10 Association of Research Libraries Research Library Issues 295 — 2018 colleagues in the US. The portion of the Canadian ARL professional population that is 65+ rose from just 0.4% in 2005 to 4.7% in 2015. Figure 4 Retirements Delayed, Not Eliminated Whatever the cause or causes of ARL’s delayed retirements, there is no evidence that people are foregoing retirement altogether. Retirement is a deeply rooted, cultural, even biological phenomenon, one that is affected but not determined by changing economic conditions or legal provisions.10 Among ARL professionals, a paltry 1.5% continued to work at 70 and beyond, and while that represents an uptick relative to 2005, this is still a very small number of individuals. With eventual retirement a virtual certainty, what can we expect in the five years between 2015 and 2020? It is not possible to derive a formal retirement rate from the ARL demographic data, but given that net changes to the size of cohorts are fairly consistent from one data set to the next, rough estimates are possible. For example: in any given year, the percentage of the population aged 60–64 drops by about half relative to the 65+ cohort five years later.11 Applying this calculation to the 2015 population, we would expect that by 2020, 689 vacancies will arise from the 60–64 cohort alone.