This article ir republished from and in accordance with the policy of “VoxEU.org”
Dr. David Hummels
Professor of Economics at Purdue University
Postdoctoral Research Fellow at Yale University’s Department of Economics
Dr.Jakob R. Munch
Asian Dynamics Initiative Professor of International Economics in the Department of Economics, University of Copenhagen
Associate Professor of Economics, and Director of Graduate Studies, at Economics Department, Purdue University
With stagnating wages and lingering unemployment, income inequality is back in the headlines. Is globalisation to blame for this inequality? Is more education a solution? This column argues that focusing on university education misses important effects. It presents evidence that wage effects vary markedly among those with degrees depending on their specific skill sets, and that globalisation can often benefit workers
Fuelled by concerns over rising income inequality, Occupy Wall Street has grown into a global movement in slightly over 2 months, with protests i over 900 cities worldwide. Protestors have been criticised for lacking a specific set of policy demands, but in this the protestors are hardly alone.
There is no clear consensus among either academics or policy makers as to the specific causes of inequality – some point to technological change, others such as Feenstra and Hanson (1997, 1999) to the role of offshoring in boosting the demand for skilled labour, still others to regulation or corporate malfeasance.
Nor is there consensus regarding the policies likely to ameliorate inequality. Many suggest minimising inequality by increasing educational attainment. This seems perfectly sensible. If inequality results from a rising college wage premium, then boosting educational attainment should increase the number of households benefiting from that premium while also allowing supply to catch up to rising demand in order to slow growth in the premium itself.
However, there is a growing concern that college isn’t enough. Since 2000, inflation-adjusted earnings in the US have fallen for every educational group except for doctors, lawyers, MBAs, and PhDs (Wessel 2011). This has contributed to the perception that today’s college degrees may simply lead to jobs “that don’t exist or don’t pay middle-class wages” (Krugman 2011). Again, offshoring may be a culprit. Firms in India and China have moved up the value chain from textiles and apparel into sectors that intensively use college-educated workers, such as advanced electronics,
alternative energy, and computer software. Making matters worse, US college tuition has grown twice as fast as inflation, leaving students with heavy debt loads to go with their bleak employment prospects. And the US is not alone, as governments throughout Europe are looking to stem the flow of budgetary red ink by boosting tuition fees.
Does offshoring increase inequality?
Early work on offshoring and college premium in the 1980s focused on industry-level data and examined average wage bills. In Hummels et al. (2011) we examine changes in the Danish labour market from 1995-2006, a period of rapid globalisation for Denmark. In our data, we see the entire population of Danish workers and their characteristics including education, occupation, and wages. The worker data are matched to the data on
the firms in which they are employed, including detailed data on the trading behaviour of those firms. This allows us to identify changes in the wages individual workers are paid when their employers offshore production or increase exporting. The key element in our approach is the ability to see firms changing their trading behaviour due to external shocks, that is, changes in the foreign supply of inputs, or changes in the foreign demand for their output. This gives us confidence that the wage changes we identify are due to globalisation and not due to technological change or other factors affecting the firm.
When we focus on workers who remain
employed with the firm, we find that offshoring raises the college wage
premium, both by increasing wages (elasticity +3.6%) for college-educated
workers and lowering wages (elasticity -1.6%) for workers without a college
education. Over the 12 years of our sample, a firm that increases offshoring at
the same annual rate (5.5%) as Denmark as a whole would raise wages for
college-educated workers by 2.8% and lower wages for other workers by 1.3%.
At first pass, our findings appear
consistent with the older literature – that trade raises the college wage
premium and with it, inequality. But we can go further to understand other
mechanisms at work.
Which college degrees are valuable in a global economy?
Our data report the occupations of
Danish workers, and this allows us to measure the on-the-job requirements for
four categories of knowledge and skills that are closely related to college
degrees. When we examine the effects of offshoring on wages for different
knowledge groups, we find that offshoring has the largest positive effect on
occupations that require communication and language (premium of +4.4%),
followed by social sciences (+3.7%), and maths (+2.7%). The premium for natural
sciences and engineering is close to 0. This may seem curious given the policy
emphasis on STEM (Science, Technology, Engineering, and Maths) education in
many advanced economies, but if science and maths are universal languages, jobs
requiring them can be done anywhere with an educated workforce.
The primacy of communication in our
results is broadly consistent with findings on immigration and economy-wide
task specialisation by Peri and Sparber (2009) and Ottaviano et al. (2010) that
highlight natives’ comparative advantage in language and communication tasks.
We find that offshoring firms not only retain workers doing communication-intensive
tasks, but that these workers actually become more valuable when offshoring
rises. Why? One possible explanation is that offshoring increases interactions
between domestic workers and foreign workers from different cultural
backgrounds, and those differences raise the cost of communicating within the
firm (Lazear 1999). Having domestic workers with finely tuned communication and
language skills and social scientists with knowledge of other cultures and
societies may be useful in overcoming offshoring-induced costs of
A rising tide that lifts all boats
Does globalisation leave all
non-college-educated employees behind? In short, no. Globalising firms often
increase both offshoring and exporting activity. This is not coincidental –
offshoring may lower costs or increase firm productivity in a way that makes
the firm more competitive on world markets. We find that exports are a rising
tide that lifts all boats, as exporting firms increase wages for all types of
workers. For non-college-educated workers, the (positive) exporting wage
elasticity is larger than the (negative) offshoring wage elasticity. This
implies that if a particular firm increases its offshoring and exporting at the
same rate, these workers enjoy net wage gains. For manufacturing as a whole,
50% of these workers enjoy positive wage gains on net from globalisation. The
bad news is that 10% of non-college educated employees suffer wage losses more
than -1.5% per year. This corresponds to workers in firms that offshored
production but without a corresponding increase in exports.
How about laid-off workers?
Being laid off is painful, and it
turns out that offshoring makes lay-offs even more traumatic. Figure 1 plots
the change in earnings for college-educated workers who were displaced from
employment in year “0” (measured in Danish Krona – DKK). The solid line shows
mass lay-off displacement due to offshoring and the dotted line shows mass
layoff displacement due to other reasons (e.g. failing firms going out of
business). One year after displacement, college-educated workers laid-off from
offshoring firms lose 64,000 DKK, or 15% of pre-displacement earning. These
losses are more than double the loss for other laid-off college-educated
workers, and the earnings losses are more persistent.
Figure 1. Earnings after lay off, high skilled (DKK)
Figure 2. Earnings after lay off, low skill (DKK)
Figure 2 tells the same story for non-college educated workers, but the effects are even more pronounced with displaced workers losing 21% of their pre-displacement earnings. Offshoring increases earnings losses after lay-offs and makes earnings losses more persistent.
Why? Workers displaced due to offshoring are much more likely than other displaced workers to remain unemployed or out of the labour force, and offshored workers are more likely to switch industries in order to regain employment. This suggests that globalisation shocks are reducing demand for these workers’ skills, not only within the offshoring firm but throughout the Danish economy. In contrast, a worker displaced from a failing firm may find it relatively easy to find employment with a successful competitor because similar jobs are still available within the Danish economy.
The debate over globalisation and income inequality is often viewed through the lens of education – college-educated workers benefit from globalisation while other workers are hurt. While on average this may be true of Denmark, we think a more nuanced view is in order. Not all college-educated workers benefit – those laid off from offshoring firms suffer large and persistent earnings losses, and among those who remain employed wage changes vary considerably depending on knowledge sets. Not all workers without a college education lose – those fortunate enough to be employed in firms with rapidly expanding exports see sharp wage increases, a force large enough to yield wage gains due to globalisation for half of manufacturing workers.
All this suggests that getting a degree is perhaps less important than what you study and who you work for. Rather than focus on educational attainment as an end in itself, policy might better focus on specific skills that are valuable in a globalising world. Unfortunately, policymakers appear to be leaning the wrong way with their emphasis on the so-called STEM disciplines. We find that rising offshoring leads firms to value these knowledge sets less than communication, language, and social science skills.
But the most profound earnings effects are found in workers displaced due to offshoring. Unlike the normal job gains and losses that occur as firms grow and fail, globalisation shocks can result in sharp reductions in demand, economy-wide, for specific tasks. Re-attaching to the workforce may then require a more fundamental retraining of these workers. It remains an open question whether the programmes in place to retrain workers are adequate to the task.
Feenstra, RC and Gordon H Hanson
(1997), “Foreign Direct Investment and Wage Inequality: Evidence from
Mexico’s Maquiladoras”, Journal
of International Economics, 42:371-393.
Feenstra, RC and Gordon H Hanson (1999), “The Impact of Outsourcing
and High-Technology Capital on Wages: Estimates for the US“, Quarterly Journal of Economics, 114:907-940.
Hummels, David, Rasmus Jørgensen, Jakob R Munch, and Chong Xiang (2011), “The Wage Effects of
Offshoring: Evidence from Danish Matched Worker-Firm Data”, NBER
Working Paper 17496.
Krugman, Paul (2011), “Degrees and Dollars”, The New York Times, 6 March.
Lazear, Edward (1999), “Globalisation and the Market for Team-Mates”, Economic Journal, 109:C15-C40.
Ottaviano, Gianmarco IP, Giovanni Peri, and Greg C Wright (2010), “Immigration,
Offshoring and American Jobs”, mimeo, University of California Davis.
Peri, Giovanni and Chad Sparber (2009), “Task Specialization, Immigration and Wages”, American
Economic Journal: Applied Economics, 1:135-169.
Wessel, David (2011), “Only Advanced-Degree Holders See Wage Gains”, Real Time Economics Blog, Wall
Street Journal, 19 September.