A big issue I personally have with the concept of Keynesian economics is that, in practice, it involves way too much discretionary action by people who often really shouldn’t have the kind of economic power they do. The whole thing is part of a wider discretionary vs. rules-based economic policy debate, and something on which I find myself in remarkable agreement with the freshwater school in economics, who often don’t seem to make much sense on other grounds. There is a good argument to be made that when stuck in a liquidity trap (as long as something like that is even the big problem it is often portrayed to be) fiscal policy is the go-to economic tool to help get the economy back on track. Most of the standard arguments against this generally move somewhere between the either claiming it is totally ineffective or that it is so effective that the stimulus always proves to become permanent after a while. Both arguments are mostly wrong. But the real issue I see is that discretionary fiscal spending is, basically by definition, highly uncertain. The recent crisis has shown that when push comes to shove, politicians of all stripes will dig up their own copy of The General Theory and start throwing money at the problem. Yet, even though in theory it would be fairly easy to imagine a fixed rule which would be followed in these situations, in practice both the size and the direction of this throwing of money is highly arbitrary.
This is also one of the reason why modern automatic stabilizers have proven to be so effective. They are in general fairly permanent institutions, which means that politicians can’t screw up too bad even if they wanted to. Sometimes they are tweaked a bit, such as extending the period over which unemployment benefits are paid out, but at their core they remain a key source of economic certainty particularly at times when certainty is lacking the most. An in contradiction to what many commentators on the conservative side of the spectrum generally suggest, recent experience shows that these automatic stabilizers should be much stronger and much more pro-cyclical than they are at the moment, not less so. Unemployment benefits, for instance, do not make unemployment worse, as is generally suggested, but prevent the entire economy from collapsing even more than it did in 2008 and beyond. And if these benefits are tied to people having to actively search for another job, as is often the case, they are might even help make the jobs market more flexible rather than less flexible by preventing people from dropping out of the labor force entirely.
Predictable, rules based policy making becomes arguably even more important – and at the same time easier to pull off – in the realm of monetary policy. Predictability is the whole reason why concept like the Taylor Rule, NGDP level targetting or the Evan’s rule are so attractive – if correctly followed, they greatly reduce the degree of policy uncertainty and thus economic volatility in general. Forget what might be taught in introductory economics concerning central banks having to “surprise” the market in order to actually move the economy. A central bank having to surprise anyone is a central bank failing miserably at doing its job. Europe, for instance, continues to be stuck in recession-like circumstances because the ECB consistently has stated it does not care about the European economy. If it did, Inflation wouldn’t be far below target while unemployment in the double-digits. So while the ECB actually doing it’s job would indeed prove to be surprising, the only reason it would have any effect at all is because it has failed to do so for so long.
The whole concept of central bank credibility is strongly related to this – it means nothing else than the central bank actually doing what it has promised to do, and doing so consistently. Actually, that’s not entirely correct. In reality it means that the market has fairly coherent views on how the central bank will react in different situations, and the central bank doesn’t do anything to prove the market wrong. Credibility is enormously important for price stability, particularly in the presence of issues like the time inconsistency problem of monetary policy, yet in the current situation it seems to actually prove to be a big issue. The ECB has credibly stated that it won’t do more, and even though by ever imaginable measure it should do much more the market believes it and reacts accordingly, and thus in some sense it is the ECB’s high level of credibility that is hurting it. Adhering to and actually following a clearly voiced rule, however communicated, also has the added benefit that it makes the jobs of central banks a lot easier by having the market do much of it for them. That’s the whole point of people trying to make the argument that we need a higher inflation target to get out of this mess and prevent us from falling into this kind of mess in the future: if the central bank can credible promise to do whatever it takes to bring about the higher level of inflation, the market will accept this, and given it’s never a good idea to try and fight a central bank when it has said it wants to print money, will most likely help to bring about a higher level of prices without the central bank actually having to do all that much – and will then also help keep it there.