Authorities or legislators charged with making the decisions have a right to know what an economist thinks, not about their likes and dislikes, but about the effects a particular measure, policy, or institution would be likely to have if it were adopted. Hence, the economist should frankly state what he would recommend if the constraint of “political feasibility” were removed, that is, if he could assume the absence of political resistances he thought were preventing the acceptance of certain proposals.
— Fritz Machlup, “Why Economists Disagree,” Proceedings of the American Philosophical Society 109, 1 (1954), p. 3.
I was having trouble making use of this quote, but some recent work by Robert Murphy has given it some relevance. Paul Krugman, Brad DeLong, et. al., justify their focus on fiscal policy during 2007–09 by arguing that conventional monetary policy had “shot its bolt,” and it was unlikely that unconventional monetary policy would be adopted in sufficient force (nevermind the fact that the Fed’s program to buy very large amounts of mortgage backed debt assets, as the financial crisis unfolded, was unconventional). Economists can turn around and make the same argument today, as to why advocating “QEinfinity” might be better than pushing for fiscal stimulus.
As Machlup argues above, though, the economist should always advocate what he or she finds ideal. But, I’ll take a few steps beyond Machlup; the economist should be frank and clear not just because the politician should know the various consequences of different actions, but for the sake of clarity and consistency in and of itself.