The survey coverage for the 2015 Survey of Public Pensions: State- and Locally-Administered Defined Benefit Data includes public pension systems administered by state and local governments throughout the nation. For Census Bureau statistical purposes, a public pension system is one that is financed by a separate accounting fund of the administering government, excluding pay-as-you-go insurance plans. It must have some type of assured revenue stream or dedicated revenue source other than appropriations from the administering government. Defined benefit’ plans provide a specified monthly benefit in retirement. Retirement benefits are calculated through a plan formula that considers such factors as salary and service. This is different from other plans that do not promise a specific amount of benefits at retirement. Retirement income based on the employee or the employer (or both) contributions to the employee's individual account that are invested on the employee's behalf and depend on investment performance.

Other criteria exist for membership, such as funding and organization. A public pension system's members must consist of current or former public employees who are eligible for inclusion in the Employment component of the Census of Governments (CoG). A pension system must have at least one separate identifiable fund within a recognized government unit, and it must be funded completely or partially with public contributions. A pension system must also be recognized as a government unit (as defined by the Census Bureau) that provides revenues, expenditures, financial assets, and membership information for public pension systems.
Each state- and locally-administered pension system is considered an agency of the corresponding state government; however, the information in this publication reflects only the pension system portion of revenues, expenditures, and assets.

The sampling frame is the 2012 Census of Governments file updated with births, deaths, and mergers since fiscal year 2012. There were 299 state-administered pension funds and 6,000 locally-administered defined benefit public pension systems, all of which are represented here.

For further information, please refer to the 2015 Survey of Public Pensions 2015 Survey Methodology [PDF, 221KB].

Data Collection - Collection period and methods

Dates of Collection:The following are important dates in the data collection process.

Mailout: Octoberr, 2015

Follow-up Mailout: November, 2015

Data in these files are based on information obtained from the 2015 Annual Survey of Public Pensions: State-Administered Defined Benefit Data. Forms were mailed to the 299 state-administered funds and 6,000 locally-administered defined benefit pension systems in the public pension system universe. Staff contacted non-respondents through a follow-up mail-out and follow-up telephone calls. Approximately 45.2% of the respondents chose to submit their data via the web.

Data for survey year 2015 encompass fiscal years that ended between July 1, 2014, and June 30, 2015 (FY2015), and do not reflect data for the entire calendar year of 2015. There are exceptions to the fiscal year rule for the state pension systems in Alabama. Michigan, and Texas. For systems in these states, the fiscal year moves beyond the June 30 cutoff. The data for survey year 2015 covers the fiscal year ending August 31, 2015 for Texas and September 30, 2015 for Alabama and Michigan. These exceptions are made to better align the pensions data with the State Finance Survey.br />

The Census Bureau collects these data by law under Title 13, U.S. Code, Sections 161 and 182.

For further information, please refer to the 2015 Annual Survey of Public Pensions: State-Administered Defined Benefit Data 2015 Survey Methodology [PDF, 221KB].

Data Processing - Editing, imputation, and sampling error

Editing

Editing is a process that ensures survey data are accurate, complete, and consistent. Efforts are made at all phases of collection, processing, and tabulation to minimize errors. Although some edits are built into the Internet data collection instrument and the data entry programs, the majority of the edits are performed after the case has been loaded into the Census Bureau's database.

Edits consist primarily of two types: consistency and a ratio of the current year's reported value to the prior year's value. The consistency edits check the logical relationships of data items reported on the form. For example, if a value exists for the number of retirees receiving benefits because of age or length of service then there must be a value reported for the amount paid. The current year/prior year edits compare by item code the data reported for the current year with data reported for the prior year. If data fall out of acceptable tolerance levels, the item is flagged for review.

For both types of edits, the edit results are reviewed by analysts and adjusted when needed. When the analyst is unable to resolve or accept the edit failure, contact is made with the respondent to verify or correct the reported data.

Imputation

Not all respondents answer every item on the questionnaire. Effective with the 2012 Census of Governments: Finance—Survey of Public Pensions: State- and Locally-Administered Defined Benefit Data, the Survey improved upon item imputation methodology to fill in missing data based on grouping similar pension systems together. Both partial and full nonrespondents were imputed using either a mean growth rate, median growth rate, adjusted mean, mean, or direct substitution. These methods are applied to certain variables based on research conducted in 2012.

The imputations were based on either a prior year annual survey or the most recent Census of Governments. All but six missing variables (Z90, Z95 and four relatively new variables: Z13, Z14, Z15, and Z16) were imputed using one of the following methods: cell median or donor distribution of Z81, cell mean, or reported prior year or census year data which was multiplied by a growth factor. If the non-respondent to Z90 and Z95 does not write in anything in the Other categories for Z90 and Z95, we impute those variables to be zero.

Tabulations

After the data were edited, the survey data was aggregated to yield the viewable and downloadable files that are available on the website.

Revisions

The 2015 Annual Survey of Public Pensions: State- and Locally-Administered Defined Benefit Data released data for Fiscal Year 2015 on June 14, 2016. Users should note that this release also includes revisions to Fiscal Years 2012, 2013 and 2014. The revised data are accessible through all viewable and downloadable data files on the survey's website.

Sampling Error

Because all 299 state government pension funds are included, these data are not subject to sampling error or any sampling variability. Of the 6,000 local government-administered pension systems, 1,774 were sampled in 2015. Sampling error is available as a coefficient of variation (CV).

Data Quality - Response rates and nonsampling errors

Nonsampling Error

Although every effort is made in all phases of collection, processing, and tabulation to minimize errors, the survey is subject to nonsampling error, such as the inability to obtain data for every variable for all units, inaccuracies in classification, mistakes in keying and coding, and coverage errors.

While the data records are ultimately from state pension sources, the classification of finances among the different categories is entirely the responsibility of the Census Bureau. Therefore, classification might not reflect the actual classification or presentation as requested by the various state pension respondents or what is presented in a state pension system's own financial statements.

Although the original sources for pension statistics are accounting records of governments, the data derived from them are purely statistical in nature. Consequently, the Census Bureau statistics on government pensions cannot be used as financial statements or to measure a pension system's fiscal condition.

The Census Bureau program develops these data to measure the economic activity of state and local governments in general. The definitions used in Census Bureau statistics about governments can vary considerably from definitions applied in standard accounting reports.

Overall Unit Response Rate

The unit response rate for the 2015 Annual Survey of Public Pensions: State- and Locally-Administered Defined Benefit Data was 55.1 percent. The response rate was calculated as the number of responses received divided by the number of parent governments mailed minus the number of governments that were determined to be out of scope.

Total Quantity Response Rate

The total quantity response rate for the 2015 Annual Survey of Public Pensions: State- and Locally-Administered Defined Benefit Data was 97.2 percent. The total quantity response rate was calculated as the value of "total holdings and investments" reported divided by the estimated total value of "total holdings and investments" of those units mailed, minus those systems that were determined to be out of scope.

For further information, please refer to the 2015 Annual Survey of Public Pensions: State- and Locally-Administered Defined Benefit Data 2015 Survey Methodology [PDF, 221KB].