# R&D investment threshold: is it worth funding?

> Free R&D investment calculator: find the minimum benefit an R&D programme must deliver to justify its cost, using a published real-options method.

URL: https://www.buildresil.com/tools/rd-investment-threshold
Last updated: 2026-11-17

Enter what the R&D will cost and what it would take to launch. See the minimum benefit that makes the spend worthwhile.

## How much must R&D deliver to be worth funding?

**This R&D investment calculator shows the minimum value of benefits an R&D programme must be expected to unlock for its cost to be justified.** It treats R&D as what it is: paying now for the right, but not the obligation, to launch later if the results and the market are good. The method was published by Dr. David Goldenberg and Dr. Jonathan D. Linton, a co-founder of Building Resilience, who used it to test whether nuclear fusion research is worth its cost [1].

It answers a question boards, investors and public funders ask: *what would this have to be worth for the spend to make sense?* That is often easier to judge than a single forecast of future profits.

## Try the calculator

<div class="tool wide">
<form class="form" data-rd-tool novalidate aria-label="R&D investment threshold calculator">
  <div class="tool-presets tool-js"><button type="button" class="btn-ghost" data-preset="fusion">Nuclear fusion example</button><button type="button" class="btn-ghost" data-preset="product">New product example</button></div>
  <label>R&D spend, valued today<input name="spend" inputmode="decimal" value="70"><small>Everything you expect to spend on the R&D, in today's money.</small></label>
  <label>Cost to launch, if you go ahead<input name="launch" inputmode="decimal" value="43813"><small>What it would cost to build, launch or roll out at the decision date. Use the same currency unit throughout.</small></label>
  <label>Months until the launch decision<input name="months" inputmode="decimal" value="576"></label>
  <label>Risk-free rate, % per year<input name="rate" inputmode="decimal" value="3.5"><small>The yield on a government bond with the same term as the time to the launch decision.</small></label>
  <label>Uncertainty (sigma), % per year<input name="vol" inputmode="decimal" value="18"><small>How much the value of the benefits could swing each year. We favour the sigma of stocks listed in the same industry.</small></label>
  <label>Your estimate of the benefits, valued today (optional)<input name="benefits" inputmode="decimal" value=""><small>Add it to see whether your estimate clears the threshold.</small></label>
</form>
<div>
<div class="tool-result" data-rd-result aria-live="polite">
<p class="eyebrow">Minimum value of benefits needed</p>
<p class="tool-figure">1,075.2</p>
<p>With the nuclear fusion inputs (in US$ billions), today's value of the benefits must be at least <strong>1,075.2</strong>, about <strong>15.4 times</strong> the R&D spend.</p>
</div>
<div data-rd-sensitivity></div>
<noscript><p>The calculator needs JavaScript. The worked example below shows the same calculation.</p></noscript>
</div>
</div>

The calculator runs in your browser. Nothing you enter is sent or stored.

## What do the inputs mean?

| Input | What to enter | Fusion example |
|---|---|---|
| R&D spend | Today's value of all planned R&D spending | US$70 billion (2002 dollars) |
| Cost to launch | The cost of putting the result into use, at the decision date | US$43.8 trillion to build fusion power plants (2050 dollars) |
| Months until the decision | When you would decide whether to launch | 576 months (48 years, 2002 to 2050) |
| Risk-free rate | The yield on a government bond with the same term as the time to the decision, % per year | 3.5% (the historical average Treasury bill rate used in the study) |
| Uncertainty (sigma) | How much the value of the benefits could swing per year: the sigma of stocks in the same industry | 18% (S&P electric utility index) |

The result is the **minimum value of benefits**, valued today, that the launch must be expected to bring. Compare it with your own realistic estimate. If your estimate is higher, the R&D is worth funding under these assumptions; if it is lower, it is not.

### How should you choose sigma?

Sigma is the measure of uncertainty in options theory. The higher the sigma, the greater the risk and the greater the potential profit. Because the right to launch is worth more when the upside is larger, a higher sigma lowers the minimum benefit needed, and a lower sigma raises it.

We favour using the sigma of stocks listed in the same industry as the benefits. For nuclear fusion, that meant electrical utilities. Utility stocks have a low sigma, so the estimate is more conservative: it sets a higher bar for the R&D to clear. If the case is justified with a low sigma, it is justified with more room to spare.

## Worked example: is nuclear fusion research worth it?

In 2006, Goldenberg and Linton applied this method to the planned public spending on nuclear fusion [1]. With the inputs above, fusion must be expected to save at least **US$1.075 trillion** in operating costs, valued in 2002, to justify the planned R&D. Spread over projected world energy use, that is about **2.4¢ per kWh** (2002 dollars) cheaper than other ways of generating power.

They then tested the result across 14 scenarios for plant uptime, adoption rate and construction costs. The required saving ranged from **0.6¢ to 2.5¢ per kWh**, against an estimated fusion operating cost of about 1¢ per kWh. Their conclusion: the planned fusion research was financially justified across a wide range of reasonable assumptions.

| Scenario (plants introduced over 50 years, 80% uptime) | Cost to launch (US$ billion) | Minimum benefits (US$ billion) | Saving needed (¢/kWh, 2002) |
|---|---|---|---|
| Base case: 0.4% adoption a year, costs rise with inflation | 43,813 | 1,075.2 | 2.4 |
| Faster adoption: 1% a year | 109,535 | 1,868.5 | 1.7 |
| Plant costs stay flat | 7,334 | 400.2 | 0.9 |
| Faster adoption, flat costs | 18,335 | 653.1 | 0.6 |

*Source: Goldenberg and Linton (2006), Table 1. The calculator reproduces each of these results.*

## When is this useful?

- **Funding a technology programme:** setting a bar the programme must clear, before arguing over a single forecast.
- **Pilots and new products:** deciding whether a pilot is worth running when the full launch is far off and uncertain.
- **Public and long-term R&D:** energy, infrastructure and research programmes where benefits arrive decades later.
- **Stage gates:** checking at each review whether the case still clears the bar.

For decisions with several stages, or a choice between competing technologies, the same idea extends to more detailed models. See [decision support under uncertainty](/decision-support).

## What this model leaves out

- It assumes one launch decision at a fixed date, not a series of stages.
- It assumes the value of the benefits moves smoothly with a constant sigma, and that the risk-free rate stays constant.
- It values the right to launch. It does not include benefits that are hard to price, such as lower emissions or energy security, which the fusion study treated as extra reasons in favour.
- The result is only as good as the inputs. Test a range, not a single number: the table under the result shows how the minimum changes with sigma.

## How the model works

The calculator uses the Black–Scholes option model [2], run in reverse. Normally the model takes the value of an asset and gives the value of an option on it. Here the R&D spend is taken as the price of the option, and the model solves for the value of the benefits that makes that price fair. This is the approach of Goldenberg and Linton (2006) [1]. The calculation has been checked against every row of their published table.

## Want to use this on a real decision?

We help organizations frame decisions like this, estimate the inputs and test them across many futures. [Talk to us](/contact) or read more about [decision support under uncertainty](/decision-support).

## References

1. Goldenberg, D., & Linton, J. (2006). Nuclear fusion R&D. *Energy Risk*, January 2006, 2–5.
2. Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. *Journal of Political Economy*, 81(3), 637–654. [doi:10.1086/260062](https://doi.org/10.1086/260062)

## Questions people ask

### How do I know if an R&D project is worth funding?

Compare the R&D spend with the minimum value of benefits it must unlock. R&D buys the right, but not the obligation, to launch later, so it should be valued as an option rather than with a standard net present value. If your realistic estimate of the benefits, valued today, is above the minimum, the spend is justified under your assumptions.

### Why not just use net present value (NPV) for R&D?

NPV assumes you will go ahead whatever happens. R&D is different: you launch only if the results and the market turn out well, and stop otherwise. That flexibility has value, and NPV leaves it out, so it tends to undervalue R&D and long-term innovation.

### What is sigma, and how should I estimate it?

Sigma is the measure of uncertainty in options theory: how much the value of the benefits could swing each year. The higher the sigma, the greater the risk and the greater the potential profit. We favour using the sigma of stocks listed in the same industry. For nuclear fusion that meant electric utilities, whose stocks have a low sigma, which makes the estimate more conservative.

### Is this calculator investment advice?

No. It shows what your own assumptions imply, using a published method. It is not a complete appraisal or a funding recommendation. For a real decision, the inputs need careful estimation and the results should be tested across a range of scenarios.


---
Building Resilience · 226-386-4343 · SupplyChain@buildresil.com · https://www.buildresil.com
