Dr. Erica Groshen, former commissioner of the U.S. Bureau of Labor Statistics, and Paul Famolari with the South Carolina Department of Employment and Workforce join the podcast to discuss enhanced wage records and the role they can play in helping states better understand their workforce and evaluate job training programs.
Our focus on this podcast is enhanced wage records and the role they can play in helping states better understand their workforce, evaluate job training programs and show the return on investment for various educational programs.
Collecting enhanced employment data involves asking employers for additional when they file regular reports with their state unemployment office, such as information about hours worked and establishing consistent coding for similar job types.
Joining the podcast to discuss the issue are Dr. Erica Groshen, former commissioner of the U.S. Bureau of Labor Statistics and now senior economics adviser at the Cornell University School of Industrial and Labor Relations, and Paul Famolari, the assistant executive director for unemployment for the South Carolina Department of Employment and Workforce.
Groshen explained how collection of this information in a standardized manner can help policymakers and researchers see trends in the workplace and know which job programs are paying off for workers. They also can show how new technology is affecting workers, such as which workers appear most vulnerable to losing their jobs to AI.
Famolari discussed how the program has worked in South Carolina, what lawmakers were hoping to learn when they passed the enabling legislation and some of the challenges his agency ran into in trying to implement the law.
Resources
ES: (00:12):
Hello and welcome to “Our American States,” a podcast from the National Conference of State Legislatures. I'm your host, Ed Smith.
PF: (00:20):
The more granular level of data that is offered by enhanced wage records really has the potential to bring significant benefits to just about every single person in this state.
ES: (00:33):
That was Paul Famolari, the assistant executive director for unemployment for the South Carolina Department of Employment and Workforce. He is one my guests on this podcast along with Dr. Erica Groshen, former commissioner of the U.S. Bureau of Labor Statistics and now senior economics adviser at the Cornell University School of Industrial and Labor Relations.
Our focus on this podcast is enhanced wage records and the role they can play in helping state policymakers better understand labor market outcomes for people who graduate with a bachelor’s degree and also those who complete a job training program.
Collecting enhanced employment data involves asking employers for additional information when they file regular reports with their state unemployment office, such as information about hours worked and establishing consistent coding for similar job types.
Groshen explained how collection of this information in a standardized manner can help policymakers and researchers see trends in the workplace and know which college degrees and job programs are paying off for workers. They also can show how new technology is affecting workers, such as which workers appear most vulnerable to losing their jobs to AI.
Famolari discussed how the program has worked in South Carolina, what lawmakers were hoping to learn when they passed the enabling legislation and some of the challenges his agency ran into in trying to implement the law.
There are two acronyms used in this podcast that might not be familiar to all our listeners. One is SOCC, or standard occupational classification code. The other is NAICS, the North American Industry Classification System.
Here’s our discussion, starting with Dr Groshen.
Dr. Groshen, welcome to the podcast. Great to have you.
EG: (02:21):
Thank you for having me. It's an honor and a pleasure to join you and to speak to your audience.
ES: (02:27):
A lot of folks in our audience, legislators, legislative staff, they're pretty sophisticated folks, but they might not be too familiar with enhanced wage records. I have to say that when I began to study some background on this, I was not very familiar with that either. And I wonder if you could just start by explaining what these are and how they're used, and perhaps more to the point, what does enhancing them mean?
EG: (02:50):
Well, you're right. It's jargon. This is pure jargon. Really, it's very simple. These are the records, the files, the data that employers keep so that they can pay their workers. I mean, that's what they are. They have the information that appears on your pay stub, and that's reported to the IRS, the unemployment insurance system, and various other state and government agencies for regulatory and statistical purposes. And at their most basic, they have workers' ID. So your name and social security number will be in there, not necessarily printed out every time, but it's in there. What pay period are we talking about? The wages that were paid to the worker, the name of the company, things like that. States require the employers to submit wage records for their employees, for all of the employees in a particular state, basically for the state to run the unemployment insurance system.
EG: (03:56):
Because if you're then laid off, they have to be able to check whether or not you're eligible. And so the states maintain these files so that on the chance that you're laid off, they can check on this. So it helps them to determine eligibility and benefits. These are really valued by people doing other kinds of work, not just the unemployment insurance system, by the agency I used to have, the Bureau of Labor Statistics and other agencies, because these are the records that are going to be the most accurate, the most complete and fairly timely because all of the employers are doing it and they're doing it in a similar format. And so it's easy to know what to ask of the employers. This is really trustworthy information. So it's transactional information that's available quarterly with just a few months lag. So because it's there and because of the growth of computerized records all around, over the past 30, 40 years, state's use of their own UI wage records has really grown dramatically far beyond just for unemployment insurance purposes.
EG: (05:16):
And basically a lot of evidence-based policy is now based ... Comes out of this, right? That's really taken off. So that includes evaluating the impact of government policies, tracking economic trends, guiding economic development. This is a resource that the states can call upon to decide how to target particular populations, how to deal with problems, see what's working, what's not. And the more state that the states have used their records, the more they want from them. So then they say, "Oh, if we only had this information, then we could look at that program." So sometimes they can merge that information in. Other times they would have to go back and ask for it. When we talk about enhanced records, we're saying enhanced from what the unemployment insurance system was asking for before. So right now, for example, the enhancements that are most valued are first of all, putting them in a common format to make them truly interoperable.
EG: (06:28):
And then really key, not on there now, but many states are working on adding it is occupation and/or job title. Because if you're talking about technological change, if you're talking about work and training needs, then you need to know what the person is doing, what kind of training they need. So occupation job title. Another important element is hours because what matters for a lot of questions was what's the hourly pay, not how much they earn over the whole period. They're both interesting, but you need the hours too. Start and stop date ends up mattering a lot because if you're only getting the number for a quarter, then you don't know how long the person was employed in a very clear sense. Location of work and residence matters a lot because commuting patterns matter for economic development, you want to know where people are living and where they're working.
EG: (07:29):
Those are just some of the enhancements, but when people talk about enhancements, it will at least include those things.
ES: (07:35):
So, as we're talking here with policymakers, both in the executive and legislative side of state government, I wonder what kinds of questions do these enhanced wage records help people who are researchers or policymakers answer, particularly about career pathways as compared to the usual data. And I guess as a corollary to that, to what degree can you talk about the value of these things being hooked to education records? So we know you have this kind of degree and this kind of job.
EG: (08:07):
The real value, I've kind of mentioned this already, is of course that these are really accurate. They're coming from records, this isn't a survey of the workers. And there's some important things you can only learn from surveying workers, but this information is really great for getting at wages exactly, right? Because when you're asked your salary, you round, you may not remember exactly, but this is right off the records, right? And because it covers everybody, you can get much more detailed geographic and demographic if you merge in demographic information. So, age of worker or gender, their training, right? And you can track it over time. So you can see trajectories, which is really important because sometimes a snapshot doesn't really tell you how much difference a program or something is made. So ,what kinds of questions can you answer with this sort of data? Well, which workers are being exposed to AI, for example, artificial intelligence, right?
EG: (09:17):
Which are the workers that are likely to be most at risk in your city, in your state, right? There was a study of where do New Mexico's oil and gas extraction workers come from when the energy prices go up and where do they go to when energy prices are down? So, what occupations do they go to? Do they go out of state? What industries do they go back to or enter that sort of thing. Generally, who's leaving and coming into the state to work? Are they high wage, low wage? Where were they educated? When were they educated? Are commuting patterns changing? So that transportation folks really want to know about that. And so do school districts. There's been studies on federal research grants and patents. How do those influence economic development? As a labor economist, I often say this, but I think it's true that there's almost no question you can think to ask that doesn't have a labor component.
EG: (10:18):
How is this affecting people's jobs? How's this affecting the demand of labor, the supply of labor? So those are some examples. The California Policy Lab is a good example of a place that's making a lot of use of these wage records. So I was just looking, actually, since you asked, at the one study that they did in 2022, and it looked at 10 different work and training programs in the state. And just from this data, they could see which ones seemed to have an effect on employment compared to people who weren't in the program who were otherwise similar, and which ones raised wages. And they could look at both of them, and they went back for, I think it was five years, they went back in of the 10 that they had, the one that really boosted wages the most was certified apprenticeships. As far as employment, the ones that boosted employment most were treatments for dislocated workers.
EG: (11:30):
So, in some ways that makes sense. Dislocated workers may not know how to find a job anymore. It's been a while since they looked for one. They may not know which industries are growing, whatever. So that really helps them. It doesn't necessarily raise their wages that much, but at least they get into the new job faster, right? The apprenticeship is very much about getting people in a position to learn on the job skills.
ES: (11:58):
I wonder if you could expand a little more on, you pointed out California, do you see patterns in other states of what's sort of emerging and how states are using this data to do exactly that, evaluate the effectiveness of the programs they have?
EG: (12:12):
There are all sorts of efforts in this direction. I was just at a conference that the Kohler Initiative sponsored, which is, this is a nonprofit group that is trying to help states exploit their data, combine it with federal data and other data they may have and answer burning policy questions. Gee, there were so many of these papers, all on different things. There was take up of SNAP benefits, right? Food stamps, who takes them up, which outreach methods seem to not only get to the families, but then what happens to the families afterwards? Housing was a big thing too. How can getting people into housing help their employment, right? And which paths to housing is most useful for that? There's just a big push for the states more and more to use the data that they have to help guide these decisions.
EG: (13:17):
And this is really foundational. Wage records turns out are a part of, as I said before, almost every question that you ask.
ES: (13:27):
Well, it's so interesting that what you're saying is really these data are all there. It's just a matter of how they've been collected, how they've been formatted, how they've been distributed. Given that, it sounds like a very, very promising field for not only labor economists, but for policymakers who want to try to enhance the employment of their constituents. But are more states picking up on this, beginning to put these systems into place? I talk elsewhere on the podcast with someone from South Carolina about what they're doing there, but how widespread is this? And is it growing?
EG: (14:02):
Oh, it's definitely growing. It's not only growing because the states very much want to do it, but federal mandates implicitly require this now. So, for example, the federal government has passed a work requirement for recipients, for some recipients of SNAP benefits and for Medicaid. Well, how's the state going to enforce that? They can only do it with records like these. So, they're going to have to tap into the records for that, right? And then think about also, we haven't even talked about workers who are in non-traditional work. So right now, they're actually not covered in these work records, but during the pandemic, we extended unemployment insurance benefits to people from gig jobs. They're doing a lot of cleanup and they're finding that there was fraud there and it's not surprising because states didn't have any records to check people's assertions against. So, the next time we have a recession, there will be, I'm sure, another push to grant benefits to people in gig work, which makes sense because they're the people most likely to be living hand to mouth anyway.
EG: (15:22):
If the states haven't built up a record on people's gig work, then they're going to be in the same position again. And so these records are not only hugely important now, but their coverage is likely to grow, which will make them even more valuable. Other examples are ... Oh, occupational licensing. There are a lot of people who say that occupational licensing is actually detrimental in a lot of situations, right? So how does a state decide what should be licensed and what shouldn't be licensed? Well, having this kind of data, merging in the information on licensing can help you look at what's going on with the employment in states that have licensing and don't have licensing, and allow the states to make an informed choice about which way they're going to go.
ES: (16:23):
Thinking about our audience, state lawmakers, others interested in state policy, what's their role? What would you say to them in terms of what they can do to encourage the collection, or not the collection, so much as the distribution and uniformity of this data, and how it can help them?
EG: (16:43):
There's actually a lot that the states can do, and I would argue really need to do. So we're entering a world, the information age, we're in it, but it's really at its infancy, and we're realizing all the power of these data. We've digitized an awful lot of information. We're digitizing more and more of it. And to get the real value of the information, we need to make it more interoperable, right? We need to be able to put it together from a lot of different places and see the patterns. There are two problems with doing that. One is that if they're not collected in the same way, we may get confused. Actually, let me say, there are three problems. So one is if it's not done in the same way, then we may confuse ourselves. We may see a pattern that's just an artifact of some ways.
EG: (17:46):
Some companies were reporting this information one way and other companies were reporting information the other, or state by state differences too. Another problem is that it can actually be pretty burdensome on the employers to submit all this information, particularly if they have to submit it slightly differently for one request than another. And so a large employer can have 175 different reports that it has to submit because remember the states, for some of our federal programs, it's the states that are asking for it and they ask it in different ways. And then the third problem is some of this information is sensitive either to the workers or to the companies, part of their secret sauce, what occupations they have and what they pay and how they manage their workforce. And it turns out there's a solution to all three of those things. And that solution would be adopting common standards, data standards, and technology standards.
EG: (18:54):
And this is not a new concept. You go back to the Bible. What was one of the first thing that you see in the Bible when the Israelites were making their new society, right? Standard weights and measures. If you're going to exchange things between people, arms leads transactions where you don't actually know the other person that well, then it's much easier to do that if you agree what an inch is or what a foot is or what a pound is. And the same thing is true of data. If we all agree what a wage is, what an hour is. So that's the data step. And then we attach to it a way of transmitting it that's safe. A company keeps the data abiding by those technology standards, giving permissions to certain government agencies to use it and for their purposes, what they use it for, right?
EG: (20:01):
That's auditable and then they can say, okay, we are protecting, say to the workers, we are protecting your data in this way, and that's clear and auditable. And if a user of the data doesn't abide by the agreements, then the company can withdraw permission for that user to have those data going forward. So if we move to that world, then the burden goes down on the employers because the requests come in the form of this particular field that you're already collecting and doing, and the quality of the data goes up. Everybody who uses it knows they're using comparable data, and that their access is assured as long as they abide by all of the requirements, what they can and can't do with it. The data subjects are protected, again, by the protections that are built into the whole system. And then you need, when you're going to have this, you need a governance system for it, but we have some of those already.
EG: (21:11):
So, there are already public private partnerships that set standards for interoperable electrical equipment and interoperable plumbing equipment and paper sizes and paint grades and things like that. There will be some governance mechanism that says, okay, wage records, this is what they are. And when they need to change, that group gets together and they have their big confab and they decide what belongs in a wage record, what the fields are going to be, what are the optional parts. And so that's what we need to move toward. And when we get there, there are going to be huge opportunities for the states and for the cities and the counties and all of that because then they're going to be able to use software that's much more off the shelf because the data fields will be similar. They'll be able to replicate what other states have done that that's of interest to them too much more easily.
EG: (22:15):
They will know what data is easy for employers to provide and what is hard for employers to provide. And if they have a new need that they didn't have before, there will be a governance mechanism to go to petition to say, look, this field should be in there now and here's the reason why.
ES: (22:38):
Well, it certainly sounds like this whole effort has tremendous upside potential and I really appreciate you walking us through all this. Anything else you'd like to share with our audience before we sign off?
EG: (22:51):
Oh, sure. So actually, I want to tell people that there are efforts in this direction and one in particular that's really focused on this is being run by the U.S. Chamber of Commerce Foundation. It's called JEDX, the Jobs and Employment Data Exchange. And right now they developed a system that can be used. They are piloting it in Arkansas and South Carolina to make sure it's ready for prime time or more ready for prime time. And so that's a part of getting this in, that there's a group from the industry side, which I think is really important for credibility and for ensuring that this is going to be manageable. The statistical agencies are very much working with them and supportive of this effort. The Bureau of Labor Statistics has a related program called their Wage Records Program. Thirty-four states have already joined this, so that leaves 16 outside, those 16 I hope will be joining it soon.
EG: (24:06):
What that does is BLS actually takes the state records as they're submitted and after the fact, puts them in as uniform a way as it can, and then the states can exchange those records with each other. And that's pretty important because an awful lot of labor markets span state lines. So think about our big cities. Most of our big cities are near state lines. People go across them all the time, right? It expands the opportunities for research. It also, when more and more states have joined, then this gives BLS, which has not had access to these records in the past, the opportunity of building those into some of its programs, and that will have the really important beneficial effect of allowing BLS to put out much more geographically granular data than it has in the past. That's a big opportunity. So, I would encourage states to reach out to find out about JEDX, to get their state joining the BLS Wage Records Program.
EG: (25:18):
They can also think about setting up something like the California Policy Lab, working with an educational institution to get some of their researchers to ... These really creative researchers who are used to crunching a lot of data and working on policy issues, have them all set up so that when a new question comes up, you've got a team ready to go forth and do the work on it. State of Washington has some groups like that too. Two other things I would say is that it's a lot of this kind of research, both for program and policy evaluation, for our statistical system and for long-term research on consequences and causes of policy issues. The funding for that has been really dramatically reduced, not just in the past year, although the past year has been problematic, but it was declining even before that. So if the states really want this, they have to be sure that their congressional delegations, their senators and their Congress people are speaking up for these data and these programs, because otherwise it could disappear.
EG: (26:36):
If you're not actively on the hill arguing for these things, they can go away.
ES: (26:44):
Well, certainly, yes. NCSL is well positioned to help with that sort of thing. And I think that your message about states joining together so that they can learn from each other, so it's not only sort of overlapping districts, but what someone's doing in Arizona might be something people in Michigan might learn from and vice versa. So I think that's all really good advice, and I thank you for that. And thank you so much for taking the time to be a guest.
EG: (27:13):
Oh, it's my pleasure. And thank you again for the opportunity to talk to you and your audience, and I look forward to hearing all the cool things the states are doing going forward.
ES: (27:27):
I'll be right back after this with Paul Famolari from South Carolina.
Speaker 3 (27:37):
When it comes to podcasts, only one organization NCSL keeps a focus on the people, policy, and politics of state legislatures. Each episode of our podcast offer behind the scenes insights into the legislative process. First, the program Our American States provides in- depth discussions twice a month on key state policy issues shaping our nation, or explore bipartisan dialogue with our special series Across the Aisle, where diverse political perspectives converge to create constructive conversations in state legislative chambers. And for history buffs, we've also produced a six-part series called Building Democracy: The Story of State Legislatures. Listen to learn more about the creation and development of the first branch of government in the United States. Stay connected, stay informed. Subscribe to these programs now on your favorite podcast platform. Learn more at ncsl.org.
ES: (28:52):
Paul, welcome to the podcast.
PF: (28:55):
Thank you for having me, Ed. It's a pleasure to be here.
ES: (28:58):
So Paul, I wonder if you can start out by telling us a little bit about South Carolina's experience with enhanced wage records. And what questions were lawmakers trying to answer when they expanded wage reporting?
PF: (29:11):
The idea of adding SOC [standard occupational classification] codes to unemployment insurance wage records, it was batted around about 10 or so years ago, but at that time there were some concerns that the requirements would be too burdensome on the business community. Since then, however, momentum grew for the benefits that could come from the collection of this additional information, especially as lawmakers started to grapple with how the state would find the workers with the skills necessary to fill the thousands of jobs that were open statewide, particularly as they peaked coming out of the pandemic in 2022. At that time, there were over a hundred thousand job openings in South Carolina. Our business was booming coming out of the pandemic, but businesses were becoming more and more vocal about the challenges that they were facing in Finding qualified workers really became a constant refrain on their part. And additionally, the state really didn't have the means to provide timely and complete information about how post-secondary graduates were meeting the needs of South Carolina's economy.
PF: (30:17):
While we were able to provide information on the number of students graduating with degrees in a particular field, there was limited information about whether those graduates were, one, staying in South Carolina, two, working in South Carolina, and three, working in a field related to their area or area of training. So we were really lacking that level of granular detail. Those factors, I think, along with a couple of other legislative interests, particularly a desire to better assess or create a return on investment of the various degree and certification programs here, as well as a desire to provide better information to students, parents, and teachers, so they can make more informed decisions about a student's future. That really drove the creation of the statewide Education and Workforce Development Act in 2023. And that directed my agency, the Department of Employment and Workforce, to begin the mandatory collection of both a standard occupational classification code or SOC code, and the hours worked for each individual as part of our quarterly UI wage record collection.
PF: (31:29):
Once that bill was signed into law, we started working immediately on implementation and we've been running ever since.
ES: (31:36):
Well, I have to say that I've done a lot of podcasts that have either been about workforce issues or have touched on workforce issues over the last several years. And I think you just hit on most of the topics that have come up on those because this is something that clearly all over the country is a great concern to people. One thing we know about legislation, every piece of legislation everywhere is it's often easier to pass it than implement it. So, my question is, what were the biggest challenges you faced in South Carolina as you tried to implement these enhanced wage records? And I mean, were they technical computer stuff, were the administrative, dealing with employers, things I can't think of?
PF: (32:19):
Yeah, that's a great question. And truthfully, it's a little bit of everything that we encountered after the implementation. We were pretty fortunate at the outset that we had a little bit of a leg up on the technical side. We modernized our unemployment insurance tax system in 2018, and that was round about that time that the state was first considering adding SOC codes to UI wage records. So, at that time, we had the, as it turns out, foresight to build that capability into our system and so that we would be able to add additional required elements to our UI wage reports if we needed to down the line. And that turned out to be really beneficial when the legislature finally got around to passing this. So though there was an expense for our vendor to incorporate SOC codes into the benefit system or into the tax system rather, we were able to deploy the functionality pretty quickly.
PF: (33:14):
The legislation was signed by the governor in May of 2023, and it was effective starting with first quarter wage reports of 2024, which wouldn't be due until April of 2024, but we soft launched the capability to collect SOC code information voluntarily just a few months after the bill was signed. And so when employers came in to file their third quarter 23 reports, they saw the field for SOC codes, and if they wanted to, they could provide us with one, although it wasn't required. So we had that already. And I think that was a huge win for us because it limited our focus on the technical side. And that led us pivot to what I think was the biggest challenge, which was getting the business community buy-in and getting them educated on the requirements, especially because a lot of the people who might be at a high level within a business that might really appreciate all the benefits of SOC codes and hours worked being added to the wage records.
PF: (34:12):
They're not the folks who are actually doing the paperwork every quarter. So educating those folks became a really significant priority and a really big focus of our efforts. So we rolled out a pretty aggressive communications plan, and that covered everything from creating a dedicated SOC code page on our website that had a host of resources, direct resources from the Bureau of Labor Statistics of what SOC codes are, what each one is and their detailed definitions. We created some dashboards and tools on the website that would allow an employer to enter in, say, their industry or their NAICS [North American Industry Classification System] code. If they knew their NAICS code, they could enter that in and the dashboard would return the most common SOC codes associated with it. We really kept adding information to that page so that we could always direct employers to it so that they could engage in as much self-service as possible.
PF: (35:08):
And then we would add to it over time. We eventually created a frequently asked questions section on that, and we would add questions and answers as we learned from the community and got feedback. We also created a very in- depth customer service representative training because we knew we were going to get calls. And so we trained a handful of staff to be really knowledgeable about the topic and be able to walk employers through the requirement. We developed direct correspondence to all the business community. We took advantage of our yearly unemployment insurance tax rate setting process to include a detailed flyer in every single tax rate notice that went out to all 165,000 businesses. And then biggest thing is a ton of speaking engagements. I would talk to anyone and everyone I could about the requirements and how it would benefit the business community. And so whether it was Society of Human Resource Managers, our CPA association, we worked closely with the state and also local chapters of the Chamber of Commerce.
PF: (36:14):
Big one was the Payroll Reporting Consortium, your Equifaxes and ADPs and all those folks. They were huge partners in this. And we wanted to get their feedback about how we could make this successful for everyone. And what I learned over time was that the key was to focus on the value that SOC codes and hours worked would provide to the businesses themselves, that the granular level of detail would allow businesses to get a clearer sense of what's going on in their industry, both at a state level as well as within their local communities. And I would tailor that pitch depending on the group I was talking to. I think the biggest and most commonly voiced concern from the business community surrounded confidentiality. Employers were worried that all of a sudden everyone would know how much or maybe how little they were paying their employees and that that would somehow create a reputational issue or anything along those lines.
PF: (37:07):
And so as a result, I built into my presentations and would always emphasize the very strong confidentiality requirements of state and federal law regarding unemployment insurance data and that we were just not going to be sharing this data. Any of the reports or dashboards or tools that we would ultimately be able to create from this data would not allow anyone to identify who that business was. It would all be aggregated information, and that really went a long way to calming people down. I think those were the big challenges that we encountered and some of the ways that we worked on addressing that.
ES: (37:43):
Well, that's good, I think, for other states to know. It makes sense that that's going to be the biggest hurdle, but it does make a lot of sense. Now that you've had some experience with it, the big question, what have you learned? How does it help you understand your workforce and workforce needs?
PF: (38:01):
I think we're really excited. We're already learning things and I think we're going to continue to learn so much. We have a division within our agency, a labor market information group, or LMI for shorts, and they've really jumped into exploring career pathways in particular, starting off with healthcare, which I think is a big driver of the state's economy and perhaps every state's economy at this point. And having SOC codes is going to help inform stakeholders of the variety of career pathways that people actually take rather than relying on existing crosswalks between degrees or occupations or anecdotes. That was one of the examples I would always cite in my presentations that prior to SOC codes, we may know that a hundred people worked in a hospital, but we didn't know what those hundred people did. We didn't know necessarily how many of those were doctors, nurses, administrative staff, attorneys, HR professionals.
PF: (38:56):
But now with the SOC codes, we're going to have a clearer sense of what kinds of occupations are making up particular industries and how can people advance from position to position up the career ladder through the course of their lives. And I think that's going to be a real big benefit to the state. Of course, we are just at the beginning of this research. It's likely going to take us a couple of years of soft code collections to really be able to fully examine the early, mid and late career movements along various pathways. But we're hopeful that over time, we'll be able to see not only the occupations that individuals are moving between, but also the resulting wages that they earn to determine if people are moving laterally or possibly getting promotions or even returning to school to be able to move into higher skilled positions.
PF: (39:43):
And of course, with better data sharing with our partners and other state agencies like education or the technical college system, you might be able to see, oh, this person started out at an entry level position, they went and got a certificate, and now they got promoted into a different position, making more money. And that's great information to better assess the efficacy of our education system. There are also challenges though in fully mapping out career pathways in addition to needing more years of data. For instance, what we've learned is that not all businesses classify employees in the same way. So, you may have a wide variety of reporting of standard occupational codes for individuals who are ultimately performing the same underlying job functions, and that may result in occupational webs rather than a traditional career pathway. It may also show that occupation titles that tend to perform similar duties but may be more lateral movements.
PF: (40:43):
So, we're trying to figure out ways that we can kind of clean that up, and that's something that our LMI team plans on monitoring and exploring as we get more quarters of data available to us.
ES: (40:55):
Well, I think your example about the health care industry is great because most of us who are not very sophisticated at this, look at that industry. It's a $5 trillion piece of our economy. Oh, there's X number of jobs in it, but as you're pointing out, not knowing what those jobs are and the more granular data would be really helpful. Sort of to your last point there, how is this changing the way states think about, or your state thinks about ,unemployment? It's not just people are employed or unemployed, it's how much they're making, where they're working, how their earnings are progressing, all that kind of thing. Is it going to change the way the state interacts with workforce programs and that sort of thing, do you think?
PF: (41:40):
Oh, I think absolutely. Employment outcomes are one of the very first things that we started looking at after the passage of that legislation two years ago or three years ago rather. And we're really starting to dive into the data that we've gotten from our first eight quarters of the requirement being effective. There's been a lot of great work collaboratively between Do and our educational partners at both the secondary and post-secondary levels to share data, to answer some of those kinds of questions. So last year, we began sharing data with our State Department of Education, which covers K through 12, to determine what high school students who do not move on to college are doing. Previously, we produced a supply gap analysis that compared the number of graduates from post-secondary institutions to the projected demand for those types of workers in South Carolina. Now that was a great start, but there was a large segment of the population that does not have a higher education degree and understanding where those individuals were working using the enhanced wage records that we now get, allowed for that supply gap analysis to be beefed up and to be extended further to the high school graduate level.
PF: (42:54):
We can actually now use actual high school graduates experiences and compare that to demand for various occupations to identify areas of potential shortfall. I think another highlight of the SOC codes has been the ability it's provided us to provide more insight on whether individuals are working in field, particularly for our state technical college system graduates who may have gone there specifically for a particular industry or certificate. Our in-house research economist has shared with me that there are some interesting patterns beginning to emerge in certain occupations that have been in shortage. For example, we have a dental hygienist shortage in the state, but if you look at the graduates in that field and what their occupation is currently in our wage records, you see that a very high degree of retention in South Carolina and a high degree of staying in field. So those folks who are getting that dental hygienist certificate are staying in South Carolina and they're staying as hygienists five, 10, and even 15 years after they graduate.
PF: (43:56):
So, one possible policy response for lawmakers might be, "Well, we should be putting more people through a dental hygienist program if we want to increase that supply." And you can contrast that with something like an associate's degree in childcare management. By that 15th year postgraduation timeframe, fewer than a quarter of those graduates are still working in childcare occupations, at least according to the initial data that we have. So, a potential policy response there would thus be quite different from that of the dental hygienist shortage. Maybe you don't want to funnel as many people through that program if retention is ultimately your goal. So those are the sorts of things that I think we want to be looking at and will be expanding going forward, but right now with eight quarters worth of data, that's some of the stuff that our LMI team and our research economists are starting to discover.
ES: (44:48):
So often when you implement legislation, you discover that there are some limitations to what you're collecting. And I wonder, as you collect this data, set up the system, are you finding additional information you'd like? Are there gaps in there that you'd like to fill in?
PF: (45:06):
There are definitely some limitations. We were very lucky in that the percentage of employers providing SOC codes has grown each of the last eight quarters. So every quarter has been a little bit more than the quarter before. So we started off with the first quarter of 2024 and we had a little over 50% of the state's employers were providing SOC codes for employees. We now have more than 80% of employers providing SOC codes with the fourth quarter 2025 reports that came in. And we now have a SOC code associated with nearly 85% of our reported employees. That's fantastic. However, just because an employer provided a SOC code does not necessarily mean they provided an accurate one, and that's something that we're looking into. So for example, we've noticed a tendency of employers to overutilize generic SOC codes, like the ones ending in 99, just kind of a catchall bucket for various occupations.
PF: (46:03):
For instance, the SOC code 51-9199, it's a very popular code that our manufacturing industry is reporting a ton of, and it equates to basically production workers, all others. So, we know that they are doing something in production in a manufacturing firm, but we don't know what kind of production work they're doing. So, we're trying to figure out ways that we can fine tune that. And our LMI shop has been doing some outreach to those employers who we believe may be overrepresenting certain SOC codes. And we've been offering up virtual meetings with their HR staff or whoever's responsible for reporting to see if we can assist them in better identifying the SOC code for their workers. We're also seeing an over reporting of management occupations versus supervisors. Those are two distinct SOC code categories. And so while federal BLS data may show that about 5½% of South Carolina workers are in some type of management occupation, our most recent quarter had about 9 1/2%.
PF: (47:05):
So, it's pretty significant difference between the state level data and what BLS would expect us to have. And we suspect that that's likely due to employers putting down a management occupation when it's really a frontline supervisor. And that's educating the employer community about the distinctions within the SOC codes about that. And then also while the initial focus of our efforts was on SOC codes, we also had rolled out the requirement at the same time to provide hours worked. And this may be a lesson learned on my part. Intuitively, I assume that people would understand hours worked in a quarter more naturally than they might SOC codes. And so that's why we focused the vast majority of our outreach on SOC code education. We have really good compliance with employers providing hours worked, but the accuracy of that I think has been a little bit less or a little bit worse than with SOC codes.
PF: (47:57):
I think many employers are putting down zero hours worked because maybe they're trying to reflect the person's salaried or they're putting down an eight or 40 because they're thinking hours per day or hours per week and our instructions are hours for the quarter. So, if you're making 40 hours, multiply that by 13, those are the hours that you provide. And oddly, we have a whole lot of people who are putting completely impossible numbers of hours, thousands of hours worked in a quarter. So, we do have a lot of work with the community still ahead of us to kind of smooth that out. But you asked specifically about additional information. I don't know how likely it is from a policy level, but I do know that our research economist has always wished that we could have added a location so that you could have a county code for each individual.
PF: (48:51):
There are 46 counties in South Carolina. You assign one through 46 to each county and an extra for out of state, and that would give us even more data. And we've got some very large employers in South Carolina that are nationwide. We've got hundreds of Walmart locations, for instance. They report thousands of employees, but we don't know where any of those people actually are working. So that's sort of a wish list item from a research economist standpoint.
ES: (49:17):
I had to laugh when you were talking about employment codes. I worked in the newspaper business for a long time, and you probably know this, but most people probably don't know that newspapers are classified as manufacturing enterprises. So, they could have put in that manufacturing enterprise for me as a reporter or editor, and I guess that would have technically filled the bill. You can see how easily that could get confusing and maybe not so useful. Now thinking of that, you're sitting down at a meeting with some people from another state, legislators or people who work in the labor department. What would you tell them about your experience so far? What would you tell them that ... Was it worth the effort? Is the information you're getting helping you in terms of how you are able to approach the issues that you deal with as a result of this?
PF: (50:03):
Well, I would tell them absolutely that it's very worthwhile information to have. And I think it's already born some very positive fruits and very promising research avenues to look down. I think that's only going to increase over time, both as we get more data and we get more accurate data as we work with the community. But I think I would break down our experiences into two broad categories. First, from a strategic perspective, like I mentioned, I would emphasize all of the benefits of it to various stakeholders in the state, from businesses, students, teachers, policymakers of getting SOCC data, getting hours work data. I shared the experience in my presentations that I'm a parent, I have a 14-year-old, figuring out what her future is going to look like and having better access to more accurate, granular level workforce data. That would be really reassuring to know that I'm providing good guidance to my daughter as she goes down her career pathway.
PF: (51:03):
And the more granular level of data that is offered by enhanced wage records really has the potential to bring significant benefits to just about every single person in this state, somehow or other. And it's important to have those benefits constantly at the forefront of your policymaking discussions. And then if you do pass legislation to require this at the forefront of your conversations with your stakeholders within the UI program to get as much compliance as possible. The second broad category though is there are some unique challenges, I think, of using the unemployment insurance program for this kind of data collection. The UI program, for one, only covers traditional employees, so you're not going to capture independent contractors, the self-employed, gig economy workers. Additionally, each state carves out exemptions from UI coverage that might exclude a traditional employee for a variety of policy reasons. So, selling real estate, I think is exempt from South Carolina.
PF: (52:01):
I know, for example, if someone who goes and gets their CDL, their commercial driver's license, and gets a job transporting cars from one dealership to another dealership, those people aren't covered under the UI program. They're not considered employees, even though by a traditional definition, they would be. So, they're not included in our records. And if you were looking to see the effectiveness of a particular program, CDL certification or training, that person's not going to show up in our data. I think further, we talked about confidentiality requirements. I think that's just setting expectations for education and workforce partners. They see that you're collecting additional information on these wage records. You've got social security number, you've got wage information, now you've got SOC codes, now you've got hours worked. Everyone wants a piece of that data and all the things that they could do with it. There are very stringent confidentiality requirements that do restrict the release or sharing of the information that we collect as part of the unemployment insurance program, and we can't really release anything that would identify a particular individual or business.
PF: (53:07):
So, level setting and making sure everyone understands that there are certain limits and really we're going to be sharing aggregate information. Additionally, the UI program is funded by federal funds almost exclusively. We get a federal administrative grant from the Department of Labor, and it's got certain requirements about what are allowable costs. And I think a lot of the work involved with SOC code or hours worked or other enhanced wage record elements and promoting compliance is not work that can necessarily be covered by the UI administrative grant. And we were very fortunate that we had some state funds that we could tap into to cover those expenses, but I would definitely advise states that if they're going down this road, be sure that you provide the agency implementing this with state funds to cover the expected costs, both of the initial implementation, as well as for a couple of years after post-implementation because as we're experiencing now, your outreach rep efforts, your data cleanup efforts, that's all going to continue for years to come after you initially implement the functionality.
PF: (54:11):
And I think that's particularly the communication outreach. That was the most impactful work that we've done in generating compliance, and that's going to keep going for years. And then lastly, a plug to any state that is in the process of monetizing either their UI tax or benefit system, encourage states to be forward thinking and think about the potential of new information you may want to collect and be sure that you work with your vendors to give yourself as much future flexibility as possible when you're designing these systems so that you can easily adapt them when the time comes or when the legislature makes some changes to your program.
ES: (54:52):
Well, that all sounds like great advice, Paul, and thanks so much for sharing the perspective of South Carolina's experience with this. And take care.
PF: (55:00):
It's my pleasure. Thank you, Ed. I really appreciate the opportunity to share.
ES: (55:07):
I've been talking with Dr. Erica Groshen and Paul Famolari about enhanced wage records and their effect on state policymaking. Thanks for listening.
Search for NCSL Podcasts wherever you get your podcasts. This podcast, “Our American States,” dives into some of the most challenging public policy issues facing legislators. Our occasional series, “Across the Aisle,” features stories of bipartisanship. Also, check out our special series, “Building Democracy,” on the history of Legislatures.