A Practical Guide to Investment Tax Incentives

4. Monitoring

Abstract

Monitoring key metrics related to tax incentives – including on take-up, performance of beneficiaries and the value of the tax benefit – is key to confirm compliance and inform policy evaluation. Effective monitoring requires building capacity to track how incentives are used in practice, improving data infrastructure and data sharing across government agencies, and easing any compliance burdens on firms.

Monitoring involves tracking how incentives are used in practice, supporting tax compliance, and collecting quality data for policy evaluation. Data on compliance, take-up, characteristics and performance of beneficiaries, and value of the tax benefit are key understanding if tax incentives are contributing to stated goals and at what costs. Monitoring compliance with the terms of the incentive promotes effective application of tax laws. Data collection can also inform evaluations of the policy (see Evaluation). Monitoring can be carried out by multiple government agencies, requiring clearly assigned responsibilities and established procedures for coordination and data sharing.

Policymakers face a trade-off between collecting more detailed data for better evaluation, and administrative and compliance costs for governments and taxpayers. Data needs should be clearly assessed and balanced with compliance costs for tax incentive beneficiaries; reporting requirements or procedures that are too complex can disadvantage smaller or less established firms. Governments should consider data infrastructure and data analysis capacity when considering additional monitoring. Existing data sources should be leveraged where possible, though improving monitoring capacity in the medium- to long-term can bring benefits. Table 4.1 sets out a practical framework to improve monitoring of tax incentive use in practice.

Table 4.1. Key steps for improving tax incentive monitoring

Key steps		Recommended Actions
1. Determine what to monitor
	Compliance1	Track compliance with tax incentive conditions, informed by a compliance risk assessment, to minimise abuse, ensure proper use of public funds and more effective policy
	Take-up	Track which firms claim the tax incentive and their characteristics to understand how the policy works in practice
	Cost2	Report on itemised cost per incentive for tax expenditure reports (see Evaluation)
	Performance & potential outcomes	Monitor performance of beneficiaries (e.g. employment, local sourcing), non-beneficiaries, and other outcomes that might be linked to the policy to provide input for evaluations
2. Establish process3
	How data is collected	Assess if most relevant information is already collected in tax returns or other reporting, and what coverage gaps exist
	Data format4	Aim for sufficient disaggregation, standard data formats and digital datasets
	Who collects	Provide tax authorities access to all relevant monitoring data, consider how other agencies (e.g. investment promotion agencies) can support monitoring
	How data secured & shared5	Ensure procedures and infrastructure in place to allow secure data sharing
3. Build capacity
	Consider most appropriate means to expand data collection6
	Explore how to draw from other data sources to fill monitoring gaps

4.1. Determine what to monitor

The extent of monitoring will depend on the capacity of the jurisdiction, with higher-capacity jurisdictions likely to be able to gather more sophisticated data. At a minimum, monitoring should establish whether firms are complying with the terms of the incentive. Policymakers can also track metrics on take-up, to inform evaluations of whether tax incentives are successful at supporting their stated policy goal. Practitioners with sufficient capacity can also gather data on cost, firm performance and other potential outcomes.

4.1.1. Compliance

Tracking compliance with the terms of the incentive is essential for policy effectiveness, to minimise abuse and promote tax compliance and ensure proper use of public funds. Even in cases of capacity constraints, tax officials can seek to identify which incentives might have the highest risk of non-compliance and the highest revenue consequences, in order to best target resources (OECD, 2004[1]).7 Firms might be non-compliant because of efforts to exploit “grey” areas of the law, or beneficiaries might not reach outcome targets, which may reflect shortfalls in the design of the incentive. For example, loss-making firms that receive tax incentives might carry over benefits past legal requirements to do so. Table 4.2 provides examples of tax incentive non-compliance and suggested improvements to design or implementation to minimise mistakes or policy misuse. Improving tax incentive compliance should be accompanied by wider tax compliance and enforcement capacity, as well as legal statues that set sanctions on non-compliant firms to recover revenue forgone and discourage future or other policy abuse.8

Table 4.2. Identifying reasons for non-compliance can support policy reform

	Example	Potential improvement
Required outcomes not met	Outcome (e.g., job creation) were not reached or sustained	Reflect whether outcome conditions can be replaced by better design; refine outcome conditions (if not measurable, or too ambitious given other constraints); consider if targeting firms able to reach outcomes
Different interpretations of requirements	Firm claims expenses that government deems are not eligible, or argues it has met conditions that are open to interpretation (e.g., 51% of activities must be related to high technology)	Set clear and specific definitions of qualifying expenditure, income, and other requirements
Mistakes	Errors in reporting or claiming9, under-reporting	Clarify procedures, guides to support implementation
Abuse	Deliberate misuse of the policy, e.g. inflating eligible expenditure, applying the incentive to ineligible goods or services, claiming operations in locations (e.g. Economic Zones) that did not take place, carry-forward of losses past limits, tricking fulfillment of outcome conditions	Promote wider tax compliance including targeted audits to discourage abuse, as well efforts to combat base erosion and profit shifting (implementation of the minimum standards set by the OECD/G20 Inclusive Framework on BEPS).

Policymakers should consider which agency is best placed to support compliance verification. Confirming compliance can require technical knowledge for example in the area of R&D spending. Some countries involve specialised government agencies to verify compliance in areas where technical knowledge is required; for example, regarding determining if a project or eligible expenditure meets the definition of R&D spending (Box 4.1, Example 1). 10

4.1.2. Take-up

Tracking which firms claim the incentive is key to understanding how the policy works in practice and can inform improvements to incentive design. Relevant data on take-up includes the number of firms that claim and characteristics of beneficiaries. Table 4.3 provides a non-exhaustive list of potentially relevant data on the firm and project, as well as performance metrics (see below). Understanding which firms use the incentive can also reveal if there are unintended biases in the tax incentive design, for example, if it disproportionately benefits different firms types.

4.1.3. Cost

Tracking the amount of the benefit claimed is essential to understanding the costs of incentives. This includes data on the tax liability after the tax incentive is applied, and the taxpayer’s baseline tax base and rate (before claiming the incentive) (see Evaluation stage).11 In some countries, tax returns only require companies to report tax liability after accounting for all incentives, or do not separate tax incentives from standard deductions (available to all taxpayers). This makes tax expenditure reporting challenging, which is a key part of evaluation. Countries may also only track tax incentive use per project, rather than per firm. Ensuring unique tax identification numbers are linked to projects is key to attribute total costs per firm, which can be important to understand if recipients are concentrated among a small number of companies, for example.

4.1.4. Performance & potential outcomes

More sophisticated monitoring can track data that could help assess if the policy achieved its stated goals, including regarding spillovers. Monitoring performance of beneficiaries and other outcomes that might be linked to incentives can provide data for evaluations and assessing effectiveness. Performance metrics can be related to the stated policy goals, and could include data related to the firm, such as employment (jobs created and retained), as well as macroeconomic measures (Box 4.1, Example 2).

To support evaluations, it is useful to collect information on firm performance before the introduction or use of the incentive. For example, to better track job creation, firms could be required to report on total employment prior to receiving the benefit. Data should then be gathered over several periods after the incentive is granted, as effects may appear or peak at different times depending on the measure (Tinits and Fey, 2022[2]). Simple before-and-after comparisons will not, however, provide robust causal evidence of an incentive’s impact, as firms may have planned to expand employment regardless, delayed recruitment to meet eligibility conditions, or hired new workers while dismissing others—resulting in no net job creation. Collecting ex-ante information is therefore only a first step towards understanding impact; credible assessments require applying appropriate evaluation techniques (see Evaluation stage). Similarly, investment incentives can be evaluated more effectively when information is collected on both incentivised and non-incentivised investments—particularly when incentives target specific activities such as R&D or emerging technologies.

Table 4.3. Relevant data on tax incentive beneficiaries for evaluation

	Data points	Possible data sources
Firm information	Tax identification number	Tax returns
	Sector	Tax returns, business registries
	Date of incorporation	Business registries
	Ownership: foreign, domestic	Tax returns, business registries
Tax & Financial data	Gross and net income	Financial statements
	Taxable profit	Tax returns
	Tax regime, tax rate	Tax returns
	Taxes due & paid	Tax returns
	Tax incentives claimed (itemised)	Tax returns, financial statements
Performance metrics	Firm size: revenue, turnover, employees	Financial statements
	Amount of investment in period	Financial statements
	Key expense items: payroll, depreciation, etc.	Financial statements
	Employment & job quality: new hires, % employees above minimum wage, % employees on long-term contracts	Social security filings, national surveys
	Location of project	Business registries
	Local sourcing	Firm surveys
	Exports	Customs declarations

Policymakers could also track data beyond tax incentive recipients, with policy evaluation in mind. Assessing whether incentives have a causal impact may require information on comparable firms that did not benefit from them, to serve as a benchmark for what outcomes could have been in the absence of the measure. Such comparison groups can be constructed statistically – for instance, by identifying firms with comparable characteristics that did not claim the incentive – but this requires leveraging broader data sources, such as tax returns or other filings covering the wider population of taxpayers, rather than relying solely on information collected through incentive applications or beneficiary surveys.12

Box 4.1. Determining what to monitor: country examples

1. Verification of compliance might be carried out by a specialised agency. In Norway the decision for eligibility for the R&D incentive is taken by the Norwegian Research Council; in France, some applications are referred to the Ministry of Education and Science (EC, 2020[3]).

2. Evaluators for a tax credit aimed at quality job creation in the United States suggested tracking (among other data points): average and median wage of new employees and all employees; percentage of new and total employees with access to benefits; average annual spending with in-state vendors (to track linkages to local suppliers); change in gross state product; and change in number of major business headquarters in the state (OPEGA Maine State Legislature, 2020[4]).

4.2. Establish monitoring processes

With required indicators in mind, governments should designate how this information is collected, which agency or agency collects it, and, if relevant, how data is shared across agencies. This might require new laws permitting data sharing, as well as concurrent reforms digitising data, and ensuring data collection, storage and transfers meet confidentiality and security requirements. If tax enforcement capacity is low, or reporting requirements frequently change, firms also might not comply with reporting requirements, limiting monitoring. When considering how to collect data, policymakers should assess whether taxpayers are already submitting required information to other sources, and how procedures can be eased to collect the required data at lower costs.

4.2.1. Consider the most appropriate means to collect data and seek standard data formats

The necessary data for monitoring might already exist or may need to be collected. Below are the main datasets for monitoring tax incentives, and potential benefits and challenges with these sources.

• Tax returns may contain much of the relevant information for some basic forms of monitoring, depending on level of detail of forms and the structure of the CIT system. To track the full population of beneficiaries and calculate foregone revenue, it is important that all incentive recipients file tax returns, including those that are tax exempt, such as in economic zones, and those subject to special regimes, e.g. for extractive industries.

• Financial statements can provide complementary information in tax returns, particularly for tracking metrics on firm performance, such as investment and employment.

• Applications can include pertinent data on tax incentive beneficiaries depending on what information is required, including on intended performance of firms (e.g. investment size, tax relief, employment goals, R&D spending, etc.). However, this data is only available for tax incentive beneficiaries (not non-beneficiaries), and may only show self-reported projections that may require verification if goals are realised ex-post.

• Reporting ex-post, for example as part of requirements for incentive beneficiaries, may also provide key data related to incentive beneficiaries and policy goals. As with applications, data will be restricted to incentive users only, and might face biases from self-reporting, particularly if audit procedures are not in place or are not robust. Evaluators might require information from several years after the tax incentive was received, if some outcomes take time to materialise, requiring that firms submit reports consistently.

• Surveys may be less suited to collected information for evaluation, due to potential biases in survey design, and often high administration costs. To have time series data, which is important to track progress, surveys will have to be consistently issued and thoroughly completed. Surveys might be most useful to understand challenges related to how an incentive is used in practice, and other information not routinely collected. This could include for example, if requirements for eligibility and potential value of benefits could be improved (Pew Charitable Trusts, 2018[5]). Where other data is not available, surveys can be useful in this regard (Box 4.2, Example 1).

Information in any dataset should be sufficiently disaggregated to allow for relevant analysis. This includes, for example, separate line items in tax returns for each incentive, which is important to understand how each benefit reduces effective tax rates and for estimating revenue foregone. Detailed sectoral or location data may be important where incentives are targeted by location or sectors (Box 4.2, Example 2).

Forms should use standardised terms and common identifiers, and digital filing where possible, to support data interoperability. This is key to allow taxpayer data to be merged with other datasets, such as administrative or survey data. Forms should require use of standard sector classifications. It is also important that practitioners produce documentation (including metadata) for each dataset (OMB, 2016[6]). Among the many benefits of digitising data forms, online filing can ensure more complete datasets by requiring taxpayers fill out all mandatory information (Box 4.2, Example 3).

4.2.2. Set procedures for who collects data, and how data is shared & secured

The agency or agencies that track the required information to monitor tax incentive use should ideally be set in the conception phase. Depending on how incentives are granted, multiple authorities may be involved in monitoring or granting some or all incentives. For example, in some countries, firms in economic zones may submit tax returns only to economic zone authorities, rather than to the national tax authority. An investment promotion agency might be responsible for granting tax incentives based on applications submitted by firms. In these cases, it is important to set clear procedures for which government body is responsible for collecting what information, and how relevant information is shared.

Tax authorities should have access to all relevant monitoring data, for all incentive beneficiaries. This is particularly important for tracking compliance and informing audits, as well as for accurate tax expenditure reports, and other evaluations often conducted by the Ministry of Finance. In some cases, other government agencies might already be collecting relevant data or could be well placed to do so, which highlights the importance of data sharing.13

Procedures and infrastructure should be put in place to allow secure data sharing. Sharing relevant monitoring data across agencies might be hindered by confidentiality restrictions or administrative barriers. Countries may allow for sharing of other agencies’ data with the tax authorities, as well as specific access upon request for external research (see Box 4.2, Example 4) (OMB, 2016[6]). Reducing procedural barriers, including the using of standard formats or classifications, using common identifiers and strengthening data infrastructure (Heady and Mansour, 2019[7]). To ensure confidentiality, appropriate data protection procedures should be in place, including, as necessary, ability to anonymise or encrypt datasets. Some countries have set up secure physical data labs to allow for data analysis (see Box 4.2, Example 5).

Box 4.2. Establishing monitoring processes: country examples

1. South Africa has used surveys to inform improvements to how incentives are implemented for example, and to glean specific challenges faced by small investors.

2. In the United States, evaluators could not assess take-up or performance of individual Empowerment Zones, as tax forms did not require incentives recipients to specify where projects were located (GAO, 2022[8]).

3. One empirical study in China found that the government’s electronic tax platform improved take-up of a reduced tax rate for high-technology enterprises, primarily by bridging information gaps. The platform also allows the government to collect data on firm financial and tax information (Lin, Ma and Zhang, 2025[9]).

4. For examples on sharing administrative data across agencies in the United States see (OMB, 2016[6]).

5. South Africa, Uganda and Zambia have set up data labs to support monitoring and analysis (UNU Wider, 2026[10]).

4.3. Build capacity for data collection

Governments should seek to improve capacity and expand data collection to support monitoring in the medium- to long-term. This includes improving data infrastructure and statistical capacity. Here, maturity models or other frameworks could support assessments of current capacity, constraints, and areas to focus.14 Digital adoption can support tax administrations including for assessing and mitigating compliance risk and data security and analysis (OECD, 2020[11]; OECD, 2021[12]; ADB, 2022[13]; Okunogbe and Santoro, 2023[14]).

Jurisdictions should consider the most appropriate means to expand data collection. While modifying required information on tax returns might be the most direct and consistent, practitioners will have to establish which additional information is most appropriate to inform monitoring on all incentives. Modifying tax returns also has costs, including updating guidance on forms, and other support to taxpayers.

References

[13] ADB (2022), Launching A Digital Tax Administration Transformation, Asian Development Bank, Manila, Philippines, https://doi.org/10.22617/TCS210343.

[21] CIAT (2022), Data governance for tax administrations. A practical guide, Inter-American Center of Tax Administrations – CIAT, https://www.nto.tax/sites/default/files/resources/Data%20Governance%20for%20Tax%20Administrations.%20A%20Practical%20Guide%20%282022%29..pdf (accessed on 19 February 2025).

[3] EC (2020), Mutual Learning Exercise Administration and Monitoring of R&D tax incentives Horizon 2020 Policy Support Facility, European Commission, Brussels.

[18] FERDI (2018), Tax Expenditure Assessment: From Principles to Practice Methodological guide, Foundation for Studies and Research on International Development, Paris, https://ferdi.fr/dl/df-PSDXZuxEEh8LihnEhHUoFkjx/tax-expenditure-assessment-from-principles-to-practice-methodological-guide.pdf (accessed on 14 February 2025).

[8] GAO (2022), “Economic Development: Status of Recommendations on Empowerment Zones and Other Selected Community Investment Initiatives”, No. GAO-23-106113, United States Government Accountability Office (GAO), Washington, DC, https://www.gao.gov/products/gao-23-106113 (accessed on 27 January 2025).

[25] GAO (2010), “New Markets Tax Credit”, Report to Congressional Committees, No. GAO-10-334, United States Government Accountability Office (GAO), Washington, DC, https://www.gao.gov/assets/d10334.pdf (accessed on 18 March 2025).

[24] Granger, H., K. McNabb and H. Parekh (2022), “Tax expenditure reporting in Rwanda and Uganda: challenges, practical guidance and lessons learnt”, ODIWorkingPaper, ODI, London, https://cdn.odi.org/media/documents/ODI_Working_paper_Tax_expenditure_reporting_in_Rwanda_and_Uganda.pdf (accessed on 18 September 2024).

[7] Heady, C. and M. Mansour (2019), Tax Expenditure Reporting and Its Use in Fiscal Management: A Guide for Developing Economies, IMF, Washington, DC.

[19] Junquera-Varela, R. et al. (2022), “Digital Transformation of Tax and Customs Administrations”, Equitable Growth, Finance and Institutions, World Bank Group, Washington, DC, http://documents.worldbank.org/curated/en/099448206302236597 (accessed on 19 February 2025).

[20] Junquera-Varela, R. and C. Lucas-Mas (2024), Revenue Administration Handbook, World Bank, Washington, DC, http://hdl.handle.net/10986/41090” (accessed on 14 February 2025).

[9] Lin, C., K. Ma and X. Zhang (2025), “Where technology meets tax: The impact of digital tax administration on tax incentive take-up”, Economics Letters, Vol. 247, p. 112118, https://doi.org/10.1016/j.econlet.2024.112118.

[16] OECD (2024), Tax Administration 2024: Comparative Information on OECD and other Advanced and Emerging Economies, OECD Publishing, Paris, https://doi.org/10.1787/2d5fba9c-en.

[26] OECD (2022), Digital Transformation Maturity Model, OECD Publishing, Paris, http://www.oecd.org/tax/forum-on-tax-administration/publications-and-products/digital-transformation-maturity-model.htm (accessed on 18 March 2025).

[22] OECD (2022), “Tax Compliance Burden Maturity Model”, OECD Tax Administration Maturity Model Series, OECD, Paris, http://www.oecd.org/tax/forum-on-tax-administration/publications-and-products/tax-compliance-burden-maturity-model.htm (accessed on 19 February 2025).

[12] OECD (2021), Supporting the Digitalisation of Developing Country Tax Administrations, OECD Publishing, Paris, https://doi.org/10.1787/8600724a-en.

[11] OECD (2020), Tax Administration 3.0: The Digital Transformation of Tax Administration, OECD Publishing, Paris, https://doi.org/10.1787/ca274cc5-en.

[15] OECD (2017), The Changing Tax Compliance Environment and the Role of Audit, OECD Publishing, Paris, https://doi.org/10.1787/9789264282186-en.

[23] OECD (2015), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris, https://doi.org/10.1787/9789264239012-en.

[1] OECD (2004), Compliance Risk Management: Managing and Improving Tax Compliance, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/topics/policy-issues/tax-administration/compliance-risk-management-managing-and-improving-tax-compliance.pdf (accessed on 18 February 2025).

[28] OECD/UNDP (2024), Tax Inspectors Without Borders: Annual Report 2024, OECD Publishing, Paris, https://doi.org/10.1787/8968125d-en (accessed on 18 February 2025).

[14] Okunogbe, O. and F. Santoro (2023), “The Promise and Limitations of Information Technology for Tax Mobilization”, The World Bank Research Observer, Vol. 38/2, pp. 295-324, https://doi.org/10.1093/wbro/lkac008.

[6] OMB (2016), “Barriers to Using Administrative Data for Evidence-Building”, White paper, Office of Management and Budget, Washington, DC, https://obamawhitehouse.archives.gov/sites/default/files/omb/mgmt-gpra/barriers_to_using_administrative_data_for_evidence_building.pdf (accessed on 12 February 2025).

[4] OPEGA Maine State Legislature (2020), “Maine Capital Investment Credit (MCIC): A Complicated Response to Federal Bonus Depreciation that Is Unlikely to Significantly Encourage Capital Investment in Maine”, Tax Expenditure Review: report to the Government Oversight Committee and Taxation Committee, No. Report No. TE-MCIC-17, The Office of Program Evaluation and Government Accountability of the Maine State Legislature, Augusta, https://documents.ncsl.org/wwwncsl/Fiscal/evaluationDB/MaineCapitalInvestmentTaxCredit.pdf (accessed on 13 September 2024).

[27] Pecho, M. et al. (2024), “Managing Tax Incentives in Developing Countries”, Technical Notes and Manuals, No. 2024/007, IMF, Washington, DC, https://doi.org/10.5089/9798400289590.005.

[5] Pew Charitable Trusts (2018), How States Can Gather Better Data for Evaluating Tax Incentives, The Pew Charitable Trusts, Philadelphia, https://www.pewtrusts.org/en/research-and-analysis/issue-briefs/2018/06/how-states-can-gather-better-data-for-evaluating-tax-incentives (accessed on 13 February 2025).

[2] Tinits, P. and C. Fey (2022), “The Effects of Timing and Order of Government Support Mechanisms for SME Exports”, Management International Review, Vol. 62/2, pp. 285-323, https://doi.org/10.1007/s11575-022-00465-2.

[10] UNU Wider (2026), Secure research data labs: unlocking evidence-based policy in Africa through data, UNU Wider, https://www.wider.unu.edu/about/secure-research-data-labs-unlocking-evidence-based-policy-africa-through-data (accessed on 14 January 2026).

[17] World Bank (2024), Tax Expenditure Manual, World Bank Group, Washington, DC, http://documents.worldbank.org/curated/en/099062724151636908 (accessed on 14 February 2025).

Notes

← 1.  On setting a process for compliance risk management, see (OECD, 2004[1]; OECD, 2017[15]; OECD, 2024[16]). On developing a compliance programme for tax incentives see (Pecho et al., 2024[27]).

← 2.  For an overview of data requirements and sources to estimate tax expenditures, see (World Bank, 2024[17]; Heady and Mansour, 2019[7]; FERDI, 2018[18]).

← 3.  For process and data requirements for TE reporting, see (Heady and Mansour, 2019[7]))

← 4.  For guidance on digitalisation of tax administration in developing countries, and options for tools and assistance, see (OECD, 2021[12]; ADB, 2022[13]).

← 5.  For a summary on privacy, security and transparency concerns, as well as digital transformation of tax administration more broadly see (OECD, 2020[11]; Junquera-Varela et al., 2022[19]).

← 6.  For guidance on building capacity for data science in tax and customs administrations see (Junquera-Varela and Lucas-Mas, 2024[20]); for practical advice on improving data governance for tax administration see (CIAT, 2022[21]).

← 7.  A compliance risk assessment can support this identification. For guidance and good practice examples on the process of compliance risk management see in (OECD, 2024[16]; OECD, 2004[1]; OECD, 2017[15]).

← 8.  There exist resources to support countries seeking to improve compliance capacity, including Tax Inspectors Without Borders, a joint initiative of the OECD and the UNDP, which provides practical, hands-on assistance on current audit cases and related international tax issues (OECD/UNDP, 2024[28]). For a maturity model to support self-assessment of tax compliance burden see (OECD, 2022[22]).

← 9.  For examples of common mistakes made in claims for R&D tax incentives, see (EC, 2020[3])

← 10.  Governments can refer to internationally recognised definitions for R&D set out in the Frascati Manual (OECD, 2015[23]). However, even with clear definitions, confirming that a firm’s activities meet this definition might sometimes be referred to specialised agencies.

← 11.  For more on information required to assess tax expenditures see (World Bank, 2024[17]; Heady and Mansour, 2019[7]) and for application in Rwanda and Uganda (Granger, McNabb and Parekh, 2022[24]). Additional information on tax expenditure cost estimates available in Evaluation.

← 12.  It may also be useful to collect data on projects that were not completed, or filed for bankruptcy, particularly for incentives that aim to support riskier projects, to understand which percentage of supported projects were viable (GAO, 2010[25]).

← 13.  For example, the tourism ministry might require registered hotels to submit data on occupancy rates; this could be pertinent to track performance of tax incentives geared towards supporting the industry.

← 14.  For example, the OECD Digital Transformation Maturity Model aims to support tax administrations to assess and expand their digital maturity, including for data management (OECD, 2022[26]).