When did the world start getting more equal? The Great Divergence and the Great Convergence.

Richard Baldwin, 瑞士洛桑国际管理发展学院 (IMD) - 商学院 16 August 2024, Factful Friday.

While the world income distribution is incredibly unequal, inequality has been falling since about 1990 when the Great Divergence flipped over into the Great Convergence. Most of the extra equality, however, comes from a convergence of inter-nation differences. Income inequality within many nations, especially the US and Britain, is rising, but not fast enough to offset the massive inter-nation convergence in average incomes.

Introduction.

Income is very unevenly distributed across the planet. For instance, more than 80% of humans earn incomes that are below the US poverty line—even adjusting for international price differences. But since the second big transformation of globalization happened around 1990, incomes have become more equal.

When did rising income equality start? Excellent question and the topic of today’s Factful Friday.

The cheat sheet chart.

The chart below shows the Gini Coefficient of inequality among nations. This ranges from zero when there is perfect equality (everyone earns the same) to 100 when one person has all the income.

The chart shows that inequality rose steadily from the opening of modern globalization in 1820 (O’Rourke and Williamson 2020) up to 1988. This is what is often called the Great Divergence (Pomeranz 2000). This was the phase where economists learned to think about modern globalization and its impact on nations and within nations.

Inequality declines after 1988, first slowly and then rapidly. The level of global inequality is now back to where it was at the end of the 1800s, around the time that the Second Industrial Revolution started. This is what I called the Great Convergence in my 2016 book of the same name (Baldwin 2016).

The radical change in the evolution of inequality and its global scope very strongly suggests that globalization is not a single process—a point that I’ve been flaming on about since Baldwin (2006).

Cross-cutting inequality.

The next chart presents a pair of fascinating facts.

• As inter-nation inequality was rising from 1820 to 1988, intra-nation inequality was stable or falling.
• When inter-nation inequality was falling from 1988 to 2018, intra-nation inequality was rising.

There are some explanations for this based on the nature of the automation that I put forth in my 2019 book, The Globotics Upheaval: Globalization, Robotics, and the Future of Work. The basic idea is that from 1870 to 1970, automation meant mechanization and it gave more power to manpower in industry, but did little to help higher-education, higher-income workers. Clerks in 1900 were working in ways that weren’t too different to those in 1800, but factory workers in those two years worked in radically different ways and with radically different tools. But importantly, the mechanization of industry did not—as industrial robots are doing today—remove humans from the process. Today’s automation is creating better substitutes for hand workers (e.g., industrial automation), and better tools for head workers (software and hardware, the Cloud, etc.). That’s why automation is associated with rising inequality.

This phenomenon is sometimes called the skills twist, where technology favors particular sets of skills. The assertion is that in the 1820-1988 period, the skills twist favored manual workers, while in the 1988-2018 period it favored mind workers.

Changes across vs within nations.

The Gini Coefficient is an old workhorse in the study of inequality—tried and true. If you want a single number to summarize global inequality, it’s likely the leading contender. But using a single number tempts even very careful thinkers to over generalize.

The next pair of charts show the income changes with much finer resolution (but only for the post 1988 period). This is the famous Elephant chart that Branko Milanovic presented almost ten years ago in his brilliant 2016 book, Global Inequality: A New Approach for the Age of Globalization (Milanovic 2016). This so-called “growth incidence’ chart looks at how much income growth was enjoyed by people at various income levels. This looks at all humans, one by one, ignoring their nationality. To keep things manageable, the individuals are lumped together into twenty groups.

In other words, the chart shows how much people’s incomes grew according to how rich they were in 1988. Specifically, the chart lays out the world population on the horizontal axis from the poorest to the richest. The poorest are listed on the left side and the richest on the right. For example, those who were halfway up the income distribution in 1998 would be included in the point labeled “50” (short for the fiftieth percentile).

The aim is to see what drove the greater global equality. If every income class grew by the same amount, there would be no change in global inequality. One thing is plain as a pig in a pool hall; that is not what happened.

Globalization’s second unbundling, i.e. the Great Convergence, had very uneven effects across the global income distribution.

• A global middle class was created during the Great Convergence; this is why globalization is popular in most emerging economies.

This can be seen in point A being so high. People who started in the middle of the world income rankings in 1988 experienced growth that was faster than all segments of the income distribution. In other words, the medium poor were getting richer faster than the rich, so inequality fell.

• The really rich also did well (point C on the chart); this is one reason Davos men and women think globalization has been a fine thing since 1990.

The groups that suffered—in the sense that their incomes stagnated—were people at point B in 1988—basically the middle class in rich nations.

As I explain in Baldwin (2016), these outcomes are in line with the basic predictions of my three cascading constraint view of globalization. In particular, globalization’s second unbundling involved G7 manufacturers moving their firm-specific knowledge to a handful of developing economies and combining it with low wage labor. This sparked industrialization and growth in the emerging markets on the receiving end of the industrial knowhow, but it undermined the monopoly that G7 labor used to have on the use of G7 technology. Thus, globalization offshored middle class jobs, lowered the prices of the things they were still made in advanced economies, and created indirect wage competition for factory jobs. This impact was amplified by the fact that ICT also was creating better substitutes for manual workers in G7 countries. The double-whammy—extra competition with low-wage workers abroad and industrial robots at home. This helps explain why the MAGA voters in America are so mad.

A second point about the growth bulge at point B. The middle of the income distribution enjoyed good growth for one of two reasons. Either they were in one of the rapidly industrializing nations (China, India, Korea, etc.), or they were in emerging markets that experienced commodity-led income takeoffs. The global elite won since the global value chain revolution, and the ICT revolution more broadly, allowed them to sell their know- how to a wider audience. The people who were poor and stayed poor.

This pattern disappeared around 2008.

All change from 2008.

The chart below shows that this second-unbundling outcome has changed since 2008—which is when trade in goods started to deglobalize. This is shown with the orange line. What we see here is that as the offshoring of manufacturing stages came to a slowdown, the forces of automation and globalization allowed the poor to start to catch up in the sense that they experienced faster growth in this period than their richer citizens.

The new growth-incidence curve (in orange) is good news for global equality as it is the poorest that are enjoying the fastest growth. More research is needed to figure out what accounts for this rapid growth of the poorest. In general, people who are that poor are not very connected to the global economy as they tend to work in non-trade services, or subsistence agriculture. Maybe it has to do with the fall in commodity prices that happened when the commodity Super-cycle ended?

It is not too hard to imagine what changed. Around 2008, trade in goods, including manufactured goods, slowed around that time so the globalization element of the shock diminished. Moreover, the automation shock was coming off the boil as industrial robots were already widely spread.

Moreover, the offshoring phase—which, in my view, was critical to the industrial knowhow transfer driving the Great Convergence—clearly slowed or stopped in 2014.

Summary and Concluding remarks.

So, the world started getting more equal on the whole at the end of the 1980s, but that was driven by equalization of national average incomes. Inside the advanced economies, the opposite was happening. Countries like the US were becoming more unequal internally as the world was becoming more equal externally.

The fact that the external and internal inequality went in opposite directions before and after 1990 calls for a lot of explanation. I’ve pieced together a number of threads of logic on the cross-cutting trends in Baldwin (2019), but I’ll put them in some other Factful Friday.

And that’s it for today’s factful Friday!

References

(Warning: some of the details of some of these references may have been invented by ChatGPT, but the research behind them is real.)

Baldwin, R. (2019). The globotics upheaval: Globalization, robotics, and the future of work. Oxford University Press.

Baldwin, R., & P. Martin (1999). Two waves of globalization: Superficial similarities, fundamental differences. In H. Siebert (Ed.), Globalization and labor (pp 3–58). Mohr Siebeck Verlag.

Bénétrix, A. S., O'Rourke, K. H., & Williamson, J. G. (2012). The spread of manufacturing to the periphery 1870-2007: Eight stylized facts (Working Paper No. 18221). National Bureau of Economic Research. https://www.nber.org/papers/w18221

Ebrey, P. B., Walthall, A., & Palais, J. B. (2009). East Asia: A cultural, social, and political history (2nd ed.). Houghton Mifflin.

Fairbank, J. K., & Goldman, M. (2006). China: A new history (2nd ed.). Harvard University Press.

Frank, A. G. (1998). Reorient: Global economy in the Asian age. University of California Press.

Gernet, J. (1996). A history of Chinese civilization (2nd ed.). Cambridge University Press.

Keay, J. (2000). India: A history. Grove Press.

Keay, J. (2009). China: A history. Basic Books.

Maddison, A. (2007). Contours of the world economy 1-2030 AD: Essays in macro-economic history. Oxford University Press.

Milanovic, Branko (2016). Global Inequality: A New Approach for the Age of Globalization (Cambridge: Harvard University Press, 2016).

Milanovic, Branko (2024). “The three eras of global inequality, 1820–2020 with the focus on the past thirty years” (World Development 177 (2024) 106516.

O'Rourke, K. H., & Williamson, J. G. (2002). When did globalization begin? European Review of Economic History, 6(1), 23-50. https://doi.org/10.1017/S1361491602000023

Pomeranz, K. (2000). The great divergence: China, Europe, and the making of the modern world economy. Princeton University Press.

Spence, J. D. (2013). The search for modern China (3rd ed.). W. W. Norton.

Wallerstein, I. (1974). The modern world-system (Vol. 1). Academic Press.