Although libraries experienced an average annual increase of 4 percent in operating funds from 1996 to 2005,² an analysis of PLFTAS data indicates decreases in both library expenditures and a redistribution of expenditures during fiscal year 2008. Indications are that individual libraries are beginning to experience a shift of expenditures away from collections and staff to other expenditures (e.g., technology, utilities, building maintenance).³

When comparing anticipated FY2007 operating expenditures reported in the 2006–2007 PLFTAS study libraries with the actual expenditures reported in this year's study, it quickly became apparent that anticipated expenditures were not realized. Overall operating expenditures fell short by 15.5 percent, and they varied by specific expenditure type from those anticipated by as much 20 percent:

• 20 percent below anticipated expenditures for salaries
• 0.8 percent below anticipated expenditures for collections
• 12.5 percent above anticipated expenditures in other areas

When considered by source of funding, average expenditures missed or exceeded anticipated levels as follows:

• Local/county missed anticipated levels by 22.2 percent.
• State exceeded anticipated levels by 0.8 percent.
• Federal exceeded anticipated levels by 28.6 percent.
• Fines/fees missed anticipated levels by 22.5 percent.
• Donations/local fundraising exceeded anticipated levels by 136 percent.
• Grants, including private grants, exceeded anticipated levels by 19.9 percent.

A smaller percentage of overall expenditures has been attributed to salaries and more to collections and other expenditures than anticipated. Considerably more funding came from soft, nontax sources than not. The most notable shifts occurred with salaries in local/county support and donations/local fundraising (25.8 and 522.5, respectively), and the unexpected increase in federal funding directed to salaries and collections (see figure 15). The increase in federal support may be a result of specific Library Services and Technology Act (LSTA) subgrants to libraries for specific projects and could be expected to readjust in future years.

Libraries again reported using “soft” funding sources—fines/fees, donations, and grants—to support salaries and other expenditures more than collections. Libraries report a use of soft, or nontax, revenue in both the current fiscal year (FY2007) and anticipated expenditures for the next fiscal year (FY2008) that is above the national average. The national average for funding operating expenditures from other sources of revenue (nontax sources) was 8.4 percent, as reported in the Public Libraries in the United States: Fiscal Year 2005 data.⁴ However, libraries reporting in the 2007–2008 PLFTAS study have indicated using nontax sources for operating expenditures at 12 percent in FY2007, with 10 percent anticipated in FY2008—as much as 3.6 percent higher than the national average. The increased reliance on nontax revenue reported in the PLFTAS studies may be an indicator of what national IMLS data will show when FY2006 and newer data are published.

Overall, the use of nontax revenue reported in the 2007–2008 PLFTAS study were:

• About 37 percent was used to pay for staff salaries in FY2007; about 33 percent is anticipated in FY2008.
• About 20 percent was used to fund collections in FY2007; 23 percent is anticipated in FY2008.
• About 42 percent was used to support other expenditures in FY2007; 44 percent is anticipated in FY2008.

Building on the data first collected last year, library systems were asked to estimate expenditures for FY2008 in four categories—salaries, outside vendors, hardware/software, and telecommunications.

Estimated expenditures for technology-related expenditures declined slightly between FY2006 and FY2007, approximately –3.9 percent overall (see figure 14). Declines in expenditures from local/county and state sources were about 11 percent, and about 26.3 percent more than anticipated was spent from federal sources. Expenditures from soft funding sources (nongovernment) declined most noticeably from FY2006 in donations and grant funding categories. For FY2007–2008, overall anticipated expenditures were expected to decline by less than 1 percent (–0.3 percent). Although local/county funding sources were anticipated to continue declining, notable increases were anticipated from state and federal sources. Also, more revenue from fees/fines was to be directed to technology-related expenditures, with less reliance on donations and grants.

It is important to remember that, although salaries for technology staff may be supported heavily from soft funding sources, those salaries are a small proportion of overall salary expenditures for libraries. On the other hand, technology expenditures (hardware/software, outside vendors, and telecommunications) may constitute a significant proportion of “other expenditures” as a category. As such, it may be that libraries continue to use local tax support to pay traditional and ongoing expenses, such as programs, utilities, and transportation. Reliance on nontax revenue to support basic technology hardware and telecommunications expenditures is a reflection of creating a revenue stream from soft-funding sources to build and support new services. This is especially evident in suburban and rural libraries.

Figures 16 and 17 present the estimated changes reported by public libraries for technology-related expenditures by funding source and type of expenditure for FY2006–2008. Although there were declines in the use of certain funding sources to pay for technology, these tables show the item-level variations with overall expenditures. For instance, although use of local/county funding sources to pay for technology declined by 0.2 percent, expenditures increased for software and telecommunications and declined for salaries and hardware. Further, even though overall use of donations to pay for technology declined, libraries reported spending 135.5 percent more on software from this funding source in FY2007 over FY2006. In FY2007–2008 considerably more funds in all categories are anticipated to be directed to software expenditures, and more state and federal funds and fee/fines are expected to be used to pay technology staff salaries.

The unknown impact of a shift of local/county revenue away from funding telecommunications costs, together with a growing reliance on soft funding sources and state and federal support for these expenditures, is worrisome. It will be interesting to see what libraries report next year to determine if this shift continues across types of funding sources, even though overall expenditures have fluctuated very little during the last three fiscal years.

Rural libraries anticipate the most improvement in all funding categories for technology expenditures between FY2007 and FY2008. Interestingly, suburban libraries—a segment of the library community that historically has had strong and stable local funding—anticipated continued declines in overall local/county support into FY2008, as well as declines in donations and grants directed to technology expenditures. Suburban libraries did, however, anticipate improvement in state and federal funding directed to technology expenditures.

Urban libraries continue to show fairly steady improvement in local/county support for technology and to anticipate improvements in their use of state and federal funding for these expenditures. Urban libraries anticipate a significant increase in the use of fees/fines for technology and a decline or very modest improvement in using other soft funding—donations and grants—to pay for these expenditures. Consult the full study report for detailed expenditure tables by metropolitan status.

Presenting library finance data in parallel with other findings reported in this study—by metropolitan status (rural, suburban, and urban) and poverty ranges—provides a useful context for understanding public access computing services. However, the finance data tell a somewhat different story when viewed through the lens of community size.

Figure 18 presents the average anticipated FY2008 technology-related expenditures from all revenue sources by population-served ranges. One of the more surprising findings from this recalculation is the nearly equal distribution of expenditures across all technology categories for public libraries in the largest communities. The smaller the community, the more expenditure distributions shift toward salaries and away from telecommunications and outside vendor expenditures as a proportion of overall technology-related expenditures.

Individual expenditures by specific funding sources and population-served ranges present even greater detail (figure 19):

• Libraries serving more than 500,000 residents reported 45 percent of telecommunications costs were paid from local tax revenue and 55 percent from federal.
• Libraries reporting in this population-served range reported no state tax support and no use of soft revenue sources (e.g., fee/fines, grants, etc.) to pay for telecommunications costs.
• Libraries serving 100,000–499,999 residents reported telecommunications costs paid 55.7 percent from local tax revenue, 21.7 percent from state, and about 16 percent from federal sources. The remaining 6.6 percent of costs were paid from soft revenue sources.
• Libraries serving 25,000–99,999 residents reported telecommunications costs paid 73.4 percent from local tax revenue, 13.9 percent from state, and 8 percent from federal sources. The remaining 3 percent of costs were paid from soft revenue sources.
• Libraries serving 10,000–24,999 residents reported telecommunications costs paid 68.5 percent from local tax revenue, 10.9 percent from state, and 8.9 percent from federal sources. The remaining 10 percent of costs were paid from soft revenue sources.
• Libraries serving fewer than 10,000 residents reported telecommunications costs paid 55.7 percent from local tax revenue, 12.7 percent from state, and 9.8 percent from federal sources. The remaining 21.8 percent of costs were paid from soft revenue sources.

Perhaps the most interesting finding was that, when asked specifically about technology-related expenditures anticipated in FY2008, reliance on specific types of funding sources skewed from those estimates reported for salaries and “other expenditures.” Specifically, the larger the library service area, the more likely libraries were to report that technology expenditures were paid from local or state revenue sources, not paid from soft revenue sources (e.g., fee/fines, grants, etc.), even though libraries reported relying on these funds for “other expenditures.” Figure 20 presents these distributions of nontax revenue sources. The first two columns present what libraries reported for anticipated FY2008 operating expenditures that would include the detailed technology-related expenditures, and the next four columns present anticipated FY2008 operating expenditures for some specific technology-related expenditures.

Public libraries serving the smallest communities are far more likely to rely on soft revenue sources to acquire and pay for outside vendors, hardware/software, and telecommunications than larger libraries. Some good news is visible in the population-served analysis—for libraries with technology staff, a significant proportion of financial support is from local, state, or federal tax revenue. This is not particularly surprising since these libraries were more likely to report not having dedicated IT staff. However, for libraries serving 100,000–499,999, which benefit from strong local tax support for library expenditures, the level of support from nontax revenue is notable and will require closer review in the study's next year of data collection (2008–2009).

Hardware/software and telecommunications expenditures were the most likely to rely on nontax revenue. Very few libraries reported grants as a source of support for either hardware/software or telecommunications expenditures but did report reliance on fee/fines, donations/local fundraising, and private foundation support for these particular expenditure categories. The smaller the population-served range, the greater the reliance on private foundation support for hardware/software (30.9 percent versus 8 percent). This can be attributed to private foundation hardware strategies targeting high poverty and rural communities. Libraries serving 100,000–499,999 reported the lowest level of private foundation support, about 5 percent.

Libraries in the smallest communities also reported a higher proportion of donations being directed toward hardware/software and telecommunications costs than did other population-served ranged. Although the funding amounts were considerably less than other population-served ranges, libraries serving fewer than 10,000 residents reported 21.6 percent of support coming from donations. Libraries serving 100,000–499,999 reported relying on about 6 percent of donations to fund hardware/software and telecommunication expenditures.

The study team anticipates doing more analysis and making a closer review of the fiscal data by population served ranges to begin understanding the nuances between the long-standing metropolitan status and poverty categories against the population-served ranges.

A solid understanding of revenue by source and expenditure by type is critical to making strategic decisions for funding and sustaining quality technology access in libraries. There are a number of issues, however, that impact how well, and at what level of detail libraries are able to capture the expenditure data needed to make meaningful decisions. The top three are: working with finance staff to get what you need; estimating costs for hardware, telecom and personnel; and, embracing the things you can't control.

Finance staff: It may seem a simple matter to identify the people who can provide you the information needed to plan technology-based services, but that is not always the case. Many libraries do not manage their accounting activities, and work with larger fiscal offices in communities and on campuses. Further, how expenditures are attributed in accounting systems vary considerably and may make it nearly impossible to tease out specific expenditures. Some suggestions:

• Be clear about what you are looking to track. Work with fiscal staff to organize records for simplified reporting.
• Be clear about how often you want the information (e.g., monthly, annually, etc.).
• Be flexible.

Estimating costs: Sometimes all you can achieve is a “best guess” regarding the costs for hardware, telecom and technology personnel. This is especially true when understanding the cost of ownership of public access computing services (e.g., electricity, maintenance, etc.). Perhaps a greater challenge is knowing when to stop calculating costs – micro estimates may be unnecessary when a “ball park” figure based on reliable individual costs will do.

Estimating costs is difficult in instances where hardware, telecom or technology-based services are paid by units or agencies outside of the library. These include payments made “on behalf of” the library, or perhaps as part of a larger contract in which the library is a part (e.g., telecommunications contracts, online database licenses, etc.). Another example of instances where estimates may be needed is with long-term contracts, such as multiyear leasing contracts (hardware) or multiyear telecommunications licenses or software licenses. In some cases, all that can be done is an estimate of costs based on the number of library buildings, number of computers connected, or hours a library is open for service.

Embracing what you cannot control: Sometimes an estimate is all that is possible. Focus on what you have and be consistent in what your collect and methods of estimating expenditures. For instance, if you are using building-based estimates as the most reasonable model for presenting expenditures, do that consistently from reporting period to reporting period. Don't change methods of measurement year to year – longitudinal estimates are most meaningful.

Just as in working with fiscal staff – be clear about what you want to know, be clear about how you determine expenditures, and repeat the method over multiple reporting periods.