RESEARCH DOSSIER
International Tax Competition
Abstract. International tax competition is strategic fiscal interaction among jurisdictions seeking to influence the location of mobile investment, reported corporate income, and related economic activity. This monograph examines its theoretical foundations, cross-border fiscal externalities, tax-rate measures, international rate trends through 2026, asymmetries, competition for real investment and accounting income, distributional incidence, empirical identification, policy evaluation, and international cooperation. It synthesizes selected scholarly studies with official comparative statistics and minimum-tax institutional documentation. Transparent hypothetical calculations separate private location incentives from domestic and collective revenue effects. The central analytical finding is that the welfare consequences of competition depend on the mobility of the particular tax base, the location of economic rents, the financing and value of public expenditure, and the feasibility of alternative policy instruments. A stable headline tax rate is compatible with continuing strategic interaction through tax bases, incentives, and expenditures. The study is an interpretive public-finance synthesis within Offshore Economics; it does not claim an original econometric estimation or a newly established theory.
Keywords: International tax competition; fiscal externalities; corporate income taxation; effective tax rates; foreign direct investment; profit shifting; tax incidence; tax incentives; offshore economics; global minimum tax; international tax coordination. JEL classification: H25, H26, H77, H87, F23
Introduction
International tax competition is the strategic interaction of fiscal jurisdictions whose tax decisions alter the location, volume, legal attribution, or net return of internationally mobile economic activity. The interaction occurs when a government chooses its rates, bases, incentives, or administrative practices in light of the fiscal choices of other governments and the expected responses of taxpayers. A tax reform becomes an instance of international competition when its cross-border incentives and expected foreign responses enter the decision environment. International differences in taxes are observable outcomes; strategic interaction is the economic mechanism to be investigated.
This dossier studies that mechanism as a component of the global offshore system. Corporate groups can organize production, financing, intellectual assets, investment ownership, and accounting across jurisdictions. Governments retain the authority to determine taxes within their legal competence, yet each decision has consequences for foreign tax bases. Geographic dispersion of corporate activity therefore contains an endogenous fiscal dimension. Some jurisdictions seek investment that employs people and purchases local inputs; others may seek financial intermediation, taxable accounting income, or corporate registration. The economic payoffs and costs differ even when the statutory instrument is identical.
The central research question is how international tax competition changes the distribution of public revenue, investment, and economic rents when tax bases have heterogeneous mobility. Four subsidiary questions organize the analysis. What tax measures correspond to particular business decisions? Why do governments face incentives to change them? Which empirical observations distinguish tax-induced relocation from profit reallocation or common economic shocks? Under what conditions can cooperation improve the outcome, and how does it change the instruments available to national governments?
The unit of analysis is the taxing jurisdiction as a fiscal decision-maker, with firms, investors, workers, and public institutions as transmission mechanisms and affected parties. Corporate income taxation receives primary attention because it links cross-border business activity to public revenue and because it admits a well-developed theoretical and empirical literature. Personal taxation, commodity taxes, and transfer taxation enter where they illuminate the logic of strategic interaction. Legal doctrine, specialized avoidance channels, investment incentive treatment effects, and the detailed global minimum-tax reform have independent research objects within the program. Their relevant economic implications are integrated here without duplicating those objects.
The study combines literature synthesis, comparative public finance, examination of official statistical definitions, and explicitly hypothetical calculations. It does not present newly collected firm-level data, replicate published regressions, or claim a newly established theory. Its contribution is an analytical assessment of policy interactions and evidentiary boundaries. The institutional discussion is bounded by materials identified through November 18, 2026; individual jurisdictions' current statutes and implementation dates require separate primary-law research.
The economic object and its boundaries
The conventional corporate income tax is a levy on an institutionally defined measure of corporate profit. A government can affect the resulting liability by changing the nominal rate, deductible expenses, timing of allowances, loss relief, exemptions, credits, rules for cross-border income, or the administration of those rules. Each instrument changes the after-tax payoff associated with some set of decisions. The intensity of competition is consequently a property of the relevant decision margin and its institutional environment.
Consider a corporation that has already constructed a factory. Its choice of whether to expand the plant depends on the incremental net return to new capital. Its decision about a second, mutually exclusive factory can depend on the discounted profit from each feasible location. Its decision about the fiscal attribution of income already generated can depend on the tax incurred on an additional unit of taxable income in alternative jurisdictions, as constrained by applicable law. The corporation's choices over financing, operational activities, and location of taxable profit interact but remain analytically distinct.
The government faces an equally differentiated decision environment. A reduction in tax on new equipment may affect the scale of investment more directly than a reduction in the rate applied to profits already earned. A broadly applicable rate cut has implications for inframarginal firms as well as mobile entrants. A geographically targeted incentive may redistribute activity within the same country. A rule change governing cross-border profit can alter revenue without generating new machinery or labor demand. Each instrument needs a counterfactual specifying which of these movements would have occurred at unchanged policy.
Keen and Konrad's (2013) survey of tax competition brings these distinctions into the theoretical literature by addressing physical capital movement and jurisdictional profit allocation within a common subject. The resulting research object has a wider scope than a comparison of offshore financial centers. The offshore economy is particularly informative because it separates legal, financial, and accounting locations from physical activity, making the economic channels easier to distinguish conceptually.
A competition question also differs from a tax-compliance question. Lawful changes in the location of investment, eligible tax preferences, disputed arrangements, administrative noncompliance, and criminal evasion carry different evidentiary and legal implications. A low tax burden is not itself evidence of wrongdoing. Conversely, legality alone does not answer whether a policy generates fiscal spillovers or an inefficient allocation of resources. This study evaluates economic consequences while reserving legal classification to applicable law and established facts.
Strategic interdependence can exist without an explicit announcement that a government seeks to attract foreign firms. Budget forecasts may already anticipate changes abroad. Domestic political actors can compare their tax environment with competitors and alter proposals. Ministries can negotiate alternative incentives when one instrument becomes constrained. Inference about strategy must nonetheless distinguish documented policy intentions, observed policy correlation, and theoretically predicted responses. The existence of mobile capital creates the opportunity for competition, not automatic proof that every rate change was caused by it.
Historical development and the meaning of a race to the bottom
Modern theoretical analysis began with the implications of geographically mobile resources for decentralized fiscal choices. Zodrow and Mieszkowski (1986), in their study of local public goods, demonstrate how taxation of mobile capital in their competing-local-government setting can reduce the public services supplied to residents. The international extension changes the scale and political institutions of the problem: national governments possess differentiated tax systems, currencies, economic structures, public expenditures, and powers to enter international agreements. The underlying fiscal interdependence nevertheless survives.
The contemporary offshore system intensified scholarly attention to the mobility of accounting income. Hines and Rice (1994), in Fiscal Paradise, documented unusual profitability patterns among historical US multinational affiliates in tax havens using data for 1982. Their evidence belongs to that corporate population and period. It demonstrates why identifying economic activity solely through reported corporate income can be misleading; it does not yield a current global share or a universal optimal tax rate for small jurisdictions.
During the late twentieth and early twenty-first centuries, governments increasingly treated the tax system as one component of the international investment environment. The policy debate developed alongside concern about the loss of taxable income. The OECD's 1998 report on harmful tax competition explicitly addressed the erosion of national tax bases and the distribution of tax burdens while proposing guidelines against harmful preferential regimes. Its institutional significance lies in defining an area for cooperation. Its normative characterization should be read together with its scope and historical setting, rather than treated as a contemporary designation of every jurisdiction with a low rate.
The phrase race to the bottom can describe a process of mutually reinforcing reductions in fiscal burdens on a mobile base. It becomes a scientific proposition only when the relevant burden, participating jurisdictions, time horizon, causal mechanism, and benchmark outcome have been specified. The observation that major economies lowered headline rates over several decades is compatible with strategic rate reductions, domestic simplification, changes in the composition of tax bases, and general changes in taxation philosophy. A descriptive time series cannot isolate these explanations.
The phrase also contains several possible endpoints. A race in one statutory rate need not culminate at zero. Inframarginal domestic profits, location-specific rents, administrative feasibility, tax-funded infrastructure, and political preferences can sustain positive tax rates. Even a temporary period of stable rates can characterize an equilibrium with substantial cross-border tax effects. A government may increasingly use tax relief for particular investment activities while leaving the general rate unchanged. Competition may shift from the nominal rate to another feature of the tax system.
The empirical question is therefore not whether a single global average continually falls. It is how governments' policy instruments respond to one another and how the aggregate welfare consequences differ from the policy choices that would prevail if each fiscal authority accounted for the full set of affected jurisdictions. The historical record motivates this question without determining its answer.
Fiscal externalities and the benchmark economics
The basic fiscal externality can be illustrated with a familiar public-finance identity. Let a jurisdiction's taxable corporate base B depend on its own tax rate t, the foreign rate s, and other conditions. Its gross revenue from the simplified tax is R = tB(t,s). Holding s and other policies fixed, its marginal revenue from raising t is B + t(∂B/∂t). When a higher domestic rate reduces the domestic base, the second term reduces the direct revenue gain. A foreign fiscal effect arises if the domestic policy changes the foreign tax base.
In a closed, simplified two-jurisdiction world with perfectly conserved aggregate taxable income, a unit of reported profit lost by the first jurisdiction is gained by the second. If a rate increase in jurisdiction A causes reported income to move to B, A experiences lost tax base and B experiences increased base. The income reassignment can occur without a change in output. A national decision-maker that counts only domestic receipts omits the foreign revenue consequence. When the base consists of productive investment, the effects additionally include changes in wages, prices, output, and the benefits of public expenditure.
Suppose initially A and B each report taxable profit of 100 monetary units and charge 25%. Each receives 25; aggregate revenue is 50. A reduces its rate to 15%. With no relocation, A would collect 15 and B 25, totaling 40. Now assume, solely for illustration, that 80 units of taxable income migrate from B to A while underlying activity stays constant. A collects 27 on its new base of 180; B collects 5 on its residual base of 20. The combined revenue is 32. A gains 2 relative to the initial fiscal position, B loses 20, and the combined loss is 18. The example requires fixed worldwide income, no behavioral change beyond reassignment, and no cross-border top-up or other countermeasure. It is an accounting exercise, not a calibrated prediction.
Consider instead a smaller response of 20 units. A collects 18 and B collects 20. A sacrifices 7 relative to the original 25, even though it attracts income. The effect on combined receipts is a loss of 12. In this example A needs more than 66.67 units of additional base for the 15% rate to exceed its initial revenue. The condition follows from 0.15(100 + x) > 25. The attraction of tax base and the profitability of attracting it are separate economic tests.
The global welfare comparison requires more than revenue arithmetic. Public receipts fund services, transfers, debt repayment, or other uses that have social value and opportunity cost. Firms' increased after-tax profits also belong in a comprehensive welfare evaluation, with the relevant distributional weights and ownership geography. Cross-border relocation might improve total productive efficiency, diminish it, or leave it unchanged. Even if world taxable profit is conserved, worldwide real income and the distribution of its claims can change.
A policy can be individually rational in the revenue calculation while producing a collectively expensive equilibrium. If B responds by cutting its own rate, the original advantage of A may diminish. The value of government coordination in this benchmark comes from bringing the foreign revenue and resource effects into the choice. Its value depends on whether cooperation can be enforced and whether the participating governments' welfare objectives genuinely encompass the affected populations.
Fiscal externalities can also operate in the opposite direction. A country's restriction on a distortionary preference may expand the tax base of another country, or competition can induce more efficient services. The sign and magnitude of total welfare effects depend on the base, fiscal objective, and real economic transmission. The arithmetic above isolates one plausible source of divergence; it establishes neither a necessary loss from all competition nor a universally preferred tax rate.
Government behavior, strategic interaction, and equilibrium
A country's treasury is neither a representative multinational enterprise nor a unitary global planner. Fiscal objectives arise from expenditure commitments, the political distribution of gains and losses, legal constraints, and the government's assessments of how investors respond. The textbook assumption that government maximizes the welfare of its residents helps identify externalities but is not itself an empirical description of policymaking. A government may weight public revenue, employment creation, investment announcements, political support, and regional development differently.
In a simple simultaneous-move description, every government chooses a feasible tax policy while anticipating that firms respond to the collection of policies. A Nash equilibrium is a profile at which no government can improve its specified objective by a unilateral deviation, given others' choices. This concept does not require countries to agree with the outcome or communicate with one another. It identifies a consistency condition for mutually responsive policy decisions.
The strategic effect depends on the objective and the relative mobility of the tax base. Tax decisions are strategic complements when an increase in foreign taxation makes a domestic increase more desirable under the government's best-response behavior. They are substitutes when the foreign increase makes a domestic reduction more desirable. A foreign reform can relieve pressure on the domestic government, induce renewed efforts to attract activity, or change the composition of optimal policies. The sign of a best-response derivative cannot be assumed from the word competition alone.
Devereux, Lockwood, and Redoano's (2008) study of corporate rate interaction separates real capital decisions related to marginal effective taxation from reported-profit decisions related to statutory rates. This distinction provides an empirical rationale for examining more than one policy variable. A single observed best-response coefficient can conceal several channels that have different effects on welfare.
The timing of national reforms also matters. A government can announce an incentive that is effective only after legislation or after a future investment threshold. A mobile firm may delay an irreversible investment while comparing proposals. A legislature may be unable to commit successors to unchanged rates. Competition then involves credibility and expectations as well as contemporaneous tax schedules. The fiscal consequences of a tax promise and a tax payment can occur in different budget periods.
There are limits to the use of game-theoretic language. A stylized policy game abstracts from constitutional procedures, domestic elections, heterogeneous firms, and the difficulty of observing other governments' beliefs. An observed policy response is not automatically proof of equilibrium, and an equilibrium of a simplified model is not a forecast of rates. The theoretical benchmark is useful when the approximations and the empirical identification problem are kept explicit.
Tax-rate measures and business decisions
Tax competition must be measured on the same margin as the policy's intended effect. The statutory corporate income tax rate, abbreviated STR, describes the generally applicable legal percentage for taxable profits under specified assumptions. It is a useful benchmark for the taxation of an additional unit of reported profit, subject to special regimes, tax-base definitions, cross-border rules, and the facts of the taxpayer.
A forward-looking effective marginal tax rate (EMTR) describes a tax wedge on a marginal project at the point where its pretax return just makes investment worthwhile. Its value depends on the assumptions governing financing, depreciation, interest, inflation, and assets. A marginal investment may qualify for a deduction or credit that changes its tax cost independently of the headline rate. An effective average tax rate (EATR) summarizes tax on a hypothetical profitable investment and helps analyze a discrete choice among mutually exclusive alternatives. Devereux and Griffith's (2003) work on location decisions develops the relevance of the discrete-choice problem for evaluating tax policy.
Neither forward-looking index is an observed company tax payment. The OECD's effective tax rate chapter (2026b) explicitly computes measures using hypothetical standardized investments and tax legislation parameters. Their value is consistency across jurisdictions. Their limitation is the abstraction from the particular facts of firms. An observed accounting effective tax rate is typically computed from reported tax expense and pretax income for a defined population and period; a cash tax rate uses payments rather than expense, often with material differences in timing. Negative earnings can complicate percentage interpretations. A separate internationally agreed minimum-tax computation can use another income denominator and covered-tax measure.
An elementary project comparison shows why the choice of metric matters. Suppose projects A and B each produce discounted pretax profit of 100. Under a deliberately simplified immediate and final tax calculation, the first faces a 25% tax and a project-specific credit of 12, producing liability of 13. The second faces a 15% tax with no credit, producing liability of 15. Project A has 87 of after-tax profit; B has 85. The project with the higher statutory rate yields the higher modeled payoff here. Real-world credits can have refundability rules, eligibility conditions, timing effects, and interactions with other taxes; none of those are represented in this illustration.
The same example cannot establish that A has a lower EMTR. Marginal projects and profitable discrete projects can respond differently to tax provisions. The credit could be a lump-sum amount tied to an existing asset and thus have no effect on further investment, or it could vary proportionally with new spending. To calculate the marginal wedge, one needs the policy's treatment of the incremental project, not merely the total liability in the example.
For revenue accounting, a tax-rate measure should be paired with the appropriate base measure and a timing convention. The corporate tax-to-GDP ratio compares receipts with domestic output and can remain stable despite rate cuts if bases expand, corporate profitability rises, or the composition of income changes. It can increase during an exceptional profit boom without any improvement in tax enforcement. Rates and revenue data are complements; using either in isolation weakens causal analysis.
The international rate record through 2026
The OECD's statutory corporate tax statistics (2026a) provide a dated empirical benchmark. In its reported regional comparison, the mean rate for OECD members fell from 32.3% in 2000 to 24.2% in 2026. For 36 Latin American and Caribbean jurisdictions the corresponding comparison was 26.9% and 21.1%. In 2026, the report gives means of 26.6% for 27 African jurisdictions, 20.6% for 36 jurisdictions in Asia and the Pacific, and 24.2% for OECD members. The broad covered-jurisdiction mean was around 21.2% throughout 2020–2026, indicating recent stability after the earlier long decline.
| Jurisdiction grouping | 2000 average STR | 2026 average STR |
|---|---|---|
| OECD members | 32.3% | 24.2% |
| Latin America and Caribbean (36 jurisdictions) | 26.9% | 21.1% |
| Africa (27 jurisdictions) | — | 26.6% |
| Asia and Pacific (36 jurisdictions) | — | 20.6% |
Source: OECD (2026a), reported regional means. Dashes indicate values not transcribed for this comparison. These jurisdiction averages are not weighted by global investment, corporate income, or a multinational group's actual exposure.
Several methodological consequences follow. First, a mean assigns the small and large jurisdictions their specified statistical weights; it cannot represent the rate faced by a dollar of internationally mobile profit unless the weights reproduce that exposure. Second, different regional means are compatible with different preferences, fiscal structures, and distributions of firm activity. Third, historical comparisons require attention to jurisdiction coverage and special regimes. Fourth, the period of stability overlaps major changes in tax cooperation but does not identify their causal effect. Many other post-2020 changes could affect the same national policies.
Policy evaluation should distinguish changes in the distribution of statutory rates from changes in effective taxation. A particular country's statutory rate may be unchanged while its accelerated depreciation rules alter the modeled marginal tax burden. Similarly, a concentration of low effective taxation in certain activities can persist within a jurisdiction that features a conventional general rate. The overall tax system includes multiple treatments that a single number cannot display.
For longitudinal research, the preferred design is a panel of consistently defined rates, tax-base features, incentive eligibility and economic outcomes, accompanied by records of each reform's announcement and commencement. The analytical question concerns the response of policy and economic activity to changes in the feasible opportunity set. The statistics establish observable levels and trends; their interpretation requires an empirically identified behavioral mechanism.
An editorial caution is necessary when comparing the values above with another release of the OECD database. A newly added country, change in inclusion criteria, correction to historical rates, or different averaging period could yield a different series. Subsequent research should archive the precise database extract alongside code that reproduces aggregate values. The present comparison transcribes selected published values rather than reconstructing the database.
Real investment, discrete location, and immobile rents
Real investment competition concerns the location or scale of productive assets and associated economic activity. A government that cuts tax in the expectation of a larger tax base anticipates more than a relocation of accounting entries. It may expect construction, employment, training, demand for domestic suppliers, innovation spillovers, and a greater capacity to produce. These are empirically testable consequences. Some may arrive long before the project generates taxable profits; others can persist after a preference ends.
An investment decision reflects a bundle of location attributes: infrastructure, access to buyers, complementary inputs, labor skills, political and regulatory predictability, financing conditions, and taxation. A tax advantage can change the ranking of otherwise comparable locations. It is less likely to compensate for a serious deficiency in a necessary production input. Therefore, the effectiveness of tax competition depends partly on its interaction with public expenditure. A government that finances a tax incentive by reducing the quality of an important service can undermine the investment value of its own offer.
Inframarginal returns offer another distinction. A company that would choose a country even at the original tax rate obtains a windfall if the rate is reduced for all taxpayers. That effect may be an intended redistribution, but it is a fiscal cost rather than evidence of induced investment. Location-specific rents, such as those associated with scarce natural resources, dense supplier networks, or a major domestic consumer market, may be less mobile than generic capital. A tax on those rents can have different efficiency effects from a tax on the return required to induce a new project.
Consider a hypothetical local project that generates 120 units of gross present-value profit under one location's infrastructure and 100 in another. If both face a uniform 25% tax, the after-tax comparison is 90 versus 75. Cutting the rate on the second project to 10% produces 90, merely bringing it to parity. A decision-maker must compare the expected additional local economic value with the public revenue foregone and any associated spending. If the underlying location advantage is uncertain, a larger subsidy may purchase a high probability of relocation but at a high fiscal price. The numbers illustrate ranking effects rather than investment forecasts.
Klemm and Van Parys (2009), in their empirical study of tax incentives, report results from a dynamic panel of more than 40 countries in Latin America, the Caribbean, and Africa for 1985–2004. Their evidence associates lower corporate rates and longer tax holidays with foreign direct investment measures, while comparable increases in aggregate private capital formation or growth were not established by that analysis. The result is historically and methodologically bounded. It directs attention to the outcome being measured: financial investment positions, physical investment, and national growth are different variables.
In the author's assessment, the economically relevant test for a targeted tax instrument is additional productive activity relative to a credible no-incentive baseline, not the gross value of the announced investment. Cost-effectiveness should be measured over the full life of the instrument. The government's policy portfolio includes the infrastructure and institutions that may generate durable location advantages. Investment competition is a joint question of taxes, public goods, and the distribution of economic rents.
Reported profit, offshore intermediation, and measurement
An internationally organized corporation may have a production location, a legal owner of assets, a financing location, a residence jurisdiction for tax purposes, and an accounting location for particular profits. These positions can coincide, overlap, or diverge. International tax competition changes the fiscal return associated with that institutional geography. The government benefits from taxable profit only to the extent that the profit enters its legal base and generates receipts.
Reported-profit mobility creates a particularly important externality because the relevant base can move faster than physical production. A rate differential can affect the allocation of income reported by affiliated firms without proportionately changing aggregate employment or capital stock. The literature addresses several channels of international corporate tax avoidance; Beer, de Mooij, and Liu (2018) organize the evidence in their review of avoidance channels. The present analysis relies on the scope of that review rather than transferring its numerical effect estimates into an unverified global forecast.
Tørsløv, Wier, and Zucman's (2023) research on missing profits provides an example of the use of international accounting data to analyze the geographic attribution of corporate profit. Their research page includes the final journal article and replication materials. The importance of the method for tax competition is that recorded output, trade, and factor income may themselves be affected by the geographic allocation of profits. Consequently, regressions involving those outcomes can inherit measurement problems related to the phenomenon they seek to explain.
Financial investment statistics also require care. Damgaard, Elkjaer, and Johannesen (2019) estimate in their analysis of the FDI network that close to two-fifths of global foreign direct investment might reflect positions in corporate entities lacking substantial links to the local economy. This is an estimate from a particular reconstruction, not a timeless proportion or an accusation of misconduct by individual entities. Their analysis emphasizes both the composition of FDI positions and the location of the ultimate investor.
A balance-sheet illustration clarifies the measurement issue. An investor in country A establishes a holding company in B, which in turn owns an operating company in C. An immediate-counterparty dataset may register A-to-B and B-to-C positions even though the operating investment is in C and the ultimate investor is in A. Summing both positions as new factories would be a conceptual error. The holding company can perform a legitimate function, and its local economic activities may vary. The researcher needs ownership structure, employment, assets, and income data to distinguish the roles.
For the assessment of international tax competition, changes in aggregate FDI should be compared with physical capital expenditure, establishment counts, employment, and wage payments. Changes in reported profit should be compared with value added, assets, and the ownership structure of firms. Even these measures have limitations because intangible-intensive production may legitimately generate high profits with modest tangible assets. A statistical mismatch is evidence for further inquiry; it is not dispositive evidence of a legal violation.
A revenue-seeking jurisdiction may rationally accept a small share of a large volume of mobile income while a larger economy loses receipts on profits associated with its productive activity. An economic analysis must identify that mechanism without presuming that the jurisdiction's financial services are all artificial or that reported profit is independent of legitimate value creation. The strength of the offshore example is precisely that multiple kinds of economic activity and fiscal location coexist.
Asymmetric jurisdictions and the political economy of mobility
Governments differ in population, economic scale, capital endowments, dependence on corporate receipts, access to natural resources, and bargaining power. Competition among unequal jurisdictions can produce different preferred fiscal strategies. A small state can sometimes acquire a large tax base relative to its original revenue, while a large state applying the same rate change would sacrifice substantial receipts from firms whose activity is already established. The absolute and relative magnitudes matter.
Kanbur and Keen (1993) analyze tax coordination and country size in a two-country commodity-tax setting. Their analysis illustrates the importance of asymmetry for the inefficiency of noncooperative policy. The particular behavioral assumptions belong to their model; the application to corporate taxation is an analogy about relative mobility and fiscal scale, rather than a direct transfer of estimated coefficients.
Consider two jurisdictions with baseline taxable corporate bases of 1,000 and 20 units respectively. At a 20% tax each receives 200 and 4. The smaller jurisdiction reduces its rate to 10% and attracts 35 units from the larger. It now earns 5.5 instead of 4, an increase of 37.5% in its original corporate-tax revenue. The larger receives 193 instead of 200, a decrease of 3.5%. Combined revenue goes from 204 to 198.5. The example assumes transferable reported profit, no changes in productive activity, and no response from the larger jurisdiction. It illustrates how the same reallocation translates into very different proportional political incentives.
A government representing a small population may give substantial weight to a specialized export service sector that generates fees, professional employment, and fiscal receipts. Such specialization can also increase exposure to regulatory change or a concentrated revenue base. Larger economies possess advantages grounded in consumer markets, industrial ecosystems, and the size of existing taxable activity. They may be able to sustain higher rates on some location-specific returns, but they are not uniformly insulated from mobile accounting income.
Different exposures make international bargaining distributional. A coordinated minimum can benefit a large source jurisdiction by reducing the attractiveness of profit reassignment while imposing adjustment costs on a jurisdiction that has relied on low taxes. Even a policy that increases combined revenue can be difficult to sustain if the gains accrue to one group and adjustment costs to another. Negotiations over collection rights, transition periods, and technical capacity reflect this difference in economic interests.
Asymmetry also affects the meaning of a nominally identical policy instrument. A grant of one million dollars can be fiscally minor in a large country and significant in a small one. Conversely, a large country's public budget can provide a more generous investment subsidy in absolute terms. Rules limiting one form of competition may favor states able to use alternative instruments. The evaluation of coordination must include that distribution of practical policy capacity.
Developing economies and administrative constraints
Developing economies often face an investment financing problem, constraints on tax administration, demands for essential public services, and a narrower set of reliable fiscal instruments. International investment can contribute to productive capacity, export activity, and economic diversification. A government seeking such outcomes must weigh tax preferences against the alternative uses of public funds and the sensitivity of the tax base to cross-border decisions.
Crivelli, de Mooij, and Keen's (2015) study of corporate tax spillovers examines a panel of 173 countries across 33 years and distinguishes spillovers associated with real choices from those associated with avoidance. Its results identify material base exposure for developing economies. The evidence is a macro-level analysis, not a direct measure of every country's current revenue loss. The working paper's year is 2015; the migration date embedded in the website path does not change the bibliographic year.
Tax administration is itself a scarce economic resource. Monitoring the taxable base of multinational groups may require access to financial information, accounting expertise, cross-border cooperation, and effective resolution of disputes. A complex incentive can demand more administrative resources than a simple standardized provision. Benefits that appear substantial under perfect administration may prove difficult to target in practice. Firms that face uncertainty about eligibility may discount the value of an announced preference, diminishing its intended investment effect.
A government also faces the risk of negotiating under incomplete information. A firm generally has more detailed information than the government about its alternative project locations and expected profitability. An incentive agreement can transfer rents to projects that would have entered anyway. Competitive pressure can make it difficult for a government to distinguish a credible alternative location from a strategic bargaining claim. Public evaluation, eligibility rules, and ex post assessment can reduce these information costs without eliminating them.
Fiscal consequences may persist beyond the annual budget cycle. A long-term preference can limit the tax base after a project becomes profitable or after public infrastructure costs have been incurred. Conversely, a project can produce beneficial employment and supplier activity during a period of low corporate receipts. The appropriate evaluation uses expected discounted fiscal and economic outcomes, including the cost and maintenance of public services.
International cooperation can expand effective administrative capacity through common information, compatible rules, and mechanisms that reduce redundant verification. Yet a complex multilateral system can also consume resources and introduce fixed costs that weigh disproportionately on smaller administrations. The appropriate choice depends on which constraints dominate. The long-run fiscal value of an investment can be improved through institutional capacity even where the initial private attraction of the location depends partly on favorable taxes.
Tax incidence and the distribution of economic burdens
A corporation's statutory liability is assigned to a legal entity. Economic incidence refers to changes in individuals' real incomes and consumption opportunities brought about by the tax and market adjustment. Depending on market structure, the economic burden can fall on shareholders, workers, consumers, suppliers, or owners of immobile property. The incidence of one jurisdiction's tax policy can extend across national borders because the affected firm's ownership, output market, and production inputs may be international.
In a conventional small open-economy benchmark with very mobile capital and relatively immobile labor, a tax on the marginal return can reduce domestic capital intensity. Under suitable production assumptions, the reduced capital stock lowers labor productivity and wages. This result follows from the assumed mobility and production conditions. Where capital consists of location-specific rents, expected returns exceed the internationally required return, or production is difficult to relocate, shareholders may bear a greater share. Imperfect competition, wage bargaining, and market power create additional effects.
Fuest, Peichl, and Siegloch (2018) examine German wage incidence using changes in municipal corporate taxation and linked employment information. Their published abstract reports an estimated worker burden of about half of the total tax burden in that setting. Its significance here is direct evidence that statutory corporate incidence and economic incidence can differ. The estimate belongs to the institutional environment and empirical design of that study; it is not a fixed global coefficient. Transferring the value to a different country, tax type, or time horizon would require independent justification.
The beneficiaries of international tax competition may include foreign shareholders and domestic workers, but those same categories may face costs under another combination of effects. Suppose a reduction delivers a windfall to foreign owners without changing local labor demand while reducing funding for public education. Domestic household welfare may deteriorate despite the corporation's higher after-tax return. Conversely, if the reduction generates genuinely additional capital that raises local earnings and the benefit exceeds the fiscal cost, domestic residents may gain.
Two distributional questions thus need separate answers. The first concerns who receives the gains from a tax reduction on existing or prospective private income. The second concerns who bears the financing consequences of lost public revenue. Governments can compensate through other taxes, lower spending, or borrowing. A decline in corporate receipts can lead to a larger tax burden elsewhere even while business conditions improve. The choice depends on domestic fiscal institutions, the composition of expenditure, and the time horizon.
International comparisons should therefore incorporate household and ownership data rather than inferring welfare from corporate tax receipts alone. A government objective of increasing GDP differs from an objective of increasing the real disposable incomes of residents. In a country with significant cross-border ownership, the distinction is material. The economics of tax competition concerns the distribution of claims on income alongside the allocation of productive resources.
Empirical identification of strategic tax competition
A research design must first specify its estimand. One question is whether a country changes its statutory rate in response to a foreign rate. A different question is whether foreign tax changes move the domestic base. Another is whether induced investment produces a net social benefit. These require distinct outcome variables and identification strategies. A paper estimating one effect cannot establish the others without additional assumptions and evidence.
In cross-country policy panels, researchers may model a domestic rate as a function of neighboring or economically similar jurisdictions' rates, domestic conditions, and common trends. The definition of the relevant foreign policy set is crucial. Geographic proximity can matter for cross-border consumer movement, while firm ownership, industry overlap, transportation access, and trade networks can define corporate competitors. Using observed investment links as weights can introduce endogeneity if tax policy itself changes those links.
Simultaneous determination creates a reflection problem. Domestic policy affects foreign policy while foreign policy affects domestic choices; both may respond to a common shock. An ordinary regression of domestic on foreign rates may capture a shared macroeconomic trend rather than strategic behavior. Country fixed effects help absorb stable institutional differences; time fixed effects can absorb common temporal movements. Neither removes every country-specific shock correlated with policies. Instrumental-variable approaches require credible exclusion restrictions, which become particularly demanding when foreign fiscal shocks also influence domestic trade or investment.
Reform-based designs compare treated and comparison jurisdictions or firms across time. A difference-in-differences design requires a suitable counterfactual trend and consistent treatment timing. Anticipation, overlapping reforms, policy spillovers into nominal controls, and changes in group composition can undermine interpretation. Event-study plots can reveal some timing patterns, but a visually flat set of prereform estimates does not prove all identifying assumptions. A threshold-based comparison may provide useful variation where treatment eligibility changes at a documented cutoff, subject to selection, sorting and possible simultaneous policies.
The relevant empirical literature therefore needs to be read at the level of its target population, data coverage, statistical model, and identifying assumptions. A government decision observed at the national level may be influenced by domestic institutions that vary from the municipal setting used in another study. Firm-level accounting data can mismeasure taxable income; aggregate tax receipts may include distinctive payments that distort intertemporal comparison. The researcher should assess whether the outcome approximates the underlying economic mechanism.
A credible empirical program would link four classes of observations: statutory rules and effective dates; exposures to foreign policy; firm or establishment outcomes; and public revenue. A rate-response analysis could then be compared with direct responses of investment, employment, profit, and tax paid. Reconciliation between these datasets is preferable to the use of one proxy for all channels. Where identification remains weak, descriptive evidence and conditional theory can be reported separately.
The present monograph does not estimate policy reaction functions or causal treatment effects. It uses published abstracts for selected original studies and substantive official HTML sources for policy and statistical facts. Replication would require the underlying data, code, methodological appendices, and a preregistered specification strategy. These limits constrain the strength of its empirical conclusions and are recorded in the evidence ledger.
Revenue competition, fiscal accounting, and policy evaluation
For a government contemplating a rate reduction or an investment incentive, the first measurable fiscal consequence is the change in tax receipts relative to a reference policy. The government's actual net budget effect also includes administrative expenditures, additional services consumed by the project, infrastructure construction and maintenance, other taxes arising from new activity, and effects on borrowing. The time profile can be critical because initial expenditures and later receipts occur in different periods.
Imagine a project incentive that costs 10 units annually in forgone tax over three years and generates additional local tax receipts of 6 units in each of those years. Assume that associated local labor or supplier income produces a further 2 units of public receipts annually and administrative cost is 1. The annual net fiscal cost is 10 − 6 − 2 + 1 = 3. With a 5% discount rate and year-end payments, the present value of this cost is approximately 8.17: 3/1.05 + 3/1.05² + 3/1.05³. The arithmetic includes only assumed fiscal flows, not household or firm welfare. Its purpose is to show why a projected investment headline answers neither the government budget question nor the comprehensive welfare question.
If the same project would have located in the jurisdiction without the incentive, the apparent new tax receipts may simply be receipts that existed in the counterfactual. The true incremental receipts would then be lower, making the preference more costly. Conversely, if its presence creates durable additional activity after the incentive expires, a three-year comparison omits later gains. An honest evaluation identifies both the time horizon and behavioral counterfactual.
Tax expenditures present another measurement question. A benchmark tax system is needed to define revenue foregone, and that benchmark can vary across governments. A mechanical difference from the benchmark assumes an unchanged base; an economic revenue estimate includes the behavioral changes attributable to reform. Direct expenditure and tax relief can also differ in timing, risk allocation, and budget visibility. Comparable evaluation should examine their economic function and specific legal mechanics.
An alternative objective of maximizing gross corporate receipts is not equivalent to maximizing social welfare. A government can collect more by sustaining an economically distortionary policy; it can also sustain a lower tax burden to attract an activity that creates public benefits beyond immediate receipts. The fiscal test is an essential input to welfare analysis, rather than a complete substitution for it.
Public investment evaluation provides the appropriate conceptual language: specify the objective, compare realistic alternatives including an unchanged-policy scenario, quantify material fiscal and resource consequences, identify risks, and assess distribution. A decision about international competition rests on the same discipline, augmented by the anticipated responses of other countries and internationally mobile firms.
Cooperation, harmonization, and competitive discipline
Tax cooperation includes measures that make national systems more compatible, reduce cross-border abuse, share relevant information, coordinate administrative processes, and constrain selected policy choices. Harmonization can extend to comparable definitions of income, deductions, minimum burdens, or allocation mechanisms. The two terms describe overlapping activities rather than a single predetermined degree of centralization.
The International Monetary Fund's 2019 corporate taxation policy paper emphasizes sustaining and advancing international cooperation in response to problems in the global corporate tax system. The normative question concerns which forms of cooperation address identified fiscal externalities at acceptable economic and administrative cost. Coordination is not necessarily a single uniform tax system. It can target a particular type of tax preference while preserving substantial national variation elsewhere.
The efficiency case for a coordinated rule depends on the externality it internalizes. In the simplified revenue example, reciprocal constraints on base-attracting rate cuts could allow the participating governments to preserve revenue without changing global output. Competition over real activity entails additional considerations. A coordinated rate might prevent some projects from receiving subsidies that would have offset genuine market failures, or it might reduce grants that redistribute activity among countries with little net productive gain. Uniformity can therefore create benefits and costs on different margins.
A critique of coordination deserves serious attention. When governments are inefficient, politically captured, or spend revenues on policies of low public value, tax competition can impose fiscal discipline. Mobility can create pressure to provide services more efficiently, simplify administration, and maintain a predictable policy environment. National experimentation can reveal which tax instruments achieve better economic outcomes. The strength of these arguments depends on the actual domestic political and administrative institutions. It should not be assumed that all revenue is wasted or that every external constraint is beneficial.
Other economic benefits of coordination arise from the reduction of uncertainty and duplicated compliance. A firm that must interpret inconsistent rules across multiple countries incurs costs even when its activity is fully compliant. Compatible definitions and effective dispute resolution can reduce these costs. At the same time, international agreement may generate a new layer of reporting and administration. Assessing institutional design requires comparing the costs before and after coordination rather than treating harmonization as intrinsically simpler.
A cooperative arrangement must also remain feasible for its participants. Governments can disagree over allocation of the tax base, the scope of permitted preferences, and who collects the additional tax. If a state's expected benefits from participation are low, implementation may be politically fragile. Transitional support, clear definitions, and credible mechanisms of mutual assistance can matter alongside the headline rule. Coordination shapes the strategic environment; it rarely eliminates all policy discretion.
The global minimum tax and competitive substitution
The OECD (2021) GloBE Model Rules describe a coordinated approach to minimum taxation of large multinational groups. Their economic importance lies in conditioning the value of a low local effective tax burden on the treatment of the same income elsewhere. The rules use a jurisdiction-specific calculation. In the OECD's minimum tax overview, the minimum rate is 15%; an additional tax can apply to the relevant excess profit in a jurisdiction where the computed rate falls below that level, after specified adjustments and a substance-based income exclusion.
The collection architecture includes a qualified domestic minimum top-up tax (QDMTT), an income inclusion rule (IIR), and a backstop undertaxed profits rule (UTPR). In simplified terms, the low-tax jurisdiction can have priority in collecting the additional tax through a qualified domestic rule; the other provisions can assign collection rights elsewhere according to the agreed ordering and applicable law. The common approach is implemented through national law. The existence of model rules does not mean that every country has enacted identical provisions or that every firm is in scope.
A simplified numerical example isolates the fiscal incentive. Assume a covered group has GloBE income of 100, covered taxes of 5, no adjustments, and a substance-based exclusion of 20. The jurisdictional computed rate is 5%. Applying the 10 percentage point top-up rate to excess profit of 80 gives an illustrative additional liability of 8, before any other applicable adjustments or rules. Combined covered tax plus illustrative top-up is 13. This is not a universal 15% tax on total income; the example demonstrates why the excess-profit base and effective-rate denominator need to be preserved. Actual GloBE calculations are more complex.
If a country otherwise imposes very little tax and another authority can collect an additional amount, the private advantage of the first country's low burden can be smaller than under a system without the coordinated rule. The domestic fiscal authority may prefer to collect a qualified top-up itself rather than leave the same fiscal resource abroad. Economic responses depend on whether a group is in scope, the relevant jurisdictional tax computation, expected rule application, available safe harbors, and the costs of changing business structure.
Cooperation does not remove the incentives to attract productive activity. Governments can still compete through public administration, infrastructure, labor supply, grants, and legally permitted tax incentives. The design of incentives becomes a question of whether the tax advantage survives the global calculation and whether the alternative instrument delivers genuine local value. The OECD's January 2026 Side-by-Side package overview identifies safe harbors that include treatment of some substance-based incentives and qualifying minimum-tax systems. Consequently, the evolving institutional treatment matters for comparisons. This dossier does not provide a current country-by-country safe-harbor or eligibility determination.
Contreras, Hugger, and Zawisza (2026), in their early study of MNE responses, use group-level financial and ownership data and compare firms around the EUR 750 million revenue threshold with a difference-in-differences strategy. Their design illustrates how early evidence can be assembled around a policy eligibility boundary. The present manuscript does not report their numerical estimates because the tables and robustness appendices were not independently inspected. The distinction between first-year responses and durable long-run adjustment remains central.
The author's interpretation is that a coordinated minimum alters the marginal fiscal advantages of certain tax instruments and can change the composition of jurisdictional competition. It does not mathematically guarantee the disappearance of fiscal rivalry. A higher tax burden on previously low-taxed excess profits can coexist with intensified efforts to attract activities using expenditures or more targeted treatments. Analysis should measure the resulting mix of tax and non-tax instruments and their budgetary consequences, rather than judge the full reform solely from the number of jurisdictions changing general rates.
Research implications for the global offshore system
International tax competition connects the institutional geography of the offshore system to the fiscal objectives of states. The central economic problem is the coexistence of several locations: where production occurs, where investors reside, where assets are legally held, where profits are recorded, and where tax is collected. A study of the offshore system that collapses those locations into one country label obscures the mechanisms of competition.
Some offshore services facilitate investment, risk allocation, joint ownership, financing, and access to specialized professional institutions. The observable presence of cross-border structures therefore has multiple interpretations. The value of a jurisdiction's fiscal strategy can depend on legitimate domestic value added, fees, and tax receipts as well as the benefits and costs created abroad. Evidence about low reported tax or high financial positions alone cannot distinguish these functions.
For comparative research, the most informative unit may be a jurisdiction-by-activity observation rather than a jurisdiction-wide binary classification. Corporate financial intermediaries and operating establishments can face different tax instruments. Industry characteristics influence the mobility of capital and profit. The government policies relevant to a bank, an intangible-intensive company, and a physically anchored extractive operation may have different consequences even under the same national law. Disaggregated evidence is required to identify the margin.
The author's synthesis yields a practical interpretation of statistical anomalies. A large change in corporate income without a corresponding change in employment or assets is a hypothesis-generating observation; it motivates investigation of accounting attribution and real business changes. A new factory with publicly funded supporting infrastructure generates another pattern that calls for net-welfare evaluation. Both are relevant to tax competition. They demand different evidence and distinct counterfactuals.
Future research should compare the first-round effects of rate and incentive changes with subsequent changes in policy instruments. Persistent competition after a period of stable headline rates could be evaluated with data on tax-base provisions, refundable credits, grants, administrative certainty, public spending, and productive outcomes. Public expenditure needs to be incorporated into the research design as a complementary and potentially substitutable policy instrument. Such investigation would test a plausible mechanism rather than claim that competitive substitution occurs everywhere.
The most durable contribution of international public finance to offshore economics is methodological: fiscal attractiveness is an equilibrium outcome involving domestic policy, foreign responses, firm choices, and the distribution of tax collection. Scientific analysis should preserve that interaction at every level of aggregation. It becomes possible to connect quantitative measures of international activity with evidence about the institutional arrangements that generate them.
Conclusion
International tax competition arises from the mutual dependence of national fiscal choices when economic activity and taxable income can respond to international differences in tax treatment. Its consequences follow from the particular mobility of real investment, reported profit, and the incomes of other taxpayers. The relevant tax burden is therefore specific to the decision margin under examination. A general statutory rate, a standardized investment tax wedge, a company's cash tax rate, and an internationally coordinated jurisdictional minimum-tax rate convey different information.
The established literature identifies reasons decentralized policies can diverge from coordinated outcomes. A government can benefit from the movement of taxable activity while imposing a revenue cost on another jurisdiction; the benefits and costs differ with national scale, existing tax bases, and the location of economic rents. The resulting equilibrium may be individually rational for participating governments and still impose collective costs. The existence of an externality provides a basis for evaluating cooperation, subject to domestic political institutions, economic efficiency, distribution, implementation costs, and the actual use of public revenue.
Empirical research must distinguish strategic policy causation from coincident policy movements and financial measures from productive investment. Descriptive tax-rate statistics establish important historical patterns, including the stabilization of broad averages after their long decline. They do not establish that tax competition has ended. The available literature and institutional evidence direct attention to the full portfolio of tax-base rules and incentives, real activities, financial attribution, and the allocation of public receipts.
The welfare of residents is a wider object than the attractiveness of a country to a firm. Evaluation must include the additional economic activity caused by tax policy, public costs, market adjustments, and incidence. The effects of competition for highly mobile accounting income differ from those of competition for a long-lived productive establishment. The fiscal position of a specialized small jurisdiction can be affected differently from that of a large source economy. That asymmetry is fundamental to the feasibility and distributional design of international cooperation.
The global minimum-tax architecture changes selected incentives by linking local tax treatment to coordinated collection mechanisms. Its economic implications depend on coverage, implementation, the treatment of particular incentive instruments, and subsequent government and taxpayer behavior. It is an evolving institutional reform whose realized long-run effects require direct empirical investigation. The general economic problem of strategic competition survives changes in its instruments.
The monograph's original contribution is the integration of these strands into a disciplined comparative analysis of policy choice, measurement, fiscal externalities, incidence, and instrument substitution in the global offshore system. It establishes interpretive distinctions and derives transparent conditional examples from familiar economic reasoning. It makes no claim to a new named theory, original causal estimate, or verified discovery beyond the evidence reviewed. A fuller empirical assessment requires replication-ready data, access to complete scholarly texts and methods, and jurisdiction-specific institutional verification.
References
Beer, S., de Mooij, R. A., & Liu, L. (2018). International corporate tax avoidance: A review of the channels, magnitudes, and blind spots. IMF Working Paper No. 18/168. International Monetary Fund, Washington, DC. DOI: 10.5089/9781484363997.001.
Contreras, L., Hugger, F., & Zawisza, T. (2026). MNE responses to the global minimum tax. OECD Taxation Working Papers No. 77. OECD Publishing, Paris. DOI: 10.1787/0bc9aac9-en.
Crivelli, E., de Mooij, R. A., & Keen, M. (2015). Base erosion, profit shifting and developing countries. IMF Working Paper No. 15/118. International Monetary Fund, Washington, DC. DOI: 10.5089/9781513563831.001.
Damgaard, J., Elkjaer, T., & Johannesen, N. (2019). What is real and what is not in the global FDI network?. IMF Working Paper No. 19/274. International Monetary Fund, Washington, DC.
Devereux, M. P., & Griffith, R. (2003). Evaluating tax policy for location decisions. International Tax and Public Finance, 10(2), 107–126. DOI: 10.1023/A1023364421914.
Devereux, M. P., Lockwood, B., & Redoano, M. (2008). Do countries compete over corporate tax rates?. Journal of Public Economics, 92(5–6), 1210–1235. DOI: 10.1016/j.jpubeco.2007.09.005.
Fuest, C., Peichl, A., & Siegloch, S. (2018). Do higher corporate taxes reduce wages? Micro evidence from Germany. American Economic Review, 108(2), 393–418. DOI: 10.1257/aer.20130570.
Hines, J. R., Jr., & Rice, E. M. (1994). Fiscal paradise: Foreign tax havens and American business. Quarterly Journal of Economics, 109(1), 149–182.
International Monetary Fund. (2019). Corporate taxation in the global economy. IMF Policy Paper No. 2019/007. Washington, DC. DOI: 10.5089/9781498302197.007.
Kanbur, R., & Keen, M. (1993). Jeux sans frontières: Tax competition and tax coordination when countries differ in size. American Economic Review, 83(4), 877–892.
Keen, M., & Konrad, K. A. (2013). The theory of international tax competition and coordination. In Handbook of Public Economics, Vol. 5, pp. 257–328. Elsevier.
Klemm, A., & Van Parys, S. (2009). Empirical evidence on the effects of tax incentives. IMF Working Paper No. 09/136. International Monetary Fund, Washington, DC. DOI: 10.5089/9781451872835.001.
OECD. (1998). Harmful tax competition: An emerging global issue. OECD Publishing, Paris. DOI: 10.1787/9789264162945-en.
OECD. (2021). Tax challenges arising from digitalisation of the economy – Global Anti-Base Erosion Model Rules (Pillar Two): Inclusive Framework on BEPS. OECD/G20 Base Erosion and Profit Shifting Project. OECD Publishing, Paris. DOI: 10.1787/782bac33-en.
OECD. (2026a). Statutory corporate income tax rates. In Corporate Tax Statistics 2026, chapter 2. OECD Publishing, Paris. Report DOI: 10.1787/73af6222-en.
OECD. (2026b). Corporate effective tax rates. In Corporate Tax Statistics 2026, chapter 4. OECD Publishing, Paris. Report DOI: 10.1787/73af6222-en.
OECD. (2026c, January 13). Global minimum tax: Understanding the Side-by-Side package. Technical webinar and event overview. OECD.
OECD. (n.d.). Global minimum tax. OECD topic overview. Accessed November 18, 2026.
Tørsløv, T. R., Wier, L. S., & Zucman, G. (2023). The missing profits of nations. Review of Economic Studies, 90(3), 1499–1534. DOI: 10.1093/restud/rdac049. Author-hosted research page.
Zodrow, G. R., & Mieszkowski, P. (1986). Pigou, Tiebout, property taxation, and the underprovision of local public goods. Journal of Urban Economics, 19(3), 356–370.