Posts filed under “Data Analysis”
As we have previously discussed, inflation remains contained to the rest of the world ex-USA. Globalization be damned, there may be rapid price rises in most of the world, but there is no inflation in the U.S., thanks to a combination of hedonic adjustments and the absurd focus on the core rate.
Whenever I hear the phrase "excluding volatile food and energy" I just laugh. Can a pricing group be considered volatile if it merely goes up each month in an orderly fashion — for years and years?
That’s not volatility, thats a trend.
One way to actually measure how absurd the US core inflation measure is to look at what has happened to the spread between headline CPI and Core CPI. If Core CPI is understating inflation, than the spread should be widening. If it is accurate, the overall ratio between the two should be relatively steady.
What does the data show? The spread has increased substantially since the US adopted an ultra low rate/easy money policy under Greenspan (now affiliated with bond giant PIMCO). Since the easy money policy of the 1990s, and the rate slashing of the 2000s, it is no coincidence that the spread between the headline number and the core has grown dramatically.
If you want to trace this widening spread back to its origins, it coincides with implementation of Boskin Commission changes in CPI. (About as intellectually dishonest analysis of Inflation as has ever been penned — its goal was to reduce Social Security payments and avoid bankrupting the US Treasury — not measure inflation accurately). Since then, the spread between the core and headline data have only grown further apart.
This simply reflects the government’s BLS inflation data diverging from reality.
Core CPI flatlined over the past 8 years because that is how it was constructed — to not show inflation. However, the absurdity of the adjustments in inflation measures is revealed in the widening spread between Core and Headline:
Notice how tightly the two data series coincide (top chart) in the latter half of the 1980s and all of the 1990s? See how that starts diverging in 2001?
Bill King points out: "Targeted inflation may be the headline CPI, or a
derived core inflation measure. In either case central banks should be
aware of the sources of error and bias in their country’s CPI"
If you really want to econo-geek out on this stuff, there are several good resources for taking apart how this data is constructed:
Core Inflation Measures and Statistical Issues in Choosing Among Them
IMF Working Paper April 2006
PCE Inflation and Core Inflation
Julie K. Smith
July 6, 2006
But the bottom line is that the US measures of Inflation, especially the core levels, are constructed to diverge from reality, and understate price rises. That is seen in the headline CPI/Core spread.
The Payroll numbers are out, and they are not particularly pretty:
88,000 new jobs were created in April, according to BLS. This is the weakest job gain since November ’04. Cumulative revisions for prior months were to the downside by 26,000.
As expected, losses were in Manufacturing (19k), Retail (26k) and Construction (11k). The weakness in Construction has been very uted, implying that the full impact of the housing slow down has yet to be fully realized.
Biggest gains were had in Services (116k), Education and Health (53k), Gov’t (25k) Professional (24k) and Leisure/Hospitality (22k).
Temporary help jobs fell for a 3rd month (January was flat) making 4 consecutive months of no gains. Temp help tends to lead employment gains, and this weakness can be read as a future forecastor of employment.
We don’t pay close attention to the Household survey, (the self reported number is very volatile) but the drop of -468k was an eyebrow raiser.
To put this into some context, of those 317k new jobs hypothesized by BLS, 49k of those supposed jobs are in construction. Now what are the odds of that?
While Wall Street celebrates the upcoming recession, let me remind you that this economy requires about 150k new monthly jobs to merely keeep up with population growth.