Decompose CO2 emissions into four driving factors — population, affluence, energy intensity and carbon intensity. Compute emissions, solve for a missing factor, or compare two scenarios.
CO₂ = P × (GDP/P) × (E/GDP) × (CO₂/E)
Honesty note: the Kaya identity is an accounting decomposition, not a causal or predictive model. It always holds exactly by definition; it cannot tell you which factor causes emissions. Illustrative presets are labeled as such.
Compute CO₂ from the four factors
Enter the four Kaya factors. Try the presets: a world calibrated to the Global Carbon Budget 2024, or a small illustrative city.
Total CO₂— Mt
Total CO₂ (gigatonnes)—
Per capita—
Implied total energy—
Factor chain—
World 2024 preset: total calibrated to the Global Carbon Budget 2024 fossil total of 37.4 GtCO₂; the four factor inputs are illustrative, not measured statistics.
Solve for a missing factor
Set an emissions target and the three factors you know; the calculator solves the fourth exactly by rearranging the identity.
Required factor—
Compare two scenarios
Enter a base scenario and an alternative. The change in emissions is split across the four factors with a log-share allocation whose parts always sum exactly to the total change.
Base emissions—
Alternative emissions—
Change—
Effect per factor = change × ln(factor ratio) / ln(total ratio). The four effects sum exactly to the total change by construction.
How the math works
CO₂ (Mt) = P × GDPpc × E/GDP × CO₂/E ÷ 10¹²
With population in persons, GDP per capita in USD, energy intensity in megajoules per USD and carbon intensity in grams of CO₂ per megajoule, the product is grams of CO₂; dividing by 10¹² converts to megatonnes. Solving for a missing factor divides the target by the product of the other three (with unit conversions). Scenario decomposition uses a log-share allocation, effecti = ΔCO₂ × ln(ri)/ln(R), where ri is each factor's alternative/base ratio and R is their product — the parts sum exactly to the total change.
Sources. World totals: Global Carbon Budget 2024 (Global Carbon Project, Earth System Science Data) — 37.4 Gt of fossil CO₂, 41.6 Gt including land-use change — via Carbon Brief's analysis. The identity itself: Kaya, Y. (1989). Preset factor values are illustrative and chosen so their product matches the published total.
Frequently asked questions
What is the Kaya identity?
The Kaya identity (Yoichi Kaya, 1989) is an accounting identity that breaks total CO2 emissions into four driving factors: CO2 = Population x (GDP / Population) x (Energy / GDP) x (CO2 / Energy). It is a decomposition, not a causal model: it always holds exactly, but it does not prove that any one factor causes emissions.
What are the four factors of the Kaya identity, and what units does this calculator use?
Population (millions of people), GDP per capita (USD per person), energy intensity of GDP (MJ per USD), and carbon intensity of energy (g CO2 per MJ). Results are shown in megatonnes (Mt) and gigatonnes (Gt) of CO2.
Can you show a worked example of the Kaya identity calculation?
Yes. Take 100 million people x $10,000 GDP per capita x 5 MJ per USD x 60 g CO2 per MJ. Multiply: 100,000,000 x 10,000 x 5 x 60 = 3 x 10^14 grams of CO2, which equals 300 megatonnes (Mt) of CO2, or 0.3 Gt. This calculator reproduces that exact result in Mode 1.
How do I solve for a missing factor, for example the carbon intensity needed to hit an emissions target?
Use Mode 2: enter your emissions target and the three factors you know, then pick the factor to solve for. Example: with a target of 1,000 Mt CO2, 100 million people, $20,000 GDP per capita, and 6 MJ/USD energy intensity, the required carbon intensity is 83.33 g CO2 per MJ. The calculator rearranges the identity algebraically, so the answer is exact.
How does the scenario comparison decompose an emissions change into per-factor effects?
Mode 3 computes CO2 for a base scenario and an alternative scenario, then splits the change between the four factors with a log-share allocation: each factor's effect equals the total change times ln(ratio_i) / ln(total ratio). The four effects always add up exactly to the total change. Example: base 36,000 Mt vs alternative 35,645.4 Mt gives a change of -354.6 Mt (-0.99%), split as population +884.55, GDP per capita +4,219.27, energy intensity -2,986.93, and carbon intensity -2,471.49 Mt.
Is the Kaya identity a predictive model of emissions?
No. It is an accounting identity, so it is true by definition and cannot be falsified or used alone to forecast. It is useful for structuring scenarios: you decide how each factor might evolve, and the identity tells you what emissions those assumptions imply. Treating any single factor as the 'cause' of emissions is a misuse.
What data does the World 2024 preset use?
The preset's total is calibrated to the Global Carbon Budget 2024: 37.4 Gt of fossil CO2 (41.6 Gt including land-use change), reported by the Global Carbon Project and summarized by Carbon Brief. The individual factor inputs (population 8.2 billion, GDP per capita $13,500, energy intensity 4.35 MJ/USD, carbon intensity ~77.67 g/MJ) are illustrative values chosen so their product equals the published 37.4 Gt total; they are not separately measured statistics. Per-capita result: about 4.56 tonnes of CO2 per person.
Why can GDP grow while emissions fall?
Because the identity is multiplicative. If energy intensity (MJ per USD) and carbon intensity (g CO2 per MJ) fall faster than population and GDP per capita grow, total CO2 declines. This is what economists call decoupling: growth driven by efficiency and clean energy rather than by burning more fossil fuel per dollar of output.
What are the limitations of the Kaya identity?
Three main ones. First, it says nothing about causation or policy levers; it only reorganizes the arithmetic. Second, the four factors interact in reality (for example, cheap clean energy changes GDP), which the identity cannot represent. Third, results are only as good as the inputs: mixing nominal and PPP GDP, or fossil-only vs all-source CO2, changes the answer.
How is this different from a carbon footprint calculator?
A carbon footprint calculator estimates emissions from specific activities, like diet or flights. The Kaya identity works at the economy level: it decomposes a country's or the world's total emissions into four macro factors. Try our food footprint tools, meat-footprint and vegan-footprint, for the activity-level view.
Does this calculator send my data anywhere?
No. Everything runs locally in your browser in a single HTML file; there are no network requests, no cookies, and no tracking. Your inputs never leave your device.
Can I use the results for policy or investment decisions?
The calculator is an educational tool. Its math is exact given the inputs, but real decisions also need verified statistics, uncertainty ranges, and domain expertise. The illustrative presets are not substitutes for official data from sources like the Global Carbon Budget, the IEA, or national statistics offices.
We use cookies and third-party services (including Google AdSense) to personalize content, analyze traffic, and serve relevant ads. By clicking "Accept", you consent to our use of cookies as described in our Privacy Policy.
Report an issue
2 taps, no typing needed — diagnostics are sent automatically.