RESEARCH DOSSIER

Offshore Finance and Tax Incidence

Abstract. This Research Dossier examines the distribution of tax burdens and benefits generated by offshore financial arrangements in an economically interconnected world. It distinguishes legal liability, tax remittance, accounting location and economic incidence, and relates the mobility of reported income to the mobility of productive factors and ultimate ownership. The analysis integrates foundational public-finance theory with evidence from corporate taxation, international profit shifting, labor markets, consumer price adjustment, offshore household wealth, enforcement and global minimum taxation. It explains the distinct implications for shareholders, workers, consumers, landowners, financial intermediaries and public budgets, with attention to source, residence and host jurisdictions. Conditional numerical illustrations demonstrate how equivalent changes in corporate tax payments can produce different distributions according to market adjustment and fiscal closure. The study finds that changes in reported profit and tax revenue do not identify ultimate household incidence without complementary information on effective exposure, behavioral responses, ownership and government policy. It offers a comparative scholarly synthesis rather than a new theoretical system or a new global econometric estimate.

Keywords: offshore finance; tax incidence; corporate income taxation; profit shifting; capital mobility; economic rents; fiscal closure; labor markets; consumer prices; offshore wealth; beneficial ownership; international taxation

Introduction

Offshore finance changes the distribution of taxation by changing where financial claims, taxable income, corporate activity and ownership rights are located, and by changing the choices available to the people and organizations that own them. A corporate tax is remitted by a legal entity, but the ultimate economic consequences may reach employees, shareholders, consumers, creditors, landowners, intermediaries and beneficiaries of public services. The offshore dimension connects these consequences across jurisdictions. A lower effective liability for a multinational can increase the value of its equity in one country while altering wages and tax-financed services in another. An offshore arrangement can also facilitate genuine cross-border investment and reduce financing frictions without generating any relevant tax advantage.

The scientific object of this dossier is the distribution of the economic burden and benefit of changes in taxation that are induced by cross-border financial arrangements. Its question is distinct from the measurement of government revenue losses, the legal determination of residence or source, the mechanics of profit shifting, the concentration of wealth and the institutional design of international tax competition. Each of these subjects is necessary to incidence analysis, but each has a different dependent variable. The significance of offshore finance for economic distribution emerges when the observed or hypothesized change in liability is connected to changes in real income, asset valuations, employment, consumption opportunities and public provision.

Two comparisons are essential. The first concerns an arrangement that is available to one economic actor but unavailable to another: the tax advantage of access. The second concerns a reform that changes the tax treatment of that arrangement for participants and nonparticipants: the incidence of policy. In the first comparison, access may be correlated with wealth, size or multinational status. In the second, reforms can change investment and contracting decisions in ways that extend beyond initial participants. Answers depend on the relevant base, tax rule, time horizon and fiscal counterfactual.

The argument developed here is that revenue attribution and incidence require different evidence. Administrative, financial and corporate records establish changes in tax bases or liabilities. Economic models and linked outcome data establish who gains or loses through behavioral responses and market adjustment. Fiscal closure then determines what happens to the public side of the distribution. The analysis connects these bodies of evidence without converting a measured volume of shifted profits into a presumptive wage, price or equity effect. The method is a critical narrative synthesis with conditional, explicitly hypothetical numerical examples; it does not estimate a new global incidence coefficient.

Concepts and measurement boundaries

Tax incidence is the distribution across economic agents of changes in welfare or resources caused by a tax or tax change, relative to an explicit counterfactual. Statutory incidence identifies the person or entity on which the legal obligation rests; remittance identifies the actor transferring funds to the government. Economic incidence concerns real purchasing power, economic rents, net asset returns and publicly financed benefits. A source-country affiliate that pays tax from its bank account is a remitter; the persons affected through distributions, wages, prices and transfers are the bearers. Fullerton and Metcalf's Tax Incidence (2002) organizes the analytical distinction between remittance and the distribution of burden through economic adjustment.

Offshore finance in this study means financial arrangements that separate some relevant jurisdiction of intermediation, holding, contracting or booking from the residence or principal activity of the associated economic actor. Offshore financial services include custody, pooled investment, funding, insurance, risk management and corporate administration. Some arrangements are entirely transparent and fully taxed; different arrangements may reduce legally due taxes or facilitate illegal concealment. Evasion requires a breach of applicable obligations; neither the existence of a foreign account nor the use of an offshore company establishes such a breach.

For causal analysis, an observation must specify at least the following characteristics: the ultimate owner or beneficiary, the tax residence of each relevant party, the jurisdiction of booking, the jurisdictions of real activity and of the final customer, the taxable event and the applicable counterfactual. An intercompany payment can be recorded in several sets of accounts. It may be a deduction for one affiliate and income for another, or eliminated in consolidation. Neither the gross transaction nor the total assets of an offshore entity should be counted as a newly created tax base.

Stocks and flows have different units. Offshore assets at the close of a year are stocks; dividends, interest, taxable capital gains, profit and taxes during that year are flows. A ratio of financial wealth to annual gross domestic product is a scale comparison, not a tax-to-income rate. An effective tax rate requires a described numerator, denominator and timing rule; a cash tax rate can differ from an accrual-based rate through payment schedules, loss relief and deferred taxes. The average tax rate on existing earnings is also distinct from the marginal tax wedge that governs a prospective investment. An analyst who merges these quantities cannot infer the distribution of the economic consequences.

The relevant incidence object may be a small marginal reform, a large structural regime change, a prohibition on a specific arrangement or the availability of a previously inaccessible intermediary. A partial equilibrium price response and a general equilibrium welfare response need separate descriptions. In the former, other markets are held fixed. In the latter, wages, investment locations, returns, prices and government budgets adjust together. Both are useful if their scope is explicit.

The foundations of incidence analysis

A transparent starting point is the competitive market for a taxable transaction. Let the absolute elasticity of demand be d and the elasticity of supply be s. With a small per-unit tax, a locally linear competitive approximation yields a fraction s/(s+d) borne by buyers through their gross price and d/(s+d) borne by sellers through their net price. This expression follows by differentiating the market-clearing condition and the tax wedge between the two prices. It holds at the chosen equilibrium under the stated assumptions; it is not a universal formula for corporate income taxation.

For example, if supply has elasticity 3 and demand has absolute elasticity 1, buyers bear three quarters of a small unit tax and sellers one quarter. A tax collected from the seller and a tax collected from the buyer create the same wedge in a frictionless competitive market under equivalent enforcement. Offshore arrangements change this elementary result when they alter the ability to avoid the wedge, the available supply of intermediation, the market power of the parties or the cost of geographic substitution. They may also change the population on which the tax is effectively enforced.

Corporate income taxation requires attention to the return on investment, economic profit and the mobility of factors. Harberger's corporation income tax study (1962) established a foundational general equilibrium approach in which tax-induced reallocation between sectors affects capital returns and factor prices. A general equilibrium result follows from assumptions about production, substitutability, mobility, ownership and demand; a tax paid by a corporation is not distributed solely by inspecting the corporation's accounting entries.

Consider a firm with a marginal project producing a normal return and a separate stream of location-specific rent. Taxing the normal return can change the set of projects that satisfy investors' required returns. Taxing the rent from an established unique resource can reduce equity value while leaving activity nearly unchanged. If the rent can itself be shifted for accounting purposes, enforcement changes the net share retained by owners and the costs of the arrangement even when real production is unaffected. If a tax base includes both types of return, the response must be decomposed.

The burden of existing capital assets is especially sensitive to timing. A permanent unexpected increase in tax on a fixed stream of future economic rent reduces the present value of that rent at announcement. The owner at the moment of the announcement can bear the valuation loss even if subsequent purchasers remit future taxes. Buying the asset after its price adjusts does not reproduce that original loss. Continuing rents, changes in risk and expected growth must be distinguished from one-time capital losses. The same reasoning applies to values of commercial land in business districts that depend on expected demand from financial institutions.

Open economies and the mobility of tax bases

International capital mobility extends incidence beyond domestic industries. In a benchmark small open economy with an internationally given required net return, perfectly mobile capital and immobile labor, a local tax that raises the cost of marginal capital can reduce the domestic capital stock. If workers' marginal product depends positively on capital, labor productivity and wages can decline. This is a conditional comparative static. It assumes a specified tax on the marginal return, an outside opportunity available without relevant friction, and no offset through other policy, market power or immobile rents.

The mechanism is different when multinational profits can be recorded abroad without moving productive assets. If an enterprise can lawfully reduce effective tax on the existing normal return or rent while keeping the location of employment fixed, the sensitivity of real activity to the statutory rate may fall. When a reform closes such a margin, it can restore taxation with little immediate capital relocation, or it can expose marginal projects to a larger effective wedge and alter activity. An observed change in reported profit alone cannot identify which alternative obtains.

Gravelle and Smetters (2001), in their open-economy analysis, show why the limiting assumptions matter. Imperfect substitutability of domestic and foreign products in their two-economy calibration substantially constrains the domestic wage channel; some parameter configurations instead transfer part of the burden abroad. The relevant question is therefore how mobile each productive factor and each accounting base is, conditional on the size and structure of the market.

There are at least four distinct margins of mobility. A corporate entity or financial claim can change its recorded jurisdiction; a financially integrated group can alter contractual relations among legally separate units; physical capital can be installed in another location; and workers or consumers can themselves relocate or substitute across employers and products. The first two margins may respond quickly because they require organizational or financial adjustment, while productive facilities and skilled labor relationships often embody sunk costs. Time horizons and institutional restrictions determine which are active.

Open-economy tax incidence also has an ownership dimension. When a domestic enterprise is substantially foreign-owned, a reduction in its rental income can fall on nonresident households. Conversely, resident savers may be exposed to foreign corporate taxation through diversified investment funds and pension claims. National income accounting and domestic welfare analysis therefore require ultimate ownership, not simply the jurisdiction of the business. A source country may tax foreign owners while using the proceeds for domestic services; a residence country may grant credits or exemptions that alter its own revenue and its residents' burdens. These effects are distributions across distinct populations, not components of a single national wage coefficient.

Tax competition can modify these results when governments respond to each other. A unilateral increase in one country's effective rate changes location incentives relative to alternatives; a coordinated increase can reduce those outside options. This changes the relevant elasticity and hence the expected incidence. The strategic response of governments is conceptually distinct from how households and firms share the burden of a tax already enacted. The analysis in this dossier will hold policy in other jurisdictions fixed where using a unilateral example and will disclose the alternative explicitly when analyzing coordination.

Corporate offshore arrangements and financial frictions

The tax consequences of offshore corporate finance arise through the interaction between real economic activities, financing and tax rules. A group may establish foreign subsidiaries to serve local customers, pool investors, isolate commercial risks or obtain access to funds. It may also face incentives to recognize taxable income in different jurisdictions or to alter the allocation of debt and returns. A difference in statutory rates is insufficient to establish the amount of tax saving: deductions, withholding, credits, exemptions, loss positions, transfer pricing, substance requirements and the residence tax system all matter.

Desai, Foley, and Hines's Demand for Tax Haven Operations (2006) documents the heterogeneity of US multinational haven use under the historical tax regime they study. Their analysis cautions against treating foreign activity and haven affiliates as independently assigned. Scale, organization and location choices can jointly determine both use of offshore facilities and tax outcomes. The results concern that historical regime; a contemporary analysis must allow for subsequently changed rules.

A simple allocation example isolates the direct effect. Suppose a multinational initially earns 100 units of taxable profit in jurisdiction A at a 25% tax rate. Under a hypothetical compliant change in allocation, 40 units are instead taxed in jurisdiction B at 5%, with no residence-country top-up, no withholding and no real production change. A's liability becomes 15, B's becomes 2, and the group's total tax falls from 25 to 17. The source-country revenue difference is 10, worldwide net revenue difference 8, and the direct private tax saving 8. If the arrangement requires 2 units of real additional resource cost, the group's private net saving is 6. If that payment is a fee to an adviser, the fee is the adviser's gross receipt, not a distinct 2 units of global welfare loss; only the real opportunity cost of the services and any associated distortions can be counted as social costs. Each amount has a distinct economic recipient or use.

Suppose the group passes 1 unit of its private saving into higher employee compensation and 1 into lower prices on existing output. Holding quantities and all other effects fixed, the residual owner's gain is 4. The resulting direct allocations are hypothetical; the initial accounting cannot establish the two pass-through amounts. In a real study, separate data on compensation, prices and owners would be necessary. If A replaces its lost revenue with a 10-unit levy on domestic households, or reduces services by the same budgetary amount, those effects belong in the fiscal counterfactual. B gains 2 of revenue and may use it differently from A. The worldwide loss of 8 cannot be equated with the local loss of 10 or the owner's gain of 4.

This illustrative case also distinguishes tax-induced financial engineering from genuine reduction in financing cost. A universally accessible intermediary that reduces the cost of pooling international investors can improve project funding even with identical effective taxation. A preferential arrangement that changes relative tax liability can alter competitive advantage even if its intermediation technology is identical to that available domestically. Real efficiency and tax-induced redistribution can coexist. To evaluate both, estimate the funding or risk-sharing benefit separately from the change in tax payments, and include the resources consumed in obtaining each benefit.

A firm with accumulated losses may have little current incentive to shift a deduction into the loss jurisdiction, but the present value of future relief can still be material. Likewise, deferral produces a time-value advantage even when nominal lifetime liability is unchanged. In a stylized example, delaying payment of 10 for five years at an annual discount rate of 5% reduces its present value from 10 to approximately 7.84. The present-value benefit is about 2.16; the nominal liability is still 10. A comparison that reports only cash taxes in the first year will overstate the change in lifetime liability if the later payment is certain. Conversely, if subsequent collection is uncertain, that probability belongs in the valuation.

Shareholders, economic rents and financial intermediaries

Shareholders initially receive the residual of corporate revenues after payments to employees, suppliers, creditors and government. When a tax reduction changes this residual without changing prices, investment or contracts, shareholders benefit. The location of the booking entity does not determine the location of the people who benefit. Ownership may be held through an investment fund, trusts, retirement accounts, insurers or other institutions. The appropriate unit is the beneficial economic claim subject to its actual sharing rules.

Economic rents arise from market power, valuable intellectual assets, scarce licenses, limited land, location-specific productivity or other advantages. A levy on a fixed rent can reduce the owner's return without eliminating the activity that generates it. Offshore access may reduce the effective tax paid on these rents and increase their capitalized value. Where the market is competitive, the first beneficiary of a persistent tax privilege may be the person selling the right to a scarce opportunity, rather than the future purchaser who pays the capitalized price. Tax benefits associated with a long lease or scarce regulated license may thus affect sellers, buyers and workers at different moments.

Consider a rent stream of 10 units per year that is expected to persist without growth or risk change. At a discount rate of 5%, its pre-tax present value is 200. If a permanent tax changes the annual after-tax rent from 10 to 8, the value becomes 160. The existing owner loses 40 in capital value, equivalent to the discounted difference in future after-tax rents. Subsequent investors who buy for 160 receive the new fair return. Adding the 40 valuation effect to the same discounted 40 future tax-induced reduction would double count the owner's loss. This example assumes a constant perpetuity, no behavioral adjustment and an unchanged discount rate.

Financial service providers can share the benefit of offshore arrangements through fees and markups. Where financial expertise is scarce or legal access is differentiated, advisors may charge more than marginal resource cost and capture part of the surplus. Competitive entry can erode those rents and reduce fees. Investors who can obtain a comparable service at home may be sensitive to fee differentials; investors who require specialized services or confidentiality may be less price sensitive. Enforcement or disclosure reforms can change the cost structure of providers, the number of firms serving a market and the allocation of fees between investors and intermediaries.

Debt holders can be affected indirectly through changes in solvency, risk or refinancing costs. A higher tax on a leveraged enterprise with a binding financing constraint may increase default risk and reduce bond values. Contractual covenants, seniority and renegotiation determine the consequences for lenders. A tax saving that strengthens a balance sheet can benefit existing creditors if default probabilities fall, while new creditors price the improved position when extending funds. Counting the entire equity-value response and the entire debt-value response requires a consolidated firm-value accounting rather than treating both securities as separate underlying economic assets.

These channels are particularly relevant in offshore finance because the value of a business can depend on claims on several jurisdictions. Ownership data should look through chains of holding companies wherever permitted, distinguish claims that duplicate the same underlying asset, and avoid identifying legal title alone with ultimate economic exposure. Measurement may remain incomplete where beneficiaries are unobserved. In such cases the analysis should show an ownership allocation range or explicitly leave part of the burden unattributed.

Labor markets and wage incidence

Workers can bear corporate taxation through lower wages, lower hours, employment changes, slower investment and reduced labor productivity. They can also gain from tax-financed public services. Gross wage changes, total labor compensation changes and the welfare of workers are distinct. A worker who remains employed at an unchanged nominal wage may still face a reduction in real purchasing power; a worker who changes employer may have a different wage and relocation cost. A firm's wage bill combines employment, hours and compensation per unit of labor.

Fuest, Peichl, and Siegloch (2018) estimate wage effects from roughly 6,800 tax changes in a panel of German municipalities spanning twenty years. Their German municipal tax study uses linked employer-employee records and event-study and difference-in-differences approaches. The reported burden on workers is approximately one half in their setting, with differences across worker groups. The authors emphasize labor institutions and profit-shifting opportunities. These estimates concern their observed changes, jurisdictional system and worker sample; the paper's available abstract does not establish a universal share for offshore multinationals.

Two channels help interpret this evidence. In a bargaining channel, taxation can change the pool of economic rents divided between a company and its employees. If workers receive a negotiated portion of the net surplus, changes in tax payments can alter compensation even when installed capital remains fixed. In a productivity channel, investment changes alter capital available per worker and hence marginal product. The two channels need not respond at the same speed. Contract renegotiation can shift the division of rents, whereas investment changes may accumulate over years.

The offshore margin complicates both channels. A tax reduction on book profits might improve profitability available for bargaining without changing physical capital. An additional tax on an arrangement may cause a multinational to reduce discretionary payroll or relocate selected functions. A firm already able to offset an increase through lawful cross-border planning may exhibit a smaller immediate wage response than a domestic firm facing the same increase in the statutory rate. Yet a reform aimed directly at that planning opportunity may change the multinational's effective rate more than that of the domestic firm. An empirical analysis therefore needs the actual exposure to the reform, rather than using multinational status alone as proof of a common treatment.

Labor mobility is uneven. High-skill financial employees in an internationally connected occupation may be able to change jurisdictions or employers more readily than workers providing local services. That difference can change bargaining and labor supply elasticities. A financial center losing a specialized business function can experience a combination of direct employment reduction and indirect changes in demand for housing, transport and services. A source country gaining activities can experience the opposite effects. Geographic movement changes who appears in the local wage data; a study of incumbent workers may therefore answer a different question from a study of the average wages of people currently resident in the location.

A useful empirical decomposition tracks wage effects per continuing worker, entry and exit of workers, hours, employment and employer composition. It should separately identify changes in compensation from changes in capital income of employees who own stock or participate in pension plans. Combining these effects at the household level requires a consistent population and a common deflator.

Spatial equilibrium, property and location-specific burdens

Business tax changes can alter local labor demand, land values, rents, the number and type of establishments and commuting patterns. Some gains that appear as higher wages are offset by housing cost increases. Conversely, a decrease in land prices may harm existing homeowners or landlords while making entry cheaper for future households. The distinction between incumbent and entrant is necessary when the geographical distribution of finance changes.

Suárez Serrato and Zidar (2016) develop a spatial approach in their study of state tax cuts, using US state changes in tax rates and apportionment with imperfect mobility of firms and workers. Their analysis makes inframarginal firms and location-specific profitability central to how local policy affects owners and employees. It also connects business activity to property values. The approach is useful for examining financial centers, because agglomeration, professional networks and limited local property supply constrain adjustment.

The original estimates require a correction. Malgouyres, Mayer, and Mazet-Sonilhac's published comment (2023) identifies problems subsequently addressed in the authors' response. In their corrected incidence estimates, Suárez Serrato and Zidar (2023) report 38.1% for firm owners, 35.0% for workers and 26.9% for landowners; some uncertainty intervals widen after correction. These point estimates sum to 100. They depend on the model and the US state setting and are suitable as evidence that all three channels can matter, not as calibrated parameters for international financial centers.

The geographical concentration of offshore finance can turn local property into an important recipient of tax-induced rent. If new firms enter a center with a constrained commercial-property supply, demand may capitalize gains into office rents. If residential construction is constrained, higher financial-sector wages and migration may increase housing rents; workers may retain only part of their nominal gain. Property owners benefit from appreciation if their ownership predates the transition. If financial-sector activity falls, the reversal can hurt investors who bought based on the expectation of continuation.

Housing costs, land values and occupancy are different variables. A rise in office rents can be associated with a rise in the capital value of commercial property, but summing the two as independent wealth gains is incorrect unless the flow and asset accounting are carefully separated. Likewise, a rise in the number of employees in a center measures economic activity, not directly the wellbeing of original residents. New residents may receive substantial gains while previous residents experience a larger housing burden.

A study of incidence in a financial center should examine local employment and wage distributions, housing rents and prices, business formation and exits, land ownership, and municipal revenue. Demand from unrelated domestic sectors, monetary policy, zoning and infrastructure can influence the same outcomes. A credible comparison requires evidence on these confounders. Without it, the observed development of a center can establish a descriptive association but cannot identify a tax policy's independent effect.

Consumers, market power and price pass-through

Consumers can bear corporate tax changes when firms alter product prices, service fees, quality, availability or contract terms. For a tax proportional to pure economic profit in an idealized static model with unchanged deductible costs, a single firm maximizing after-tax profit chooses the same output as before: multiplication of the objective by a positive constant preserves its maximizer. Actual corporate taxes can affect marginal costs, entry, organizational form, financing, investment and competitive structure. The idealized neutrality condition therefore requires careful examination.

Baker, Sun, and Yannelis's Corporate Taxes and Retail Prices (2020, revised 2023) studies US product prices using variation between production and sales locations. Their working paper reports a net-of-tax price elasticity of 0.24 and an approximate consumer burden of one half in the study's setting. This is working-paper evidence for a particular identified reform margin. The authors' use of the geographic separation between production and consumption emphasizes the importance of estimating price changes without conflating them with local demand shocks.

Offshore finance may change the price channel by altering the effective cost faced by a multinational enterprise, especially when competing firms do not have comparable opportunities. A reduction in effective corporate tax can lower the cost of entering new markets or allow a company to sustain lower prices while expanding output. It can also increase after-tax profits at unchanged prices if the market structure allows the firm to retain the advantage. The same difference can influence competitive entry: if smaller domestic rivals face different effective tax burdens, a tax-induced cost differential may affect concentration and long-run consumer outcomes. In a concentrated market, changes in competition can eventually dominate initial price pass-through.

Financial products present an additional layer of contractual complexity. Consider two investment funds with identical underlying assets but different tax and operating structures. Investors care about expected after-tax returns after management fees and expenses. A lower tax burden at the fund or investor level can be retained by investors or absorbed through higher fees. A transparent low-fee cross-border structure can benefit investors by reducing double taxation or other costs. An opaque arrangement may impose extra legal and governance costs. The relevant comparison is across economically equivalent risk-adjusted investments with equivalent investor tax obligations.

Price data must distinguish wholesale from retail prices, quantities, taxes included in listed prices, and changes in product characteristics. A 1% rise in a product price need not correspond to a 1% welfare loss to each buyer, because consumption quantities and alternatives can change. For lifetime financial contracts, an up-front fee and an annual asset-based management fee are not directly comparable without discounting and expected asset growth.

It is tempting to combine separately estimated wage burdens, consumer burdens and asset-owner burdens into a numerical global allocation. This is invalid when studies use different tax shocks, populations, denominators, horizons and counterfactuals. A worker's real wage already incorporates effects of the price index used to deflate it; subtracting the entire measured price effect a second time from the same real income would double count part of the impact. A single welfare calculation should use nominal incomes and a compatible price index or use real income measures without repeating the same price channel.

Profit shifting and the geography of reported income

Profit shifting changes the jurisdictional distribution of taxable profit and may distort the geographic distribution of measured factor income. Tørsløv, Wier, and Zucman's The Missing Profits of Nations (2023) uses foreign-affiliate statistics and discrepancies in multinational versus domestic profitability to estimate that 36% of multinational profits are assigned to tax havens through shifting in their study. The study also considers implications of reallocating shifted profits for national accounts and factor shares. The published percentage refers to its historical data and estimation procedure, not to a current tax incidence fraction.

The method is important because accounts can report extremely high profitability in locations with comparatively little labor or productive capital. Such a pattern is a signal requiring investigation; it is not conclusive evidence for the exact amount of economically unjustified profit in every affiliate. Returns on intangible assets, holding activity, sector composition, genuine business functions and accounting conventions can affect the same ratios. A macro estimate reconstructs the aggregate distribution under stated assumptions and need not adjudicate the legal correctness of individual transactions.

There are two distinct distributional questions. The first is whether offshore allocation changes the true distribution of income by reducing tax or shifting investment and compensation. The second is whether the statistical allocation of an unchanged global profit stream changes the measured distribution of factor shares across countries. If reported profits leave a source jurisdiction while wages stay there, the ratio of employee compensation to measured local value added may rise mechanically. A reallocation exercise can reveal an underlying change in measured capital income without establishing a change in workers' actual compensation. Accurate gross domestic product and national income accounting is necessary for incidence analysis but does not replace it.

The degree of development of source countries also matters. Johannesen, Tørsløv, and Wier (2020), in Are Less Developed Countries More Exposed to Multinational Tax Avoidance?, investigate reporting behavior using international firm data. Their reported regressions distinguish tax incentives for shifting from firm and country characteristics and find stronger reported-profit responsiveness in less developed country groups. The result concerns the sensitivity of reported profits; its interpretation must respect differences in data coverage, unobserved corporate characteristics and the identification strategy. It provides evidence for heterogeneous exposure, not a direct estimate of the burden falling on low-income workers.

An affected tax jurisdiction's policy response determines the consequences of lost revenue. A government with limited access to alternative tax bases or affordable borrowing may adjust spending differently from a government with deep capital markets and diverse revenue instruments. The same change in reported profit can have different welfare effects because the feasible fiscal counterfactuals differ. An international incidence study should therefore evaluate the local marginal benefits of spending and marginal costs of substitute financing, rather than equating revenue lost with a uniform burden on residents.

Household offshore wealth and horizontal equity

Individuals can hold assets across borders for diversification, access to investment markets, currency risk management, migration or business purposes. The economic incidence of taxation depends on whether the resulting income or gains are reported and taxable under applicable rules. A fully declared cross-border portfolio can have the same residence tax burden as an equivalent domestic portfolio, even if income is recorded by foreign intermediaries. It can also benefit from different source-country withholding or relief provisions. Analysis requires the counterfactual applicable to the owner and asset.

Alstadsæter, Johannesen, and Zucman's Who Owns the Wealth in Tax Havens? (2018) constructs historical offshore wealth estimates from macro financial statistics. The study's global magnitude is approximately one tenth of world annual GDP in its reference period, with substantial geographic heterogeneity. It separates assets entrusted to offshore custodians from foreign securities held through domestic custodians. These are estimates of wealth, not directly of unpaid tax; not every offshore asset is concealed.

For household taxation, converting a wealth stock into a liability requires a taxable return, realization rules, applicable rates, exemptions, reporting status and residence. Suppose a person owns 1,000 units of assets yielding 50 units of taxable income over a year. If a 20% income tax applies with no relief, liability on that income is 10 units. Taxing the full wealth stock at 20% would instead be a 200-unit liability under an entirely different base. If only 20 units of the annual income are actually taxable after allowances, the liability is 4, subject to the given assumptions. The existence of the 1,000-unit holding alone does not determine which of these numbers is relevant.

Horizontal equity compares similarly situated taxpayers. A difference in tax burdens among households with the same underlying income can arise through legally differentiated treatment, access to specialized advisers, reporting behavior or enforcement constraints. If the difference is correlated with wealth, distributional consequences can persist even when statutory rates are formally progressive. Asset mobility and access to cross-border financial expertise become relevant to effective progressivity.

Alstadsæter, Johannesen, and Zucman (2019) investigate this relationship using Tax Evasion and Inequality. Their Scandinavian evidence combines leaked offshore records with administrative data and information from random audits. They report especially high evasion relative to taxes owed at the very top of the wealth distribution, around 25% for the wealthiest 0.01% in their study. Ordinary random audits detect substantially less evasion. The evidence illustrates how hidden assets can undermine measurement of both tax compliance and wealth concentration. The estimate is specific to the population, time period and combined data strategy; it does not establish an equivalent rate for wealthy residents of every country or for all offshore holders.

An income distribution study that observes filed returns alone may omit concealed financial income. In this situation, estimated effective tax rates at the top can be biased because taxable income, wealth and actual compliance are measured imperfectly. Incorporating estimated hidden income without estimating associated taxes consistently also yields a misleading ratio. The approach requires integrated treatment of both the income denominator and tax numerator, with intervals for unobserved amounts. Distributional rankings may themselves change when concealed assets are imputed.

Pension wealth and investment funds complicate the inference that all capital-income taxation falls on rich households. Many workers have claims on financial assets through retirement institutions. The economic allocation depends on the distribution of those claims, whether returns are defined by asset performance or contractual pensions, and the time horizon over which participants are affected. Direct offshore secrecy advantages tend to require a different access mechanism from ordinary pooled foreign investment. The incidence of each must be examined with separately identified ownership and reporting data.

Enforcement, information and changes in compliance

Offshore tax enforcement can increase the expected liability of people who had previously failed to report taxable income. It can also impose compliance costs on people already meeting their obligations and change the products offered by foreign financial intermediaries. The distribution of these consequences depends on the enforcement instrument's scope, the probability of detection, the sanctions, the availability of legal remedies and the cost of compliance.

Johannesen, Langetieg, Reck, Risch, and Slemrod (2020), in Taxing Hidden Wealth, study US enforcement efforts initiated in 2008. Their published estimates associate enforcement-driven disclosures with an annual increase of $2–4 billion in reported capital income and an estimated $0.6–1.2 billion in additional annual tax revenue. These quantities are changes in disclosed income and estimated revenue associated with the intervention, not a complete measure of the lifetime welfare losses of account holders or all costs of enforcement. They should also be distinguished from older working-paper estimates, which use different reported numerical ranges.

A direct enforcement effect falls on the person whose unreported liability becomes payable. Its welfare effect can be larger or smaller than the collected amount depending on penalties, the timing of payment, legal costs and previous expectations. If an account holder anticipated enforcement, the market value of relevant claims may already incorporate some expected cost. Changes in expected sanctions can alter behavior even without contemporaneous collection.

An enforcement policy can induce substitution. Investors may move deposits to a different intermediary, convert financial assets to other forms, report previously concealed holdings, change residence lawfully or alter investment choices. Observing a fall in deposits at a center does not distinguish these outcomes. The composition of offshore banking balances, the reporting status of each owner and changes in declared wealth or taxes are necessary to estimate the distributional consequences. Aggregate cross-border deposit data alone cannot separate a repatriation from a redirection of ownership or custody.

There are legitimate privacy and administrative costs associated with financial transparency. A policy that imposes administrative duties on intermediaries can lead to fees, service withdrawal or exclusion of smaller customers when fixed compliance expenses exceed expected revenue. Those effects are economically distinct from the loss of a previously available advantage from evasion. The appropriate evaluation considers the unlawful tax advantage separately, identifies actual resource costs and measures the extent to which compliance burdens are passed through to compliant customers.

Tax enforcement also creates indirect effects through public revenue and norms of compliance. If enforcement increases perceived fairness for taxpayers with similar incomes, acceptance of the tax system may change; establishing a behavioral effect would require independent evidence. For economic incidence, a conservative appraisal can treat this as a potential channel and avoid assigning a quantitative welfare value without an empirical identification strategy. The immediate accounting links remain changes in liabilities, costs, cash receipts and public uses.

Fiscal closure and unequal public consequences

Tax burdens and benefits depend on what happens to government revenue. Fiscal closure specifies the adjustment mechanism by which a government accommodates a change in revenue or expenditure. Standard incidence calculations can hold government spending fixed and vary another tax, or hold other taxes fixed and vary spending, or allow borrowing. Each choice produces a different distribution.

Suppose an offshore corporate arrangement reduces receipts of government A by 100. Under a revenue-neutral scenario, A introduces a substitute tax of 100 on consumption. If it is fully passed through under the assumed market structure, consumers finance the replacement. Under a fixed-tax scenario, A cuts 100 in public expenditure. The welfare consequences depend on which programs are cut and the value recipients place on them; the expenditure amount is not automatically equal to the welfare loss. Under debt financing, the immediate reduction in tax collections affects borrowing and future debt-service obligations. The eventual distribution depends on interest rates, who holds the debt and future tax and spending decisions. In any case the alternative scenario includes real administrative and financing costs where relevant.

An offshore tax advantage can therefore change effective progressivity through two mechanisms. It can concentrate private savings among households owning internationally mobile capital. It can also induce a revenue replacement that falls on a less mobile tax base or reduce public transfers and services that are especially valuable to different groups. Neither mechanism is implied by the arithmetic of lost revenue alone. The effect on inequality requires joint observation of ownership and fiscal adjustments.

Official distributional accounts sometimes allocate corporate tax payments across households using fixed incidence assumptions. For example, the US Congressional Budget Office's household income distribution analysis (2024, income year 2021) assigns 75% of corporate income taxes to capital income and 25% to labor income in its methodology. These are distributional accounting assumptions used for that analysis, not newly estimated causal shares for offshore arrangements. If a study uses them as a baseline, it should explicitly test alternative allocations and avoid presenting the result as direct evidence identifying the beneficiaries of cross-border profit shifting.

A stylized combined account makes the sensitivity visible. Suppose tax enforcement raises 100 in corporate taxes, and allocation A assigns 75 to capital owners and 25 to labor. Allocation B assigns 40 to capital owners, 35 to labor, 15 to consumers and 10 to landowners. If the government distributes 100 equally across five equally sized household groups, each group receives 20 in gross transfers. The net effect on each group still depends on its share of capital, labor, consumption and land ownership. Suppose the highest group owns 60% of capital, receives 30% of labor income, accounts for 25% of consumption and owns 50% of land. In scenario A its allocated private burden is 45 plus 7.5, or 52.5, and net transfer-adjusted burden 32.5. Under scenario B its allocated burden is 24 plus 10.5 plus 3.75 plus 5, or 43.25, and net burden 23.25. The two distributions differ despite identical tax receipts. All parameters are artificial and used solely to show how to compute a conditional allocation.

The decomposition does not capture excess burden from changed economic behavior or differentiated marginal benefits from public services. A household may simultaneously be a worker, owner, consumer, tenant and beneficiary, and its roles must be combined using the same economic unit. If replacing revenue involves an expenditure cut rather than a transfer, welfare analysis requires assumptions about the value of foregone spending that are explicitly separate from accounting incidence.

For global comparison, the residence of both private beneficiaries and government recipients matters. A revenue loss in a low-income source jurisdiction and a tax receipt in a high-income residence jurisdiction can offset in global money accounting while changing the distribution of public resources. A worldwide aggregate can therefore conceal meaningful shifts between populations. Evaluating distributive consequences requires the distinction between source, residence and host government, even when no new worldwide tax liability is created.

Host jurisdictions and competing claims to income

Financial centers are economies with employees, residents, land, firms, public budgets and external investors. Offshore tax incidence concerns them as recipients of economic activities and income as well as locations in which profits or assets are recorded. A jurisdiction that collects low taxes on a large external base may obtain receipts that are significant relative to its domestic economy. Other revenues can originate in licensing, professional services, employment, consumption and property. These quantities must be distinguished from the global tax advantage retained by multinational owners.

In an idealized host jurisdiction, an offshore corporate service contributes 3 units in fees paid to providers, 1 in additional income tax receipts and 2 in local wages. If the 3-unit service fee includes the 2 units of wages and 1 unit of profit, adding the fee and wages as separate components of value added would overstate domestic income. If 1 of the profit is paid to nonresidents, the distribution of local national income differs from production measured domestically. A public budget receipt can finance local goods and services but need not accrue directly to private residents.

When analyzing a reform that removes part of a center's tax advantage, the second-order effects matter. Reduced bookings may have little impact on local productivity if the structures use minimal local resources, while reductions in genuine expertise-intensive activity may affect labor, property and service exports. If replacement work becomes available in other sectors, the long-run cost can differ substantially from the immediate loss of sector income. Employment concentration, retraining, housing supply, and the ability of businesses to change offerings influence realized outcomes.

Source and residence authorities may also disagree about the tax claim on income earned through a network. From the incidence perspective, the government's allocation rule matters because receipts accrue to different populations, and the enterprise faces the resulting total effective liability. If a residence tax credit offsets the full increase in source tax, private liability at the group level may remain unchanged although government revenues are redistributed. If no credit is available, the same source increase may instead raise aggregate group liability and induce market effects. Double-taxation relief and collection priority therefore shape the incidence of cross-border taxation independently of the legal debate about the appropriate jurisdiction of each claim.

A rigorous comparison avoids assigning a unitary national interest to every actor in an offshore center. Financial professionals, external owners, domestic consumers, tenants and landowners may receive different portions of the gains or losses. The jurisdiction's net fiscal benefit is one component of the distribution. The interests of source-country workers and service users can differ from those of its capital owners. Neither side of the cross-border comparison is a homogeneous economic agent.

Minimum taxation, coordinated reforms and transition effects

A coordinated minimum tax changes the relative advantage of reporting profits in low-tax jurisdictions. Its distributional effects depend on treatment of covered income, deductions and exclusions, implementation by individual jurisdictions and the rules determining which authority collects additional taxes. Tax competition can persist on other margins such as investment subsidies, non-tax amenities and the treatment of business functions. An incidence analysis must use the actual rule configuration for the population and period being studied.

Hugger and colleagues' Global Minimum Tax and MNE Profit (2024) models the prospective effects of international minimum taxation. The authors estimate reductions in shifting incentives and low-taxed profit and an increase in corporate tax revenue under their modeled conditions. These are ex ante results. They do not determine the amount that will be collected in each jurisdiction under all possible subsequent policy choices, or how that amount will be distributed among private households.

A simplified example illustrates the distinction. A group has 100 units of profit subject to a 5% effective tax under one regime. Assume a hypothetical minimum rate of 15% applies to the entire 100, with no exclusions, credits, adjustments or behavioral responses. Existing tax is 5 and additional tax is 10. If the host jurisdiction collects the full additional 10, total liability becomes 15 and the additional receipts accrue there. If the parent jurisdiction collects the same amount instead, aggregate group liability is still 15 and the initial private accounting effect is identical, while public recipients differ. Actual minimum-tax rules are more complex: this example is an incidence comparison of simplified collection alternatives, not a calculation of liability under existing law.

The private effects can change over time. In the short run, an unexpected reform may reduce the value of an existing pool of after-tax profit. In the medium run, firms can adjust investment, location, pricing and internal allocation of functions; governments may modify policies and incentives. Over a longer horizon, changes in the stock of productive capital and market structure can generate effects for workers and consumers. The transition also determines whether firms incur sunk reorganizational costs, which may differ from the recurring effect of the tax.

The availability of recent evidence does not eliminate this distinction. Contreras, Hugger, and Zawisza's MNE Responses to the Global Minimum Tax (2026) investigates early implementation using group-level data and a difference-in-differences comparison around the EUR 750 million revenue scope threshold. The study examines effective tax rates, investment and employment, including anticipation. Its early ex post design addresses observed responses of firms near a specific eligibility margin; neither threshold-based evidence nor projected revenue by itself identifies a global household incidence profile. No numerical effect is attributed here beyond the verified description of the study design.

A coordinated reform can alter the distribution of economic rents by narrowing the tax advantage enjoyed by firms with preferred access to particular arrangements. If the additional burden is mostly a tax on existing rents, shareholders may bear much of its immediate effect. If the reform substantially increases the marginal cost of productive investment, factor and product-market adjustments can matter more. A stronger effective burden for large firms relative to smaller firms may affect competition in either direction: it may reduce a prior tax advantage, or it may introduce a new differential depending on each firm's tax and operating position. This relationship must be investigated, not assumed from the statutory threshold.

The assessment of reforms should match outcomes to mechanisms. Changes in effective tax rates and reported profits test an accounting and compliance channel. Changes in capital spending and operating assets test real investment. Employment, compensation and sales prices help identify the distribution across workers and consumers. Government collection and expenditure data determine fiscal closure. All must be studied over compatible horizons, using comparison groups affected by different exposure rather than assuming identical treatment of all enterprises.

Evidence architecture and empirical identification

A complete empirical incidence study would assemble several mutually compatible sources of information. Corporate tax records establish assessed liabilities, tax base and payments. Consolidated company accounts establish the aggregate financial position and distributions to claimants. Affiliate-level reports establish the jurisdictional allocation of reported profit and activity. Payroll, employment and price records describe changes in the real outcomes of workers and consumers. Property transactions and lease data help measure local capitalization. Ownership and household income data allocate exposures across people. Public finance data describe how revenues and expenditures adjust.

The Country-by-Country Reporting Statistics in OECD (2026) illustrate the reach and limitations of corporate reporting data. The edition reports fiscal-year 2023 information from 60 submitting jurisdictions covering more than 9,400 multinational groups; five other jurisdictions reported receiving zero reports. It describes the distribution of reported profit, taxes and economic activities, and discusses high-level indicators of base erosion and profit shifting. These aggregates are valuable for comparing reported accounting geography with activity measures. They do not map ultimate shareholders, individual employees' wages or consumers' welfare changes. Submission and aggregation limit coverage, and historical reporting periods differ from the publication year.

Specific empirical strategies answer different questions. Event studies can describe how outcomes evolve before and after a tax change, with pre-event coefficients providing evidence relevant to the parallel-trends assumption. Differences-in-differences designs can exploit differential exposure across groups, provided their untreated outcome trajectories are credible comparators. Threshold-based designs can study changes in behavior around a statutory boundary, while accounting for manipulation, anticipation and other policies at the same threshold. Instrumental-variable strategies can use plausibly exogenous changes in relative tax incentives or exposure to a reform, but the exclusion restriction is substantive and must be defended. Structural models can translate estimated responses into welfare measures, conditional on maintained functional forms and parameters.

A generic specification clarifies the object. Let Y(i,t) be a firm, worker, consumer or property outcome at time t; X(i) be predetermined exposure to a policy; Post(t) identify the post-policy period; and Z(i,t) contain appropriate predetermined or exogenous covariates. A baseline panel model with unit and time fixed effects includes an interaction X(i) × Post(t). Its coefficient describes the differential post-policy change associated with exposure under the identifying assumptions. If Y is log nominal wage, the result is an approximate proportional wage response. If Y is effective tax rate in percentage points, the result is a rate response. These coefficients cannot be interpreted as comparable fractions of one tax burden without conversion to common levels, populations, counterfactuals and units.

The use of an offshore affiliate as an explanatory variable creates selection issues. Firms choose multinational structures, financing arrangements and business functions. High-productivity firms may enter foreign markets and also differ in wages, markups and profitability. Simple comparisons of firms with and without an offshore affiliate risk attributing pre-existing differences to offshore access. Similarly, owners of offshore wealth may differ in risk tolerance, income, business ownership and migration histories. Selection influences both the availability of offshore arrangements and the outcomes measured.

A stronger design uses a policy change with different predetermined exposure and assesses trends before the change. It can compare affiliates within multinational groups, employers exposed to different effective tax wedges, consumers of otherwise comparable products, and locations with different initial concentrations of affected business functions. The treatment may change at the firm-group rather than subsidiary level. Standard errors and spillover analysis must reflect the level of policy assignment and the interdependence of observations. A reform affecting a large group can change prices, employment and competition for nominally untreated firms, violating a simple no-interference assumption.

The difference between a marginal treatment effect and a worldwide total is central. If a study estimates a wage response to German municipal tax changes, it estimates that policy margin under its institutional context. A US state reform may change local apportionment incentives as well as the tax rate. A global minimum tax changes the outside option supplied by low-tax jurisdictions. The three empirical shocks differ. A responsible synthesis uses them to identify plausible channels, not to form an average of published percentage burdens.

Incidence estimates should report uncertainty from sampling, parameter choices, data missingness and counterfactual definition. A sensitivity analysis can vary labor mobility, capital ownership, pricing and fiscal alternatives. It should report the extent to which conclusions remain stable. When access to full text, replication files or underlying confidential administrative records is limited, the specific source and claim assessed should be recorded. This review identifies and compares reported research findings rather than replicating their regressions or pooling their estimates.

Analytical synthesis and comparative illustrations

The preceding analysis supports an organized sequence of substantive questions for an offshore incidence assessment. First, identify the change in the legal and effective tax position relative to a specified alternative. Second, separate movement of reported income from changes in real activity, risk and financing. Third, follow the associated changes in payments and asset values through all economically relevant claimants. Fourth, account for government budget adjustment, including the division of receipts among jurisdictions. Fifth, translate exposure into household outcomes while avoiding duplication between factor, product and asset markets. This is a practical synthesis of established incidence and international taxation reasoning, not a proposed new named model.

Consider three idealized arrangements with an identical 12-unit reduction in group tax liability. In arrangement X, the reduction concerns existing rent, and neither production nor prices nor compensation changes. The owners receive the initial 12. In arrangement Y, the reduction increases investment and the value of long-term contracts; at the chosen period owners receive 5, employees 4 and consumers 3, with all amounts expressed as hypothetical direct gains over that period. In arrangement Z, financing becomes less costly and an intermediary charges 2 for services, with contractual gains distributed across borrowers and investors; the private group net tax-and-fee saving starts at 10 before additional behavioral responses. These cases are intentionally stylized. They show why identical tax changes need not have identical effects, and why a provider's fee income and the opportunity cost of the services must be accounted for separately.

The first-order private gain in X can be capitalized into asset prices if it is persistent and unexpected. In Y, the reported labor gain must be measured against an appropriate employment and compensation baseline; the consumer gain requires price and quantity information. In Z, the gain from a lower cost of financing is distinct from the tax-related gain; an empirical study would estimate both and identify the owners receiving net returns. If government revenue falls by 12 and is replaced by an increase in another tax or a cut in expenditure, all three cases require a fiscal adjustment in the same units where feasible. No aggregate welfare ranking follows from the direct private benefit alone.

A comparative design for three jurisdiction types illustrates additional variation. In a high-income market economy with diversified ownership and substantial institutional investors, offshore corporate advantages can reach numerous households through investment claims; effects may remain concentrated if the largest holdings are held by high-wealth households. In a lower-income source economy with an administratively narrow tax base, a reduction in corporate receipts may change the feasible composition of expenditure and taxation substantially. In a specialized host center, genuine financial activity and tax-induced recording attract potentially different labor, property and government gains. These are analytical cases, not asserted descriptions of individual countries. Scholarship must investigate each jurisdiction using its actual institutions and evidence.

A relatively mobile factor is more capable of responding to a poorly coordinated tax change, but the consequences depend on whether mobility is physical, contractual, accounting-based or a matter of financial ownership. Moving recorded profit without moving productive assets can change the effective tax while leaving current output relatively stable. Moving productive capital changes factor productivity and local demand. Moving ownership claims reallocates the recipients of future returns without necessarily moving productive property. A policy can simultaneously affect several margins, and their directions can oppose one another. An aggregate quantity labeled offshore capital cannot distinguish them.

An analyst can check the sign and units of each channel before attempting a complete welfare estimate. An increase in liability lowers the group's cash resources at fixed revenue and spending; its ultimate allocation need not equal that cash change once behavior changes. A government receipt is a benefit to a public budget but becomes a household welfare effect only through subsequent allocation. In consolidated global accounting, one government losing a receipt while another gains it creates a redistribution of fiscal resources; summing the two as separate losses would overstate the global effect. Market adjustment creates changes in output and resource use beyond these transfers.

The original contribution of this dossier is the explicit integration of these accounting boundaries with the literature on open-economy incidence, household ownership and compliance, spatial incidence, consumer price responses and fiscal closure. This integration provides a checkable explanation for why widely reported offshore tax-base estimates cannot directly answer a household distribution question. Its substantive value lies in the distinctions and comparative reasoning, not in a claim to have independently established a universal empirical law.

Limitations and research priorities

The evidence base contains foundational theory, firm and worker studies, studies of wealth and enforcement, official reporting data and assessments of coordinated policy. The studies span different time periods, tax instruments and institutional settings. Some original papers could be inspected only through publisher abstracts or indexed excerpts because full-text or direct retrieval was unavailable. The sources and scope of each verification are recorded in the accompanying evidence ledger; the quantitative claims in this manuscript are restricted to the verified portions. No replication data were executed, no new microdata were estimated, and no confidential taxpayer information was processed.

The principal remaining scientific challenge is integrated measurement. Firm-level offshore exposure and changes in effective tax rates would have to be linked to workers, products, financial claimants and property outcomes, and to the public use of collected revenue. International comparability is difficult because tax systems and data definitions differ, beneficial ownership records can be incomplete, and tax shocks often coincide with other economic reforms. Cross-jurisdiction spillovers further complicate comparison groups. A future study should test which portions of the burden are shifted across ownership, labor and product markets over short and long horizons using the same policy shock and common denominator.

An important empirical priority is the distribution of access to tax advantages within sectors. The observed advantage of a multinational firm may differ from that of a domestic competitor with otherwise similar productive opportunities. Reforms can change this differential. Measuring the effect requires information on investment opportunities, industry structure, actual effective tax liabilities and competing firms' outcomes. Industry-level changes in concentration should not be attributed to offshore finance without evidence distinguishing these determinants from changes in technology, trade, demand or regulation.

A second priority is the incidence of legitimate financial intermediation costs and transparency rules on compliant investors and users. Closely linked data could determine whether compliance fees are absorbed by intermediaries, passed to customers or accompanied by service changes. An economic evaluation should measure the tax compliance benefits, administrative costs, substitution between financial products and resulting effects on households separately.

A third priority is the distribution of additional public revenues across countries and within their populations. The collection of an international top-up tax by different jurisdictions can leave a group's total liability unchanged but redistribute fiscal resources. Identifying the marginal recipients of government funds and the policies displaced by reform is therefore indispensable. A worldwide distribution analysis that allocates all public revenue identically across individuals regardless of jurisdiction omits a central feature of the offshore system.

The analysis also requires institutional continuity. Historical evidence about corporate tax rules should not be treated as a description of current law; recent reports based on earlier fiscal years should not be interpreted as contemporaneous evidence. Future research should document effective dates, transitional rules, thresholds, data vintages, reporting delays and definitions. The comparison between older case studies and new coordinated reforms is valuable precisely because the mechanisms, rather than fixed historical numerical shares, are the subject of inference.

Conclusion

Offshore finance modifies tax incidence through the separation of ownership, taxable income, financial intermediation, productive activity and governmental collection across jurisdictions. The economic consequences of that separation depend on the relevant effective tax wedge, the available margins of adjustment, the ownership of economic claims and the response of public budgets.

The reviewed theoretical and empirical literature establishes the relevance of multiple channels. Corporate taxation can affect capital owners, workers, consumers and landowners; the relative strength of those effects varies across instruments and institutional environments. Offshore profit shifting can redistribute booked income and tax receipts without an equivalent relocation of productive activity. Household offshore wealth can generate compliance and distributional concerns where income or assets are concealed, while legitimate cross-border holdings can perform ordinary functions of diversification and investment. The incidence of reforms directed at these arrangements depends on actual exposure, not simply on the presence of an offshore account or affiliate.

The most consequential analytical conclusion is that changes in tax collections, measured shifted profits and private welfare are separate quantities. A tax saving can be retained by an owner, shared with workers and customers, capitalized into scarce assets or partly captured by service providers. A source-country revenue loss can be offset by collections elsewhere, by substitute taxation or by public spending adjustments. Each outcome has its own recipients, units and counterfactual. The consolidated worldwide accounting result is a necessary condition for a coherent analysis; it does not specify the geographic or household distribution of gains and losses.

A scientifically credible account of the offshore system therefore needs both international tax-base evidence and incidence evidence. It must link the tax event to real outcomes, distinguish the mobility of the booked base from that of productive factors, assign financial claims to their ultimate economic owners, and evaluate public fiscal closure. Published estimates from different studies can inform each mechanism without yielding a universal burden share. The strength of this approach lies in the ability to state precisely what is measured, what is inferred and what requires further evidence. It supports economic evaluation of offshore arrangements and international reform while preserving the distinction between observed evidence, conditional examples and authorial synthesis.

References

  • Alstadsæter, Annette, Niels Johannesen, and Gabriel Zucman. 2018. Who Owns the Wealth in Tax Havens? Macro Evidence and Implications for Global Inequality. Journal of Public Economics 162: 89–100. DOI: 10.1016/j.jpubeco.2018.01.008. Source
  • Alstadsæter, Annette, Niels Johannesen, and Gabriel Zucman. 2019. Tax Evasion and Inequality. American Economic Review 109 (6): 2073–2103. DOI: 10.1257/aer.20172043. Source
  • Baker, Scott R., Stephen Teng Sun, and Constantine Yannelis. 2020. Corporate Taxes and Retail Prices. NBER Working Paper 27058, revised March 2023. DOI: 10.3386/w27058. Source
  • Congressional Budget Office. 2024. The Distribution of Household Income in 2021. Washington, DC: Congressional Budget Office. Source
  • Contreras, Lucía, Felix Hugger, and Tom Zawisza. 2026. MNE Responses to the Global Minimum Tax. OECD Taxation Working Papers 77. Paris: OECD Publishing. DOI: 10.1787/0bc9aac9-en. Source
  • Desai, Mihir A., C. Fritz Foley, and James R. Hines Jr. 2006. The Demand for Tax Haven Operations. Journal of Public Economics 90 (3): 513–531. Source
  • Fuest, Clemens, Andreas Peichl, and Sebastian Siegloch. 2018. Do Higher Corporate Taxes Reduce Wages? Micro Evidence from Germany. American Economic Review 108 (2): 393–418. DOI: 10.1257/aer.20130570. Source
  • Fullerton, Don, and Gilbert E. Metcalf. 2002. Tax Incidence. NBER Working Paper 8829. Cambridge, MA: National Bureau of Economic Research. DOI: 10.3386/w8829. Source
  • Gravelle, Jane G., and Kent A. Smetters. 2001. Who Bears the Burden of the Corporate Tax in the Open Economy? NBER Working Paper 8280. DOI: 10.3386/w8280. Source
  • Harberger, Arnold C. 1962. The Incidence of the Corporation Income Tax. Journal of Political Economy 70 (3): 215–240. DOI: 10.1086/258636. Source
  • Hugger, Felix, Ana Cinta González Cabral, Massimo Bucci, Maria Gesualdo, and Pierce O’Reilly. 2024. The Global Minimum Tax and the Taxation of MNE Profit. OECD Taxation Working Papers 68. Paris: OECD Publishing. DOI: 10.1787/9a815d6b-en. Source
  • Johannesen, Niels, Patrick Langetieg, Daniel Reck, Max Risch, and Joel Slemrod. 2020. Taxing Hidden Wealth: The Consequences of US Enforcement Initiatives on Evasive Foreign Accounts. American Economic Journal: Economic Policy 12 (3): 312–346. DOI: 10.1257/pol.20180410. Source
  • Johannesen, Niels, Thomas Tørsløv, and Ludvig Wier. 2020. Are Less Developed Countries More Exposed to Multinational Tax Avoidance? Method and Evidence from Micro-Data. World Bank Economic Review 34 (3): 790–809. Source
  • Malgouyres, Clément, Thierry Mayer, and Clément Mazet-Sonilhac. 2023. Who Benefits from State Corporate Tax Cuts? A Local Labor Markets Approach with Heterogeneous Firms: Comment. American Economic Review 113 (8): 2270–2286. DOI: 10.1257/aer.20201753. Source
  • OECD. 2026. Corporate Tax Statistics 2026. Paris: OECD Publishing. Chapter 7, Country-by-Country Reporting Statistics. DOI: 10.1787/73af6222-en. Source
  • Suárez Serrato, Juan Carlos, and Owen Zidar. 2016. Who Benefits from State Corporate Tax Cuts? A Local Labor Markets Approach with Heterogeneous Firms. American Economic Review 106 (9): 2582–2624. DOI: 10.1257/aer.20141702. Source
  • Suárez Serrato, Juan Carlos, and Owen Zidar. 2023. Who Benefits from State Corporate Tax Cuts? A Local Labor Market Approach with Heterogeneous Firms: Reply. American Economic Review 113 (12): 3401–3410. DOI: 10.1257/aer.20230208. Source
  • Tørsløv, Thomas, Ludvig Wier, and Gabriel Zucman. 2023. The Missing Profits of Nations. Review of Economic Studies 90 (3): 1499–1534. DOI: 10.1093/restud/rdac049. Source