Labor markets in the United States are being affected by the changing nature of work, including the decline of paid labor in the production of goods and services. Some would say that a shift is underway that will lead to a “world without enough work”. While there are many factors that are contributing to this change, technology is and will continue to play an important role. Some estimates suggest that up to 47% of jobs in the United States are at high-risk of being eliminated as a result of automation.
In mid 2017, a study was launched to understand the impact of automation on Connecticut’s job market. This effort utilized the results of recent studies and information that was available from the Connecticut Department of Labor to determine the level of potential job losses, its timing, and whether new job creation could keep pace with the changes in the job market.
The following is a summary of the conclusions that were reached based on an analysis of 648 occupations (using the United States Department of Labor Bureau of Labor Statistics Standard Occupation Classification system) that are represented in Connecticut’s job market.
- Of a total of 1,692,200 Connecticut jobs, 39% or ~660,000 jobs are at high-risk
(>75% probability) of being eliminated as a result of automation.
- The jobs that fall into the high-risk category represent $28.4 billion in wages,
which is 29% of the total for the state.
- The risk of job loss by income level indicates that jobs at the lower end of the income spectrum tend to fall into higher-risk categories, while higher paying jobs fall into the lower-risk categories.
- The effect of automation varies by labor market area, with Norwich – New London – Westerly RI having the highest percent of jobs at high-risk and Bridgeport – Stamford the lowest. The following is a summary of the impact of automation on Connecticut’s labor market areas.
- This analysis didn’t include the small labor market areas, i.e. Enfield (45,200 jobs), Torrington (33,300 jobs) and Willimantic – Danielson (27,300 jobs).
- An analysis of the occupation groups based on variables related to perception and manipulation, creative intelligence, and social intelligence, indicates that their risk profiles varies widely, with some groups at very high-risk of being impacted by automation. The following is a summary of the composite risk scores for the occupation groups that were examined.
- While it is difficult to separate the role of automation from other considerations
(such as the state’s job generation capability, desirability as a place to do
business, contribution of “legacy” businesses to employment, trends related to
new business formations, workforce readiness and productivity, and quality of life
issues), it is the confluence of these factors that will shape the future, with
technology assuming a progressively larger role over time.
- While residents of certain labor market areas benefit from their proximity to other
labor markets, e.g. Bridgeport – Stamford and Danbury (Northern New Jersey,
New York City and Westchester County New York), Hartford (Springfield), and
Norwich – New London (Providence), no attempt was made to assess the
automation risk in these locales.
- The timing of the impact of automation is not certain, but there is evidence in
recent employment statistics and WARN (Worker Adjustment and Retraining
Notification) data that the impact is already being felt, esp. in industries whose
employees’ principal activities fall into high-risk occupation group, e.g. retail
- It is possible that Connecticut will feel significant impact within the next 8 – 10
years, and new job creation may not be able to keep pace with the jobs that are
- While the range of the impact of automation varies by region and occupation
group, it is likely that the safety net programs will be overwhelmed by the
magnitude of the change that is underway.
- In addition to the impact on social programs, the loss of wages related to the
elimination of jobs in the high-risk category will impact tax revenues and the
ability of state and local government to fund essential services, and meet
pension, retiree health and debt obligations.
At this time, there is limited, specific research that is available to assist policy makers and planners understand the impact of automation on Connecticut. In addition, the political dialogue at the state and local level may not be sufficient when you consider the critical issues related to the future sources of job growth in an era of increased automation, the need to build new digitally oriented economic clusters, the effectiveness of existing economic development programs in the creation of higher paying / “sticky” jobs, and steps that should be taken to prepare the state’s workforce to effectively compete for jobs in the economy of the 21st century.
While the transition to a “world without enough work” is inevitable, steps can be taken to slow the impact, but there is a need to focus on this challenge now. It is not clear how this will end, but we have a choice, i.e. we can manage or fall victim to the transition that is underway. One thing is certain; we need to prepare for a fundamentally different world.