Editor’s note: Prof. John Quinterno is the Visiting Professor of the Practice in the Sanford School of Public Policy at Duke University, and also the founder and principal of South by North Strategies Ltd., a research consultancy specializing in economic and social policy. Recently Quinterno expressed concerns about the impact of Apple’s new campus in RTP and Google’s new engineering hub in Durham. WRAL TechWire asked him for his analysis of the NC jobs projections from the Department of Commerce.
DURHAM – The employment projections prepared periodically by the NC Department of Commerce are a useful but highly misunderstood data product.
The basic idea is to project certain important aspects of the labor market for a future point of time using standardized methods developed by the federal government. A projection is not a forecast.
Rather, a projection makes assumptions that allow for the identification of long-term trends against a neutral backdrop, while a forecast attempts to predict a value at a certain point in the future.
That said, plausible employment projections can help economic developers in the crafting of industrial recruitment strategies, workforce developers in the targeting of programs, higher education institutions in the development and refinement of programs and curricula, high school career counselors in their work helping students think about their futures and individual working people in their career planning. All of those are positive things.
At the same time, the projections are based on any number of unworldly assumptions.
The assumptions allow for the math to work, but they differ sharply from reality. For instance, the models assume a state of full employment. In other words, the assumption is that overall labor supply and labor demand balance and the only unemployment that exists is frictional unemployment or just short-term unemployment that occurs when someone enters the labor force for the first time or decides to switch jobs.
Other forms of unemployment, like that caused by a recession or when, say, the main factory closes in a community are assumed away. Similarly, the models don’t acknowledge changes in the business cycle (recessions and expansions), nor do they allow for unexpected shocks like wars, natural disasters, or, especially pertinent today, pandemics. The models also assume that broad social, demographic, and economic trends will continue along the current trend line.
Another issue is that the employment projection consists of a set of six different models that relate to different important aspects of the labor market.
Each sub-model measures different concepts in different ways. For instance, industrial employment relates to the number of positions on the payrolls of individual establishments, not the number of distinct workers. Labor force data, in contrast, reflect discrete human beings. Industrial classifications relate to what business establishments produce, occupational classifications to the tasks individual persons perform. A common mistake in using these models is to mix the industrial and occupational data in an effort to identify labor shortages in certain industries. The numbers don’t support such claims.
Given the nature of projections, they can differ widely from real-world developments.
For instance, the newest set of models makes no allowance for the COVID-19 pandemic, just as models developed in the early 2000s made no allowance for the “Great Recession.” In reviewing national projections for the 2008-2018 period, the US Bureau of Labor Statistics found, among other things, that the model was right as to which industry would grow 81 percent of the time. The model also overestimated such factors as overall economic growth and labor force growth.
Turning to the newest projections for North Carolina, they seem reasonable based on the state of the world that existed pre-pandemic. The ideas that most industrial growth is projected to occur in the service sector and that most occupational growth is projected to originate from replacement rather than growth make sense.
Similarly, the idea that occupations requiring postsecondary education are projected to have faster growth rates than other fields while most actual jobs are projected to be in occupations that don’t require post-secondary education makes sense. And, the leading industries projected for each sub-state area make sense, as well.
I would imagine that actual trends will not mesh with the model’s assumptions owing to the pandemic, which already has upended many demographic and labor force assumptions. As a result, I would take the newest projections with a heaping mountain of salt. Hopefully, the dust will have settled and models can be updated for the projections slated for release circa 2023.
Even if we assumed away the pandemic and looked at the world as it existed in 2018, the baseline trends point to a labor market that, statewide, is projected to grow too slowly to accommodate every person who wants or needs a job.
It also points to a labor market in which projected growth is heavily concentrated in lower-paying fields that tend to have few benefits, offer little security, and are increasingly precarious in nature. The projections also point to the increasing geographic disparities within the state as different regions are projected to grow differently.
From a policy standpoint, based on the projections, state and federal policymakers should pay much more attention to such issues as job quality, wage growth, social insurance, and equitable development, both geographically and in relation to different social groups.
At the state level, for instance, policymakers should consider how to set higher floors (higher minimum wage, paid leave standards, Medicaid expansion, retirement access, unemployment insurance), limit inequality (progressive taxation), and invest in upward mobility (all levels of education).
The pandemic only has turbocharged those issues.
Prof. John Quinterno is the Visiting Professor of the Practice in the Sanford School of Public Policy at Duke University, and also the founder and principal of South by North Strategies Ltd., a research consultancy specializing in economic and social policy. Recently Quinterno expressed concerns about the impact of Apple’s new campus in RTP and Google’s new engineering hub in Durham.