8 INVESTMENT AND SAVINGS

Table 8.3 Savings ratios

Householdsa		Nationalb

	1985 1990					1995	1998	1985	1990	1995	,ry7
Australia			7.3 6.8			1.3	2.5	18.9	17.3	17.6	18.4
Austria			8.3 13.7			13.4	8.6	23.2	26.3	22.4	23.p
Belgium			14.0 17.1			19.3	11.6	15.0	21.0	22.5	22.4
Canada			13.3 9.7			7.4	1.2	19.6	16.4	17.8	18.4
Denmark			... 11.4			5.4	5.8	14.9	17.8	20.4	20,9
France			14.0 12.5			14.3	14.1	18.9	21.5	19.8	20.0
Germany 11.4					13.8	11.6	11.0	22.0	24.9	20.6	20,7
Italy			18.9 18.2			13.1	11.5	21.6	19.6	20.6	20.4
Japan			15.6 12.1			13.4	13.6	31.7	34.6	30.8	31.1
Netherlands 0.1					5.8	1.8	-0.6	24.3	26.0	24.7	27.1
Spain			11.2 10.5			11.7	10.5	20.6	21.7	21.3	21.0
Sweden			2.3 -0.6			8.2	1.2	17.5	17.7	16.7	16.1
Switzerland 5.7					12.2	9.9	9.0	29.8	33.1	28.5	29.0
UK		10.7 8.1				10.2	7.8	17.6	14.3	14.3	14.9
USA			7.1 5.2			4.7	0.5	17.6	15.6	16.2	17.4
EU15			... ...			...	...	20.1	21.0	19.6	19.9
OECD			... ...			...	...	21.1	21.0	20.4	21.1

a Net household savings as % of disposable personal income.
b Gross national savings as % of GDP.

Source: DECD.

Figure 8.3

Gross national savings

% of GDP, 1997

31.1
	29.0
	27.7

23.022.4

21.1

U

40

30

zo

,o

0						J1
	ac		a	61 is	F (.O ac a~act avi cce J~h as aaa J5P ae

~aQ				aop ~1ac Pte`'` ~yA	04'	SQ 466 ~e~F ~ ya	~, PJy~ 	`ac	ys~

S`

Source: OECD.

21.0 20.9 20.7 20.4 20.019.9

INDUSTRY AND COMMERCE

"If economists were any good at business, they would be rich men
instead of advisers to rich men."
Kirk Kerkorian

Partly for historical reasons, there is an abundance of economic indicators relating to industry and commerce in general and manufacturing in particular. These provide useful, timely signals of activity, but it should be borne in mind that their coverage is relatively narrow.

Interpretation

Industrial and commercial indicators provide a guide to both the output and expenditure measures of GDP. These should be used in conjunction with cyclical indicators (page 51) to assess overall economic activity.

Output. The main clues to output include manufacturing and industrial production, car production and sales, construction spending, housing starts and wholesale and retail sales.

Table 9.1 shows the relative importance of the main sectors. Note that services account for 60-70% of output in the major industrialised countries, but they are relatively neglected among the commonly followed indicators of output. Employment (page 61), consumer spending (page 83) and government consumption (page 31) also provide clues to services.

It is "value added" that matters - the value of output less the cost of raw materials and other inputs. This is what manufacturlng and industrial production figures measure, but the other indicators are before the deduction of input costs.

Expenditure. The main indicators of industrial and commercial expenditure are fixed investment and investment in stocks (see Chapter 8).

Fixed investment figures are available only after a lag. This chapter will show that business conditions, construction spending









9		INDUSTRY AND COMMERCE

and housing starts provide a useful advance guide; so do man,,.
"'

facturers' production of capital goods and net imports of capitl goods. Manufacturing, wholesale and retail inventories provide clues to changes in total stocks.
	Investment figures should be used with consumer spending, go.. ernment consumption and exports and imports to assess total GDp on an expenditure basis. This chapter's briefs on retail and vehicle sales provide clues to consumer spending and imports. Manufac_ turers' orders from abroad are indicative of export demand.

Coverage

Many figures such as orders and productivity can be related to the whole economy, but for convenience are often produced for manufacturing alone. It is important to be clear about the coverage of any figures you use. Where they relate to manufacturing alone, consider whether other sectors of the economy might be moving differently.

Key indicators

The following briefs are arranged broadly from top to bottom: from the whole economy through to individual sectors, and from manufacturing through wholesaling to retailing.

Table 9.1 Output by sector, 1997

of GDP

Average annual % change,
1990-97

Agri- Mann-					Ind-	Serv-	Agri-	Ind-	Serv-
culture fact					ustry	ices	culture	ustry	ices

Australia			3	26	14	71	1.1	2.5	4.4
Austria			1	30	20	68	-0.7	1.3	2.2
Belgium			1	27	18	72	1.6	0.7	1.3
Canada			3	27	5	66	1.2	1.8	1.8
Denmark			4	27	19	69	1.7	1.9	1.4
France			2	26	19	72	0.4	0.1	1.8
Germany			1	...	24	44	0.8	.,.	2.5
Italy			3	31	20	66	1.3	0.8 -	1.1
Japan			2	38	24	60	-1.6	0.7	2.1
Netherlands				3	27	18	70	3.7	1.2	2.3
	Spain			3	...	18	25		-2.5	-0.4	-131
	Sweden			4	34	23	63		-1.9	-0.7	-0.1
	Switzerland				4	27	10	64	...	...	w
	UK			2	31	21	67
	USA			2	27	18	71		2.0	4.3	1.9

Sources: World Bank.

102

BUSINESS CONDITIONS; INDICES AND SURVEYS

Fire 9.1

structure of production
industrial countries' average, gb of GDP, 1997
3.

Sources of growth
Annual average % change, 1990-97

7.9

Agriculture			Industry	services

BUSINESS CONDITIONS; INDICES AND SURVEYS

Measures: Anecdotal evidence of business climate.
Significance: Valuable early warning of changes in economic cycle.
Presented as: Index number or percentage balance of companies
	optimistic or pessimistic.
Focus on:		Trends.
Yardstick:		Watch for indicators of rising or falling confidence. The US
	IvAPM index has a breakpoint of 50 (see text).
Released:		Monthly or quarterly, one month in arrears; rarely revised.

Overview

Surveys provide valuable evidence of perceptions and expectations relating to business conditions, usually in manufacturing. Responses are subjective but they give early signals of changes in economic trends. (See briefs on Orders, Inventories, Prices, Retail sales, and so on, for comments on individual parts of the surveys.)
	Private-sector bodies conduct surveys in some countries (for example, confederations of industry in Australia, Britain, Finland; FFo in Germany), but some government agencies do the work (for example, Statistics Canada, the Bank of Japan). Many surveys are quarterly, although where they are monthly some questions are asked only 3-4 tees a year (for example, capacity use in France and Germany).

Ainerica

The US National Association of Purchasing Managers (NAPM) Publishes monthly indices covering factors such as the state of
INDUSTRY AND COMMERCE

order books, inventories, production and prices, and a composite index of industrial conditions.
	A composite index reading above 50 indicates an expanding manufacturing sector, a figure below that indicates a contracting manufacturing sector. A reading below 44 is taken as a sign of a declining economy overall.
	Also useful are the US Federal Reserve's report on region trends (known as the tan or beige book) and the Conference Board's surveys of business conditions.

Britain

The Confederation of British Industry (cm) conducts several surveys. Its monthly and quarterly surveys of about 1,250 industrial companies provide guides to manufacturers' expectations, mostly for the four months ahead. Its monthly surveys of 500 distributors, conducted in association with the Financial Times, provide information about current and expected conditions in 15,000 retailing and wholesaling outlets.
	The CBI surveys are presented in percentage balance of companies reporting orders up less those reporting them down. The OECD uses a similar style. Interpretation is not as clear-cut as with the NAPM surveys, but trends provide a good indication of perceived conditions. The CBI confidence measure is reckoned to provide a guide to corporate earnings nine months in advance.
	British economists also watch the monthly indices published by the Chartered Institute of Purchasing and Supply (ciPS). The cms produces separate surveys of manufacturing and service companies. As with the NAPM index in America, a crns index above 50 indicates expanding output; below 50 implies contraction.

BUSINESS CONFIDENCE
According to Dun & Bradstreet's latest quarterly survey of sales expectations, business confidence continues to ebb around the world. The number of business people expecting sales to rise in the fourth quarter exceeded those expecting them to fall by 36 percentage points. The net balance of optimists is at its lowest for almost three years. Only in three of the countries shown - France, Germany and Australia - have businessmen become cheerier this quarter. In Taiwan, where optimists outnumbered pessimists by 31 points in the fist quarter, they now trail by 23 points. In Britain the net balance of optimists has shrunk from 56 to ii points.
The Economist, December 5th 1999

104

INDUSTRIAL. AND MANUFACTURING PRODUCTION

Sales expectations
Net % of businessmen expecting higher sales

1998:
a Qt

i Qz

6003
Q4

lop 		per

Spurt: Dun 8 Bradsheet.

vc a~i yc

teat		OevA ..~i

so

40

zo

0

20

'17 acanaNei.

INDUSTRIAL AND MANUFACTURING PRODUCTION

Measures: Value-added output of mines and manufacturing companies. Significance: Indicator of industrial activity.
Presented as: Index numbers in volume terms.
Focus on:		Trends in volume terms.
Yardstick:		OECD average manufacturing production grew by 1.2% a
	year during 1990-95. Industrial production grew by 1.3%
	during the same period but by 3.3% in I 995-98.
Released:		Monthly, at least one month in arrears; revised.

Coverage

There are two main series, which are usually released together.

Manufacturing production. This is the value-added output of manufacturing companies. Some countries such as Italy and Spain include mining and quarrying.
Industrial production. This is manufacturing production plus
the supply of energy and water, and the output of mines, oil
wells and quarries. It generally excludes agriculture, trade,
transport, finance arid all other services.

National coverage varies widely. Canada includes services relat111g to mineral extraction; Germany covers construction; Portugal leaves out clothing, furniture and printing; Sweden excludes utilities; Switzerland omits mining and quarrying.
	In total the index in Japan covers about 60/D of industry, Spain Found 70%, Germany and Italy 80/D, and the Netherlands, Su.oden and Britain up to 100~. France conducts a monthly sur-

105
9 INDUSTRY AND COMMERCE

vey covering 65% and a fuller quarterly survey covering gsy industry. Australia and Switzerland have quarterly indices only.

The importance of manufacturing and industry

Manufacturing. Broadly speaking, manufacturing ranges from about 20% of GDP in America to about 30% Japan (see Table 9.1) . Manufacturing is tiny

in Germany and

	in some African states, but it rises to 30-40% of GDP in developing countries such as Puerto Rico and Taiwan. Table 9.2 shows the importance of various manufacturing sectors across 15 countries.

Industry. The difference between manufacturing and industrial production accounts for about 10% of GDP on average. This takes total industrial production to about 30% of American GDP and about 40% of Japan's (see Table 9.1).
	The gap is more significant where the energy producing sector is large, but even OPEC members' industrial production rarely exceeds about half of GDP.

Interpretation

Industrial production as a whole is broadly indicative of the state of the economic cycle. The output of industries producing capital goods and consumer durables tends to be squeezed most during a downturn. Some countries produce separate figures for various sectors (see Motor vehicles, page 111). Steel production is a useful lead indicator since it is an input to many other industries.

Figure 9.2

Industrial production
Annual average % change, 1995-98

3.8		3.8
nn

1.2			3.1 	3.0

2-00		1.9

	a		a ~e .~5~
~0~oa ~~a ~`1 h~a ~ce 0.4

CAPACITY USE AND UTILISATION

Table 9.2 Structure of manufacturing, % of total, 1996

Food,		Machinery
beverages			Textiles 	& transport
& tobacco			& clothing equipment Chemicals	Other

Australiae				19
Austria				19	6	19
					8		49
			16				16	8	42
	Belgium						18	7	7	15	53
	Canada						14	5	30	9	42
	Denmark						23	4	23	11	38
	France						13	12	30	52	42
	GermanYa						10	4	41	12	33
c							Italya	8	13	34	8	37
11
	Japan						9	4	39	11	37
	Netherlands						24	3	25	14	35
	Spain						18	7	27	11	38
	Sweden						9	1	34	11	44
'							Switzerlanda	10	7	31	9	42
		14					5	31	13	37
	USA						12	5	33	12	38
c							a 1991.

Source: World Bank.

CAPACITY USE AND UTILISATION

Measures: Extent to which plant and machinery is in use. Significance: Indicator of output and inflationary pressures. Presented as: Percentage of total capacity.
Focus on:		Absolute level and trends.
Yardstick:		80-905yo; more may be inflationary, less indicates room for
Released:		growth.
	Monthly, I-3 months in arrears; revised.
TOota1 Overview

which pacify. Capacity is a vague, hard to measure concept
	varies over time and according to economic conditions.
The term is generally used to refer to the sustainable maximum
output that could be produced by existing (installed) manufactur-
	plant and machinery, although sometimes other factors such as labour are taken into account. Either way "normal capacity" wiu vary if working practices are changed, even if only by increasIng the Working week by one hour. (See Productivity, page 49).
9 INDUSTRY AND COMMERCE

Capacity use. Capacity use is usually defined as output divided by sustainable maximum capacity. Sustainable maximum output is lower than temporarily attainable peak production, so use of (sustainable) capacity can and occasionally does exceed 100
some industries.
	Overall capacity use does not go as high as 100/o because different companies reach their peaks at different stages of the economic cycle. Also bottlenecks in one industry restrict supplies and therefore output in another: in major cyclical peaks, a shortage of metals can limit output of consumer durables and business machinery, and restrain capacity utilisation in those industries.

Capital investment. Figures for capital investment are sometimes taken as indicators of changes in capacity, but investment is usually measured gross, with no allowance for scrapping. Even where allowance is made, the net capital stock may not be representative of output potential. Moreover, the increase in capacity derived from a unit of new capital is affected by factors such as new technology, an increasingly shorter life for capital equipment, a longer work week for capital, restructuring of industry and the closure of plants.

Interpretation

Strong economic growth with high capacity use suggests inflationary pressures. However, if demand is expected to remain buoyant and if interest rates are low, producers may invest in new plant and machinery.
	If the economy is expanding, but there is low capacity use and no evidence of recent new investment, it may be that the economy is recovering from recession.

American capacity utilisation

The US Federal Reserve is one of the few national official bodies to produce estimates of capacity use and its figures are undoubtedly the most detailed and consistent available, but not even the Fed conducts surveys of capacity or capacity utilisation. Instead it uses data from the Bureau of the Census and various trade orgar
-

sations to estimate capacity in various industries. Overall capacity utilisation is industrial production divided by the capacity index.
	The Fed defines capacity as the realistically sustainable m~ mum capacity, rather than unsustainable short-term peak capacity,
	American capacity use exceeded 90% during the Korean and Vietnam wars, but it has been below 90% since 1967. A figure Of 88% overall would indicate severe strains (but 88% would be

108

MANUFACTURING ORDERS

normal in certain individual industries which operate a high level of capacity use, such as paper and pulp).

Surveys

In most other countries measures of capacity use are derived from survey evidence (see Business conditions, page 103). For example, the cBt s monthly survey asks British manufacturers if their capacity is satisfactory in relation to orders. Defining "capacity" and "satisfactory" is left to respondents. About half include factors other than physical capacity and the interpretation of satisfactory has changed over time.
	Accordingly, survey results are better for examining short-term rather than long-term changes in capacity use.

MANUFACTURING ORDERS

Measures: New orders received by manufacturing companies. Significance: Indicator of output in the near term. Presented as: Money, index or percentage balance form.
Focus on:		Trends.
Yardstick:		Total orders can change by 10-20%. Generally the higher
	the better, but watch for bottlenecks and inflation.
Released:		Monthly, I-2 months in arrears; revised.

Overview

The US Commerce Department publishes new orders received by manufacturing companies and their unfilled orders in current dollars, with a breakdown between orders for durable and non-durable goods.
The German Federal Statistical Office publishes volume index numbers for manufacturing orders, distinguishing between domestic and foreign orders.
The Confederation of British Industry publishes survey results indicating the percentage of respondent companies which think that order books are satisfactory less the percentage of companies reporting unsatisfactory order books. This subjective assessment of total and export orders is broken down into value and volume terms.

Manufacturing orders (including machine tool orders) ripple through the economy. An order for a washing machine may prompt an order for a metal pressing which in turn will provoke order for sheet steel. Each order will reflect the output price,

109
9 INDUSTRY AND COMMERCE

while manufacturing output and GDP will rise by the value-added component only. Orders are therefore much more volatile than manufacturing production.

Total orders

Buoyant order books indicate upward pressure on employment and output over the next few months. This is basically good, but it may suggest a rise in inflation if unemployment is low; capacity use is high; there is an order backlog (orders are high in relation to shipments or sales); and/or inventories are low. Strong orders will also tend to increase imports of materials and intermediate goods, but this may be offset by exports (see below).
	Orders provide an early signal of changes in the economic cycle. A rise in orders may indicate the end of a recession; a fall may indicate that the cycle is peaking.
	Where orders are presented in value terms they should be adjusted for inflation. Producer prices can be used. If orders rise by 6% in value and producer prices increase by 4%, the volume of orders is about 2% higher. It is important to ascertain that this adjustment does not reflect, say, a movement in refined oil prices which are erratic and can falsely suggest changes in the volume of new orders. There may also be other blips. In the American series the transport component jumps by perhaps 40% in a month when new aircraft are ordered. Try to use a series which excludes misleading influences.

Domestic orders

Domestic orders by sector are indicative of the structure of home demand. Bigger orders for capital goods suggest more investment activity and more output in the future. Machine tool orders sometimes receive special attention: an increase is encouraging because they are used to manufacture yet more machines which in turn make more goods.
	Defence orders tend to reflect political decision-making. they are identified separately, as they are in America, subtract them from capital or durable orders to get a better feel for underlying demand.
	Orders for durable and capital goods react to changes in the economic cycle ahead of other orders. Orders for consumer goods are obviously indicative of trends in the consumer sector.

Export orders

Orders from overseas reflect export competitiveness (see page

MOTOR VEHICLES

154) and the economic cycle in overseas markets. A rise is good, but see comments above about inflation. If orders are booked despite a stronger currency, profits may be squeezed and repeat business may not be sustained.

1VIOTOR VEHICLES
Measures: Activity involving cars and tn.icks.
Significance:		Indicator of manufacturing production as well as consumer
	demand.
Presented as: Number of units.
Focus on:		Trends.
Yardstick:		Volatile, but a 12-month change of more than a few
	percentage points is worrying.
Released:		Monthly; more frequently in America.

Overview

There are three main series relating to motor vehicles:

 manufacturers' production;
 sales; and
 registrations of new vehicles with national licensing authorities.

Registrations and sales are broadly similar; the difference mainly reflects the method of data collection.
	The figures usually relate to the number of units, that is, volume rather than value. They may not reveal very much about the pattern of demand for cheap or expensive vehicles or about changes in quality. However, they are a useful indicator of manufacturing production; demand for a durable good which is vulnerable to the economic cycle; competitive pressures, especially between domestic and overseas producers; and import and export trends.

Interpretation

Vehicle sales are a reasonable leading indicator of economic activity,. A vehicle purchased by an individual is classed as consumption expenditure. The same vehicle purchased by a business is investment spending. Generally, then, figures for cars are suggestive o f consumption; light vans and trucks are indicative of
vestment.

	Production and sales may not tally due to changes in stocks (see Wholesale sales and stocks and Retail sales and stocks, pages 116 and 117) and net exports.
9 INDUSTRY AND COMMERCE

Seasonality. There is a marked seasonal pattern in car sales, with turnover bunched into January and, if different, the start of a new model year or registration plate year identifier. Discounting price wars can also shift demand into a different month from normal. Short-term figures should be interpreted with care. Take 2_3 months together and compare with similar periods of earlier years.

Table 9.3 Motor vehicle markets, 1998 '000

Australia
Austria
Belgium
Canada
Denmark
France
Germany
Italy
Japan
Netherlands
Spain
Sweden
Switzerland
UK
USA

Source: EIU.

Production		New registrations

310 nil 317
1,086 nil 2,600 5,348 1,378 8,056 255
1,940 325
	nil
1,752
6,077

CONSTRUCTION ORDERS AND OUTPUT

488 296 452 673 162 125
3,736
2,364 46,868 543
1,192 253 295
2,247
6,333

Measures: 		Activity in the construction sector.
Significance:		Indicator of new investment and future output
Presented as: Value of orders, volume of construction.
Focus on:		Trends in volume terms.
Yardstick:		In OECD countries, the growth rate of capital formation in
	non-residential construction averaged 0. I % a year betwe~t
	1990 and 1995.
Released:		Monthly, at least one month in arrears.

Overview

There are several main series, including the following.

		Orders (volume in Britain; value in Japan and America).
		Permits issued (number in Belgium and France; value 'n
	Australia, Canada, Germany, the Netherlands).

CONSTRUCTION ORDERS AND OUTPUT

The value of work put in place (Germany, America).
Value-added in money and real terms (with GDP figures for
expenditure on investment for many countries).

Construction covers buildings and infrastructure (such as roads ad pons). Orders are sometimes based on contracting work. This may include projects such as oil rigs which really belong in manufacturing production under steel fabrication. The figures usually exclude residential dwellings, but this should be checked.

Significance

Construction work is fixed investment. This boosts current period GDP and lays the basis for future economic growth. Its share of GDP in different countries is shown in Table 8.2.
	New factories and offices provide a direct foundation for higher economic output. New infrastructure improves social welfare and generally boosts productivity. The only real exception to the "future output" rule is investment in new dwellings, but this still brings benefits (see Housing, page 114).

Interpretation

Seasons and cycles. Construction work is highly seasonal and cyclical. Data are frequently (but not always) seasonally adjusted, but it is wise to take 2-3 months together and compare them with the same periods in earlier years. It is important to look out for the adverse effects of a wet or freezing month.
	Construction activity is sensitive to expectations of future demand and to interest rates. Low interest rates increase the return from investment and encourage capital spending, especiallY when they are coupled with strong demand and high usage of existing capacity. (See also Investment, page 90.)

Orders. Construction orders signal demand for building materials and labour over the coming months (and years - depending n the size of the individual projects). Knock-on effects include 'plications for service industries such as architects and surveyors, manufacturers of industrial plant to go into new factories, and providers of office fittings and furnishings.
	Orders are often given on a value basis. These should be adjusted for inflation to arrive at the volume change. For example, if the value of orders rises by 10% and inflation is 8%, real growth is about 2%. Producer prices might be used for deflation if there is no obviously better indicator to hand. Figures sometirnes cover construction permits; these provide no guide to the
9 INDUSTRY AND COMMERCE

size of the projects, and they might not be translated into

structlon activity if economic conditions change. 		c 011.

Output. Construction output in volume terms helps you to judge the effect on total output. Construction accounts for about 5% of GDP, so a 10% rise in construction value-added contributes around 0.5% t0 GDP.
	A large construction project spread over several months or years adds a little bit to output in each of several periods. (See above for knock-on effects.)

HOUSING STARTS, COMPLETIONS AND SALES

Measures:		Number of new houses begun and finished; sales of new
	and existing homes.
Significance:		Indicator of construction activity; industrial and consumer
	demand.
Presented as: Number of units per month.
Focus on:		Trends.
Yardstick:		Volatile; number of starts can change by 40% a year. In OECD
	countries, the growth rate of gross capital formation in resi
	dential construction averaged 1.3% between 1990 and 1995.
Released:		Monthly, at least one month in arrears.

Overview

There are three main series indicating:

 the number of residential dwellings started;
 the number of residential dwellings completed; and		`
 the number of residential dwellings sold.

Each series applies to a given period, usually one month. They are often seasonally adjusted, but it is wise to take 2-3 months together and compare with the same periods in earlier years since house building is highly sensitive to the weather.
	Figures usually distinguish between public and private dwellings. Activity in the private sector is a good guide to underlying activity, but there are also knock-on effects fii public-sector building.

Starts

A housing start is counted on the date that foundations begun (not when the site is cleared). It implies a given 1evej

HOUSING STARTS, COMPLETIONS AND SALES

demand for construction materials and labour over the next few months and a housing completion at the end of that period.
If the existing housing stock is old, there may be an element of
placement building. Generally, however, housing starts (and completions) are closely linked to population growth rates, earnings and employment, and interest rates.

Completions

A housing completion implies a house sale, a new mortgage advance and increased consumer demand for carpets, furnishings and other durables - possibly accompanied by extra consumer credit.

Sales

Housing sales are linked to the level of completions, but turnover of existing homes is more important. House prices are highly relevant. People are more inclined to move home and buy rather than rent when house prices are rising and are expected to provide capital gains. Housing turnover is also stimulated by incomes rising relative to house prices and by lower mortgage rates which encourage borrowing.

Effect on GDP

The construction of new houses is classed as fixed investment in residential dwellings. This accounts for only a few percentage Points of GDP (see Table 8.2), so arithmetically a 10% rise in house building may add less than 0.5% to total output.
The sale of existing houses is a transfer of production scored m earlier periods and does not itself add to output, but real estate, legal and financial fees and commissions do, as does demand for new furnishing and other durables.

HOUSE-BUILDING
New homes are popping up everywhere in the rich world. After a slump in the fourth quarter of last year, housing starts in America, Britain, Germany and Japan have risen rapidly in the past three quarters. The biggest housing boom has been in Britain: in the third quarter of this year there were 11.4% more housing starts than in the same period of 1995. Residential building has also had a boost in Germany and Japan, despite the two countries' sluggish economic growth so far this year. In Germany new orders for homes have risen by 10.8% in the past year; during the same period,
INDUSTRY AND COMMERCE

housing starts in Japan have risen 9%. Housing starts have grown least in America, rising by only 4.5% in the year to the third quarter.

The Economist, December 7th 1996

Housing starts		A~

First quarter 1992=100

Germany*		ISO
ermany*

,.o

Britain
	Japant
120

o

united statee		,an

	_.							r	...	~	...	~	..

	1992						93	91	95	96
		Naw orders reaefvad for residential building.
Source: Datastream.						trout-quarter moving average.

,30

WHOLESALE SALES OR TURNOVER, ORDERS AND
STOCKS

Measures: Most common indicator measures sales by wholesalers. `' Significance: Indicator of demand.
Presented as:		Monthly index numbers.
Focus on: 		Rates of change in volume of sales.
Yardstick:		More volatile than retail sales; look for gains of 3-4%~ a
	year.
Released: 		Monthly, I-2 months in arrears; revised.

Overview

Wholesale sales (called wholesale turnover in Germany) are 20 important link in the supply chain. Wholesalers channel import5 and domestically produced or processed goods through to fIrw users. Where stocks and orders are available, these provide a useful check on trends. A fall in wholesale sales or a rise in wholesale inventories suggests or confirms slack in business and red demand.

RETAIL SALES OR TURNOVER, ORDERS AND STOCKS

	mere are relatively few figures on the services sector, but wholesale sales provide extremely loose indicators of demand from, for example, hotels and catering establishments. They are also indicative of demand for business goods including, in some countries, building materials. Wholesale sales are a reasonable signal of consumer demand (but retail sales are better - see below).

Value and volume

It is important to watch the volume of sales, particularly for durable goods (an early indicator of demand pressures). Where figures are in value terms they can be converted to volume using wholesale or producer prices. For example, if producer prices increase by 3% and sales value rises by 5%, volume is up by about 2%.

Effect on GDP

Wholesale sales are included in the distributive trades sector of GDP- Their direct contribution is wholesalers' value added (income from sales less the cost of purchases and other inputs).

RETAIL SALES OR TURNOVER, ORDERS AND
STOCKS

Measures: Most commonly sales by retailers. Significance: Indicator of consumer demand. Presented as: Monthly index numbers.
Focus on:		Rates of change in volume of sales.
Yardstick:		3% a year is reasonable; any lower and the economy might
	slow down. More than 4-5% suggests overheating.
Released:		Monthly, at least one month in arrears.

Coverage

Retail sales figures provide an important and timely indicator of spending at retail outlets. The series is known as retail turnover in Germany department store sales (with more limited coverage) in Italy and Japan; and hypermarket sales in Spain. French figures Cover only department and chain stores, mail order firms and hypermarkets. Sweden's series excludes alcohol, pharmaceuticals, c Mostd petrol.
to		series include vAT or sales taxes and many are in volume
gW s' ~	but the Dutch, Japanese, Italian, Spanish, Swedish and

are in value terms only (that is, before adjustment
9 INDUSTRY AND COMMERCE

for inflation).
	Headline series in North America are also in value terms b~ in dollars. Canadian retail sales figures included federal sales tax up to January 1991 but have since excluded the goods and ser. vices tax. Australia produces a monthly value and a quarterly volume index; both in dollars.

Basis. Information is presented by type of business, rather thby commodity. For example, food sales appear under at least t~,o subheadings: food retailers and mixed businesses. Where an establishment sells items outside its main classification, such as sales of food by a petrol filling station, these are usually excluded from the statistics. Credit sales are treated as a sale, valued at the date of the transaction at the total credit price including charges levied by the retailer.

Interpretation

Retail sales cover up to half of total consumer spending although there is not a direct correlation between the two since some items sold by retailers are bought by businesses. Nevertheless, retail sales are a key indicator of consumer confidence and demand.
	It is important to focus on volume increases. Where figures are in value terms they can be converted to volume using consumer prices. Arithmetically, a 1 % rise in retail sales adds roughly 0.3% to GDP, all else being constant.

Seasonality. Retail sales data are usually seasonally adjusted, but they should be interpreted with care. Promotional sales price discounting, warm weather or an expectation of increases in sales taxes can encourage consumers to bring forward their spending. An upward blip may not be sustained.

Cyclical variations. In times of financial stress consumers cut back on non-essential spending, which results in a decline or less rapid growth in retail sales. Spending on durables (items with a life of over one year, such as washing machines) goes first. For example, in Britain during the 1991-92 recession, the volume of food sales rose by 3% while new car registrations fell by 20%.
	Although retail sales are very nearly coincident with GDP, retail sales figures are published with a much shorter lag. They therefore provide an early indication of economic trends.
	A downturn in retail sales could lead to lower wholesale sales' slacker factory orders, an accumulation of stocks and, eventually' a cut in production.

RETAIL SALES OR TURNOVER, ORDERS AND STOCKS

Stocks and orders. Where figures for retail stocks and orders are also available they provide useful advance warning. For example, excess retail demand might show up first in a fall in the trio of retailers' inventories to sales. This spells trouble if wholesalers' and manufacturers' stocks are also low and there are constraints on manufacturers ability to increase output (high capacity use and low unemployment). Look out for price inflation and a surge in imports.
