How does OA contribute to your academic reputation?
The first significant scholarly repository, arXiv.org, was started by high-energy physicists, but quickly expanded to include other scientific disciplines. Today, archiving in arXiv.org is practically a necessity for physicists. Archiving provides the physics community with faster communication, better access, and the opportunity for an unofficial peer-review, even before the work has been submitted to a journal. Unfortunately, journal publishing is still important for official purposes (promotion, tenure, etc.) and so libraries are still forced to buy physics journals.
Another example is PLoS. PLoS peer-reviewed journals charge the authors a fee in return for publishing, and papers are published under Creative Commons licences. PLoS journals probably wouldn't have been as successful (PLoS Biology had the highest IF in its category in 2009) if libraries were asked to pay for them. It pays off for authors to pay PLoS for the prestige of publishing in such high-impact journals.
Eysenbach (2006) found that immediate-OA papers in PNAS, in comparison with those who weren't immediately available in OA, were more likely to be cited earlier and more often. However, Davis et al., who studied papers from 11 journals published by the American Physiological Society, haven't found an increase in citations for OA papers in the first years after publication. They did find that these papers had significantly more downloads and unique visitors, which means OA papers reach wider audience.
Open Access also allows scientists to build reputation in places and among people who otherwise wouldn't have been able to read their researches (the MIT OCW is very popular in developing countries). Given how important citation metrics are to a scientist, publishing Open Access seems to be a very logical step, career-wise.
Willinsky, John (2010). Open access and academic reputation Annals of Library and Information Studies, 57, 296-302
Davis PM, Lewenstein BV, Simon DH, Booth JG, & Connolly MJ (2008). Open access publishing, article downloads, and citations: randomised controlled trial. BMJ (Clinical research ed.), 337 PMID: 18669565
Eysenbach, G. (2006). Citation Advantage of Open Access Articles PLoS Biology, 4 (5) DOI: 10.1371/journal.pbio.0040157