As my clinical psych education and training advances, I come with more science:
Of all clinical trials evaluating the efficacy of anti-depressants, 31% were never published. Of studies showing the positive effects of anti-depressants, 37/38 studies were published. Of studies showing negative or questionable results of anti-depressants, 3 were published, 22 were not published, and 11 were published in a way that (in the opinion of the authors of this study) put a positive spin on the results. When the FDA reevaluated all anti-depressant studies, the overall effects were found to have been inflated by drug companies about 32%. All things considered, the overall effect size (a measure of improvement, in approximate units of standard deviations) of anti-depressants is 0.31, a "moderate" effect. This is the approximate effect size consistently achieved from placebo across nearly all studies of this type.
Citation: Turner, E. H., Matthews, A. M., Linardatos, E., Tell, R. A., & Rosenthal, R. (2008). Selective publication of antidepressant trials and its influence on apparent efficacy. New England Journal of Medicine, 358(3), 252-260.
In one of the largest and most methodologically sound randomized control trials comparing anti-depressant medication (one of the SSRIs) with cognitive therapy and behavioral activation (a component of behavioral therapy), all were found roughly equivalent for low severity depression (this is not to say they ARE equivalent - it may be that the study had insufficient power to detect a difference due to the relative low improvements someone with mild depression can obtain - this is also known as a floor effect). For those with severe depression, behavioral activation achieved significantly higher rates of remission than both cognitive therapy and antidepressant medications, which were roughly equal. On continuous measures (in contrast to remission or no remission), behavioral activation was not significantly different from anti-depressants, which were both superior to cognitive therapy.
Cite: Dimidjian, S., Hollon, S. D., Dobson, K. S., Schmaling, K. B., Kohlenberg, R. J., Addis, M. E., ... & Jacobson, N. S. (2006). Randomized trial of behavioral activation, cognitive therapy, and antidepressant medication in the acute treatment of adults with major depression. Journal of consulting and clinical psychology, 74(4), 658.
Several studies have found that anti-depressants are no different from placebo for low severity depression. For high severity depression, anti-depressants are shown more effective. The thing to note, however, is that the data suggest not that anti-depressants are more effective for more severe depression, but that placebo just becomes less effective for high severity depression. I don't have a citation for this as it was talked about in several classes and I forgot to ask for one.
All in all, seek CBT (cognitive behavioral therapy) or behavioral therapy for depression. These are generally short (~12 sessions) and are just as if not more effective than medications. The improvement is also long-lasting for most (some will relapse in all treatments), whereas medications work only as long as you take them. Considering the extreme side effects of anti-depressant medications and the cost of long-term use, there is no question which is more beneficial.