Detailed research-gap map

The two themes are complementary: the first asks whether governments can measure what distinct spending instruments accomplish in distinct places; the second asks whether the tax designs that are distributionally defensible are also understood and accepted by the people who bear them. The entries below are ranked by tractability and potential contribution. Each reported gap is limited to a limitation, inconsistency, or future-research need stated in the cited literature.

Theme 1: Geographic and instrument-specific evidence for green public expenditure

1. Build comparable expenditure measures before estimating “effectiveness”

What is established. Environmental public-expenditure research has examined allocation and public action, but the literature reviewed to date is concentrated in Europe and Latin America, relies predominantly on qualitative case studies, and commonly blends performance with descriptive accounts of spending. The review specifically calls for better transparency, accounting organization, and indicators to permit deeper analysis. [citation_1xf0or]

What is missing. The immediate gap is a harmonized, auditable way to distinguish operating expenditure, capital investment, grants, subsidies, procurement, public R&D, and tax expenditure, while recording the environmental objective, beneficiary, implementation stage, and geographic level. Without this measurement layer, cross-place estimates can confuse genuine policy effects with different budget classifications or disclosure practices.

Why it matters. A low aggregate level of spending is not itself evidence that more spending is warranted; effectiveness against environmental priorities is the relevant criterion, and conservation expenditure can have short-run costs with long-run benefits. [citation_1xf0os]

Research direction. Construct a panel for a defined set of national and subnational governments using budget microdata and procurement records. Reconcile functional classifications with climate-budget tags, then validate the coding against program documents. Report comparability diagnostics alongside outcome estimates rather than treating budget totals as interchangeable treatment variables.

2. Compare instruments rather than treating “green expenditure” as one treatment

What is established. Green public procurement is an important but distinctive demand-side instrument. Its literature has largely focused on implementation impacts; comparative analysis against other environmental tools in terms of efficiency and innovation remains limited, alongside restricted geographic coverage. [citation_1xf0td] A separate welfare-economic assessment concluded that procurement needs much more research on environmental effects and especially market competition. [citation_1xf0ow]

What is missing. Evidence is too thin on when procurement, grants, subsidized finance, public R&D, and direct public investment are substitutes, complements, or inferior alternatives for a common environmental objective. The omission is instrument-specific: an estimate for aggregate environmental spending cannot establish the relative cost, innovation effect, supplier concentration, or distributional incidence of a procurement rule.

Why it matters. Current findings are not uniformly favorable. One analysis concludes that procurement may be counterproductive unless it requires an environmental standard beyond firms’ prevailing technology and is anchored to defined environmental objectives. [citation_1xf0tf] More recent EU evidence finds positive associations between procurement and SME green products/processes but mixed results for whether greater SME-sector tangibility strengthens that effect. [citation_1xf0tg]

Research direction. Within one sector—such as building retrofits, buses, or renewable-power equipment—compare matched jurisdictions that deploy procurement criteria, grants, or R&D support. Estimate emissions, lifecycle cost, entry and award concentration, supplier innovation, and benefits by firm size. Pre-specify the common outcome metric and compare the instruments at equal public cost.

3. Identify local implementation conditions and multi-level spillovers

What is established. Reviews identify a gap in understanding localized factors shaping green procurement in developing countries and call for richer qualitative analysis. [citation_10w77v] The literature also describes unresolved tensions between uniform policy requirements and local flexibility, and among environmental, economic, and social objectives. [citation_1xf0tc]

What is missing. The unresolved unit of analysis is not merely “country.” It is the interaction among fiscal capacity, procurement-market depth, administrative capability, contract type, and the level of government that implements the program. A national green-budget rule can therefore have different effects across regions or municipalities, yet those differences are rarely designed into comparative evaluation.

Why it matters. A recent large-scale analysis of Chinese tender documents found a central–local gap in the presence of green requirements and uneven development across procurement dimensions. [citation_17y9dm] This supports treating implementation intensity as an empirical exposure, rather than assuming that a national policy produces uniform practice.

Research direction. Use staggered rollout, mandate thresholds, or changes in centralized purchasing rules across local governments. Combine tender-level data with local supplier networks and outcome data, estimate spatial spillovers, and use interviews to explain mechanisms that administrative data cannot identify. Stratify the design prospectively by fiscal capacity and market concentration.

4. Move from short-run associations to causal, long-horizon portfolios

What is established. A 27-country EU study reports that absolute environmental-protection expenditure was positively associated with green total factor productivity, whereas expenditure intensity was not; individual spending categories were not significant in the study’s disaggregated analysis. [citation_181sin] The authors state that the study cannot establish definitive causality, has a short observation period, excludes private environmental investment, and may not observe effects that require horizons longer than two years. [citation_181sin]

What is missing. The gap is credible evidence on lag structures, additionality, and private-investment responses for expenditure portfolios rather than annual totals or isolated budget lines. This is particularly important when effects operate through infrastructure, learning, or clean-technology diffusion.

Why it matters. A country-panel study likewise found no significant independent relationship between environmental spending and its proxy for green investment, and it did not support a strong causal claim about complementarity between spending and carbon pricing. [citation_1gwjmp] These findings make a short-run, single-instrument inference a weak basis for expenditure design.

Research direction. Assemble 10–15 years of program-level commitments, disbursements, and completion dates; exploit predetermined eligibility rules or financing shocks; and estimate event-study effects on emissions, innovation, firm investment, and fiscal costs. Model public spending and carbon-price exposure jointly, while reporting sensitivity to delayed effects and to alternative measures of private green investment.

Theme 2: Reconciling distributional contradictions and public acceptance of environmental taxes

5. Link measured tax incidence to perceived fairness and support in the same households

What is established. The distributional consequences of carbon taxes depend on energy use, the indicator used to measure distributional effects, and household characteristics. Lump-sum transfers are presented as better suited to mitigating regressive effects for lower-income households, while reductions in labor taxes benefit higher-income households more. [citation_1fjoxk] Separately, the acceptance literature consistently associates support with information about policy effectiveness and revenue use, trust in government, and perceived fairness of cost sharing. [citation_1ooj65]

What is missing. The linked evidence is missing: the existing research agenda identifies that acceptance studies have not tied trust and tax-morale predictors to actual incidence outcomes. [citation_1xf0lv] Consequently, a household can be objectively compensated yet remain opposed, or face a genuine loss yet support the policy for environmental reasons; separate incidence simulations and opinion surveys cannot adjudicate between those cases.

Why it matters. This is the most tractable high-value gap because it directly explains why a distributionally progressive design may not secure legitimacy, and why a popular design may not protect those most exposed.

Research direction. Link household expenditure or administrative-income data to a repeated policy survey. Simulate direct and indirect tax burdens and each recycling package at household level, then randomize transparent information on the household’s own net position, climate effects, and revenue destination. Estimate whether objective incidence, perceived incidence, trust, and environmental concern independently predict support and actual policy compliance.

6. Explain the revenue-recycling contradiction rather than assuming a universal “fair” package

What is established. A review finds that people are highly concerned about unequal burdens yet often prefer revenue for environmental projects over compensation for inequitable outcomes; better understanding of carbon pricing can shift preferences toward fairer outcomes. It recommends combining support for vulnerable groups with environmental-project funding. [citation_109qyy] However, a Swiss conjoint experiment found no average acceptability gain from redistribution; progressive redistribution was more favorable only under particular policy baselines and among some political groups. [citation_1xf0rz]

What is contested. Revenue recycling is not a single treatment. The evidence points in different directions because the comparison category, existing tax regime, political orientation, stated purpose, and information supplied can change the response. Treating a rebate, labor-tax cut, targeted transfer, and earmarked investment as equivalent “recycling” obscures the mechanism.

Why it matters. This contradiction is central to policy design: compensation may reduce material regressivity, while earmarking may build credibility and environmental purpose. A package that pursues only one can fail either equity or acceptance.

Research direction. Run a preregistered, multi-country conjoint experiment with a fixed tax base and price path. Randomize the allocation among universal dividends, targeted transfers, labor-tax reductions, transit or retrofit investment, and mixed packages; disclose each respondent’s modeled net position; and measure both policy support and perceived procedural and distributional fairness. Include an explicit status-quo condition where a carbon tax already exists.

7. Standardize how acceptability is elicited and test whether survey support predicts behavior

What is established. Carbon-tax acceptability estimates diverge across elicitation methods, policy designs, and contextual features, which complicates interpretation and challenges the validity of pooled reviews and meta-analyses. In a within-subject study, explicit surveys produced lower acceptability than implicit model-based measures, and preferred tax levels differed substantially from maximum acceptable levels. [citation_1xf0ry]

What is missing. The field needs a shared measurement protocol that separates support for a specified policy from willingness to bear its cost, and that assesses whether either survey response predicts real choices after implementation. The current evidence supports context-specific inference, not a universal acceptance coefficient. [citation_1xf0ry]

Why it matters. Information effects themselves are inconsistent: a Swiss panel study reports positive, negative, and null effects in earlier work, and found an information effect only among less-educated respondents in the full analysis. [citation_1xf0t2] A measurement design that conceals this heterogeneity can generate misleading advice about communication campaigns.

Research direction. Field the same respondents across explicit support questions, incentive-compatible willingness-to-pay tasks, and choices with actual or consequential payment implications. Recontact respondents after a tax change and compare stated support with voting, enrollment in rebates, energy use, or other observable behavioral responses.

8. Replace generic trust measures with institution- and process-specific mechanisms

What is established. The systematic review of 60 studies calls for research that moves beyond generic trust-in-government questions to responsiveness, reliability, integrity, openness, and fairness; it also calls for work on economic perceptions, tax morale, and fiscal-psychology or theory-of-planned-behavior variables. [citation_1vhz6d]

What is missing. Existing studies often establish that trust matters without identifying which government behavior changes support: credible revenue accounting, delivery of compensation, transparency of emissions results, participation in rule-setting, or enforcement even-handedness. The mechanism remains too broad to translate into an institutional reform.

Why it matters. Qualitative evidence from the United Kingdom identifies lack of trust in revenue use, difficulty understanding a tax shift, and a desire for incentives alongside penalties as design problems, rather than simple hostility to environmental taxation. [citation_1xf0s7]

Research direction. Pair a survey experiment with administrative evidence on tax-revenue execution and program delivery. Randomize verifiable disclosure of revenue flows, independent auditing, eligibility rules for compensation, and opportunities for local participation. Test which intervention changes specific trust dimensions, perceived fairness, and support, and whether effects differ for rural households, low-income households, and high-exposure workers.

Priorities for a new study

A joint household-incidence and acceptance design is the clearest near-term contribution because it answers the stated missing link between objective burdens and the predictors of acceptance. Instrument-specific expenditure evaluation is the higher-data, higher-payoff agenda: begin with one program, one sector, and one geographic implementation setting, then build a comparable portfolio design rather than estimating aggregate green spending. Across both themes, pre-registration, transparent coding of fiscal instruments, distributional reporting beyond income alone, and follow-up long enough to observe implementation are more valuable than adding another cross-sectional correlation.