The attached link is a release of the second annual survey of Small Business
owners surveyed by Chase Bank, which "surveyed CEOs and senior executives on
the prominent Inc. 500/5000 list of fastest growing U.S. small businesses about
their outlook on the economy, employment, general growth, and critical issues
to their business, such as access to capital and cash flow management". As can
be seen the results are far more bullish than the prevailing news-flow about
corporate management would suggest, particularly those of small businesses.
Even though Chase's survey selected companies that were growing faster than
their peers this survey certainly challenges the view that no substantial
progress has been made since last summer and that US small businesses are
incapable of participating is the current recovery.
Chase Bank Survey of small business owners

Looking at this morning's array of macro data we were struck by how much better
European Sentiment indexes are than their US counterparts at present. For
instance September's EC Eurozone Economic Sentiment Indicator came in at 103.2,
up from 102.3 in August and the best reading since January 2008 (a poll taken
just prior to the Jerome Kerviel inspired sell-off). Even though this index is
an amalgam of both consumer and industrialist sentiment it is still striking
how much better it has fared than the US consumer sentiment surveys (or for
that matter the NFIB Small Business Optimism index), particularly given that
there is no evidence that Europe overall has fared substantially better than
the US over the last two years (perhaps they are simply more used to economic
malaise). Indeed, as the attached chart shows the spread between the Eurozone
Sentiment index and the Conference Board index reached a record -54.70 in
September. From our perspective this underlines the degree to which US consumer
sentiment indexes remain in "recovery denial". As we have remarked before this
is far more of a problem for the current administration, which is facing
imminent mid-term elections, than for the US equity market. For the latter the
potential for consumer confidence to rebound (and with it a willingness to
allocate funds to the equity market) is a potentially bullish factor going
forwards. - eurousconfidence.gif

As we explained in yesterday's "Speculator Extra", September's brisk equity
market rally has been met with extreme scepticism by the consumer sentiment
polls. The Conference Board Consumer Sentiment Index mirrored the University of
Michigan Poll by falling to 48.50, the lowest reading since February. The
"Present Situation" index fell slightly to 23.10 from 24.90 last month (this
poll suggests things are as bad today as they were at the depth of the
recession which just shows how surveys of consumers' views differ from what
they are actually experiencing) but the "Expectations" survey fell much more
sharply from 72 to 65.40, the lowest reading since July 2009.
.
From our perspective this downshift in sentiment is encouraging since it
represents another step towards completing the corrective phase that we have
been enduring for the last few months. While not at an extreme low (apart from
the Current situation index), the survey certainly suggests that any uptick in
reported economic activity and/or continued advance by the equity market would
constitute a surprise to consumers. Ultimately it is the difference between
reality and what people were expecting it to be that influences market action,
and this month's survey shows that the "good news" bar has been substantially
lowered by US consumers.

Some interesting data out of the UK this morning that suggests an acceleration
of retail activity may be taking place. Attached is a chart of the CBI
Retail-Sales index which is compiled based on a survey of large UK retailers
(61 took part in September's poll). The level of the index represents net
positive responses (sales are growing) and has a theoretical range of -100 to
+100. As can be seen September's reading came in at a very high +49, well above
August's +35 and blowing the consensus estimate of +25 out of the water. This
is the highest monthly reading since May 2004 and pulls the trailing 6 month ma
(red) up to a respectable +17. Perhaps more interestingly retailers'
expectations are also recovering strongly. The expected sales index reached 47,
the highest level since October 2001. We have become used this cycle to
corporate management being very cautious about projecting the recovery in
activity into future growth. This has the effect of delaying inventory rebuild
and therefore depresses production. At least in the UK retail sector this
problem seems to be behind us and it will be interesting to monitor comments
made in the upcoming earnings season to see if this trend start to be more
widely apparent. - cbisalessep10.gif - cbiexpectedsep10.gif

During the corrective phase we have sporadically tracked the Positive and
Negative Volume index for the SPX index. In recent months these measures have
shown a consistent tendency for trading volume to diminish on positive days and
build up during negative sessions (see attached chart). This factor has
been one of the technical hurdles that we felt needed to be overcome before
this corrective phase could be put behind us and there are finally some
signs that this may in fact be occurring.
.
As can be seen on the attached chart, during the September surge the
Positive volume index (blue, lower chart) has managed to make sustained
progress. This is quite different from what we say during the prior strong
bounces that took place in mid June and August, and suggests that something
different may be happening this time around. We are yet to see a similar
response by the Negative volume index, which should start to lose its value in
a sustained advance but it should be noted that there have been much fewer
negative sessions to measure volume in recent weeks, and at least this index's
inexorable progress has been halted. As such we would view this data to be an
indication that the market's internal dynamics may finally be shifting and will
continue to monitor going forwards..


(See attached file: D-SPX_Index.gif) - D-SPX_Index.gif

The sharply lower mortgage rates that have been in place this summer
continue to draw in applicants for refinancings at a rapid pace. This
week's MBA refinance index came in at 4357.40, a drop of 0.9% from last
week. Although this data has moderated from the peak seen in late August
(5085.30) the length of time spent at or above the 4000 level already marks
this period as historically significant and the 10 week ma of the index has
no reached 4442, a figure that has only ever been surpassed in the 2001 to
2003 "cashout bonanza" and 2009 collapse in rates. It is therefore reasonable
to assume that this wave of refinancing will have a meaningful effect on
consumer income statements going forwards (yet another subsidy of consumption
by the nation's savers), although clearly this will be less than in earlier
periods when refinancing meant also receiving a check for tens of thousands of
dallars.
.
Looking ahead we would now expect the pace of refinancings to start to moderate
sharply since spikes in activity typically only last for 8 to 12 weeks before
moderating sharply as the pool of new applicants tends to get exhausted. This
has important ramifications for the US Treasury market which should be expected
to see an equally sharp reduction in MBS duration hedging over the next few
weeks. In our opinion this is likely to be a much more important factor
than the hoped for re-entry of the FRB into the Treasury market (other than
to replace maturing assets) and for this reason we consider it likely that
longer term yields will move higher in the weeks ahead even if the FRB
launches QE2.


(See attached file: D-MBAVREFI_Index.gif) - D-MBAVREFI_Index.gif

Federal Open Market Committee Sept. 21 Statement: Full Text

Federal Open Market Committee Sept. 21 Statement: Full Text
2010-09-21 18:14:34.647 GMT

US Housing start data showed a sizeable increase in starts from 541K in
July to 598K in August. Although this is some distance above consensus
estimates of 550K, as we have explained before the seasonal adjustments are
now very much moving in favor of the housing data provided nominal activity
remains stable at its current low level. This seems to be what happened in
August since we note that the Non-Seasonally Adjusted Single Family starts
for August were almost unchanged at 39.7K units (annualized 476.4K)
compared to July's reading of 40K. In other words the normal seasonal
factors are far less important at the current remarkably low level of
activity and will not fully reassert themselves until we break out of the
current doldrums.
.
Looking at our preferred metric of Single Family Permits we see no sign of
a pick-up in activity with August's data almost unchanged at 470K. We have
started to use a 36 month ma to track this data which helps underline the
brutal long term nature of the current cycle (activity has never remained
at these levels for anything like the current period of time) and also
should act as a useful breakout indicator once this cycle has run its
course. We would not expect this to happen prior to the 2011 spring selling
season. Neverthelss, the message of the 36 month ma is that the radical
overproduction of housing between 2002 and 2007 has now been largely
corrected which will eventually allow a much more normal level of activity
to reassert itself (the 50 year average is for this data is 1070K permits).


(See attached file: D-NHSPA1_Index.gif) - D-NHSPA1_Index.gif

The NAHB US Homebuilder sentiment index remained at a very low 13 in September
suggesting that the level of activity in the new home market remains at or near
its recent record low level. This is hardly surprising since we would not now
expect to see any significant improvement in activity prior to the start of the
2011 selling season next spring time. Even so it may be that some of the
seasonally adjusted new sale and construction data show a moderate improvement
over the winter period since the very low but essentially static level of
actual "raw" activity means that seasonal adjustment process will have had an
outsized effect on reports in the busier spring and summer months. What is
perhaps more important to track is activity at the larger public homebuilders
which we would expect to dominate any recovery in activity thanks to their
continued access to capital and store of cheap land bought over the last 24
months. As we have pointed out before, the NAHB survey is dominated by smaller
builders (447 took part in September's survey) and although this survey size
helps make its data statistically valid, for the reasons stated above it may
not prove the best guide of the early stages of recovery this time around. -
nahbsep10.gif

We have sporadically commented on Taiwanese export data over the last 18
months since this series gives some insight into changes in economic
activity across geographic regions, particularly in the technology sector
which Taiwanese manufacturing concentrates in. We therefore note with
interest that Taiwanese exports continue to expand very rapidly overall,
with August 2010 marking a new all time high of $34.9 bln, a rise of 23.3%
over the last 12 months. What is perhaps more surprising is that while the
first 12 months of recovery leaned very heavily on exports to Hong Kong and
China, since March the US and Europe have been much greater factors. Indeed
exports to Honk Kong and China have now only risen by 8.84% over the last
12 months while US exports have risen 29.38% and European exports have
risen 25.77%. We would not use this data as anything other than a "rough
and ready" indication of activity but its message seems to be that global
growth in trade is starting to lean much more heavily on the traditional
centers of economic activity than was the case 12 months ago.


(See attached file: M-TWEOTTL_Index.gif)
(See attached file: D-TWEOUS_Index.gif) - M-TWEOTTL_Index.gif -
D-TWEOUS_Index.gif

Despite a significantly better tone to the equity market and a couple of weeks
of firmer economic data the University of Michigan Consumer Sentiment Index
fell to 66.6 in September, the lowest reading since August 2009. This survey
suggests that the US consumer is thoroughly fed up with the current state of
affairs (as was also reflected in this week's primary elections in many states
and municipalities) but this should not be taken as evidence that consumption
is likely to be affected or that the US consumer is under any more duress than
it was at the start of this correction 5 months ago. We would not expect to see
significantly stronger sentiment readings until economic data unambiguously
signals that the fears of recent months were unfounded and employment data in
particular registers further sharp improvement.
.
Meanwhile it is worth noting the discordance between recent AAII investor
sentiment polls and that of the University of Michigan Consumer Sentiment
index. These two data series typically reflect similar opinions about the state
of the markets and the economy respectively (see attached chart) but as we note
above, this month's dip in consumer sentiment shows respondents were utterly
unmoved by the bounce in the equity market. Our sense is that the AAII poll is
drawing off an ever dwindling base of the retail population that still ACTUALLY
CARES about the state of the US equity market while an ever greater number have
simply cut back equity allocation or fully divested and are far less enthused
by a rally of this magnitude.


(See attached file: M-.AAII_Index.gif) - M-.AAII_Index.gif -
unimichigansep10.gif

One of the brightest pieces of financial data in recent months has been the
speed with which US credit card delinquency rates have diminished. As
recently as November 2009 the 30 day+ delinquency rate stood at 5.79%, but
the data has fallen each month since then with August's reading of 4.42% b
being the lowest delinquency rate since October 2008, and equivalent to the
level seen in mid-February 2004 at the start of the last recovery. This is
interesting since initial claims for unemployment remain at a historically
high level (the newly unemployed are in theory the most like to generate
new credit card delinquency) and as can be seen on the attached chart are
considerably worse than they were in early 2004. Indeed the degree to which
delinquency rates have plunged in recent months while claims have remain
stuck in the high 400K's is quite unusual given the traditionally close
match between these data series. No doubt some of this anomaly can be explained
by lower interest rates and also lower credit usage (debit card usage has
recently topped credit card usage for the first time) but we have wondered
for sometime about the reliability of the survey based claims data and this
discrepancy does give some pause for thought. In any event the credit card data
suggests that the US consumer is rapidly repairing its balance sheet and is
able to maintain the current pace of consumption going forwards.



(See attached file: M-DELQUS30_Index.gif) - M-DELQUS30_Index.gif

September's Philly Fed Index came in as expected with the reported level of
-0.7 not differing meaningfully from consensus of +0.5. This report suggests
that Manufacturing activity in this are stalled in late summer which is in
contrast to the national ISM report for August which still indicated robust
growth in activity. As we have commented before the Philly Fed data tends to be
considerably more volatile and give a larger number of false signals in both
directions and so we are not particularly concerned by the last 2 month's weak
data given that they have not been confirmed by the more senior and reliable
ISM data.
.
In terms of the detail of this month's report New Orders remained in slightly
negative territory at -8.10 (-7.10 in August), Shipment activity fell to -7.10
(-4.50) while Inventories shrank sharply at -16.70 (-11.90). The Employment
index moved back into marginal positive territory at +1.80 (-2.70). We doubt
that this data will have much impact on the mood of the market going forwards.
- phillyfedseptember2010.gif

The local Chinese equity market (SHASHR index) remains very much the
problem child of global equity markets. In large part this can be explained
by the fact that Chinese monetary authorities arre almost alone in remaining in
tightening mode as they struggle to deal with the effects of the explosion
in lending and real estate investment that took place over the last two
years. The news that significantly tighter capital controls on local banks
are under consideration has not been well received by the local equity
market. The SHASHR index has fallen back from challenging key resistance at 2820
and is now resting on important support in the 2700-2720 range.
.
The investor calendar may exert further pressure on Chinese equities since
their performance has been dramatically worse than those of other emerging
markets and the overall MXEF index (see attached chart). Quarter end
allocations are therefore likely to be substantially negative, as they have
been for most of the last 6 months. If support at 2700 does give way then a
fall back to the strong support that exists in the 2500-2600 range would be
indicated. Concern surrounding China has somewhat moderated in recent
months, but a fall of this magnitude would probably reignite global
investors' fears and potentially act as a catalyst for the short term pullback
in global equities that seems fairly likely to occur one way or another in the
coming sessions.
.


(See attached file: D-SHASHR_Index.gif) - D-SHASHR_Index.gif

The last 4 weeks have seen a remarkable turnaround in investor sentiment.
On August 26th the spread between AAII Bulls and Bears stood at -28.3, the
third lowest since this recovery rally started in March 2009. This weeks
poll showed that Bulls had risen to 50.89 while bears had fallen to 24.26,
creating a spread of +26.63, the largest positive reading since May 2008.
The 4 week move of +55.36 appears to be the sharpest improvement of
sentiment since April 2004.
.
Clearly this raises the odds of a short term pullback in the equity market,
particularly with the senior US equity indexes already pressing against
strong resisitance. Given the highly volatile readings that this poll has
generated in recent months a retesting of support for the SPX index in the
1040-1050 range is probably all that would be required to collapse
sentiment back to an extreme negative reading and set us up for further
gains. If we have any complaint it is that the correction of 2010 has yet
to record an ultra-low reading that has historically marked the end of a
long correction. The 10 week ma of this measure (we believe that sentiment
has to stay at an extreme for several weeks to have any reliable meaning)
mad its low on Septemer 3rd at -10.87 which strikes us as slightly too high
to definitively mark capitulation and (cruciallly) liquidation of positions
by retail investors. However, this is hardly a scientific measure, and the
fact that US investment flows have been so negative over the last 24 months
for US equities may mean that slightly higher readings than is historically
normal will accompany the end of the current correction. Our sense
therefore remains that we are far closer to the end than the begining of
this particular episode.
.
It is also worth considering that higher investor sentiment could actually
be seen to be a positive over the medium to longer term. One notable factor
about the powerful 2009/10 rally was how low sentiment remained given the
powerful performance of the equity market. The 10 week ma of this measure
never topped 10.68 (recorded in early January 2010), hardly higher than the
average reading of 8.53 since this poll was created in 1987. Other
post-recession rallies typically saw much higher sentiment readings (see
for instance the perfromance of the 10 week ma in 1991, 1993, 2003 and
2004) that remained in place for multiple weeks. We would therefore expect
the eventual breakout of the current trading range to be accompnied by
considerably better sentiment readings than we have seen in recent years
and would expect this to catch a number of observers by surprise as it
occurs.

(See attached file: D-.AAII_Index.gif)
(See attached file: W-.AAII_Index.gif) - D-.AAII_Index.gif - W-.AAII_Index.gif

With the JPY/USD rate approaching the 1995 all time low of 79.75 the
Japanese authorities finally intervened in currency markets last night with
the apparent intention of keeping the cross above 82 (see chart). Typically
initial currency intervention is effective since it causes a substantial
unwind of speculative positions but the marginal impact of subsequent
intervention tends to diminish over the coming months. We have never been
great believers in using currency strength to determine much other than the
conversion rate for goods, services and financial assets, although it
cannot be doubted that Japanese exporters have been somewhat hampered by a
stronger JPY in recent months. Far more interesting to us are the monetary
implications of a prolonged period of currency intervention by the Japanese
authorities since this would imply a substantially faster growth of their
monetary base than has been seen since the early 2000's.
.
The Bank of Japan was the only major global central bank to avoid radically
expanding its monetary base following the collapse of Lehman exactly 2
years ago. As can be seen on the attached chart the total base has only
grown by 9% since September 2008, compared to the doubling of the US,
European and UK bases and the massive monetary growth seen in China. A
prolonged period of currency intervention would involve creating new Yen to
purchase USD (and possibly EUR), and unless sterilized lead automatically
to faster growth of the monetary base. This is something that we had been
hoping to see for several months and frankly had grown tired of waiting
for. Obviously one day's intervention hardly changes matters but there is a
potential symbolism to last night's activity that should not be ignored.


(See attached file: D-JPY_Curncy.gif)
(See attached file: M-JNMBMOB_Index.gif) - D-JPY_Curncy.gif -
M-JNMBMOB_Index.gif

Prior to the unfolding of the correction in April he greatest danger in
global markets appeared to be in those countries whose strong economic
conditions were encouraging a rapid tightening of monetary conditions. We
chose to use Australia as the best market to follow as an indicator and
were justified in our assumptions as activity slowed appreciably and the
equity market suffered a strong correction in the face of a series of rate
hikes by the RBA.
.
This would almost certainly have continued to take place had it not been
for the fact that a slew of weaker than expected US economic data led to an
abandonment of any "exit strategy" by the FRB and ECB in 2010 and caused
other central banks to pause in the face of what appeared to be a
substantial loss in economic momentum in the world's largest economy. After
4 months of stasis for the local RBA Cash Target rate (green line on chart)
which was halted at the historically low (for Australia) level of 4.5% it
appears that local activity may once more be perking up. Some evidence for
this was offered by last nights release of the NAB Business Confidence
Index which bounced strongly to +11, its best reading since April. Should
we be correct in our assumption that the US recovery remains on track the
halting of global monetary tightening may be accompanied by a
re-acceleration of activity in some of the global "hot spots" and we will
be watching data for evidence of this quite closely in the weeks ahead.



(See attached file: D-NABSCONF_Index.gif)

July's Business Inventory report is a strong piece of data which goes some
way to repairing the damage done by the prior 2 reports. Overall
Inventories grew by 0.95%, well above the consensus estimate of 0.7%.
June's data was also revised 0.2% higher to 0.46%. This is the largest
monthly inventory rebuild since July 2008 and is therefore the first piece
of evidence to suggest that US Manufacturers are starting to address the
very low level of inventories that have been in place for the last 18 months
(the Inventory/Sales ratio remains at a meagre 1.26%). Perhaps more important
the Manufacturer sales moved back into positive territory at 0.7% following
June's sharp drop of -0.45%. This was something that we very much hoped would
occur and further supports our argument at the time that the reported drop in
sales had much more to do with the manner in which official data typically
fluctuates around a defined trend than an actual drop in Manufacturing Sales.
At the very least it would appear that fears of a Manufacturing slowdown are
overdone and if in fact Manufacturers have finally started to rebuild their
inventories (we would not yet jump to that conclusion on one month's data)
things may start to look surprisingly positive by Q4 2010.


(See attached file: M-MTIB_Index.gif) - M-MTIB_Index.gif

The macro pendulum continued to swing back towards positive data with this
mornings release of robust August retail sales data. According to the
Census Bureau estimates total sales rose 0.4% (consensus 0.3%) while the
July number was trimmed by 0.1% to 0.3%. Strength was most obvious in the
non-Auto data which saw sales rise 0.6% (consensus 0.3%). Although we have
little faith in the veracity of this data-series on a month-to-month basis
(its erratic nature is clearly visible on the attached chart) trend is
more important and the consecutive positive July and August reports have
kept the 12 month ma (red) rising at a normal pace for an economic recovery
even if it appears that some of the very powerful data released in Q1 2010
overstated actual activity. Against a backdrop of a market which has
largely priced in economic torpor and the possibility of a relapse into
recession positive reports such as this can have an impact on sentiment,
and we highlighted yesterday the fact that the retail sector seems to have
recovered some of its attraction to investors in recent weeks. Of course
the official data needs to be reflected in actual sales reported by public
companies (ultimately the only data that can really be trusted) but our
suspicion is hardening that the macro-concerns of recent months have been
significantly overdone.

(See attached file: M-RSTATOTL_Index.gif) - M-RSTATOTL_Index.gif

The change in sentiment surrounding the US economy led to a particularly
difficult 5 months for the retail sector which fell 22% from its April high
to July low. The pain of these losses was magnified by the fact that this
sector was only widely embraced in early February 2010 (prior to this time
it had been one of the favorite sectors to short despite its strong
performance) and the fact that an allegedly slowing US economy would result
in several quarter of lackluster sales.
.
These fears appear to have been overdone. Most retailers reported positive
sales for Q2 2010 and few warnings have been issued for summer activity,
which in any case is the least important portion of the calendar. With the
"back to school" season now underway, to be followed by the key holiday
shopping period the true strength (or lack thereof) of consumer activity
should become apparent. Interestingly this sector has comfortable
outperformed the overall SPX during the August/September rally, the first
period of sustained outperformance since April. The sub-index has also
managed to break comfortably above its July and August highs although it
remains well below its June high (the SPX came within 8 points of its June
high this morning). At the very least we would judge that this sector's
period of under-performance has been completed and we would be alert to the
possibility that retail may be one of the prime beneficiaries of a recovery
in US economic data going forwards.


(See attached file: D-S5RETL_Index.gif) - D-S5RETL_Index.gif

We resume publication where we left off 2 weeks ago with a consideration of
the US economic data cycle. This, we have argued, is distinctly different
from the actual economic cycle since data and consensus estimates tend to
oscillate many times during any single economic cycle. Attached is a chart
of the Citigroup US Economic Surprise Index which has taken a decided turn
for the better during our absence. When we last wrote in late August the
indicator was below -64, a reading that was below the minimum that we thought
sufficient to complete the cycle (-50) but potentially signalling further
disappointment ahead. Ironically this proved to be the exact low for this
indicator as a broad collection of better than expected data was released in
areas including employment, housing and inventories.
.
As of Friday the indicator had risen powerfully to -16.90, marking a
distinct improvement in the overall tone of data. Of course it should be
recognized that this indicator remains in negative territory, meaning that
over the last 90 days data has on the whole disappointed, but the turn in
the direction strikes us as important. As can be seen on the chart this
indicator gives very few false signals provided a reasonable moving average
is employed (we show a 4 week ma on the chart). We would therefore expect
US economic data to continue to improve RELATIVE TO CONSENSUS and push into
positive territory sometime in the next few weeks. Given the fact that the
last 5 months of weakness has largely been driven by a widespread fear of
US macroeconomic factors the importance of the data-pendulum moving back
into positive territory is hard to overstate. If this were to occur against
a backdrop of another strong quarter for corporate earnings we would
imagine this would be sufficient to put the difficult and violent trading range
of mid-2010 behind us. Nevertheless in the shorter term the market remains
vulnerable to reversal at the top end of its trading range and we would
expect to see at least one more dip back into support, but perhaps we
should start to look at the macro-calendar as a source of short term relief
rather than pain going forwards.



(See attached file: W-CESIUSD_Index.gif) - W-CESIUSD_Index.gif