These articles are part of exploration that started in late 2015 and early 2016. It was a confusing time: autonomous vehicles and Boston Dynamics were frequently in the news and OpenAI was just formed to make sure that the “march towards AGI was safe.” UBI was actively being discussed by Sam Altman and Ray Kurzweil was promoting 10x / exponential thinking. At the same time, economists like Robert Gordon were arguing that we were actually living in a time of low progress.
I tried to develop my own perspective on what was happening, with key conclusions presented at this link. Greater detail supporting other conclusions can be found in Part I and Part III.
Aging populations are an increasing share of the overall population, creating challenges.
Noted in Part I, weak general innovation is one driver of low economic growth. Changes affecting the makeup of our labor force will provide further drags on productivity Two factors are an aging labor force and a “skills challenge” for remaining labor attempting keeping up with dynamic changes in technology.
Aging labor forces have a tremendous effect on productivity. On average, every 10% increase in the share of a state’s population over the age of 60 reduced per capita growth in gross domestic product by 5.5%. This is explained by two factors. First (and most intuitively), as more workers retire, the labor force grows more slowly. This explains one-third of the 5.5% growth hit. But the (surprising) bigger effect is through reduced productivity—that is, output per hour—of the remaining workers. Mentorship matters! Experienced employees make their less-experienced reports more productive.
This effect of aging on productivity will be even more pronounced as McKinsey estimates that we are entering an era of slow-growing or even shrinking labor forces (see image below). To maintain historical rates of GDP growth, it’s estimated that the “aging advanced” economies would need to increase productivity growth by about 60 percent of historical levels, to about 1.9 percent annually. This is a rate that hasn’t been reached since the 1960s. As an extreme example, the Southern Europe cluster will face an even steeper challenge: these economies would need to double their 0.7 percent rate of productivity growth of the past 20 years to sustain growth in GDP per capita.
When these productivity challenges are coupled with declining labor force participation rates, the ultimate effect is a deceleration in per capita GDP growth. Future generations could have a lower rate of improvement in living standards than their parents—and in some cases, such as in the Southern Europe cluster, the next generation could be poorer than their parents.
For younger workers, capturing emerging areas of economic growth require new skills.
What about the remaining labor force? Here we find significant mismatches between what employees can do and what firms need to be done. In most advanced economies, unemployment rates for the least-skilled are two to four times those of the most highly skilled workers, whether the economy is in recession or recovery. The effects of falling demand for low-skill labor have been especially harsh for younger workers. Today, 75 million young people (aged 15 to 24) who are not in school or college are unemployed, account for 38 percent of the world’s unemployed.
For skilled labor, firms are finding a lack of talent—a “STEM shortage.” This is reflected in the time required to fill roles. In a Brookings study, the duration of advertising for a STEM occupation that typically requires a graduate degree is 25 days at the median, 50 days average, and 93 days at the 80th percentile. These search times are much longer than the average for all non-STEM U.S. vacancies, for which the median is just five days, the average is 33, and the 80th percentile is 64. The question of how to tackle these problems—adapting an economy with a large aging population with significant dynamism in the most valued skills—will be a challenge of the next few decades.
Skills in digital growth will require new education models that are promising, but unproven.
There has been much documentation around occupational skill gaps: for example, in manufacturing, in STEM, and in digital literacy. As a result, many non-traditional workforce development programs have been created in recent years, including online courses (MOOCs) and bootcamp models. Tech training is touted by workforce development policymakers as being at the leading edge of trends such as skills-based hiring, non-traditional learning, and rapid education models. Bootcamps are considered by some to be at the forefront of workforce development more generally.
Yet the field is new, and the lack of comprehensive and concrete data on outcomes and transparency makes much of the tech training hype difficult to verify. Additionally, policymakers and employers may be overwhelmed by the abundance of new training programs. Questions still linger about whether these programs are successfully creating a skilled tech workforce that meets employers’ needs, much less if these programs are fulfilling the promise of economic mobility for participants and creating a workforce that is diverse in gender and race.
“The rubber hits the road at employment. All this great stuff means nothing if you don’t hire the people at the end.” – Barbara Chang, Code to Work
As a relatively new field, complete data does not exist to evaluate which programs are most effective, making it difficult for employers, participants, and funders to determine their success. The data most programs release do not show whether graduates take jobs in the field, or whether they are still employed several years later. As a result, employers may be missing good job candidates due to limited resources for recruitment and job seekers must rely on anecdotes and unverified statistics to make decisions about whether to participate in a program and which one to choose.
What does it all mean
Aging populations may be a significant headwind to growth in the subsequent decades. To maintain and exceed current rates of productivity growth, the population will need to more intensively adopt digital skills. Job skills, retraining, and effective certification are particularly important areas of public-private partnerships.
Additional research, perspective, and detail can be found in the following supporting articles:
- Intro: Creating a general framework for current opportunities and challenges.
- Part I: We are in a time of slow productivity and strong incumbents.
- Part III: Technology requires deeper integration into our work lives.
My goal in putting this together was to clarify my own thinking - and to present it in a way that can be challenged and deepened by others. I would be grateful to learn what may resonate (or differ) from your personal perspectives and work.