Tuesday, November 7, 2017

Trade, Technology, and Job Disruption

Both technological developments and international trade can disrupt an economy, and in somewhat similar ways, but many people have very different reactions to these forces. To illustrate the point, I sometimes pose this question:

There's a US company which has developed a new technology that allows them to make a certain product more cheaply. This company hires some additional workers, but the other firms trying to make that same product don't have the technology, so they lay off workers or even go bankrupt. Should step be taken to ban or limit the use of this new technology?

Pause for thought. The usual reaction that emerges from the discussion is that we can't hope to freeze technology in place. Ultimately, we don't want to be a society with lots of workers who light gas streetlamps, or who operate telegraphs or who plow fields with oxen. Sure, it's important to have social policies to cushion the transition to new industries, but overall, we need to be facilitating new technology rather than blocking it.

All of which is fair enough, but here's the kicker. Now you discover that the "new technology" from the US firm is that it is importing more cheaply from a foreign provider. The same disruption of the US  labor force is occurring, but as a result of an expansion of international trade rather than as a result of technology. Personally, my response to the economic disruption of trade is essentially same as my response to the economic disruption of technology: that is, I believe in assisting the transition for dislocated workers no matter the reason behind the dislocation. But for many people, their reaction to economic disruption is different depending on whether the underlying cause is technology or trade.

There arguments are renewed and refreshed in a couple of recent publications. J. Bradford DeLong has written "When Globalization Is Public Enemy Number One" in the most recent issue of the Milken Institute Review (Fourth Quarter 2017, 19:4, pp. 22-31). Also, the World Trade Report 2017 from the World Trade Organization is centered on the theme, "Trade, Technology, and Jobs."

As a starting point, here's a figure from DeLong's paper about the rise of globalization. The red line shows the sum of exports and imports compared to world GDP. The first explosion of globalization starting in the 19th century, and the more recent rise of globalization, are both readily apparent.
Delong Bradford Chart1
But of course, a rise in trade isn't the only economic change taking place. Brad points out that the fall in blue-collar and manufacturing jobs was well underway back in the 1950s and 1960s, well before globalization had restarted in force--because of changes in technology  Indeed, I've written before about "Automation and Job Loss: The Fears of 1964" (December 1, 2014). Brad readily admits that the shock of increased trade with China starting around 2001 was an important event, and of course the Great Recession had a powerful effect on jobs too. But overall, he writes:

by his calculations. only a very minor part of the decline in blue-collar jobs since 1948 is about international trade: it's mostly about technological change, and to some extent about the rising strength of economies in other parts of the world and misjudgments of macroeconomic policy by the US government.
"To repeat, because it bears repeating: globalization in general and the rise of the Chinese export economy have cost some blue-collar jobs for Americans. But globalization has had only a minor impact on the long decline in the portion of the economy that makes use of high-paying blue-collar labor traditionally associated with men. ... Pascal Lamy, the former head of the World Trade Organization, likes to quote China’s sixth Buddhist patriarch: `When the wise man points at the moon, the fool looks at the finger.' Market capitalism, he says, is the moon. Globalization is the finger."
Given that comment from Lamy, it is perhaps unsurprising that the World Trade Report 2017 takes a position similar to DeLong. There are roughly a jillion examples of how technology both improves productivity but also can also disrupt job markets. The report summarizes:
"By making some products or production processes obsolete, and by creating new products or expanding demand for products that are continuously innovated, technological change is necessarily associated with the reallocation of labour across and within sectors and firms. Such technology-induced reallocations affect workers differently, depending on their skills or on the tasks they perform. ICTs tend to be used more intensively and more productively by skilled workers than by unskilled workers. Automation tends to affect routine activities more than non-routine activities, because machines still do not perform as well as humans when it comes to dexterity or communication skills. ... [T]he labour market effects of technology are relatively more favourable to skilled workers and to workers performing tasks that are harder to automate."
What about the worry that technology will lead to a dramatic reduction in the total number of jobs? Obviously, this prediction is not an extrapolation from history. The US and world economy have been experiencing technological growth in a serious way for a couple of centuries, and there is no long-run downward trend in the total number of jobs. Why is that? The report offers these reasons (citations
omitted):

"The view that the new technological advances in artificial intelligence and robotics will not lead to a `jobless future' is based on historical experience. Although each wave of technological change has generated technological anxiety and led to temporary disruptions with the disappearance of some tasks and jobs, other jobs have been modified, and new and often better jobs have eventually been developed and filled through three interrelated mechanisms.

"First, new technological innovations still require a workforce to produce and provide the goods, services and equipment necessary to implement the new technologies. Recent empirical evidence suggests that employment growth in the United States between 1980 and 2007 was significantly greater in occupations encompassing more new job titles. 
"Second, the new wave of technologies may enhance the competitiveness of firms adopting these technologies by increasing their productivity. These firms may experience a higher demand for the goods or services they produce, which could imply an increase in their labour demand. Several empirical studies ... find that the adoption of labour-saving technologies did not reduce the overall labour demand in European countries and other developed economies. 
"Finally, ... the upcoming technological advances may complement some tasks or occupations and therefore increase labour productivity, which could lead to either higher employment or higher wages, or both. The new workers and/or those benefitting from a pay rise may increase their consumption spending, which in turn tends to maintain or raise the demand for labour in the economy. Recent empirical evidence suggests that the use of industrial robots at the sector level has led to an increase in both labour productivity and wages for workers in Australia, 14 European countries, the Republic of Korea and the United States."
It's of course impossible to prove that future patterns will be similar. But the historical evidence suggests that finding ways to stimulate and work with technology is a better path to prosperity than trying to limit or block it.

In the discussion of trade and jobs, the report readily admits that trade (like technology) causes economic change and dislocation. After a substantial discussion of the empirical evidence, here are some conclusions from the report:
"First, evidence consistently shows that the welfare gains from trade are considerably larger than the costs. Effects on aggregate employment are minor and tend to be positive. The net effect on welfare depends on the magnitude of adjustment costs and trade gains. But existing evidence evaluates costs to be just a fraction of the gains.
"Second, the debate over the labour market effects of import competition needs to be qualified. While some manufacturing jobs may be lost in some local labour markets, other jobs may be created in other zones or in the services sector. When researchers take these effects into account their findings suggest a positive overall effect of trade on employment. Similar results are found when input-output linkages are taken into account or when the response of the labour supply to increased real wages is accounted for. Clearly, those who lose jobs because of import competition are not necessarily the same workers who get new jobs in exporting firms, because they are likely to have different skillsets or limited labour mobility. These adjustment costs need to be taken into account, but without losing sight of the overall picture. 
"Third, there is evidence that export opportunities are associated with employment growth. In developing countries, improved access to foreign markets has contributed to the movement of workers away from agriculture and towards services and manufacturing, as well as away from household businesses toward  firms in the enterprise sector, and away from state-owned firms toward private domestic and foreign-owned firms. Although more should be done to understand how labour markets in least-developed countries (LDCs) are affected by trade opening, there is evidence that the involvement of LDCs in GVCs [global value chains] has been a vehicle for developing employment opportunities.
"Fourth, trade offers opportunities for better-paid jobs. A significant share of jobs is related to trade, either through exports or imports, and both exporters and importers pay higher wages. This is because trading is a skills-intensive activity. International trade requires the services of skilled workers, who can ensure compliance with international standards, manage international marketing and distribution, and meet the demanding standards of customers from high-income countries; and trade leads to the selection of more productive firms and provides firms with an incentive to upgrade their technology. There is evidence that better access to foreign markets benefits exporting firms and thus their workers. This in turn positively affects regions where these firms are located, as well as occupations that are intensively used by these firms.
"As regards the evidence on the impact of trade on wage dispersion, there is evidence that by increasing the demand for skills, trade contributes to wage differences between high- and low-skilled workers. ... It is also worth noting that most of the existing analysis fails to account for the fact that most of the gains from trade opening come through a reduction in prices. Workers are also consumers. Trade impacts their well-being not only through changes in the wage received, but also through changes in the price of the goods that they consume. Given that most of the gains from trade opening through the consumption channel accrue to lower-income groups, failing to account for the income-group specific price changes overestimates the impact on wage disparity."

For some additional discussion of concerns that technology (or trade) would decimate the number of jobs, see: