Madhavi Sunder's framing questions are pitching this discussion quite high -- appropriately so, given what's at stake. As my grasp of the debate is not, unfortunately, as "panoptical" as hers or Prof Merges', I can only try to contribute by bringing in a more limited, different perspective.
Coming to IP not through a formal training in law but rather through debates over scientific authorship and the history of patenting by scientists, I am not directly invested in either supporting or critiquing economic conceptualizations of IP. From my disciplinary point of view, that so much discourse about IP is framed by the logic and assumption of neoclassical economics does not look like a "natural" fact, but rather the result of some kind of history. Still, that does not imply that economics would necessarily provide bad or unacceptably narrow conceptualizations of IP. While I find no reason to attribute them the status of a "master narrative," different economic perspectives on IP can function as tools that, depending on who uses them and how, can produce results ranging from illuminating, sharp, and thought-provoking to flat and predictable as only ideological scholarship can be.
I'd say that the way one views the relationship between economics and IP depends on how "scientific" one takes economics to be. For instance, in my neck of the disciplinary woods, scholars like MacKenzie, Callon, Latour, and Lepinay have argued, in different ways, that economics does not simply describe the dynamics of the economy (the way a science would describe its object). It rather develops models that, once adopted by powerful actors like the finance industry, start functioning more like prescriptions than descriptions of the markets' behavior. "Economics perform the economy," says MacKenzie. While these perspectives acknowledge the powerful effects of economics (as they argue that economics is part of the construction of the economy), they also take away from it its status of 'science' and replace it with that of "builder" or "performer".
I wonder whether we could apply some of these perspectives to the relationship between economics and IP. One could argue, for instance, that while the discourse of economics has indeed shaped much of the doctrinal and policy directions in which IP has been articulated, it has not done so because it was able to model and describe the motivations and behavior of authors and inventors in a "scientific" manner, but rather because that kind of modeling has been used and turned into practice by powerful political and corporate IP stakeholders (and also, perhaps, by individual authors who, with so much economic imagery swirling around in media and public discourse, have come to conceive of their own creative work as a process of "maximization of intellectual property"). MacKenzie has shown that the Black-Scholes-Merton formula for pricing options ended up structuring (rather than just describing) the behavior of securities markets. Maybe the economics talk about IP has performed the same kind of "self fulfilling prophecy". That does not men that that kind of talk is either right or wrong. It is simply constitutive. Which also means that it could be constituted differently -- even very differently.