The GTED gathers official information published by governments on the revenue they forgo through tax expenditures. The current version includes reported revenue forgone estimates for 119 jurisdictions. A new visualization linking each country’s reported revenue forgone to the GRADE model allows users to estimate what the same revenue could mean for children, mothers and communities between 2026 and 2035.
Tax expenditures are public spending delivered through the tax system. Unlike direct spending, they are rarely reviewed annually, and their cost is less visible in the budget. The question we set out to answer is the next one: what could the same revenue achieve if it were spent on public services instead?
Even a modest reform could make a real difference. Suppose governments in sub-Saharan Africa curtailed only their corporate income tax expenditures and collected just half of the revenue these provisions forgo: about US$1.8 billion a year across 23 of the region’s 31 reporting countries that have such provisions (constant 2015 US$). Spent as governments have spent additional revenue in the past, we estimate this could avert about 42,000 deaths of children under five and 3,300 maternal deaths over 2026–2035. It could also provide 2.1 million additional years of schooling for children and, by 2035, give 724,000 more people basic drinking water and 1.4 million more basic sanitation.
Why connect tax expenditures to children, their families and their communities?
Tax expenditure reports usually stop at the cost: revenue forgone, which can be normalized as a share of GDP or of tax collected. That figure is essential, but it says little about what is given up. A budget line of 2% of GDP means something very different in a country where one child in ten dies before age five than in one where survival is near universal.
GRADE makes that link. It estimates how outcomes such as child and maternal survival, school attendance, access to water and sanitation, and social protection have historically changed with government revenue per capita, with the quality of governance moderating the response. Connecting GRADE to GTED allows us to express revenue forgone in terms that matter for children’s rights: lives, years of schooling and access to basic services.
The visualization shows opportunity cost. It does not recommend removing any provision: some tax expenditures serve important purposes.
How the visualization works
For each country we average three recent years of reported revenue forgone, using government estimates of past years wherever they exist and projections where they do not. Values are converted to constant 2015 US dollars and expressed as a share of government revenue, the form GRADE uses. GRADE allows a five-year lead-in, and outcomes are reported for 2026–2035.
Users work through four steps:
- Choose countries: one country, or a group, by region or income level.
- Choose which tax expenditures to include: a ready-made scenario (for example, provisions with no stated policy objective, or business investment incentives), or a custom selection using GTED’s categories of tax base, tax, type, beneficiary, policy objective and time frame.
- Set the recovery rate: revenue forgone is a static estimate that assumes no change in behavior, so users can test what happens if governments recover only part of it.
- Choose outcomes: survival, education, basic services, social protection, nutrition and the health workforce, with services shown for all people, children or women of reproductive age.
The results update as settings change. GRADE assumes additional revenue is spread across the budget as governments have spent in the past, so the outcomes are simultaneous effects of the same revenue, not alternative uses of it.
What the results show
The gap between where the money is and where the gains are is stark. Low- and lower-middle-income countries report US$140 billion a year in tax expenditures, 4% of the global total, yet that revenue could avert an estimated 1.27 million under-five deaths over 2026–2035.
Country results tell the same story at national level. Kenya reports tax expenditures worth US$2.1 billion a year, equivalent to 11% of government revenue. Spent on public services, that revenue could avert an estimated 17,800 under-five deaths and 2,100 maternal deaths over 2026–2035.
Reading the results with care
For various reasons, revenue forgone estimates are far from perfect and hence, need to taken with a grain of salt.
- Revenue forgone is static. It assumes there is no change in behavior, so the revenue a policy change would raise may be lower. The recovery slider lets users test this.
- Figures are not strictly comparable across countries. Governments define their benchmark tax systems against which revenue forgone is calculated differently. Moreover, as shown by the Global Tax Expenditures Transparency Index (GTETI), the scope and quality of TE reports varies significantly across jurisdictions, and a country that reports few provisions will appear to forgo little.
- The figures show gross, not net, revenue. Where a tax expenditure serves a genuine purpose, removing it may require a compensating measure, such as a direct transfer to the households it was meant to help, which would reduce the net revenue available.
- Projected baselines beyond the latest observed data are GRADE’s mechanical extensions, not forecasts, and the timing of reform is illustrative.
What does this mean for tax expenditure policy?
Three lessons follow for anyone who designs, reports on or reviews tax expenditures.
First, the cost of tax expenditures should be reported alongside what they displace. A figure in percent of GDP invites comparison with other countries; a figure in children’s lives and years of schooling invites comparison with the services a government could otherwise provide.
Second, review should start where the opportunity cost is highest. Provisions with no stated objective, and those in countries where public services are thinnest, carry the largest human cost. Reviews that weigh each provision’s stated purpose against that cost can target reform where it matters most.
Third, transparency makes this analysis possible. Every figure in the visualization can be traced back to the GTED records behind it, and a companion workbook shows each step from the database to the GRADE inputs and results. More governments reporting would strengthen both tax policy and the estimates that rest on it.
The visualization is designed for country advocates, finance ministries, parliamentarians and researchers. Choose a country, select the provisions you want to examine, set a recovery rate you find credible, and the visualization shows what that revenue could mean for children, families and communities. We welcome feedback as we develop it and will update it as new GTED and GRADE data becomes available.