Free traders are up in arms following the decision of the United States to impose import tariffs to protect its sickly steel industry. They point to the retaliatory import restrictions imposed by the European Union as a sign that the tariffs risk sparking a trade war. The hardest-hit nations, they suggest, will inevitably be developing ones. And they argue that tariffs remove the incentive for uncompetitive operators to reform, thus prolonging rather than solving the industry’s problems.

These arguments are sound, but there is little chance that the US government, or any other, would allow companies to fold on the scale required to make the steel industry more efficient. For it has been in trouble for decades, and despite quota restrictions, tariffs, and billions of dollars in government subsidies—usually justified as an interim measure to revive the industry—little has changed. Steel companies around the world still have the capacity to produce, each year, at least 200 million more tons of coil, plate, wire, rods, and other products than customers want, with disastrous effects on prices.

That is why we suggest a new way forward: a multilateral steel agreement to cut global capacity by some 15 percent, to about 900 million tons, within three to five years. Yes, our approach would require government-guaranteed loans, but steel producers would also share the pain. And, crucially, our plan would ensure the destruction of obsolete and uneconomic capacity—the only way to reach a long-term solution for the industry. Most rescue plans, including the latest capacity reduction proposals from the Organisation for Economic Co-operation and Development (OECD), fail to ensure that capacity will fall permanently. Instead, managers tend to mothball plants, only to reopen them once prices pick up again.

Here is how our plan would work: If market forces are doing their job, excess capacity dwindles as weaker players are bought by stronger, more competitive ones, which keep valuable assets and dispose of the rest. This kind of restructuring doesn’t occur in the steel industry, because of so-called legacy costs. The biggest of them in the United States—which among steel-producing nations has some of the highest costs—take the form of ruinously expensive health care and pension benefits, for employees and retirees, that were negotiated by steel unions active in the significantly downsized integrated mills. Because no one can afford to inherit these unfunded costs, mergers don’t happen.

The US government is afraid to let inefficient producers fold because it might have to pay for their health care and pension benefits

So why not let inefficient producers fold? Because the government fears it might be forced to foot the bill to cover the promised benefits. US officials estimate the legacy costs of the 30 or so US steel mills currently under bankruptcy protection at about $21 billion. Luckily for the government, US bankruptcy laws allow the producers to keep operating and meeting these obligations while being protected from creditors. But this provision also means that excess capacity is rarely destroyed. Meanwhile, in Europe more than $1 billion of environmental cleanup liabilities associated with integrated plants, as well as regional social pressures, also help prevent the industry from rationalizing more quickly.

Developing countries too worry about the legacy costs of consolidating the steel industry (and other industries as well). Many Central and Eastern European plants, using obsolete technology, keep tens of thousands of workers on the payroll. Modern plants could make do with 80 to 90 percent fewer of them. The restructuring of Poland’s steel industry, for example, has lagged because of concerns about the cost and social implications of laying off so many workers.

To break the deadlock, we propose the formation of special companies whose purpose would be to destroy capacity and jump-start a restructuring process. Steelmakers that chose to participate in the scheme would own these companies. They would hand over their obsolete or uneconomic assets for destruction, and in return the companies would assume some, though not all, of the legacy costs tied to these assets. (Some of the costs would rest with the steelmakers as a penalty for not having managed their assets better.) The steelmakers would also hand over viable assets that they couldn’t afford to run profitably or invest in but that the special-purpose companies could modernize and improve operationally before selling. Loans guaranteed by governments would provide the initial funding.

In the United States and Western Europe, the two regions we studied in depth, some $12 billion would be needed in all—65 to 75 percent for the former, the rest for the latter. This sum would be supplemented by an industry-wide surcharge of $2 to $5 on each ton of steel shipped and an additional surcharge of $4 to $7 a ton for participating steelmakers. If only about half of them in each region participated in the scheme, these sums would cover the assumed legacy costs as well as the cost of destroying plants and of restructuring.

Because legacy costs and the amount of uneconomic or obsolete capacity differ from region to region, the special-purpose companies should be established regionally. Ideally, the first participants would be in Western Europe and North America, which are big importers, as well as either Japan, on the one hand, or Russia, Ukraine, and Kazakhstan, on the other—big exporters with similar capacity problems. Three big steel regions would suffice to get the process going, and their participation would encourage others to set up special-purpose companies.

Uncompetitive steelmakers would have great incentive to participate. Carrots would include the reduction of their legacy costs and the ability to sell some of their assets. The stick—and this is crucial—would be the abolition of all other government support for the industry. Failing companies could no longer expect to survive on government handouts—direct or indirect subsidies, high tariffs, and all sorts of nontariff barriers—which often amount to more than $30 a ton in all.

Would such a scheme serve the interests of the more competitive companies and countries? We believe so. Developing countries that currently enjoy the competitive advantage of low labor costs would be able to export into higher-cost countries, without the prospect that trade sanctions might be imposed. And these competitive countries would also have an opportunity to shut down obsolete capacity. Moreover, cost-competitive companies, wherever they might be, would have the best position to begin buying up the assets that became available, thereby positioning themselves as potential global leaders.

Without doubt, many problems would have to be overcome in putting such a plan into action. To give one example, to forestall free riders—countries that take advantage of stabilized prices without reducing their own capacity—the scheme would require rigorous enforcement by the World Trade Organization. For starters, the WTO would have to monitor compliance. It would also have to allow the imposition of trade sanctions on WTO members that chose not to participate and still subsidized steel. And it would have to impose tariffs or quotas on non-WTO members that declined to make efforts to reduce uneconomic or obsolete capacity.

Extensive, perhaps unprecedented, cooperation among governments would be necessary to overcome other organizational and institutional challenges. As a starting point, any initiative to solve steel’s problems must be based on a model such as this—one that will work both economically and financially, bringing capacity closer to demand and hence prices to a level that could sustain the industry. The challenges are great. But governments and the industry should rise to them, for the risks of maintaining the status quo are greater still.