BY SAKSHAM KHOSLA
Saksham Khosla is a first-year International Development concentrator at SAIS and an editor at SAIS Perspectives. He previously conducted research on social protection, governance, and financial inclusion in India at the Carnegie Endowment for International Peace and Carnegie India.
The World Bank’s World Development Reports (WDRs) are its flagship annual research publication: each draws upon a wide-ranging academic literature and new data to provide an in-depth analysis of an issue central to economic development. The World Development Report 2019: The Changing Nature of Work, released this month, examines how technology is transforming the skills and business models underpinning employment, and how governments can make investments in human capital and social protection to prepare their citizens for the future.
Given the direct relevance of the WDR 2019’s focus to SAIS Perspective’s 2018-19 theme of “Work in a Changing World,” editor Saksham Khosla interviewed Dr. Simeon Djankov, the director of the WDR 2019 and former deputy prime minister of Bulgaria, about the report’s key findings. This interview has been edited and condensed for clarity.
Perspectives: While writing the WDR, did you encounter any popular misconceptions among the development community about the changing nature of work and its implications for international development?
Dr. Djankov: When we started writing this study, there were at least four views that we thought our data would corroborate, but it did not. One of the values of the WDR 2019 is that we bring new data to some of these stylized facts in development.
First, there is a stylized fact that the so-called “fourth industrial revolution” is characterized by a much faster speed of technology adoption and hence productivity gains. We find that is not the case. So far, this has not yet resulted in very significant productivity gains anywhere in the world. So, it is not the pace of technology adoption, it is the scale with which this adoption affects particular sectors. That has lots of implications because you need the argument for much faster technology adoption to hold for much higher rates of unemployment. Once you do not have faster rate of technology adoption, then the next question is whether we are heading towards a future of work where people will be out of a job, replaced by robots. And we do not find that to be true.
Overall, the world is creating thirty to forty million jobs a year—so it is not about how many jobs are being created or displaced, but about geography. Where do these jobs go? We find that the vast majority of new jobs basically go to East Asian countries, like Vietnam, Laos, and Indonesia—and not to the rest of the world. So, the question of employment is more of a distributional issue rather than one of declining employment growth.
The third stylized fact that we show is incorrect is rising income inequality. We find that in the latest data, using standard measures of income inequality like Gini coefficients, over the past decade inequality has actually has risen in very few countries, like the US, UK, Australia, South Africa, Turkey, and Bulgaria. It’s really a handful among the 110 or so countries that we measure.
In a way, these three things go together. Once you show that technology is not at a different pace than before, then unemployment growth also cannot be at a different rate than before, and hence inequality also cannot be so dramatically different.
Another fact that is somewhat more related to advanced economies is the size of the gig economy. Some of the business school literature suggests 15-20% of the workforce in rich countries is heading towards a gig economy form of employer-employee relations. But we find that even in economies like the US and Germany, it is below half a percentage point of the workforce. In terms of aggregate implications for the job market, it is trivial. These are the four stylized facts that we thought were correct when we started this work and found that they were not.
Perspectives: Along with the distributional variation in new employment, we are also seeing a rising trend of deindustrialization. How might developing countries provide meaningful jobs at scale given that labor-intensive manufacturing is likely to play an increasingly smaller role in job creation?
Dr. Djankov: I think this is the most troubling finding of this WDR. If you asked any development economist how Africa, Latin America, or South Asia would develop, the traditional answer would have been that as East Asian countries become richer, then the jobs would move from there to India, Egypt, or Nigeria. We find in the report that this is not the case. Whatever manufacturing jobs are being created actually go into East Asia. Nothing flows from there to rest of the developing world. This is a very significant challenge for development economics. We do not see new jobs being created in high population-growth regions.
Then there are two questions: can you actually attract such jobs, and what do you need to do to attract them? Our answer is that you need to have public investment in human capital and basic infrastructure. Because what East Asia has that these other regions do not have is a highly educated workforce; the Human Capital Index that we developed shows that clearly. Even for the same per capita income, these countries have a larger highly educated population, and much better basic infrastructure such as roads, ports, and airports.
In some sense, Africa, Latin America, the Middle East and South Asia need to have the right balance of public investment. But that takes time. Can they do something else in the meantime? The report is not too optimistic in that regard because for particular sectors to develop, say online banking in Africa, you need broadband. Just for sub-Saharan Africa to have reasonable broadband coverage, we calculate that it would need approximately $5 trillion dollars of investment. Governments need to make some hard choices on where to start.
Perspectives: The WDR articulates a new social contract comprising more flexible labor regulations and higher levels of social assistance. How can this vision be realized in countries that have limited social protection, weak bureaucracies and state capacity, or inadequate tax bases to expand public spending?
Dr. Djankov: The vast majority of countries lack an efficient social protection system. So many people are working in informal jobs. For all practical purposes, social protection does not exist at all for the vast majority of the world’s population. Once we start there, we can say that the current system works if you have a formal and rich economy but it does not work otherwise. Can you have an alternative to that?
One optimistic answer in our report is that yes, you can. New technologies, especially e-payment technologies, have allowed many middle-income and poor countries to reach more people. And since the government and the private sector have this data, a new social insurance system can be developed with it, and also perhaps a new minimum income. This is a technological possibility that did not exist five years ago and certainly not ten years ago.
The next question is: how do you fund it? We have a spectrum of suggestions on how government can collect more revenue. Governments in emerging markets can collect more local taxes like property tax in larger cities and impose carbon taxes, and use those proceeds to subsidize basic incomes.
At the global level, many digital platforms actually do not pay taxes anywhere – not in their own country, nor in any other country where they conduct their operations. Globally, how can we come up with a system that collects corporate taxation from everybody? Politically, this is going to be very difficult because you need consensus for international tax treaties. At least in the near-term, we do not see that happening. Instead, we demonstrate how some large emerging economies like Korea, Russia, Argentina, and Brazil have recently started legislating such taxation with relative success. We suggest that the world more closely at that option.
Perspectives: With the growing power of digital platforms, are we seeing the emergence of a new political economy of development?
Dr. Djankov: We know, in principle, the components of a new social contract. We document that you need to allocate resources to basic education and health to ensure early childhood development, and then from there pay attention to the acquisition of new skills to prepare citizens for the work of the future. What we lack—and that is where this relates to political economy—are the sources of financing such expenditure.
It is not coincidental that there is no international treaty on taxing platforms because they have significant political power both in particular countries and globally. These platforms have managed in countries like the US and the UK to say to the regulators: you cannot regulate us because we are so new and so entrepreneurial that you do not know what to regulate. At some level this is true, but how long can you not regulate?
This is the dimension of political economy that standard measures of income inequality and also power do not capture because they do not consider wealth inequality. The reasons they do not do this is that the super-rich hide their wealth in tax-free locations. We simply do not know where this wealth this and how much it is. But this wealth corrupts the global political process, so the real political economy question is how can different countries around the world agree that the top 0.1% cannot hide their profits and gains in tax-free locations. In the WDR, we document that annually this accounts for something like 6-8% percent of global GDP, which is huge. If this is equally divided around the world, you suddenly have enough for health and enough for education. You need, however, either a large regional bloc like the European Union or ASEAN to say that all of us are going to regulate in this particular way—or for the WTO, IMF, or World Bank to basically establish a global standard. Otherwise the lobby power of the platform companies is just too large for any one country.