We think the charts below illustrate the situation pretty plainly, but nevertheless we'll explain. The first chart shows rolling initial jobless claims from September through August for 2009/10, 2010/11 and 2011/12. On the left half of the chart we show the September through February period and on the right half the March through August period. Looking back over the past three years, a very consistent pattern has emerged. Seasonally adjusted jobless claims see ALL of their improvement from September through February and then worsen from March through August. We've quantified this by showing the trend line slope equations for each series. In 2009/10 September through August saw average monthly improvement of 4.7k fewer initial jobless claims. This moderated to 2.8k in 2010/11 and to 2.5k in 2011/12, but the trend was highly consistent in each of the past three years. Conversely, the March through August period saw claims rise 0.7k/month in 2009/10, 0.8k/mo in 2010/11 and 0.3k/mo in 2011/12. We show the strength of the relationship between claims and the market each week in this note, so we think it's worth noting that the employment data has just entered a period of tailwind, after having been facing a headwind for the last six months.
In the second chart, we show the same dynamic, but for Nonfarm payrolls (private). The trend is the same. Tomorrow's report on August NFP will bring to a conclusion the headwind data, and usher in the next six months of tailwind data. In case anyone's curious why this is the case, it's principally a byproduct of errors in the government's seasonal adjustment factors stemming from the dramatic deterioration in the economy immediately following Lehman Brother's collapse. This distortion will be with us for two more years, becoming slightly less significant each year and then will disappear entirely three years from now (government models use a 5-year lookback).
SNAP: Food Stamp Participation Growth, While Trending Lower, Ticks Up In June (the Most Recent Month)
The number of people receiving government assistance to buy food in the U.S. reached 46,670,373 people in June, 2012, or 14.8% of the population. In terms of households, there are 22.4 million receiving assistance, or roughly 19.3% of U.S. households. In June alone, 173,612 additional people joined the program vs. May 2012. That's a lot. The number of people on food stamps has risen 75% since June of 2007. Said differently, in the last five years, an additional 20 million people have begun receiving food stamps. For reference, that's just slightly more than the entire population of New York State (19.5 million).
If there's any silver lining here, it's that rate of growth (YoY) among food stamps participants has been slowing. We show that in the black line in the chart below. Although it's worth pointing out that the most recent data was actually a sequential acceleration on a YoY basis.
Claims: The Data
Initial claims fell 9k to 365k last week. After incorporating a 3k upward revision to the prior week's data however, claims fell 12k. Rolling claims rose 0.25k WoW to 371.25k. On a non-seasonally adjusted basis, claims fell 4k.
We also saw some modest improvement in the YoY change in the rolling NSA series, which came in at -8.2% versus -7.7% in the prior week. As a reminder, this is the series we use to gauge actual underlying improvement in the jobs picture. It's still improving.
The 2-10 spread was flat WoW with both 2 and 10yr treasuries falling 4bps. The 2-10 spread is at 1.37%, which brings its QTD level down 18 basis points vs 2Q12. Net interest margin comparisons will still be down linked-quarter.
Financial Subsector Performance
The table below shows the stock performance of each Financial subsector over multiple durations.
Joshua Steiner, CFA
Having trouble viewing the charts in this email? Please click the link at the bottom of the note to view in your browser.