Friday, November 20, 2009
Weekly Indicators
- by New Deal democrat
Both the monthly data and the high frequency weekly data were mixed, but with a generally bullish bias.
The monthly data included a bad housing report, mediocre increases in industrial production and capacity utilization, a decent but not outstanding LEI, an mixed Empire State index, and a strongly positive Philly manufacturing index.
Meanwhile, as to the weekly data:
The BLS reported new jobless claims remained the same as last week on an SA basis at 505,000. On a 4 week average basis, this series continues to decline.
Edmunds automotive said that on a preliminary basis,
Rail traffic continued to improve -- not just in a relative sense to last year, but in an absolute sense as well. This is bullish. Last week commenter Olephart correctly observed that I had posted a graph which didn't really make my point. Cyclical traffic frequently starts to decline in October, but sometimes not until now. As of mid-November, it is not just flat, but actually turned up last week.
The Daily Treasury Statement as of November 18 showed $85.8 Million in withholding taxes paid month-to-date, compared with $86.7 Million last year. For the last week, this year's totals have been generally equal to or slightly ahead of last year's. Since this data series tends to lag the end of recessions by one quarter, this may be the month it finally turns.
Oil retreated from its flirtation with $80, as of midday Friday trading at $76.50.
Both the monthly data and the high frequency weekly data were mixed, but with a generally bullish bias.
The monthly data included a bad housing report, mediocre increases in industrial production and capacity utilization, a decent but not outstanding LEI, an mixed Empire State index, and a strongly positive Philly manufacturing index.
Meanwhile, as to the weekly data:
The BLS reported new jobless claims remained the same as last week on an SA basis at 505,000. On a 4 week average basis, this series continues to decline.
Edmunds automotive said that on a preliminary basis,
U.S. new-vehicle sales are expected to rise in November from a year earlier, adjusted for two fewer sales days this year .... [C]ar sales are seen rising 3.8% in the U.S. from a year ago. Without the adjustment for there being 23 selling days this year and 25 last year, the auto Web site's estimate is 4.5% below last year's sales and down 15% from October...."The ICSC reported that same store sales fell -0.1% last week, the second minimal decline in a row. On a YoY basis, sales are up 2.4%, and
The overall annualized sales rate in November is projected to be 10.3 million, down slightly from October's 10.43 million rate.
ICSC Research expects same-store sales for November could beShopperTrak an increase of 0.7% in mall retail sales compared with a year ago, and also up 7.5% from the previous week.
as strong as up 5 to 8 percent.
Rail traffic continued to improve -- not just in a relative sense to last year, but in an absolute sense as well. This is bullish. Last week commenter Olephart correctly observed that I had posted a graph which didn't really make my point. Cyclical traffic frequently starts to decline in October, but sometimes not until now. As of mid-November, it is not just flat, but actually turned up last week.
The Daily Treasury Statement as of November 18 showed $85.8 Million in withholding taxes paid month-to-date, compared with $86.7 Million last year. For the last week, this year's totals have been generally equal to or slightly ahead of last year's. Since this data series tends to lag the end of recessions by one quarter, this may be the month it finally turns.
Oil retreated from its flirtation with $80, as of midday Friday trading at $76.50.
When will the economy add jobs? November update
In September I took a long look at Leading Indicators for job growth, and concluded that they have historically turned in the following order:
(1) Real retail sales bottom and turn.
(2) Initial Jobless claims turn.
(3) The ISM manufacturing index turns above 50, i.e., signals actual growth.
(4) Industrial Production turns.
(5) ISM manufacturing index is above 53, ISM employment is at -5 or above, initial jobless claims are at least a sustained 16%-20% off peak, and both Industrial Production and Real retail sales have advanced at a rate of 2.5% or more year-over-year from the bottom.
In September, (1) through (4) had apparently already happened. Number (5) is all about the strength of the turns. Based on the strength of the Leading Economic Indicators, I concluded with a prediction that payrolls would most likely hit bottom and turn around in November or December, +/- 1 month.
I. Initial Jobless Claims
In the first installment of the series, I looked at initial jobless claims. I noted that (1) in those recessions and recoveries where jobless claims fell steeply, peak unemployment occurred within 2 months of the point where jobless claims fell 12% from the peak; but (2) in those recessions and recoveries where jobless claims fell slowly, peak unemployment occurred not at the 12% mark, but only when new jobless claims were more than 16% less than peak claims, and stayed more than 16% off for at least 3 months thereafter.
In September, initial claims were still less than 16% off their highs. That has changed dramatically. Jobless claims have fallen substantially, and as of today the 4 week moving average stands at 514,000, more than 21% off the peak of 658,750, over 7 months ago. Here is a graph comparing this recession with the two previous "jobless recoveries" and the V shaped deep recession of 1982 in terms of the first 7 months of post-peak initial jobless claims:
In the last month, our recession/recovery has become more comparable to the 1982-3 recovery than previously. As you can see, in the two "jobless recoveries," new jobless claims failed to penetrate the 20% off level except for a brief instance in summer 2002 (coinciding with a brief positive jobs number). Should the present trend continue this indicator predicts actual job growth this month or next.
[Note: In this regard I am at odds with such esteemed luminaries as Berkeley Prof. Brad DeLong and Calculated Risk, who believe that jobless claims must fall all the way to 400,000 before jobs will actually be added, based on the last two "jobless recoveries." I disagree, based on the recoveries from the deep 1970s and 1980s recessions. All I can say is that the data is already not behaving in accord with their analysis. Nevertheless, I would be glad to be proven wrong provided it occurs quickly!].
II. ISM Manufacturing
I next looked at the ISM Manufacturing Index, and concluded that the 53 level (50 is the dividing line between expansion and contraction) is the point where jobs began to be added in the very strong recovery after 1982, as well as during the week recoveries of 1992 and 2002. Further, a reading over 54 on the index has always coincided with actual job growth.
Additionally, whenever the hiring vs. firing sub-index was -5 or higher (i.e., no more than 5% more employers plan to fire than hire) and rising, where other evidence indicates a recession is ending, that has always indicated net employment growth was imminent, at least on a temproary basis; and also, whenever current staffing intentions were 65+. and hiring plans were 15+, that has always coincided with positive jobs numbers in the BLS survey, including during and after the "jobless recoveries" of 1992 and 2002.
This record has now been broken. The ISM surged in October to 55.7, and the employment index is at +53.1:
Despite this indication of actual net hiring in the manufacturing sector, the October BLS jobs data claimed 61,000 job losses in manufacturing -- the most in four months! -- and 190,000 total job losses. Everybody knows that the BLS significantly revises their initial numbers, and I suspect the report of job losses in manufacturing is going to be substantially revised.
III. Industrial Production
In part 3 of the series I noted that industrial production tends to peak a median +2 months before payrolls, and to trough at the end of recessions a median +1 month before payrolls. Of the 10 troughs since World War 2, in 8 of them industrial production troughed within 2 months of the payrolls number. Further, the only times that industrial production has led employment growth by a relatively long period of time, it has also shown weak growth -- less than 5% a year. In more typical V shaped job recoveries, it has grown at a rate of 10% or more a year.
In October, industrial production continued to move positively. It has grown 2.9% in the last 4 months, or in other words indeed growing at a rate of more than 10% a year. Here is a graph comparing that growth with growth in the first 4 months of the V shaped recovery of 1983 vs. the "jobless recoveries" of 1992 and 2002-03:
This, like the trend of initial jobless claims, is at levels consistent with actual job growth this month or next.
Three Negative or Flat Developments
I. ISM Non-Manufacturing
One unique factor of this "Great Recession" is how strongly services jobs have been hit, much moreso relative to manufacturing than in any post-WW2 downturn, as shown on this graph:
In the last few months, however, the percentage of job losses in services compared with manufacturing has, if anything, increased. Indeed, while the ISM manufacturing report has been strong, the ISM non-manufacturing report has been consistent with the tepid growth in real retail sales. I did not include this in my original analysis, in part because the data does not have a long record at all -- less than 15 years. That being said, here is a graph showing that in the 2002-03 expansion, normed at 100 for an ISM Nonmanufacturing reading of 60, showing that job growth in services did not occur until that point in 2003:
As of October, this index has stalled at just above 50, and the employment subindex fell to 41.1 (meaning substantial layoffs). Based on this sparse record, one would have to say that the prospects of job growth in any services sector in the next few months look remote, at best.
II. GDP Growth
I have previously shown in a discussion of "Okun's law" that over the last decade or more it has taken 2% YoY growth in GDP for even 1 net job to be added. Here's an updated graph demonstrating that relationship (with YoY GDP growth shown in blue, payrolls in red):
This suggests that, while 3rd quarter GDP was initially reported up at an annual rate of 3.5%, which surprised most people (it may be revised downward somewhat due to increased imports), at least two more quarters of such growth must take place for YoY GDP to reach +2%. The Leading Economic Indicators at this point do indicate positive GDP growth in the 4th quarter, carrying over at least into part of the first quarter of 2010, but after that, the jury is out -- especially with $80 Oil.
III. Real Retail Sales
A major revision of recent data by the Census Bureau calls this element of my analysis into question. I described real retail sales as the Holy Grail of Leading Indicators for job growth, noting that his consistently turned at both tops and bottoms, an average of 3-5 months before job growth or losses turned. When real retail sales stay flat, they generate a lot of noise, and a longer period between the turn in sales and payrolls. Strong turns in sales generate reliable subsequent moves in payrolls in subsequent months. In general, with regard to recoveries, an increase of about +2.5% a year is necessary to reliably generate a subsequent move in real retail sales.
On a three month smoothed average basis, real retail sales bottomed in April of this year. Despite October's growth of 1.4%, nasty revisions downward of August and Septmber totaling -1.3% mean that real retail sales are up only about 0.9% off that low, for an annual rate of 1.7%. In short, the revisions keep real retail sales within the range of "noise." The past few months may be the beginning of a solid upward trend, or continued revisions may mean the series stays flat with no growth whatsoever.
Here is a graph comparing the trend in Real Retail Sales since its smoothed bottom in April with the first 7 months of growth during the expansions of 1983, and 1992, and the first eight months of growth in 2001 (September and October 2001 show extreme volativity for obvious reasons so I also included November):
This is simply not consistent with job growth yet.
Conclusion: BLS revisions and John Maynard Keynes
John Maynard Keynes famously justified a change in opinion by saying, "When the facts change, I change my opinion. What do you do?" Well, a critical fact in my analysis has changed.
Three indicators -- ISM manufacturing, Initial Jobless Claims, and Industrial Production -- are now at levels typically associated with actual job growth in previous recoveries. All three of these series exceed their growth in the two "jobless recoveries" of 1992-3 and 2002-3.
The "Holy Grail," however, Real retail sales, is only trending sideways or slightly higher albeit for unique reasons (cash for clunkers). The substantial downward revisions in real retail sales in August and September call into question when the overall jobs number will turn. While industrial production is indeed having a "V" shaped recovery so far, growing at a rate similar to 1983; by contrast, Real retail sales' anemic growth is similar to -- and even weaker than -- its pattern in the last "jobless recovery," below the 3% annual growth typically associated with job growth. The services sector is stagnated with continuing substantial job losses, and GDP growth, while impressive, isn't impressive enough to think that jobs will actually turn positive by year's end.
At the end of the day, the Leading Economic Indicators either work or they don't. If they work, then the coincident indictors of real income, real retail sales, and employment must start to trend upward shortly. What may be developing is a bifurcated economic expansion, in which jobs in manufacturing are added back quickly (hence, not a "jobless recovery") but services continue to shed jobs as consumers continue to repair their balance sheets and shop discerningly for bargains (thus a "job loss recovery"). This tends to move the turning point for jobs into January-March (January at least being consistent with my original prediction).
Finally, as I said in my concluding installment in September and have reiterated since: let me be the first to acknowledge that this is not a scientific truth or certainty, but a best estimate based on a logical review of existing data with a long history that accommodates both traditional and "jobless" recoveries.
(1) Real retail sales bottom and turn.
(2) Initial Jobless claims turn.
(3) The ISM manufacturing index turns above 50, i.e., signals actual growth.
(4) Industrial Production turns.
(5) ISM manufacturing index is above 53, ISM employment is at -5 or above, initial jobless claims are at least a sustained 16%-20% off peak, and both Industrial Production and Real retail sales have advanced at a rate of 2.5% or more year-over-year from the bottom.
In September, (1) through (4) had apparently already happened. Number (5) is all about the strength of the turns. Based on the strength of the Leading Economic Indicators, I concluded with a prediction that payrolls would most likely hit bottom and turn around in November or December, +/- 1 month.
Two months later, how is that playing out? A major revision in data calls into question one large element of the analysis. There are three very positive developments, and three negative or flat. Details below.
The Positive DevelopmentsI. Initial Jobless Claims
In the first installment of the series, I looked at initial jobless claims. I noted that (1) in those recessions and recoveries where jobless claims fell steeply, peak unemployment occurred within 2 months of the point where jobless claims fell 12% from the peak; but (2) in those recessions and recoveries where jobless claims fell slowly, peak unemployment occurred not at the 12% mark, but only when new jobless claims were more than 16% less than peak claims, and stayed more than 16% off for at least 3 months thereafter.
In September, initial claims were still less than 16% off their highs. That has changed dramatically. Jobless claims have fallen substantially, and as of today the 4 week moving average stands at 514,000, more than 21% off the peak of 658,750, over 7 months ago. Here is a graph comparing this recession with the two previous "jobless recoveries" and the V shaped deep recession of 1982 in terms of the first 7 months of post-peak initial jobless claims:
In the last month, our recession/recovery has become more comparable to the 1982-3 recovery than previously. As you can see, in the two "jobless recoveries," new jobless claims failed to penetrate the 20% off level except for a brief instance in summer 2002 (coinciding with a brief positive jobs number). Should the present trend continue this indicator predicts actual job growth this month or next.
[Note: In this regard I am at odds with such esteemed luminaries as Berkeley Prof. Brad DeLong and Calculated Risk, who believe that jobless claims must fall all the way to 400,000 before jobs will actually be added, based on the last two "jobless recoveries." I disagree, based on the recoveries from the deep 1970s and 1980s recessions. All I can say is that the data is already not behaving in accord with their analysis. Nevertheless, I would be glad to be proven wrong provided it occurs quickly!].
II. ISM Manufacturing
I next looked at the ISM Manufacturing Index, and concluded that the 53 level (50 is the dividing line between expansion and contraction) is the point where jobs began to be added in the very strong recovery after 1982, as well as during the week recoveries of 1992 and 2002. Further, a reading over 54 on the index has always coincided with actual job growth.
Additionally, whenever the hiring vs. firing sub-index was -5 or higher (i.e., no more than 5% more employers plan to fire than hire) and rising, where other evidence indicates a recession is ending, that has always indicated net employment growth was imminent, at least on a temproary basis; and also, whenever current staffing intentions were 65+. and hiring plans were 15+, that has always coincided with positive jobs numbers in the BLS survey, including during and after the "jobless recoveries" of 1992 and 2002.
This record has now been broken. The ISM surged in October to 55.7, and the employment index is at +53.1:
Despite this indication of actual net hiring in the manufacturing sector, the October BLS jobs data claimed 61,000 job losses in manufacturing -- the most in four months! -- and 190,000 total job losses. Everybody knows that the BLS significantly revises their initial numbers, and I suspect the report of job losses in manufacturing is going to be substantially revised.
III. Industrial Production
In part 3 of the series I noted that industrial production tends to peak a median +2 months before payrolls, and to trough at the end of recessions a median +1 month before payrolls. Of the 10 troughs since World War 2, in 8 of them industrial production troughed within 2 months of the payrolls number. Further, the only times that industrial production has led employment growth by a relatively long period of time, it has also shown weak growth -- less than 5% a year. In more typical V shaped job recoveries, it has grown at a rate of 10% or more a year.
In October, industrial production continued to move positively. It has grown 2.9% in the last 4 months, or in other words indeed growing at a rate of more than 10% a year. Here is a graph comparing that growth with growth in the first 4 months of the V shaped recovery of 1983 vs. the "jobless recoveries" of 1992 and 2002-03:
This, like the trend of initial jobless claims, is at levels consistent with actual job growth this month or next.
Three Negative or Flat Developments
I. ISM Non-Manufacturing
One unique factor of this "Great Recession" is how strongly services jobs have been hit, much moreso relative to manufacturing than in any post-WW2 downturn, as shown on this graph:
In the last few months, however, the percentage of job losses in services compared with manufacturing has, if anything, increased. Indeed, while the ISM manufacturing report has been strong, the ISM non-manufacturing report has been consistent with the tepid growth in real retail sales. I did not include this in my original analysis, in part because the data does not have a long record at all -- less than 15 years. That being said, here is a graph showing that in the 2002-03 expansion, normed at 100 for an ISM Nonmanufacturing reading of 60, showing that job growth in services did not occur until that point in 2003:
As of October, this index has stalled at just above 50, and the employment subindex fell to 41.1 (meaning substantial layoffs). Based on this sparse record, one would have to say that the prospects of job growth in any services sector in the next few months look remote, at best.
II. GDP Growth
I have previously shown in a discussion of "Okun's law" that over the last decade or more it has taken 2% YoY growth in GDP for even 1 net job to be added. Here's an updated graph demonstrating that relationship (with YoY GDP growth shown in blue, payrolls in red):
This suggests that, while 3rd quarter GDP was initially reported up at an annual rate of 3.5%, which surprised most people (it may be revised downward somewhat due to increased imports), at least two more quarters of such growth must take place for YoY GDP to reach +2%. The Leading Economic Indicators at this point do indicate positive GDP growth in the 4th quarter, carrying over at least into part of the first quarter of 2010, but after that, the jury is out -- especially with $80 Oil.
III. Real Retail Sales
A major revision of recent data by the Census Bureau calls this element of my analysis into question. I described real retail sales as the Holy Grail of Leading Indicators for job growth, noting that his consistently turned at both tops and bottoms, an average of 3-5 months before job growth or losses turned. When real retail sales stay flat, they generate a lot of noise, and a longer period between the turn in sales and payrolls. Strong turns in sales generate reliable subsequent moves in payrolls in subsequent months. In general, with regard to recoveries, an increase of about +2.5% a year is necessary to reliably generate a subsequent move in real retail sales.
On a three month smoothed average basis, real retail sales bottomed in April of this year. Despite October's growth of 1.4%, nasty revisions downward of August and Septmber totaling -1.3% mean that real retail sales are up only about 0.9% off that low, for an annual rate of 1.7%. In short, the revisions keep real retail sales within the range of "noise." The past few months may be the beginning of a solid upward trend, or continued revisions may mean the series stays flat with no growth whatsoever.
Here is a graph comparing the trend in Real Retail Sales since its smoothed bottom in April with the first 7 months of growth during the expansions of 1983, and 1992, and the first eight months of growth in 2001 (September and October 2001 show extreme volativity for obvious reasons so I also included November):
This is simply not consistent with job growth yet.
Conclusion: BLS revisions and John Maynard Keynes
John Maynard Keynes famously justified a change in opinion by saying, "When the facts change, I change my opinion. What do you do?" Well, a critical fact in my analysis has changed.
Three indicators -- ISM manufacturing, Initial Jobless Claims, and Industrial Production -- are now at levels typically associated with actual job growth in previous recoveries. All three of these series exceed their growth in the two "jobless recoveries" of 1992-3 and 2002-3.
The "Holy Grail," however, Real retail sales, is only trending sideways or slightly higher albeit for unique reasons (cash for clunkers). The substantial downward revisions in real retail sales in August and September call into question when the overall jobs number will turn. While industrial production is indeed having a "V" shaped recovery so far, growing at a rate similar to 1983; by contrast, Real retail sales' anemic growth is similar to -- and even weaker than -- its pattern in the last "jobless recovery," below the 3% annual growth typically associated with job growth. The services sector is stagnated with continuing substantial job losses, and GDP growth, while impressive, isn't impressive enough to think that jobs will actually turn positive by year's end.
At the end of the day, the Leading Economic Indicators either work or they don't. If they work, then the coincident indictors of real income, real retail sales, and employment must start to trend upward shortly. What may be developing is a bifurcated economic expansion, in which jobs in manufacturing are added back quickly (hence, not a "jobless recovery") but services continue to shed jobs as consumers continue to repair their balance sheets and shop discerningly for bargains (thus a "job loss recovery"). This tends to move the turning point for jobs into January-March (January at least being consistent with my original prediction).
Finally, as I said in my concluding installment in September and have reiterated since: let me be the first to acknowledge that this is not a scientific truth or certainty, but a best estimate based on a logical review of existing data with a long history that accommodates both traditional and "jobless" recoveries.
Regarding "Facts"
Invictus has a piece just below where he highlights a fairly typical event: a public figure making up facts as they go along. Unfortunately, this is more and more of a regular event.
At first I found this existed on the right side of the political aisle. The reality is extremely conservative economic doctrine -- the whole "let the market work it out" argument -- just doesn't work as planned. The basic problem is any system needs rules to survive and prosper along with referees. Imagine a sporting event without a ref. No one would go because eventually it would break down into a fight. In addition, we need rules to create a structure. That's just the way things work.
But we're also seeing this more and more on the left side of the political aisle. As the recovery has continued -- as the data has turned more and more positive -- there are more and more allegations that the data is rigged -- at least the data that is positive. People go on to quote negative data as if it is gospel. And economists are ignorant bastards -- until they're bearish. Then they are to be trusted completely. In other words, people on the left are now guilty of what they accused the Republicans of -- making up facts to fit a preconceived world view.
What we are seeing more and more is the existence of multiple realities where people can find "facts" to fit their view. Think tanks on both sides of the aisle are happy to spin in any direction.
The problem is no one is living in reality. Anything that conforms to their world view is golden; anything that doesn't conform to their world view is corrupt. And as a result of all of this, we're not going to get anywhere.
At first I found this existed on the right side of the political aisle. The reality is extremely conservative economic doctrine -- the whole "let the market work it out" argument -- just doesn't work as planned. The basic problem is any system needs rules to survive and prosper along with referees. Imagine a sporting event without a ref. No one would go because eventually it would break down into a fight. In addition, we need rules to create a structure. That's just the way things work.
But we're also seeing this more and more on the left side of the political aisle. As the recovery has continued -- as the data has turned more and more positive -- there are more and more allegations that the data is rigged -- at least the data that is positive. People go on to quote negative data as if it is gospel. And economists are ignorant bastards -- until they're bearish. Then they are to be trusted completely. In other words, people on the left are now guilty of what they accused the Republicans of -- making up facts to fit a preconceived world view.
What we are seeing more and more is the existence of multiple realities where people can find "facts" to fit their view. Think tanks on both sides of the aisle are happy to spin in any direction.
The problem is no one is living in reality. Anything that conforms to their world view is golden; anything that doesn't conform to their world view is corrupt. And as a result of all of this, we're not going to get anywhere.
Forex Fridays
We're still seeing an incredibly bearish overall trend in place. Note that we are still seeing a trend of lower highs (A) and lower lows (B). Also note the bearish orientation of the EMAs (D) -- all are moving lower and the shorter are below the longer.
However, there is still the possibility of a double bottom emerging (C). Note the momentum increased on the second bottom and the RSI printed a higher total. Fundamentally, it's still very difficult to see what the catalyst for a rebounding dollar would be. The US is printing tons of debt and our overall interest rates are still very low. While we have printed a stronger GDP, the economy is still on shaky ground.
Thursday, November 19, 2009
Business As Usual: Make Stuff Up
During Tim Geithner's somewhat contentious appearance before Congress today, some Republicans took every opportunity to get their shots in. Congressman Michael Burgess (Asshat-TX), was no exception. He got his sound bite in -- "I don't think you should be fired, I think you never should have been hired," or something to that effect. Whatever, he's entitled to his opinions, but not his own facts.
It's what the Congressman said immediately before that that intrigues me, as he and Geithner duelled on the merits of different economic policies, Burgess being all about "cut taxes and get out of the way." He said, and I quote:
“When I came here in 2003 we were in a jobless recovery. Tax relief was passed in May of 2003 and as a consequence, by July of that year we were adding jobs at a significant [ed. note: he stresses "significant"] rate. It [referring to tax relief] seems to have worked fairly well.”
Watch it here at 3:57 into the video.
Well, if by "significant" he means a 25,000 add in July 2003 followed by a
-42,000 loss in August, then he's on to something. Otherwise, he's just a lying sack of shit (whose office was oddly unable to answer my simple question as to what the hell he was talking about):
Business as usual: Just make stuff up.
P.S. Congressman, if you're out there, just drop your response in the comments section.
Today's Market
I've spent a fair amount of time explaining why I don't like the current market. This is another post along those lines.
Click for a larger image
A.) On November 9 prices gapped higher, printing a strong bar. That's the kind of event bulls like.
B.) The prices clustered in a very narrow range, printing three very weak candles. This is a terrible way to follow-through from a strong up day.
C.) Prices can't gtet above the 111 level in a meaningful way. They try but just can't get above the level with any momentum. This is followed by today's action which was a sell-off.
The EMA picture is still bullish. But notice that prices just aren't moving higher with any conviction. Combine with with the weak performance by the Tranports and microcaps and I'm just not impressed.
Click for a larger image
A.) On November 9 prices gapped higher, printing a strong bar. That's the kind of event bulls like.
B.) The prices clustered in a very narrow range, printing three very weak candles. This is a terrible way to follow-through from a strong up day.
C.) Prices can't gtet above the 111 level in a meaningful way. They try but just can't get above the level with any momentum. This is followed by today's action which was a sell-off.
The EMA picture is still bullish. But notice that prices just aren't moving higher with any conviction. Combine with with the weak performance by the Tranports and microcaps and I'm just not impressed.
Housing Starts Drop
From the Census:
However, let's put that number in visual perspective:
Click for a larger image
The total annual pace of housing starts (1 unit) has been between 476,000 -511,000 for the last five months. So far, that looks like a low-level range. In addition,
Click for a larger image
Total starts have been in a roughly 100,000 unit range for 10 months.
Right now, it looks like we're bottoming.
Privately-owned housing starts in October were at a seasonally adjusted annual rate of 529,000. This is 10.6 percent (±8.7%) below the revised September estimate of 592,000 and is 30.7 percent (±8.3%) below the October 2008 rate of 763,000.
However, let's put that number in visual perspective:
Click for a larger image
The total annual pace of housing starts (1 unit) has been between 476,000 -511,000 for the last five months. So far, that looks like a low-level range. In addition,
Click for a larger image
Total starts have been in a roughly 100,000 unit range for 10 months.
Right now, it looks like we're bottoming.
October Leading Indicators
As I predicted a couple of weeks ago, October Leading Economic Indicators (and revisions to September) came in at +0.3, the seventh positive reading in a row. This suggests that economic growth will continue through this quarter and the first quarter of 2010 as well.
To repeat what I said then: typically, even in the last two "jobless recoveries", jobs began to be added to the economy when the YoY LEI was up +5% or better. This month will replace the awful -1% of October 2008, meaning that for the last 7 months, the LEI is up 5.9%, and up 4.2% YoY. If the LEI simply print flat for November and December, the YoY growth will be +5.0%, consistent with jobs being added in December or January.
In view of this week's poor housing permits number, that should prove interesting.
To repeat what I said then: typically, even in the last two "jobless recoveries", jobs began to be added to the economy when the YoY LEI was up +5% or better. This month will replace the awful -1% of October 2008, meaning that for the last 7 months, the LEI is up 5.9%, and up 4.2% YoY. If the LEI simply print flat for November and December, the YoY growth will be +5.0%, consistent with jobs being added in December or January.
In view of this week's poor housing permits number, that should prove interesting.
Sage Wisdom For Investing -- In Anything
I oft mention David Rosenberg, and he oft mentions one of the wisest men on Wall St., Bob Farrell. Farrell is a legendary, Hall of Fame market maven and technician who plied his trade for decades at Merrill Lynch. When he left (he still writes a subscription-only newsletter), he penned his Market Rules to Remember, which I'd posted a long while ago over at Blah3 and will share with the Bonddad crowd now. These rules are truly timeless, and applicable to investing in just about anything. Without further ado:
1) Markets tend to return to the mean over time.
2) Excesses in one direction will lead to an opposite excess in the other direction.
3) There are no new eras -- excesses are never permanent.
4) Exponential rapidly rising or falling markets usually go further than you think, but they do not correct by going sideways.
5) The public buys the most at the top and the least at the bottom.
6) Fear and greed are stronger than long-term resolve.
7) Markets are strongest when they are broad and weakest when they narrow to a handful of blue chip names.
8) Bear markets have three stages -- sharp down, reflexive rebound, and a drawn-out fundamental downtrend.
9) When all the experts and forecasts agree -- something else is going to happen.
10) Bull markets are more fun than bear markets.
Print them out and keep them handy. You'll be glad you did.
Intial Jobless Claims: 505,000
The BLS reported that for the week ending Nov. 14, seasonally adjusted initial jobless claims were 505,000. Last week's number was revised slightly higher to 505,000 as well.
"The 4-week moving average was 514,000, a decrease of 6,500 from the previous week's revised average of 520,500."
Unadjusted, there were 479,295 new claims, a decrease of 53,132 from the week before, and well below the 513,000 initial claims in the same week last year.
In unadjusted terms, this was the best new claims number, relative to normal seasonal adjustment, in over a year. The 4 week moving average is now about 21% lower than the peak of 658,750 on April 3 of this year. Needless to say, the continuing decline in the number of new claims bodes well for the jobs outlook.
"The 4-week moving average was 514,000, a decrease of 6,500 from the previous week's revised average of 520,500."
Unadjusted, there were 479,295 new claims, a decrease of 53,132 from the week before, and well below the 513,000 initial claims in the same week last year.
In unadjusted terms, this was the best new claims number, relative to normal seasonal adjustment, in over a year. The 4 week moving average is now about 21% lower than the peak of 658,750 on April 3 of this year. Needless to say, the continuing decline in the number of new claims bodes well for the jobs outlook.
Thursday Oil Market Round-Up
A.) Prices are still contained by a trend line
B.) The EMA picture is still bullish -- the shorter EMAs are above the longer EMAs. But also notice the 10 day EMA is more or less horizontal, indicating a flat line short term trend. The longer terms trends (20 and 50 day EMAs) are still positive. Finally, note that on several times over the last few weeks prices have used the 200 day EMA as technical support.
A.) Momentum is decreasing but not crashing. This is standard in a consolidation pattern.
B.) The A/D line is still positive, indicating we have not seen a huge outflow of money from the market.
Wednesday, November 18, 2009
Today's Market
Yes, the SPYs are rallying. BUT
A.) Notice that prices have not been able to get above 111.50. Also note the candles are getting smaller. And
B.) There is low volume. If people are so excited about this market then they should be stampeding into the market, thereby increasing volume.
And if the market is so strong, why aren't the Transports confirming the rally?
A.) Or the microcaps -- which are still hitting resistance at the EMAs.
This is not a broad rally which is very disconcerting.
Congress and White House Look At Jobs Bill
From the NY Times:
With Congressional Democrats in near-panic amid forecasts that unemployment will remain high through next November’s midterm elections, a party leader said on Thursday that the House will pass a new “jobs bill” before Dec. 18.
Senate Democrats likewise are weighing options. And the signals from Congress follow by a day the White House’s announcement that President Obama will follow his “Forum on Jobs and Economic Growth” on Dec. 3 with a “Main Street Tour” starting the next day in Allentown, Pa., and continuing to other hard-hit places in coming months.
With more than half of last winter’s $787 billion package of tax cuts and stimulus spending still in the pipeline, Representative Steny H. Hoyer, the Democratic majority leader from Maryland, said the new measure should not be called another stimulus bill.
“I don’t want it to be as broad as that,” he said. “I want it to be very targeted on jobs.”
He indicated that the legislation might include money for public jobs, which many liberals have advocated; tax credits to employers for new hires, an Obama campaign proposal that was shelved early this year amid concern that businesses might game their payrolls; and additional spending for infrastructure and road projects.
First, this is an overall good idea. With unemployment at 10.2% every little bit helps.
But it's also important to remember exactly what has happened in the economy and where we are in the cycle to understand exactly what is going on. And no -- the following is not an endorsement of bad times; it is simply an explanation of the facts without a judgment attached.
At the end of last year and the beginning of this year the US was losing jobs at a rate of 600,000/month. That lasted for 5 months. Or to put it another way, the US lost 3 million jobs in 5 months. That is almost half of all the jobs created during the last expansion. That tells us the severity of the economic situation was indeed severe. Employers simply cut everybody they could and then some. In addition, we have also learned that the BLS has added an additional 800,000 jobs losses to the official job loss total. These will be added in February. This further indicates we were in an extremely severe economic contraction.
Currently the economy is back from the brink. We printed a solid GDP number last quarter and the rate of job losses continues to decrease. In addition, the pace of initial unemployment claims continues to move lower. All of these facts tell us we're moving in the right direction. BUT -- and this is very important -- remember where we were last year at this time. There were a lot of people talking about depressions and deflationary spirals. These are the most severe economic events we can experience. The repercussions of these events last a long time.
Does this statement imply that I am unsympathetic to the unemployed? No. I have never advocated (and will never advocate) that we decrease or cut off unemployment benefits or show any less sympathy for those who have lost their jobs. That is not the point of the above recitation of historical facts. The point is things are moving in the right direction. Recoveries -- especially from near financial collapses -- don't happen overnight. As the article states we've only just started to use the stimulus money. We've just printed our first quarter of positive GDP growth. By this time next year things should be better.
October CPI: the end of the Deflationary Bust
- by New Deal democrat
This morning's CPI came it at +0.3%, slightly higher than estimates. YoY CPI is -0.2%. This is undoubtedly the swan song for deflation this year. Almost certainly we will find that this month prices increased about 2.0% YoY.
All things considered, that's a good thing. Here's a graph showing YoY inflation for consumers (blue), finished goods (red), and commodities (green):
When all three are rising, green more than red more than blue, that's bad and almost always triggers a recession (always in the presence of an inverted yield curve 1 year previously). That's because price pressures in crude goods can't be passed on to producers, who in turn can't turn them on to consumers, so there are cutbacks, triggering a recession. All three then proceed to fall, as demand slackens.
When demand hits bottom, crude increases can be passed on to producers with room to spare (so profits increase), and similarly producers can pass on increases to consumers. Increased demand and increased profits lead to economic expansion and hiring new employees. (Graphically this means all three lines rising, with blue higher than red higher than green.)
As I have repeastedly noted, this was also true of the Great Depression and the 1920's booms and busts.
In short, the K.I.S.S. signal -- a positive yield curve, a rising rate of price changes, and cpi exceeding ppi -- together with the positive LEI of the last few months, indicate the economy will show growth for the 4th quarter.
This morning's CPI came it at +0.3%, slightly higher than estimates. YoY CPI is -0.2%. This is undoubtedly the swan song for deflation this year. Almost certainly we will find that this month prices increased about 2.0% YoY.
All things considered, that's a good thing. Here's a graph showing YoY inflation for consumers (blue), finished goods (red), and commodities (green):
When all three are rising, green more than red more than blue, that's bad and almost always triggers a recession (always in the presence of an inverted yield curve 1 year previously). That's because price pressures in crude goods can't be passed on to producers, who in turn can't turn them on to consumers, so there are cutbacks, triggering a recession. All three then proceed to fall, as demand slackens.
When demand hits bottom, crude increases can be passed on to producers with room to spare (so profits increase), and similarly producers can pass on increases to consumers. Increased demand and increased profits lead to economic expansion and hiring new employees. (Graphically this means all three lines rising, with blue higher than red higher than green.)
As I have repeastedly noted, this was also true of the Great Depression and the 1920's booms and busts.
In short, the K.I.S.S. signal -- a positive yield curve, a rising rate of price changes, and cpi exceeding ppi -- together with the positive LEI of the last few months, indicate the economy will show growth for the 4th quarter.
A Positive Economist
FYI: there are people who see positive things in the economy. (I realize he's probably just a corporate shill who is relying on compromised data, but anyway...)
Wednesday Commodities Round-Up
A.) Prices consolidated in a triangle pattern from roughly mid-October to mid-November.
B.) Prices moved out of this price range by gapping higher and printing a very strong candle. Also notice the volume surge that took place.
Also note that prices are still above the long-term trend line. Finally, we still have a very bullish EMA orientation -- the shorter EMAs are above the longer EMAs, all the EMAs are moving higher and prices are above all the EMAs.
A.) Like copper, agricultural commodities formed a triangle consolidation pattern from roughly mid-October to mid-November.
B.) Price broke out of this pattern yesterday by printing a strong bar on high volume.
The EMA picture is a bit more muddled. Prices have been coalescing around the 200 day EMA and the shorter EMAs are below the 200 day EMA. In addition, notice the EMAs are in a tight range. Ideally in a bull market we'd like to see a more bullish orientation for the EMAs.
Tuesday, November 17, 2009
Today's Market
A.) The QQQQs may be forming a broadening top formation.
B.) but the EMA picture is still very strong -- the shorter EMAs are above the longer EMAs, all the EMAs are rising and prices are above all the EMAs.
C.) Momentum was leaving the index, but
D.) We've seen a recent increase in overall momentum and
E.) If we're really topping out, why aren't we seeing money flow out of the market? Instead, we're seeing the A/D line hold steady.
Empire State and Retail Sales Recap
From the NY Federal Reserve:
Here is the relevant chart:
Notice the index is currently at levels associated with expansion. Also note the index has rebounded from the extremely low levels we saw earlier this year. SilverOz has noted that part of most statistical rebounds we are seeing is due to the extreme readings we saw earlier this year; that is, part of what we are seeing is a standard rebound. I think that's accurate. The economy literally hit the brakes at the end of last year/beginning of this year. However, I think there are continuing signs the economy is recovering. Here are additional charts from the report:
Notice the overall trend for both overall business and new orders is positive.
Also note the employment component of the report is also getting better.
Retail Sales Increase:
From the Census Bureau:
Here are some very important observations:
Simply put, sales are moving in the right direction. Also note the alot of the increase was car sales - in a post cash for clunkers world. That tells us there is still demand for autos out there without government stimulus.
Here is the relevant chart:
A while ago I noted that we can look at the recent data in the following way:
A.) The complete contraction in any spending and
B.) The recovery where the pace of month over month percentage changes returned to more normal levels. That appears to be where we are now.
Short version, both of these pieces of data are positive.
The Empire State Manufacturing Survey indicates that conditions for New York manufacturers improved in November, but at a somewhat slower pace than in October. The general business conditions index fell 11 points, to 23.5. The indexes for new orders and shipments posted similar declines. Pricing pressures eased, with the prices paid index positive but lower than last month and the prices received index rising to a level just below zero. Employment indexes fell from October’s elevated levels, remaining slightly positive. Future indexes conveyed an expectation that activity and employment would improve in the months ahead and that both input and selling prices would increase significantly.
Here is the relevant chart:
Notice the index is currently at levels associated with expansion. Also note the index has rebounded from the extremely low levels we saw earlier this year. SilverOz has noted that part of most statistical rebounds we are seeing is due to the extreme readings we saw earlier this year; that is, part of what we are seeing is a standard rebound. I think that's accurate. The economy literally hit the brakes at the end of last year/beginning of this year. However, I think there are continuing signs the economy is recovering. Here are additional charts from the report:
Notice the overall trend for both overall business and new orders is positive.
Also note the employment component of the report is also getting better.Retail Sales Increase:
From the Census Bureau:
The U.S. Census Bureau announced today that advance estimates of U.S. retail and food services sales for October, adjusted for seasonal variation and holiday and trading-day differences, but not for price changes, were $347.5 billion, an increase of 1.4 percent (±0.5%) from the previous month, but 1.7 percent (±0.5%) below October 2008. Total sales for the August through October 2009 period were up 1.5 percent (±0.3%) from the same period a year ago. The August to September 2009 percent change was revised from -1.5 percent (±0.5%) to -2.3 percent (±0.3%).
Here are some very important observations:
Excluding the 7.4% increase in auto sales, retail sales rose 0.2% in October, the data showed. Sales excluding autos have risen for three months in a row and in five of the past six months.
Simply put, sales are moving in the right direction. Also note the alot of the increase was car sales - in a post cash for clunkers world. That tells us there is still demand for autos out there without government stimulus.
Here is the relevant chart:
A while ago I noted that we can look at the recent data in the following way:
A.) The complete contraction in any spending and
B.) The recovery where the pace of month over month percentage changes returned to more normal levels. That appears to be where we are now.
Short version, both of these pieces of data are positive.
Industrial Expansion, Services Stagnation (and Weekly Indicators)
- by New Deal democrat
The release of October retail sales yesterday and industrial production this morning make for interesting bookends to the economic situation.
This morning Industrial Production for October was reported up a mere 0.1%. Capacity utilization was reported up 0.2%. I tend to ignore the latter number, since it generally tracks the former, and capacity utilization has been in decline literally for decades, reflecting America’s relative industrial decline. Industrial production, however, is an excellent coincident indicator for industrial growth or weakness. Whether October's meager increase means a slowing down of the trend of expansion, or is just one month's noise, is impossible to tell. In any event, this means that since its bottom at the end of the second quarter, industrial production has grown 2.9% in four months (or about 12% a year), which is the best rebound from a recession trough since 1982. In other words, so far manufacturing is having a V-shaped recovery.
On the other hand, while yesterday’s retail sales figures appeared great at first blush, up +1.4% from September and topping estimates substantially, the downside was the very nasty revisions to August and September. August was revised down from +2.7 to +2.2. September was revised down from -1.5 to -2.3. In other words, those two months together went from +1.2 to -0.1. The efficacy of “cash-for-clunkers” as anything other than a momentary blip appears to have been entirely revised away. Ouch! Thus September may have actually made a new low in real retail sales for the recession. Over the longer term, since April, real retail will be about +0.8 instead of +2.0 as it may have otherwise appeared. The services economy isn’t having a V-shaped recovery; in fact it is barely having a recovery at all.
The above contrast fits perfectly with the ISM manufacturing and non-manufacturing data. Manufacturing moved into expansion first, and is already expanding faster than it has coming out either post-1982 recession. Non-manufacturing, however, is barely expanding at all. Employment in manufacturing has already started to increase, according to the ISM manufacturing report, but employment in services is continuing to fall, and actually fell off at a worse rate last month. This analysis also seems to dovetail well with Invictus’ take on the divergence between large vs. small employers. I’ll have more to say about this divergence in an extended post in a couple of days.
So we have two bookends for the economy: industrial expansion, services stagnation. The recovery from the recession is all about selling goods to foreign consumers, chiefly Asians, whose standard of living is improving. On the other hand, the American consumer is no longer the engine of global growth, but the caboose. His/her standard of living is in decline, and will only turn around when the structural forces which led to that decline have been abated.
One final note: the October growth in retail sales was not surprising, if you’ve been following my “Weekly Indicators” each Friday. I have been tracking these items precisely because they give high volume real-time information. So when automakers reported sales up 10% from September to October, and when the ICSC reported ever-improving week over week sales followed by good monthly same store sales for October, it appeared likely that the retail sales number would oblige, and it did.
From Bonddad:
From the NY Federal Reserve:
Here is the relevant chart:
Notice the index is currently at levels associated with expansion. Also note the index has rebounded from the extremely low levels we saw earlier this year. SilverOz has noted that part of most statistical rebounds we are seeing is due to the extreme readings we saw earlier this year; that is, part of what we are seeing is a standard rebound. I think that's accurate. The economy literally hit the brakes at the end of last year/beginning of this year. However, I think there are continuing signs the economy is recovering. Here are additional charts from the report:
Notice the overall trend for both overall business and new orders is positive.
Also note the employment component of the report is also getting better.
Retail Sales Increase:
From the Census Bureau:
Here are some very important observations:
Simply put, sales are moving in the right direction. Also note the alot of the increase was car sales - in a post cash for clunkers world. That tells us there is still demand for autos out there without government stimulus.
Here is the relevant chart:
A while ago I noted that we can look at the recent data in the following way:
A.) The complete contraction in any spending and
B.) The recovery where the pace of month over month percentage changes returned to more normal levels. That appears to be where we are now.
Short version, both of these pieces of data are positive.
The release of October retail sales yesterday and industrial production this morning make for interesting bookends to the economic situation.
This morning Industrial Production for October was reported up a mere 0.1%. Capacity utilization was reported up 0.2%. I tend to ignore the latter number, since it generally tracks the former, and capacity utilization has been in decline literally for decades, reflecting America’s relative industrial decline. Industrial production, however, is an excellent coincident indicator for industrial growth or weakness. Whether October's meager increase means a slowing down of the trend of expansion, or is just one month's noise, is impossible to tell. In any event, this means that since its bottom at the end of the second quarter, industrial production has grown 2.9% in four months (or about 12% a year), which is the best rebound from a recession trough since 1982. In other words, so far manufacturing is having a V-shaped recovery.
On the other hand, while yesterday’s retail sales figures appeared great at first blush, up +1.4% from September and topping estimates substantially, the downside was the very nasty revisions to August and September. August was revised down from +2.7 to +2.2. September was revised down from -1.5 to -2.3. In other words, those two months together went from +1.2 to -0.1. The efficacy of “cash-for-clunkers” as anything other than a momentary blip appears to have been entirely revised away. Ouch! Thus September may have actually made a new low in real retail sales for the recession. Over the longer term, since April, real retail will be about +0.8 instead of +2.0 as it may have otherwise appeared. The services economy isn’t having a V-shaped recovery; in fact it is barely having a recovery at all.
The above contrast fits perfectly with the ISM manufacturing and non-manufacturing data. Manufacturing moved into expansion first, and is already expanding faster than it has coming out either post-1982 recession. Non-manufacturing, however, is barely expanding at all. Employment in manufacturing has already started to increase, according to the ISM manufacturing report, but employment in services is continuing to fall, and actually fell off at a worse rate last month. This analysis also seems to dovetail well with Invictus’ take on the divergence between large vs. small employers. I’ll have more to say about this divergence in an extended post in a couple of days.
So we have two bookends for the economy: industrial expansion, services stagnation. The recovery from the recession is all about selling goods to foreign consumers, chiefly Asians, whose standard of living is improving. On the other hand, the American consumer is no longer the engine of global growth, but the caboose. His/her standard of living is in decline, and will only turn around when the structural forces which led to that decline have been abated.
One final note: the October growth in retail sales was not surprising, if you’ve been following my “Weekly Indicators” each Friday. I have been tracking these items precisely because they give high volume real-time information. So when automakers reported sales up 10% from September to October, and when the ICSC reported ever-improving week over week sales followed by good monthly same store sales for October, it appeared likely that the retail sales number would oblige, and it did.
From Bonddad:
From the NY Federal Reserve:
The Empire State Manufacturing Survey indicates that conditions for New York manufacturers improved in November, but at a somewhat slower pace than in October. The general business conditions index fell 11 points, to 23.5. The indexes for new orders and shipments posted similar declines. Pricing pressures eased, with the prices paid index positive but lower than last month and the prices received index rising to a level just below zero. Employment indexes fell from October’s elevated levels, remaining slightly positive. Future indexes conveyed an expectation that activity and employment would improve in the months ahead and that both input and selling prices would increase significantly.
Here is the relevant chart:
Notice the index is currently at levels associated with expansion. Also note the index has rebounded from the extremely low levels we saw earlier this year. SilverOz has noted that part of most statistical rebounds we are seeing is due to the extreme readings we saw earlier this year; that is, part of what we are seeing is a standard rebound. I think that's accurate. The economy literally hit the brakes at the end of last year/beginning of this year. However, I think there are continuing signs the economy is recovering. Here are additional charts from the report:
Notice the overall trend for both overall business and new orders is positive.
Also note the employment component of the report is also getting better.Retail Sales Increase:
From the Census Bureau:
The U.S. Census Bureau announced today that advance estimates of U.S. retail and food services sales for October, adjusted for seasonal variation and holiday and trading-day differences, but not for price changes, were $347.5 billion, an increase of 1.4 percent (±0.5%) from the previous month, but 1.7 percent (±0.5%) below October 2008. Total sales for the August through October 2009 period were up 1.5 percent (±0.3%) from the same period a year ago. The August to September 2009 percent change was revised from -1.5 percent (±0.5%) to -2.3 percent (±0.3%).
Here are some very important observations:
Excluding the 7.4% increase in auto sales, retail sales rose 0.2% in October, the data showed. Sales excluding autos have risen for three months in a row and in five of the past six months.
Simply put, sales are moving in the right direction. Also note the alot of the increase was car sales - in a post cash for clunkers world. That tells us there is still demand for autos out there without government stimulus.
Here is the relevant chart:
A while ago I noted that we can look at the recent data in the following way:
A.) The complete contraction in any spending and
B.) The recovery where the pace of month over month percentage changes returned to more normal levels. That appears to be where we are now.
Short version, both of these pieces of data are positive.
Subscribe to:
Posts (Atom)