Slumping productivity in the midst of rapid technological innovation presents a paradox. Information and communication technology (ICT) is experiencing a boom, as evidenced by impressive advances in computers and software, smartphones, and robotics. Yet, productivity growth has slowed in most major economies. For example, in the United States, the leader of ICT innovation, labor productivity since 2005 has grown at less than half the rate of the preceding decade.
Weak productivity growth has contributed to slow economic growth. At the same time, the benefits of economic growth have been shared unequally, with income inequality rising in most major economies. As productivity is the main engine of economic growth, persistently slow productivity growth may entail reduced potential for remunerative jobs and rising living standards. The current concern and ensuing debate about productivity, therefore, is no surprise.
One side of the debate questions whether technological innovation is now less consequential. “Techno-pessimists” see new ICT innovations as less impactful in terms of economy-wide benefits than were past technological breakthroughs, such as electricity, internal combustion engines, and jet turbines. In contrast, “techno-optimists” see ICT advances as transformative in their potential to drive productivity growth, though the impact may be subject to lags and occur in waves. While both of these arguments are hotly debated, the real issue moving forward is how the challenges seen by the “pessimists” could be addressed to capture the opportunities envisioned by the “optimists.”
Another part of the debate focuses on measurement: is the productivity slowdown real or illusory? If prices do not reflect true quality improvements in products based on new technologies, real output is underestimated. Moreover, standard measures of productivity are based on GDP, which includes only market production and excludes consumer surplus. Internet-based services, such as Google and Facebook, generate substantial utility for consumers above their virtually zero market price. Thus, measured output growth may fail to capture important benefits that ICT innovations entail.
Recent research finds that the gains from new technologies are indeed underestimated for the reasons mentioned above. However, it also finds that mismeasurement can explain only a relatively small part of the actual slowdown in gains. In large part, the slowdown is real. The productivity slump is not an illusion. The key questions, then, are what exactly has caused the slowdown in productivity growth in the United States and other major economies and what can be done about it.
Part of the slowdown in productivity growth can be explained by cyclical factors related to the 2008 global financial crisis that saw sharp declines in aggregate demand and disruptions in financial markets. But the slowdown, which started before the crisis, also reflects structural factors. Productivity growth should recover as economies emerge from the long shadow cast by the crisis, but it will fall short of pre-crisis levels unless the structural impediments are addressed.
At the root of the issue, the productivity slowdown may reflect growing inequality between large, innovative firms and the rest of the firms in the market. Firm-level evidence shows that productivity growth remains fairly robust among firms that are at the technological frontier, but is slowing considerably in the majority of smaller firms. The widening productivity gap between frontier and lagging firms suggests that new technologies are diffusing less effectively across economies. Business dynamism, as reflected in firm startup frequency, waned while market concentration rose in many sectors. Thus, while innovation itself may not have weakened, the spread of innovation certainly has.
Broader diffusion of gains from new technologies are hampered by skill constraints and mismatches. Education and training have not kept pace with technological change. Supply has responded slowly to demand for new and higher-level skills, effectively limiting productivity growth and contributing to rising wage inequality. Labor market rigidities hinder the reallocation of labor and skills to changing demand, further compounding the challenges.
Increasing income inequality, in turn, has depressed aggregate demand and added to the risks of “secular stagnation”, wherein GDP and productivity growth are trapped at low levels by chronically deficient demand. Additionally, increasing inequalities reduce the capacity of lower-income groups to invest in skills, thereby compounding the problem. Rent capture and winner-take-all markets can entrench inequalities and blunt the competitive edge of markets.
Weak investment is also part of the productivity slowdown story and the increased economic stagnation risks. Innovations are typically embedded in capital and need new investment to diffuse across economies and firms. Complementary investments in infrastructure have increasingly fallen short of needs. In most advanced economies, investment rates dropped sharply after the financial crisis, but again, they were already declining even before the crisis began.
Looking ahead, the future could be bright—but only if policies rise to the challenge. Further advances in digital technologies—cloud computing, 3D printing, the internet of things, artificial intelligence—hold much promise. There is also unexploited potential for further productivity gains from existing ICT innovations. Moreover, the environmentally sustainable growth agenda offers a new world of opportunities for innovation and investment, particularly in energy-related sectors.
Integral to unlocking this potential are regulatory and institutional reforms meant to spur innovation at the frontier and promote its diffusion throughout the economy. This includes reforms in product, labor, and risk capital markets to provide more competition, a level playing field, and flexibility in responding to dynamic shifts in technology.
Corporate and personal income tax reforms will be key in incentivizing investment and innovation, as well as capital and labor efficiency. At the same time, tax and public spending reforms will need to be part of a strategy to address rising income inequalities. Instituting taxes on carbon emissions could help further the promise of green technologies, while also providing an efficient way to raise public revenue.
Additionally, education and training programs need to be reformed to improve access to a dynamic job market by focusing on helping individuals learn skills that will remain relevant in the face of technology’s shifting demands. This will require innovations in skill-building programs, including by adapting education and training systems to take advantage of the digital age, to foster new skills at scale, and to support their renewal through lifelong learning. Social insurance programs will need to respond to increased labor mobility and reallocation, as well as needs involved in retraining.
Governments’ macroeconomic policies should support a rebound in investment, underpinned by structural reforms that remove impediments to channeling desired savings toward productive national and global investment opportunities. Especially at a time of historically low interest rates, there is a strong case for mobilizing and using fiscal space to boost public investment in infrastructure in ways that also catalyze private investment. Increased public investment in research and development can work alongside private investment to bolster innovation and enhance its economy-wide impact.
However, against this backdrop of necessary policy reforms are worrisome developments in the global economy. Many countries are witnessing a rise in national populism and protectionist sentiments. The policy response to the challenges posed by open trade and technological change should not be to limit these forces, as they are key drivers of productivity and growth. Rather, governments should aid productivity and growth through retraining and transitional support programs, as well as by using fiscal tools to promote a fairer sharing of the gains from trade and technological progress.
The debate on productivity measurement and techno-pessimism versus techno-optimism will continue. But the slump is real and more attention should be given to how policies need to respond, which is ultimately the more productive debate.