The key finding:
We ﬁnd that the eﬀect on exit from unemployment occurs primarily through a reduction in labor force exits rather than through exit to employment (job ﬁnding). This is important because it implies that extended beneﬁts do not delay the time to re-employment substantially and so do not have large ﬁrst-order eﬃciency consequences. The major eﬀect of extended beneﬁts is redistributive, providing income to job losers who would have exited the labor force otherwise (consistent with Card et al. 2007).
The first sentence suggests that when unemployment benefits run out people are not taking a job but are completely giving up and exiting the labor force. They simply don't believe there is a job available for them to take or they are qualified for. If they live in an area where the job market is weak, even in a recovery stage, this makes sense.
The last sentence suggests that people stay on the extended benefits because there is no alternative to having that income stream. They were likely to exit the labor market whether the time limit for benefits was 26 weeks or 99 weeks (the maximum at the height of the recession).
The overall conclusion is that a statistically significant number of people DID NOT use extended unemployment benefits to avoid taking a job.
My own caveat that is not addressed in the paper below but likely has some correlation, is the rise in Federal Disability claims that overlaps this time period. In other words, how many people moved from Extended Unemployment benefits to the Disability rolls. Maybe statistically significant or not. Sounds like fodder for another research paper.
Note for teachers and students. Starting on page 3 there is an excellent, basic explanation on unemployment benefits and the timelines for receiving benefits. Students ALWAYS have these detailed questions.