The Bureau of Labor Statistics (BLS) is part of the U.S. Department of Labor, which is the ultimate authority on jobs data in the U.S. Each year, the department accounts for the expansion and contraction of almost a thousand different types of jobs. And it goes a step further than that. The BLS also publishes 10-year projections for every job category, which estimate how much a given sector will grow or shrink over the coming decade.
These projections are taken very seriously, not just by economists and the media, but by millions of Americans planning their educational and professional futures. These projections can influence whether you decide to pursue a specific field or choose a certain degree type, like an associate’s degree or a bachelor’s degree. Given that BLS forecasts directly affect how people map out their career trajectories, these projections are highly consequential.
These projections have always been a forward-looking endeavor, and the conversation shifts every year or so to focus on the next round of forecasts. What gets lost is an account of how accurate the prior BLS projections were. This report, in part, makes that account.
Education Advisors analyzed the last 20 years of BLS jobs projections, calculating when and in what sectors the BLS has been most inaccurate. Read on to find out how often the highest jobs authority in the country gets it right!
Key Takeaways
- The most underestimated job category of the past decade was mathematical science occupations.
- The most overestimated job category of the past decade was personal care and service workers.
- Over the past 20 years of estimates, the BLS, on average, overestimated job growth by 5.6% each year, or about nine million jobs.
- The BLS overestimated or underestimated growth by 20% or more for 42% of all job categories.
Understanding Job Numbers
Bureau of Labor Statistics (BLS) data can be complex and challenging to interpret, especially for those unfamiliar with the terminology. Here is a simplified guide to help make sense of the data presented visually in this report:
- Job Numbers: All job numbers in tables and graphic visualizations are expressed in thousands.
- Analysis Timeline: Data is assessed in 10-year periods, with each period containing the following definitions:
- Beginning jobs: Job counts at the beginning of the period
- Projected jobs: Job count estimates made at the beginning of the period
- Actual jobs: Job counts at the end of the period
- Percent Difference: The percent difference is the difference between projected jobs and actual jobs at the end of a period.
- Job Categories: Jobs are organized into summary and line item categories. Summary categories are broader job groups, while line times are specific job types within summary groups.
- Category Focus: Most of our tables reflect summary categories only to focus on the most prominent parts of the job market, with a few exceptions that we identify in this analysis as line items.
The Most Overestimated and Underestimated Jobs of the Past Decade
If you’ve ever relied on a weather report, you know that forecasts can sometimes be wrong. Forecasting in most fields is an imperfect science, and that includes the economy. Due to unpredictability and uncontrollable factors that exist in economic forecasting, it could be said that landing a perfect job growth prediction is harder than landing a rocket on the moon — we should expect some wide margins of error.
However, job growth predictions that are grossly under or overestimated provide shocking evidence as to just how wildly wrong one of the highest U.S. economic authorities can sometimes be.
For example, the summary jobs category, transportation and material moving, accounted for 9,246,000 jobs in the U.S. in 2012. This category includes warehousing and trucking jobs, which the BLS forecasted to grow by 9% by 2022. In reality, that sector exploded by 55%, with the predicted growth off by 43%, meaning the BLS underestimated jobs growth for that sector by nearly half, which is far outside of any typical margins for error.
However, it is crucial to note that large external shocks to the economy can derail BLS predictions. Predicting the onset of the COVID-19 pandemic is not something that is within economic forecast capabilities. The unpredicted growth that the transportation and material moving summary sector saw by 2022 is likely a large result of the significant surge in e-commerce and home delivery services that became essential during the pandemic.
Job growth in healthcare support occupations, which include home healthcare services and nursing care facilities, was also severely underestimated by the BLS. As of 2022, there were 36% more jobs in this summarized sector than predicted. In other words, there were 1,879,000 more healthcare support occupations jobs by 2022 than the BLS foresaw. For an industry sector that had only 4,110,000 jobs in 2012, an extra 1,879,000 jobs not predicted is a pretty extraordinary forecast miss.
If you are considering a career in healthcare, our healthcare degree guide can help you find the right path and understand the opportunities available in this still-growing field.
Management occupations also had a better-than-expected decade, beating BLS forecasts by 32%. This broad summary category includes company and enterprise management, school management, real estate, and much more. This heavily underestimated sector added 3,077,000 more jobs than the BLS predicted between 2012 and 2022.
The BLS did not just underestimate job categories this last decade; there were many sectors that the bureau was quite bullish on that failed to meet their forecasts.
The personal care and services jobs summary category was estimated to grow by 21% from 5,376,000 jobs in 2012 to 6,499,000 jobs by 2022. Instead of growing at all, this sector experienced a contraction, losing 1,260,000 jobs by 2022, shrinking by 23%, meaning the BLS overestimated growth for personal care and service jobs by 37%.
The BLS also overestimated growth for office and administrative support occupations by 17%. This major summary category recorded 22,470,000 jobs in 2012 and was expected to grow by 7% to 24,004,000 by 2022. However, the industry instead lost 2,579,000, dropping by 11% to 19,891,000 by 2022.
Choosing a resilient career path can be challenging, but Education Advisors is here to help navigate the uncertainty. Whether you are looking to enter high-demand fields such as cybersecurity, healthcare, or psychology, we offer comprehensive resources to help you navigate your career path amidst current and future economic uncertainty. Check out our computer science degree guide to explore options in an exponentially growing field.
The Most Overestimated and Underestimated Jobs of the Past 20 Years
Looking back over most of this century, we find a great example of how major unexpected shifts in the economy can totally derail jobs forecasts. Four of the five most overestimated jobs categories since 2002 were in the 2002–2012 forecast period, which saw the Great Recession of 2008.
Between 2002 and 2012, the summary job category, wood model makers and patternmakers, saw the greatest overestimate of jobs growth by the BLS. This industry was not very recession-proof. While the BLS forecasted a modest 11% growth for this category, the sector plummeted by 78% during this time. The difference between the number of jobs the BLS had predicted by 2022 and what the actual jobs numbers were was a difference of 80%.
Baggage porters, bellhops, and concierges also suffered a massive blow during this time, with BLS predictions off by 76%. At the start of 2002, there were 248,000 jobs in this summary category, but by 2012, only 67,000 remained — a nosedive of 73% of jobs lost.
Another unfortunate collateral of the Great Recession was Industrial machinery installation, repair, and maintenance workers. While the BLS was dramatically off in their predictions for this summary sector, the 76% difference in projected and actual jobs by 2012 is clearly reflective of the unforeseen external economic shock that the Great Recession had on the U.S. economy.
Given that the Great Recession was a negative external economic shock, the jobs growth that the BLS underestimated during this time is of special interest. They are of such interest because many sectors experienced sharp contraction during the recession, and those that not only grew but outperformed BLS expectations lay insights into the true meaning of recession-proof.
The first extraordinary forecast-buster was in a BLS category largely composed of health technologists and technicians not listed in other major health technologists categories. Some of the industries that fall in this catch-all category are specialty hospitals and medical and diagnostic laboratories. This small group was projected to rise from 5,000 jobs in 2002 to 6,000 by 2012. However, the sector category exploded to 104,000 by the end of the period, beating BLS predictions by 1,637%.
Many catch-all categories, like miscellaneous health technologists and technicians, saw growth far outside the bounds of BLS predictions during this time. This unforecasted growth is in part due to the nature of these catch-all groups that often account for jobs that are not yet well-known or established in the economy.
Miscellaneous social scientists and related workers was another catch-all category that saw astounding growth during this time. In 2002, there were only 14,000 U.S. jobs in this summary sector. By 2012, there were a total of 54,000, nearly three times more than what the BLS expected the total jobs to be for this group in 2012. This category, which mostly includes social scientists working at public schools, hospitals, and research and development, beat BLS predictions by 261%.
For those interested in a career in social sciences, psychology can be a rewarding career trajectory with new opportunities emerging. Explore our psychology degree guide to understand the options available to you in this industry that have previously outperformed BLS job predictions.
What to Expect in the Coming 10 Years of Jobs Growth
The jobs growth forecast for the 2022–2032 period should sound a little less authoritative now that we’ve explored the difficulties in predicting future economies. Nevertheless, BLS jobs growth predictions are the best barometer we have to gaze into the crystal ball of future jobs growth. Still, let’s examine future forecasts for a couple of categories the BLS was previously dramatically wrong about.
By 2032, the BLS forecasts a 15% growth for the computer and mathematical summary category. The BLS previously underestimated growth in this category by 17% between 2012 and 2022. Is the 2032 bullish BLS forecast reflective of a bureau self-correcting for its prior miss or based on optimistic data? Only time will tell (and a historic BLS predictions accounting analysis).
The BLS also heavily underestimated jobs growth in the healthcare support summary category for the 2012–2022 forecast period. The bureau failed to forecast an additional 36% growth during that time. By 2023, BLS projections are for a 15% expansion in this category. Overall, the BLS projects a very modest 3% growth in total jobs over the next 10 years.
BLS jobs growth predictions are not known for their wild guesses; the bureau is a conservative forecasting agency by nature. Their jobs numbers serve as guides but not mile markers. When nearly totally unpredictable external economic shocks affect the markets, it is almost certain to send forecasts into a tailspin.
Economic forecasts are made under the assumption of ceteris paribus, which means “all other things being equal.” In essence, it is a practice of isolating variables for analysis, ignoring everything else. In reality, nothing exists in a vacuum, economically speaking, which is why BLS forecasts can be wrong so often. This necessary method is why forecasting jobs growth is an imperfect science; assumptions are required to execute the practice. Nevertheless, these predictions are the best forecasting metric for U.S. jobs growth in the world.
Summary
Our analysis underscores the challenges inherent in predicting job market trends. BLS insights are valuable, but as we have just detailed, forecasts are only just that — forecasts, and these projections can frequently be off due to unforeseen economic shifts and many other factors.
For those looking to navigate this ever-changing job landscape, education is a critical tool. Whether you’re just starting in your career, considering a change in your career trajectory, or seeking to deepen your knowledge, obtaining a degree can be your pathway to advancement. Education Advisors is your comprehensive and informative resource, helping you make informed decisions about your education and career.
Jobs in healthcare frequently beat BLS predictions. Explore our healthcare degree guide to discover degree types that align with your career goals. Computer science–based jobs growth was also often underestimated by the BLS. Dive into our computer science degree guide to explore popular study concentrations that can start your path to unlocking a growing array of high-paying and dynamic career opportunities!
Methodology
To account for the accuracy of the Bureau of Labor Statistics over the past 20 years, we assessed projections and subsequent job counts over six 10-year periods occurring between 2002 and 2022, each separated by two years. The BLS now makes its “employment projections” annually, but for most of the span between 2002–2022, projections were made every two years.
What data was included and excluded from our assessment?
BLS job categories vary slightly from year to year. Between two given periods, the BLS might change the name of a category, absorb one category into another, or remove a category completely. To account for this, we only included categories that consistently appeared across all six periods based on code rather than name alone (allowing for slight naming changes). This was a vast majority of categories that exist, more than 95% of approximately 1,000 total job categories.
How did we calculate accuracy?
We calculated accuracy by measuring the difference between the projected jobs at the beginning of a period and the actual jobs at the end of a period. So, for example, if the BLS notes in 2024 that Job Type X currently has 80,000 jobs, and it predicts that it will grow to 100,000 jobs by 2034, and then in 2034, the actual count of jobs is 110,000, that represents an underestimate of 10% because the difference between projected and actual jobs was 10%.
How to understand categories
BLS categorizes jobs either in a summary category or a line item category. Unfortunately, to complicate matters, the BLS uses several summary categories, so a different code and job category title can be “summary” but also nested under another code and category, also called summary.
For this analysis, we treated all summary categories the same. We focused primarily on summary instead of line item since that’s the best way to express broad and important trends in the job market.
One exception to note is when we report on major groups in the category, which are technically listed as summary categories in BLS data but are known to be the 22 highest-level categories that define the entire job market (excluding military jobs, for which we did not find consistent data across periods).