The discussion about Milton Friedman going on at the moment is great fun – with a bunch of people discussing whether Friedman was Keynesian and what this even means.
It all starts with Krugman attacking Friedman’s obsession with monetary aggregates.
Then the absolutely spectacular Uneasy Money blog more broadly defending the idea that Friedman was just relying on IS/LM in drag. (Although this is moderated in a follow up and then more context here).
This is used here to attack Friedman and defend Samuelson-Solow (paper). Sounds similar to this. But if this point on not estimating the Phillips Curve is correct the defense is hardly compelling IMO – authors have to be clear about how their work will be taken, and those graphs are misleading! It is good to “condition” results, but we can tell from history that these conditions were largely ignored and that Samuelson and Solow appear to have not said much to disabuse this – this is very important (even if I think the critics were themselves excessive).
Scott Sumner comes out and says Friedman wasn’t Keynesian, wasn’t as ideologically straightjacketed as these two seem to be implying, and that the sticky price models used (implicitly?) by Monetarists and Keynesians were pre-Keynesian.
This all in turn mentions this paper from Brad Delong from 2000 which discusses how monetarism transformed into New Keynesian-ism (the view I’d previously been bobbling around with).
Update: And then along comes Krugman again – although he appears a little dismissive of history before using it. A statement I’m a touch surprised about.
If you want my view in all of this – you must think I know more than I actually do. I am no help at all!
My working assumption is the following when on the internet (so not at all in academia – where I don’t know if school labels seem as useful) – so looking at these terms at the current point in time, and taking these groups as self-identified. The key difference I expect to hear as being between New Keynesian and a myriad of the groups that will term themselves “Keynesian” schools is the assumption around the long-term neutrality of money.
The key difference I expect to hear between New Keynesian and the “New Classical” schools has to do with the identification of the efficacy of monetary policy in the current environment.
The key difference I expect to hear between New Keynesian and the “monetarist” schools is a hard one as far as I can tell I only see these arguments put up as straw-men (I don’t see or meet anyone who calls themselves an old-school monetarist) – and this is where debates around whether money is endogenous suddenly appear from (from what I can tell everyone thinks it is outside of full reserve banking ). Note, I view “New Monetarist” and the “Market Monetarist” labels as essentially versions of people that call themselves New Classical and New Keynesian – just usually a bit more technical
And the key difference between New Keynesian and “Austrian” schools (and some other flavours of Keynesian schools) relies on views of the aggregation of capital and the function of the time path of interest rates on the type and duration of the capital stock. Sometimes these guys can go down the exogenous money supply root as well, but I’m not a fan of that bit.
There is undeniably much more to each argument (for example, I have no doubt that many self-avowed blogging [New] Keynesian’s would state that the Neo-Wicksellian interest rate is time varying and can be adjusted by shifts in government consumption and investment – but this point, and everything with it, is a lot for me to take in ). But as these are all valid points, it is an indication that macroeconomics cannot use (or does not have) data that can allow us to satisfactorily falsify any of these views (which, for all no-one wants to admit it, rely on a series of firm a priori assumptions). Tis a broader version of this Lucas Critique criticism of DSGE models, but to ALL macroeconomic models whether “mainstream” or not.
Note: I would indicate how important testability and empirical analysis is in this – the “long-term neutrality of money” assumption and the “trend stationarity” associated with developed economies have been tested repeatedly and have been shown to hold in many circumstances. Given the dearth of data and conditional nature of these models, this is very encouraging – but of course it is not the be all to end all.
Sidenote: And this reminded me of the first paper that really convinced me about active monetary policy (I didn’t find sticky prices compelling until I saw this – this made the idea of small micro failures having big macro impacts compelling to me) – and it also pointed to the idea that we don’t want to go “too far” from our SR eqm, or risk dropping into a pareto-inferior outcome. I missed this paper – this has been fun.
At the time there were attempts being made to partially microfound Keynesian economics by looking at rationing regimes where either goods or labour markets did not clear. Leijonhufvud was critical of this, and the earlier, emphasis on wage and price rigidity as being at the heart of Keynesian economics, both in terms of an interpretation of Keynes but also as useful macroeconomics. What I also remember, but did not fully appreciate at the time, was a good deal of discussion of the importance of intertemporal coordination failures, in which people’s expectations about long-term interest rates differ from the marginal efficiency of capital.