Basically when you say "there are studies that show X" you aren't saying anything because "studies" are cheap unless your methodology is rigorous and sound, which isn't the case a lot of the time.
http://archpsyc.jamanetwork.com/article.aspx?articleid=20429...
http://ajp.psychiatryonline.org/doi/full/10.1176/ajp.2006.16...
http://archpsyc.jamanetwork.com/article.aspx?articleid=20433...
Apparently, 91 papers passed Nussbaumer et al's assessment of the abstract (and were presumably randomized, controlled trials):
We assessed 91 full-text papers for inclusion in the review, but only one study providing data from 46 people met our eligibility criteria.
Those eligibility criteria were:
For efficacy, we included randomised controlled trials on adults with a history of winter-type SAD who were free of symptoms at the beginning of the study.
I think it may have been the focus on prevention (rather than treatment of symptomatic depression) that excluded all the other papers.
I just skimmed the paper the paper the first time and didn't notice that the authors were searching for studies of a very specific question: prevention (excluding treatment) of seasonal affective disorder (excluding any other form of depression).
> the small sample size of the only available study
The only study?
https://scholar.google.com/scholar?q=light+therapy+depressio...
There are numerous studies showing a benefit from light therapy.
It is true that the the quality of the evidence could be improved. That's a political/financial issue, not a reflection of the efficacy (or lack thereof) of light therapy. The available evidence supports the treatment.