How best to bring back manufacturing


Attitudes to manufacturing were a small but telling split in the cold war. The Soviet Union had such a focus on industry that its statisticians kept services from the country’s measure of national income. A year after the conflict ended, Michael Boskin, then the White House’s chief economist, is said to have joked it did not matter whether the “chips” America produced were made from semiconductors or potatoes. There are echoes in the present geopolitical face-off. Xi Jinping, China’s president, is so focused on hard tech that he has cracked down on consumer-tech firms.

Listen to this story. Enjoy more audio and podcasts on iOS or Android.

Your browser does not support the <audio> element.

Listen to this story

Save time by listening to our audio articles as you multitask

But Mr Boskin’s laissez-faire approach is no longer in vogue among Western policymakers. They have introduced a sweep of policies intended to “onshore” manufacturing. In July America’s Congress passed the Chips and Science Act, which will dole out $ 52bn to the chip industry over five years, mostly to subsidise domestic production. Japan and Europe are also spending big on chips. The majority of the eu’s €43bn ($ 49bn) package will subsidise “mega fabs”, or cutting-edge chip-fabrication plants. In August America also passed a climate-change package, worth nearly $ 400bn, stuffed with “made in the usa” subsidies to be spent over ten years. West Virginia is getting wind farms; electric-vehicle battery factories are coming to Ohio.

Arguments for onshoring fit into two categories. The first concern security. More than 90% of advanced chips, many needed for manufacturing weapons, are made in Taiwan—far closer to China than is comfortable for the West. The second concern economics. Advocates claim that manufacturing can create mountains of well-paid jobs. Economists are doubtful. A paper published in 2018 by Teresa Fort of Dartmouth College, Justin Pierce of the Federal Reserve Board of Governors and Peter Schott of the Yale School of Management finds that the number of jobs in American manufacturing has fallen considerably since 2000, but output has not. That is in part because American industry has become more technologically intensive and therefore productive. It is thus unlikely more high-tech factories will mean many more jobs.

But there is another, more subtle economic case for onshoring. Gary Pisano and Willy Shih, both of Harvard Business School, argue that there can be broader “spillover” benefits to innovation from having a strong manufacturing base. One way this happens is when research and development (r&d) on products is done next to the manufacturing of them. This eases collaboration between the two stages, which is especially important in the early days of new products. A working paper by Teresa Fort of Dartmouth University, Wolfgang Keller of the University of Colorado and colleagues looks at innovation among American firms. It finds that those that locate their manufacturing near their r&d produce more patents and citations. Indeed, the smaller the geographic distance between the manufacturing and innovation arms of a firm, the more innovation ensues.

What about when firms deem it better value-for-money to move production elsewhere? Another working paper by Lee Branstetter, then of Carnegie Mellon University, Britta Glennon of the University of Pennsylvania and colleagues, examines just that. In 2001 Taiwan lifted rules banning the offshoring of production to China, but only for some products. The study finds that offshoring did reduce the quantity of patents related to these products. But it also freed up resources for r&d in adjacent types of products, leading to more patents in these areas.

Mr Pisano and Mr Shih suggest that the benefits of locating r&d and manufacturing near one another depends on the type of work. For instance, when your columnist researched and wrote this article, there was no reason for him to be located near the printer or distributor of The Economist because software neatly separates the two steps. By contrast, new biotech drugs often require r&d to be near production facilities, because drug design is closely linked to the manufacturing process. Semiconductors, the subject of many of the recent round of industrial policies, lie somewhere in between these two types of work. A few firms like Nvidia, in California, only design chips and send the designs overseas to be made by other companies like tsmc, a Taiwanese chip firm. But Taiwan also has a burgeoning chip-design industry, in part because of its advanced-manufacturing prowess, which makes it easier for startups to prototype and test new ideas.

To subsidise or not to subsidise

That manufacturing sometimes boosts innovation does not justify the enormous price tag carried by subsidies. Even many economists who sing industry’s praises concur. Governments tend to be poor at picking industries and technologies to support. And as Mr Pisano notes, if there truly were big enough benefits to be found from moving r&d and manufacturing near one another, firms would do it themselves.

The case of American competition with Japan in the 1980s and 1990s offers a useful parallel. Just like now, policymakers in Washington worried about losing market share in advanced-tech manufacturing. But as Mr Branstetter and colleagues noted in a paper in 2013, a falling market share did not stop American firms from better capitalising on the software boom that followed. One difference between the countries was openness to immigration. American firms could simply draw on a bigger pool of programmers. Mr Branstetter also notes that Japan’s government incentivised hardware manufacturing, delaying a pivot to software.

In a twist of fate, such openness to foreign talent may support manufacturing at home. Another working paper by Ms Glennon finds strong evidence that restrictions on h1b visas, which are intended for employment of high-skilled foreigners, lead to more offshoring, as firms are forced to head abroad for talent. That is an inconvenient finding for the many politicians who both support domestic manufacturing and are loth to increase immigration. ?

Read more from Free Exchange, our column on economics:
How to escape scientific stagnation (Oct 27th)
Why inflation refuses to go away (Oct 19th)
Energy shocks can have perverse consequences (Oct 13th)

For more expert analysis of the biggest stories in economics, finance and markets, sign up to Money Talks, our weekly subscriber-only newsletter.