CNBCrazy

One of the first things that I noticed when I started working in the financial industry was that every trading floor has TVs on the walls and everyone has CNBC on - but no one admits to taking anything they say seriously.  It is like financial info-tainment.  Jim Cramer might be entertaining to watch but his stock picks don't actually do that well (see here and here and here) although you would expect that they might just on the basis of self-fulfilling prophecy.  Larry Kudlow still dishes out messianic supply side advice - he may be the only remaining believer in the Laffer Curve.  But like Cramer his economic predictions are not so accurate (see here here here).  And then there is Rick Santelli.  Santelli is a local institution in Chicago reporting from the floor of the CBOT on every bounce in the futures markets and dispensing his views on economic policy.  He is also (in)famous for launching the Tea Party movement with this 2009 rant .

Like Kudlow & Cramer, Santelli is entertaining but his economic policy advice is also suspect...Last week while walking past a wall mounted TV I heard this Santelli gem called Inflation Vacation Or Inflation Gestation.   His basic contention can be summarized by this passage

"...If you look at the Wiemar Republic in their hyperinflation  It was in the early 20s.  It didn't happen overnight.  I have used the analogy it's a lot like soybeans.  Ya plant em.  You wait. Conditions take some time.  You need some sun.  You need some water.  But ultimately things start to grow.  And are we in that phase or not?"

So from this passage I take away that Santelli believes that 

• there is a gestation period between initial monetary stimulus and latter (hyper)inflation
• this was what happened during the German Weimar hyperinflation of the 1920s
• the US could be facing a similar situation.  At least there is enough of a possibility to discuss it.

The first point stems from the traditional monetary theory of Hume / Mill / Fisher / Friedman & Schwartz.  If you push money into peoples hands it initially results in increased demand for final goods.  Increased demand for final goods increases the equilibrium quantity of final goods while slowly bidding up the price of final goods.  As the quantity of final goods increases so does the demand for input factors including labor.  Eventually the price of input factors are also bid up.  And you have inflation. 

The above is the general case but not always the case.  Sometimes for whatever reason people become afraid of the future and the demand for safe assets (such as currency) increases relative to the demand for final goods.  That means people value holding currency more and value buying goods less. If prices were fully flexible we would see the relative price of safe assets (including currency) increase and the relative price of goods fall (ie deflation).  However nominal prices of goods tend to be sticky and nominal wages tend to be very sticky.  If the relative prices of goods versus safe assets do not equilibrate then we end up with excess demand for safe assets (including money) and excess supply of goods (ie recession).  In this case increasing the money supply (somewhat) does not lead to inflation it just sets the relative values of currency and goods back to the levels where supply for goods is fully employed.  In this case we can see a build up of money supply which does not result in inflation.

How about Santelli's second point.  Did the Wiemar hyperinflation start with monetary stimulus which gestated over time before blooming into full fledged hyperinflation - as Santelli contends?  I would argue that it did not.  Rather the initial moderate monetary expansion led to moderate inflation.  The latter hyperinflation was due to a later geometric expansion of the money supply.

Data on Wiemar Germany is from Thomas Sargent The Ends of Four Big Inflations.  Below is plotted the

• monthly growth rate of notes in circulation (M)
• monthly growth rate of Marks Per Dollar
• monthly growth rate of Wholesale Price Index (WPI)

These figures are not annualized so when one of the series hits 100% it means that in that month the value of the series doubled.  For example in November 1922  Marks Per Dollar broke through 100% meaning the Mark lost just over half of its value in that month.

If Santelli is correct we expect to see high rates of money growth then a period of gestation and then a burst of inflation.  But that is not what the data shows.  Rather when the money supply was growing (relatively) slowly so was inflation.  And when we observe extreme inflation we observe it in parallel with extreme rates of money growth.  The other point to notice is how huge the rates of money growth in fact were.  By late 1922 the Wiemar Republic was doubling their money stock every two months.  By late 1923 they were increasing their money stock by a factor of five each month.  Just to be clear what that means 1, 5, 25, 125, 625....that is hyperinflation




To put the scale of note issuance into perspective.  The below chart shows for every 1 unit of currency in circulation as of January 30, 1921 how many units were in circulation as of other dates.

Date	Ratio
Jan-21	1.0
Jun-21	1.1
Dec-21	1.7
Jun-22	2.5
Dec-22	19.0
Jun-23	256.4
Dec-23	7,363,633,658.3
Jun-24	16,274,039,587.0
Dec-24	28,793,230,909.3

So as to Santelli's third point that the US may or may not be in for significant inflation as a result of high rates of monetary growth.  The first point to notice is that the scale of money issue is just entirely different.  Below is chart of growth rates of the Monetary Base (MB), M1, and M2 with the graph set to the same scale as the Wiemar graph.  It looks pretty similar doesn't it?



Now let's add prices CPI and 1/Dollar Index so we can observe the inflation rate



Just like the Wiemar Germany right?  Ok obviously its a ridiculous comparison.  But I was not the one who made the comparison - that was Rick Santelli.  I am just pointing out how bad a comparison it is.  Before we leave this topic - for completeness lets look at a pair of graphs in which each series is divided by its value as of December 2004 so we can see the aggregate amounts of growth over the past eight years. Today the Monetary Base (MB) is 3.5 times what it was in December 2004 as the Fed expanded its balance sheet.  However M1 and M2 have only grown by about 50% over 8 years and prices are only up by about 20% over 8 years.



 

So why do I care?  Because there is very little attempt at serious economic analysis on TV.  And the one place that does purport to run serious economic analysis all day long (CNBC) runs ridiculous economic analysis - yet it passes for serious economic analysis because it is just TV.  Hey they are just entertainers right?  Wrong. CNBC has enormous ability to influence the discussion relative to other platforms.  Put it this way; if all day you hear that cutting taxes for the wealthiest 1% will lead to an economic boom (Kudlow) and no one disputes it then eventually it might sound like a rational view.  Or if you keep hearing about how the US "could" face huge inflation (Santelli) then that idea becomes part of the policy discussion as well - even if it is ridiculous.  Don't believe me?  See here

When he's right he's right

Ark. GOP: Candidates' statements 'highly offensive'

"Fuqua, who served in the Arkansas House from 1996 to 1998, wrote there is "no solution to the Muslim problem short of expelling all followers of the religion from the United States," in his 2012 book, titled "God's Law."  Fuqua said Saturday that he hadn't realized he'd become a target within his own party, which he said surprised him.  "I think my views are fairly well-accepted by most people," Fuqua said before hanging up, "

I can see why he was surprised.  His views are fairly well accepted by "most people"...if by "most people" you mean pundits on right wing TV, radio, websites, and FOX NEWS.  See here here here here here here here here here here or hundreds of other sites.

Thinking Fast and Slow and Large

How large is large? 

Most people intuitively understand what "half" and "two times" look like.  For example if I display a medium pizza and say that I will give you half of the pizza then most people would be able to immediately picture how much I am giving you.  If I instead display a small pizza and say that the medium pizza is twice as large then most people would be able to immediately picture how large the medium pizza is.

Likewise I think most people understand one quarter intuitively as well.  If I display a large pizza and say that I will give you quarter of the pizza then most people would be able to immediately picture how much I am giving you. They would think of cutting the pizza in half and then in half again.  However if instead I display a small pizza and say that the large pizza is four times as large - well given a bit of time - I believe that most people would be able to figure out how large the large pizza is - but the image might not be quite as immediate.

But what would happen if I show you a small pizza and say that the super-deluxe pan pizza is twenty  five times as large.  I would be surprised if many people (outside the pizza making world) could immediately picture how large that is.  They would know that the super-deluxe pan pizza is much larger than the small pizza but how much larger?  I might go back to how I learned to multiply.  Twenty five is five times five.  So if I make five rows of five small pizzas that would approximately equate to the size of a super-deluxe pan pizza.  But obviously this takes a bit of thinking.

The above examples illustrate Daniel Kahneman's System 1 & System 2 processes.  Kahneman argues that there are two modes of thinking.  System 1 is fast and automatic, based on relating new information to old learned situations.  System 2 is slow and requires analytic thought.  One half, two times, one quarter are System 1 ....larger numbers require System 2.   However I contend that when most people hear someone quote a large number X they do not immediately revert to System 2 and try to figure out what the true scale of that number is - rather they stay with System 1 and just accept that number X is large and relate large to what they think of as large.

For example most children know that a million is a large number, and billion is a large number, and a trillion is an even larger number.  And we would hope that most high school students can figure out that one billion is 1000 times one million.  And that one trillion is 1000 times as large as one billion - as well as 1,000,000 times as large a as one million.  But when brought up in casual conversation I believe that most people will revert to System 1 and just know that one trillion is a big number.

Where am I going with this?  I started thinking about this after hearing the Gov. Romney's comment about cutting funds for PBS and the Corporation for Public Broadcasting (CPB) so we won't need to borrow money from China to fund our deficit.  As many pundits have pointed out afterwards that cutting funding for CPB would only be a drop in the bucket toward closing our budget deficit much less get anywhere near to funding a 20% across the board tax cut.  But our reversion to System 1 thinking leads us to gloss over the magnitudes of the numbers and instead just think large.

The actual numbers.  In 2012 Congress appropriated 445 million USD to CPB.  The 2012 federal budget had predicted receipts of 2.469 trillion USD and predicted expenditures of 3.796 trillion USD leading to a predicted deficit of 1.327 trillion USD.  Obviously 1.327 trillion is a larger number than 445 million - but ....

Imagine instead that Congress was spending 100 billion USD on CPB appropriations.  Then cutting CPB funding and twelve other programs like it could close the budget deficit.  Obviously 100 billion is a small number relative to 1.327 trillion, but stated in this way it is not a ridiculous suggestion.  But Congress is actually only spending 445 million USD on CPB.  How does 445 million that compare to 1.327 trillion?  Here is where our System 1 thinking starts foundering.  1.327 trillion, 100 billion, 445 million - these are all big numbers.  We know that 1.327 is the biggest but how much bigger is it?

Sure we can all do the math.  1,327,000,000,000 / 445,000,000 = 2,982.  So if we defund CPB and find another 2,981 budget items of the same size to defund then we could close the budget deficit. If you also want to fund a 20% tax cut that would require finding an additional (20% * 2.469 trillion / 445 million = ) 1,078 CPB sized budget items or 4,059 budget items in total to defund.  Once again 2,982  and 1,078 and  4,059 are large numbers and our System 1 again starts foundering.

Let's help our System 1 thinking put this in context...In order to close the deficit and fund a 20% across the board tax cut we would need to cut federal government spending by ((1.327 trillion USD +20%*2.469 trillion USD ) / 314 million persons = ) 5,755 USD for every person in US.   Cutting funding to CPB would account for (445 million USD / 314 million persons = ) 1.41 USD of those spending cuts.   So we need to cut 5,755 USD for every person in US and Gov. Romney suggested $1.41 in cuts per person.

To be fair...I assumed a 20% across the board revenue cut.  If you want to instead assume it is a  20% across the board income tax cut then we would need to cut federal government spending by ((1.327 trillion USD + 20%*1.165 trillion USD ) / 314 million persons = ) 4,968 USD per person.  Or if you want to assume that we just cover the deficit and forget the tax cuts then we would need to cut federal government spending by just (1.327 trillion USD / 314 million persons = ) 4,226 USD per person....of which Gov. Romney has suggested $1.41 in cuts per person.  So obviously that makes it a lot more manageable.

Ezra Klein quote

@ezraklein   "If Congress was your doctor, they'd give you too few antibiotics and, when that failed, give up on the idea of antibiotics altogether."

Good analogy.  However I would say - If Congress was your doctor they'd give you too few antibiotics and, when that failed, blame the antibiotics for your sickness.