The requests that arrive at a state workforce agency on a Tuesday afternoon are rarely the same. A machinist who was laid off wants to know if welding is still profitable. A high school counselor is attempting to determine whether the new dental hygiene program at the nearby community college truly results in employment. Sounding a little impatient, a reporter is on the phone requesting the county-by-county unemployment figures for the previous quarter. The same data, very different requirements.
This is the fundamental conflict that underlies labor market information, or LMI as its producers refer to it. In an attempt to properly map this out, a 2012 study commissioned by the Workforce Information Council divided clients into three broad categories: labor market actors and advisers, policymakers and planners, and what the report dryly referred to as value-added disseminators. Although the phrase is awkward, it conveys a real idea. Depending on who is asking, the same dataset is pulled in entirely different directions.
At one end of that spectrum are students and job seekers, who typically want something fairly tangible. What is the average salary for a paralegal in this area? which professions are genuinely expanding rather than merely appearing to do so based on what everyone’s LinkedIn feed indicates. Students and those looking to change careers often ask a slightly longer version of the same question, focusing more on whether a two-year program in HVAC repair will still be relevant in five years than on this month’s job board.
Employers have different expectations, and it’s important to observe how transactional their interpretation of LMI becomes. They want to know whether a certain skill set is becoming more difficult to find within their hiring radius, what competitors are paying, and the availability of local labor. Technically, a hospital system in Texas and a small manufacturer in Ohio are using the same federal data pipelines, but they are doing entirely different calculations on top of them.
The policy side, on the other hand, typically runs on a slower clock. Undoubtedly, government officials and economic development offices require demographic shifts and unemployment snapshots, but more importantly, they require data to support decisions that have already been partially made, such as funding allocations, training grants, and the occasional politically awkward report that no one asked to commission. Researchers and educators work in a similar field, converting unprocessed data into academic papers or curriculum decisions that might or might not be used by those who truly need them.
Strangely enough, career counselors and workforce professionals likely have the most difficult jobs in this entire ecosystem. They are supposed to take complex statistical tables and transform them into a useful tool for a nineteen-year-old choosing a major. This calls for combining hard data with soft, local knowledge—the kind of information that doesn’t appear neatly in any spreadsheet and is instead obtained through years of discussions with employers who never bothered to complete a formal survey.
Observing this system from the outside reveals how little coordination there is between the organizations that generate the data and the individuals who use it. According to that same 2012 report, the majority of state LMI offices continue to collect client feedback in rather haphazard ways, with sporadic surveys here, focus groups there, and nothing particularly systematic. Better web analytics and more interactive portals may have contributed to some of that change in the years since. However, there is still a perception that the gap between what is produced and what people actually need hasn’t shrunk as much as people would like to acknowledge.
