International trade is traditionally thought of as goods crossing borders. Trade in services, however, is becoming increasingly important for high-income countries and its role is likely to grow substantially over the next years (Francois and Hoekman 2010).
Trade in services and goods differ along several critical dimensions (WTO 2010).
- Goods must physically cross borders, so the means of transportation matter.
- Services are intangible and thus do not need to cross borders in a physical sense.
- Goods’ characteristics are generally observable before purchase, and goods can be produced, moved, stored and consumed in different locations and time.
- Typically, services are not storable, their characteristics are not observable before purchase, and production and consumption often coincide spatially and temporally.
- Trade in goods is subject to tariffs that discriminate across countries.
- Trade in services is not subject to any tariff, but rather to technical barriers1 that rarely discriminate across countries.
Why are trade patterns so similar when goods and services are so different?
Despite these differences, recent studies using firm-level trade data2 surprisingly find that both types of trade share many qualitative features and only few, if any, differences. In new research (Ariu 2012) I use a more detailed dataset on Belgian firms for the1995-2005 period to go deeper into the comparison of trade in goods and in services.
Trade is rare, even more so for services
It is a well-known fact in the literature that among all operating firms, only a tiny minority – around 4% - participate in international trade in goods (Bernard et al. 2007). This low percentage highlights that international trade is a very selective process and only very few firms can consider operating in foreign markets. In Ariu (2012) I find that among all operating firms, only 1% participate in international trade in services. This suggests that trade in services is even more elitist than trade in goods.
Our results point at higher fixed costs as the cause of this low participation rate for trade in services. A more severe selection process allows fewer, but bigger and more productive firms to enter the export market. This mechanism might be induced by the relatively higher restrictions that are still in place for trade in services, such as visa requirements, special national certifications or professional associations (Breinlich and Criscuolo 2011).
Another interesting aspect of trade participation is that 5% of exporters and 8% of importers trade both goods and services. Despite representing a small number of firms, these account for more than 30% of total trade. Therefore, even if firms have to pay higher fixed costs to start trading both goods and services, selling them together might generate more value than trading only one of them. This fact can be an indication that there can be complementarities in exporting them jointly. Examples of this phenomenon are vehicles and insurance, machinery and technical services such as repairs or maintenances.
Trade frequency and size
Firms trading goods tend to make a lot of shipments over one year, and the value of each tends to be relatively small. On the contrary, firms trading services make very few transactions and their size tends to be relatively big. This difference remains true even when we focus only on firms trading both goods and services. The difference in transaction size and frequency may be due to the fact that goods can be easily divided while services are not divisible. Therefore, firms exporting services tend to do it all at once, while firms trading goods divide the supply into different shipments and customise the process following customers’ needs.
Entry and exit in foreign markets
Every year, 43% of firms that export services are “new” exporting firms, i.e. firms that were not exporting services previously. The entry rate is thus very high. This percentage is only 31% for goods’ exporters. Yet, 36% of firms involved in export of services will not export the following year (this