VoxEU Column Environment

Environmental Engel curves

Rich countries pollute less partly because people in richer countries consume a less pollution-intensive bundle of goods. This column investigates whether this results from consumer preferences or economy-wide changes. Within a country, the environmental Engel curve is concave – meaning that richer households, while polluting more, consume a less pollution-intensive bundle. Over time, this accounts for half of the decrease in rich household pollution, with the remainder being due to price changes and environmental regulations.

Over the past 30 years US GDP per capita has nearly doubled while pollution emitted by US manufacturers has declined. Ozone emissions are down 33%, nitrogen dioxide 60%, sulfur dioxide 81%, and carbon monoxide 84%.1 Some analysts point to this cleanup as evidence that economic growth alone will automatically improve the environment. For example John Tierney wrote in The New York Times that “the richer everyone gets, the greener the planet will be in the long run.” Others caution that environmental improvement does not necessarily follow from income growth, but depends on environmental policies and institutions (Yandle et al. 2004).

Claims that richer countries pollute less typically cite the seminal paper by Grossman and Krueger (1993) demonstrating that many pollutants appear to increase with national income in poor countries, but decrease with national income in rich countries. A graph of this relationship – with pollution on the vertical axis and income on the horizontal – makes an inverted-U shape, and has come to be called an environmental Kuznets curve (EKC). But EKCs are just correlations. They don’t demonstrate that income growth causes environmental improvement – at least not directly. Richer countries could pollute less because they enforce stricter environmental regulations or offshore the manufacture of pollution-intensive goods. Neither of these would represent a direct or automatic effect of income growth on global pollution.

To analyse the relationship between national income and pollution, economists decompose it into three components:

  • Increases in the overall size of the economy (scale),
  • Changes to the pollution intensity of production (technique), and
  • Changes in the overall mix of activities (composition).2

Several recent papers have shown that most of the cleanup of manufacturing in developed countries has come from changes in production techniques. But of the remaining change in composition, a significant fraction comes from people in richer countries choosing to consume a less pollution-intensive mix of goods and services.3 This leaves open the question of whether richer people choose less polluting goods, or richer countries pass regulations making polluting goods more expensive.

In Levinson and O’Brien (2015) we use Environmental Engel Curves (EECs) – or income expansion paths – to separate these two parts of the composition change in which residents of richer countries choose cleaner goods:

  • The direct and perhaps automatic portion coming from richer people’s preferences, and
  • The indirect portion due to economy-wide changes in richer countries.

Unlike EKCs, EECs are meant to be causal. They depict structural relationships between people’s incomes and the goods and services they choose, holding all else equal.4

Figure 1 shows a set of EECs estimated for one representative local air pollutant – particulate matter less than ten microns (PM10). We begin by matching categories of household consumption in the Consumer Expenditure Survey (CEX) to emissions intensities in the 1997 Trade and Environmental Assessment Model (TEAM).5 Then for each household in the CEX we calculate the total amount of PM10 emitted while producing the goods and services that the household consumes.6 We then separate the CEX households into 50 income groups and plot the average level of pollution associated with producing those households’ goods and services. The top curve in Figure 1 uses the 1984 CEX; the bottom curve uses the same 50 household income groups and production emissions intensities applied to the 2002 CEX.

Figure 1. Environmental Engel curves for particulate matter, 1984 and 2002

Notes: Income and consumption are adjusted for inflation using the core CPI. Food and energy consumption are adjusted separately using the corresponding CPIs. Pollution per household in both years is estimated using 1997 pollution intensity of production data.
Sources: Calculations based on the NBER extracts of the Consumer Expenditure Survey (www.nber.org/data/ces_cbo.html) and the Trade and Environmental Assessment Model (Abt. Associates, Inc.).

Three phenomena are apparent in Figure 1.

  • First, unsurprisingly, richer households are responsible for more pollution.
  • Second, EECs are concave, meaning that richer households consume less pollution-intensive mixes of goods and services.
  • Third, the EECs in Figure 1 shift down over time. Households represented in 2002 are responsible for less pollution than their 1984 counterparts with similar real incomes. This downward shift does not result from improvements in technology or abatement because we have fixed the pollution intensity of production for both years. Instead, the shift reflects a change in consumption due to some combination of changing prices, regulations, or other economy-wide effects.

EECs like the ones in Figure 1 can be used to disentangle the two parts of income growth’s effect on the pollution-intensity of household consumption:

  • The automatic part from richer people’s preferences and
  • The indirect part from richer countries’ prices or regulations.

Movement along concave EECs represents pollution changes due to innate preferences by richer households for less pollution-intensive mix of goods and services, holding all else equal. The downward shift in the EECs over time represents other society-wide changes such as prices or regulations. According to our estimates, these two effects each explain about half of the reduction in U.S. pollution that comes from shifts in consumption towards less pollution-intensive goods and services. Half of the change related to shifts in consumption comes from movements along concave EECs and half comes from shifting those EECs down over time.

So while there is evidence for some direct, automatic effect of income growth on pollution, that effect comprises only half of the household consumption composition change. And decomposing income growth into movements along and shifts in the EECs represents just one aspect of the environmental consequences of economic growth. The largest portion of the cleanup in the US has come from changes in production technology. Nevertheless, isolating the consumption-related compositional changes in pollution suggests that household-level composition changes have more than offset the increased pollution from growing household incomes, and this is due in roughly equal parts to a direct effect of income growth via consumer preferences and indirect economy-wide changes such as prices and environmental regulations.


Abt Associates Inc. (2009), “Trade and Environmental Assessment Model: Model Description”, Prepared for the U.S. Environmental Protection Agency National Center for Environmental Economics. Cambridge, MA.

Brunel, C (2014), “Pollution Offshoring and Emission Reductions in EU and US Manufacturing”, Working paper. Washington, DC: Georgetown University.

Copeland, B, and M S Taylor (2005), Trade and the Environment: Theory and Evidence. Princeton, NJ: Princeton University Press.

Grossman, G, and A Krueger (1993), “Environmental Impacts of a North American Free Trade Agreement”, in Garber, P M (ed.), The Mexico-U.S. Free Trade Agreement, Cambridge, MA: MIT Press.

Leontief, W (1970), “Environmental Repercussions and the Economic Structure: An Input-Output Approach”, Review of Economics and Statistics 52(3): 262–271.

Levinson, A (2009), “Technology, International Trade, and Pollution from U.S. Manufacturing”, The American Economic Review 99(5): 2177–92.

Levinson, A (2015), “A Direct Estimate of the Technique Effect: Changes in the Pollution Intensity of US Manufacturing, 1990–2008”, Journal of the Association of Environmental and Resource Economists 2(1):43-56.

Levinson, A and James O’Brien (2015), “Environmental Engel Curves”, Working paper 20914. Cambridge, MA: National Bureau of Economic Research.

Shapiro, J and R Walker (2015), “Why is Pollution from U.S. Manufacturing Declining? The Roles of Trade, Regulation, Productivity, and Preferences”, Working paper 20879. Cambridge, MA: National Bureau of Economic Research.

Tierney, J (2015), “Use Energy, Get Rich and Save the Planet”, The New York Times, 20 April 2009. Accessed 6 February 2015.

Yandle, B, M Bhattarai, and M Vijayaraghavan (2004), “Environmental Kuznets Curves: A Review of Findings, Methods, and Policy Implications”, Research Study 02-1 update. Property and Environment Research Center.


1 See http://www.epa.gov/airtrends/index.html for the pollution data. Real GDP per capita and personal consumption expenditure are from the Bureau of Economic Analysis via the Federal Reserve Bank of St. Louis: http://research.stlouisfed.org/fred2/.

2 Grossman and Krueger (1993), Copeland and Taylor (2005).

3 See Levinson (2009, 2015), Brunel (2014), and Shapiro and Walker (2015).

4 Engel curves are named for Ernst Engel, a German economist writing in the mid-1800s who studied the degree to which household food expenditures increase with income. They have since been applied to many different categories of consumption and form the basis for ‘equivalence scales’ that are used to determine eligibility for means-tested entitlements, such as food stamps and Medicaid.

5 Abt Associates (2009).

6 We use a Leontief (1970) input-output analysis – along with benchmark IO tables published by the US Bureau of Economic Analysis – to include pollution from manufacturing the products consumed, inputs to those products, and inputs to those inputs ad infinitum up the supply chain.

1,785 Reads