Agency was to gather intelligence data, as part of its identification projects. Some work has been started, but more is still required to complete the intelligence gathering.

4.46 We looked at identification projects completed between April 2009 and March 2011 to determine whether the Agency had adequate methods in place to find and compel non-filers to file their returns.

Case Study 4.1—Examples of projects

In 2010, the Agency started a project that required a number of cities to provide a list of individuals and companies that had received revenues other than salaries. This would allow the Agency to check that the recipients’ obligations under the Income Tax Act were being met. The information received from the cities would be matched against Agency databases. This would allow non-filers and non-registrants to be identified and pursued by field officers to file their returns and register their businesses.

Agency staff collected advertisements for home renovations and construction from publicly available sources, such as telephone directories, newspaper listings, flyers, and websites. This information was used to see whether those businesses filed returns and were registered for GST/HST.

4.47 We found that the Agency undertook some projects during the audit period, but its choice of projects was not based on a comprehensive analysis of the types of non-filers it might be missing. The Agency’s own analysis concludes project activity is ad hoc and has not benefited from a more systematic approach.

4.48 We examined whether the identification projects achieved the desired results. We also considered whether the Agency cancelled or revised projects that were not meeting intended targets. We found that the majority of completed projects met or exceeded their desired results. However, we noted that one region was not consistently meeting its targets. The Agency has made efforts to address these issues at the regional level.

4.49 Agency guidance on projects suggests that, when compared to other types of field work, identification projects find individuals with higher average amounts of taxes assessed. We looked at identification projects and other field work, by comparing the ratio of salaries to assessed tax dollars, and we found no difference. This suggests that identification projects are no more successful in raising tax revenue than any other type of field work. However, they are important to the success of the program, because they find the unknown non-filers and send an important compliance message to taxpayers.

Tax services offices rarely follow the new project approval process

4.50 Tax services offices are expected to spend a portion of their local budget on project work each year. They may also request additional funding for special compliance projects. In both of the years under audit, the Agency budgeted around 15 percent of its field salary dollars for identification projects.

4.51 The Agency developed a project approval process, with a set of rules, for submitting potential projects for approval. It developed the process to ensure that it retained knowledge and experience obtained during identification projects, and to encourage more rigour at the local level when projects are being selected. For example, the process identifies the need for

  • establishing a business case,
  • performing sample testing,
  • setting objectives,
  • establishing a strategy,
  • mapping the process, and
  • having the project approved.

4.52 We reviewed the applications for project approval by Headquarters, to determine whether Agency staff followed the new rules, which came into effect in July 2010. We found that most projects that were submitted lacked important information. The most common omissions were business case and sample testing, as well as approvals from local managers. These missing pieces of information are critical to demonstrate the need for the project as well as to assist in measuring its success. Without sample testing, it is difficult to determine the scope of the non-compliance staff are trying to address. Without approvals at the local level, managers may be unaware of any issues or may disagree with the scope and intent of the project.

Planning work

4.53 With a diverse population to identify and detect, planning becomes a key factor in the process of identifying non-filers. Assessing what worked in the past, incorporating lessons learned, and considering the results of research all play a part in developing a business plan. In addition, certain business sectors or geographic areas may be more problematic than others. The Agency needs to consider what to measure when developing a plan. To ensure success, the Agency must set appropriate standards in advance.

4.54 The overall objective of the Non-Filer/Non-Registrant (NF/NR) program is to achieve high levels of filing compliance and registration. Managers need to know whether the Agency’s programs and choices lead to desired changes in taxpayer behaviour, thus achieving this objective.

The annual planning process focuses on production targets

4.55 Guidance from the Organisation for Economic Co-operation and Development (OECD) refers to the importance of analyzing compliance behaviour, determining treatment strategies, and planning and implementing those strategies. Agency documents have identified non-filers and the underground economy as among the highest risks the Agency must address.

4.56 We reviewed planning documents for the period between 1 April 2009 and 31 March 2011. We did not find an integrated plan that outlines the work of the NF/NR program as a whole. We saw budget documents, output targets, and emails. We found that, each year, the Agency used the same approach, which focused on obtaining a set number of tax returns or registrations and a target amount of assessed taxes. Because the Agency did not prepare and evaluate an integrated plan, it did not review the program’s overall status.

4.57 During the years under audit, the internal reports on the automated program production had an error in how production results were calculated. The Agency has informed us that controls are now in place to make sure that these reports are reliable.

4.58 In our 2005 audit (November Report of the Auditor General of Canada, Chapter 3, Canada Revenue Agency—Verifying Income Tax Returns of Individuals and Trusts), we recommended that the Agency develop and implement a strategy to better use certain types of information returns, such as documents that report the sale of shares. This could be another useful source of information in the matching process for identifying potential non-filers. The Agency agreed with this recommendation but has not yet fully implemented it.

4.59 The NF/NR program has undertaken a long-term initiative to improve its operations. This initiative is based on an internal review of the Agency’s offices across Canada, which is the foundation for change in areas such as risk management. It is an example of positive steps the Agency is taking, through the use of research, to improve its planning and program operation. This initiative is in the early stages of implementation, and it does not address past recommendations or other research agenda issues.

The Agency considers the cost of achieving filing compliance when setting objectives

4.60 Risk management involves balancing risk with mitigation and takes into account the concept of diminishing returns. This requires the Agency to focus on obtaining compliance in cases where there is greater tax potential, and should consider what resources are necessary to achieve that level of compliance. We note, however, that the NF/NR program also recognizes the need to send a deterrent message to certain sectors, to encourage compliance.

4.61 Cost-benefit analysis aids in selecting the appropriate approach to workload and general planning. We looked at the way the Agency considers this analysis when it sets objectives and again, throughout the year, when priorities or budgets change.

4.62 We looked at Agency planning documents and found that, each year, it conducted a cost-benefit analysis before it began the automated and manual processes to assign non-filing and non-registration files. In fact, this concept was the foundation for many of the annual work plans.

4.63 We reviewed three cases in which additional funds were injected into the program partway through the fiscal year. Agency documentation demonstrated that a clear cost-benefit analysis was conducted in only one of the three cases.

Measuring results and making improvements

4.64 The Canada Revenue Agency’s risk management policy, guidance from the Organisation for Economic Co-operation and Development (OECD), and non-filer/non-registrant (NF/NR) operations manuals highlight the importance of regularly learning from past experience, relevant research findings, and action plans for the purpose of continuous improvement.

4.65 In addition to our 2005 audit, we conducted an audit of the NF/NR program in 1994 (Report of the Auditor General, Chapter 31, Revenue Canada—Ensuring Fairness of the Income Tax System: Detection of Non-Filers and Special Investigations).

4.66 Some of the Agency’s internal audit and evaluation reports also addressed the NF/NR program, including the following:

  • 2005—Contract Payments Reporting System Program Evaluation Study
  • 2007—Non-Filer/Non-Registrant Program—Internal Audit Report
  • 2008—GST/HST Registration Compliance Evaluation Study
Performance indicators measure only routine program activities

4.67 The OECD notes that revenue collection bodies (organizations that administer and enforce national tax laws) have traditionally measured their performance through the day-to-day activities that result from the various work streams they administer. However, while such measures represent an important part of revenue bodies’ reporting systems and help with their day-to-day management, they do not offer insights into the impact of their activities. Positive trends are sometimes assumed to reflect improvements in taxpayers’ compliance, but this may not be true. The Compliance Risk Management Process (Exhibit 4.2) links taxpayer behaviour to strategies. Strengthened performance monitoring would incorporate this concept and help the Agency to make this link.

4.68 Previous audit and evaluation recommendations, which the Agency accepted, included the need for performance measures that provide more meaningful information. In our 1994 audit report on non-filers, we found that the focus on files obtained and taxes assessed did not provide information on whether taxpayer behaviour had improved. The Agency’s 2008 GST Registration Compliance Study found that the yearly production targets that headquarters set for the NR workload

do not demonstrate the extent to which the program is having the desired effect of addressing registration non-compliance or identifying the relative risk posed by the non-registrant population.

4.69 The Agency’s 2007 Internal Audit Report concluded that its existing program performance measures are sufficient to assess routine program activities. However, according to the internal audit, the NF/NR program needed to develop performance measures and indicators to accurately report its outcomes. Doing so would provide a complete performance picture of its portfolio to support the Agency’s strategic decision-making.

4.70 We reviewed published performance measures, as well as those available internally to Agency staff, for the two years under audit. We looked for the trend analysis and behavioural impact that were being incorporated into management information.

4.71 We found that the Agency uses the following indicators to assess the program

  • amount of taxes assessed,
  • amount spent on salaries,
  • number of returns filed, and
  • number of GST/HST registrations obtained.

For the two years under audit, the NF/NR program exceeded its target of $2.4 billion for taxes assessed; it assessed $2.8 billion in each of the 2009–10 and 2010–11 fiscal years.

4.72 The Agency also reports the rate of voluntary compliance in its Annual Report to Parliament, as a measure of the overall level of filing compliance in Canada. Individual filing compliance is quite high; in each of the years under audit, 92.8 percent of returns were filed on time, exceeding the target of 90 percent. Corporations are less compliant; the target of filing 90 percent of returns on time has not been met for a number of years (for example, 85.1 percent were filed on time for the 2010–11 fiscal year and 85.5 percent for 2009–10). (Exhibit 4.3)

Exhibit 4.3—The Agency’s target and compliance rates

Line graph showing the compliance rates for registering businesses and filing tax returns

[text version]

Source: Canada Revenue Agency’s annual reports to Parliament, 2005–06 to 2010–11

4.73 In its annual report to Parliament for the 2008–09 fiscal year, the Agency committed to reporting the percentage of taxpayers that it identified as having outstanding tax returns after the Agency initiated an action. It did not report these percentages for the 2009–10 or 2010–11 fiscal years.

4.74 The Agency does report annually on GST registration compliance. However, it reports this compliance as a percentage based on GST filing compliance, since the actual number is unknown. In our opinion, this is not a meaningful measure because there is no relation between those who file their GST returns and those who are not registered for GST. The Agency’s annual report for the 2008–09 fiscal year indicated that the next report would include a percentage measure of the effectiveness of the actions that the Agency took to identify businesses that are required to register for the GST/HST. No measure was included in subsequent annual reports.

4.75 The Agency continues to struggle to develop measures that demonstrate its effectiveness in addressing filing or registration compliance.

4.76 Recommendation. The Canada Revenue Agency should follow through on its previous commitments to develop meaningful performance measures and indicators.

The Agency’s response. Agreed. Performance indicators should evolve and adapt to changes in the program. Existing indicators continue to provide relevant information on program activities, expenditures, and results. New indicators, however, can provide even more meaningful information if the systems to collect and report on the data can be created. System changes will be governed by Agency and Government of Canada funding and expenditure policies. Beginning in 2012–13, the Agency will, in line with the Auditor General’s recommendation, review the relevant performance indicators to ensure they further support program reporting requirements.

The Agency has not integrated research results into planning to improve the program

4.77 The Agency has been conducting research on non-filer behaviour for many years. Understanding why certain taxpayers do not file their returns is essential to developing effective action plans. The Agency has identified some actions it could take to increase non-filer compliance, such as changes to the contents of letters sent to non-filers (to be implemented for the 2012–13 fiscal year) and standard reminders from Agency staff when they respond to taxpayer queries about their obligations to file tax returns.

4.78 We reviewed the Agency’s use of “data mining” (sifting through large amounts of data to find useful information) to improve the way it selects cases for non-filer enforcement action. One research project looked at the accuracy of the risk assessment when compared to the actual amount a non-filer declared on a tax return, regardless of Agency intervention. Overall, data mining shows promise.

4.79 In 1994, we concluded that to conduct compliance research and generate more leads, which would help it to find non-filers who are not on the tax roll, the Agency needed to continue expanding its use of internal and external sources of information. We also concluded that the Agency needed to evaluate how effective it was at convincing non-filers to file tax returns.

4.80 We looked at whether Agency enforcement action has an effect on voluntary compliance. While the Agency has not conducted any research about the effect of its NF/NR program on the Canadian public, its research on the non-filer population indicates that individuals who were subject to enforcement action were more likely to file their returns in future years. However, the impact fades over time, and future voluntary compliance is lowest among taxpayers who required more severe intervention, such as prosecution. This measure could be used to track the impact of the NF/NR program.

4.81 We looked at Agency research about non-filers and non-registrants to determine whether any of the research findings had been integrated into the program planning during the period we audited.

4.82 We found the Agency had a multi-year plan to conduct research, but it was not always clear how this plan fit into the NF/NR program’s overall plans. We saw no modifications to the annual planning process to incorporate research findings; the process was primarily a repeat of the previous year’s actions. An Agency’s program evaluation of GST/HST Registration Compliance, in 2008, noted similar concerns about risk and research.

4.83 Recommendation. The Canada Revenue Agency should include its research findings in its planning and reporting processes, to effectively manage the risk of non-filers and non-registrants.

The Agency’s response. Agreed. The Agency has made significant progress in conducting research and gathering business intelligence through the acquisition of data, technology, and the Agency research agenda. The Agency recognizes that there are opportunities to improve how this intelligence is incorporated into the planning and reporting processes. By 2013–14, the Agency will develop and implement a formal process to achieve this goal.

Conclusion

4.84 The Canada Revenue Agency has adequately addressed non-compliance concerning the filing of tax returns and GST/HST registration, although we identified some areas for improvement. In particular, it does not know the impact of its actions on taxpayer compliance.

4.85 The Agency has identified suitable methods to address filing and registration non-compliance. The planning process has some weaknesses, because it is not integrated with research and audit findings and tends to rely on past methods, although it does include the cost of applying particular methods.

4.86 The Agency uses a risk-scoring model and data-mining research to choose files to pursue through an automated process, but it has no way of knowing whether it excludes the right files. Field staff use various methods to choose files, in addition to the automated risk-scoring system that does not always identify the files with the greatest potential for obtaining a tax return and assessing taxes.

4.87 Identification projects successfully find non-filers about whom the Agency has no previous information.

4.88 The Agency’s performance measures do not provide information about the impact of the Non-Filer/Non-Registrant (NF/NR) program or on whether the program met its objective of achieving high levels of compliance. Rather, they measure the results of routine program activities. In addition, the Agency has made limited progress in implementing our previous recommendations or on those from its own internal audits and program evaluations. Overall, the NF/NR program works, but there is room for improvement.

About the Audit

All of the audit work in this chapter was conducted in accordance with the standards for assurance engagements set by The Canadian Institute of Chartered Accountants. While the Office adopts these standards as the minimum requirement for our audits, we also draw upon the standards and practices of other disciplines.

Objectives

The overall audit objective was to determine whether the Canada Revenue Agency adequately addressed filing non-compliance (non-filers) and registration non-compliance (non-registrants). For the purposes of this audit, we defined “adequately” and “addressed.” Adequately meant in a measureable way, with targets and indicators reflecting the selection and identification process, and whether the Agency knew at each stage (at Headquarters and in tax services offices and tax centres) if the process was effective. Addressed meant concentrated on or directed its activities toward.

We conducted our audit with the following lines of inquiry, each with its own sub-objective:

  • Identifying, selecting, and pursuing non-filers and non-registrants. To determine whether the approach to addressing non-filers and non-registrants was implemented as designed, at Headquarters and in the tax centres and tax services offices.
  • Planning. To determine whether the Agency had designed an adequate approach to addressing non-filers and non-registrants.
  • Monitoring and reporting actions to address non-filers and non-registrants. To determine whether the Agency monitored and reported its actions in addressing the inventory of non-filers and non-registrants.

Scope and approach

The audit focused on the program areas concerning non-filers and non-registrants with the Taxpayer Services and Debt Management Branch. In particular, we looked at how they planned the work, used previous audit findings and research, and selected files, according to a risk assessment. We also considered how the Agency monitored and reported on the risk of non-filing and non-registration.

During our audit, we requested and reviewed documents and reports from the Agency. We also analyzed data relating to identification projects and file selection. Our review of applications for identification projects included the random sampling of 40 files over the two-year period of our audit. We chose our sample in accordance with statistical sampling methodology, to provide a high level of assurance that the Agency was following procedures when approving projects. A sample size of 34 is sufficient to project the entire population of 62 with an expected error rate of 0 percent, a confidence interval of +10 percent, and a confidence level of 90 percent.

To confirm our understanding of the information we received, we interviewed Agency staff at various levels and locations. We visited eight tax services offices, three tax centres, and one regional office and conducted interviews with regional program advisors, technical advisors, managers, field officers, team leaders, and assistant directors. At the Agency’s headquarters, we met staff in the Enterprise Risk Management Branch, Strategy and Integration Branch, and the Taxpayer Services and Debt Management Branch.

We did not examine Agency activities related to

  • payment non-compliance,
  • non-filing of GST (as opposed to non-registration), and
  • criminal activities of the underground economy.

Criteria

To determine whether the Agency had designed an adequate approach to address non-filers and non-registrants, we used the following criteria:
Criteria Sources

The Agency’s plan identifies suitable methods to compel filing and registration compliance.

  • Corporate Risk Inventory Update, Canada Revenue Agency (CRA), 2010
  • Corporate Risk Inventory, CRA, 2009
  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, Organisation of Economic Co-operation and Development (OECD)

The plan incorporates past experience, relevant research findings, and action plans for the purposes of continuous improvement.

  • Enterprise Risk Management (ERM) Policy, CRA, 2006
  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD
  • Compliance Review II, CRA
  • Non-Filer/Non-Registrant (NF/NR) Tax Services Office (TSO) Operations Manual

The Agency’s plan compares the cost of applying the methods designed to mitigate the problem of non-filers/non-registrants with the benefit.

  • Enterprise Risk Management (ERM) Policy, CRA, 2006
  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD
  • Corporate Business Plan 2009–2010 to 2011–2012, CRA

The Agency implements the work plan.

  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD
  • Canada Revenue Agency Act

The Agency identifies and selects inventory in a manner that takes into account the cost to achieve filing compliance.

  • Enterprise Risk Management (ERM) Policy, CRA, 2006
  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD
  • Corporate Business Plan 2009–2010 to 2011–2012, CRA

The Agency validates its method to select inventory.

  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD

The Agency has adequate methods in place to compel non-filers to file their returns and non-registrants to register their businesses.

  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD
  • Income Tax Act
  • Excise Tax Act
  • Canada Revenue Agency Act

The Agency has measures to assess that its actions achieve the desired results as defined by the program objective.

  • Corporate Risk Inventory Update, CRA, 2010
  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD
  • NF/NR TSO Operations Manual

The Agency reports whether its actions achieve the desired results as defined by the program objective.

  • Canada Revenue Agency Act
  • Auditor General Act
  • NF/NR TSO Operations Manual
  • Guidance Note: Evaluating the effectiveness of compliance risk treatment strategies, OECD

Management reviewed and accepted the suitability of the criteria used in the audit.

Period covered by the audit

The audit covered the period from April 2009 to March 2011. Audit work for this chapter was substantially completed on 5 January 2012.

Audit team

Assistant Auditor General: Marian McMahon
Principal: Vicki Plant
Director: Heather Miller

Jeff Graham
Suzanne Moorhead
Rodney Newcombe
Stuart Smith
Sarah Winton

For information, please contact Communications at 613–995-3708 or 1-888–761-5953 (toll-free).

Appendix—List of recommendations

The following is a list of recommendations found in Chapter 4. The number in front of the recommendation indicates the paragraph where it appears in the chapter. The numbers in parentheses indicate the paragraphs where the topic is discussed.

Recommendation Response
Identifying non-filers and non-registrants

4.29 The Agency should determine the effectiveness of both its file selection and rejection processes for the total population of potential non-filers and non-registrants. (4.20–4.28)

Agreed. The Agency will develop a methodology for determining the effectiveness of the file selection and rejection processes. It is expected that this methodology will be applied to the non-filer/non-registrant population in 2013–14; and, in 2014–15, the file selection and rejection processes will be adjusted, as required, based on the evaluation of the testing results from previous years.

4.37 The Canada Revenue Agency should implement a robust risk assessment system to prioritize its selection of files. (4.30–4.36)

Agreed. As noted in this report, the Agency’s current effort towards improving file selection through data analysis is encouraging. The Agency is committed to developing and applying new risk assessment methodologies and business intelligence solutions to improve the selection of both low- and high-complexity files. The Agency will continue this work in 2012–13, so that a more robust risk assessment process can be implemented by 2015–16.

Measuring results and making improvements

4.76 The Canada Revenue Agency should follow through on its previous commitments to develop meaningful performance measures and indicators. (4.67–4.75)

Agreed. Performance indicators should evolve and adapt to changes in the program. Existing indicators continue to provide relevant information on program activities, expenditures, and results. New indicators, however, can provide even more meaningful information if the systems to collect and report on the data can be created. System changes will be governed by Agency and Government of Canada funding and expenditure policies. Beginning in 2012–13, the Agency will, in line with the Auditor General’s recommendation, review the relevant performance indicators to ensure they further support program reporting requirements.

4.83 The Canada Revenue Agency should include its research findings in its planning and reporting processes, to effectively manage the risk of non-filers and non-registrants. (4.77–4.82)

Agreed. The Agency has made significant progress in conducting research and gathering business intelligence through the acquisition of data, technology, and the Agency research agenda. The Agency recognizes that there are opportunities to improve how this intelligence is incorporated into the planning and reporting processes. By 2013–14, the Agency will develop and implement a formal process to achieve this goal.

 


Definitions:

Taxpayer—According to the Income Tax Act, any person who resides in Canada, whether or not they are liable to pay tax. For tax purposes, a person may be an individual, a corporation, or a trust. (Return)

Underground economy—Typically involves commercial activity that is unreported for tax purposes. Underground economic activity is particularly prevalent in industry sectors where cash transactions are common, such as hospitality, automotive repairs, and construction, including home renovation.

Source: Canada Revenue Agency (Return)

Organisation for Economic Co-operation and Development (OECD)—An organization consisting of member countries. It conducts research and creates agreements, standards, and recommendations to end poverty through economic growth and financial stability. One of the OECD’s focus areas is taxation. (Return)

Demand to file—The formal notification of a taxpayer’s requirement to file an income tax return—based on subsection 150(2) of the Income Tax Act. (Return)

Subsection 152(7) of the Income Tax Act—A provision of the Act that gives the Agency the authority to assess the tax payable where no return has been filed by a taxpayer. Agency staff calculate an amount they believe is an accurate estimate of a taxpayer’s income. The taxpayer is able to revise the amount of tax assessed by filing a corrected income tax return. (Return)

 

PDF Versions

To access the Portable Document Format (PDF) version you must have a PDF reader installed. If you do not already have such a reader, there are numerous PDF readers available for free download or for purchase on the Internet: