Should the Fed look at the nominal GDP?

There I was taking my morning walk listening to the newest episode of The Weeds which talked about inflation.  It was a good episode talking about how it can be difficult to measure actual inflation and how it can be a difficult gauge for the Federal Reserve to use since it takes a while to collect information meaning that the picture of the world it provides is always out of date.  And distorted too since information from different sectors arrive at different times.

And the calculations are hard too.  If Verizon goes and starts offering unlimited data plans does this mean that data is cheaper, for less inflation?  Or are limited and unlimited plans not actually comparable that way?  I wrote a little on this way back in 2012 but this stuff is hard.

This was all a big problem for all of us back in 2008 when the Lehman Brothers was collapsing.  According to the Fed's most recent data high commodity prices were continuing to cause higher inflation than they desired over the last few months.  To quote:

On the inflation front, overall consumer prices rose rapidly for a third straight month in July but then edged down in August, because of a sharp drop in energy prices. Core consumer price inflation remained elevated in July and eased somewhat in August.

So while they pumped money into the banking system they didn't lower overall interest rates and took the unprecedented step of paying banks interest on their excess reserves so they wouldn't lend it out.  However, what was actually happening that month was that inflation had cratered and the Fed's information was out of date.

The Fed could see that growth was slowing but deflation is another symptom of a depression and as far as the Fed knew that hand't happened.  They had nominal, that is not adjusted for inflation, data about things like sales and wages but controlling the nominal wasn't their job, controlling inflation was.

If you want to know why the Fed uses inflation I think Noah Smith does a good job in this post from just today of explaining why inflation makes sense as a very neat and important theoretical concept.  When people started doing central banking of course it made sense to work in terms of inflation and real GDP, which are theoretically quite distinct.   But in terms of control of the actual banking system in real time perhaps it would be better to target nominal GDP - inflation multiplied by GDP in theory but in practice something we can measure quickly.  Also something that exists in the real world in a concrete sense unlike inflation which can be approximated in a variety of ways all of which mostly but not quite agree with each other.

Also, one reason to care about inflation or deflation is that it grows or shrinks the size of people's obligations compared to their real income streams.  If someone has to pay a certain dollar amount of rent every month but suddenly that nominal value is worth more in real terms they might not be able to come up with that money any more and get kicked out.  If we're talking about the salary of an employee the same principle applies and they may be fired.  

But if the real economy is growing very quickly then most of the people affected will be able to come up with the extra real value they need to pay and despite the value of rent or wages being increased out from under them they'll be able to manage.  And in the opposite situation where a lot of real wealth goes away due to an oil crisis or something it's better that the misery be spread out by a burst of inflation reducing everyone's wages than that the situation be compounded by some people losing their jobs and the loss in real wealth being compounded.

There are other reasons to want stable inflation as well besides the destructive effects of deflation.  Economists debate them but as far as I can tell from the sidelines its better to have a stable NGGP growth than stable inflation, though really you'd prefer both.  And I really wish that when talking about Fed responsiveness this played a larger role in the conversation.