to Elsevier Title Level Pricing: Dissecting the Bowl of Spaghetti

INTRODUCTION This study will explore the issue of pricing opacity associated with prices paid by academic libraries that have recently unbundled from the Elsevier Big Deal journal package. Additionally, this study will provide metrics for assessing the fair market value (FMV) of unbundled journal packages. The pricing metrics will assist academic libraries in negotiations of subscription and open access agreements. METHODS Pricing information was gathered from five academic libraries. The data was analyzed to arrive at two key metrics (adjustment from list price and the average cost per journal) for establishing comparables, i.e., prices paid by similarly sized institutions, to assess the collective FMVs for unbundled Elsevier journal packages. RESULTS & DISCUSSION The study results show that significant variations existed in the way institutions were charged for content. Additionally, the comparables show wide variations among institutions when measured by the overall adjustment from list price and the average cost per journal. CONCLUSION The pricing metrics developed in this study, adjustment from list price (ALP) and average cost per journal (ACJ), will help libraries assess their final net prices for individual journal subscriptions. The results will be useful to administrators, collection development personnel, and negotiating teams in understanding the prices paid by other institutions for unbundled journal packages to determine FMVs.


FMV Metrics (ALP and ACJ ) varied widely among institutions
3. What is the total adjustment from list price (ALP) by institution? 4. What is the average cost per journal (ACJ) paid by institution?
The final net price was below the published list price for three institutions (Iowa State, UNC-Chapel Hill, and West Virginia). For two institutions (Florida State University and Institution A), the final net price was above the published list price and resulted in a premium paid above the list price. The average ALP for all institutions totaled -3.6% (discount from list price). The minimum ALP totaled 7.2% (premium above list price). The maximum ALP totaled -14.6% (discount from list price). The standard deviation (variation from the average) was +/-9.8% (see Table 4 and Table 4.1).
The ACJ across all institutions was $4,793. The minimum and maximum ACJ were $3,673 and $6,239, respectively. The standard deviation (variation from the average) was +/-$1,109 (see Table 4 and Table 4.1). The average number of unbundled subscribed titles was 308. The minimum and maximum number of unbundled subscribed titles were 150 and 405, respectively (see Table 4 and Table 4.1).

Unbundled Savings
5. Do libraries achieve savings when moving from the Big Deal to title by title subscriptions?
All institutions in our sample realized significant savings from unbundling. The average unbundled savings was $795,200. The minimum and maximum unbundled savings were $369,000 and $1,104,000, respectively (see Table 5 and 5.1).   (1) A positive (+) number represents the premium paid over list price. A negative (-) number represents the discount received from list price.

Implications for Academic Libraries
Similar to research on Big Deal pricing (Bergstom, Courant, McAfee, & Williams, 2014), our findings demonstrate pricing discrimination by Elsevier for its unbundled, title by title pricing to libraries. Price discrimination is demonstrated across both metrics used in the analysis, ALP and ACJ. The differences being charged by Elsevier are troubling and show an opaque journal market devoid of FMV concepts. This has allowed Elsevier to construct its final net price from a confusing array of pricing components, leaving similar institutions paying widely varied prices for the same content. With its -14.6% ALP, Iowa State paid only $1,803 for Acta Biomaterialia in 2020 while Institution A, with a 7.2% ALP, paid $2,262 for the same subscription.
The pricing data reviewed in this study shows that the components Elsevier uses to arrive at a final net price unnecessarily add complexity and opaqueness to title level pricing. Absent comparable metrics like ALP and ACJ, Elsevier's variable pricing components make it very difficult for a library to know whether they are paying anything close to FMV. Using ALP and ACJ, libraries can get at least some idea of how their pricing compares to their peers. However, the ultimate solution is for Elsevier to establish title level pricing  that is simple and transparent. This would allow easy pricing comparisons between institutions, and we highly recommend Elsevier move in this direction.
The wide ranges of ALP and ACJ encountered across our small sample suggest libraries have an opportunity (if not obligation) at the negotiating table to achieve more favorable pricing. The 21.8% ALP difference between the sampled institutions getting the best and least favorable pricing is significant. This range can be considered the playing field across which negotiations will establish title level pricing. Libraries can achieve significant savings by negotiating an ALP towards the favorable end of this range. The ALP range found in this study also shows how much work Elsevier has to do if they are going to undo, rather than perpetuate, their pricing spaghetti.