Explore Harvard's Nieman network Nieman Fellowships Nieman Lab Nieman Reports Nieman Storyboard

Are regulations killing jobs? Here’s how to find out.

ASK THIS | July 16, 2012

NYU Law School’s Institute for Policy Integrity has put out a tip sheet for reporters to get to the bottom of assertions that government regulations, such as ones by EPA, are killing jobs. And also to check on claims to the opposite – that regulations are creating jobs.

By Michael Livermore
With unemployment rates slow to recover after the crash of 2008, regulations—specifically the environmental kind—have been scapegoated as a major roadblock. Journalists confronted with claims of jobs “killed,” “crushed,” and otherwise tortured by EPA rules might find it difficult to quickly determine their truthfulness. Likewise, competing statements about jobs that regulations could “create” are somewhat murky.  
To help clarify what to look for when discussing jobs and regulations, the Institute for Policy Integrity at NYU School of Law put together a tip sheet for reporters. In it, we suggest questions and framing terms that can create a baseline standard for evaluating these claims.

One problem with claims about jobs and regulations is that sometimes they are made without supporting documents. If journalists hear claims like jobs are being “killed” or “created” by a certain regulation, the first question should be: What study does that statement rely on?  
Once a study is cited, the real work begins.  
  1. The first question reporters must ask when reporting on a specific study that makes claims about jobs and regulation is: Does your study describe net job effects?

    At issue here is the scope of the study. If the scope is small—only accounting for one or two sectors or a small region—the results cannot be attributed to the national economy. It might be true that a regulation causes some number of layoffs in one part of the country, but they might be (and usually are) offset in another part of the country.  

    So claiming that a rule “kills” jobs is inaccurate. Jobs may transition, shift, or move as a result of a rule, but unless the study captures all the consequences nationwide of implementing environmental protection, it cannot properly state jobs are “killed.” 

    The same is true of the opposite side of the coin—unless studies can show the results of a regulation “create” jobs across the economy, the most that can be claimed is that jobs are shifted, or transitioned into a new or different sector or region.  

    The shortcut for reporters is this: To correctly use phrases like jobs “gained” or “killed,” a study must consider all employment repercussions and offer a net result. If you can’t tell from the write up, ask.  
  2. The next clarifying question for job impact studies is: What kind of jobs are we talking about?  

    Employment-impact studies have a tendency to conflate new, permanent positions—those that will be filled for the foreseeable future—with employment years—single years of fulltime employment for one person. If a study tries to describe employment years as jobs “created” or “killed,” it misleads its readers into thinking new, permanent positions are either gained or lost.  

    Of course no one wants to see layoffs in a downturned economy, but it is important to know what kind of layoffs is predicted, if any. Temporary unemployment and long-term unemployment have very different economic consequences, obviously, and must be considered differently. But most of the models used to try to predict employment effects do not deal with this distinction, and that can create confusion for readers.  

    To clear the air on these two points, journalists can ask for the job numbers reported to be separated into new permanent jobs and employment years, and for layoffs to be separated into estimates of long- and short-term.  
  3. Everyone learns in science class that if an experiment cannot be replicated, its results cannot be fully trusted. In the context of economic studies on jobs and regulations, if certain assumptions aren’t made public, the “experiment” cannot be re-run and the results must be questioned. So reporters should ask: Have the model, modeling choices, data sources, and assumptions of the study been made public?

    As we say in our tip sheet: The results of job impact models are sensitive to model structure (which includes the equations used in the design of the model and the underlying assumptions of the model) as well as the data inputted into the model. All models make simplifying assumptions. Changing the underlying assumptions of a model can dramatically change model results. 

    The punch line for journalists is that any analysis that does not make its modeling choices public should be questioned just as you would a scientific study that doesn’t reveal its experiment process. 
  4. The most important question when examining these kinds of job-impact studies is: Where do hiring and layoff s fit within the larger cost-benefit analysis of a regulation?

    The point of putting public health and environmental protections in place is to protect public health and the environment. Those goals bring with them many economic costs and benefits that have nothing to do with jobs. In many cases, these other elements amount to much greater consequence than a rule’s effects on employment.  

    An example from our tip sheet:  EPA estimates that its rule to reduce mercury contamination from power plants will generate benefits of about $90 billion, as well as costs of about $10 billion. EPA predicts that 8,000 workers would be hired permanently and about 50,000 temporarily, to comply with the rule. Compared side by side, it’s easy to see that jobs are only a small portion of the overall economic picture.

    Job impacts are important for the individuals affected and should be given appropriate weight in the decisionmaking process, but they shouldn’t be made to seem like the only factor.
Reporters covering these kinds of studies should ask about the comparison between the job impacts and the overall economic impact of a regulation.  
The bottom line is, if regulations ever created the kind of massive job “destruction” that is often claimed, we would be in quite a pickle, considering the number of environmental regulations in place.  Instead, these rules end up not having a big impact on employment at all.  

The NiemanWatchdog.org website is no longer being updated. Watchdog stories have a new home in Nieman Reports.