In June, the Department of Commerce released Department Administrative Order (DAO) 216-26, titled "Disclosure Avoidance for Statistical Products." This order established new policies around how we protect the confidentiality of individuals and organizations when we publish statistics.

The Census Bureau has always faced a challenge: how to publish useful statistics while protecting the confidentiality of the data and the people behind the numbers. In recent years, including after the 2020 Census, we used a technique called differentially private noise infusion. This method adds random “noise” – data additions or subtractions – to data to make it harder to identify individuals.

While this approach was meant to protect confidentiality, it led to concerns that adding noise made the statistics less accurate and less trustworthy. The new policies – developed by the Department in conjunction with the Office of the Undersecretary for Economic Affairs and with technical guidance from the U.S. Census Bureau – aim to address these concerns. The order is clear: the top priority is to provide the public with accurate statistical information.

We know that our data users have lots of questions. We are currently working to develop clear guidelines on how the Census Bureau will implement the order’s policy requirements. Here’s where we are now.

The administrative order allows two methods of disclosure avoidance:

1. Coarsening. Reducing the detail of data (for example, grouping ages into ranges or rounding). The new order names coarsening as the preferred disclosure avoidance method for all statistical products.
2. Suppression. Withholding certain data from publication. This is permitted only as a last resort.

Under the administrative order, a third method of disclosure avoidance called noise infusion is no longer allowed.

3. Noise infusion. Adding random changes to data values. This includes statistical methods such as record swapping and the use of synthetic data.

Going forward, we will rely solely on coarsening and suppression methods. Coarsening and suppression can reduce the amount and detail of data published, especially for small geographic areas or population groups. These are the places where confidentiality risks are highest. We are exploring ways to meet the data needs of these communities while protecting their information.

The order applies to all statistical products disseminated by the Census Bureau. It covers any information produced by our staff or contractors using data protected under the confidentiality provisions of Title 13.

However, not all research or presentations that use our data are considered “disseminated” by the Census Bureau. For example, if an external researcher produces a report from a Federal Statistical Research Data Center or if Census Bureau employees produce a conference presentation, journal article or working paper, they may use other disclosure avoidance methods with a disclaimer that their views are their own.

The order also applies only to new products going forward. Products that have already been published, such as the 2020 Census or the 2024 American Community Survey (ACS), won’t be affected. Internal microdata files with noise already infused can still be used if no additional noise is added to new products derived from them.

Most Census Bureau products use a variety of disclosure avoidance methods, and not all of them rely on noise infusion. For many products, the new policy won’t affect release schedules. For those impacted, we are working to minimize delays, especially for high-priority data:

• 2030 Census. Decennial census data are the hardest to protect due to the data’s scale and detail. This new order has shifted the research agenda, but our key objectives remain the same: better assessment and communication of disclosure risks and better understanding of data users’ needs. We plan to release a demonstration data product in mid-2027 for public feedback that will inform decisions for census testing and ultimately the 2030 Census.
• ACS. We’re researching new disclosure avoidance methods. Transitioning to methods compliant with the new order will occur gradually, with a full transition by 2029.
• Demographic surveys. Many of our demographic surveys use noise infusion methods. We will perform disclosure risk assessments for these products and research and identify new methods that are compliant with the order.
• Economic indicators. Most of our economic indicators are protected by suppression and won’t be affected. The Quarterly Financial Report is affected and will switch back to suppression.
• County Business Patterns. These data have historically used noise infusion, and we are exploring alternatives.
• Longitudinal Employer-Household Dynamics (LEHD). These products use state unemployment insurance data as inputs. Products produced with noise infusion required by states in our data-sharing agreements will continue until these agreements are updated.

The new order marks a positive shift in how the Census Bureau balances confidentiality, accuracy and data availability. The Census Bureau will continue to regularly engage with stakeholders to shape our future products and policies to ensure our products meet your needs. Implementing the policy will involve tradeoffs and resource implications but will allow us to meet our most important goal: ensuring accurate and objective statistics.

By working together, and by prioritizing accuracy and transparency, we will continue to provide the high-quality statistics our nation has relied on for decades.