Because competition is never perfect, markets themselves are never perfect. I take that as an argument in favor of markets, not against them, because the incentives involved in market-like competition—rewards that roughly track achievement—spur us to try, try again to achieve and to see failures and setbacks not as final results but only as steps in an ongoing iteration.
Market imperfection holds in marketplaces of ideas as well: Sometimes good ideas fall through the cracks, while inferior ideas get buoyed along; sometimes important figures in the history of thought can lapse into obscurity while figures of lesser achievement can continue to be read, studied, or venerated. I am sure you can think of examples of all these cases in your own discipline, as I can in mine.
These imperfections are opportunities. When they are recognized, they can be exploited by knowledge entrepreneurs both for private gain and for public benefit. Thus if, say, a junior professor (re-)discovers a forgotten idea in a historical figure, it can not only lead to publications that can help earn tenure but it can also enrich the current conversation by saving us from having to reinvent the wheel, from repeating mistakes that have already been located and diagnosed, etc.
Academics are professionals in the seeking-truth business, which, put otherwise, means they are professional players in the marketplaces of ideas. I think the fact of market imperfection imposes a special obligation on them. This obligation arises because the limits of human understanding entail that we cannot always recognize when market imperfections obtain. We do not always know which ideas, opinions, or judgments will stand the test of time (and which will not); we do not always know which figures’ work will lead to important truths and whose will turn out to be blind alleys; and we do not always know what areas of investigation will occur to us in the future that will lead to new, unexpected, and as-of-now-unpredictable avenues of thought.
How will we discover these imperfections in the marketplaces of ideas? John Stuart Mill provides an apt suggestion: “Precisely because the tyranny of opinion is such as to make eccentricity a reproach, it is desirable, in order to break through that tyranny, that people should be eccentric” (On Liberty, chap. 3). Mill’s claim that “people should be eccentric” translates in the setting of the academy, I submit, into the imperative to be the other side. When we find places where conformity of opinion is regarded with more deference than reasons for the opinion, there we find a symptom of market failure. If, then, we are under obligation to do what we can to uncover the truth—which I argue is, or should be, one central principle in the academic’s code of professional ethics—then we are under obligation to challenge the opinion.
As Mill says, this spirit of intellectual confrontation and challenge is so essential to the progress of our understandings that “if opponents of all-important truths do not exist, it is indispensable to imagine them and supply them with the strongest arguments which the most skillful devil’s advocate can conjure up” (On Liberty, chap. 2). This is very difficult to do, both for the holders of received opinions and for their challengers. The holders may have invested a substantial amount of intellectual capital in the opinion(s) or worldview(s) being questioned, and the natural reluctance to fight to hold on to one’s intellectual investments regardless of the objective strength of the challenge is formidable. The challengers, for their part, find that they are often punished, not rewarded, for their efforts, and they can soon come to understand exactly what Mill means by “tyranny of the majority.”
But the nature of our jobs as academics requires that we be willing to risk those difficulties, for the cost of silencing opinions or closing off areas of debate may include sacrificing truth itself. There may be times that we qua parents or qua citizens or qua members of this or that organization, association, or community may be willing to sacrifice the unfettered search for truth for the sake of the stability and comfort that comes with conformity, but qua academics we cannot do that—without betraying a central aspect of what defines us as academics.
The marketplace of ideas is imperfect. Fashions, conventions, bigotries, and errors are as much a part of it as are the disinterested pursuit of the truth, the objective weighing of reasons and evidence, and the willingness to follow the argument wherever it leads. For this reason, Mill was right when he articulated one half of the moral obligation of the academic:
If there are any persons who contest a received opinion, or who will do so if law or opinion will let them, let us thank them for it, open our minds to listen to them, and rejoice that there is someone to do for us what we otherwise ought, if we have any regard for either the certainty or the vitality of our convictions, to do with much greater labor for ourselves.
Thus we must nurture, encourage, and charitably receive people proposing ideas contrary to our own. The other half of the academic’s duty, I submit, is that, when we see no one forthcoming willing to challenge our most dearly held beliefs, then we ourselves are called upon to adopt the position—as fully and and as faithfully as we can. This stance of the contrarian, which has a long and venerable history, and which indeed may be what initiated the intellectual tradition that we in the West inherit, may sometimes issue in personal cost; but that is a price we academics must pay if we are to remain true to its highest ideals.
PILEUS 
One of the reasons why the marketplace of ideas is imperfect is that the checks on error are imperfect. If a politician tells voters what they want to hear, rather than the truth, the politician is in many cases more likely to be re-elected.
The inefficiencies in the political market are clear, but the same problem applies to much of academia. What determines whether you get published in high-profile venues? Not just whether you have good arguments or evidence, but whether the “packaging” appeals to other academics, i.e., whether it’s something that they want to hear. Certainly, it works that way in the social sciences, and I imagine the humanities too.