Economic complexity: analysis and predictions for automation and income inequality
The effect that technological change will have on income inequality globally can be better understood and predicted by using economic complexity
Inequality
Inequality is present in today’s development discussions for two reasons. First, it has been argued that inequality has both increased and decreased since the Industrial Revolution, with conflicting datasets and term definitions. Using World in Data’s (2017) measure of inequality, inequality between country since then, following a surge in the last century, while intra-country inequality has increased in the last decades of the 20th century.
Decrease in intercountry inequality, increase in intracountry inequality. World in Data (2017)
This trend has continued over the last decade: while it is argued that inter-country inequality has decreased in recent years, as argue those who see a decrease in the Global Gini index (The Economist, 2012), the reality is that this index only measures the average for all countries, independent of size or absolute variations1. The reality is that inequality has become one of today’s most sensitive “social, economic and political challenges of our time” (The Economist, 2012).
The second cause for concern is the potential for massive inequality arising from technological change, in what’s commonly referred to as the “Fourth Industrial Revolution”. While technology has great potential to shape employment opportunities and education for good, it can also create a “winner-takes-all” global economic scenario that pushes low-skilled workers and low-income nations out of competitive positions, thus pushing up inequality levels further. This would mean developed nations bringing back manufacturing and industrial jobs from overseas due to their technological advances, reducing the need for low-skill labor. Thus, as Driemeier and Nayyar (2017) argue, these changes will challenge traditional economic growth models, concluding that the risk of rising inequality in the coming decades is high.
Manufacturing opportunities and patterns shifts with each industrial revolution. Driemeier and Nayyar (2017)
This poses a big risk to societies across developed and developing nations. According to the World Economic Forum (2017) there is existing data indicating that unequal societies tend to be more violent, have higher recidivism, suffer from worse mental health and obesity and have lower life expectancies than those with more equal societies. The Fourth Industrial Revolution will fundamentally challenge the idea of work as a way to find meaning in life, and could bring about more social exclusion (World Economic Forum, 2017).
The Future of Work
Economists and policymakers are hard at work to find out what countries might do to be better positioned to face the challenges that will come with technological change. One of the most interesting lines of research is that of “economic complexity”, which measures to what extent knowledge is embedded in the economy of a country (Observatory of Economic Complexity, 2017). Extensive research has been done into economic complexity as a way of measuring the resilience of a country’s economy: complexity has been proved to be a better predictor for future growth than traditional indicators of GDP, labor and capital (World Policy, 2011), but also for income inequality (Hartmann, Guevara, Figueroa, Aristarán, Hidalgo, 2016).
Economic Complexity
Ricardo Hausmann, one of the leading thinkers in the economic complexity field, recently pondered this question: “What does the future of work hold in store, and how should we prepare for it?” (Hausmann, 2017). The piece seemed to point to the use of complexity to understand how a country might be better prepared for these future challenges:
“We do know that most countries should focus on ensuring that they can master every new technology and exploit every new opportunity that comes their way” Hausmann, 2017
I will now break down some of the ways in which complexity might help us understand these questions, some short fallings and some new lines of inquiry.
On one hand, economic complexity can be a useful tool to understand the exposure of different countries to technological changes, and especially to automation, which appears could the most important factor leading to higher inequality. While there are no studies directly linking both, there are three concepts that could be a starting point for research into this relationship.
First, the framework Driemeier and Nayyar (2017) create to analyze the impact of new technologies on employment and income outline three pillars for future growth: Competitiveness, Capabilities and Connectedness. These are related to the definition of economic complexity, as, from a network perspective, these three levers determine a country’s ability to produce and trade goods and services with one another. This framework is useful to understand the present interactions between nations and to attempt to infer new trends and behaviors.
Secondly, complexity gives an insight into how well a country might be able to adapt to the “servification” of industries. According to Driemeier and Nayyar (2017), there is a trend towards a heavier involvement of services into the manufacturing processes, through activities like marketing, engineering or research. Developed countries with manufacturing activities are taking the lead in this, with countries like Australia, where agricultural exports are nearly one-third embodied services (Driemeier & Nayyar, 2017), which can provide new opportunities for employment in the country.
Lastly, growing economies usually require an investment in education to increase the professional opportunities for the citizens of a given country. As observed in the case of South Korea’s development, the shift towards more complex and higher value-add manufactures and services was sustained through investment in education that prepared students for the new economic conditions. The nature of the employment that will be available will require the performance of nonroutine and cognitive tasks, according to Driemeier and Nayyar (2017). Thus, complexity could predict the educational opportunities that might provide higher income for young professionals.
On the other hand, the relationship between economic complexity and the inequality caused by technological advancement may not be directly linked. The first question is whether economic complexity indicators represent a lever in themselves, or whether they solely reflect other levers that affect a country’s economic performance, such as the strength of its institutions, the political stability or geopolitical factors.
The first challenge to this relationship is that, so far, complexity has been quite successful at predicting growth according to traditional industrial growth models, such as those of South Korea, Taiwan, China and other Asian Tigers. This growth took place over a time period of growing demand for manufactures, low interest rates and low barriers to entry to industry for nations with low-wage employees. However, Driemeier and Nayyar (2017) argue that this manufacture-led growth model is obsolete and cannot be used as an example for increased growth and economic complexity. Thus, complexity might not adapt well to these new models.
Secondly, by design, the Economic Complexity Index only takes into account manufactured products that are traded between nations, which leaves some important holes in the methodology (such as services and domestic products): there is some new research to include services in the mix, break down the complexity of each individual product, and to extend the definition of complexity, which will improve the predictive power of complexity over time (Stojkoski, Utkovski, Kocarev, 2016).
Lastly, many argue that the main effect technological change will have on international trade will be shortening the global value chains for products and services, thus limiting a country’s ability to specialize and add complexity to existing items. An example of how this would happen is with new technologies such as 3D printing, which could reduce the number of raw materials need for a final product, and could eliminate the need for globally sourced products, as it would be centralized in developed countries with access to that technology (which would in turn affect the transportation and logistics industry).
Analyzing both sides contributes to an understanding of what new lines of inquiry we might want to explore. In order to better understand the relationship that there could be between complexity and inequality, I propose two ideas.
Firstly, I would look more in depth into the “Complexity — Automation Country Paradox” I found in Driemeier and Nayyar’s (2017) data. When looking at the percentage of jobs that are at high risk of automation per country, I found that the countries above the 6% mark are some of the most complex economies in the world (such as Germany or Japan, per Observatory of Economic Complexity, 2017), while many at the lower end of the ranks are not complex economies at all (Lao, Vietnam or Kenya).
The Complexity-Automation Country Paradox through the % of current jobs at risk of automation. Driemeier and Nayyar (2017)
This could be caused by the more complex countries having better access to the technologies that would lead to automation, such as 3D Printers, smart factory machinery and artificial intelligence applications, thus making substitution more likely developing countries. However, this trend would put the less complex countries in a non-competitive position, thus decreasing their incentives to produce those products. This is an insightful paradox because the plausible answers will be informed by assumptions about how countries will interact in the future to create employment and competitiveness opportunities.
Manufacturing sectors by feasibility of automation. Driemeier and Nayyar (2017)
The second proposed line of research would be linking specific products to their probability of automation. This would allow researchers to run simulations based on product-price competitiveness, which would allow analysts to understand how technological innovations might affect specific products over time and the countries that are able to produce them at a competitive price point today. This could be explored along with the paradox in order to test how product complexity might affect national complexity, and make predictions about the effect on inequality.
Predictions
When it comes to predicting the relationship between technological change and inequality, there are 5 points I argue will shape the future:
1. Manufacture-led development will be increasingly difficult as a path for development
2. New niches for developing countries: new services and internal/regional markets
3. Different levels of automation will depend on industry and country
4. Striving for greater economic complexity will likely allow countries to better adapt to technological changes
5. Meaning and fulfillment, more than economic indicators, are the end goal for citizens and must be included in future analyses
1. Manufacture-led development will be increasingly difficult as a path for development
The macroeconomic growth models used by the Asian Tigers in the 20th century will no longer be available for developing countries that are looking for greater economic growth per Driemeier and Nayyar (2017), given the new trends in manufacturing and services development. Because the developed countries will be able to attract industrial investment due to their higher access to technology, developing countries will not be able to gain a competitive advantage in the production of these goods. As the diagrams show, developed countries will gain an advantage over the manufactures previously done by developing nations, but the competitiveness of developing service providers will be explored thanks to greater access to technology
Global Economic Matrix — Traditional Model
Global Economic Matrix — New Model with lower competitiveness for developing nations
2. New niches for developing countries: new services and internal/regional markets
Developing nations will still have a few markets that will remain accessible for competition: regional and internal manufacture markets will be a source of growth for developing nations, as evidence by the increase in exports among African countries in the last decade, and new services that will be provided thanks to new access to technology.
Exports among African nations have increased in the last decade. Driemeier and Nayyar (2017)
3. Different levels of automation will depend on industry and country
Some industries are revising their predictions for automation and substitution of human labor, given that, while economically feasible, it will be long until automated labor is competitive to substitute very low skilled labor in some industries, such as textiles or mineral resources. However, and as aforementioned, this will not be a channel for substantial growth, as once salaries increase with greater training and complexity, the products will cease to be competitive
Impact of Automation per industry. Driemeier and Nayyar (2017)
4. Striving for greater economic complexity will likely allow countries to better adapt to technological changes
Using the 3 Cs framework (Competitiveness, Capabilities, Competitiveness), countries will likely position themselves better for the arrival of automation. However, the important question is whether developing countries should pursue higher economic complexity as a lever for greater resilience: developing countries can push for greater importance of the service sector in their economy given greater technological reach and higher incomes (“servification” will play a larger role in manufacturing, as seen in the diagram below), but it remains unclear how their productivity will evolve. More research is needed to conclude this question, as mentioned in the previous section.
Capabilities and Connectedness in trade with completion satisfaction levels . Driemeier and Nayyar (2017)
5. Meaning and fulfillment, more than economic indicators, are the end goal for citizens and must be included in future analyses
Demographic changes will change the traditional social structures and values of developed nations. This, added to economic changes and the high probability of higher inequality, will pose challenges to the life meaning and fulfilment of individuals, which was often realized through work. Governments will play a larger role in this aspect of citizens’ lives, and it must therefore be included in the economic analyses conducted in the future.
In conclusion, technological advancements will likely bring about benefits for society, but there is a high probability it will also bring about negative effects through higher income inequality. Economic complexity has proved to be a valuable method to analyze the effect technological change will have in the economy and society, though there is a need for more research to pinpoint more accurate actions governments can take to become more resilient to technological change.
Resources
● Driemeier, M.A., Nayyar, D. (2017) Trouble in the Making? The Future of Manufacturing-Led Development. World Bank Publication. Retrieved from www.worldbank.org/en/topic/competitiveness/publication/trouble-in-the-making-the-futur e-of-manufacturing-led-development
● Hartmann, D., Guevara, M., Jara-Figureoa, C., Aristarán, M., C.A., Hidalgo (2016) Linking Economic Complexity, Institutions and Income Inequality. MIT Media Lab. Retrieved from https://arxiv.org/pdf/1505.07907.pdf
● Hartmann, Jara-Figueroa, C., Guervara, M., Simoes, A., Hidalgo, C. (2016) The structural constraints of income inequality in Latin America. Retrieved from https://arxiv.org/pdf/1701.03770.pdf
● Hausmann, R. (2017) Making the Future Work for Us. Project Syndicate. Retrieved from https://www.project-syndicate.org/commentary/technology-future-of-work-by-ricardo-haus mann-2017–09
● Stojkoski, V., Utkovski, Z., Kocarev, L.( 2016) The Impact of Services on Economic Complexity: Service Sophistication as Route for Economic Growth. Retrieved from journals.plos.org/plosone/article?id=10.1371/journal.pone.0161633
● The Economist (2017) For richer, for poorer. Retrieved from www.economist.com/node/21564414
● World Economic Forum (2017) The Fourth Industrial Revolution. Top Link We Forum. Retrieved from https://toplink.weforum.org/knowledge/insight/a1Gb0000001RIhBEAW/explore/dimension/a1Gb00000027vYmEAI/summary
● World in Data (2017) Global Economic Inequality. Retrieved from https://ourworldindata.org/global-economic-inequality
● World Policy (2011) It’s Complicated”: How Economic Complexity Predicts Growth. Retrieved from www.worldpolicy.org/blog/2011/10/31/“it’s-complicated”-how-economic-complexity-predi cts-growth
