Where Does the Money Go? Creating Transparency in Libraries’ Operational Budgeting

Operating and personnel budgets encompass the majority of spending in academic libraries, but are often decided with little process or participation. Transparent budget processes for assigning this funding provide the opportunity for widespread clarity and understanding of how budgets are used, and how to acquire funding. The budget model described offers one way to help ensure that every voice in the library has the opportunity to share their ideas for improvements, programs, personnel, and events.


Introduction
Libraries evaluate and strategize their collections budgets as a matter of course. After all, our collections are, on some level, the reason libraries were created, and their fundamental purpose. However, collections budgets, encompassing such categories as print collections, online materials, datasets, special materials, and publishing costs, account for only part of the overall budget of a library. On average, academic libraries spend 46% of their budgets on collections, 42% on personnel, and the remaining 12% of budgets are devoted to operating costs (Association of Research Libraries 2019). Operating budgets include the non-collection elements of library expenditures, such as renovations, office supplies, professional development, and software/hardware. This 12% represents a small percentage of overall budgets, but also represents the potential investment for innovation and impact on overall strategic priorities and staff morale (Mosley 2004). While a diverse range of articles have been published regarding the allocation and assessment of collections budgets in libraries, few exist describing how operating budgets are assigned and distributed (Gregory 2019;Bernhardt, Hinds, and Strauch 2016;Martin 1993;Robinson and Robinson 1994). Frequently, the allocation of these budgets is assigned to one administrator. If this assignment is not paired with a clear request process, it can often result in confusion, distrust, and frustration. Transparent budget processes provide the opportunity for widespread clarity and understanding of how budgets are used, and how to acquire funding (Mosley 2004).
The library operating budget can come from a variety of sources. Some universities may allocate a specific amount to the library each year to operate their facilities, but the funds can often come from other sources. Those could be gift funds, endowments, or unspent salary (funds assigned to unfilled positions that can be redistributed elsewhere until the position is filled).
Libraries without adequate assigned operating budgets may choose to leave positions unfilled for some period of time in order to utilize that funding for operations. Since these funds tend to be the only part of the budget that may be less restricted in how they are spent, the allocation of operating budgets should be of high interest to libraries' personnel (Mosley 2004).

Institutional Context
Historically, requesting funds for operations within the authors' institution was largely unstructured. For many years, if there were needs for operational funding for furniture, renovation, outreach, exhibits, events, or anything else that was not personnel or library materials, the process was to make a request via an email message to the associate director or associate dean.
Each of the 26 department heads was responsible for asking for what their departments needed.
An assertive department head might submit many requests and draft extensive justifications, though they were not required to do so. Some department heads operated under the belief that there was no money for these needs, and opted not to request anything, not even pencils. All types of requests (including professional travel, furniture, promotional items) rolled in throughout the year, when things were thought of or needed. Administrators made decisions based on the information available to them at the time. This information varied by time of year, including what requests were pending, or what had already been approved, but by necessity could not include a thorough knowledge of what was going to be needed further into the fiscal year.
Since the requesting process was unstructured, the response process was equally unstructured. Department heads did not know the total amounts of what was requested, funded, not funded, or generally, why a particular funding decision was made. The overall narrative was one of scarcity, a condition that employees began to internalize. Personnel expected that if they wanted hot food, they needed to use their personal money to provide their own microwave for their workspace; if their office chair broke, they sat on a broken chair; if their desk drawers would not stop rolling open, they taped them shut. The result of the scarcity mentality was to discourage requests for materials, food, exhibits, training, and other resources that employees needed to effectively do their job. The scarcity narrative was not entirely fictitious -there is no such thing as plenty of money, and the library is located in a state with extremely poor state funding for its universities -but the lack of process inhibited the possibility of identifying the opportunity cost (Myers and Kambhampati 2019). By agreeing to fund one request, what was effectively being denied later? And without the strategic assignment of funding, department heads could not identify how to best align their requests to the priorities of the institution. The long-term effect of this approach was that departments with leaders who were willing to assertively request resources became the haves and the departments with less assertive leaders became the have nots. This 'haves and have nots' mentality also created a perception that those areas that received funding were priorities, even if another area was officially identified as a strategic priority. Additionally, this approach hamstrung the ability of the administration to fund worthy ideas if they were not requested sufficiently early in the fiscal year or if the ideas had the misfortune of arriving on the heels of some other recently funded, high-cost request. In particularly lean years in which overall budget cuts translated to high fiscal pressures, the lack of clear process made it challenging to choose the most strategic requests for the limited available funding.
Personnel requests were handled strictly on the basis of refilling positions. Regardless of the level of the position, most personnel lines were thought of as belonging to a specific department. When a position became vacant, the department head could either fill it as it had been before, or redeploy that position to do different work within their department. New positions were not awarded, nor were vacant positions transferred from one department to another. For departments with relatively high levels of staffing, this facilitated their ability to be flexible and responsive, and to take on new projects when old ones were complete. Conversely, for departments with relatively low levels of staffing, that shortage became insurmountable, and inhibited their options for adding new projects, rethinking existing roles, or in some cases meeting the most basic expectations. As with the other elements of the budget, this process complicated the libraries' abilities to respond strategically to changing priorities, and resulted in departments with staff that did not match their level of need on both the high and low sides. Department heads had to request permission to refill librarian positions, which was routinely granted, but did not have access to any information about what other department heads were requesting or planning for their open positions. In addition, information about what was approved was generally not common knowledge until the position was posted publicly for applications. In this model, positions generally could not be planned strategically, or in relation to one another, unless they reported to the same department head. With two exceptions (cataloging and reference), most departments consisted of fewer than five or six people. Therefore, even positions that were closely related generally reported to separate departments. This approach led to a balkanization of responsibilities, with a high degree of overlapping functions across multiple departments and positions, and numerous lost efficiencies, or unrealized opportunities for rethinking roles.

Opening the Budget Process
A library reorganization, combined with sudden, nearly complete turnover in the administration, created a natural opportunity to rethink the budget approach. There are a variety of budget systems for distributing funding, but the challenges faced in the institution were an outgrowth not so much of how money was distributed, but how it was requested in the first place (Linn 2007;Dugan and Hernon 2018;Hallam and Dalston 2005). In the reorganization, the number of departments was reduced from 26 to 8, making it possible to consolidate requests into fewer groups. A new strategic plan made it easier to identify and align requests with strategic priorities. In service of being strategic about hires and positions, the approach to personnel vacancies changed completely. All vacated position lines reverted automatically to the dean rather than continuing to belong to the department. The result was that positions could be redeployed either back into what they were already doing, or if there was a more strategic need elsewhere, the money for the position could be allocated to a different department or a different kind of position.
The first step in opening the budget process was providing each department head a small discretionary fund of $500, with specific parameters around how it could be used. This small change was perceived favorably by most department heads, who used the funds for professional development, retreats, and team-building events. Administration then asked department heads to submit all requests for the full fiscal year before the fiscal year began. The first results contained a wide range of detail, ranging from no justification at all to intricate histories and excessive justifications, and a range of format for submissions including documents, spreadsheets, and even slideshows. Attempting to reconcile this information was arduous and time-consuming, and still did not result in a clear picture of the needs of the organization.
During this same time, the dean initiated a new, annual internal small grant fund for Library Innovation, Service, and Entrepreneurship. This program funded grants from the University Libraries to underwrite creative and innovative experiments in library operations and services.
The overall goal of the fund was to enhance, wherever possible, library service to the academic community. Providing this alternative and predictable way for libraries' personnel to fund creative ideas increased the number of ideas shared with administration and improved libraries' services and programs. For example, this small grant program funded the purchase of a custom-built mobile library bicycle for use in outreach programming that has proven very successful (University Libraries 2017). This program also improved library spaces, as units proposed minor renovations that were outside the scope of the grant, but were eventually funded through other means. The first round of funding requests illustrated the need for additional improvement to the budget request system.
The small grant program began hosting annual presentations from funding recipients, in which they shared their results and impact with the libraries' personnel. The presentations offered an opportunity for everyone to learn about successful proposals and their impact on the libraries' strategic goals. They also provided the libraries' leadership an opportunity to decide if the project fit with the organization's strategic directions, and if the pilots should be made permanent. These funding processes encouraged information sharing for writing better justifications, as well as encouraging collaboration and communication. These processes also gave people across the organization at every level the opportunity to submit ideas for improvements, and related funding to go from idea to pilot, and possibly long-term implementation. Together this small grant program and a clearer budget process provided multiple avenues for individuals to seek investment for ideas and funding for programming, professional development, and other strategic needs.

Defining Transparency
High trust workplaces result in more productive, engaged, and committed employees, and transparency is an important element of establishing organizational trust (Shockley-Zalabak, Morreale, and Hackman 2010; Covey and Merrill 2006). Transparency, however, means different things to different people, and emphasizing the transparency of a process can result in widely differing expectations for what that process will include (Garrett and Vermeule 2006).
To library administration, budget transparency meant that the process for requesting funding was clear and understandable; the requirements for budget requests were made consistent across all departments; it was obvious who was requesting funds and why; it was explicit who was making decisions about allocations; and the timeline was predictable. Over time, efforts to prioritize budget transparency expanded to include allowing all department leaders access to the contents of all submitted budget requests and subsequent funding decisions. When presenting and discussing the process, library administration routinely emphasized the value of transparency.
For some, however, transparency in a budget meant that each employee understands the process for budget requests, knows what was requested, has input into what was approved, and agrees with the final decisions. These expectations crossed from transparency into decisionmaking. Transparency thus defined would indeed prove impossible for most organizations, since without unlimited budget, some budget requests will inevitably be denied. Library administration found that routinely describing the process as transparent inadvertently created the expectation that budget allocation would be delegated more widely than was possible in practice and under the organization's fiscal reality. Nevertheless, transparency into what was requested, who was approving or denying requests, timeline for approval, and what was approved or denied, remained a core priority for implementation of a budget system.

Planning for a Budget Request System: Process and Tool
In order to build a clearer picture of overall requests and an effective budget request system, it was first useful to define what was needed for all participants in the budget process.
The first attempt at a more open budget process, while messy, did illustrate that specific parameters were required to clarify needs and strategic priorities. Additionally, any process required those evaluating the budget requests to be able to review, compare, and contrast requests. In order to align requests with strategic priorities, it was also helpful to have requests articulate how each item aligned with strategic priorities. These parameters were the basis for developing a budget request database tool (see Figure 1). In the absence of any off-the-shelf budgeting software that would have met the organization's needs, the libraries developed an inhouse budget database tool that has been through six iterations. In the first iteration, the tool began with required fields for submission of each request and over time these fields expanded.
In the first iteration, department heads input all requests on behalf of their department members, and while they could see a record of everything they submitted, they could not see the requests of any other department head. The budget request process had a clear structure, but was still mostly a black box.

Figure 1: Budget Request Form
This first iteration of the budget database allowed administrators to prioritize and allocate funding to items that would further the libraries' strategic objectives, which subsequently improved the alignment of the strategic priorities and individual responsibilities. Departments were better able to demonstrate their impact, or potential, in each strategic area. Over time, elements of the budget database and its related processes were improved to continually augment responsiveness and clarity. An annual presentation given by administrators provided context and guidance for how to best submit and justify budget requests. Eventually, department heads were given access to be able to see not only their own requests, but all requests submitted, in addition to what decisions had been made for each request. At last, the budget black box had a window.
Categorization of the different requests, along with strategic priority alignments, permitted administrators to assess and compare requests that aligned to the same strategic priorities or that fell into the same categories. This allowed administrators to assess both the possibilities of new requests, as well as the opportunity costs of what might be funded instead. The process still needed to enable a predictable, universal timeline for submitting budget requests so that all requests could be considered together and answered consistently. The tool needed to allow requestors to assign prioritization, so that they had the ability to communicate to the administration which proposals were the most important to their department. The tool also needed to allow administrators to identify personnel requests across all departments, in order to plan effectively for the allocation of limited personnel budget.
After early iterations of employing a systematic approach to budget requests, additional needs emerged. The budget request timeline aligned with a July-June fiscal year, with the majority of the requests due in May. The initial process offered one budget request round for the entire year. However, some funding needs were unknown or unexpected at the start of the fiscal year, which required a mechanism for submitting requests more than once per fiscal year. Budgetary needs could not consistently be anticipated for a full year, especially for travel and events. It made sense, then, to add two supplemental request rounds in September and January. It additionally became obvious that the process would be more efficient and effective if all personnel had the ability to submit requests, thus reducing the telephone game effect of permitting requests only from department heads. The outcome of addressing this need was to create a protocol in which all requests were initially funneled to department heads, who then decided whether to reject or support the request, prioritized it, and submitted it to the administration.
There were nine possible categories of requests: • Computer, Equipment, Software In the first year, the tool only required a business justification, but requestors found it difficult to include sufficient detail within the 150-character limit, so the description and title fields were added in order to separate the questions of "what" and "why." All new staff and faculty position requests also required the submission of a position description outside the database to administrators. It remained a challenge to fit all the details into the character-limited request, but the form also included a robust question and answer function to allow for more discussion between administration and department heads. These exchanges of questions and answers were attached to, and remained part of, the official request documentation.
One of the most important parts of any request was the priority assignment. The department head assigned level of priority for each of their requests. Priority ranged from 1-4 with 1 being the highest priority, and 4 the lowest. The expectation was that the majority of requests would be in the 2-3 range, and that top priority would be reserved for truly mission critical requests.
Establishing these norms around prioritization took a few rounds, and they were reiterated in each cycle to remind department heads to avoid the temptation of assigning top priority to everything.
If a leader assigned the same priority to all requests, then the administrators would make the determination of what was important or critical themselves, and their choices of prioritization might not match the desires of department heads.
Once all the requests were submitted, administrators downloaded them from the database for deeper analysis. Requests were, unsurprisingly, significantly higher than the amount of funding available each year (see Figure 2), and thus the analysis of these requests needed to be robust.
This step necessarily required a large time commitment on the part of the administration. It was important that personnel understood that each request was read carefully and considered seriously. If there was a sense that this step of the process was done too quickly or not robustly, it would negate much of the value of the added transparency.

Figure 2: Budget Requests and Responses by Fiscal
Year approved denied deferred/holds

Demonstrating Deliberation and Transparency in Responses
There were four possible responses to each request. The first two, "approved" and "denied," were easy to understand, but too limiting in a year-long process. Therefore, those standard responses were supplemented with two additional categories: "hold" and "defer." "Hold" meant that while funding was not available for this request at the time, it was possible that there might be funding later in the fiscal year. These "hold" decisions were reviewed at the next budget deadline and/or mid-spring and moved to one of the other response categories. The second: "defer," was for requests that had merit but could not be funded in that fiscal year. Individuals were encouraged to resubmit the following year.
In addition to the responses, each request also included a clarifying note. A note for an approved request may only have stated something like, "work with the budget officer to proceed with this request." Notes for the other responses provided context for the decision, such as, "while this is a good idea for an event we are approving a different request instead due to strategic priorities" or, "while we cannot fund this year we think this is a good idea and ask you to resubmit next year when you know more about [a specific related detail]" or, "we will re-examine this category of requests in the spring when we have a better idea of how many one-time funds we have available for furniture." These notes became one of the most important parts of this final step. They ensured that for answers that were not 100% approved, people could gain an idea of why a decision was made. It could also help open a dialogue if there was anything that the administrators misinterpreted in the request.
These responses were added via an uploaded spreadsheet so that all the responses were delivered at the same time. Typically, administrators sent an alert that a particular category, or all requests, would be getting a response on a particular day/time, so that department heads could be prepared to share this information with their department members. The first version of the alerts only shared the status, but none of the notes, amounts, or a direct link to the original request in the database. Since not all department heads opened the database to review the notes or approved amounts, confusion reigned regarding why some requests were not approved. In some cases when requests were approved, but only partially, department heads did not know that funding was partial, nor did they pass that information along to the requestor. The first step was to add a link to the request and its response, but even in that case individuals did not always follow the link and some of the related confusion continued. Therefore, the notice now contains a response, the latest note, the amount of funding, and a link to the full record.
After the first year, personnel in human resources, finance, and event support staff were given permission to see the contents of the database. This allowed the personnel in charge of taking action on approvals to have access to the full request and response, as well as any instructions the requestor had received. These staff in turn made changes to related documentation. For example, travel approval forms were revised to include a field for the budget request identifiers. In addition, the budget officer was given the ability to add account codes, to ensure that those implementing the budget requests charged the right account.
For everyone who submitted a request, this final step was the most important in the process, since it was the moment in which they discovered whether or not their request had been funded. It was also the step in which they were likely to have additional questions, or want to know more about why a particular decision was made. By utilizing the Q&A through the form, it became possible to continue the conversation in the database, ensure that everyone was heard, and those responses were recorded and shared. Since all department heads had access to the notes, decisions, and other information in the database, this also helped build a better sense of how and why decisions were being made, and the types of questions and concerns being raised across the libraries.
Every fall, the administrators gave an update to the entire library detailing the number and amount of initial requests, along with the rates and amounts of approvals, denials, holds, and deferrals. This helped people to better understand how their individual requests fit in the broader context. These presentations were an important component of ensuring that administration provided multiple opportunities and strategies for people to offer input into, and ask questions about, budget process and practice. As demonstrated by the numerous changes over time, this was and continues to be an iterative process. Individuals invested time and excitement into writing these proposals, and it was important that administrators devoted that same time and understanding into the decisions.

Areas of Continued Challenge
No budgeting system is perfect, and there were several elements of this budget system that have been unpopular. Some of the most consistent negative feedback from the libraries' personnel submitting budget requests has fallen into a few main areas:

Speed
The request system funneled requests from individuals to their department heads and then from department heads to their associate deans. The response then traveled back from the dean and associate deans to the department head, and back to the individual. The process of limiting requests to three times per year could result in several months passing between the time the request was submitted to when it was answered. Some personnel perceived this approach as too slow, and inhibiting spontaneity. This problem was most acute as related to events; it was common for people to conceive of an event they wanted to hold very close to when they want to hold it, and the budget process could hamper that effort. While there were advantages, as described above, to being able to view budget requests holistically and clearly understand the opportunity costs, the necessary related drawback was delay in responses to new requests. Even in the assigned rounds of budgeting, it took quite some time to carefully read and evaluate hundreds of requests, which also slowed the speed with which individuals received responses.

Small Dollar Requests
Some personnel bristled at the requirement of requesting funding for even small amounts, again particularly related to events. These personnel often interpreted the necessity of requesting small amounts as a lack of trust, or a lack of respect for their ability to make responsible decisions.
As with some other elements of this process, there was benefit to be gained from looking at all requests together. Since there was not enough funding for all event requests, some must be chosen over others. The advantage of administrators being able to compare requests in the same category came at the cost of individual decision making, such as which events or how many to hold. Additionally, before the existence of this budget process, there was no systematic institutional funding for events at all, and many personnel provided food or other events costs out of their own pockets. That absence of funding was consistently felt as a lack of support, and creating this process was an attempt to provide event support while still managing costs and expectations. So while the new budget process added much more event support than there was before the process existed, there remained less funding, and less control, for events than individuals desired.

Transparency
Any process in which some people get money and others do not is likely to be felt more negatively by those whose requests are denied. Since the administrators set an initial goal of making the process as transparent as possible, and more transparent than most other budgeting processes in similar institutions, it was necessary to embrace the fact that it was not possible to meet everyone's expectations for what transparency really meant. Occasionally, individuals felt concerned that the decision about their budget request was personally motivated, and that concern typically translated to requests for more transparency. This underlined the importance of explaining why a decision was made and the expansion of the process to include this in the open documentation.

Administration Decides
In a process where all of the budget decision making was handled by administration, it was perhaps not surprising that a common complaint was that administrators did not sufficiently understand the requests, and therefore should not be making the decisions about whether to fund them. Administrators were sometimes a few steps removed from the person making the request, and some personnel were concerned that the request was lost in translation when it was funneled through their department head. Administrators frequently asked detailed questions in order to improve their understanding of each request, but unfortunately, those questions were sometimes interpreted as lack of trust, rather than an effort to thoroughly understand. Once the administrators fully understood the requests, it was easier to make strategic decisions such as which events would yield the most impact, or why additional student workers were critical for a project.

Inflexibility
The desire to request an amount of money to distribute at will was evident in various ways.
Many people wished that they could simply request funds, and decide for themselves how to spend it. The drawback of that approach, for the administrators, was that it hampered their ability to ensure that money was being spent on institutional priorities, and that in some cases the funding available was restricted to particular uses. It was also common for individuals to request money for one thing (for example, furniture), and then to decide they did not want what they requested, and assume they could spend that same amount on something different (like events or training). Additionally, personnel wished that their approvals could roll from one fiscal year into the next. For example, if they received approval for a particular event that for some reason did not occur, they would prefer not to ask again the following fiscal year. Since each year represented a new budget and new fiscal reality, it was not possible for administrators to approve requests that crossed the fiscal year boundary.
To be sure, many of these concerns have merit. In particular there was strong desire among the personnel to receive a budget amount and determine the spending themselves. There are some structures, and some budgets, in which such an approach might function well. This institution had a long history of haves and have-nots in different departments, an unusually small budget for both personnel and operations for a research library of this size, and an organizational structure that generated very uneven budget needs across departments. Given that context, this process attempted to strike a balance between need and input in a way that simply distributing unassigned dollars would not. As with collections budgets, institutions have a responsibility to strike a balance between broad and departmental based funding that should be determined based on organizational and budgetary needs.

Conclusion
Operational and personnel budgets provide an opportunity for libraries to foster innovative ideas and practices, support mechanisms for employee community and professional development, and to promote our spaces, resources, and services to our communities. While in academic libraries this may be a small portion of the total budget, the impact can be wide-ranging.
Institutions should devote time and attention to ensuring that these resources are realizing the maximum benefit. Models such as the one described above offer one way to help ensure that every voice in the library has the opportunity to have their desires, dreams, and ideas considered.