Robert Waldmann brings it up. Tobin delivered it in 1971, and it was published in 1972.
Pointer from Mark Thoma.
As I remember it, Tobin suggested thinking of an economy with two industries and wages rigid downwards. Suppose that demand shifts away from industry X to industry Y. Because wages do not fall in industry X, you get unemployment there. Because wages do rise in industry Y, the overall rate of inflation goes up. Note, however, that with more inflation, the real wage in X falls, which means less unemployment there than otherwise.
This simple story gives you an explanation for both stagflation and the Phillips Curve. The point that Waldmann is making is that macroeconomists did not need to take the detour that they took in the 1970s. They could have stayed on the path that Tobin laid out for them. My thoughts:
1. It is amazing how much better you can do if you break up the GDP factory into two industries. I think you can do even better with more disaggregation, but the modeling would be much hairier.
2. I agree that macro would have done better to follow this path. However, macro still would not be very good. The problem of too many plausible causal factors chasing too little data is insurmountable. See my science of hubris paper, as well as the recent Paul Romer screed.
3. The sociology-of-economists question of how macro remained (and continues to be) stuck for so long is quite interesting. See Daniel Drezner’s piece (for which I also thank Thoma). As you know, my explanation is that Stan Fischer became the Genghis Khan of macro.
Of course, the 1970s disaproved the Philips Curve, yada yada yada. However we have to remember the 1970s had the highest growth labor force growth in the post WW2 era as the Boomers were reaching adulthood and entering the workforce. My guess is Macro does need to include more national trend impacts.
So yes, macro is so new societyand economic changes will always throw it a curveball.