Enabling A Just Transition to Net-Zero Emissions: Distributional Impacts of Carbon Pricing in Taiwan

Taiwan is building a carbon-pricing system — the Climate Change Response Act passed in 2023 authorizes a carbon fee, and rates are now being set. Carbon pricing is widely regarded as the most efficient tool for cutting emissions, but it carries a fairness risk: the cost can shift from industry onto households, and often lands hardest on those least able to pay. This completed NSTC project (2023–2024) asked a deceptively simple question for Taiwan — who actually bears the burden? — and used the answer to design a fairer policy. It pairs the E3 Center's engineering-and-modeling approach with an environmental-sociology co-investigator, so the analysis speaks to both efficiency and equity.

Methodologically, carbon pricing reaches households through two channels. Direct emissions come from a household's own electricity and transport-fuel use; indirect emissions are embedded in the goods and services it buys. The project quantifies both by coupling an environmentally-extended input-output (EEIO) model — built on Taiwan's 63-sector Input-Output Table — with the 2022 Survey of Family Income and Expenditure, which records income, spending, and rich socio-demographic detail for roughly 16,500 households. Multiplying each household's total emissions by a benchmark carbon price (US$40/tonne) gives its carbon cost, and expressing that as a share of spending gives its carbon burden. Quantile regression then isolates how each characteristic shifts the burden across the whole distribution — not just at the average — so the high-burden tail can be characterized directly.

Three results reframe the debate. First, Taiwan's carbon pricing is only mildly regressive — and the choice of yardstick matters: clearly regressive on an income basis (Suits Index −0.12) but only slightly so against expenditure (−0.011). Second, and more important, income alone cannot identify the vulnerable: differences in burden within income groups run 3–5× larger than the gap between rich and poor deciles, so socio-demographic profiling is essential. Third, the burden is driven by gasoline, not electricity, and concentrates in three household types — larger households; car- and scooter-dependent households outside the well-served metropolitan north; and those in detached, low-rise, large-footprint homes.

Finally, the project shows the regressivity is fixable, and affordably so. Recycling revenue as population-weighted lump-sum rebates turns the system progressive (Suits Index +0.30); achieving a distribution-neutral outcome needs only about 37% of revenue on a per-household basis, and a rebate targeted at the lowest earners restores neutrality using just 13.4% — freeing the remainder for other decarbonization spending. The result is a compact, transferable framework that lets policymakers see who is exposed and choose a compensation design that keeps Taiwan's net-zero transition a just one.