Politicians claim regulation hurts small businesses. When you look at real-world data, the truth is more complicated

Politicians like Harris and Trump often lament that small businesses are victims of overly burdensome government regulations. However, our research found that medium-sized businesses are actually hit hardest by the costs of regulatory compliance. Furthermore, agencies rarely assess whether the rules they create achieve their intended goals ex-post—or the real cost of regulation to businesses and the economy.

Public officials should start taking a more data-driven approach to regulation. To this goal, we built an employment-data-driven model that estimates the overall labor and capital costs of regulatory compliance for U.S. companies, using data on more than a million businesses from the U.S. Bureau of Labor Statistics and the Occupational Information Network, or O*NET, a government website that describes the tasks involved in each occupation using information provided by private sector employees and employers. While there are many costs involved in adhering to regulations, we look at one of the biggest: paying employees to manage regulatory compliance (and related equipment expenditures). 

About a third of all U.S. occupations include some regulation-related work, according to O*NET’s description of tasks performed by each occupation and our analysis of the extent to which these tasks are tied to regulatory compliance. These include activities like reviewing regulations, conducting internal audits, and modifying monitoring processes to ensure regulatory compliance.

Some workers devote most of their time to handling regulations: compliance officers, environmental specialists, and lawyers, for example. Other workers practice compliance as one component of their profession, such as nurses who deliver care following regulatory guidelines regarding patient safety and privacy.

We found that regulation-related tasks account for up to 3.3% of total labor costs for the average American establishment—costing all companies combined a total of $239 billion in 2014 ($289 billion including equipment). To give a sense of the scale, U.S. gross business income taxes amounted to $353 billion in the same year. Yet regulatory burdens vary widely across firms of different sizes. Midsize firms with around 500 employees spend, on average, about 47% more on regulatory compliance costs per employee than small firms—and 18% more than large firms. 

Why regulation hits mid-sized firms the hardest

Some regulations only apply to medium to large-size businesses and exclude smaller businesses. This can facilitate entry by small firms but creates a big hill to climb as these businesses wish to expand. 

Meanwhile, large firms benefit from economies of scale that allow them to hire dedicated compliance specialists and put systems in place for managing compliance and reporting efficiently. These practices are more cost-effective than spreading the task among many nonspecialized employees. Large firms may also have a hand in shaping the rules to limit competition.

The uneven impact of government regulations creates a hump that companies must climb as they grow. The high regulatory compliance costs faced by midsize companies could prevent their growth, or even deter them from trying to grow. Companies sometimes bunch below the size threshold at which onerous regulations would begin to apply.

Many regulators do have good intentions, and many regulations do provide benefits, including protections for employees, consumers, and the environment. A model driven by real-world data can help weigh these benefits against the costs. While government agencies (roughly) estimate the costs and benefits of regulations before enacting them, they rarely conduct follow-up analyses to validate their estimates. Follow-up research sometimes finds that regulations that initially seemed promising ultimately turned out to impose costs exceeding the benefits. For example, the Sarbanes-Oxley Act of 2002 was intended to prevent accounting fraud by imposing personal and potentially criminal liability on senior executives when firms’ accounting statements are inaccurate. Work by our group and others suggests that complying with this regulation is time-consuming for executives, lowers stock prices, and likely costs investors more than it saves them in reduced accounting fraud, although other assessments are more optimistic.

Agencies could also test whether individual regulations are truly worth the expense through pilot programs. Before adding a new sanitation requirement for restaurants, for example, a city’s public health department might first roll out the regulation to a subset of establishments. The agency could then track how much the costs of compliance go up, and whether rates of food-borne illness go down relative to restaurants that are not operating under the new regulation. 

This sort of targeted analysis has a precedent. In 2015, the Securities and Exchange Commission implemented a new rule on tick sizes (the smallest increment by which the price of stocks or other exchange-traded instruments can move) for only a subset of publicly traded companies and then tracked how the regulation affected stock trading to inform future public policy changes.  

Our analysis suggests that the conventional political wisdom about which businesses are most affected by regulations is wrong. The smallest firms do not bear the greatest burden, nor do the largest, but rather mid-sized firms with the potential to grow. There may be opportunities to reduce burdens on mid-sized firms by scaling back regulations that produce limited public benefits, or through targeted tax relief.

More must-read commentary published by Fortune:

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.