Rules, Discretion and Noble Lies
There are lies that help people accomplish goals and hold one another responsible for their actions. People who believe that success is the result of effort and not luck are more likely to work hard, and thus succeed. Yes, much of success is due to luck, but the false belief enables people to succeed.
The Fed can’t perfectly control the price level (or any other nominal variable for that matter). There are a lot of shocks that are more powerful than the Fed, at least in the short run. But what standards should we hold the Fed to? What optimal beliefs the Federal Reserve Board should hold to do a good job stabilizing the economy?
Outcome = (Factors the Fed controls) + (Factors the Fed doesn’t control)
I believe that the Fed has powerful control over the long run, and less control over the short run. If the Fed wanted 100% inflation over the course of a decade, that would be be easier than causing 2% in a month. But if short run shocks cause inflation to be 1% higher than the target, the Fed adjust their actions to compensate, and eventually they will be able to bring the price level in line with their long run objective.
Which institution will create less misbehavior, rules or discretion?
If the Fed were to be controlled via a rule, that rule would be determined by Congress. Congress, to put it gently, does not consist of macroeconomic experts. Actual monetary policy history is a parade of failure. There have been many dumb ideas that have held sway at various points in history. Once you make something a law, it fixes that idea for a long time. Our current “dual mandate” is a good example, yet no one has bothered to change the rule.
If a rule were good, a discretionary policy could just choose to follow the rule of their own accord. The Fed just does whatever the consensus is among macroeconomists, so not having a rule just means it’s decided by the sort of people who would come up with the rule anyway. Why not continuously determine what the best policy should be, in real time?
What does the central bank control? No one knows! Legally, the Fed has a dual mandate of inflation and unemployment. But no one these days believes that the Fed can actually target unemployment. They might as well have a “color of canaries” target, or “flavor of watermelon” target. Everyone knows this, but every time they miss their never-specified inflation target, they just mumble something about unemployment and people let them off the hook!
When things go wrong, the Fed says “We’re doing discretionary policy so you can’t really criticize us.” Imagine if everyone were so lucky at their jobs! You go to a restaurant and not only won’t they tell you what kind of food you’re going to get, when it tastes bad, you can’t complain because the cook cites his “cooking discretion”. That’s not good enough for the world’s most powerful economic actor. Any change in a target itself makes the central bank less effective because their most important goal is stability itself.
My main complaint about discretion is that it makes talking about macroeconomic general equilibrium impossible. People write all kinds of nonsense about how certain actions will affect macroeconomic variables: increasing capital taxes won’t reduce spending, government spending will increase inflation, every single statement made about real interest rates, etc. The internet is full of it. If the central bank had a target, I could say “that won’t happen because the central bank will/won’t offset it. Now, if I say “assuming the central bank targets inflation, blah blah blah”, people retort with “yeah, well they don’t target inflation all the time so my nonsense is really true!”. B.S. hides in randomness.
It’s not a trivial matter because Congress is a big purveyor of and victim of, B.S.
Source (modified for compactness).
ARRA did not improve the supply side of the economy, so any positive effect was purely from the demand side. Furthermore, not many people were hired because of ARRA (see also here and here), so the entire weight of the claim of lower unemployment is the secondary spending caused by the bill – the additional spending from people who got money from the bill.
Now, try to make the claim above in a world with an inflation target. It’s impossible. Fiscal stimulus works in a fixed inflation world iff it improves the supply side of the economy. Either it makes efficient investment, or increases TFP, or some other microeconomic reason why unemployment should go down. With fixed inflation, if the government moves the IS curve left, Y decreases, leading to lower employment.
The only way the fiscal stimulus works is if a.) the central bank leaves the inflation target, b.) The fiscal authority increases inflation and c.) The central bank never would have increased inflation back to the trend line in the absence of fiscal action. That chain of events can only happen in a “discretion” world.
I’m not a believer in the “zero lower bound” theory. A central bank can always inflate. If you don’t think you can do it, give me the printing press for a few days and I’m sure I can work something out.
There are good reasons for not wanting to hand control over monetary policy over to a fixed law, however, pure discretion can be mere chaos in practice. What if the central bank had to follow a rule, but the rule they followed was up to their discretion? A central bank could pick an inflation target, an NGDP target, a price of apples target, whatever, and then they would have to hit that target until they declared a new target. You get most of the flexibility of discretion, but the predictability, stability, and clarity of a rule based policy.