Inequality in the Digital Era by Zia Qureshi 3

WORK IN THE AGE OF DATA

BBVA OPENMIND DATA, IDEAS, AND PROPOSALS ON DIGITAL ECONOMY AND THE WORLD OF WORK BBVAOPENMIND.COM Work in the Age of Data 2 Inequality in the Digital Era by Zia Qureshi 3

We are living in an era of mounting socie- tal discontent and political divisiveness. In many countries, social disaffection with economic outcomes is up sharply, stoking pop- ulist and nationalist sentiment. Increasing income inequality is one important reason behind this sociopolitical tumult. We are also living in an era of major technological change, led by the digital revolution. Today’s technological changes—advances in computer systems and software, mobile telephony, digital platforms, robotics, cloud computing, artificial intelligence, and cyber-physical systems—are arguably unpar- alleled in their scope and speed. Inequality in Are these two megatrends of our time connected? The answer is yes. Digital technologies are reshaping the world of business the Digital Era and work in profound ways. Policies have been slow in adapting to the new dynamics. The interaction between technological change and market conditions as influenced Zia Qureshi by the prevailing policy environment has been a key factor driving income inequality higher. Disruptions caused by technological change have added to business and worker anxieties. A more unequal distribution of income, however, is not an inevitable consequence of a digitizing world. Outcomes that are more inclusive are certainly possible with better, more responsive policies.

Rising Income Inequality amid Booming Digital Technologies

Income inequality has risen in practically all major advanced economies since the 1980s, a The digital revolution is transforming period of a rising boom in digital technologies (fig. 1). It has risen particularly sharply at the economies. Potential economic gains from top end of the income distribution. Wealth digital technologies are enormous, but with new inequality is still more acute, roughly twice opportunities come new challenges. Within as high as income inequality. The increase in economies, income and wealth inequalities inequality has been especially marked in the have risen as digitization has reshaped markets United States. Over a two-decade period end- and the world of business and work. Inequalities ing in 2015, US disposable income inequality, have increased between firms and between as measured by the broadest measure of in- workers. The distribution of both capital and equality (the Gini Index), increased by more labor income has become more unequal, than 10%. The income share of the richest 1% and income has shifted from labor to capital. has more than doubled since the early 1980s, to around 22%. The share of the top 1% in Technological change, however, is not the sole wealth has risen to around 40%. Those with reason for the rising inequalities. Policy failures middle-class incomes were squeezed and the have been an important part of the story. typical worker saw largely stagnant real wages Policies will need to be more responsive to the over long periods. Higher inequality has been new dynamics of the digital economy to achieve associated with a decline in intergenerational outcomes that are more inclusive. economic mobility (Chetty et al., 2017). Work in the Age of Data 4

This article focuses on advanced econo- full potential to boost productivity (Brook- at the technological frontier. However, it has mies, but the rise in income inequality is not ings Institution and Chumir Foundation, slowed considerably in the vast majority of confined to this group. In emerging econo- 2019). Developments in income distribution other, typically smaller firms, pulling ag- mies, income distribution trends are more and productivity have been linked by shared gregate productivity growth lower. Between mixed but many major emerging economies dynamics. 2001 and 2013, in OECD economies, labor also witnessed a rise in income inequality. productivity among frontier firms rose by In the two largest emerging economies, around 35%; among non-frontier firms, the China and India, inequality has increased Transformations in the World of Business increase was only around 5% (Andrews et appreciably.1 al., 2016).2 Aggregate labor productivity In the cauldron of political debate, much Digital technologies are altering business growth in OECD economies in the decade of the blame for the rise in income inequal- models and how firms compete and grow. to 2015 was only about half of that in the pre- ity and underlying business and job dislo- They are reshaping market structures. ceding two decades. The growing inequality cations is heaped on globalization—often Change affects all markets, from produc- in productivity performance between firms from both ends of the political spectrum. tion and commerce to finance. The manner not only depressed productivity growth, but The backlash against globalization threat- in which the new technologies deploy across also caused income disparities to rise. ens a retreat into economic nationalism and industries and firms has important impli- A weakening of competition is one im- inward-looking policies. Globalization has, cations for their economic impact and the portant reason for these adverse productiv- indeed, been a factor behind rising inequal- distribution of rewards. ity and distributional dynamics. Barriers to ity. However, a much bigger factor has been competition and related market frictions are technological change. Uneven Diffusion of New Technologies and preventing a broader diffusion of the new Not only is the proverbial economic pie Widening Gaps between Firms technologies and causing a persistent rise in being shared more unequally, it has also been How technological innovation diffuses productivity and profitability gaps between growing more slowly, adding to social dis- within economies and interacts with mar- firms. Evidence for OECD economies shows content. Paradoxically, productivity growth ket conditions matters greatly for both pro- that in industries less exposed to competi- in major economies has slowed rather than ductivity growth and income distribution tion, technological innovation and diffusion accelerated during the boom in digital tech- (Comin and Mestieri, 2018; OECD, 2018a; are weaker, inter-firm productivity diver- nologies. This has slowed overall economic Aghion et al., 2019). The benefits of the gence is wider, and aggregate productivity growth. Research finds that the same inter- new technologies have not been diffusing growth is slower (Cette et al., 2016; Égert, action between technological change and widely across firms. They have been cap- 2016). Studies of the United States and Eu- policy failures that contributed to higher tured for the most part by a relatively small ropean economies also find that lower com- income inequality also explains why the number of larger firms. Productivity growth petitive intensity in markets depressed in- new technologies have not delivered their has been relatively strong in leading firms vestment in new productive capital, as firms

1975 1980 1985 1990 1995 2000 2005 2010 2015

0.38

0.34

0.30

0.26 Gini Index of Disposable Income of Gini Index

Year USA Italy United Kingdom Canada Spain France Fig. 1. Rising income inequality: major advanced economies Japan Germany (Source: OECD Income Distribution Database) Australia Inequality in the Digital Era by Zia Qureshi 5

Between 2001 and 2013, wielding increased market power invested more unequal, with a relatively small num- in OECD economies, labor less and made a lot more on existing capi- ber of firms reaping supernormal profits. In productivity among “frontier tal through higher markups and increased the United States, for example, the ninetieth firms” rose by around 35%; stock buybacks (Gutiérrez and Philippon, percentile firm earned a return on invested among other firms, the 2017; Égert, 2018). capital reaching around 100% in 2014, which increase was only around 5% The erosion of competition is reflected was more than five times the return earned in a variety of indicators: rise in market con- by the median firm; this ratio was around 2 centration in industries, higher markups about twenty-five years ago (Furman and showing increased market power, and cor- Orszag, 2018). The uneven distribution of porate ossification with declining business returns on capital was particularly marked dynamism as measured by new firm forma- in technology-intensive industries. There tions. These trends are observable broadly is also evidence of low churning among across advanced economies but have been high-return firms, with a large proportion particularly marked in the United States. of such firms persistently achieving high The share of top four US companies in total rates of return. sales has risen since the 1980s in each of the Markets have shifted toward more mo- major sectors covered by the US Economic nopolistic structures, giving rise to higher Census (Autor et al., 2017). The rise in mar- economic rents (Krugman, 2016; Stiglitz, ket concentration is greater in industries 2016; Summers, 2016). The share of “pure that are more intensive users of digital tech- profits” or rents (profits in excess of those nologies. Markups over marginal cost for US under competitive market conditions) in publicly traded firms have nearly tripled, total income in the US economy rose from with the rise concentrated in high-markup 3% in 1985 to 17% in 2015 (Eggertsson et al., firms gaining market share (De Loecker et 2018). As monopoly profits boosted the mar- al., 2018). The share of young firms (five ket value of corporate stocks and produced years old or less) in the total number of US large capital gains, the share of total US stock Establishments like the legendary firms has declined from about one-half to market value reflecting monopoly power department store Harrods in London, which one-third (Decker et al., 2017). (“monopoly wealth”) rose from negligible had previously replaced small retailers, are now losing market share to online With increased market power, the dis- levels to around 80% over the same period megastores tribution of returns on capital has become (Kurz, 2018). Work in the Age of Data 6

There has been much Winner-Takes-Most Dynamics and Competi- three times as fast as economic activity in innovation in financial tion Policy Failures recent decades. The rapid financialization services based on the new Why is market power rising and competition compounded the inefficient and unequal technologies, but much of it weakening? First, the new technologies are outcomes resulting from decreased compe- has focused on areas such as contributing to increased market concen- tition in markets (OECD, 2015; Philippon, trading and asset management tration by altering competition in ways that 2016). In the credit boom that preceded the that primarily benefit the well- produce “winner-takes-most” outcomes. global financial crisis, the lion’s share of the off Digital technologies offer first-mover ad- credit went to households rather than firms, vantages, scale economies, network effects, boosting stock and real estate markets rath- and leverage of “big data” that encourage the er than productive investment—an alloca- The largest US firm in rise of dominant firms—and globalization tion of credit with negative implications for 2017, Apple, had a market reinforces the scale economies by facilitat- growth, stability, and income distribution. capitalization forty times as ing access to markets worldwide. The rise There has been much innovation in financial high as that of the largest US of “the intangible economy,” where assets services based on the new technologies. A firm in 1962 (AT&T), but its such as software and intellectual property large part of it, however, has been focused total employment was only matter more and more for economic success, on areas such as trading and asset manage- one-fifth that of the latter has been associated with a stronger tenden- ment that primarily benefit the well-off and cy toward the emergence of dominant firms do not have first-order effects on economic (Haskel and Westlake, 2017). Digitization also productivity. allows firms controlling big data to extract Rewards in the financial sector rose more of the consumer surplus through in- sharply relative to the real economy. In the creasingly sophisticated algorithmic pricing United States, the financial sector captured and customization of offerings. an outsize share of profits—35–40% of all The winner-takes-most dynamics have corporate profit in the years leading to the been most marked in the high-tech sectors, financial crisis. A sizable part of these high as reflected in the rise of “superstar” firms profits reflected rents in an increasingly con- such as Facebook and Google. Increasingly, centrated sector: the top five banks’ share of however, they are affecting economies more banking assets increased from 25% in 2000 broadly as digitization penetrates business to 45% in 2014. In European countries, fi- processes in other sectors, such as transpor- nancial sector workers on average account- tation, communications, finance, and com- ed for one in five of the top 1% of earners merce. In retail trade, for example, the big even though they accounted for only one box stores, which previously had replaced in twenty-five of the total workforce (Denk, mom and pop outlets, are now losing market 2015). Financial wealth boomed but bene- share to online megastores such as Amazon. fited mainly those at the top; in the United Second, failures in competition policies States, the top 1% of the wealth distribution have reinforced the technology-driven dy- held half of stock and mutual fund assets in namics producing more concentrated mar- 2013, and the top 10% held more than 90% ket structures. These include weaknesses in (Wolff, 2014). antitrust policies, flaws in patent systems that act as barriers to a wider diffusion of innovations, and regulatory acts of omission Transformations in the World of Work and commission (deregulation unsupported by competition safeguards, and regulations Just as transformations in the world of busi- that restrict competition). Related factors ness caused by digitization-driven techno- include an increase in overlapping owner- logical change have been a key factor influ- ship of companies that compete by large encing the distribution of capital income, institutional investors, rise in rent seeking, technology-driven transformations in the and firm behavior showing greater adeptness world of work have been a key factor influ- in erecting barriers to entry through product encing the distribution of labor income. differentiation and other means. Rising Wage Inequality and Falling Labor Financialization Income Share Digital technologies have been instrumental Across OECD economies, increased inequal- in the financialization of economies, rein- ity in firm productivity and profitability is forcing the impetus from financial sector de- mirrored by increased inequality in labor A worker grinds metal at a workshop in regulation. In OECD economies, credit and incomes. As profitability gaps widened be- Mumbai other financial intermediation has grown tween firms, so did wage gaps. Rent sharing Inequality in the Digital Era by Zia Qureshi 7

also contributed to wider wage differences be- power strengthened, worker bargaining pow- temporary or part-time contracts and own-ac- tween firms. Better-performing firms reaped er weakened with a decline in unionization count employment. a higher share of total profits and shared and erosion of minimum wage laws. As the demand for skills has shifted, sup- part of their supernormal profits with their These developments reinforced the effect ply has been slow to adapt. Education and workers. Increased fissuring of the workplace of labor-substituting technological change on training have been losing the race with tech- through outsourcing played a role as well, the distribution of income between labor and nology (Goldin and Katz, 2008; Autor, 2014). with noncore activities typically employing capital. Production shifted toward firms and Shortages of higher-level skills demanded by low-skill workers farmed out to other firms, processes using more capital (tangible and in- the new technologies have prevented a broad- cutting such workers from the rent sharing. tangible) and less labor. The largest US firm in er diffusion of the innovations across firms. Between-firm wage inequality rose more in 2017 (Apple) had a market capitalization forty Workers with skills complementary with industries that invest more intensively in dig- times as high as that of the largest US firm the new technologies have been clustered ital technologies. Overall, wage inequality has in 1962 (AT&T) but its total employment was increasingly in leading firms at the techno- risen sharply in the past couple of decades only one-fifth that of the latter (West, 2018). logical frontier. Across industries, skills mis- and much of that rise is attributable to in- The shift of income from labor to capital in- matches have increased: in OECD countries, creased wage differences between firms (Song creased overall income inequality, as capital on average around one-quarter of workers et al., 2019). ownership is highly uneven.3 report a mismatch between their skills and While workers in firms at the technolog- International trade and offshoring also those required by the job (Adalet McGowan ical frontier earned more than those in oth- contributed to the shift in income toward and Andrews, 2017). er firms, gains from higher productivity at capital by putting downward pressure on Imbalances between skills demand and these firms were shared unevenly, with wage wages, especially of lower-skilled workers in supply have fueled income inequality, by growth lagging behind productivity growth. tradable sectors. Overall, research shows that increasing the wage premia on higher-level Wages rose in the better-performing firms but globalization has played a significant role in skills (Autor, 2014; Hanushek et al., 2015). by less than the rise in productivity. For most the decline of the labor-income share. How- The skill premium rose in all major econo- other firms, limited wage growth reflected ever, it also shows that globalization’s role has mies, especially over the 1980–2000 period. limited productivity growth, although even at been much smaller than that of technological The rise has been particularly sharp in the these firms wage growth tended to fall short change and related outcomes. IMF research United States: those with a postgraduate of the meager gains in productivity (OECD, finds that, in advanced economies, techno- degree could expect to earn around 215% of 2018b; Schwellnus et al., 2018). In the United logical change has contributed about twice the wages received by those with only a high- States, net labor productivity increased by as much as globalization to the decline in the school education in 2016, compared to around 72% between 1973 and 2014, while real hour- labor-income share (IMF, 2017a). 155% in 1980.5 The rise in nonstandard work ly compensation of the median worker in- arrangements imparted more flexibility to creased by only 9% (Bivens and Mishel, 2015). Shifts in Labor Demand, Job Polarization, the labor market. However, it probably also The decoupling of wages from produc- and Skills Mismatches contributed to increased earnings inequality tivity contributed to a shift in income distri- Technology has been the dominant force as nonstandard jobs (especially at lower skill bution from labor to capital. In the past cou- in reshaping the demand for labor. Digital levels) typically paid less than standard jobs. ple of decades, most major economies have technologies and automation have shifted de- experienced both increasing inequality of mand toward higher-level skills. Globalization labor earnings and declining labor-income has exerted pressure in the same direction. Weakening Redistributive Role of the shares. In the United States, for example, Demand has shifted, in particular, away from State the percentage share of labor in total income routine, middle-level skills that are more vul- dropped from the mid-60s around 2000 to nerable to automation, as in jobs like clerical As technological change interacted with de- the mid-50s around 2015. work and repetitive production. Job markets velopments in product, financial, and labor Increased market concentration has have seen an increasing polarization, with the markets to drive income inequality higher, played a role in the shifting of income from employment share of middle-skill jobs falling making the distribution of both capital and labor to capital as it reallocated labor within and that of higher-skill jobs, such as techni- labor income more unequal and shifting in- industries to dominant firms with supernor- cal professionals and managers, rising. The come from labor to capital, the state’s role in mal profits and lower labor-income shares employment share of low-skill jobs has also alleviating the inequality of market incomes (Autor et al., 2017). Dominant firms not only increased but mainly in nonroutine manual arising from the interplay of these forces acquired more monopoly power in product jobs in services such as personal care that are weakened. In advanced economies, taxes and markets to increase markups and extract hard to automate. Between 1995 and 2015, the transfers reduce market income inequality higher rents but also monopsony power share of middle-skill jobs in total employment on average by about one-third: in 2015, the to dictate wages in the labor market (CEA, fell by about 9.5 percentage points in OECD average Gini Index for disposable income 2016; Azar et al., 2017). A new phenomenon economies on average, while the shares of in these economies was 0.31 compared with has been the fast-expanding digital labor high-skill and low-skill jobs rose by about 7.5 0.48 for market income. Between 1985 and markets—online jobs platforms such as Task and 2 percentage points, respectively.4 A con- 1995, fiscal redistribution offset about 60% Rabbit and Amazon Mechanical Turk. Here current development has been the rise of the of the increase in market income inequality too, employer concentration has been high “gig” economy, with more workers engaged in advanced economies. Between 1995 and (Dube et al., 2018). While employer market in nonstandard work arrangements, such as 2010, it hardly offset any (OECD, 2016). Work in the Age of Data 8

IMF research finds that, Fiscal redistribution declined because Revitalize Competition for the Digital Age in advanced economies, of reduced progressivity of personal income Competition policies should be revamped for technological change has taxes and lower taxes on capital as well as the digital age to ensure that markets contin- contributed about twice as tighter spending on social programs as ue to provide an open and level playing field much as globalization to the countries took steps to rein in fiscal defi- for firms, keep competition strong, and check decline in the labor-income cits and rising public debt. In OECD econo- the growth of monopolistic structures. This share mies, the average top personal income tax includes regulatory reforms and stronger an- rate fell from 62% in 1981 to 35% in 2015. In- titrust enforcement. The winner-takes-most Innovations such as ternational tax competition resulting from dynamics associated with digital technologies capital mobility led to a large fall in corpo- is raising new challenges for competition poli- mobile financial services, rate income tax rates as well. The average cies, including how to address market concen- digital platforms, equity corporate tax rate in advanced economies tration resulting from tech giants that resem- crowdfunding, and blockchains fell from around 45% in 1990 to 26% in 2015 ble natural or quasi-natural monopolies. Once have much potential (IMF, 2017b). in dominant positions, firms can entrench themselves by erecting a variety of barriers to entry and taking over rising competitors. The Harnessing Technology for More beneficiaries of an open, competitive system Inclusive Growth often work to close the system and stifle competition, necessitating reform to “save capital- The rise of the digital economy has pushed ism from the capitalists” (Rajan and Zingales, income inequality higher. At the same time, 2003; Krugman, 2015). Competition policy the potential of the new technologies to spur also needs to become more global to address productivity growth has not been fully re- cross-border issues posed by multinational alized. However, this should not provoke tech giants that affect market concentration despair, much less a negative backlash. and competition in many countries. Most dynamic economic change is inher- Proprietary agglomeration of data, as in ently disruptive, creates winners and losers, digital platforms, is an increasingly import- and entails difficult transitions. Technolo- ant source of competitive advantage. Regu- gy—and globalization—are no exceptions. lations pertaining to digital platforms, own- They are key forces that drive economic ership of data, how user data are handled, progress. Advances in digital technologies and privacy protections matter increasingly hold great promise to boost productivity for competition. There has been more action and economic growth, create new and bet- on this agenda in Europe than in the United ter jobs to replace old ones, and enhance States, an example being the General Data human welfare. Policies have a crucial role Protection Regulation (GDPR) introduced in to play in ensuring that the potential gains Europe in 2018. are captured effectively and inclusively— Enhancing competition is also important by improving the enabling environment for in financial markets, to address issues such as firms and workers to broaden access to the increased concentration, interconnectedness, new opportunities that come with change and rent seeking. It would spur better use of and to enhance capabilities to adjust to the advances in digital technology to expand the new challenges. Unfortunately, policies range of financial services and reduce their and institutions have been slow to rise to cost, open new gateways to entrepreneur- the new challenges of the digital economy. ship, and democratize access to finance. In- Indeed, they have often exacerbated rather novations such as mobile financial services, than ameliorated the outcomes. digital platforms, equity crowdfunding, and Policies to reduce inequality are often blockchains have much potential. Young Fin- seen narrowly in terms of redistribution— Tech firms are in the vanguard in the appli- tax and transfer policies. However, there is cation of such innovations. A challenge for a much broader policy agenda of “predistri- policy-makers is to foster the growth of these bution” that can make the growth process new entrants into the financial industry while itself more inclusive. Much of the reform managing associated risks. agenda to achieve more inclusive outcomes from technological change is also an agen- Improve Innovation Ecosystem for Wider da to achieve stronger growth outcomes, Technology Diffusion given the linked dynamics between the re- Intellectual property regimes need to be bet- cent rise in inequality and the slowdown in ter balanced so they reward innovation but productivity. also foster wider economic impacts. “The Inequality in the Digital Era by Zia Qureshi 9

copyright and patent laws we have today look (Shambaugh et al., 2017). This underscores Many breakthrough more like intellectual monopoly than intel- the need to revitalize public research pro- innovations developed lectual property” (Lindsey and Teles, 2017). grams and ensure broad access to their dis- commercially by private firms Arguing that patents are locking in incum- coveries. Many breakthrough innovations originate from government- bents’ advantages rather than spurring the developed commercially by private firms supported research. Recent hoped-for bursts of innovation, some have originate from government-supported re- examples include Google’s called for a complete dismantling of the pat- search. Recent examples include Google’s basic search algorithm, key ent system (Boldrin and Levine, 2013). That basic search algorithm, key features of Apple features of Apple smartphones, would be too radical an approach. What is smartphones, and even the Internet itself. and even the Internet itself needed is a fundamental reexamination, to Governments should consider how to give change excessively broad or stringent pro- taxpayers a stake in such profitable outcomes tections, align the rules with current reali- from publicly supported research, not least ties, and give freer rein to competition. Long to replenish public R&D budgets. Here, the patents may have been appropriate for phar- tax system has an important role to play. maceutical innovations, which involve pro- Infrastructure that supports digitization tracted and expensive testing, but the case is should be strengthened. Despite progress, less clear for advances in digital technologies the digital divide remains wide. Even in ad- that have much shorter gestation periods and vanced economies, population remaining typically build on previous innovations in an offline could be as high as one-fifth (ITU, incremental fashion. 2016). Most sectors of the US economy are Government research and development less than 15% as digitized as the leading sec- (R&D) spending focuses on supplying the tors (McKinsey, 2015). public good of basic research, which often produces knowledge spillovers that benefit Invest in Skills for a Changing World of Work the economy at large. Yet, it has been declin- Advances in digitization, robotics, and artifi- An underwater room installed by the Airbnb ing. In the United States, government R&D cial intelligence have led some to draw up dire online accommodation platform at the spending has fallen from 1.2% of GDP in the scenarios of massive job losses from automa- Aquarium of Paris for a contest in which early 1980s to half that level in recent years tion (a “robocalypse”). However, experience winners spent a night sleeping with sharks Work in the Age of Data 10

The wealth dynamics of recent with past major episodes of automation shows keep workers in existing jobs, to forward-look- decades paint a picture of that as technological change made some old ing policies that encourage reemployment, private riches and public jobs redundant, it generated new ones by including innovative unemployment/wage poverty. While private wealth creating new roles and tasks and spurring insurance mechanisms, retraining, and place- has soared, public wealth has economic growth. How technological change ment services. declined, hobbling the capacity impacts employment must be seen as a dy- Other barriers to worker mobility and of public policy namic adjustment process of old jobs giving competition in labor markets, such as the way to new ones (Acemoglu and Restrepo, ever-increasing professional licensing re- 2018; World Bank, 2019). Looking ahead, not quirements and noncompete covenants in only will the skill needs of jobs continue to worker contracts, should also be addressed. evolve, but the composition of employment Well-functioning labor market institutions— will evolve as well, with more people work- collective bargaining, minimum wage laws, ing independently—including as microen- labor standards—are important to ensure trepreneurs in an expanding “crowd-based that workers get a fair share of economic re- capitalism” enabled by digital platforms, as turns, especially at a time of rising market exemplified by Uber and Airbnb (Sundarajan, power of dominant firms. 2016; Brynjolfsson and McAfee, 2017). Social contracts will need to be over- The main issue is that the nature of work hauled. Benefits such as pension and health is changing, and the main policy challenge is care, traditionally based on formal long-term to equip workers with nonroutine, creative, employer-employee relationships, need and higher-level skills that the new technol- to be made more portable and adapted to ogies demand and to support workers during evolving work arrangements, including the the adjustment process. Traditional formal expanding gig economy. Here, several pro- education must be complemented with new posals have been put forward, including a models and options for reskilling and lifelong universal basic income currently being pi- learning. As the old career path of “learn- loted in some jurisdictions, a negative in- work-retire” gives way to one of continuous come tax up to a certain income threshold, learning—a process reinforced by the aging and social security accounts that pool work- of many economies’ workforces—the avail- ers’ benefits and are portable across jobs. ability and quality of continuing education Reform options will need to be considered must be scaled up. This will demand innova- in a context where many social security sys- tions in the content, delivery, and financing tems already face financial sustainability of training, including new models of pub- challenges. lic-private partnership. It will involve exper- Pursuing labor market and social pro- imentation, and learning from what works, tection reforms as a package will have the such as the apprenticeship system in Ger- advantage of capturing reform synergies many. The potential of technology-enabled and easing the adjustment for workers. solutions, such as online learning platforms, For example, in 2017, France implemented must be harnessed, supported by a stronger reforms to its job protection laws to boost foundation of digital literacy. labor market flexibility combined with the A strong commitment to improving access introduction of a portable “personal activ- to affordable and quality education, including ity account” that enables workers to accrue skills upgrading and retraining, for the eco- rights to training across multiple jobs. nomically disadvantaged is also vital. Even in an advanced economy such as the United Reform Tax Systems States, almost two-thirds of workers do not Tax policy is often seen as presenting have a college degree. Gaps in higher educa- trade-offs between efficiency and growth tion attainment by family income level have on the one hand and equity on the other. widened rather than narrowed (Turner, 2017). Trade-offs do exist, but there are win-win opportunities for reform. In labor-income Revamp Labor Market Policies and Social taxation, reducing the tax wedge for low- Protection wage workers through greater use of op- Labor market policies and social protection tions such as earned-income tax credit arrangements must be reformed to improve can boost labor force participation as workers’ ability to change jobs. This means well as improve distributional outcomes. shifting the focus from backward-looking pol- Countries may consider shifting part of icies, such as the stringent job protection laws the financing of social benefits to gen- in many European economies that seek to eral tax revenue to avoid overburdening Inequality in the Digital Era by Zia Qureshi 11

social security contributions and labor- will need to be more responsive to change, income taxation (OECD, 2017). Such a shift which will only intensify as advances in ar- in financing may also be needed to extend tificial intelligence and other innovations social security coverage to those working take the digital revolution to another level. independently or in short-term or other New thinking and policy adaptations will atypical contracts. The changing nature of be needed in areas such as competition pol- Zia Qureshi is a Visiting Fellow at the work will require more attention to horizon- icies, innovation systems and knowledge Brookings Institution. He also advises and consults for several other organizations. tal equity in taxes and transfers for workers diffusion, infrastructure underpinning the Qureshi’s research and commentary in different types of work arrangements. digital economy, upskilling and reskilling cover a broad range of global economic In capital income taxation, recent prog- of workers, social protection regimes, and issues, including a current focus on how technology is reshaping the ress under OECD/G20 processes on interna- tax policies. The era of smart machines will economic agenda. He has recently led tional cooperation to curb tax base erosion demand smarter policies. research projects at Brookings on the and profit shifting should enable national The politics of reform is inevitably com- implications of technological change for productivity, growth, jobs, income tax authorities to make better use of cor- plex. Reform may seem even more daunting distribution, and related policies. He has porate taxes that have been driven lower in the current political environment. One published widely on global economy and in recent years by international tax compe- thing reform action should not be para- development. Prior to joining Brookings, Qureshi worked at the World Bank and tition for mobile capital. In a period when lyzed by, however, is continued trite debates the IMF for thirty-five years, holding corporate profits have been high, boosted by about conflicts between growth and equity. leadership positions across a wide range rents associated with increased market pow- Research has increasingly shown this to be of responsibilities, including serving as Director, Development Economics, at the er, the optimal policy would be to tax profits a false dichotomy. World Bank. He also served as Executive at relatively high rather than low rates. In The dominant part of the agenda for Secretary of the IMF-World Bank Joint an increasingly networked global economy change to make technology—and glo- Ministerial Development Committee, and represented the World Bank at and fast-expanding digital commerce, in- balization—work better for all lies at the major international fora, including the ternational cooperation on tax matters will national level. Reforms are needed at the G20. He led several World Bank and IMF be even more important. international level as well so that rules of flagship publications. He holds a DPhil in Economics from Oxford University, where Making better use of wealth taxes can engagement between countries in trade and he was a Rhodes Scholar. improve both the efficiency and equity of other areas are fair. Not only must past gains tax systems. Wealth taxes are underutilized in establishing a rules-based international and have not kept pace with the surge in system be protected from the recent rise of wealth. High wealth inequality is a key nationalist and protectionist sentiment, but driver of intergenerational persistence of new disciplines and cooperative arrange- income inequality. Thomas Piketty’s work ments must be devised to underpin the next on inequality (Piketty, 2014) has attracted phase of globalization led by digital flows. controversy, but one key proposal—to find a better way to tax wealth—certainly has mer- it. The wealth dynamics of recent decades paint a picture of private riches and public poverty. While private wealth has soared, public wealth has declined, hobbling the capacity of public policy.6 There is scope to recover some of the lost tax progressivity without hampering economic growth (IMF, 2017b). Higher progressivity does not necessarily mean sharply raising marginal tax rates. A more efficient way is to reform the assortment of regres- sive and distortive tax expenditures that characterize most tax systems—and curb tax evasion.

Conclusion

Digital technologies are transforming the world of business and work. A key challenge for policies is to harness the potential of these technologies to produce more robust and inclusive economic growth. Policies Work in the Age of Data 12

Notes Education, and the Rise of Economics Department Working —Krugman, Paul. 2016. “Robber Earnings Inequality among the Paper, No. 1225. Paris: OECD. Baron Recessions.” The New York 1. While income inequality has Other 99 Percent.” Science —Dube, Arindrajit, Jacobs, Times, April 18, 2016. risen within many countries, 344(6186): 843–851. Jeff, Naidu, Suresh, and Suri, —Kurz, Mordecai. 2018. “On which is the focus here, inequality —Autor, David, Dorn, David, Katz, Siddarth. 2018. “Monopsony in the Formation of Capital and between countries has fallen, Lawrence, Patterson, Christina, Online Labor Markets.” NBER Wealth: IT, Monopoly Power and thanks to faster-growing emerging and Van Reenen, Jon. 2017. Working Paper Series, No. 24416. Rising Inequality.” Working Paper economies that are narrowing “Concentrating on the Fall of the Cambridge, MA: National Bureau 17–016. Stanford, CA: Institute of the income gap with advanced Labor Share.” American Economic of Economic Research. Economic Policy Research. economies. However, technological Review 107(5): 180–85. —Égert, Balázs. 2016. “Regulation, —Lindsey, Brink, and Teles, Steven. change poses new challenges —Azar, José, Marinescu, Ioana, Institutions and Productivity: 2017. The Captured Economy: to this process of convergence and Steinbaum, Marshall. 2017. New Macroeconomic Evidence How the Powerful Enrich as automation erodes the “Labor Market Concentration.” from OECD Countries.” American Themselves, Slow Down Growth, comparative advantage based on NBER Working Paper Series, No. Economic Review 106(5): and Increase Inequality. Oxford, low-cost, low-skilled labor that 24147. Cambridge, MA: National 109–113. UK: Oxford University Press. has propelled many emerging Bureau of Economic Research. —Égert, Balázs. 2018. “Regulation, —McKinsey Global Institute. 2015. economies’ growth. —Bivens, Josh, and Mishel, Institutions and Aggregate Digital America: A Tale of the 2. Frontier firms in this estimate Lawrence. 2015. “Understanding Investment: New Evidence Haves and Have-Mores. McKinsey are defined as the top 5% firms the Historic Divergence Between from OECD Countries.” Open & Company. with the highest labor productivity Productivity and a Typical Economies Review 29(2): —OECD. 2015. “Finance and within each industry. Non-frontier Worker’s Pay.” Economic Policy 415–449. Inclusive Growth.” OECD firms cover all other firms. The Institute Briefing Paper No. 406. —Eggertsson, Gauti, Robbins, Economic Policy Paper, No. 14. estimate covers firms in twenty- Washington, DC. Jacob, and Getz Wold, Ella. 2018. Paris: OECD. four OECD countries. —Boldrin, Michele, and Levine, “Kaldor and Piketty’s Facts: The —OECD. 2016. OECD Economic 3. The role of uneven capital David. 2013. “The Case Against Rise of Monopoly Power in the Outlook. Volume 2016, issue ownership and returns on capital Patents.” Journal of Economic United States.” NBER Working 1, chapter 2 on “Promoting as sources of inequality has Perspectives 27(1): 3–22. Paper Series, No. 24287. Productivity and Equality: A Twin been particularly emphasized —Brookings Institution and Chumir Cambridge, MA: National Bureau Challenge.” Paris: OECD. by Thomas Piketty in his 2014 Foundation. 2019. Productive of Economic Research. —OECD. 2017. A Fiscal Approach bestseller (Piketty, 2014). Equity: The Twin Challenges —Furman, Jason, and Orszag, for Inclusive Growth in G7 4. OECD Employment Database. of Reviving Productivity and Peter. 2018. “A Firm-Level Countries. Paris: OECD. 5. US Bureau of Labor Statistics Reducing Inequality. Report. Perspective on the Role of Rents —OECD. 2018a. The Productivity- data. Washington, DC. in the Rise in Inequality.” In M. Inclusiveness Nexus. Paris: OECD. 6. Most major economies —Brynjolfsson, Erik, and McAfee, Guzman (ed.), Toward a Just —OECD. 2018b. OECD Economic experienced this pattern. In the Andrew. 2017. Machine, Platform, Society: Joseph Stiglitz and Outlook. Volume 2018, issue United States, while net private Crowd: Harnessing Our Digital Twenty-First Century Economics. 2, chapter 2 on “Decoupling of wealth increased from 326% of Future. New York: Norton. New York: Columbia University Wages from Productivity: What national income in 1970 to about —CEA (Council of Economic Press. Implications for Public Policies?” 500% in 2015, net public wealth Advisers). 2016. “Labor —Goldin, Claudia, and Katz, Paris: OECD. fell from 36% to -17% (Alvaredo et Market Monopsony: Trends, Lawrence. 2008. The Race —Philippon, Thomas. 2016. al., 2018). Consequences, and Policy between Education and “Finance, Productivity, and Responses.” Washington DC: The Technology. Cambridge, MA: Distribution.” Paper prepared for White House. Harvard University Press. Brookings-Chumir research project Select Bibliography —Cette, Gilbert, Lopez, Jimmy, —Gutiérrez, Germán, and on The Technology-Productivity- and Mairesse, Jacques. 2016. Philippon, Thomas. 2017. Inequality Nexus, October, 2016. —Acemoglu, Daron, and Restrepo, “Market Regulations, Prices, and “Investment-less Growth: An —Piketty, Thomas. 2014. Capital Pascual. 2018. “The Race between Productivity.” American Economic Empirical Investigation.” Brookings in the Twenty-First Century. Machine and Man: Implications Review 106(5): 104–108. Papers on Economic Activity. Fall Cambridge, MA: Harvard of Technology for Growth, Factor —Chetty, Raj, Grusky, David, 2017: 89–169. University Press. Shares and Employment.” Hell, Maximilian, Hendren, —Hanushek, Eric, Schwerdt, —Rajan, Raghuram, and Zingales, American Economic Review Nathaniel, Manduca, Robert, Guido, Wiederhold, Simon, and Luigi. 2003. Saving Capitalism 108(6): 1488–1542. and Narang, Jimmy. 2017. “The Woessmann, Ludger. 2015. from the Capitalists. New York: —Adalet McGowan, Muge, and Fading American Dream: Trends “Returns to Skills Around the Crown Business. Andrews, Dan. 2017. “Labor in Absolute Income Mobility Since World: Evidence from PIAAC.” —Schwellnus, Cyrille, Pak, Market Mismatch and Labor 1940.” Science 356(6336): European Economic Review Mathilde, Pionnier, Pierre-Alain, Productivity: Evidence from 398–406. 73(C): 103–130. and Crivellaro, Elena. 2018. PIAAC Data.” Research in Labor —Comin, Diego, and Mestieri, —Haskel, Jonathan, and Westlake, “Labour Share Developments Economics 45: 199–241. Martí. 2018. “If Technology Stian. 2017. Capitalism without over the Past Two Decades: The —Aghion, Philippe, Akcigit, Ufuk, has Arrived Everywhere, Capital: The Rise of the Intangible Role of Technological Progress, Bergeaud, Antonin, Blundell, Why has Income Diverged?” Economy. Princeton, NJ: Princeton Globalisation and ‘Winner-Takes- Richard, and Hémous, David. American Economic Journal: University Press. Most’ Dynamics.” Economics 2019. “Innovation and Top Income Macroeconomics 10(3): 137– —IMF. 2017a. World Economic Department Working Paper, No. Inequality.” Review of Economic 178. Outlook. April 2017: chapter 3 1503. Paris: OECD. Studies 86(1): 1–45. —Decker, Ryan, Haltiwanger, on “Understanding the Downward —Shambaugh, Jay, Nunn, Ryan, —Alvaredo, Facundo, Chancel, John, Jarmin, Ron, and Miranda, Trend in Labor Income Shares.” and Portman, Becca. 2017. Lucas, Piketty, Thomas, Saez, Javier. 2017. “Declining Business Washington DC: IMF. “Eleven Facts about Innovation Emmanuel, and Zucman, Gabriel, Dynamism, Allocative Efficiency, —IMF. 2017b. Fiscal Monitor. and Patents.” Washington, DC: 2018. World Inequality Report and the Productivity Slowdown.” October 2017: chapter 1 on The Hamilton Project, Brookings 2018. World Inequality Lab. American Economic Review “Tackling Inequality.” Washington Institution. —Andrews, Dan, Criscuolo, 107(5): 322–326. DC: IMF. —Song, Jae, Price, David, Chiara, and Gal, Peter. 2016. —De Loecker, Jan, Eeckhout, Jan, —ITU (International Guvenen, Faith, Bloom, Nicholas, “The Best versus the Rest: The and Unger, Gabriel. 2018. “The Telecommunication Union). 2016. and von Wachter, Till. 2019. Global Productivity Slowdown, Rise of Market Power and the ICT Facts and Figures 2016. “Firming Up Inequality.” Quarterly Divergence Across Firms, and Macroeconomic Implications.” Geneva. Journal of Economics 134(1): the Role of Public Policy.” OECD Working Paper. —Krugman, Paul. 2015. 1–50. Productivity Working Paper, No. —Denk, Oliver. 2015. “Financial “Challenging the Oligarchy.” —Stiglitz, Joseph. 2016. 5. Paris: OECD. Sector Pay and Labour Income The New York Review of Books, “Monopoly’s New Era.” Project —Autor, David. 2014. “Skills, Inequality: Evidence from Europe.” December 17, 2015. Syndicate, May 13, 2016. Inequality in the Digital Era by Zia Qureshi 13

—Summers, Lawrence. 2016. “Corporate Profits are Near Record Highs. Here’s Why That Is a Problem.” The Washington Post Wonkblog, March 30, 2016. —Sundarajan, Arun. 2016. The Sharing Economy: The End of Employment and the Rise of Crowd-Based Capitalism. Cambridge, MA: MIT Press. —Turner, Sarah. 2017. “Education Markets: Forward-Looking Policy Options.” Hutchins Center Working Paper No. 27. Washington DC: Brookings Institution. —West, Darrell. 2018. The Future of Work: Robots. AI, and Automation. Washington, DC: Brookings Institution Press. —Wolff, Edward. 2014. “Household Wealth Trends in the United States, 1962–2013: What Happened over the Great Recession?” NBER Working Paper Series, No. 20733. Cambridge, MA: National Bureau of Economic Research. —World Bank. 2019. World Development Report 2019: The Changing Nature of Work. Washington, DC: World Bank. ZIA QURESHI HOW TO CITE THIS ARTICLE Qureshi, Zia. “Inequality in the Digital Era.” In Work in the Age of Data. Ma- drid: BBVA, 2019.

ARTICLE IN SPANISH LISTEN TO THE ARTICLE HERE • La desigualdad en la era digital.

READ THE FULL BOOK • Semiglobalization and Strategy for a Post-Crisis World, Pankaj Ghemawat The Multiple Faces of Globalization. • Work in The Age of Data • El trabajo en la era de los datos • Development and Strategies for Fighting Poverty, Abhijit V. Banerjee Frontiers of Knowledge.

• The New Economics and Politics of Globalization, Douglas R. Nelson The Age of Perplexity: Rethinking the World We Knew.

ALL THE OPENMIND COLLECTION TITLES