MACRO FOR MARKETS mfglobal.com 1 Disappointing Employment Report The June employment report was much weaker than expected. While disappointed, we caution against simply extrapolating. Yes, there has clearly been some loss of momentum in the last few months, but (a) we see good reasons to believe that the latest employment data exaggerate the falloff in the underlying trend, and, (b) there are also good reasons to expect momentum to turn more positive in coming months. That said, risks are rising that we will have to pare our growth projections. The 22,000 per month average gain in payrolls in the last two months is down sharply from a 179,000 per month pace in the first four months of the year. In our view, that degree of slowing is much greater than seems consistent with other data, including the employment indexes in the ISM surveys and jobless claims. We continue to expect the unemployment rate to trend lower in the next year and a half, although we have adjusted our forecast slightly to reflect the higher than expected level in June. We now forecast a decline to 8.8% at the end of 2011 (instead of 8.6%) and 7.9% at the end of 2012 (instead of 7.8%). At 9.2%, the June reading was up from a low of 8.8% in March but still down from 9.6% in Q4 of 2010 and 10.0% in Q4 of 2009. Preview: Lots of Key GDP Input Data + Fed Chairman Testimony We will review our estimate for Q2 real GDP growth after some key input data are reported in the upcoming week—most important will be foreign trade, federal outlays, retail sales, and inventories. Currently, risks appear firmly on the downside relative to our 3.5% estimate for the annualized pace, although we expect many forecasters will have to raise their estimate in the week ahead. The consensus appears to be close to 2.0%. Also of note: the PPI and CPI, claims, industrial production, and the Michigan sentiment survey. The Fed chairman will testify. The weakness in payrolls in the last two months has contrasted with the relative strength being signaled by the ISM employment indexes. * Based on a 2003-09 regression of payrolls changes vs. the two individual ISM employment indexes; the regression implies a 26% weight for manufacturing and 74% for nonmanufacturing. Note: Shaded bars represent periods of recession. Source: Bureau of Labor Statistics, Institute for Supply Management, and MF Global Economic Analysis | US JAMES F. O’SULLIVAN STEPHANIE S. CHENG Chief Economist Economist+1 212 589 6479+1 212 589 6373 [email protected][email protected]CONTENTS Pg. 2 | Di sappointing Employment Report Pg. 6 | Forecast Summary Pg. 7 | Data Preview Pg. 14 | Calendar JULY 8, 2011 INSTITUTIONAL USE ONLY MF Global Weekly Report 30 35 40 45 50 55 60 -840 -630 -420 -210 0 210 420 01 02 03 04 05 06 07 08 09 10 11 12 Payroll s e x census Mod el esti ma te for p ayr olls based on regression of ISM employment data* ch, 000s per month, sa Jun
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Disappointing Employment ReportThe June employment report was much weaker than expected. While disappointed, we caution
against simply extrapolating. Yes, there has clearly been some loss of momentum in the last few
months, but (a) we see good reasons to believe that the latest employment data exaggerate the
falloff in the underlying trend, and, (b) there are also good reasons to expect momentum to turn
more positive in coming months. That said, risks are rising that we will have to pare our growth
projections.
The 22,000 per month average gain in payrolls in the last two months is down sharply from a
179,000 per month pace in the first four months of the year. In our view, that degree of slowing is
much greater than seems consistent with other data, including the employment indexes in the
ISM surveys and jobless claims. We continue to expect the unemployment rate to trend lower in
the next year and a half, although we have adjusted our forecast slightly to reflect the higher than
expected level in June. We now forecast a decline to 8.8% at the end of 2011 (instead of 8.6%)
and 7.9% at the end of 2012 (instead of 7.8%). At 9.2%, the June reading was up from a low of
8.8% in March but still down from 9.6% in Q4 of 2010 and 10.0% in Q4 of 2009.
Preview: Lots of Key GDP Input Data + Fed Chairman TestimonyWe will review our estimate for Q2 real GDP growth after some key input data are reported in the
upcoming week—most important will be foreign trade, federal outlays, retail sales, and
inventories. Currently, risks appear firmly on the downside relative to our 3.5% estimate for the
annualized pace, although we expect many forecasters will have to raise their estimate in the
week ahead. The consensus appears to be close to 2.0%. Also of note: the PPI and CPI, claims,
industrial production, and the Michigan sentiment survey. The Fed chairman will testify.
The weakness in payrolls in the last two months has contrasted with the relative strength
being signaled by the ISM employment indexes.
* Based on a 2003-09 regression of payrolls changes vs. the two individual ISM employment indexes; the regressionimplies a 26% weight for manufacturing and 74% for nonmanufacturing.Note: Shaded bars represent periods of recession.Source: Bureau of Labor Statistics, Institute for Supply Management, and MF Global
The June employment report was much weaker than expected, in
contrast to somewhat stronger than expected data in the last few
weeks. Payrolls were reported up just 18,000 in June following a
downward-revised 25,000 increase in May (revised from 54,000).The unemployment rate rose to 9.2% from 9.1%. While
disappointed, we caution against simply extrapolating. Yes, there
has clearly been some loss of momentum in growth in the last few
months, but (a) we see good reasons to believe that the latest
employment data exaggerate the falloff in the underlying trend,
and, (b) there are also good reasons to expect momentum to turn
more positive in coming months. That said, risks are rising that
we will have to pare our growth projections.
How disappointing was the report? Along with weaker than
expected readings for payrolls and the unemployment rate, the
workweek fell 0.1 hour and average hourly earnings were only
flat. Moreover, the rise in the unemployment rate was despite a272,000 decline in the labor force; the household survey
employment measure was reported down 445,000.
Some Positive Signs, too
More positively, jobless claims fell in the latest week (to 418,000
from 432,000) and the employment component of the
nonmanufacturing ISM survey rose slightly in June from an
already high level in May (to 54.1 from 54.0), even as the overall
nonmanufacturing ISM index slipped (to 53.3 from 54.6). The
weakness in the employment report also contrasted with the
pickup signaled by the ADP survey (which showed a 157,000 rise
in private payrolls in June following 36,000 in May).
Within the employment report, one positive takeaway was the
pickup in hours worked and wage income in Q2 as a whole. The
index of private sector hours worked rose at a 3.3% annual rate in
Q2, up from a 2.0% pace in Q1 and 1.9% in all of 2010 (Q4/Q4).
Total private sector wage income (combining average hourly
earnings and hours worked) rose at a 5.1% annual rate in Q2, up
from a 4.3% pace in Q2 and 3.7% in 2010 (Q4/Q4).
Were the Last Two Months Really that Weak?
On that smoothed, quarterly average basis, the data look
encouraging. Unfortunately, the last two months alone look much
less encouraging, which raises the question: How much has the
trend deteriorated in the last couple of months?
The 22,000 per month average gain in payrolls in the last two
months is down sharply from a 179,000 per month pace in the
first four months of the year. In our view, that degree of slowing is
much greater than seems consistent with other data, including the
employment indexes in the ISM surveys (see bottom chart) and
jobless claims. (And also the ADP report.)
Bond yields continue to track the surprises in the economic data.
The weaker than expected employment report reversed some—
but not all—of the recent turnaround in the economic surprise
index calculated by Citigroup.
The economic surprise index, calculated by Citig roup, reflects “weighted standarddeviations of data surprises relative to Bloomberg median expectations in rollingthree-month windows”Source: Citigroup Global Markets Inc., Federal Reserve Board, and Bloomberg
The weakness in payrolls in the last two months has contrastedwith the relative strength being signaled by the ISM employmentindexes.
* Based on a 2003-09 regression of payrolls changes vs. the two individual ISMemployment indexes; the regression implies a 26% weight for manufacturing and
74% for nonmanufacturing.Note: Shaded bars represent periods of recession.Source: Bureau of Labor Statistics, Institute for Supply Management, and MFGlobal
Could the 179,000 average in the first four months of the year
have been exaggerated on the strong side, with some of the
slowing in the last two months reflecting “payback”? Yes, that is
possible. For the first half of 2011 as a whole, gains averaged
126,000 per month, which was a pickup from 103,000 per monthon average on an ex-census basis in the second half of 2010,
56,000 per month on that basis in the first half of 2010, and
-196,000 per month in the second half of 2009. In turn, we expect
the second half of 2011 to show clear further improvement
relative to the first half, consistent with the unemployment rate
continuing to trend lower.
The unemployment rate also showed improvement in the first half
of 2011 as a whole. At 9.2%, the June reading was up from a low
of 8.8% in March but still down from 9.6% in Q4 of 2010 and
10.0% in Q4 of 2009.
We expect the unemployment rate to continue to trend lower in
the next year and a half, although we have adjusted our forecast
slightly to reflect the higher than expected level in June. We now
forecast a decline to 8.8% at the end of 2011 (instead of 8.6%)
and 7.9% at the end of 2012 (instead of 7.8%).
Realization of our forecast will likely require a resumed downtrend
in jobless claims soon. At 425,000, the latest four-week average
in claims is down from as high as 440,000 in May, although it is
still up from just under 400,000 in February and March. Note,
though, that the level is still down from the 448,000 per week
average in the second half of 2010, when the trend in payrolls
gains was around 100,000 per month—consistent with the
weakness in payrolls in the last two months being exaggerated.
Payrolls gains averaged just 22,000 per month in May and June, down from a 179,000 per month average from Januarythrough April. The 126,000 average for first half of the year was still a pickup from a 103,000 per month (ex-census) average inthe second half of 2010 and 56,000 in the first half of 2010.
000s per month, sa unless noted 2010 10H1 10H2 11H1 11Q1 11Q2 APR 2011 MAY 2011 JUN 2011
PAYROLLS 78 110 47 126 166 87 217 25 18
CENSUS 2010 -1 54 -57 0 1 -1 -3 0 0
EX CENSUS 2010 80 56 103 126 165 88 220 25 18
GOVERNMENT -19 40 -79 -31 -26 -37 -24 -48 -39
EX CENSUS 2010 -18 -14 -22 -31 -27 -36 -21 -48 -39
The ISM employment indexes are also suggesting that the
weakness in payrolls in May and June was exaggerated. Based
on a regression of the data for most of the past decade, the June
readings were consistent with a trend in payrolls gains of more
than 200,000 per month. While those indexes also overstatedstrength in the first four months of the year, they were quite
accurate in 2010 and 2009.
Don’t Blame the Seasonal Factors
Note that we are not attributing the weakness in payrolls to overly
aggressive seasonal adjustment factors. (Apparently, some
analysts have used that explanation.) While the June seasonal
factors were more aggressive this year than last year, the change
reflected a later cutoff point for the sample (June 18 vs. June 12)
in a month when not seasonally adjusted private payrolls rise
sharply. (Specifically, applying last year’s seasonal factors would
have boosted total payrolls by 59,000 and private payrolls by
75,000.)
In short, we have no excuses for the much weaker than expected
report other than to say that the data can be volatile and even two
months combined are not necessarily representative of the trend.
Also, other data are starting to suggest that momentum is turning
positive again, with a number of factors that have contributed to
weakness now fading—including Japan supply chain effects, the
Q1 spike in oil prices, and some extreme weather. Among the
positive signs have been a decline in jobless claims from their
high, stronger-than-expected chain store sales in late June, and
strength in the equity market (even with a selloff after the
employment data).
The y/y pace in average hourly earnings appears to have
stabilized at just under 2%. It was 1.9% in June.
Source: Bureau of Labor Statistics
At 9.2%, the June reading for the unemployment rate was up
from 9.1% in May and 9.0% in April, but it was still down sharply
from 9.6%, on average, in Q4. The broader, and more cyclical,
U-6 underemployment series rose to 16.2% in June from 15.8%
in May; it is down from 16.9% in Q4.
*Along with the total unemployed from the official unemployment rate (reflectingindividuals who say they have looked for a job in the last month), the broader U-6rate also includes (1) all "marginally attached" potential workers, and (2) workersemployed “part time for economic reasons,” as a percent of the civilian labor forceplus all “marginally attached” potential workers. Marginally attached potentialworkers are persons who currently are neither working nor looking for work butindicate that they want and are available for a job and have looked for worksometime in the past year. Persons employed part time for economic reasons arethose who say they want and are available for full-time work but have to settle for apart-time schedule.Note: Shaded bars represent periods of recession.Source: Bureau of Labor Statistics and MF Global
At 425,000, the latest four-week average in claims is up from just
under 400,000 in February and March but down from 440,000 in
May and a 448,000 average in the second half of 2010 (when thetrend in payrolls gains was around 100,000 per month).
Source: Department of Labor
5.0
8.5
12.0
15.5
19.0
3
5
7
9
11
79 84 89 94 99 04 09 14
Un emp loyment rate (l) U-6* rate (r)
percent, bo th scales
Jun
270
370
470
570
670
Jan-08 Jul-08 Jan-09 Jul-09 Jan-10 Jul-10 Jan-11 Jul-114-week average Weekly
More broadly, in addition to an unwinding of some of the factors
that have contributed to slowing recently, part of our relative
optimism has reflected our assessment that the headwinds that
have been holding back the recovery are fading with time. We
continue to expect the net result will be more improvement in theunderlying trend, but, of course, we will continue to review the
details of our forecast as more data are reported.
The index of hours worked rose at a 3.3% annual rate in Q2, up
from a 2.0% pace in Q1. While productivity growth can fluctuate
significantly, the pattern in hours worked looks consistent with a
pickup in GDP growth in Q2 after what appears to have been
exaggerated weakness in Q1. More direct input data for
estimating Q2 GDP will be released in the week ahead.
Source: Bureau of Economic Analysis and Bureau of Labor Statistics
Most immediately, we will review our estimate for Q2 real GDP
growth after some key input data are reported in the upcoming
week. Currently, risks appear firmly on the downside relative to
our 3.5% estimate for the annualized pace, although we expect
many forecasters will have to raise their estimate in the weekahead. The consensus appears to be close to 2.0%.
The overall ISM indexes looked consistent with only about a
2.8% annual rate for real GDP growth in Q2, although the 1.9%
annual rate reported for GDP growth in Q1 was much weaker
than suggested by the ISM data—the ISM data looked
consistent with about a 5.0% pace in Q1. The pattern has raised
the potential for some payback in GDP for exaggerated
weakness in Q1.
*Economy-weighted average of manufacturing and non-manufacturing indexes(calculated by Haver Analytics)Note: Shaded bars represent periods of recessionSource: Institute for Supply Management, Haver Analytics, Bureau of EconomicAnalysis, and MF Global
The Conference Board’s CEO confidence index, which mainlytracks large businesses, probably slipped to a still-high level in
Q2, consistent with the pattern already reported in the Business
Roundtable survey. The Q1 reading was the highest since Q2 of
2004. In general business confidence has recovered much more
than consumer confidence—at least for large firms.
Business confidence has likely held up better than consumer
confidence in recent months—at least for large businesses. The
Q2 Conference Board CEO index will be reported on July 11.
Source: The Conference Board
NFIB SMALL BUSINESS SURVEY (TUE, JUL 12, 07:30)
sa MAR APR MAY JUN
OPTIMISM INDEX 91.9 91.2 90.9
HIRING PLANS (net %) 2 2 -1 3
CREDIT HARDER TO GET (net %) 8 9 10
Source: National Federation of Independent Business (NFIB)
The NFIB small business optimism index was historically low
even before the declines in the last few months. It probably
started to recover in June. A rise in the hiring plans component
has already been reported.
GDP has consistently shown less weakness than implied by the
NFIB index since the recovery began (see chart below). In part,
the divergence likely reflects larger credit crunch effects for small
businesses than large businesses, although we believe the NFIB
index has also overstated weakness in the small business sector
recently.
The relationship between GDP growth and the NFIB index has
broken down in the current cycle, with much less weakness in
growth than implied by the index.
Note: Shaded bars represent periods of recession.Source: National Federation of Independent Business (NFIB), Bureau of Economic
Analysis, and MF Global
WEEKLY STORE SALES (TUE, JUL 12, 07:45/08:55)
JUN 18 JUN 25 JUL 2 JUL
WEEKLY ICSC, %w/w, sa -0.7 2.9 1.5
WEEKLY ICSC, %y/y 2.2 3.0 3.5
REDBOOK, %y/y 4.2 2.5 5.2
APR MAY JUNJU
THRU
WEEKLY ICSC, %m/m, sa 2.5 -2.3 0.1
REDBOOK, %m/m, sa 1.3 -2.7 0.9
MONTHLY ICSC, %m/m, sa 0.3 -2.6 1.8
WEEKLY ICSC, %y/y 3.0 3.0 2.7
REDBOOK, %y/y 5.1 3.9 3.8
MONTHLY ICSC, %y/y 8.5 5.4 6.9
Note: monthly data are based on the retail industry's fiscal calendar, the fiscal monthof July ends on July 30.Source: International Council of Shopping Centers, Instinet, and MF Global
-10
-5
0
5
10
74
84
94
104
114
80 83 86 89 92 95 98 01 04 07 10 13
NFIB index (l) Real GDP (r)
ind ex, sa
Q1
May
%q/q, saar
20
36
52
68
84
100
116
22
31
40
49
58
67
76
02 03 04 05 06 07 08 09 10 11 12
Con ference Board: CEO confidence (l)Con ference Board: consumer expectations (r)
YEAR-AGO LEVEL -717.0 -799.7 -935.6 -1004.0 -1004.0 *Fiscal year starts in October.Source: US Treasury, Congressional Budget Office, Bloomberg, and MF Global
The June budget report will provide some important input data fo
adding up Q2 GDP; weakness in defense spending looked
exaggerated in Q1.
RETAIL SALES (THU, JUL 14, 08:30)
JUN EST
sa MAR APR MAY CONS MF
TOTAL (%m/m) 0.8 0.3 -0.2 x -0.2
EX AUTOS 1.2 0.5 0.3 x -0.1
EX AUTOS & GAS 0.8 0.3 0.3 x 0.5
EX AUT, BLDG MATS, & GAS 0.6 0.3 0.2 0.5
MOTOR VEHICLES & PARTS -1.4 -0.7 -2.9 -0.6
FURNITURE & FURNISHINGS 2.7 -1.2 -0.7
ELECTRONICS AND APPLIANCES 2.7 -1.2 -1.3
BLDG MATS, GARDEN EQUIP 3.0 0.5 1.2 0.5
FOOD & BEVERAGE 0.3 1.2 -0.5HEALTH & PERSONAL CARE 0.6 -1.0 0.8
GASOLINE STATIONS 3.8 1.4 0.3 -3.5
CLOTHING & ACCESSORY 1.0 0.1 0.2
SPORT, HOBBY, BOOKS, MUSIC 0.7 -0.3 -0.4
GENERAL MERCHANDISE 0.4 0.6 -0.1
MISCELLANEOUS -2.0 2.0 2.1
NONSTORE RETAILERS 0.4 1.3 1.2
FOOD & DRINKING PLACES 1.1 -0.8 0.6
TOTAL (%y/y) 7.5 7.3 7.7 7.9
EX AUTOS 6.9 6.8 8.2 8.
EX AUTOS & GAS 5.3 4.9 6.2 6.3
EX AUT, BLDG MATS, & GAS 5.3 5.7 6.1 6.2
Source: Bureau of the Census, Bloomberg, and MF Global
Retail sales likely fell again in June, dragged down by a decline in
auto sales and price-related weakness in the gasoline
component. (The data are nominal.) Excluding autos and gas,
sales probably picked up.
We estimate our June forecast is consistent with total real
consumption increasing by less than 1% at an annual rate in Q2,
a slowing from the 2.2% pace in Q1. We expect growth to re-
accelerate in Q3, helped by the boost to real spending power
from declining gasoline prices.
-12
-7
-2
3
8
13
18-8
-6
-4
-2
0
2
4
07 08 09 10 11
Non auto consumer goods import price (l)Core goods CPI (l)Real broad dollar index (r)