...

Patent Term Trade-Off: Market Diagram and Timeline of Gains (2D)

The textbook supply-and-demand picture of the Nordhaus patent-term model, plus a swimlane timeline of protected years, inventor profit, consumer gain and deadweight loss. Educational economics, not legal advice.
Cost cut chosen by inventor-
Inventor net gain (PV)-
Consumer gain (PV)-
R&D spending-
Deadweight loss (PV)-
Total social gain G-
Welfare-optimal term T*-
G(T) / G(T*)-
Inventor payback year-
Price while protected → after-
Present value ($M) vs term T: orange total G, gold inventor, green consumers; cyan line T*. Dot: current term.
Cumulative PV over calendar time: gold inventor (starts at minus R&D), green consumers, orange total. Cyan line: payback year.
Independent numerical checks
Each line compares two separate calculations; a FAIL would show here.
The science

This is the textbook economics of patent length (Nordhaus, 1969), reduced to one equation system. It is an educational model of an incentive trade-off, not legal advice and not a description of any real patent system or case.

Setting

Linear demand q = A − p, old unit cost c0, a = A − c0. An inventor pays R&D C(ε) = κε²/2 to cut unit cost by ε. While the patent lasts (T years) rivals still have the old cost, so the patentee can charge at most c0 and earns ε·a per year. After expiry price falls to c0 − ε.

f(T) = (1 − e^(−rT)) / r PV of 1 $M per year for T years Inventor: Π(ε) = ε·a·f − κε²/2 => ε* = a·f / κ Consumers (after expiry): (aε + ε²/2) · (1/r − f) Deadweight loss while protected: ε²/2 per year

Optimal term

Total gain G(T) = Π(ε*) + consumers. Setting dG/df = 0 gives

3r·f² + (2κr − 2)·f − 2κ = 0, T* = −ln(1 − r f*) / r

A longer term raises ε* (more innovation) but delays the price cut and enlarges the deadweight loss. The optimum balances the two.

What the checks do

  • ε*: the closed form against a brute-force grid search over the inventor's own profit (20,001 points).
  • T*: the closed form against a grid search over T of a year-by-year numerical sum of discounted welfare (it uses none of the closed forms for f or the consumer term).
  • Deadweight loss: numerical integral of the demand curve between c0 − ε and c0 against ε²/2.

Honest limits

  • Only one inventor, one product, no follow-on invention, no patent races, no licensing, no trade secrets, certain demand and a quadratic R&D cost chosen for tractability.
  • Valid for non-drastic innovation (ε ≤ a). The sliders keep κ ≥ f(T) so this holds. All dollar values are illustrative.
  • Real patent terms (about 20 years in many systems) come from legislation and treaties, not from this formula; the model only shows why an interior optimum exists.

2D view: top-down / diagram rendering with sprites rendered in Blender (Cycles) from the 3D page's model parts; all numbers come from the same solver as the 3D page.

Made with 🟠 Blender