When Fossil Industries Retreat, What Happens to the Wage Structures They Leave Behind?

Climate policy is designed to cut emissions, accelerate clean investment and move economies away from fossil fuels, but decarbonization is also dismantling a production model that, in many countries, shaped wages, industrial employment and regional prosperity for decades. A new cross-country study suggests those distributional effects may be more deeply rooted in economic structure than in carbon itself.

Published in Economies, "The Distributional Footprint of Carbon: Economic Development, Emissions, and Income Inequality Across 79 Economies (1995–2022)" examines whether the carbon-intensive character of production is linked to how income is distributed within countries.

Authored by Sadagat Ahmadova, Jeyhun Mahmudov, Rahiba Abdulhasanova, Flora Alasgarova and Konul Buyuker, the research reverses the direction of a familiar climate-economy question. Instead of asking whether inequality influences emissions, it investigates whether production systems associated with high emissions also leave a measurable imprint on inequality.

Carbon-intensive growth left a distributional legacy

The researchers find a striking pattern: within countries, periods of higher per-capita CO₂ emissions are associated with lower measured income inequality after controlling for income levels and common shocks. In the baseline model, a doubling of per-capita emissions is associated with roughly a three-point decline in the Gini index. The relationship also survives when the analysis uses consumption-based emissions or primary energy consumption instead of production-based emissions.

The finding does not mean pollution promotes equality, and the authors explicitly reject such an interpretation. Instead, emissions appear to capture characteristics of production systems that have historically supported relatively productive jobs for workers without university education, large-scale industrial employment, collective bargaining and regional supply chains. Mining, heavy manufacturing, utilities, construction and freight can create a thicker middle of the earnings distribution when wages and productivity are relatively high across a broad workforce.

Such structures have distributional consequences beyond the factory gate. Large industrial plants often anchor local economies, support suppliers and service businesses, and create bargaining institutions that compress wage differences. When those activities contract, workers may shift into lower-productivity services, union coverage can weaken and losses can become geographically concentrated, raising the risk that inequality widens even if aggregate economic output eventually recovers.

The research, hence, adds a labour-market dimension to the just-transition debate. Climate policy is not simply replacing one energy source with another; it can alter the institutional architecture through which income is generated and distributed. The challenge is particularly acute when decarbonization affects regions where industrial employment, local tax bases and relatively well-paid non-graduate jobs depend heavily on fossil-energy-intensive production.

The strongest signal comes from energy-intensive production, not carbon itself

The researchers test whether the relationship is genuinely specific to carbon by separating the carbon intensity of the energy mix from the total level of energy use. If carbon itself were driving the distributional relationship, economies using more carbon-intensive fuels should display a stronger inequality effect even after overall energy consumption is held constant.

The evidence does not support that stronger hypothesis. Energy use per person remains strongly associated with lower inequality, while the carbon intensity of the fuel mix carries only a weak and non-robust relationship. The fuel-level results also run against the study's original expectation: oil shows the largest estimated association, gas an intermediate one and coal the smallest, despite coal being the most carbon-intensive fuel and historically linked to large, fixed industrial workplaces.

The authors reinterpret per-capita emissions primarily as an indicator of energy- and capital-intensive production, rather than evidence that carbon content itself has a unique distributional effect. This distinction considerably strengthens the analytical credibility of the paper because the researchers do not stretch a failed hypothesis into a supportive conclusion. Their narrower claim is that economies organized around large-scale fossil-energy use appear to have developed labour-market structures associated with more compressed income distributions.

Decarbonization still targets sectors according to their emissions exposure through carbon pricing, standards, regulation and border measures. Industries historically associated with high energy use may therefore face restructuring because of carbon policy even if their distributional characteristics arose from broader production technologies rather than from the carbon content of fuel itself.

Richer economies may face sharper adjustment

Development level substantially changes the strength of the relationship. In the higher-income half of the sample, the estimated association between emissions and inequality is roughly twice as large as in the lower-income half. The study reports coefficients of −4.363 for higher-income economies and −2.094 for lower-income economies, while the interaction analysis shows the relationship becoming stronger as income rises.

The pattern fits the institutional history of mature industrial economies. Carbon-intensive sectors in many richer countries developed alongside comparatively high wages, organized labour, established bargaining systems and regions built around manufacturing, mining or energy production. In poorer economies, informality, early-stage industrialization and weaker labour institutions can dilute the connection between industrial structure and household income distribution.

The study also finds that the association is stronger for market-income inequality than for disposable-income inequality. Since market income is measured before taxes and transfers, the result suggests the relationship is rooted more deeply in how earnings are generated than in how governments redistribute them afterward. Fiscal policy appears to cushion part of the underlying movement, but it does not fully explain the pattern.

Several limitations temper any attempt to generalize the findings globally. Europe and Central Asia account for 44 of the 79 economies, while the sample includes only one South Asian economy and two Sub-Saharan African economies under the World Bank regional classification. Inequality data also combine different survey concepts, and the proposed mechanisms involving wages, unions and regional employment are not directly tested with sector-level labour-market data.

A just transition has to replace institutions, not only energy

Retraining programmes, regional development funds, income support and recycling of climate-policy revenues become more than compensatory measures when industrial decline can weaken wage-setting institutions and entire local labour markets. The research points toward a broader conception of transition policy in which employment quality and regional economic resilience sit alongside emissions targets.

For governments and multilateral institutions, the study suggests that counting green jobs will not be enough. Policymakers also need to ask whether replacement industries offer comparable productivity, wage progression, bargaining power and geographic accessibility. A clean-energy investment boom can expand output while still leaving former industrial regions behind if new opportunities emerge elsewhere or rely on different skills, employment models and ownership structures.

The evidence is also more consistent with a slow structural relationship than a short-term feedback loop. Five-year dynamic estimates retain a negative emissions–inequality association, while annual causality tests find no significant predictive relationship in either direction. Structural change in production, employment and bargaining institutions appears to unfold over years rather than respond mechanically to annual movements in emissions.

So, a credible just-transition strategy has to answer a more demanding question than how many green jobs will be created. It has to ask where those jobs will be located, who can realistically move into them, whether they offer comparable earnings and security, and whether regions dependent on older industries can build viable economic bases before existing ones erode.