Interpreting PMI data

Exploring the inter-relationships of selected economic indexes from the PMI surveys

Confidential | Copyright © 2017 IHS Markit Ltd

Interpreting PMI data

The boom-bust cycle

It helps to first consider a typical, albeit extremely simplified, pattern of events in a classic “boom-
bust” business cycle. During a period of economic expansion it is typical for employment to rise (and
unemployment to fall) and the demand for raw materials to increase. If employment and demand for
raw materials rise at suitably fast rates it is then common for skill shortages and supply-chain
bottlenecks to develop.
When demand exceeds supply, prices tend to rise. Wages and salaries and raw material prices will
therefore begin to increase. Retail price inflation may then pick up as higher costs are passed on to
the consumer.
The standard economic policy prescription for rising high street inflation is an increase in central bank
base rates which, by raising borrowing costs to business and the consumer, restrains demand. Prices
and economic growth then tend to grow at slower rates. When demand has slowed sufficiently,
interest rates may be lowered again, thus stimulating economic growth.

Diagram 1: Boom-bust cycle

Economic
Interest
growth
rates fall
stimulated

Increased
employment
Economic
and demand
growth slows
for raw
materials

Skill
shortages
Interest
and supply
rates rise
problems
develop

Inflation Staff wages

picks up as and
higher costs raw material
are passed prices rise
to customers

Confidential | Copyright © 2017 IHS Markit Ltd ` |2

Interpreting PMI data

PMI survey variables

The PMI surveys provide time-series variables relevant to a number of important stages in the above
cycle, allowing analysts to ascertain the pace of economic growth, to see whether demand and
supply imbalances are taking hold, and to see if prices are consequently rising. Most crucial is the
fact that the PMI survey variables are available well in advance of comparable official data.
Several of the indicators are directly comparable for both manufacturing and services. For example,
“business activity” in services is the direct equivalent of manufacturing “output” and “incoming new
business” for the service sector is the direct equivalent of manufacturing “new orders”. Some indexes,
on the other hand, are not produced for both sectors.
Many of these variables will of course be of importance not just to economic policy makers but also to
those monitoring corporate performance and profitability. The behaviour of the variables is analysed
on the following pages, with examples given for the eurozone manufacturing sector.

Manufacturing Services
 Output  Business Activity

 New Orders  Incoming New Business

 Employment  Employment

 Input Costs  Input Costs

 Output Prices  Prices Charged

 Backlogs of Work  Business Outstanding

 Export Orders  Business Expectations

 Quantity of Purchases

 Suppliers’ Delivery Times

 Stocks of Purchases

 Stocks of Finished Goods

 Future Output

Confidential | Copyright © 2017 IHS Markit Ltd ` |3

Interpreting PMI data

Diagram 2: Boom-bust cycle with PMI survey variables identified

New Orders /
New Business

Output / Activity

Economic
Interest
growth
rates fall
stimulated
Employment

Increased
employment
Economic
and demand
growth slows
for raw
materials

Quantity of
Purchases

Skill
shortages
Interest
and supply
rates rise
problems Backlogs of
develop
Work / Business
Outstanding

Inflation Staff wages

picks up as and
higher costs raw material
are passed prices rise Suppliers'
to customers
Delivery Times
Output Prices /
Charges
Input Costs

Confidential | Copyright © 2017 IHS Markit Ltd ` |4

Interpreting PMI data

Output, new orders & backlogs

Changes in new orders generally drive growth of economic output. PMI data can also help identify
deviations in this relationship, such as leads and lags. These often occur due to delays in the
adjustment of production to demand, especially in the manufacturing sector.
In the service sector, output is measured by business activity and new orders are measured through
changes in incoming new business.

Chart 1: Eurozone Manufacturing

sa, 50 = no change on previous month
65

60

55

50

45

40

35
Output New Orders
30

25
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017
Source: IHS Markit.

Backlogs of work/business outstanding vary according to the amount of new orders/incoming new
business received by manufacturing/service sector companies. When new orders/incoming new
business rises, supply imbalances can develop and backlogs of work/business outstanding
accumulates.

Chart 2: Eurozone Manufacturing

sa, 50 = no change on previous month

65

60

55

50

45

40

35
New Orders Backlogs of Work
30

25
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

Source: IHS Markit.

Confidential | Copyright © 2017 IHS Markit Ltd ` |5

Interpreting PMI data

New orders, purchasing and supply chains

The amount of goods bought by manufacturers for use in the production process varies directly with
the volume of incoming new business.

Chart 3: Eurozone Manufacturing

sa, 50 = no change on previous month

65

60

55

50

45

40

35
New Orders Quantity of Purchases
30

25
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017
Source: IHS Markit.

Supplier performance varies according to the amount of goods bought by manufacturers for use in
the production process. When demand for inputs rises, capacity constraints develop and delivery
times lengthen (the delivery times index falls below 50), resulting in an inverse relationship between
purchasing and speed of supplier delivery.

Chart 4: Eurozone Manufacturing

sa, 50 = no change on previous month

65

60

55

50

45

40

35

30
Quantity of Purchases Suppliers' Delivery Times
25
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

Source: IHS Markit.

Confidential | Copyright © 2017 IHS Markit Ltd ` |6

Interpreting PMI data

Supply chains and input prices

When suppliers become busier and bottlenecks arise, demand exceeds supply and a “seller’s
market” results, driving up input prices.

Chart 5: Eurozone Manufacturing

sa, 50 = no change on previous month sa, 50 = no change on previous month

30 95

35 85

75
40
65
45
55
50
45
55
35
60 25
Suppliers' Delivery Times (LH) Input Prices (RH)
65 15
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017
Source: IHS Markit.

A direct relationship therefore exists between manufacturing output and raw material (input) prices.
However, changes in input prices will lag changes in output.

Chart 6: Eurozone Manufacturing

sa, 50 = no change on previous month sa, 50 = no change on previous month

80 90
Output (LH) Input Prices (RH)
80
70
70
60 60

50 50

40
40
30

30 20
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

Source: IHS Markit.

Confidential | Copyright © 2017 IHS Markit Ltd ` |7

Interpreting PMI data

Input prices and output prices

A close relationship exists between input prices and output charges, as companies tend to raise the
prices charged for their goods/services when the average cost of their inputs increases.
The differential between output charge inflation and input price inflation can provide information on
firms’ ability to pass on higher costs to clients (pricing power).
The differential between output charge inflation and input cost inflation is also a ‘barometer’ of
pressure on profit margins in the manufacturing/service sectors.

Chart 7: Eurozone Manufacturing

sa, 50 = no change on previous month sa, 50 = no change on previous month

70 90

80
60
70

60
50
50

40
40

Output Prices (LH) Input Prices (RH) 30

30 20
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017
Source: IHS Markit.

Confidential | Copyright © 2017 IHS Markit Ltd ` |8

Interpreting PMI data

Stocks of finished goods

Changes in stocks of finished goods lag changes in output…

Chart 8: Eurozone Manufacturing

sa, 50 = no change on previous month sa, 50 = no change on previous month

65 52

60 50
55
48
50
46
45
44
40
Output (LH) Stocks of Finished Goods (RH)
35 42

30 40
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017
Source: IHS Markit.

…with output driven by changes in order books less changes in stock levels.

Chart 9: Eurozone Manufacturing

sa, 50 = no change on previous month Difference

65 20
15
60
10
55 5

50 0
-5
45 -10
40 -15
Output (LH) -20
35
-25
New orders minus stocks of finished goods (RH)
30 -30
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

Source: IHS Markit.

Confidential | Copyright © 2017 IHS Markit Ltd ` |9

Interpreting PMI data

Employment
Employment directly correlates with changes in output (or business activity in the service sector). Any
divergences in the relationship provide important information on capital:labour intensity and
productivity growth.
Over time, output will tend to grow at a faster pace than employment as industry becomes
increasingly capital intensive and reduces hours worked per unit of output. If employment grows at a
faster pace than output, productivity will decrease.

Chart 10: Eurozone Manufacturing

sa, 50 = no change on previous month sa, 50 = no change on previous month

65 60

60
55
55
50
50

45 45

40
40
35
Output (LH) Employment (RH) 35
30

25 30
1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017
Source: IHS Markit.

New orders and future output

New orders and output expectations are closely correlated. If new order growth strengthens, firms are
likely to become more optimistic regarding projected output levels, and vice versa.

Chart 11: Eurozone Manufacturing

sa, 50 = no change on previous month 50 = no change over next 12 months

60 75

70
55
65

60
50
55

50
45
New orders (LH) Future Output (RH) 45

40 40
2012

2013

2014

2015

2016

2017

Source: IHS Markit.

Confidential | Copyright © 2017 IHS Markit Ltd ` | 10