Arreglos institucionales

The services sector has come to dominate the economic structure of many economies in the latter half of the 20^th century and even more so in the 21^st century, both in terms of output and employment. There is therefore a growing belief that we have now entered a ‘post-industrial society’, in which services are gradually taking over for manufacturing as a source of productivity growth and economic development. This is especially apparent in industrialised economies, where, as mentioned, the manufacturing sector seems to be shrinking. Michael Porter, one of the most well-known advocates of the post-industrial society discourse, has argued that services are where the high value is today:

We used to think of services as flipping hamburgers, now we have to think of services as rocket science. Services are where the high value is today, not in manufacturing.

Manufacturing stuff per se is relatively low value. That is why it is being done in China or Thailand (McCormack, 2006, p.1).

Also in some African countries, services based growth strategies—especially those centred on growth of the tourism and telecommunications sectors—rather than manufacturing based growth strategies, have become a more common strategy to diversify away from dependence of primary commodities (see ACET, 2014; UNECA, 2015).

There are good reasons to take the post-industrial society discourse seriously. First, given massive increases in firm size, it is now more profitable to procure some services from specialist providers rather than produce them within a manufacturing firm (the latter largely being the case in the past). Telecommunications, finance, and business services are now organised in a way that resembles the manufacturing sector, as scale economies and technological advances are more easily incorporated into these services to increase efficiency.

In some digitalised services, especially in advanced economies, marginal costs of providing an additional unit of service have come close to zero, making scale economies even more prevalent than in the manufacturing sector. For example, media streaming services, such as Netflix and Spotify, can deliver their services around the globe in a flash and have already proven their tremendous potential for scale economies. In 2016, Netflix had 93.8 million paying subscribers, while Spotify had 40 million paying subscribers. Only 6 years before that, both services had a marginal subscriber base (Statista, 2017).

Second, the revolutions in ICT and transport technology have made more services tradeable. The poster child of services-based trade success is the UK, where trade in services now account for 19.4 per cent of GDP—roughly 50 per cent higher than the world average (WDI, 2017)—thanks to growth in trade of mainly financial and business services. India is another country often cited as having achieved success through exporting services, like software, accountancy and the reading of medical scanning images. A third example is Rwanda, a country that in the last 10 years has increased its foreign exchange earnings considerably through the expansion of tourism services, such as gorilla viewing. In fact, Rwanda, and many other African countries, like Uganda, Tanzania and Tunisia, report that tourism is the top single earner of foreign exchange for their respective countries.

Third, it is not improbable that in the future we will see a type of ‘Engel’s law’ whereby the income elasticity of demand for manufactured goods on a global scale becomes significantly lower than that of services, as world incomes rise. Especially in the more affluent parts of the world, we are seeing that the consumption of services is increasing with rising incomes. One aspect is ‘servicising’ what used to be our daily tasks: people are eating out, or ordering take-away rather than cooking themselves, and we have the option of getting others to clean for us, walk our dogs, and take care of our children. In the end, the same amount of

goods is consumed, but an extra layer of services is being added to the ‘consumption basket’.

Another aspect is simply having more available income and being able to spend more on services. For example, spending money on ‘experiences’ rather than ‘things’ has become fashionable among the millennial generation.

However, there are good reasons to be sceptical of the post-industrial society discourse.

First, too often, services are equated with ‘knowledge’ work and manufacturing is stereotyped as low value ‘grunge’ work. It is dangerous to generalise in either direction. Burger-flipping, on the one hand, and the design of space rockets, on the other hand, are services that surely do not generate the same amount of value. Likewise, the assembly of an Apple iPod is a manufacturing activity that might not generate much value, but the manufacturing of its display module and its multimedia processor chip does. Manufacturing is just as much of a

‘knowledge’ sector as services is. Moreover, the amount of value that accrues to an activity in the value chain is often an outcome of market position and power, rather than the complexity of the activity. As we will see in the next chapter, the huge chunk of revenue generated in business-related services by retailers and brand names in the fast-moving consumer goods sectors (the most prominent example being clothes) has a lot to do with the global reach and market power that these companies, like Nike and Walmart, have attained over the years.

Second, it does not always make sense, or is not always possible, to look at the value and productivity generation of manufacturing and services separately. There is a massive interdependence between the two. As we have already seen, productivity growth in the services sector would not be possible without inputs from the manufacturing sector. And many services that have grown rapidly in the last few decades are heavily dependent on manufacturing firms as their customers. These include banking, communications, insurance, construction and consulting (equally, manufacturing operations run smoothly thanks to these services). Even more importantly, producer services, such as transport, design, retail, supply chain management and engineering, are by definition linked to the production of goods. A country can to some extent specialise in exporting these services, but there are still huge benefits of co-locating them within national boundaries. This means that those countries which lose their manufacturing bases are likely to lose many of their important services as well.

Furthermore, in certain manufacturing industries, it does not make sense to separate manufacturing, R&D and design activities. Pisano and Shih (2012, p.66) presents a framework that explains in which industries it makes sense to co-locate these activities. Basically, in industries where major manufacturing process innovations are evolving rapidly and can have a huge impact on the product, it makes sense to integrate R&D, design and manufacturing.

Examples of industries where the risk of separating these activities are “enormous” include biotech drugs, nanomaterials, electrophoretic displays and superminiaturised assembly.

Examples of industries where outsourcing makes sense include desktop computers, consumer electronics, active pharmaceutical ingredients and commodity semiconductors. There is an additional, but subtler and less visible ‘symbiosis’ between these activities: many of the people who work in process R&D and product design have technical manufacturing experience, obtained either through higher education or work experience.

Third, low tradability characterises most services because they require consumers and producers to be in the same location, like cleaning, grooming, public utilities or education. No one has yet invented ways to provide a haircut or house cleaning long-distance. This means that countries that rely on their services sector for economic growth will eventually struggle with trade balance constraints. Services have in fact been stuck at around 20 per cent of international trade since the 1990s. The UK has had a negative trade balance of £4 billion since 2014, because its trade surplus in services cannot make up for its trade deficit in goods (ONS, 2017). Between 2004 and 2011, in India, which, like the UK, is also supposed to be a model of service-based economic development, trade surplus in services covered only 17 per cent of its trade deficit in goods (Chang, 2014).

Fourth, the decline of the manufacturing sector is partly an illusion. Much of the apparent fall in the manufacturing sector’s share of GDP in advanced economies is due to the decline in the prices of manufactured goods, relative to the prices of services. This is thanks to faster productivity growth in the manufacturing sector. Think about how computers and mobile phones have become cheaper (holding the quality constant), compared with the cost of getting a haircut or eating out. When this relative price effect is taken into account and the shares of different sectors are recalculated in constant prices, as opposed to current prices, the share of manufacturing has not fallen very much in most high-income countries. In some of them, like the US, Switzerland, Finland and Sweden, when calculated in constant prices, it has actually increased (Chang, 2014).¹² The growth of the services sector is also a bit of an illusion. A lot of services that are now supplied by independent companies at home or abroad used to be provided in-house in manufacturing firms (for example, catering, security guards, some design

12 The relative price decline of manufactured goods is also a result of price increases of services. Because the services sector has lower potential for productivity growth, income growth in advanced economies, combined with the fact that many services are not tradable, has led to higher wages and prices in the services sector (Baumol, 1967), as well as an increased employment share (Rowthorn and Wells, 1987). Thus, there is good reason to believe that growth of the services sector has come about as a result of economic growth, and not the other way around, as indeed some people argue.

and engineering activities). In that sense, there has not been a big restructuring of the advanced economies, but more a change in the way we count/measure certain economic activities.