Scottish Prison Population Projections: February 2026

As of 1 February 2026, the prison population was 8,257. The modelling indicates a projected range for the average daily total prison population of between 8,200 and 8,800 in July 2026.

It is highly likely that the overall prison population will increase over the projection horizon.

The remand population is sensitive to a range of factors, which is reflected in the projected range of 1,450 to 2,500, compared to 2,156 on 1 February 2026. This implies that there is a realistic possibility that the average daily remand population could either increase or decrease over the projection horizon.

It is almost certain that the sentenced population will increase between February and July 2026. It is projected that the average daily sentenced population could be between 6,200 and 6,850 in July 2026, compared to 6,101 on 1 February 2026.

For both the total and sentenced populations, it is important to note that the population on 1 February 2026 is estimated to have been approximately 180–200 lower than it would have been had EER not been implemented and that the impact of EER is not sustained. 

Over the coming months, remand and sentenced arrivals, and case conclusion levels will be monitored to assess which scenario the actual prison population most closely tracks.

As previously highlighted, the key short-term drivers of prison population change are the flows into and out of the remand and sentenced populations. These flows are influenced by factors such as the court recovery programme, including the reallocation of resources from summary to solemn in April 2023. The current modelling assumes no further changes to the number of courtrooms allocated to summary and solemn procedure until at least July 2026[1].

On 3 February 2026[2], the Scottish Government announced proposed measures to help manage rising prison numbers, including changing the automatic release point for certain short-term determinate prisoners from 40% to 30% of their sentence served in custody[3]. This would not apply to those serving sentences for domestic abuse and sexual offences. It should be noted that these projections do not include the effects of these measures, as these are still subject to Parliamentary approval.

For several reasons, the microsimulation becomes less accurate over longer time horizons. This could be due to systematic errors in the model, inaccurate assumptions due to unforeseen circumstances, or limitations in the model’s ability to adapt to evolving and unforeseen circumstances. Experience to date suggests that a six-month projection window provides a reasonable balance between reliability and foresight. However, in recent months, the microsimulation has undergone several methodological changes aimed at extending the projection horizon beyond six months. We intend that future publications will include longer-horizon projections and a more detailed overview of the modelling methods used.