Among the Western Balkans countries, Serbia and North Macedonia are the most active suppliers of online labour. Most freelancers in the region work for buyers from North America and Europe. There is almost no network flow into the other direction.
Online labour participation is regionally clustered within the countries. While most of the online labour activity is generally concentrated in the capital region, more rural regions usually have much less active freelancer communities.
The figure shows the relation between demand and world market price for Web Developer projects. As economic theory predicts, demand and price are inversely related.We conclude that demand and supply on online labour platforms are closely interlinked, leading to fluctuations in the price paid for projects. These dynamics have consequences for the income potential of people who want to make a living as online freelancers and underline the importance of up-skilling.
The figure shows the share of freelancers per country in the five income groups. While the groups containing freelancers that worked on at least one project are of roughly equal size, there is, at the same time, a vast excess supply of labour: 89 % of the registered freelancers from the region did not get at least one project, and hence have earned nothing.
For the success of online labour platforms as a means of economic development, it will be paramount to activate this large pool of potential freelancers. It will therefore be necessary to better match online labour supply to the needs of customers.
The Figure shows the hourly rates of active freelancers on UpWork. Most freelancers from the region earn comparatively low wages. A minority of 12 % demand hourly rates of 30 USD or more.
The developer platform Stack Overflow allows us to gain insights about both supply and demand of digital skills. On the platform’s job board companies from around the globe advertise job openings. At the same time, contributions on the Q&A forum reflect the general skill level of the global community in specific technologies or programming languages. For both demand and supply, for example, we can observe the rise of the general programming language Python, over the last decade.
On the Stack Overflow company board tech companies from all over the globe are listed. As of March 2019, 26 companies that have an office in Belgrade are visible on the forum. According to their offices in other countries, a network of Belgrade tech companies is created. Cities are linked if their local companies share offices with other locations. The different nodes are sized by the number of companies they host. The colours of the nodes represent the skills they demand.
Digital skills, such as knowing how to program in Python or C++, have realtionships with each other. On the Stack Overflow Job Board, companies advertise their job together with a set of skills they require. From this list, a network of skills is constructed. Clearly, main programmming languages, like Java, Javascript, or Python are central nodes in the network - other capabilities are often required on top of these main skills.
Among the digital technologies, Machine Learning and Artificial Intelligence have recently got increasing public attention as it is speculated that they might disrupt many sectors of the economy. Accordingly, the accumulation of human capital in these sectors is important for future economic development. The figure displays the quarterly contributions to four important subtopics / programming languages within AI: Machine Learning, Scikit-Learn (a Python-based ML framework), Deep Learning, and TensorFlow (an important Deep Learning framework). In contrast to other Eastern European Countries (Ukraine, Russia), and in particular in comparison with the European technology leaders in the United Kingdom, the Western Balkans have contributed less to all technologies.
Oxford Data Science
Oxford Data Science is a group of Oxford and Berlin-based researchers with a strong focus on digital platform analysis.
Fabian Braesemann
Dr. Fabian Braesemann is Computational Social Scientist at Saïd Business School at the University of Oxford. His research interest is on applying data mining and machine learning techniques to measure the evolving digital knowledge economy. Dr. Braesemann has worked as a Data Scientist in marketing where he applied statistical methods on customer data to optimise marketing efficiency of retail companies. He gained his PhD at the Vienna University of Economics and Business after having finishing his studies in Economics at Humboldt University and Technical University in Berlin. Dr. Braesemann is a Research Associate at the Oxford Internet Institute.
Fabian Stephany
Dr. Stephany is a free-lance Computational Social Scientist. His research focuses on the application of data science and social science statistics in education, migration, and public policy. Dr. Stephany has been working for various actors in the international policy landscape, such as the Viennese think tank Agenda Austria and the OECD in Paris. He holds degrees in Economics and Social Sciences from Mannheim University, Universitá Bocconi Milan, Cambridge University, and Vienna University of Economics and Business. He is affiliated at the Humboldt Institute for Internet and Society in Berlin and the Wittgenstein Centre for Demography and Global Human Capital in Vienna.
Disclaimer
The designations employed and the presentation of material on our maps do not imply the expression of any opinion whatsoever on the part of the Secretariat of the United Nations or UNDP concerning the legal status of any country, territory, city or area or its authorities, or concerning the delimitation of its frontiers or boundaries. (Version: April 2019)