You are viewing a javascript disabled version of the site. Please enable Javascript for this site to function properly.
Go to headerGo to navigationGo to searchGo to contentsGo to footer
In content section. Select this link to jump to navigation


We are dedicating the March issue to the Statistics of Income Division (SOI) of the Internal Revenue Service (IRS) in the United States. The SOI provides information about the US tax system. Data collected and disseminated by the SOI are used to formulate the federal budget and evaluate existing and proposed tax laws. The data are also used as inputs by other US statistical agencies for developing survey frames, conducting economic research, and constructing the National Economic Accounts.

The manuscripts in this issue are representative of the work that goes on within the agency. Most SOI data are derived from administrative records (tax returns). The data are selected through stratified random samples and augmented with supporting information provided by tax payers with their returns. The demographic information on the returns are coded and the tax data are standardized to improve consistency and accuracy across records.

We start with the history of SOI. The history is told by Fritz Scheuren, editor-in-chief emeritus of our Journal and director of SOI from 1980 to 1994.

Fritz Scheuren received the Shiskin Award in Economic Statistics (1995) in recognition of his transformation of the organization and his efforts to modernize its statistical activities, computer hardware and software, and its costumer focus.

Barry Johnson is SOI’s current director. He has worked for the SOI most of his statistical career. He is well versed in methodological and technological issues associated with the production of federal statistics and the use of administrative data for statistical purposes. He has worked extensively on studies of the federal transfer tax system and the distribution of personal wealth.

SOI turned 100 in 2016. Barry Johnson says that it is his goal to ensure its health and relevance for at least another 100 years. He would like to make SOI and the contributions of its staff visible to a wide audience. He believes the agency needs to use Big Data techniques to augment tax return-based statistics, thereby expanding the information value of the statistics. It is his hope that all federal agencies will work collaboratively on best practices to modernize and improve the ways data are collected and information are produced and disseminated.

The manuscripts in this issue focus on:

  • 1. The ways SOI is leveraging administrative records for official statistics

  • 2. How to improve techniques for producing cross-sectional estimates from panel data

  • 3. The effects of audit on future filing behavior

  • 4. Evaluation of weighting techniques to improve small area estimates

  • 5. The impact of assumptions about mortality on wealth estimates derived from samples of estate tax data

  • 6. SOI sample designs

  • 7. Using administrative tax data for evidence-based research

  • 8. SOI’s evolving mission

The latter paper is prepared by Arthur Kennickell who was the focus of our Festschrift in March of last year.

In addition to the SOI manuscripts, you will find Katherine Condon’s interview with Nancy McBeth. Nancy has been a consultant with the Statistical Center for the Cooperation Council for the Arab Countries of the Gulf (GCC-Stat) in Muscat, Oman, since late 2014. The GCC-Stat is responsible for data collection, analysis, dissemination of economic, social, demographic, agricultural, environmental, energy, and other data through the national statistical centers and work on their classification, storage and analysis. Nancy McBeth is working on the implementation of a harmonized Population Census in 2020.

You will also find a conversation with our interview editor, Nancy Torrieri. She talks to Michael Radcliffe, a U.S. Census Bureau geographer. He is the Assistant Division Chief for Geographic Standards, Criteria, Research, and Quality and manages many of the tasks associated with defining and delineating the Census Bureau’s geographic areas. The conversation is meant to raise awareness of census geography, and its importance and relevance for users of official statistics.

Let me conclude my thanking all the contributors to this issue, but especially Barry Johnson who worked with the editorial staff to make it happen.

As always, the views and opinions expressed in the interviews and conversations are those of the intervie-

wees and do not necessarily reflect the policy or position of the Journal, nor IAOS or IOS Press.

Kirsten West