The Suits index is a statistical measure of tax progressivity, analogous to the Gini coefficient for income inequality but applied to the distribution of tax burdens relative to income shares, developed by economist Daniel B. Suits in 1977.[1] It quantifies how disproportionately taxes fall on higher- versus lower-income groups by constructing a Lorenz curve of cumulative tax payments plotted against cumulative pre-tax income, where the index value equals twice the area between this curve and the line of equality.[2] Ranging from -1 for a perfectly regressive tax (where the poorest bear the entire burden) to +1 for a perfectly progressive tax (where the richest bear the entire burden), a value of 0 denotes proportionality, with positive values indicating progressivity and negative values regressivity.[2] Widely employed in empirical tax policy analysis, the index facilitates comparisons of tax system equity across jurisdictions, time periods, or policy reforms, such as evaluating the net progressivity of combined taxes including lump-sum transfers or environmental levies.[3][4] Its additivity property allows aggregation of multiple taxes into a system-wide measure as a weighted average, aiding assessments of overall fiscal redistribution without requiring assumptions about behavioral responses to tax changes.[1]
Origins and Conceptual Foundation
Historical Development
The concept of measuring tax progressivity quantitatively emerged in the early 20th century, driven by efforts to evaluate vertical equity in fiscal systems where higher-income individuals should bear proportionally greater burdens. A foundational approach was articulated by Arthur Pigou in 1928, who defined a simple progressivity metric as the ratio of the percentage change in the average tax rate to the percentage change in income, enabling assessments of whether tax structures amplified or diminished income-based differentials.[5]Building on such rudimentary indicators, Richard A. Musgrave and Tun Thin introduced the "effective progression" index in 1948, which quantified the extent to which cumulative tax liabilities exceeded or fell short of proportional shares of income distribution up to each incomepercentile. This measure represented an early attempt to integrate distributional analysis into progressivity evaluation, influencing subsequent methodologies by emphasizing comparisons between tax and pre-tax income Lorenz curves.[6]By the mid-20th century, these tools were applied in empirical studies of tax incidence, such as Joseph Pechman and Benjamin Okner's 1974 analysis of U.S. federal, state, and local taxes using microdata to decompose progressivity across revenue sources. However, limitations in earlier indices—such as sensitivity to specific income thresholds or lack of a bounded, Gini-like scalar—prompted innovations in the 1970s to develop comprehensive, inequality-inspired metrics capable of summarizing overall system progressivity for multi-tax environments.[7]
Daniel B. Suits' Contribution
Daniel B. Suits (1918–), an American economist with a Ph.D. in economics from the University of Michigan, introduced the Suits index in his seminal 1977 paper "Measurement of Tax Progressivity," published in the American Economic Review.[1] In this work, Suits adapted the Gini coefficient methodology—originally developed for measuring income inequality—to evaluate the progressivity of tax systems by constructing a Lorenz curve for tax payments relative to pre-tax income shares.[1] This approach quantifies how tax burdens deviate from proportionality, yielding an index value between -1 (fully regressive) and +1 (fully progressive), with zero indicating a proportional tax.[8]Suits' innovation emphasized empirical tractability and interpretive clarity, arguing that the index captures the net redistributive effect of taxation through a single, bounded statistic analogous to the Gini but inverted for tax shares (where lower-income groups pay less than their income proportion under progressive systems).[1] He demonstrated its applicability to aggregate tax systems, highlighting a key property: the progressivity index for a multi-tax regime equals the income-share-weighted average of individual tax indices, facilitating decomposition analysis without requiring marginal rate data.[1] This additivity property, derived from the mathematical structure of the underlying integrals, enables policymakers to isolate contributions from specific levies like income versus sales taxes.[1]Developed independently alongside N.C. Kakwani's contemporaneous measure, Suits' formulation gained prominence for its direct linkage to observable tax incidence data and resistance to manipulation by rate changes alone, prioritizing burden distribution over nominal structures.[9] Empirical examples in Suits' paper applied the index to U.S. federal taxes, illustrating values above zero for progressive elements like individual income taxes, underscoring its utility in assessing real-world fiscal equity without assuming unitary elasticity or other restrictive behavioral assumptions.[1] Subsequent adoption in tax incidence studies, such as state-level analyses, validates its robustness, though critics note sensitivity to income definition and incidence assumptions inherent in Lorenz construction.[10]
Mathematical Definition and Calculation
Core Formula
The Suits index S quantifies the progressivity of a tax system by adapting the Gini coefficient methodology to the distribution of tax liabilities relative to income shares. It is defined as S = 1 - \frac{L}{K}, where L is the area under the tax concentration curve and K = 0.5 is the area under the 45-degree diagonal line in the unit square representing perfect proportionality.[11] This simplifies to S = 1 - 2L, yielding values between -1 (extreme regressivity, where the curve lies entirely below the diagonal) and +1 (extreme progressivity, where the curve lies entirely above the diagonal).[1] A value of S = 0 indicates proportionality, with the tax concentration curve coinciding with the diagonal.[12]The tax concentration curve, denoted T(p), plots the cumulative share of total tax payments on the vertical axis against the cumulative share of total pre-tax income p (from 0 to 1) on the horizontal axis, with individuals ranked from lowest to highest income.[1] In continuous form, L = \int_0^1 T(p) \, dp, so S = 1 - 2 \int_0^1 T(p) \, dp.[11] For progressive taxation, T(p) < p for low p (poorer groups pay less than their income share), placing the curve below the diagonal and resulting in L < 0.5, hence S > 0; the reverse holds for regressivity.[10]In discrete applications using grouped data (e.g., income quintiles), L approximates via trapezoidal integration or summation: for income shares y_i and tax shares t_i across n groups ordered by increasing income, L \approx \sum_{i=1}^n \left( t_i \cdot \left( \sum_{j=1}^{i-1} y_j + \frac{y_i}{2} \right) \right), then S = 1 - 2L.[4] This mirrors Gini computation but substitutes tax payments for income on the ordinate, ensuring the index reflects deviations from income-proportional tax burdens. Empirical calculations often draw from microdata like household surveys or tax returns, assuming full tax incidence on payers.[13]
Graphical Representation via Lorenz Curves
The Suits index is graphically depicted through a specialized Lorenz curve, known as the Suits curve, which illustrates the distribution of tax liability relative to pre-taxincome distribution. In this representation, the horizontal axis denotes the cumulative proportion of total pre-taxincome, ordered from the lowest to highest earners, while the vertical axis indicates the corresponding cumulative proportion of total tax payments borne by those income groups.[14][15] For a taxsystem where liability is exactly proportional to income, the curve aligns with the 45-degree line of equality, reflecting zero progressivity. Deviations from this line quantify the degree of progressivity or regressivity: a curve positioned below the diagonal signifies progressivity, as higher-income groups bear disproportionately more tax relative to their income share; conversely, a curve above the diagonal indicates regressivity.[7][16]To construct the Suits curve, income recipients are ranked by ascending pre-tax income, and cumulative shares are computed for both income and tax liability up to each percentile or quantile. The area under this curve, denoted as L, serves as the basis for the indexcalculation: S = 1 - 2L, where the total area beneath the 45-degree line is 0.5. This formula mirrors the Gini coefficient's structure, adapted here to measure the concentration of tax burden against income rather than inequality within a single distribution. Positive values of S (up to 1) arise when the curve bows below the diagonal, confirming progressive taxation; negative values (down to -1) occur with regressive systems.[14][17]This graphical approach facilitates intuitive visualization of tax progressivity's magnitude and direction, enabling comparisons across tax instruments or jurisdictions. For instance, in analyses of U.S. state income taxes, the curve's position below the diagonal has been observed to yield positive Suits indices, underscoring their progressive nature. Empirical plots often highlight how the curve's convexity reflects the tax schedule's marginal rates, with steeper bows corresponding to stronger progressivity at higher incomes. Limitations include sensitivity to data granularity and assumptions about income ordering, but the method's alignment with distributional Lorenz principles ensures robust interpretive power for policy evaluation.[7][15]
Key Properties and Interpretive Framework
Scale and Interpretation
The Suits index ranges from -1 to 1, providing a bounded measure analogous to the Gini coefficient but adapted for tax distributions.[18][19] A value of 0 signifies a proportional tax system, in which each income group's share of total tax payments equals its share of total income.[19][20] Positive values indicate progressivity, where higher-income groups contribute a greater proportion of taxes than their income share, reflecting a redistributive effect.[18][10] Negative values denote regressivity, with lower-income groups bearing a disproportionately larger tax share relative to income.[18][10]At the extremes, an index of 1 represents perfect progressivity, where the highest-income group pays the entire tax burden, while -1 indicates perfect regressivity, with the lowest-income group covering all taxes.[10][19] Interpretation relies on the index's derivation from the area between the Lorenz curve of income shares and the concentration curve of tax shares, normalized against the area under the 45-degree line of equality.[18] This framework allows direct comparability across tax systems or jurisdictions, though values near zero in aggregate systems often reflect offsetting effects of progressive (e.g., income) and regressive (e.g., sales) components.[19] Empirical applications, such as U.S. state-level analyses, typically yield indices between -0.2 and 0.3 for overall tax progressivity, underscoring modest net redistribution after accounting for multiple revenue sources.[19]
Additivity for Multi-Tax Systems
The Suits index exhibits additivity across multiple taxes in a system, meaning the progressivity of the overall tax burden is the revenue-share weighted average of the progressivity indices for each individual tax. Formally, for a tax system with components j = 1, \dots, m, the aggregate Suits index S satisfies S = \sum_{j=1}^m r_j S_j, where r_j denotes the share of total tax revenue attributable to tax j (with \sum r_j = 1), and S_j is the Suits index for tax j.[21][22] This property arises because the cumulative distribution of total tax payments, when plotted against the income distribution, forms a revenue-weighted combination of the individual tax Lorenz curves, preserving the linear structure of the index calculation.[23]This additivity enables straightforward decomposition of a multi-tax system's progressivity, isolating the contribution of each tax as r_j S_j, which reflects both its inherent progressivity (or regressivity) and its fiscal weight. For instance, a highly progressive income tax with substantial revenue share can offset the regressivity of consumption-based levies, yielding an aggregate index that quantifies net vertical equity.[21] Empirical applications, such as analyses of U.S. state tax systems, leverage this to evaluate how shifts in revenue composition—e.g., increasing reliance on sales taxes—affect overall progressivity without recalculating the full aggregate curve.[24]The property's utility extends to policy simulations, where marginal changes in individual tax parameters propagate predictably to the system level, aiding assessments of reform impacts on distributional outcomes. Unlike some alternative progressivity measures (e.g., Reynolds-Smolensky), the Suits index's decomposability holds under standard assumptions of additive tax liabilities across individuals, without requiring reranking for horizontal equity concerns.[25] However, it assumes tax revenues sum proportionally, which may not capture interactions like base overlaps or behavioral responses altering effective shares.[26]
Path Independence
The Suits index possesses the property of path independence when aggregating progressivity across multiple taxes in a system. This means that the overall index value for the combined taxstructure remains unchanged regardless of the sequence or hierarchical grouping in which individual taxes are aggregated—whether pairwise or in subgroups—provided the population is consistently ordered by pre-taxincome and tax payments are summed per unit. This property arises directly from the index's mathematical structure, which ensures consistency in aggregation.[7]To see why, consider the definition: the Suits index S for a tax is S = 1 - 2 \int_0^1 C(p) \, dp, where C(p) is the tax concentration curve, representing the cumulative share of total tax payments up to the cumulative share p of income earners ranked from lowest to highest income. For an aggregate tax T = \sum T_i with revenues r_i = \sum T_i for each component i, the concentration curve C_T(p) = \sum (r_i / r) C_i(p), the revenue-weighted average of the component curves. Integrating yields \int C_T(p) \, dp = \sum (r_i / r) \int C_i(p) \, dp, so S_T = \sum (r_i / r) S_i, the revenue-weighted average of the individual indices. This linear additivity implies that intermediate aggregations—such as first combining taxes 1 and 2 to form a subgroup index S_{12} = (r_1 S_1 + r_2 S_2)/(r_1 + r_2), then aggregating with tax 3—always recover the global weighted average S_T, irrespective of order.[27]Path independence facilitates reliable decomposition in multi-tax analyses, allowing policymakers to assess overall progressivity without artifacts from arbitrary groupings, such as separating direct versus indirect taxes. For instance, empirical studies of U.S. state tax systems compute aggregate Suits indices by summing payments across income, sales, and property taxes, yielding consistent results aligned with this property.[7] In contrast, non-additive indices applied sequentially may produce order-dependent outcomes due to nonlinear effects on post-tax distributions, underscoring the Suits index's robustness for complex fiscal evaluations.[4] This feature has supported its use in cross-jurisdictional comparisons, where tax bundles vary, ensuring the measure reflects inherent progressivity rather than methodological path.[28]
Applications Across Tax Types
Progressive Taxes like Income Tax
The Suits index quantifies the progressivity of income taxes by comparing the cumulative share of tax payments to the cumulative share of income, yielding a value greater than zero when higher-income individuals bear a disproportionately larger tax burden. For graduated income taxes, which feature escalating marginal rates, the resulting tax concentration curve lies above the line of equality, with the index value S = 2 \int_0^1 L_t(p) dp - 1, where L_t(p) is the tax Lorenz curve; higher S reflects steeper rate graduation and fewer exemptions favoring low earners.[10][29]Empirical applications to the U.S. federal individual income tax demonstrate substantial progressivity, though varying over time due to legislative changes; for instance, post-World War II structures with top marginal rates above 90 percent produced elevated index values, which declined amid rate cuts in the 1960s and 1980s as measured by analogous progressivity metrics tracking similar trends.[15] State-level income taxes similarly exhibit positive Suits indices; in Minnesota, the individual income tax index is calculated at 0.2539 under prevailing law, reflecting bracketed rates from 5.35 percent to 9.85 percent applied after standard deductions and credits.[30]This progressivity serves a redistributive function within broader tax systems, often offsetting regressive components like sales taxes; the index's additivity property enables aggregation, showing income tax contributions to net fiscal progressivity—for example, U.S. state studies link higher income tax Suits values to reduced overall regressivity despite flat consumption levies.[5] Policy alterations, such as broadening bases or altering exemptions, directly influence the index; exempting portions of income, like Social Security benefits, can modestly elevate S by concentrating liabilities on non-exempt higher earners.[30]Critics note that statutory calculations assume legal incidence on payers, potentially overstating progressivity if economic incidence shifts burdens via behavioral responses like labor supply reductions among high earners, though empirical validations using microdata affirm the index's robustness for statutory analysis.[15] In international contexts, progressive income taxes in OECD countries yield comparable positive Suits values, correlating with top rate levels but tempered by deductions and avoidance opportunities.[18]
Regressive Taxes like Sales and Excise
Sales and excise taxes impose uniform rates on consumption expenditures or specific goods, resulting in regressive distributions where lower-income households bear a higher burden relative to their income, as quantified by negative Suits indices. This occurs because poorer individuals allocate a larger share of income to taxable consumption, while wealthier ones save or invest more, evading the tax base.[31] The Suits index for these taxes falls below zero, with the Lorenz curve of cumulative tax payments versus cumulative income bowing below the equality line, indicating disproportionate contributions from lower earners.[32]State-level tax incidence studies provide concrete examples of this regressivity. In Minnesota, the 2007 Tax Incidence Study reported a Suits index of -0.175 for general sales and use taxes in 2004, with effective rates of 5.5% for the lowest income decile versus 1.2% for the highest.[31] Connecticut's 2019 Tax Incidence Study calculated -0.29 for sales and use taxes, noting that the first three income deciles paid the majority despite comprising a smaller income share.[32]Excise taxes on targeted goods like fuels, alcohol, and tobacco amplify regressivity, as consumption patterns skew toward lower-income groups for these necessities or habits. Minnesota's analysis yielded -0.253 for motor fuels excise in 2004 and -0.486 for cigarette and tobacco excises, reflecting extreme disproportionality.[31] A 2015 Minnesota study on alcohol beverage excises reported -0.248, confirming regressivity with 89% of collections borne by residents but disproportionately by lower earners.[33] Connecticut's excise aggregate scored -0.36 in 2019, with the four lowest deciles covering nearly 66% of payments at 0.77% average income share.[32]
These values demonstrate the Suits index's sensitivity to consumption-based structures, aiding policymakers in assessing offsets like exemptions or rebates, though unadjusted implementations consistently show negative progressivity.[31][32]
Aggregate Tax System Analysis
The Suits index applied to an aggregate tax system measures the net progressivity by integrating tax payments from all revenue sources into a unified framework, plotting the cumulative share of total taxes against the cumulative share of income. This aggregate index, denoted as S, is mathematically equivalent to the revenue-weighted average of the individual Suits indices for each tax type, where weights are the shares of total revenue contributed by each tax: S = \sum (r_m \cdot S_m) / \sum r_m, with r_m as the revenue share and S_m as the index for tax m.[21] This additivity property, inherent to the Lorenz-curve-based construction, enables decomposition of overall system progressivity without requiring a full re-estimation of the combined tax distribution curve.[4]In empirical applications, this aggregation reveals how progressive income taxes can counterbalance regressive consumption or property taxes, determining the system's overall tilt. For example, U.S. state and local tax systems frequently exhibit negative aggregate Suits indices, indicating regressivity, because sales and property taxes—typically with indices around -0.2 to -0.4—generate substantial revenue despite income taxes yielding positive indices of 0.1 to 0.3.[34]The Institute on Taxation and Economic Policy's analysis of 2022 data across all 50 states found most systems regressive, with aggregate indices ranging from -0.09 in Washington to near proportionality (approximately +0.004) in Vermont and Minnesota, where stronger income tax progressivity offsets other components.[24]This metric's utility in policy evaluation lies in quantifying shifts from tax reforms; for instance, Minnesota's 2013 tax changes raised the aggregate Suits index by about +0.01 toward zero, lessening regressivity through targeted income tax adjustments.[13] Internationally, similar aggregations in low-income countries highlight challenges, as reliance on indirect taxes often results in negative indices, though data limitations constrain precise cross-country comparisons.[18] Critics note that such analyses assume statutory incidence aligns with economic incidence, potentially overstating progressivity if behavioral responses shift burdens.[7]
Empirical Implementations and Case Studies
Early Applications Post-1977
One of the earliest empirical extensions of the Suits index appeared in a 1982 analysis of transportation financing alternatives, where it was applied to evaluate the progressivity of proposed highway funding mechanisms in the United States. The study compared options including gasoline taxes, vehicle sales taxes, and registration fees against income distributions, finding that flat-rate fuel levies exhibited regressivity with Suits index values below zero, while ad valorem sales taxes on automobiles showed mild progressivity due to higher burdens on wealthier vehicle owners. This application underscored the index's adaptability to sectoral policy debates, particularly in assessing how consumption-based taxes deviate from proportionality across income quintiles.[35]In 1984, Donald W. Kiefer further implemented the Suits index to examine tax progressivity variations across diverse household configurations, using U.S. tax data to compute S values under statutory incidence assumptions. His calculations revealed that progressivity measures differed significantly by family type—for instance, single filers faced higher effective S indices than large families with dependents, reflecting how exemptions and deductions alter cumulative tax shares relative to income Lorenz curves. Kiefer's work also tested sensitivity to alternative progressivity definitions, confirming the Suits index's robustness for isolating average rate changes from liability concentrations, though it highlighted limitations in capturing marginal rate effects for heterogeneous units.[36]These post-1977 implementations, primarily in academic policy analyses, demonstrated the index's practical value for dissecting U.S. federal and statetax structures, often yielding S values between -0.2 and 0.3 for combined systems depending on incidence shifts like forward versus backward tax pass-through. Early adopters emphasized its path independence for aggregating multi-instrument progressivity, facilitating comparisons of pre- and post-reform scenarios, such as the impacts of 1981 federaltax cuts on overall regressivity. However, applications remained constrained by data granularity, with reliance on aggregated income surveys introducing potential underestimation of top-end concentrations.[37]
Contemporary Uses in Policy Analysis
In analyses of U.S. federal and state fiscal systems, the Suits index has been used to quantify tax progressivity among working-age households, ranking them by pre-government income and comparing cumulative tax burdens to income shares. A 2025 study by Heathcote, Ventura, and Violante applied the index to 2015–2016 data, finding the federal system progressive with an index of 0.195, driven by income and payroll taxes, while state systems averaged near proportionality at -0.004, reflecting a mix of progressiveincome taxes and regressive sales and property taxes. Including corporate and business taxes slightly reduced state progressivity to -0.011, with policy implications highlighting how regressive state taxes in southern U.S. regions may attract high-income migrants and erode revenue bases.[7]The index has informed evaluations of environmental tax policies, particularly for assessing distributional impacts on low-income groups. In a 2021 Swedish study on CO2 emission reductions from passenger cars by 2030, fuel tax increases yielded Suits indices of -0.056 in large cities and -0.096 in rural areas for octiles 2–8, indicating regressivity, while intensified bonus-malus systems on high-emission vehicles were more regressive at -0.117 and -0.161, respectively, underscoring higher burdens on rural and lower-income households. Similarly, a 2024 analysis of U.S. household carbon footprints under population aging scenarios deemed a uniform carbon tax regressive with a Suits index of -0.13, necessitating revenue recycling to mitigate effects on lower-income and elderly households.[38][39]In developing country contexts, the Suits index features in World Bank assessments of tax system design for low-income nations, where it complements other metrics to gauge progressive capacity amid reliance on regressive indirect taxes. A 2023 World Bank paper referenced the index alongside Kakwani measures to evaluate progressivity approaches, noting its utility in plotting tax paid distributions against income for policy recommendations on enhancing vertical equity without deterring compliance. Such applications emphasize the index's role in informing reforms that balance revenue needs with distributional fairness, particularly where data limitations challenge precise estimation.[18]
Comparisons to Alternative Measures
Relation to Gini Coefficient
The Suits index adapts the Gini coefficient's methodology to quantify tax progressivity by measuring deviations from proportionality in the distribution of tax burdens relative to income. The Gini coefficient assesses income inequality as twice the area between the Lorenz curve—cumulative income share plotted against cumulative population share—and the 45-degree line of equality, yielding values from 0 (perfect equality) to 1 (maximum inequality). In contrast, the Suits index constructs a parallel curve using cumulative tax share on the vertical axis against cumulative income share on the horizontal axis, with taxpayers ranked from lowest to highest income; the index equals twice the area between this curve and the proportionality line (where tax share equals income share), ranging from -1 (extreme regressivity, curve above the line) to +1 (extreme progressivity, curve below the line).[1][29]This income-share axis distinguishes the Suits index from the Gini's population-share axis, enabling direct evaluation of whether taxes deviate from proportionality—a core criterion for progressivity—without conflating it with per-capita equality. For a proportional tax, the Suits curve coincides with the diagonal regardless of income inequality, yielding an index of 0; progressivity emerges when lower-income groups contribute less than their income share to total taxes (curve below diagonal), concentrating the burden on higher incomes. Empirically, the index's formula, S = 1 - 2 \int_0^1 C_t(u) \, du where C_t(u) is the cumulative tax share up to income share u, mirrors the Gini's G = 1 - 2 \int_0^1 L(p) \, dp with L(p) as the Lorenz function and p population share, but substitutes income ranking to align with tax base relevance.[40][21]Algebraically, the Suits index relates to Gini coefficients as S = G_t - G_y for an income tax, where G_t is the Gini coefficient of tax payments with individuals ranked by income, and G_y is the standard income Gini coefficient. This decomposition shows progressivity as the excess concentration of taxes beyond income concentration: proportional taxation implies G_t = G_y (identical distributions), while G_t > G_y signals progression by amplifying inequality in tax burdens relative to income. Such formulation underscores the indices' shared covariance basis—Gini as G = \frac{2}{\mu_y} \cov(y, F_y(y)) approximately, and Suits extending it to tax-income covariance—facilitating comparisons in policy analysis, though the Suits index isolates tax structure effects from demographic inequality captured by the Gini.[21][40]
Differences from Kakwani and Reynolds-Smeeding Indices
The Suits index quantifies tax progressivity through the difference between the income Lorenz curve and the tax concentration curve, yielding a coefficient S equal to twice the area between these curves, ranging from -1 (fully regressive) to 1 (fully progressive), with 0 indicating proportionality.[15] The Kakwani index employs an equivalent mathematical formulation, defined as K = C_t - G, where C_t is the concentration coefficient of tax payments (ranked by income) and G is the Gini coefficient of pre-tax income; for progressive taxes, C_t exceeds G, producing a positive K.[18] Although numerically identical under standard assumptions—both capturing the deviation of tax shares from income shares—the indices differ in interpretive emphasis: the Suits index focuses on departure from strict proportionality to income ranks, while the Kakwani index highlights the tax's role in mitigating preexisting income inequality, leading to claims of fundamental distinction in axiomatic properties despite convergent empirical results.[9][41]In contrast to these shape-focused measures, the Reynolds-Smolensky index (often referenced alongside Smeeding in inequality studies) assesses redistribution as the reduction in the Gini coefficient from pre-tax to post-tax income, RS = G_pre - G_post.[42] This decomposes into a progressivity component (akin to the Suits or Kakwani index) multiplied by the average tax rate (tax revenue as a fraction of pre-tax income), rendering it sensitive to the scale of taxation.[18][15] Consequently, the Suits index remains invariant to uniform scaling of tax rates—ideal for cross-system comparisons of structural progressivity—whereas the Reynolds-Smolensky index diminishes if average rates fall, even if the relative burden distribution (Suits value) holds constant, emphasizing actual inequality compression over intrinsic tax design.[42] Empirical analyses confirm that high Suits values correlate with positive Reynolds-Smolensky effects only when tax burdens are substantial, as seen in U.S. federal tax data where progressivity trends align but redistribution varies with revenue shares.[15]
Criticisms, Limitations, and Debates
Assumptions on Statutory vs. Economic Incidence
The Suits index attributes tax burdens to income groups based on assumptions about incidence, distinguishing between statutory incidence, which assigns the tax to the party legally responsible for remittance (e.g., consumers for sales taxes or property owners for real estate levies), and economic incidence, which accounts for the actual burden distribution after accounting for behavioral responses, supply-demand elasticities, and market adjustments. Standard applications of the index, as in Daniel Suits' original 1977 formulation and subsequent state-level analyses, predominantly adopt statutory incidence for practical data availability, treating indirect taxes like sales and excise as fully borne by final consumers based on expenditure surveys.[1][32] This approach simplifies computation by relying on reported tax remittances or consumption patterns without modeling equilibrium shifts.[43]A key limitation arises because statutory incidence often diverges from economic incidence, potentially distorting progressivity estimates. For regressive taxes such as sales taxes, statutory assumptions attribute the full burden to households proportional to spending, exaggerating regressivity since lower-income groups allocate a higher income share to consumption; economic incidence, however, may shift portions of the burden to producers via reduced producer surplus or wages if supply elasticities are low, mitigating this effect based on empirical elasticity estimates.[44] Similarly, property taxes under statutory incidence are assigned to owners, yet economic incidence frequently passes costs to renters through higher rents or to capital owners via lower returns, with studies showing modest differences in Suits index values (e.g., -0.41 under both for Connecticut property taxes in 2019, but larger variances for other levies).[32] Corporate income taxes, when included, exhibit high sensitivity to incidence assumptions, with statutory treatment on firms yielding different progressivity outcomes compared to economic allocations to shareholders or labor.[45]Critics contend that overreliance on statutory incidence in Suits index calculations neglects these dynamic effects, leading to overstated regressivity for indirect taxes and incomplete evaluations of aggregate system progressivity, as evidenced by varying index values across incidence scenarios in fiscal studies.[46] While economic incidence requires uncertain parameter estimates (e.g., elasticities from econometric models), its omission assumes no burden shifting, which contradicts causal mechanisms in tax theory where taxes alter relative prices and incentives.[26] Advanced applications attempt adjustments, such as in Minnesota's 2024 tax incidence report incorporating behavioral responses, but standard Suits implementations remain vulnerable to this critique, underscoring the index's dependence on auxiliary assumptions rather than purely empirical tax-payment data.[47]
Neglect of Behavioral and Evasion Effects
The Suits index measures tax progressivity using the observed distribution of reported tax liabilities relative to reported income shares, thereby neglecting behavioral responses to taxation that alter pre-tax income and economic activity. Such responses include reductions in labor supply, shifts toward untaxed forms of compensation, or decreased investment, which can erode the intended progressivity by changing the underlying income distribution upon which the index is calculated. For instance, empirical estimates of the elasticity of taxable income with respect to marginal tax rates typically range from 0.2 for the general population to 0.6 or higher for top earners, indicating that higher tax rates prompt substantial adjustments in reported income that static measures fail to incorporate.[48] Analyses of progressivity indices, including those akin to Suits, highlight that ignoring these channels overstates true progressivity, as taxes influence pre-tax outcomes through incentives rather than solely post-tax redistribution.[49]This static approach assumes a fixed behavioral baseline, disregarding dynamic effects where progressive taxation may discourage productivity or encourage relocation of economic activity to lower-tax jurisdictions. Studies on tax reforms demonstrate that behavioral elasticities amplify with progressivity, potentially reducing revenue yields and altering incidence in ways unreflected in the index's Lorenz-curve-based computation.[50] For example, higher marginal rates at the top can induce income deferral or conversion to capital gains, distorting the cumulative tax shares used in Suits calculations and leading to underestimation of deadweight losses associated with progressivity.[4]Regarding evasion, the Suits index relies on administrative taxdata, which systematically underreports income due to noncompliance, particularly at higher income levels where opportunities for concealment are greater. Top earners exhibit higher evasion rates through mechanisms like offshore accounts or underreporting business income, with estimates suggesting unreported income comprises 20-30% of true top-1% earnings in various contexts.[18][51] This omission biases the index by compressing measured income inequality, as evaded amounts are absent from the denominator, potentially exaggerating progressivity if taxes appear concentrated on a narrower reported base; conversely, if evasion reduces reported tax payments disproportionately, it understates progressivity. Progressive structures exacerbate evasion incentives, as the marginal benefit of hiding income rises with rates, yet the index treats reported figures as veridical without adjustment.[52] Empirical adjustments for evasion, such as those incorporating audit data, reveal that conventional progressivity metrics like Suits systematically misrepresent distributional impacts in evasion-prone environments.[53]
Data and Measurement Challenges
The Suits index, which plots the Lorenz curve of cumulative tax payments against cumulative pre-taxincome shares, demands granular microdata on both income and tax liabilities across the population to construct accurate distributions. Empirical calculations typically require sorting individual or household records by income, computing cumulative proportions, and integrating under the resulting curve, a process that is computationally feasible with standard software but heavily reliant on comprehensive datasets like national tax administrative records or household surveys.[54] Incomplete coverage, such as excluding the top or bottom deciles, can distort the index, as full-sample estimation necessitates data spanning the entire income spectrum to avoid biasing progressivity estimates toward zero or incorrect signs.[32]Data availability poses significant barriers, particularly for indirect taxes or informal economies, where tracing incidence to income groups is impeded by the absence of linked taxpayer identities or consumption records. In low-income countries, these limitations often exclude key revenue sources like value-added taxes or customs duties from analysis, yielding incomplete aggregate indices that understate overall system progressivity or regressivity.[18] Even in advanced economies, prior studies have circumscribed their scope to partial distributions—such as wage earners or specific tax types—due to gaps in merged income-tax datasets, which hampers representation of capitalincome or self-employment earnings that disproportionately affect high earners.[29]Measurement inconsistencies further complicate reliable computation, including variations in income definitions (e.g., gross versus net, realized versus accrued) and units of analysis (individual versus family), which can alter the Lorenz curve's shape and thus the index value by up to several points in empirical applications.[15] Harmonizing data across jurisdictions for cross-country comparisons exacerbates these issues, as differing survey methodologies or reporting thresholds lead to non-comparable distributions; for instance, reliance on self-reported surveys often undercaptures top-incometax contributions relative to administrative filings.[55] These challenges underscore the index's sensitivity to input quality, prompting researchers to supplement with robustness checks, such as bootstrapped confidence intervals from available microdata, though such mitigations cannot fully resolve inherent gaps in coverage or accuracy.[56]