PIATRA . INSTITUTE

frontier governance

when state control of profits helps vs. harms innovation across frontier sectors

based on 2022, IMF, Excess Profit Taxes: Historical Perspective and Contemporary Relevance

ClaudeMarch 2026·initial implementation

Frontier

100.0

Welfare

93.0

State capacity

150.0

Concentration

94.1

Shortage risk

5.9

system trajectory

metric

regime shape

policy phase space

Top-right = high frontier with high welfare

sectoral end-state

Innovation, deployment, concentration, and shortage by sector in the final year

sector deep dive

sector

Compute-intensive frontier sector with high network effects and energy demand.

best instrument

Public R&D + procurement

Best when the problem is frontier uncertainty and positive spillovers rather than monopoly pricing.

sector anatomy

monopoly-like28

frontier intensity95

network effects84

capex burden62

safety burden72

Direct profit control is least suited to sectors with high volatility and long R&D cycles. Utility-style regulation becomes more defensible as monopoly structure rises. Excess-profits taxation is strongest when rents are structural or windfall-like.

historical cases

1917-1945 · Rent capture

WWI / WWII Excess-Profits Regimes

Often good when temporary and targeted

Windfall capture during wartime can tax rents while preserving normal returns and procurement-driven expansion.

20th century onward · Monopoly discipline

Utility Rate-of-Return Regulation

Often good in natural monopolies

Electricity, water, and grids can justify regulated returns because duplicative competition is structurally wasteful.

Late 20th / early 21st century · Control failure

Broad Profit / Price Controls

Often bad in fragile economies

The state cannot infer replacement costs fast enough; shortages, black markets, and underinvestment emerge.

1980s-2000s · Efficiency vs rent

Late-20th-Century Deregulation Waves

Mixed

Can raise efficiency in some sectors, but in networked industries may entrench concentration and rent extraction.

parameter sweep

sweeping profit control from 0 to 100

sensitivity analysis · welfare index

public R&D

Δ16.100

open science

Δ7.700

energy abundance

Δ5.700

safety intensity

Δ4.000

antitrust

Δ3.800

cost of capital

Δ-3.800

monopoly regulation

Δ2.300

subsidies

Δ2.200

profit control

Δ-2.000

excess-profits tax

Δ2.000

labor bargaining

Δ1.100

procurement

Δ0.700

talent mobility

Δ0.000

baseline: 93.000 · each parameter swept min→max while others held constant

The core distinction

The correct question is not “state or market?” but rather: what kind of profit, in what kind of sector, under what informational and energetic conditions? The model is built around three distinct governance instruments that interact differently with sector structure.

Model dynamics

For each sector $s$ and year $t$ , the simulation evolves knowledge, deployment, concentration, and profitability as coupled nonlinear processes:

K_s(t{+}1) = K_s(t) + \text{impulse}_s \cdot (1 - K_s) \cdot 0.18

D_s(t{+}1) = D_s(t) + \sigma(\text{readiness}_s - 0.34) \cdot 0.10

C_s(t{+}1) = C_s(t) + \text{network}_s \cdot 0.035 - \text{antitrust} \cdot 0.038

The welfare index aggregates consumer benefit, state capacity, and resilience, penalized by inequality and shortage:

W(t) = \text{consumer} + \text{strategic} \cdot 0.14 + \text{resilience} \cdot 0.12 - \text{inequality} \cdot 0.24

When control fails

Broad direct profit control fails when the state cannot observe true costs, quality changes, replacement costs, and risk across thousands of firms in real time. The profit does not disappear; it reappears as shortage, lower quality, bribery, queuing, or black-market markup. Venezuela under the 2014 Fair Prices Law is the canonical example.

When control works

Excess-profits taxation can be efficient when it falls only on economic rent: if the profit is a windfall created by war, scarcity, monopoly position, or luck rather than by marginal effort or investment, taxing the excess does less damage. A normal return still remains. Utility-style regulation works in structural monopolies where duplicative competition is wasteful.

Frontier sectors

AI, robotics, private space, private nuclear, advanced materials, and biotech each occupy different positions in the control-frontier tradeoff space. Nuclear has high monopoly-like structure ( $\mu = 0.64$ ) making utility regulation defensible. AI has high network effects ( $n = 0.84$ ) making antitrust and excess-profits taxation more appropriate. Materials science at the speculative frontier ( $f = 1.0$ ) benefits most from public R&D and open science.

Model changelog

v1March 2026

Coupled 6-sector simulation over 35-year horizon with 13 policy parameters
Five regime presets: balanced, wartime, venture, monopoly, breakdown
Five frontier event toggles: AI shock, automation wave, cheap nuclear, cis-lunar, RTSC
Trajectory, radar, scatter, and bar chart visualizations with snapshot comparison
Sector deep-dive diagnostics with best-instrument recommendation
Parameter sweep and sensitivity analysis on welfare index