Association of Research Libraries (ARL®)

Three Years and Counting - The Economic Crisis is Still with Us
UKSG Conference 2011 and Exhibition
April 4-6, 2011
Harrogate International Centre
Charles B. Lowry, Ph.D.
Association of Research Libraries

Introduction
Much discussion of the current international economic disorder refers to it as a major disrupting force for change.  Where libraries are concerned, I have a slightly different view.  I believe that the “great recession” is accelerating the longer-term disruptive forces of technological change and the need to transform the library model by redefining its role and scope, indeed much of the work it does as part of higher education. The future place of academic libraries will not be defined mainly by managing access to the end products of scholarship (books and journals) that defined them in the 20th century, but rather by deep engagement in the process of classroom teaching and research and its outputs.  Further, I believe that the world economic crisis has had the effect of clarifying the collective mind of library leaders broadly.  There is no time to waste.  Perhaps most importantly, the economic crisis has allowed us to make hard decisions—or as Rahm Emanuel famously said "You never want a serious crisis to go to waste.  And what I mean by that is an opportunity to do things you think you could not do before."¹  Early indicators point to the fact that the “serious crisis” is not going to waste.  The evidence is that ARL libraries are grappling with the economic crisis by accelerating the rate of adaptation of the research library model.  These are facts that ought to be understood as matters of survival not only by librarians, but also in the library market place generally.  There is every evidence now after three years that this is “the new normal” and we do not expect a return to systematic and regular additions to ARL research library budgets in order to support ever-increasing prices out of proportion to inflation or to fund bragging rights about rankings.

This being the case, it is useful to think about the economic crisis as a catalyst for change and consider its impact as inversely proportional to the depth of budget reductions and their length.  The 126 members of the Association of Research Libraries (ARL) in Canada and the US, like not-for-profit higher education everywhere, have been buffeted by three years of economic crisis and all its concomitant uncertainties.

Data Collection
ARL has been collecting the data on budget allocations from its members since fiscal year 2008-09.  The first two iterations of that effort appear in the published record and presentations that are available online.²  It is worth saying that the first two efforts were “quick and dirty” albeit they produced some very useful data.  I want to make clear what the three years of data we have collected is and is not. We asked members to “report all budgeted funds that come to the library from regular institutional budget sources, and from sources such as research grants, special projects, gifts and endowments and fees for service or other fees charged directly, such as student fees.”  Thus, these data are not the same as those published in our well-known annual statistical survey of library expenditures.  In simplest terms we asked members “how much money did you get, not how much did you spend?” We did so because expenditure data can give a skewed view of the fiscal condition for member institutions due to various factors such as year-to-year carry-over funds.  So these special surveys collect information on the actual beginning of year budget allocations from three fiscal years that for convenience can be termed Year I (FY2008-09), Year II (FY2009-10) and Year III (FY2010-11).  This most recent survey concluded in March 2011 gave us the opportunity for a sweeping view of all three years, one that was more rigorous and revealing than our first two individual surveys.

In order to achieve maximum comparability, for this year’s survey we repeated the request for data from Year I and II as well as ask for information about Year III, that is FY2010-11. We refined and we believe simplified the survey to achieve better comparability for the three years by using familiar definitions that conform closely to those used to collect the annual statistics of expenditure data. Of our 115 academic library members, 80 responded (Table I).  This is about a 10% decline from the rate of responses in the first two years of the survey.  We are not sure why this happened—perhaps survey fatigue among members or perhaps it was the inevitable result of 30 months of reducing human resources. But, I do not believe that the non-responding libraries would have made the budget picture look better or worse.  This table shows a distribution that is representative of our libraries by size of expenditure and type.  We did not expect responses from the eleven non-academic library members of ARL.  So, in considering this data, keep in mind that it reflects only what is happening in North American higher education.  Again, for purposes of analysis, we found it convenient to categorize the responding libraries by their size as measured by actual expenditures from our 2009-10 statistics survey and whether they were Canadian, US Private, or US Public/State institutions.  These categories will be reflected in the discussion of findings.  I should clarify further—the categorization by size here is based on expenditures and not on budget allocations.

Total Library Budget Allocations
I want to start with the big picture, that is, an analysis of the bottom line—three years total budget allocations. The three-year survey window allows us to use the base Year I as a starting point to gauge change over the next two fiscal periods. What our current survey does not provide is base data from FY2007-08 as a comparison to FY2008-09, Year I of our survey and the beginning of the great recession.  However, our first “ARL Survey on Base Budgets, 2008-2009” conducted two years ago reported serious levels of reductions in budgets at the beginning of the downturn (Figure I). These initial cuts were made well into the fiscal year.   There were 99 usable responses to that first survey and of those, 55 indicated that they had already experienced budget reductions while 44 had not. The hard data from that first survey showed that these 55 ARL member libraries had reductions ranging from less than .04% to a high of 9.4%.  Both the mean and median reductions were 3%. So, as we examine the change from Years I to II, and II to III, remember that nearly 60% of ARL members were already starting from a Year I of reduced budget allocations. Things definitely got worse.

Table II demonstrates the trends—in Year II 46 libraries or nearly 60% had flat or reduced budgets.  In Year III there were 36 or nearly 47% with flat or reduced budgets, hardly a positive trend. Keep in mind the obvious—flat budgets are as good as cuts when price rises of both periods are taken into accounts.  Few publishers responded to the downturn by eschewing increases and I know of none who reduced prices.  Do I need to say that this is an unprecedented period of budget constraint and the magnitude of some cuts was truly draconian—over 20%?  It is also illuminating to look at the range of these cuts.  From Year I to Year II these reductions were pretty evenly divided with around half below 5% and the other half above.  From Year II to Year III fewer experienced cuts above 5% and none above 10%.

The worst-case scenario for a library was to experience three consecutive years of budget cuts. We used scatterplots to try and understand how individual institutions did over time.³ The first scatterplot (Figure II) of percent changes from 2008-09 to 2009-10 (x-axis) and 2009-10 to 2010-11 (y-axis) helps us explore whether there are consistent patterns of decline or increase in the individual library’s budget allocations for the three fiscal years. Each of the colored symbols represents a library. The majority of the institutions are clustering around zero with no indication of a clear linear trend for the totality of the libraries responding. The vertical and horizontal dotted lines mark the zero points and there is a +&- key in each corner of the figure that helps interpret this snapshot.  Libraries in the upper left section had reductions from Year I to Year II and increases from Year II to Year III; upper right show increases in both years; lower right increases then declines; and the worst case those in the lower left that experienced decreases in their budget allocations in both Years II and III.  Remember though that Figure II does not capture the change in Year I of the recession when we know that around 60% received cuts after the beginning of the fiscal year.  One telegraphic way to look at this is to note that only those libraries in the upper right section of the scatterplot did not have cuts in Year II or III.  At the same time the largest number among this group were not really prospering, but rather clustered near zero.

You can see at a glance that there are some notable exceptions—outliers that had large increases in the first period (2008-09 to 2009-10) and experienced some of the larger decreases in the second period (2009-10 to 2010-11) and a few institutions that had larger decreases in the first period (2008-09 to 2009-10) experienced larger increases in the second period (2009-10 to 2010-11). Other outliers represent increases or decreases in the 30% range, which seem extraordinary anytime.  These large changes are more apparent than real, as we found by looking at the explanations they provided for what they included in the data. For instance, two libraries added new units to the reporting, skewing the size of the multi-year change.  Another changed the budget base and calculated as increases one-time reductions that were restored.  Thus, outliers showing these extraordinary percent changes were due to anomalies in reporting and we could have opted to exclude them.

However, we wanted to present here the findings from all the surveyed institutions. As indicated, the figure also demonstrates these changes across different types of libraries (Canadian, US State, and US Private). If we were to attempt to draw the regression line for each one of these groups, you would see a slightly upward trending line for the Canadian institutions and declining lines for the US private and state institutions.  Statistically though, the relations are truly weak and affected by outliers noted earlier.

Examining all the data points, it was clear that there is no consistent pattern or relation between multi-year budget changes when all libraries were considered.  This is not surprising given the budget volatility due to adverse economic conditions. On the other hand, one can observe that Canadian libraries experienced a basically flat budget cycle during this time.  By contrast, there is an indication that the US State institutions had larger budget declines.

We also took a look at the change from 2008-09 to 2010-2011compared to the size of Year III budget allocations. The scatterplot (Figure III) depicts the overall budget change from 2008-09 to 2010-2011 (y-axis) in relation to the Year III allocations. The goal here is to see if the total change over three years relates to the budgeted allocation at the end of the period, i.e. do libraries with higher Year III budgets have larger percent changes or vice versa?   Overall, there is no strong relationship between percent budget change over three years and the final allocation in US dollars for FY2010-11.

Again, the colored symbols indicate library type.  We took a look at the relationship of budget allocation, percent change and library type and we found no relationship between budget allocation and percent changes for Canadians and US state institutions.  That is, for Canadian and US state institutions, relative budget size did not have much predictive power for the whole period in any direction up or down.  However, as you can see in Figure IV there is a relationship between budget allocation and percent change in US private institutions. The main observation is that private US institutions with the larger budgets are experiencing higher percent changes compared to private US institutions with smaller budgets.  As the ascending line indicates—the larger the budget the more the trend was up, that is was positive. I would venture the guess that this was because equities affect the endowment income of these institutions positively.  The R-squared (.33), which tells us how much of the variation in percent change is explained by the variation in budget allocation, shows that about 1/3 of the percent change in the budget is explained by the size of the budget allocation for US private institutions. The relation between percent change and budget allocation, as measured by the correlation coefficient (Pearson r) is .57 (typically interpreted as a moderate to strong relation).  Once more, a word of caution, that we should interpret these relationships with an understanding that there are outliers in the data that can affect these indicators and relationships easily.

The scatterplots paint one picture, but I want to paint another. We were curious to know what types of ARL member libraries did better or worse during this period, better being measured by increases in budgets? Was there any discernible pattern?  As I said earlier, it seemed reasonable to look at members in two ways to get an insight—size as measured by expenditures and member type as categorized by US private, US public or state, and Canadian.   In the change from Year I to Year II (Figure V), the largest proportion of increases were among Canadian members.  Whereas, in the US, 58% of private and 66% of public institutions had flat or reduced budgets.  In Year III (Figure VI), the pattern maintained with Canadian’s predominantly receiving increases, while south of the 49th parallel 58% of private universities and 47% of US public universities again experienced budgets that were flat or declined further.

Looking at this through the filter of size of expenditures (Figures VII and Figures VIII), we found no significant pattern of declines or increases based on the gross expenditures. As these rather “busy” figures illustrate, ARL libraries with expenditures less than $20 million had slightly more budget cuts from Year I to Year II, but more increases in those from Year II to Year III.  Beyond that size does not seem to have an effect on the comparable budgeting experience.  For instance the largest and presumably wealthiest of ARL libraries with budget expenditures greater than $40 million fared about the same over each of the two periods—that is, the pattern of cuts and increases in the budget allocations was similar to those libraries with smaller expenditure levels.

The key finding of our research is that ARL member libraries have experienced three unprecedented years of flat or reduced budgets beginning with FY2008-09 when 55% indicated reduced budgets.  In FY2009-10 that trend continued with 61% experiencing flat or reduced budgets from the prior year. For this year, FY2010-11 47% were faced with flat or reduced budgets. As observed earlier, this is the new norm, not an aberration.  It will have consequences for the teaching and research within higher education institutions and in the market place of scholarly communications internationally.

Dissecting the Budget Allocations
“The key point to consider for mental accounting is that money is fungible; regardless of its origins or intended use, all money is the same.”⁴  But there is the countervailing natural human tendency to protect investments we value and this is what our library managers have attempted to do. To understand the budget picture at a more granular level, we have to dissect it into its three components.  To do this we look first to the component of the annual budget that will be of particular interest in the library marketplace—the materials budget allocations. At the beginning of the downturn in 2008, member libraries indicated that they were protecting their materials budgets and taking cuts in human resources and other operations.  However, (Figure IX) 4% reported that they were taking cuts only from acquisitions budgets.  Another 20% used a mix of acquisitions and other strategies.  However, even when acquisitions were fully protected, 27% said that they still had to reduce levels of collection acquisitions due to the price rises from publishers.⁵ So, very early, we find the materials budget under pressure.  The change from Year I to Year II, further reinforced the negative impact on materials acquisitions (Table III), reflecting as it did that 51% of libraries reported flat or reduced acquisition budgets. Those few that had increases in the materials budgets clustered in the 1-4% and 5-9% ranges in Year II and Year III. What one should take away from this is that libraries moved funds from other budget categories to protect acquisitions and yet materials still took large hits three years running.  Fungible budgets help only to a degree.

We do not have to look far to identify which Peter was being robbed to pay the acquisition budget Paul.  Using expenditures as a guide, the budget components among ARL libraries (Figure X) are human resources averaging 45%.  Acquisitions average 44% and all other expenditures average 11%.  The reductions in the human resources budgets in Year II and Year III (Table IV) indicate that this category bore a much larger share of the burden than materials acquisition and that a larger number of libraries reduced personnel resources—61% in Year II and 53% in Year III.  Thus libraries took larger hits in personnel to either eliminate cuts in acquisitions or to reduce their impact.  A variety of strategies were used to reduce payroll during each of the three years (Figure XI).  Over 78%, eliminated vacant slots.  Hiring freezes were also a common strategy adopted by over 54%.  Layoffs were not uncommon—over 32% employed them at some point during this three-year period.   In spite of these and other reductions in staffing budgets (or perhaps because of them), many ARL institutions did give pay increases to their staff during these years (Figure XII).  Obviously, 2008-09 salary increases were in place when the economy turned south.  On the other hand in the next two fiscal periods a high percentage of libraries reported giving pay increases to both professional and support staff.  In the case of support staff the larger number of libraries giving increases is due to the fact that many were mandated by union contracts.    

Hazy Crystal Ball
If you pay any attention to the commentary and writings of economists, policy makers, politicians and bloviators these days, then you know that there’s an expert opinion about the future to suit your personal mood-swings on a scale from deeply pessimistic and distressed to inordinately optimistic and enthused.  Perhaps this should be called the “Eeyore and Pooh Bear indices.”  My own crystal ball, or tea leaves are far from expert, but we collected a little data that allows speculation.  We asked three questions that indicate what members think will happen in the near-term.

First, we asked about force majeure clauses, that is “a common clause in contracts that essentially frees both parties from liability or obligation when an extraordinary event or circumstance beyond the control of the parties … prevents one or both parties from fulfilling their obligations under the contract.”⁶ Typically, in library contracts we are speaking of “financial exigency” permitting the voiding of the contract.  Of 76 respondents to this question, 53 or 70% indicated that they had such clauses, but only four said they were likely or definitely going to use them.  This gives us a bit of an insight into whether there is an underlying need to reduce obligations further this year.  We asked more pointedly “Do you anticipate with some certainty that there will be reductions to the budget you have reported for 2010-2011?”  A very high 81% answered no (Table V). That is a real difference from Years I and II.  So, for the remainder of fiscal year 2010-11 we will have a fairly stable response in the scholarly communication market from research library consumers in North America.

What about next fiscal year? To get at that, we asked a more open-ended question, “If there is any active discussion/planning for further budget reductions in FY2011-12, to the extent that you can, please describe and comment on the % range of reductions.”  No doubt, the question was answered by our members with some economic context in mind.  For instance, we know that state budgets in the US are under great pressure and that the predominant political mood in the US state legislative sessions right now is to rely on budget reductions not tax increases to meet the common constitutional requirement that state governments have a balanced budget.  In Canada, the impending national elections leave an air of uncertainty. International events like the revolts in the Middle East and North Africa or the tragic impact of the Tohoku Tsunami are roiling international markets with negative effects on endowments.  These events cause uncertainty and uncertainty is bad medicine for the economic recovery and political courage.  Moreover, in the US, many state budgets have for several years depended on one-time “stimulus funds” from the Federal Government to shore up shortfalls in tax revenues, all in hopes of getting to better times.  Those revenues will not be available to paper over tax revenue gaps in FY2011-12.  Together with this, the political budget cutting mood and the slowing or stagnation in the world economy could extend and worsen the budget woes of both private and public members of ARL. On the other hand, ARL institutions are planning in a worst-case frame of mind as though they will have no new funding whereas many have it within their authority to raise tuition.  In Figure XIII we can see that 59 responded to this question and of those 68% are planning for reductions some higher than 10%.  It is realistic to be prepared.  In the current year 47% experienced budget reductions.  Although I would observe that 68% is a bit high, for more than half that number to experience continued reductions or flat budgets next year would not be at all surprising for next year.

For our members and for ARL, the ferment in higher education combined with the ongoing fiscal crisis is very much the condition of the research library environment.  I do not think that all key developments that will transform the model are going on in North America or for that matter in ARL, but many are emerging there.  Communication of trends in the international library market are important, but I think we are likely to be more successful in the future if we also maximize the exchange about the work we are doing to change the academic and research library model.  That is a topic for another paper.⁷

---------

¹   WSJDigitialNetwork, viewed on YouTube March 12, 2011 http://www.youtube.com/watch?v=_mzcbXi1Tkk. 

²   Charles B. Lowry, “’The Future Ain’t What it Used to be,’ So, ‘When You Come to a Fork in the Road, Take It,’” Rough waters: Navigating Hard Times in the  Scholarly Communication Marketplace, American Library Association, http://www.arl.org/sparc/meetings/ala09/index.shtml SPARC-ACRL Forum, July 9-15, 2009, Chicago, IL. Charles B. Lowry, “Year Two of the ‘Great Recession’: Surviving the Present by Building the Future,” Journal of Library Administration, Haworth Press, v. 51 no. 1 (January 2011) p. 37-53. 

³   I want to thank Martha Kyrillidou here for providing more advanced econometric measures that delve more deeply into the data. 

⁴   “Behavioral Finance: Key Concepts - Mental Accounting,” InvestopediaR http://www.investopedia.com/university/behavioral_finance/behavioral5.asp, last accessed March 17, 2011. 

⁵   SPARC-ACRL Forum, http://www.arl.org/sparc/meetings/ala09/index.shtml 

⁶   Wikipedia definition of force majeure http://en.wikipedia.org/wiki/Force_majeure, last accessed March 18, 2011. 

⁷   For some discussion of the changing landscape of higher education and the efforts to create deep change in the research library model see, David J. Staley and Dennis A.  Trinkle, The Changing Landscape of Higher Education,” Educause Review (January/February 2011), p. 16; Lowry, “’The Future Ain’t What it Used to be,’ So, ‘When You Come to a Fork in the Road, Take It,” p. 46-52; and “Ithaka S+R Library Survey 2010: Insights From U.S. Academic Library Directors,” http://www.ithaka.org/ithaka-s-r/research/ithaka-s-r-library-survey-2010.