Greening our economy isn’t just about what we produce—it’s also about what we consume. Sending smokestack industries off to distant shores in search of cheap labor markets to make the things we consume may lessen the carbon footprint of our own economies, but it sure doesn’t do much for the global footprint. And since there are no borders in the atmosphere, it’s really the global imprint that counts.

Take steel, for example. The mass migration of North American steel production to China certainly hasn’t lessened the industry’s global environmental footprint. If all steel plants emitted equally, it wouldn’t matter where they were located. But when it comes to full-cycle carbon emissions, all steel plants are decidedly not equal.

While China may boast some of the world’s newest and most energy-efficient steel mills, it houses a whole lot more plants whose energy-efficiency is a fraction of that of steel plants that have been shuttered on this side of the Pacific. Moreover, virtually all steel mills in China are serviced by coal-fired power—which is far and away the dirtiest form of generating power. (Coal is 80 per cent of China’s power, compared to 50 per cent for the US and only 12 per cent for Canada). Combine the gap in energy efficiency between Chinese and North American steel mills with the greater carbon intensity of Chinese power and, on average, the plants where world steel production has migrated emit one third more carbon than the plants it left.

So it’s not just about trading clean air in Pittsburgh for foul air in Beijing. The global steel industry would have emitted significantly less than it does had not so much of its production moved to China’s blast furnaces. In fact, this is true not only of the global steel industry, but of most of the heavy manufacturing that has crossed the ocean in pursuit of cheap labor.

You get some sense of the scale of the emissions trail that has accompanied this shift in economic geography when you consider that the emissions coming from China’s export sector alone are greater than the total emissions from any other economy other than the US’s.

Of course, much of that trail could be wiped away by pricing carbon emissions and applying that price to imports via a carbon tariff. Competitiveness in the steel industry and a whole range of other industries wouldn’t simply be a single-variable function of wage rates anymore, but a more complex function that includes energy efficiency and carbon intensity. And the higher the price of energy and carbon, the greater their impact on steel production costs. Instead of exporting jobs away, raising the carbon bar would bring more than a few home.

The knock against tariffs, of course, is that they distort trade. But trade is even more distorted when environmental costs are allowed to simply go unpaid.