There is bad news and good news for workers when it comes to automation, says a recent McKinsey report on the subject.

The bad news is that 30 percent of work activities in 60 percent of occupations globally could be automated. The good news is that the actual proportion of work automation will likely be lower—due to technical, economic, and social factors—and that automation will create new occupations that don’t exist today as technologies have done in the past.

At the same time, the up and coming technologies such as artificial intelligence and robotics “will generate significant benefits for users, businesses, and economies,” says the report, “lifting productivity and economic growth.”

McKinsey’s scenarios across 46 countries suggest that between zero and one-third of work activities could be displaced by 2030, with a midpoint of 15 percent. Advanced economies will be more affected by automation than developing ones. Higher wage rates provide greater economic incentive to automate.

Demand for work and workers could increase even with automation as long as economies are growing. “History shows that economies that are not expanding do not generate job growth,” said the report.

Rising incomes and consumption in developing countries, increasing healthcare needs for aging societies, and investments in infrastructure and energy will create demand for work that could help offset the displacement of workers.

At the same time, the report found that by 2030, 75 million to 375 million workers—three to 14 percent of the global workforce—will need to change occupations. “All workers will need to adapt,” said the report, “as their occupations evolve alongside increasingly capable machines. Some of that adaptation will require higher educational attainment, or spending more time on activities that require social and emotional skills, creativity, high-level cognitive capabilities, and other skills relatively hard to automate.”

In the United States and other advanced economies, demand for high-wage occupations may grow while middle-wage occupations decline, the report concluded. “Increased investment and productivity growth from automation could spur enough growth to ensure full employment,” said the report, “but only if most displaced workers find new work within one year.”

In emerging economies such as China service and construction jobs will see the most net growth, fueling the growing middle class.

Policy priorities for governments will need to include job training as well as “enhancing labor market dynamism and enabling worker redeployment.” “Another priority is rethinking and strengthening transition and income support for workers caught in the crosscurrents of automation,” the report concluded. “These changes will challenge current educational and workforce training models, as well as business approaches to skill-building.”