Preliminary Report: Analysis of State and Federal Employment Data

This is a preliminary analysis of state and federal employment records tracking journalism jobs in the San Francisco Bay Area from 2001 through 2010. Read below or download the pdf: Initial Analysis of Secondary Employment Data

Quantitative Analysis Overview and Technical Issues

This memo summarizes results of the initial phase of quantitative data compilation and analysis to assess the changes in the number of employees over time (beginning in 2011) in the industry and occupation categories that are the focus of this study. Additional data obtained through subsequent primary research activities within this study will supplement the information summarized here.

For the purposes of this study, the Bay Area was defined by NOVA to include the following counties: Alameda, Contra Costa, Marin, Monterey, Napa, San Francisco, San Mateo, Santa Clara, Santa Cruz, Solano and Sonoma. Because some data were available for Metropolitan Statistical Areas (MSAs) in greater detail than for counties alone, San Benito County (part of the San Jose-Sunnyvale-Santa Clara, CA MSA) was also included. The organization of counties by metropolitan areas is shown below. For comparison purposes, data were also compiled for California and the U.S.

Metro grouped-county area
Oakland-Fremont-Hayward, CA Metropolitan Division**
Contra Costa
San Francisco-San Mateo-Redwood City, CA Metropolitan Division**
San Francisco
San Mateo
San Jose-Sunnyvale-Santa Clara, CA Metropolitan Statistical Area
Santa Clara
San Benito* (Hollister)

1 Year 2000 data employ a different industrial classification system, so the series for this study begins with 2001.

Individual, non-grouped counties
Monterey (Salinas MSA)
Napa (Napa MSA)
Santa Cruz (Santa Cruz-Watsonville MSA)
Solano (Vallejo-Fairfield MSA)
Sonoma (Santa Rosa-Petaluma MSA)

* Not on the list of Bay Area Counties. Only 3 percent of the total employment in the two counties combined.
** Divisions of the San Francisco-Oakland-Fremont, CA Metropolitan Statistical Area

Federal and state reporting agencies typically suppress employment data, to avoid disclosing information about one or a small group of firms, in geographic areas with relatively small numbers of employed persons, and also for industries where the number of employed is small (regardless of the size of the overall employment base). To minimize this problem for this assignment, data were compiled by multi-county MSA or Metro Division where possible. Nevertheless, the presence of five ungrouped counties, some of which had relatively small employment bases, meant that data suppression was frequently encountered in compiling the employment data. Data were also limited within the set of employment figures available for recent months, to 3-digit and in some cases 4-digit NAICS categories only (see Table 1-B).

Figures were compiled for all the relevant industry and occupation categories, shown below, where the data were available for at least some counties/MSAs. However, data summaries do not necessarily reflect all of the categories (primarily an issue for industry categories), due to the lack of usable data in some cases. Similarly, although occupation data were collected for the categories shown below, not all of these are necessarily of interest within this project and are included here for discussion purposes only.

Although the intent of the quantitative analysis included compiling employment data for the most recent six months, these data were more severely affected by missing figures and general lack of detail than the annual data. Monthly figures are tabulated below, within the set of employment-by-industry figures, primarily to illustrate that the most recent monthly data available from official sources do not appear to be entirely credible. These data deficiency issues could potentially be addressed by other means (i.e., estimation by the consultants) subject to further discussions with NOVA.

Based on the data challenges discussed above alone, the information in this memo should be considered preliminary and subject to further refinement and reinterpretation.

Industry categories

NAICS         NAICS Title

51                    Information
511                  Publishing Industries (except Internet)
5111                Newspaper, Periodical, Book, and Directory Publishers
51111              Newspaper Publishers
51112             Periodical Publishers
51113             Book Publishers
515                 Broadcasting (except Internet)
5151               Radio and Television Broadcasting
51511              Radio Broadcasting
51512             Television Broadcasting
5152              Cable and Other Subscription Programming


Occupation categories

Occ code     Occupation Title

273041           Editors
274032          Film and Video Editors
273011           Radio and Television Announcers
273022          Reporters and Correspondents
273042          Technical Writers
273043          Writers and Authors

Figures obtainable for employees by occupation represent the numbers of persons employed in a particular occupation for all relevant industry categories combined, so figures for the three categories of Editors, Writers and Authors, and Technical Writers were factored downward to reflect the number of employees likely to be in the industry categories of interest for this study. (If Film and Video Editors, a peripheral category that is also affected by limited data, remains relevant to the study, a similar process will be applied to that category later in this study process.)

Initial Findings

The text and tables shown below are summaries of raw data taken from more detailed tables compiled by TNDG, which can be provided to NOVA as desired.

While employment-by-industry data show steep declines in the Newspaper Publishing industry over the time period of the study, employment-by-occupation data for Reporters and Correspondents indicate less severe losses and, for the study area, employment levels by 2009* (although trending downward) that are higher than 2001-2004 levels. This apparent contradiction raises questions of: 1) whether the occupation data are in fact accurate, 2) whether reporters have been successful in finding work in other industries (while still retaining their occupational title), or 3) whether employment losses in the industry have affected primarily other occupations besides reporters – which seems unlikely. This issue will of course be addressed in subsequent phases of this study.

Based on the industry data, the most dramatic declines in employment from 2001 through the first quarter of 2010, among the industries examined, occurred in sector 51111, Newspaper Publishers. This “maximum decline” was consistent across the three geographic levels of: 1) our study area, 2) California, and 3) the U.S, as shown in the following table (see also Figures), but was most pronounced in the 12-county study area.

* Latest year of available data. (Although the California EDD website indicates that figures exist for the first quarter of 2010, the numbers match federal data for 2009.)

Area % change 2001-2010

Study Area                   -49.41%

California                     -47.85%

U.S.                               -36.41%


The occupational data for the study-area counties show a spike in occupational employment for the category of Reporters and Correspondents in 2007, followed by a two-year decline that still leaves employment levels above 2004 levels (see Figure 2-A and Table 4). This trend is echoed only slightly in the statewide figures, which show numbers of Reporters and Correspondents trending slowly upward from 2005 to 2008, then declining at a similar rate to 2009, ending at a level above 2004. For the U.S., employment in this category is declining or level from 2001 onward (Figures 2-B and 2- C, for data by state totals and U.S, respectively). Data on employment by the corresponding industry of Newspaper Publishers show no such trends, with employment in the study area dropping rather steeply from 2006 to 2009, leveling somewhat to 2010 (1st quarter). For the same industry statewide, the trends are similar but start in 2005. For the U.S., the downward trend starts in 2003 but is less severe, in terms of overall percentage change, through 2010 (see Figures 1-A through 1-C, and Tables 1-A and 1-B).

Employment figures for the occupational category of Radio and Television Announcers, within the study area, are more volatile in the early years of the 2001-2009/10 analysis period, remaining relatively flat from 2004-2009. The state trend is somewhat similar in the early years, although more muted, and the overall trend is downward over the time span, as are the U.S. figures. For the study area, figures for the Broadcasting industry mirror the Radio and TV Announcers occupational data, with the industry, as with the corresponding occupational category, posting an overall gain over the analysis period. Statewide, the industry’s employment decreased only 5 percent over the analysis period, compared to 14 percent for the U.S. The state and the U.S. both saw small employment peaks in the industry, following declines in early years from 2001 levels, over the years 2006-2008. A similar uptick also occurred in the study area.

Figures for the two occupational categories of Editors and Writers and Authors are less volatile within the study area, over the analysis period, than either Reporters and Correspondents or Radio and Television Announcers. Figures for the state show a trough in the trend line for both Editors and Writers and Authors in 2006 (Figure 2-B), in contrast to the study area, where employment is on an upward trend.

Table 1-A. Industry Employment by Geography

Table 1-B. 2010 Monthly Industry Employment by Geography



Table 3. Geographic differences between the 2001-04 and 2005-09 data sets

Table 4. Occupation Employment Totals by Geography, 2001-2009