Do journal editors have an obligation to disclose referee reports?

Philosopher Laura Schroeter (Melbourne) writes:

In September last year, I submitted a paper to Mind. In March, I discovered by chance that a senior colleague had written a referee report on the paper recommending conditional acceptance but had raised one main concern he thought should be addressed before publication. After many promptings, Mind finally made a decision to reject the paper this September. The editorial administrator attached comments from the editor (who is not a specialist in this area) and comments from one referee (which appear to be redacted). The comments from the colleague I’d spoken to earlier were not included; when I contacted him directly, he sent me a detailed 5pp report. Mind has not responded to an inquiry about whether they have a policy about not forwarding referee reports to authors. Moreover, the official rejection letter included the following proviso:

"Please note that referees' comments are supplied in confidence for the Mind editorial board.  They are forwarded to you as confidential reports, in order to provide you with feedback concerning your submission.  The Editor asks that you respect the confidentiality of this advice."

To be clear: I respect the right of editors to make the final call on a paper, and to override referees’ verdicts. But it seems to me that a policy of not forwarding reports is a real betrayal of the crucial role journals play in the discipline as intermediaries between authors and critics. When I spend time writing detailed feedback on an argument, I expect my report to be forwarded to the author – I wouldn’t go to all that trouble if it were simply advice to an editor. Personally, I won’t be refereeing for Mind in the future now that I know that they consider reports the confidential property of the editor alone.

I had not heard of a case like this previously; I thought referee reports, unless content-free, were always sent to authors.  What do readers think?  What experiences have they had?  If you name a particular journal, you'll need to sign your name to the comment; other comments must include a valid e-mail address at least.