Many believe that the service sector cannot be relied on to sustain growth. But while an economy cannot thrive just on haircuts, for example, the ability of the service sector to generate growth is widely underestimated. The opposite is the case for manufacturing and agriculture, sectors that produce “real” things. This is mainly because demand for raw materials and goods tends to grow much less rapidly than incomes beyond a certain income level, while demand for services tends to grow more rapidly than incomes beyond that point. Moreover, technological advances and global competition have historically spurred enormous gains in the production of raw materials and goods, more so than services, which means that the job-generating capacity of the agriculture and manufacturing sectors is inherently even more limited.

In recent years, this has begun to change. With digitization, the fall of communications and computing costs, and exposure to global competition, some parts of the service sector have seen large increases in labor productivity and have begun to shed labor. Still, everywhere, it appears, the demand for more and increasingly sophisticated services remains buoyant. In fact, across all the high-income countries, nearly all economic growth is now accounted for by services.

In both advanced and developing countries, the rapid rise of the service sector continues to be met with apprehension and viewed as a symptom of economic decline. Left unchecked, this attitude could lead to bad policy. For example, advanced countries could fail to press their competitive advantages in many sophisticated services, while developing countries could overlook the importance of measures that facilitate growth of their budding service sectors. Most damaging, all countries are tempted by these attitudes to resort to hidden or open agricultural and industrial subsidies or outright protectionism.

There are undoubtedly instances where the rise of the service sector is a symptom of misguided policies—such as an overvalued exchange rate or lack of fiscal discipline or restrictions on competition—which need to be corrected, but a rapidly growing service sector is actually the norm, reflecting market forces.

A Look at the Data

Over the past two decades, the service sector has propelled GDP and employment growth in both advanced and developing countries.

In advanced countries, the service sector accounted for 90 percent of GDP growth between 1985 and 2005. In 1991, services contributed to two-thirds of GDP, but by 2008, this share had risen to nearly three-fourths. Services also accounted for all employment growth, as employment in the manufacturing sector declined. In 2008, the service sector employed 72 percent of the labor force, on average, in high-income countries—ten points more than the share it employed in 1991.1

Clearly, services are running the show in high-income countries, while manufacturing is playing a less significant role. But does this mean that the manufacturing sector is performing poorly? Hardly so. Manufacturing in most advanced countries is leaner and more productive than it has ever been. Since 1960, manufacturing output per worker has increased by over 500 percent, on average, across thirteen major high-income countries.2 Over the past 20 years, productivity growth in the manufacturing sector accounted for nearly half of total productivity gains in high-income countries (see chart). Fewer workers are now needed to produce the same amount of output. Even so, the total value added of manufacturing has increased.

If this storyline sounds familiar, that is because it is precisely what happened at the turn of the century when the United States and other advanced economies began to move away from agriculture en masse and industry took off (see chart). Rising productivity due to advances in technology and improved farming practices meant that fewer workers were needed to produce the same amount of output. But this shift hardly spelled disaster. Workers who were released from farms found new jobs in factories. Agriculture output, meanwhile, is much higher than it was in 1900 and has been climbing for a century, while the United States continues to run an agriculture trade surplus.

More surprisingly perhaps, the service sector has played almost as important a role in middle-income countries, especially in South Asia. Services accounted for 85 percent of employment growth, on average, across 20 major middle-income countries over the past 20 years.3 Their contribution to GDP in Bangladesh, India, Nepal, Pakistan, and Sri Lanka is high relative to these countries’ level of development, especially as compared to East Asia. But other figures show the continued importance of manufacturing. Over the past two decades, manufacturing accounted for nearly half of GDP growth and the value-added growth rate (5 percent) has kept pace with that in services. A significant fraction of the population—almost a quarter—also remains employed in manufacturing.

The Service Sector is Changing

Imagine trying to drive economic growth on the basis of haircuts, a classic traditional service. The technology needed to provide a haircut has not significantly changed, insulating this service from a major source of productivity gains. Haircuts must be provided in person, meaning that there is no potential to trade. Finally, they can only be provided to one person at a time, limiting the opportunity to reap economies of scale and gain from specialization.

Today’s service sector, however, is a far cry from the humble haircut. It includes traditional services provided by the hotel and restaurant industries, tourism, health care, and government but also increasingly sophisticated services in areas such as banking, accounting, insurance, and consulting. Both types of services, but especially the latter, have been affected by advances in technology that have increased the reach and speed of communications, lowered the cost of transportation, and helped break down barriers to trade.


Technology has made many services increasingly tradable and scalable. Between 1980 and 2009, the share of these more sophisticated services in total world trade grew by nearly 20 percentage points. Suppliers of these services can sell not just to their home country, but increasingly to customers in any part of the world. A larger market, in turn, means greater scalability and the cost savings and efficiencies that come with it. For their part, consumers of modern services also benefit. Businesses can take advantage of low-cost locations and build up more efficient supply chains while households and individuals pay lower prices.

A larger market for services also means more competition, which—combined with advances in technology—has spurred productivity. Between 1987 and 2005, the annual labor productivity growth rate in the U.S. service sector more than doubled (as compared to a 0.5 percentage point increase in the manufacturing sector). This means that the service sector is less afflicted by the “cost disease,” which posits that the cost of services will climb as rising wages combine with stagnant productivity (especially in the public sector) to make them more expensive over time. The price of education and health care has indeed skyrocketed, but many areas of services may no longer be as susceptible to growing costs.

Policy Implications

As the manufacturing sector continues to shrink in advanced countries, policymakers will face louder calls for protective measures. But protectionism is a bad idea for many reasons, of which we highlight two. First, it would likely provoke large markets such as China and India to enact protective measures of their own. Second, protectionism hits at the wrong target. This is because the shift away from manufacturing in advanced countries is primarily driven by technologies and consumer preferences, not outsourcing; it is a sign of progress rather than decline (see chart).

Advanced countries should harness rather than resist the economic forces that are galvanizing the service sector. In manufacturing, this means continuing to push for productivity improvements and focusing on the high value-added end of the manufacturing chain, which is capital- and technology-intensive. In the service sector, advanced countries should focus on their competitive advantage in generating ideas and innovation and making sure that their workforces have the education and skills needed to be competitive for service jobs.

Developing countries, meanwhile, should view a strong service sector as a crucial complement to a robust manufacturing and/or natural resources sector. The increased tradability of services is enabling developing countries to get a bigger slice of the services pie. But it would be wrong to conclude that they can rely only, or even mainly, on their service sector to develop. Many services require skilled labor and quality infrastructure, which developing countries are still in the process of developing. By focusing mostly on services, moreover, developing countries would neglect their competitive advantage vis-à-vis advanced countries (abundant labor and low costs) in much of the manufacturing chain, and their natural resources.

Uri Dadush is the director of Carnegie’s International Economics Program. Zaahira Wyne is the managing editor of the International Economic Bulletin.

1. Some of the employment growth in services is optical. Many jobs in the manufacturing sector—for example, those that involve designing and marketing products—are service-oriented, and there is a growing trend towards outsourcing them rather than completing them in-house. This has slightly inflated employment growth in the services sector, but it is only a small factor in what has been a real shift.

2. Figure covers the 13 countries, of 19 listed, for which data is available from 1960 to 2010.

3. Figure covers the 20 middle-income countries for which data is available, including Argentina, Brazil, China, Egypt, and Turkey.