we-solve (beta)

Econ-448 · economics · Economics

Corporate-tax incidence under global profit shifting

LLM status Partially solved unverified

Statement

Original atlas identifier: P3; entry 43. Field: Public finance and taxation. Original tractability label: Hard.

Classification: Parameterized theoretical/empirical research agenda; not a certified unsolved theorem.

Definitions and assumptions

Consider countries $j=1,\ldots,J$ linked by trade, capital ownership, and multinational production. A policy vector $p$ includes corporate rates, apportionment, transfer-pricing enforcement, and any explicitly dated minimum-tax constraints. Each structural model $m$ determines real activity, booked profits, wages, consumer prices, land rents, and ownership returns; these are measured separately. Fix equilibrium selection $s_m$ and a population partition into disjoint beneficiary groups $g=1,\ldots,G$, allowing a household's several income sources to be combined before group assignment. Let $EV_g(p;m)$ be aggregate money-metric equivalent variation for group $g$, measured against policy $p_0$, and let $R_j$ and $C_j$ be public revenue and administrative resource cost. Select a differentiable feasible path $p(\epsilon)$ with $p(0)=p_0$. Assume also that every composite $\epsilon\mapsto EV_g(p(\epsilon);m)$ is differentiable at zero when the marginal estimand below is used; differentiability of the policy path alone does not ensure this.

Formal research question

Editorial formulation: the mathematical target below is not a theorem quoted from the references.

Define marginal private incidence by \[ I_g(m)=-\left.\frac{d}{d\epsilon}EV_g(p(\epsilon);m)\right|_{\epsilon=0}. \] At kinks or equilibrium jumps, use a specified one-sided derivative only when its finite limit exists, or the discrete contrast $-[EV_g(p_1;m)-EV_g(p_0;m)]$ for two stated feasible policies $p_0,p_1$. Identify the vector $(I_g(m))_{g=1}^{G}$, its horizon dependence, and the joint response of each $R_j-C_j$. When $\sum_g I_g(m)\neq0$, normalized shares $I_g/\sum_h I_h$ may be reported, retaining signs; they need not lie in $[0,1]$. Compare global and domestic welfare changes using explicitly chosen group weights $w_g$ and public-fund values $\lambda_j$: $\sum_gw_g\Delta EV_g+\sum_j\lambda_j\Delta(R_j-C_j)$. Establish what matched firm--worker--ownership data and exogenous tax variation identify independently of restrictions on profit shifting. If multiple models fit the data, deliver incidence bounds rather than a single bearer.

Interpretation and scope

A tax dollar is a transfer to government, whereas lost production and administrative effort are resource costs; fiscal and private incidence should not be conflated. Short-run rents, long-run investment, and international relocation can generate different incidence vectors. Legal booking jurisdiction is not a welfare beneficiary and does not reveal whose wages or returns change. Local tax reforms identify local responses under their assumptions, not automatically worldwide coordination counterfactuals. Consumption-price effects and pension ownership must be incorporated when the group called “workers” also owns capital. Minimum-tax rules must be described by the legal version actually studied, with a separately checked effective date. The research extension is international joint identification and equilibrium assessment, not a presumption that classical or local incidence models have no solutions.

Source audit and status

All three original academic references are identifiable. [S1] is a classical benchmark; [S2] and [S3] are setting-specific empirical/structural results. The publisher of [S2] links a later comment and reply, which should accompany replication. These sources do not establish contemporary international incidence shares.

References

    },leftmargin=*,itemsep=0.6em]
  1. Arnold C. Harberger (1962). The Incidence of the Corporation Income Tax. Journal of Political Economy 70(3), 215--240. DOI: 10.1086/258636. https://www.journals.uchicago.edu/doi/10.1086/258636
    Locator and support limits: Article, pp. 215--240: closed-economy general-equilibrium incidence benchmark; not a multinational shifting model.
  2. Juan Carlos Suárez Serrato 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. https://www.aeaweb.org/articles?id=10.1257/aer.20141702
    Locator and support limits: Abstract and article: local workers, landowners, and firm owners; publisher links a 2023 comment and reply, so estimates merit that context.
  3. Clemens Fuest, 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. https://pubs.aeaweb.org/doi/abs/10.1257/aer.20130570
    Locator and support limits: Abstract and German municipal analysis: wage incidence; does not estimate a global minimum-tax equilibrium.

Common conventions: Source mathematical conventions

Unless an entry states otherwise, agent and firm sets are finite. State and action spaces are measurable, strategies and outcome maps are measurable, probability laws are specified, and expectations exist for the targets under discussion. Infinite-horizon discount factors lie in $(0,1)$ and discounted objectives are finite. Norms, tolerances and complexity input sizes must be specified for computational comparisons. Constants or primitives that appear locally are inputs, not empirical facts asserted by this catalogue.

Identification. A statistical model $m\in\mathcal M$ implies an observable law $P_m$ and target $\theta(m)$. At law $P$, its identified set is \[ \Theta(P;\mathcal M)=\{\theta(m):m\in\mathcal M,\ P_m=P\}. \] Point identification holds when this set is a singleton, or equivalently when $P_{m_1}=P_{m_2}$ implies $\theta(m_1)=\theta(m_2)$ for all compatible models. Sharp bounds mean bounds attained, or approached when the boundary is not attained, by compatible models. Writing this set is a definition, not a solution: the substantive task is to establish informative restrictions and derive or estimate the set. An unrestricted-confounding model may give only trivial bounds.

Inference. Identification, estimability and valid uncertainty quantification are separate questions. Fix $\alpha\in(0,1)$. A confidence procedure $C_n$ over a declared law class $\mathcal P$ has asymptotic uniform coverage at level $1-\alpha$ when \[ \liminf_{n\to\infty}\inf_{P\in\mathcal P}\Pr_P\{\theta(P)\in C_n\}\geq1-\alpha. \] For set-identified targets the coverage event must be defined separately, for example coverage of the full identified set. If an entry uses identification-indexed models instead of uniquely indexed laws, the same coverage convention is applied over those models. Pointwise limits do not establish uniform validity, and asymptotic validity does not guarantee finite-sample precision. Sample sizes, dependence structures and weak-identification sequences are part of each application.

Counterfactuals. $Y_i(a)$ is a potential outcome under a specified intervention, including its timing, implementation and versions. It is not $Y_i$ conditional on voluntarily choosing $a$. A full intervention vector permits interference; shorthand scalar treatment notation does not silently impose no interference. Assignment, selection, attrition, anticipation and measurement must be specified. A claim about an instrument's local effect is not automatically a population policy effect.

Economic equilibrium. A counterfactual model includes best responses, constraints, feasibility, market clearing, state transitions and expectations consistent with the stated information. When several equilibria exist, either a declared selector is an input or the target is set-valued. Comparisons across policy regimes must use the stated selection convention. Reduced-form relationships are not presumed invariant after an equilibrium-changing intervention.

Optimization and welfare. Optimization is over the stated feasible set. If attainment is not established, $\sup$ or $\inf$ and $\varepsilon$-optimal choices replace an asserted maximizer or minimizer. For example, with value $V=\sup_{a\in\mathcal A}W(a)<\infty$, an $\varepsilon$-optimal set is $\{a\in\mathcal A:W(a)\geq V-\varepsilon\}$ for $\varepsilon>0$. Benefits, costs, risk and health or environmental outcomes can be aggregated only using declared units and conversion weights. Interpersonal utility weights, the treatment of fiscal transfers and foreign profits, discounting, and the behavioral welfare criterion are explicit inputs. A comparative-static derivative additionally requires existence and appropriate differentiability of the selected counterfactual.

Sources. Local labels [S1], [S2], and so forth restart in every question. Locators identify the relevant abstract, model, section or result at the level actually checked; a bibliographic or abstract-level verification is not represented as inspection of an entire proof. Working papers are distinguished from later publications and changing versions. Source summaries are paraphrased. All URL checks and editorial formulations refer to the audit date stated above.

Statement licence: CC-BY-4.0 · Source file

Contributions 1

Econ-448-A1 · Partial result · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-448

Booking responses, real losses, and household ownership

Status: Partial; the full catalogue problem remains unresolved.

Abstract

An exact two-jurisdiction shifting model separates owner incidence, both governments' revenues, and real avoidance costs. A mobile-capital model produces wage incidence through another channel. Two economies with identical firm observables and different ownership structures demonstrate nonidentification of household-group incidence. The exercise supplies exact restricted results without claiming a universal corporate-tax bearer.

Policy and accounting conventions

The catalogue asks for household money-metric incidence under a feasible tax-policy path, together with public revenue and resource costs. Households belong to disjoint groups after combining wages, pensions, and direct ownership. A booking jurisdiction is not a beneficiary group.

The policy here is a marginal increase in a home corporate rate $t$ holding a haven rate $h$ fixed. We work locally where $t>h$. No current minimum-tax legislation is asserted or modeled. Quasi-linear utility makes a net income change equal to equivalent variation in a common consumption numeraire. Public revenues are recorded separately until a global welfare account with equal weights is explicitly formed.

A pure shifting channel

A firm has fixed pre-tax operating rent $\Pi>0$, net of labor and other real production costs. It books $s\in[0,\Pi]$ in the haven and $\Pi-s$ at home. Shifting consumes real resources $ks^2/2$, with $k>0$, and this cost is not deducted from either tax base in the declared convention. Owners maximize \[ U(t,s)=\Pi-t(\Pi-s)-hs-\frac{k}{2}s^2. \tag{1} \] Wages, real output, and consumer prices are fixed, isolating the shifting margin.
Theorem: If $0<t-h<k\Pi$, the unique optimum and marginal responses are \[ s^*=\frac{t-h}{k},\qquad I_{\rm owner}=-U'(t)=\Pi-s^*, \tag{2} \] \[ R_H'=\Pi-s^*-\frac{t}{k},\qquad R_F'=\frac{h}{k}. \tag{3} \] With equal global owner and public-fund weights, \[ U'+R_H'+R_F'=-s^*=-\frac{d}{dt}\left(\frac{k(s^*)^2}{2}\right). \tag{4} \]
Proof: Differentiation of (1) with respect to $s$ gives $t-h-ks$, with second derivative $-k<0$. The interior restrictions establish (2). In differentiating the optimized value, the coefficient multiplying $s^{*\prime}$ vanishes by this first-order condition, leaving $U'=-(\Pi-s^*)$. Revenue levels are $R_H=t(\Pi-s^*)$ and $R_F=hs^*$; differentiate using $s^{*\prime}=1/k$ to obtain (3). Summing gives $-(t-h)/k=-s^*$, while differentiation of shifting costs gives $ks^*/k=s^*$. This proves (4). ∎
For $\Pi=10$, $k=0.2$, $t=0.3$, and $h=0.1$, shifted profits are $1$, owner incidence is $9$, the home revenue response is $7.5$, and the haven revenue response is $0.5$. Global resource welfare falls at rate $1$, exactly as (4) requires. These are synthetic parameters.

Owner burden divided by the home revenue response is $1.2$. It is not a probability or a share constrained to $[0,1]$. Some of the discrepancy reflects foreign revenue and some avoidance resources. If domestic welfare excludes foreign owners or haven receipts, its derivative differs again. Equal global weights were a declared accounting convention, not an empirical assertion about governments' objectives.

At the boundaries, $s^*=0$ when $t\le h$ and $s^*=\Pi$ when $t-h\ge k\Pi$. One-sided derivatives replace the smooth formulas at the thresholds. This check prevents an interior formula from predicting impossible negative home tax bases. A cap or enforcement kink in an applied policy would require the same care.

A different real-activity mechanism

Suppress shifting and consider a small open economy with one unit of inelastically supplied labor and production $F(K)=AK^a$, where $A>0$ and $a\in(0,1)$. Competitive factor prices are wage $w=(1-a)AK^a$ and pre-tax capital return $aAK^{a-1}$. A source tax at rate $t<1$ applies to capital operating income. Mobile investors require after-tax return $r>0$, so equilibrium capital obeys \[ (1-t)aAK^{a-1}=r,\qquad K(t)=\left(\frac{(1-t)aA}{r}\right)^{1/(1-a)}. \tag{5} \] Consequently \[ \frac{d\log w}{dt}=-\frac{a}{(1-a)(1-t)}<0. \tag{6} \] To verify (6), differentiate the logarithm of (5), obtaining $d\log K/dt=-1/((1-a)(1-t))$, and then use $\log w=\log((1-a)A)+a\log K$. The after-tax return per unit of mobile capital remains $r$ by construction.

For $a=1/3$ and $t=0.2$, the wage semi-elasticity is $-0.625$. A one-percentage-point increase therefore gives a local approximation of a $0.625$ percent wage reduction in this model. This is neither an empirical estimate nor a prediction for a contemporary international reform.

The mechanisms differ: fixed rents in (1) place private burden on owners, whereas mobile capital in (5) adjusts until its net return is restored and reduces labor's marginal product. A realistic multinational economy may have both responses, rents, land, consumer-price effects, and adjustment costs. Booked-profit responses alone do not distinguish these real mechanisms. Short-run fixed capital and long-run mobile capital also define different estimands.

An ownership nonidentification theorem

Return to the shifting economy and partition households into employed worker households and all other households. Worker households own fraction $\omega$ of the firm through shares or pensions; their wage income remains fixed. All household wealth effects on firm demand are absent under the maintained quasi-linear fixed-demand environment. Then \[ I_{\rm worker}=\omega(\Pi-s^*),\qquad I_{\rm other}=(1-\omega)(\Pi-s^*). \tag{7} \] The groups are disjoint. Worker shareholders are not counted again in an overlapping capital-owner group.

Proposition: Wages, production, booked profits, tax rates, and both governments' corporate receipts do not identify the vector (7) when ownership is unobserved and unrestricted.
Proof: Hold every primitive in (1) fixed. In one economy let $\omega=0$; in a second let $\omega=1$. Firms maximize the same aggregate profit and choose the same $s^*$, while wages and demand are unchanged. Every listed observable therefore agrees across the economies. In the first, the incidence vector is $(0,\Pi-s^*)$; in the second it is $(\Pi-s^*,0)$. Since $\Pi-s^*>0$, the targets differ under identical observable laws. ∎
If validated ownership information instead gives $\omega\in[\underline\omega,\overline\omega]$, the sharp worker-burden interval is \[ [\,\underline\omega(\Pi-s^*),\quad \overline\omega(\Pi-s^*)\,]. \tag{8} \] Every ownership share in that range leaves firm behavior unchanged and realizes its corresponding endpoint or interior burden. This proves sharpness, not merely an outer bound.

For the numerical shifting example and an ownership range $[0.2,0.5]$, worker incidence is between $1.8$ and $4.5$, although employer wages do not change at all. Wage effects and incidence on households called workers are therefore different quantities. This remains true if ownership is indirect through retirement funds.

Revenue recycling introduces another allocation. A lump-sum rebate to workers affects their total policy equivalent variation even if corporate behavior stays fixed. An applied incidence path must state whether recycling enters household welfare or is represented separately by public-fund values.

Remaining research gap

The attempt establishes a global resource identity and sharp ownership bounds in one rent-shifting economy. It also derives a separate mobile-factor response. It does not integrate all channels into a multi-country equilibrium, identify ownership in data, estimate tax instruments, or validate the fixed-demand and mobility assumptions. Administrative costs and foreign income weights would also need evidence and explicit treatment.

The algebra is a self-contained use of standard incidence reasoning, with no claim to resolve contemporary international incidence shares. The catalogue agenda remains unresolved because determining which real and financial margins apply requires matched firm, worker, ownership, and policy evidence.

Completion Estimate

45%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt proves exact two-jurisdiction shifting and global resource identities, derives a separate mobile-capital wage response, and establishes sharp ownership incidence bounds. An integrated multicountry equilibrium with matched ownership data, consumer and land effects, administration costs, and horizon-dependent policy identification remains missing.

References

  • [sz] J. C. Suárez Serrato and O. Zidar (2016). Who Benefits from State Corporate Tax Cuts? A Local Labor Markets Approach with Heterogeneous Firms, American Economic Review 106(9), 2582--2624. The primary publication treats setting-specific worker, landowner, and owner channels; its estimates are not used as universal inputs. https://www.aeaweb.org/articles?id=10.1257/aer.20141702.

Go to contribution →