RWCG


Does nominal targeting work for other problems?
December 12, 2012, 3:51 pm
Filed under: Uncategorized

I gather the Fed announced yet more stimulus** (**which is a Latin word so you know it’s totally scientific) so I just thought I’d say that the concept of ‘NGDP Targeting’ makes me giggle.

Look, I don’t know. Maybe it’ll help. But to me it just sounds like, your car’s engine has a serious problem and as a result you can’t even get the car up to a 50mph speed, but instead of, y’know, taking the car into a shop for an overhaul to get its engine fixed, you ceremonially announce your ‘intention’ to do some iterated combination of

1) Press on the gas pedal harder
2) Relabel the numbers on your speedometer a bit higher

until the needle gets to “50”. In other words, you are ‘targeting’ a Nominal Speed of 50.

Will this work? Sure, I can’t imagine why it wouldn’t.

Will this help your car? I must admit, I have my doubts.

But it gets even worse. In their latest announcement the Fed says they’ll keep interest rates low until the jobless rate gets to such-and-such. I guess that’s fine and dandy if there is some sort of predictable hard and fast functional relationship between the jobless rate and interest rates:


jobless_rate = f(interest_rates)

Is this really what all the Smart Economists believe? Because such an announcement is retarded if it’s not the case. Isn’t it?

I mean, I wonder if I could ‘target’ an annual income of $10,000,000/year via playing the computer game Dark Forces 2: Jedi Knight on my computer for as many consecutive hours as it takes. Should I try that? Would it matter whether my announcement of such an intention was ‘credible’?

At best, such an announcement might be an announcement of an intent to do X until Y happens, when Y was gonna happen anyway. “I declare my intention to Make The Sun Rise by playing Jedi Knight.” The sun will, indeed, rise. Thanks to me? Unemployment will, likewise, get below 6.5% at some point or another. Thanks to Bernanke?

I don’t know, maybe this all makes sense and I’m just too dumb – not Smart enough – to see it. I’d say I’d read Scott Sumner’s blog more closely, overcoming the boredom that usually prevents me, until this all finally clicks into place in my brain and the sense of it all dawns on me. But that would be a lie. You know it and I know it.

About these ads

27 Comments so far
Leave a comment

“Dark Forces 2: Jedi Knight ”

Strong selection

Comment by tangentstyle

Hey, I probably put in 20-30 hours of gameplay back in the day, and my salary is indeed higher than then, so I guess it worked.

Comment by Sonic Charmer

Clearly, Romney should have nominally targeted about 5 million more votes than he did.

Comment by Matt

Perhaps he targeted votes*[money spent], and achieved his goal. The NGDP target idea is similar.

Comment by Sonic Charmer

What value would target unemployment be if the number is total bullshit to begin with?

Comment by Bob

Actually, then the whole thing might in fact make sense, as a veiled threat to whichever government statheads cook the books. ‘We’re gonna keep doing this until that number gets to X…if you get my drift.’

Comment by Sonic Charmer

We should have a hurricane Sandy every month! Imagine the increase in GDP from rebuilding New Jersey twelve times per year!

Comment by Dave

Also, your point about joblessness being a function of interest rates is awesome. If I were a Big Media Reporter, I would waste my entire career by asking the question to the Bernank:

“Mr. Chairmen, can you describe the shape of the function that relates interest rates to joblessness? A rough sketch would do nicely.”

Comment by Dave

“Look, I don’t know. Maybe it’ll help. But to me it just sounds like, your car’s engine has a serious problem and as a result you can’t even get the car up to a 50mph speed”

You do this all the time. You want to drive 50 mph, you start going up a hill, you notice your speed slipping so you press on the gas. You level off and start going over 50 so you take your foot off the gas. You start going down a hill so you maybe even use the break to keep to 50 mph.

Comment by boonton

That’s great and all if your car can actually get to 50, but not if it can’t (as in the actual Problem Setup above). Grade: Incomplete

This isn’t an idle example. I once had a car whose engine had a cracked cylinder head or some such, and had very little power as a result. And so no, pressing the gas did not help maintain my speed when I was heading up a hill. In fact, in the event, I had to put it in reverse and back/coast down the hill, and try to find a more topologically-constant path back home. Fun date.

Comment by Sonic Charmer

Jobless rate function of the interest rate? Roughly yes it is. There is, though, a min. amount of joblessness you can achieve before you start getting inflation rather than further reductions in the jobless rate….if you read carefully you’ll note that the announcement was that the Fed was looking at both the unemployment rate and the inflation rate before it would decide to raise rates.

To use your analogy of stepping on the gas pedal, if you were driving up a very steep hill in a car that didn’t have a powerful engine, you’d notice at some point giving the engine more and more gas may not get you above 40 mph. At that point you’d conclude you are at such a steep hill that 50mph is just beyond the capacity of your car to do. Likewise if 6.5% unemployment is just beyond the ability of our economy to manage, that should show up in inflation at some point which is a signal the Fed would be on the lookout for just as you’re on the lookout for your engine sounding like it’s struggling.

Comment by boonton

And suppose the jobless rate isn’t a family of monotone functions of interest rates, parametrized only by the inflation rate? I know, I know, that’s crazy talk. But just suppose?

Comment by Sonic Charmer

Care to ask that question again in a bit more down to earth language?

Comment by boonton

Naw, I’m happy with how I phrased it.

Comment by Sonic Charmer

OK, using http://en.wikipedia.org/wiki/Monotonic_function to refresh myself on monotonic functions, not seeing what your question really is. Are you asserting there’s some point where lower interest rates will cause employment to cease increasing and begin decreasing? Or are you saying that the employment level is determined by things other than the interest rate?

Well strictly speaking is a car speed a monotonic function determined only by the gas pedal? It may not be, yet if you were a passenger with a driver who wasn’t keeping up with traffic you’d tell him to ‘step on the gas’

Comment by boonton

Neither. If you read closely, I’m not the one declaring unemployment to be any kind of function of interest rates whatsoever.

Comment by Sonic Charmer

So you’re saying the interest rate has no relation to employment. Why do you feel that way? What do you think the interest rate might be linked too if it isn’t linked to employment?

Comment by boonton

You have an excluded-middle problem there.

Comment by Sonic Charmer

The gas pedal analogy covers the middle nicely. How much you’re pressing on the gas matters for how fast the car goes, but it’s not the only thing that matters.

Comment by boonton

If/when X is not the only thing that affects Y it’s pretty stupid to say you’re ‘going to do X until Y happens’ no?

Comment by Sonic Charmer

Is it? If you want your car to go faster wouldn’t you press down on the gas? If you pressed harder and noticed no increase in speed or power then you’d ask whether other variables were having more impact.

Comment by boonton

“If you want your car to go faster wouldn’t you press down on the gas?”

Not if the engine’s broken. Did you not understand the premise of the post?

Comment by Sonic Charmer

One way you’d know if something was wrong with the engine is to press on the gas and notice the car doesn’t move faster.

Comment by boonton

Where are interest rates right now? Would you characterize them as high or low?

Comment by Sonic Charmer

I would say high, but keep in mind interest rates are a relative thing. If you were going up a steep hill what would be ‘normal’ as far as the gas pedal is concerned might be pressing it 3/4 to the floor. Going downhill normal may mean even slighlty touching the gas is pressing too hard.

Comment by boonton

Interest rates are high. Got it, thanks.

Comment by Sonic Charmer

And you would say? High? Low? Can you even tell? It seems to me before you can say interest rates are too high or low you have to be working with a model that assumes interest rates impacts some other important variable in some way and will cause that variable to move to either a desired or undesirable state. If you assert that interest rates have no clear relationship to anything else, it’s really not possible to mount an argument for them being ‘too’ anything.

Comment by boonton




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s



Follow

Get every new post delivered to your Inbox.

Join 465 other followers

%d bloggers like this: