we-solve (beta)

Econ-514 · economics · Economics

Entrant versus incumbent innovation

LLM status Partially solved unverified

Statement

Original subject group: Growth and productivity

Audit classification: Explicit published open question. Complementarity of creative destruction and incumbent innovation is explicit; the derivative design is editorial. Verification concerns the cited version, not first priority or proof that this exact editorial specification remains unsolved on October 8, 2026.

Definitions and precise research target

Editorial mathematical specification.
Fix a product-line innovation model with incumbents choosing research intensity $r\ge0$ and potential entrants generating creative-destruction hazard $\lambda\ge0$. Let $V(r,\lambda;w,p)$ be incumbent expected discounted profit net of research cost, with wage $w$ and demand parameters $p$ initially fixed. Under a unique differentiable interior best response $r^*(\lambda;w,p)$, define strategic complementarity or substitutability by the sign of \[ \frac{\partial r^*(\lambda;w,p)}{\partial\lambda}. \] Identify this derivative over a declared range of entrant pressure and distinguish it from the general-equilibrium derivative when research wages, product demand, and financing costs also change. Specify incumbent own-product innovation separately from acquisition of entrants or innovation in unrelated lines. Use instruments shifting entrant innovation opportunities while satisfying stated exclusions for incumbent demand and research productivity; provide bounds if those exclusions are only partly defensible. Match incumbent research, product-quality changes, and entrant hazard, rather than inferring strategic effects from aggregate entry alone. Characterize mechanisms under which replacement risk discourages research or an escape-from-competition incentive raises it, and test whether their quantitative balance varies across technological distance. The target is a causal strategic-response function, not merely an accounting decomposition of growth into incumbent and entrant contributions.

For $V$ twice continuously differentiable, an interior optimum satisfying $V_r=0$ and $V_{rr}<0$ has $\partial r^*/\partial\lambda=-V_{r\lambda}/V_{rr}$. Its sign is therefore the sign of the cross-partial $V_{r\lambda}$; this calculus identity is not the unresolved economic question. The sign depends on primitives and empirical causal responses and need not be universal across technologies. At a boundary or with multiple best responses, define finite differences or monotonicity of the best-response correspondence instead. General-equilibrium responses additionally include $r_w^*dw/d\lambda+r_p^*dp/d\lambda$ and other endogenous prices. Entrant hazard should be measured as the arrival probability/intensity of innovations that displace the incumbent's own product, not all firm entry.

Variants and assumption changes

The following are editorial variants, not additional published open conjectures.
  1. Replacement-only benchmark. Let $V(r,\lambda)=\pi(r)/(\rho+\lambda)-c(r)$ with $\rho>0$, $\pi'>0$, and a strictly concave objective. Then $V_{r\lambda}<0$, yielding substitution locally; this is a solved conditional mechanism to compare with data.
  2. Escape and endogenous wages. Let entrant pressure change profits conditional on an incumbent's quality lead, and determine research wages in equilibrium. Identify the partial response at fixed wages separately from the total response, allowing opposite signs.

References and source audit

Complementarity of creative destruction and incumbent innovation is explicit; the derivative design is editorial.
  1. Peter J. Klenow; Huiyu Li. 2021. Innovative Growth Accounting. SectionVII, Conclusion, penultimate research paragraph on innovation arrival rates; published May2021. https://www.journals.uchicago.edu/doi/full/10.1086/712325. DOI: https://doi.org/10.1086/712325.

Common conventions: Source mathematical conventions

These conventions apply to each entry unless explicitly replaced.

Models and identification. A primitive model $m$ specifies all probability laws, feasible choices, information, equilibrium and measurement rules used by its target. The admissible class $\mathcal M$ is fixed independently of outcomes. If $P_O(m)$ is its observable probability law and $\tau(m)$ the target, its identified set at observable law $P$ is \[ \mathcal I_\tau(P)=\{\tau(m):m\in\mathcal M,\ P_O(m)=P\}. \] Point identification means that this set is a singleton. An empty set signals incompatibility of data and assumptions. Coordinate bounds need not describe the joint set. Bounds are sharp only if every retained target is attainable or arbitrarily approachable by compatible models. Finite bounds require explicit restrictions on unobserved quantities; unrestricted confounding, hidden wealth, extrapolation or arbitrary equilibrium selection can yield uninformative sets.

Causal interventions. A potential outcome $Y_i(p)$ is the outcome under a complete policy regime $p$, including timing, financing, information and market spillovers. The target population and units are fixed before treatment. With interference, use the complete assignment vector or a justified exposure mapping; individual no-interference notation alone cannot identify an economy-wide intervention. A difference of mean potential outcomes does not require the cross-world joint distribution. Conditioning on post-treatment migration, survival, enrollment or employment changes the estimand and needs separate justification.

Statistical statements. An estimator, confidence set, forecast or test specifies a sampling law, model class and loss/coverage criterion. Uniform $(1-\alpha)$-coverage means $\inf_{m\in\mathcal M}\Pr_m\{\tau(m)\in C_n\}\geq1-\alpha$, or an explicitly asymptotic version. The whole parameter space trivially has coverage, so informativeness requires a stated width, risk or power objective. A forecasting improvement is relative to a named benchmark, information set and evaluation population.

Equilibrium and optimization. An equilibrium comprises feasible plans, optimality, beliefs where needed, budget constraints and all market/resource clearing. The primitives or a stated benchmark define each correspondence $\mathcal E(p;m)$. Nonemptiness is assumed or proved, never inferred just from compact policy sets. A selected-equilibrium target uses a declared selection rule; a robust target quantifies over all permitted equilibria. Use suprema or infima unless attainment is established. An $\varepsilon$-optimal policy is feasible and within $\varepsilon$ of the finite optimum value. Report an empty feasible set separately.

Welfare and accounting. Normative utility, interpersonal normalization, social weights, numeraire, discounting and the use of public receipts are primitives. Behavioral utility need not coincide with the welfare criterion. Household welfare includes ownership income and rents once; adding profits again to compensating variation generally double-counts them. A monetary equivalent is defined by an explicit transfer and price convention, with monotonicity and range conditions. Average effects, mechanism contributions, transfers and welfare losses are different targets. Infinite sums require convergence and dynamic feasible plans require terminal or no-Ponzi conditions.

Source versions. A working-paper date, revision date and journal publication year are distinguished. A DOI for a journal version is not automatically the DOI of an earlier manuscript. Locators refer to the stated version and printed pages where specified. A metadata-only check does not verify a claim about a full-text conclusion. Every entry's source subsection narrows the provenance claim accordingly.

Statement licence: CC-BY-4.0 · Source file

Contributions 1

Econ-514-A1 · Partial result · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-514

Entrant pressure and incumbent innovation: explicit mechanisms and an equilibrium reversal

Scope

The derivative of incumbent research with respect to entrant displacement pressure is not universally positive or negative. This attempt supplies one strictly concave model with a tractable sign threshold, a research-labor equilibrium that can reverse a positive partial response, and sharp linear instrumental-variable sensitivity bounds. It does not estimate the causal response curve of actual firms. The catalogue already states the implicit-function identity; the purpose here is to identify primitives and information requirements that give it economic content.

Let incumbent own-product research intensity be $r\ge0$ and entrant displacement hazard $\lambda\ge0$. Fix discount rate $\rho>0$ and a research cost coefficient $w>0$. Suppose expected discounted net profit is \[ V(r,\lambda;w)=\frac{(a+b\lambda)r-\tfrac12cr^2}{\rho+\lambda} -\tfrac12wr^2, \tag{1} \] where $a>0$, $b\ge0$, $c>0$. The denominator captures replacement risk. The numerator allows entrant pressure to increase the profit reward to a quality improvement; this is a specified escape incentive, not an inference from entry counts. The objective is strictly concave and goes to minus infinity as $r$ grows, so its unique optimum is interior and finite: \[ r^*(\lambda;w)=\frac{a+b\lambda}{c+w(\rho+\lambda)}. \tag{2} \] The coefficient $w$ may represent the price scale of increasing marginal research inputs, rather than a literal constant wage per unit of $r$.

An explicit strategic sign threshold

Differentiating (2) gives \[ \frac{\partial r^*}{\partial\lambda} =\frac{b(c+w\rho)-wa}{[c+w(\rho+\lambda)]^2}. \tag{3} \] The numerator is independent of $\lambda$ in this benchmark. Hence replacement risk dominates for $b(c+w\rho)<wa$, escape incentives dominate for the opposite strict inequality, and the response is flat at equality. Strict concavity verifies that this derivative describes the global optimum, not just a stationary point.

For $a=c=w=\rho=1$, setting $b=0$ gives $r^*=1/(2+\lambda)$ and substitution. Setting $b=1$ gives $r^*=(1+\lambda)/(2+\lambda)$ and complementarity, with derivative $1/(2+\lambda)^2$. These examples share the same replacement denominator and research-cost curvature. A falling research response is therefore not implied by displacement risk alone once the quality payoff changes with entrant pressure. Conversely, counting more entrants does not establish that the positive $b$ channel exists.

A finite-difference version does not require estimating an infinitesimal derivative. For $\lambda_1>\lambda_0$, direct subtraction in (2) yields \[ r^*(\lambda_1;w)-r^*(\lambda_0;w) =\frac{[b(c+w\rho)-wa](\lambda_1-\lambda_0)} {[c+w(\rho+\lambda_1)][c+w(\rho+\lambda_0)]}. \tag{4} \] If feasible research is capped at $\bar r$, the best response is $\min\{\bar r,r^*\}$. Its monotonicity follows from (3), while strict derivatives can become zero at the cap and fail to exist at its boundary. A smooth interior formula should not be fitted across such a kink without acknowledging the constraint.

A complete research-input price feedback

Now endogenize the research cost scale through a reduced-form research-input market: \[ w=w_0+\chi r+\psi\lambda,\qquad w_0>0,\quad\chi,\psi\ge0. \tag{5} \] An individual incumbent treats $w$ as given; (5) represents the aggregate inverse supply evaluated at symmetric incumbent demand $r$ plus entrant demand proportional to $\lambda$. Let $D=\rho+\lambda$. Combining (2) and (5), the equilibrium research intensity is the nonnegative root of \[ F(r,\lambda):=\chi D r^2+[c+w_0D+\psi\lambda D]r -(a+b\lambda)=0. \tag{6} \] For every $\lambda\ge0$, $F(0,\lambda)<0$, $F(r,\lambda)\to\infty$, and $F_r=2\chi Dr+c+w_0D+\psi\lambda D>0$. Thus the equilibrium exists and is unique. Implicit differentiation gives \[ \frac{dr}{d\lambda}= \frac{b-\chi r^2-[w_0+\psi(\rho+2\lambda)]r} {2\chi Dr+c+w_0D+\psi\lambda D}. \tag{7} \] The denominator is positive. The new numerator includes input-market congestion and entrant demand for those inputs, neither of which is in the fixed-$w$ derivative (3).

Take $a=b=c=\rho=w_0=1$, $\chi=0$, $\psi=3$. At $\lambda=0$ the equilibrium is $r=1/2$, $w=1$. Holding $w$ fixed gives the positive partial derivative $1/4$ from (3). Along equilibrium, (7) gives $[1-(1+3)(1/2)]/2=-1/2$. Thus the same point has strategic complementarity at fixed research prices and a negative total response when entrant pressure raises input costs. The inverse supply equation is a declared market primitive, so this is an internally closed benchmark rather than a partial derivative mislabeled as general equilibrium. Endogenous product demand and finance would add further equations and need their own existence argument.

Instrumental variation with a bounded exclusion violation

Consider a separate local linear empirical specification in a prespecified technological-distance group: \[ R=\alpha+\beta\Lambda+U, \] where $R$ is incumbent own-product research and $\Lambda$ is own-line displacement hazard. Center a candidate instrument $Z$ and suppose the observable population moments are $q=E[Z\Lambda]\ne0$ and $s=E[ZR]$. An exact exclusion sets $E[ZU]=0$ and gives $\beta=s/q$. If instead $|E[ZU]|\le\Delta$, then the exact moment-compatible interval is \[ \beta\in\left[\frac sq-\frac{\Delta}{|q|},\quad \frac sq+\frac{\Delta}{|q|}\right]. \tag{8} \] Necessity follows from $s-\beta q=E[ZU]$. For any retained $\beta$, define the residual from the displayed structural equation using the observed $(R,\Lambda,Z)$ and choose the intercept to match the mean. Its moment then obeys the restriction, proving sharpness within this linear moment class. The construction does not prove that $\beta$ is a causal derivative outside the maintained linear model.

For example $q=0.1,s=0.02,\Delta=0.03$ gives $[-0.1,0.5]$; a positive naive ratio $0.2$ does not establish complementarity under that exclusion allowance. An instrument that also changes incumbent demand or research productivity can violate the intended fixed-price strategic interpretation even if it predicts entry strongly.

Sampling inference can avoid division by a weak estimated first stage: retain every candidate $\beta$ for which a simultaneous confidence region for $(s,q)$ contains some pair satisfying $|s-\beta q|\le\Delta$. On the region's coverage event this projection covers the true compatible slopes. If the region includes $q=0$ and $|s|\le\Delta$, the retained slope set can be unbounded. Imposing a finite interval solely for plotting would conceal weak identification. Tail conditions or clustered sampling assumptions are needed to construct the moment region; none is supplied by the algebra itself.

Measurement and unresolved empirical content

An incumbent's acquisition expenditure is not its own-product research intensity, and total new-firm entry is not necessarily a hazard of displacement in its product line. A credible application would match research inputs, innovations in existing products, rival quality advances, and input prices within technological-distance groups. It would identify the fixed-price response separately from wage and financing feedback, and test the maintained exclusion with relevant institutional evidence rather than with a first-stage coefficient alone.

The benchmark resolves conditional signs and provides a concrete reversal, but the actual sign function across technologies remains an empirical question. The growth-accounting source measures contributions of different innovation types; accounting shares do not identify the strategic causal derivative in (3) or (7). No universal complementarity or substitutability claim is made.

Completion Estimate

50%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt derives replacement-escape sign thresholds, an equilibrium research-input-price reversal, and sharp local linear instrument-exclusion sensitivity bounds. The full causal strategic-response function still requires matched incumbent own-product innovation and entrant displacement data, credible fixed-price exclusions, and validated heterogeneity across technological distance.

References

Go to contribution →