12 MONEY AND FINANCIAL MARKETS

EQUITIES

Last year was a rollercoaster for investors everywhere; but whereas most rich-country stockmarkets ended up ahead, many emerging markets fell heavily. Having been the world's best-performing stockmarket in 1997, Russia was the worst in 1998, losing more than 959'0 of its value in dollar terms. As investors shunned emerging markets, Latin America's stockmarkets also did badly: Venezuela dropped by 50%, while Brazil, Mexico and Argentina all fell by more than 35%. As their financial crisis stabilised, some Asian stockmarkets bounced back. Thailand, for instance, was up over 25%, while South Korea topped the league, more than doubling in dollar terms.

l.e nconomwt, January ytn

178

Stockmarkets

% change Dec 31st 1997-Dec 31st 1998 f terms

goo					so	-	o 	so

------------ - ---------------------

South Korea Greece Finland Belgium

Italy Spain Netherlands

France Germany Thailand Switzerland

United States Britain Sweden

Japan Australia

China Singapore Hong Kong Czech Republic Canada Poland Taiwan India

South Africa Malaysia

Chile

Indonesia Argentina Mexico

Brazil

Venezuela Pakistan

Turkey Russia

13

PRICES AND WAGES

"Inflation means that your money won't buy as much today as
it did when you didn't have any. "
Anon

Price indicators tell you about inflation. An increase in the general level of prices is nothing new: records from the days of the Roman Empire show rapid inflation. Since 1946 Britain's consumer prices have risen every year, but in fact inflation - in the sense of continuously rising prices - is historically the exception not the rule. Linking together various price series (of admittedly varying quality) suggests that in 1914, on the eve of the first world war, British consumer prices were no higher than during the 1660s. During those 250 years periods of rising prices were

interspersed with periods of falling prices.

Inflation has three main adverse effects.

First, it blurs relative

:n changes in gel. This disreduces ecover perfectly s investment. s to borrow

should










13 PRICES AND WAGES

Causes of inflation

There are two main theories about the causes of inflation; sup_ ply-shock and demand-pull. The reality is probably a complex mixture of the two.

Supply-shock (or cost-push). Prices are pushed up by higher wage and raw materials costs; perhaps owing to trade union power, dearer imports as a result of a weak currency, or a jump in commodity prices.

Demand-pull. Prices are pulled up when spending power (demand) is greater than the availability of goods and services. Factors which can boost aggregate demand include tax cuts, higher government spending, wage rises caused by labour shortages and an increase in consumer borrowing.

Recent experiences

Experiences with inflation range from deflation (a fall in prices experienced, for example, during the 1930s depression and by some oil-producing countries in the mid-1980s) to hyperinflation (such as when German wholesale prices rose by about 1.5 trillion
between 1919 and 1923). The most prominent recent example of deflation is Japan, where the consumer price index fell in the late 1990s. Hyperinflation is frequently associated with rapid increases in the money supply (see Chapter 12).

Figure 13.1

Inflation in industrial countries

Consumer spending deflator, annual average % change

USA
Japan
EU15
Canada
Australia

10

8

s

4		II
2
				`a.''"lk,
..,w.
	011				~	....		....	,	,.

	1985		1986 1987 1988 1989 1990 1991 1997 1993 1994 11995 1996 1997 1998
Source: OECD.

180

RECENT EXPERIENCES

industrial countries. Inflation was moderate in the industrial world in the 1960s, averaging about 3/D a year. It jumped sharply after the two oil price shocks in the 1970s before falling again in the 1980s. In the fight to tame inflation, wage and price controls have generally given way to tight monetary and fiscal policies. Inflation has generally stayed low in industrial countries during the 1990s, thanks to a combination of weak commodity prices and cautious monetary policies.

Developing countries. Inflation generally accelerated in the developing countries in the 1980s, with inflation reaching over 1,000% in some Latin American countries. The early 1990s saw inflation surge in eastern Europe and the former Soviet Union as ex-communist countries struggled to adjust to a more market economy. Countries in the former Yugoslavia also experienced high inflation after the break-up of the federation.
	Since the mid 1990s inflation has moderated in developing countries, falling from 51.8% in 1994 to 10.4% in 1998. Inflation slowed especially dramatically in Latin America - in Brazil, it dropped from over 2,000% in 1993 and 1994 to 3.5/D in 1998 and in central and eastern Europe and the former Soviet Union.

INFLATION

Consumer-price inflation in the euro area is at its lowest for half a century. Prices rose by only 0.8% in the year to Febru-

ary. America's inflation rate is 1.69'0, and prices are falling in Japan. But inflation rates vary across the euro area. According to Eurostat's harmonised measures, German inflation is

Consumer prices

6 change on a year earlier

Euro-11

		s
- 			United States

Japan  '-,v

m
	8
1980 82 84 88 88 80 92		94 9698 99

Harmonised price indices

96 change on a year earlier. Feb 1999
	0					0.5	1.0	1.5	7.0 7.5 3.0
Portugal ----------------- - ----------
		Ireland
Netherlands
		Spain I
		Italy I
	Belgium
		Finland
	Euro-11
Luxembourg
		Austria
		France
	Germany
13 PRICES AND WAGES

only 0.1% and France's 03%, compared with 1.8% in Spain and 29'0 or more in Ireland, Portugal and the Netherlands. These four countries would not qualify to join the euro if the Maastricht criteria were applied today. This divergence in inflation complicates the European Central Bank's job. Conditions in Portugal and Ireland argue for a rise in rates at its next meeting on April 8th. But conditions in the euro area as a whole suggest a cut is needed.
The Economist, April 3rd 1999

Table 13.1 Comparative inflation rates, 1998 change, year on year

Compensation per employee

Consumer prices

Producer

Implicit price deflators

Private
consumption		GDP

Australia
Austria
Belgium
Canada
Denmark
France
Germany
Italy
Japan
Netherlands
Spain
Sweden
Switzerland
UK
USA
OECDa
EU15

3.2 2.3 2.3 1.8 4.8 2.0 1.3 -0.1
-0.3
3.0
1.3
5.2
0.9
6.9
3.9
2.8
2.6

a Excluding Czech Republic, Turkey.

Source: OECD.

0.9
0.9
1.0
1.0
l.s
0.8
0.9
1.7
0.6
2.0
1.8
0.4 0.0 3.4 1.6 3.8 1.6

-0.4
-2.3
-3.7
-0.3
-2.8
-2.5
-1.2
-1.3
-2.0
-5.0
-2.1
-1.3
-2.0
-0.1
-o.l
-0.3
-1.7

Hungary, Mexico,

1.7
0.9
1.0
1.2
1.7
0.3
0.9
2.4
0.4
1.8
2.0
0.5
0.2
2.1
o.s
1.3
1.5

Poland, South Korea and

0.7 1.2 1.9 -0.4
1.4
0.7
0.9
2.9
0.4
1.8
2.2
0.8
1.1
2.5
1.0
1.3
1.8

Inflation and the economic cycle

Cycles set in train by supply shocks are evident in Figure 131 The analysis of inflation is also aided by an understanding of the effects of the economic cycle, where pressures often come through on the demand side. Prices increase less rapidly or even decline during recession when consumer spending is weak. However, at the top of the cycle when personal incomes are buoyant but businesses cannot increase their output, extra demand pulls up prices.

182

PRICE INDICATORS

pgICE INDICATORS

price indicators measure levels and changes in particular prices or groups of prices. For example, an oil price index covers one set of hydrocarbons while a consumer price index relates to a basket of goods and services purchased by households. In turn, these price series act as leading indicators of cost pressures and also signal movements in current demand.
	Price indices are sometimes called deflators when they are used to convert (deflate) figures in current prices into constant price terms.

The composition of price indices

Most price indices are weighted averages of several prices. It is difficult to choose weights for some indicators, such as commodity price indices (see page 190), although often the basis for the weighting is clear. If 20% of an average family's spending goes on food, food has a 20% weight in the index of consumer prices. This can cause problems for interpretation, since there may not be an average family. The average rate of inflation is not experienced by a newly married couple with spending dominated by mortgage payments, nor by a retired person with little expenditure on consumer durables. It also poses problems if spending patterns are changing over time.

Sequence. The briefs on wages and prices are arranged loosely In the order in which inflationary pressures are transmitted through the economy: from raw materials prices and wages through producers' selling prices and consumer prices to the GDP

Base-weighted indices may be used to measure changes over any period. The weights are the same for each year, so changes in the index reflect changes in prices only. The snag is that the weights may become outdated.
Implicit deflators reflect changes in prices and spending patterns. For example, the American GDP implicit price deflator measures the difference between current and constant price GDP. Since the weights reflect the composition of GDP in each period, changes in the index reflect movements in both prices and the composition of GDP.

The transmission of inflationary pressures

when looking at price indicators, think about their relationship to final prices.

183
13 PRICES AND WAGES

deflator. Indicators towards the end of the chapter are also si~_ nals of the state of aggregate demand.
	Prices in the early stages of production generally fluctuate more than prices at later stages. A 5% change in raw materials prices is generally less worrying than a 5% change in producers' selling prices.

Relationships. The sequence of cost pressures is clear, but it is not so simple to identify a fixed relationship between any pair of price indicators.

There are leads and lags. In general, changes in raw material
costs take longer to feed through to retail prices than movements in producers' selling prices.
The ultimate effect depends on the cost mix. For example, if raw materials prices rise by 10% and account for one-tenth of a manufacturer's costs, and if all other costs are held constant, the effect is a 1% increase in list prices.
Any movement in output prices will reflect the extent to which higher input costs are absorbed in profits or offset by improved efficiency and productivity.

Indicators to use

Consumer prices are the most rapidly available guides to "national" inflation and they can be used as yardsticks for interpreting other price indicators. For example, wage settlements above the rate of consumer price inflation suggest increasing inflationary pressures (unless they are absorbed by productivity growth - see Unit labour costs, page 204).
	Cost pressures are signalled by commodity prices, producer prices and wages and earnings. Surveys of price expectations are valuable leading indicators. The GDP deflator itself may be the best overall guide to inflation and should fluctuate least because it covers so many things, but it is available only after a time lag. Share prices and house prices are useful indicators of asset prices, which influence aggregate demand (see Consumer spending, page 83).
	For reviewing particular groups of prices or their effect on certain industries or groups of consumers, select an appropriate indicator. For example, to track the cost of capital equipment purchased by a particular industry, look for a sub-index in the producer prices or the GDP investment deflators.

Cross-references. Just about every economic indicator says something about demand pressures. Capacity use and unemploy' ment are particularly useful, as are indicators of the govern'

184

GOLD PRICE

ment's monetary and fiscal stance. See also Balance of payments and, particularly, Exchange rates (Chapters 10 and 11).

1Zates of change

Remember to distinguish between a fall in the level of prices and a fall in the rate of increase. If the inflation rate declines but remains positive, prices are still rising.
	As a rule of thumb, annual inflation of 0-3% is considered good. Double-digit percentage rises are definitely bad news. Negative inflation, although rarely experienced, is not good either since it signals deflation and - almost certainly - a contracting economy.

GOLD PRICE

Measures: Significance:

Presented as: Focus on: Yardstick:

	Market price of gold.
	Raw material and psychologically important store of
	wealth.
	$ per oz.
	Trends. Use price in soRS if possible.
	Average London spot prices were SDR267.14 in 1996,
	SDR240.62 in 1997 and SDR216.89 in 1998.
Released:		Continuously round the clock

Influences on the price of gold

The gold price reflects the interaction of supply and demand in a global market with many buyers and sellers and a free flow of information. Supply depends on production and sales from stocks while demand is influenced by gold's dual function as an industrial raw material and the ultimate store of wealth. It provides a security which cannot always be matched by paper money. Speculative demand is the major short-term determinant of price.

Supply. South Africa is by far the largest producer of gold, followed by the former Soviet Union, North America and Australia. In the past Soviet output has not shown any strong relationship to prices. Other countries' production generally edges upwards in the long run in response to higher prices. Sales of gold from stocks are important since stocks are many times greater than annual production. Even though central banks and the IriF sold gold in the 1970s and 1980s when it fell from favour as a monetaI'Y standard, they still hold 940m oz in their vaults. Uncertainty about what they might do with this may depress prices; and the gold price fell sharply in 1999 after the British government said

185
13 PRICES AND WAGES

it planned to sell some of its gold. However, central bankers in developing countries will dent their gold hoards and egos only as a last resort.

AIL PRICES

Measures: Significance:

Demand. Fabrication demand is mainly for jewellery, electronics
and dentistry. Jewellery demand accounts for over half of indus
trial and commercial use and is sensitive to price. Electronics
demand reflects the fortunes of the industry, decreasing during
recession. Use in dentistry is fairly constant, if vulnerable to		Yardstick:
replacement by man-made materials.
	Speculative and investment demand is much harder to predict
since flows are large in relation to stocks and output. There tend
to be flights into gold which push up its price during rapid infla
tion, exchange-rate turbulence or political instability worldwide.
However, these are less marked now that financial markets offer
more sophisticated hedging instruments. Indeed, gold-backed
financial instruments, such as gold options, have eroded the lure
of holding the metal itself.

Forward sales. Financial engineers have also created a wide
range of instruments which allow producers to hedge several
years' future output. This was probably the most important
influence on prices in 1990. Every time prices rose, they were
capped as forward sales pushed more metal into the world mar
ket. The mining company ItTZ estimated that by the end of the
year over 1,100 tonnes were sold forward and it expected the
total to increase to a plateau in the mid-1990s, which it did.
Currency. Since gold is generally priced in dollars it is impor
tant to distinguish between exchange-rate effects and underlying
price movements. The easiest way to do this is to convert the
price into a basket currency such as the sDR.Surprisingly, not
many people do this.
Gold as an indicator of inflation
Many economists, especially Americans, argue that gold is a use
ful indicator of inflationary pressures, but it is difficult to disen
tangle all the influences on gold prices. Baskets of commodity
prices probably make better leading indicators than gold alone.
Moreover, it does not make much difference whether gold is
included or excluded from such baskets if they are weighted
according to world production levels. If gold is given a greater
weight to reflect its psychological importance, the predictive
value of such baskets deteriorates. In other words, a commodity
price index is preferable as a leading indicator of inflation.

186

OIL PRICES

Market price of crude petroleum.
Major energy source essential to every economy; also a chemical feedstock.
$ per barrel.
Traded crude such as North Sea Brent or West Texas Intermediate.
In 1990 the oil price was around $20 a barrel. In 1999 it dipped to around $ 10 a barrel before recovering to $25 a barrel before the end of the year. Sharp increases in oil prices are potentially wonying for output and inflation in oil-importing countries - but much less so than in the 1970s.
Continuously round the dock.

Presented as: Focus on:

Released:

Prices, supply and demand

Oil prices are sensitive to supply and demand, with OPEC exports being the major determinant of short-term price fluctuations.

Demand. There are obvious seasonal variations in oil demand; consumption always decreases during the hot summer months. However, in the short term annual demand is fixed in relation to GDP. Consumption fluctuates with the economic cycle in industrialised countries and rises relentlessly in line with economic growth in less developed countries. (Since 1973 oil intensity - oil consumption per unit of GDP - has fallen owing to conservation and the substitution of other fuels.)

Supply. Oil producers can be divided into three groups: oPEC, the former communist bloc, and what the oil industry has traditionally called the free world. Oil output in the free world is Price responsive; it becomes profitable to extract oil from marginal fields only when prices are high. Within such considerations the free world normally produces oil flat out. The gap between demand and supply is therefore filled by oPEC crude. The organisation's attempts to control the world petroleum markets led to oil price rises in 1973 and 1979 and a sharp slump in 1986. In the past world market sales from producers m the former Eastern bloc did not fluctuate wildly, but this is 1kelY to change.

	In the short term OPEC exports reflect the balance between members' collective willingness to restrict output to try to control the world oil market, their individual need for revenue, and the general political situation in the oil producing countries. More

187
13 PRICES AND WAGES

ominously for the longer term, OPEC members are sitting on oil reserves which promise to outlast all other supplies.

Stocks. Whereas almost all the gold ever produced is still in existence (even if some is in orbit or on the sea bed), oil is rapidly consumed. Oil companies and some governments hold working and strategic stocks which help to prevent prices rocketing in times of temporary crisis, such as during the Iraqi invasion of Kuwait in 1990. Stocks of 100 days' forward consumption are about the highest to expect.

Table 13.2 The world oil market
Barrels per day, m

1990				1994	1995	1996

1997

Demand
DECD, of which:		41.6		44.3	44.9		45.9	46 7	46.8
	United States		20.7		21.6	21.6		22.2	22.7	23.0
	Europe	13.7			14.3	14.6		14.9	15.0	15.3
	Pacific	7.2			8.4	8.7		8.8	8.9	8.4
Non-OECD		24.9		24.2	25.2		25.9	26.8	27.1
Total	66.8			68.6	70.1		71.7	73.4	73.8
Supply
DECD	18.9			20.9	21.1		21.7	22.1	21.9
OPEC	25.0			27.4	27.7		28.4	30.0	31.0
Former Soviet Union			11.4	7.2	7.1		7.1	7.2	7.3
Other non-OECD		11.5		13.2	14.3		14.8	15.1	15.2
Total	66.9			68.6	70.1		72.0	74.4	75.3
Trade
DECD net imports		22.9		23.8	23.5		24.2	24.9	25.2
former Soviet Union net exports 3.1			2.4	2.4		2.7	2.9	3.0
Other non-OECD net exports 19.8			21.4	21.1		21.4	22.1	22.2
Prices
DECD crude oil import price
$/barrel, cif		22.3		15.6	17.2		20.6	19.1	12.6

Note: Totals may not equal sum of components owing to rounding.

Source: DECD.

Traded crude. OPEC's slippery grip on oil trade means that its official selling prices (oSPS) are not necessarily representative f market pressures. However, whereas nearly all oil was sold on long-term contracts at fixed prices in the past, most is now traded at market prices. A traded crude such as North Sea Brent blend is therefore a good indicator of market conditions.

188

OIL PRICES

Table 13.3 Effect of a $10 oil price increases change in one year

One year after shock Two years after shock

GDP level
United States		-0.2
-0.2
Japan			-0.4	-0.5
European Union			-0.2	-0.2

Consumer price level
	United States
	Japan
	European Union

Current external account (% of GDP)
	United States		-0.1
	Japan
	-0.2
European Union 		-0.1

a From around $12 a barrel to around $22.

Source: DECD.

0.4
0.5
0.6

0.6
0.7
1.1

-0.1
-0.2
0.0

OIL'S PLEASANT SURPRISE
It seems that oil-price shocks are less shocking than they used to be
Could the bad old days of stagflation be about to return? Since oeEC agreed to supply-cuts in March, the price of crude oil has jumped to almost $26 a barrel, up from less than $10 last December and its highest since the Gulf war in 1991. This near-tripling of oil prices evokes scary memories of the 1973 oil shock, when prices quadrupled, and 1979-80, when they also almost tripled. Both previous shocks resulted in double-digit inflation and global recession. So where are the headlines warning of gloom and doom this time?
	Their absence is even more striking given that, at the start of the year, many commentators (including, rather prominently, this newspaper) expected prices to fall, not rise. oeEC's agreement to cut output has so far proved more durable than many predicted. The oil price was given another nudge up this week when Iraq suspended oil exports in a showdown with the urr over sanctions. Strengthening economic growth, at the same time as winter grips the northern hemisphere, could push the price higher still in the short term.
	Yet there are good reasons to expect the economic consequences now to be less severe than in the 1970s. The sharp rise in oil prices follows an equally sharp collapse over the previous two years, when prices fell by more than half to their lowest level in real terms since before the 1973 shock.

189
190

13 PRICES AND WAGES

Even now, prices are not much higher than in early 1997.
	Moreover, in most countries the cost of crude oil now accounts for a smaller share of the price of petrol than it did in the 1970s. In Europe, taxes account for up to four-fifths of the retail price, so even quite big changes in the price of crude have amore muted effect on pump prices than in the past.
	Rich economies are also less dependent on oil than they were, and so less sensitive to swings in the oil price. Energy conservation, a shift to other fuels and a decline in the importance of heavy, energy intensive industries have reduced oil consumption. Software, consultancy and mobile telephones use far less oil than steel or car production. For each dollar of GDP (inconstant prices) rich economies now use nearly 50% less oil than in 1973. The OECD estimates in its latest Economic outlook that, if oil prices averaged $22 a barrel for a full year, compared with $13 in 1998, this would increase the oil import bill in rich economies by only 0.25-0.5% of GDP. That is less than one-quarter of the income loss in 1974 or 1980. On the other hand, oil-importing emerging economies - to which heavy industry has shifted-have become more energy-intensive, and so could be more seriously squeezed.
	The impact on the output of oil-importing countries also depends on whether oil producers save or spend their windfalls. Both in 1973 and 1979 many oeec countries already had current-account surpluses, and most of their extra oil revenues were saved. Today, many have large current-account deficits (Saudi Arabia's hit 10% of GDP last year). Cash-strapped producers are more likely to spend their windfalls on imports from rich countries.
	One more reason not to lose sleep over the surge in oil prices is that, unlike the rises in the 1970s, it has not occurred against the backdrop of general commodity-price inflation and global excess demand. A sizeable chunk of the world is only just emerging from recession. The Economist's commodity price index is broadly unchanged from a year ago. In 1973 commodity prices jumped by 709'0, and in 1979 by almost 30%.

Refining the argument
Even if the impact will be more modest than in the past, dearer oil will still leave some mark. Inflation will be higher and output lower than they would be otherwise. The oECIYs rule of thumb is that a $10 increase, if sustained for a year, would increase the inflation rate in rich economies by

COMMODITY PRICE INDICES

about half a percentage point and knock about a quarter-point off growth.
	The impact of higher oil prices varies by country too. Perhaps the biggest risk is in America, where rising oil prices may push the inflation rate higher than is currently predicted. The slide in oil prices in recent years was one of the main factors that helped to hold down American inflation, so prolonging the country's long economic expansion. That positive factor is now going into reverse. Higher oil prices have already helped to lift America's inflation rate to 2.6% in October, up from 1.5% a year ago; the latest rise in oil prices could well push it above 39'0. The core inflation rate remains relatively subdued, but headline inflation could still spill into wages and hence other prices. If it does, the Fed might be forced to raise interest rates by more than is now forecast. So orEC could yet do more damage than most people expect.
The Economist, November 27th 1999

COMMODITY PRICE INDICES

Measures: 		Changes in groups of commodity prices.
Significance: Advance warning of inflationary pressures. Presented as: Index numbers.
Focus on: Trends.
Yardstick: Prices fluctuate wildly; hope for level trends.
Released: 		Daily by Reuters; weekly by The Economist; others monthly.

Significance

Commodities are unprocessed or semi-processed raw materials used for food or in the manufacture of other goods. Commodity prices in general are important lead indicators of cost pressures. Prices of metals and, to a lesser extent, non-food agriculturals are also indicative of the level of demand in the industrialised countries.
	Monetarists note with glee the correlation between high commodity prices and liquidity in the late 1980s, and the monetary contraction and commodity price falls of the early 1990s.

Price instability

Analysis by the World Bank shows that price instability for Sugar, the least stable commodity price, is over 11 times greater than that for oranges, the most stable. Moreover, commodity Price fluctuations have increased sharply since the 1960s. Quite apart from the eightfold increase in the real price of oil during the 1970s, the prices of other commodities moved sharply. For
13 PRICES AND WAGES

example, food prices fell while timber prices rose in the 1980S. Many economic problems can be traced to these shocks and the policy responses to them.

Influences on prices

Commodities may be divided into three broad groups depending on whether their prices are influenced mainly by demand, supply, or both.

Demand. The prices of industrial raw materials, such as metals and minerals, fluctuate in response to changes in demand, reflecting mainly economic conditions in the industrialised countries. Recession brings lower demand and weaker prices. Supply tends to be more stable and predictable.
	Demand for metals declined as the industrial countries introduced materials-saving technology during the 1970s and 1980s. Between 1990 and 1993, The Economist commodities dollar index for metals fell by over 30 percentage points. By around mid1994 it had recovered all that lost ground and more, before it dropped in 1995 to slightly below its 1990 level.

COMMODITY PRICE INDEX Our dollar-based metals index fell by almost 50% from its peak in January 1995 to the lowest in more than five years in March 1999. But it has since risen by 3196,

Metals

The Economist metals index

LME prices
i

terms !o increase
~pfar 15th 1999 to Nov 30th 7999
`..:0 70 20 30
	-- - ---- - ------

Copper
	Zinc

192

COMMODITY PRICE INDICES

economies are recovering. Nickel has done best, gaining 559'0 since March, because the stainless-steel industry is booming. Prices of aluminium and copper have risen by about 30%, but the threat of excess supplies next year has ended their rallies. Tin prices have been boosted by a recent stoppage of Chinese exports. Zinc prices have been healthy, even though consumption in Japan is the lowest for 24 years: demand elsewhere has been up. Stocks of lead continue to grow, weighing prices down.

The Economist, December 4th 1999

Supply. Food prices are influenced most heavily by unplanned changes in the supply side. For example, world vegetable oil prices depend significantly on the effects of weather on the American soyabean crop and on policies relating to American stockpiles. Food prices fell in the late 1980s and early 1990s as production and stocks recovered from the 1988 drought in America which cut output and pushed up prices.

Supply and demand. Non-food agricultural products such as cotton and rubber are vulnerable to changes in both supply and demand.

Index composition

Creating an ideal commodity price index is intellectually testing because:

commodities are not comparable - 1 tonne of coffee is quite different from 1 tonne of copper;

Figure 13.2

The Economist commodities index
Dollar terms, 1990=100

1990 91 q2 q3 96 97

qt
sourte: The Economist.

Metals

	Non-food agriculturals
__- All items

	q2 q3 q4 93 q2 q3 q4				94 q2 q3 q4	95 q2 q3 q4	96 q2
	ql				q1	ql	qt

193
13 PRICES AND WAGES

		they are difficult to value - for example, only 2.5% of rice pro
	duction is traded on international markets; and
		relative prices are distorted by large fluctuations in individual
	commodity prices.

Despite these problems there are many indices combining the prices of several commodities. Apart from The Economist corn_ modifies price index, the most widely followed indices are prepared by the IMF, the ulr, the World Bank and the American Commodity Research Bureau (cRB). They differ in three main ways.

The basket contents. The Economist index includes only commodities which are freely traded on open markets. This excludes items such as iron ore and rice which have a big weight in the other main indices. The Economist also omits oil and precious metals such as gold which account for a quarter of the cite index. The two previous briefs suggest that gold and oil are important commodities but they are subject to special factors which may make them less valuable as simple cost indicators.

The basket weights. The Economist index is designed to measure cost pressures in industrial countries; its constituents are weighted according to their share in OECD imports. The UN, IMF and World Bank indicators are intended to monitor the terms of trade in developing countries; the constituents are weighted to reflect shares in developing countries' exports. The citB index just gives equal weight to all components, which understates the importance of industrial commodities and so makes it less useful as a leading indicator of inflation.

The currency. Commodity prices are most often quoted in dollars. As a result, currency fluctuations will move indices even when there is no underlying change in commodity prices.

COMMODITY PRICE INDEX
Our all-items dollar index slumped by 16.5% in 1998 and is now at its lowest level for five years. Both industrial and food prices have been hit by falling demand in the wake of the Asian and Russian financial crises. The metals index has fallen by 45% from its peak in January 1995. A sustained recovery in metal prices is unlikely without cuts in output. Copper is at a 12-year low, and nickel at an 11-year low. Soft commodity prices subsided as the El Nino weather system faded; the food index is now a third below its peak in May

194

1997. Surpluses pushed wheat prices to 20-year lows and sugar to ten-year lows.

COMMODITY PRICE INDICES

The Economist, January 2nd 1999

Commodity prices

The Economist all items index

-~W
ms, Dec 30th 1997.100

Selected commodities

	'$ terms, / -decrease Dec 39th 1997
~fbec 29th 1998
35					30	25	20	15

Bottom line

The Economist index is hard to beat for tracking the interrelation between cost pressures and inflation; see the weekly commentary in The Economist. A rise in the index may signal higher inflation, but the final outcome depends on monetary conditions and supply and demand in other factor markets such as labour.
	A uN, IMF or World Bank index should be used if you are interested in the way that changes in commodity prices affect the external balances of developing countries: a fall in commodity Prices implies a deterioration in exporters' trade balances.

195
13 PRICES AND WAGES

EXPORT AND IMPORT PRICES; UNIT VALUES

Measures: Significance:

Presented as: Focus on: Yardstick:

	Prices of traded goods.
	Helps identify cost pressures, potential exchange-,-ate
	problems and changes in competitiveness.
	Index numbers.
	Changes in unit values (see text).
	A positive rate of increase, as close to zero as possible, is
	good. Check the ratio between import and export prices
	(see Terms of trade, page I 57).
Released:		Monthly with trade figures; quarterly with GDP.

Prices. Import or export price indices, including the deflators released with GDP data, capture changes in both the prices and the composition of a country's external trade.

Unit values. Indices of import or export unit values, which are essentially fixed-weight price indices, highlight changes in prices only. This is useful for reviewing cost and competitive pressures. However, since the weights get out of date when the structure of trade changes - as it inevitably does -unit value indices are best for analysing short-term trends.

Export prices or values. Compare export price indices with domestic price indicators, such as the producer price index for home produced goods, to get a feel for the way that manufacturers are passing on cost pressures to foreign buyers; or perhaps being constrained from doing so by international competitive pressures.

Import prices or values. Use import price indices to judge external cost pressures. Import prices that are rising faster than domestic prices are a clear warning of imported inflation.

Erratic items. Prices of raw materials can fluctuate widely. It is often sensible to look at price indices which exclude these more erratic items to identify underlying pressures.

Terms of trade. One of the most useful ways of looking at import and export prices is by examining the ratio between them, which is known as the terms of trade (see page 157).

196

PRODUCER AND WHOLESALE PRICES

PRODUCER AND WHOLESALE PRICES

Measures: Prices of goods at the factory gate. Significance: Leading indicator of cost pressures. Presented as: Monthly index numbers.
Focus on: 		Percentage changes.
Yardstick:		OECD average producer prices rose 5.5% a year during the
	1980s and 3.6% a year during the period I 990-95.
Released: 		Monthly, at least one month in arrears.

Wholesale price indices (WPIS) cover prices charged at the first stage of bulk distribution and generally include import prices. WPIs were first introduced to measure prices of raw materials.

Producer price indices (PPIS) track prices of home produced goods at the factory gate. Most PPIS cover output prices of goods, although some countries also prepare input price indices for raw materials purchased by industry. In principle, input prices include transport to the factory and output prices are exworks, although such prices cannot always be identified neatly.
	PPIS shed light on cost pressures affecting domestic production and are more useful than WPIS. Most major countries now produce PPIS but Switzerland produces only a wholesale prices index, as do, for example, Austria, Greece and Norway. The indices cover manufacturing and, sometimes, agriculture (Belgium).

Index construction

The indices are compiled on the basket principle with weights reflecting the output of each contributor relative to the total. For example, if lace-makers account for 1% of total industrial production, lace prices have a 1 % weight in the index.
	Weights are generally updated at 5-10 year intervals to take account of the changing structure of industry. Data are acquired by surveys, usually of major companies. The Swiss index covers 600 price series, the American index over 3,000, and the UK index over 10,000.

	In many cases returns are collected continuously so that the prices are effectively monthly averages rather than those applying on just one date. Even so, changes one month will not be reflected fully until the following month.

	Taxes such as vAT are usually excluded. Excise On tobacco and alcohol are treated as manufacturers' are included, so the index moves if they are changed.

duties such as costs and

197
13 PRICES AND WAGES

Figure 13.3

Changes in producer prices
% change, 1998

oa ~		va			c	c'~	oa	a
ha1 vJ				c~aa a~e	``aa x		ac	ua J	av1	xaa eaFa ~saa acaa DLO ~	P
	~z Jz ,aa		P~ SQ ~aQ 5.t~ `c-			v	c.t~V	P	c. O J	JS
o								-
		-0.4					-0.3	-0.3	-0.1	-0.1
-1		~~ ~Uuu~
-1.3 -1.3 1.2
_2	-1.7
-2.1		-2.0 -2.0
-2.5
-2.3
-3			2.8
-q			-3.7

-5
-51
.0

Source: DECD.

The cycle

Prices are generally order prices with list prices adjusted by government statisticians to allow for "normal" discounts in each industry. This is fine when the economy is stable, but the Ppi may overstate cost pressures when above-average discounts are offered during a recession. Conversely, the Ppi understates cost pressures when inflation is rapid; deliveries may be at prices which were negotiated perhaps several months earlier and which are much below current order prices scored in the index.

Use

Producer output prices and consumer prices tend to follow the same path, with producer prices fluctuating more widely. However, producer prices generally increase less rapidly than consumer prices (compare Tables 13.4 and 13.7).
	Producer input price indices vary more widely than output prices in response to movements in both commodity prices and exchange rates and are a useful guide to raw materials cost pressures. See the comments on pages 183-185 about the way teat pressures feed through the system.

Seasonal adjustment. nPls are rarely seasonally adjusted. Even if they are, the seasonal adjustment can be suspect, especially on the input side since, for example, exchange rates do not follow a neat seasonal path. Comparisons over 12 months provide a safer guide to trends, while underlying producer price pressures call

198

SURVEYS OF PRICE EXPECTATIONS

often be better identified by examining PPIS excluding items such as food prices, which tend to bump around erratically.

Table 13.4 Producer prices
Annual average /D change

1970-74 1975-79

1985-89 1990-94			1995	1998

Australia							8.3	11.0	6.7	8.4	3.8	-0.4
Austria							2.4	3.5	-1.5	0.4	0.4	-2.3
Belgium							8.1	3.4	-1.8	-0.1	2.2	-3.7
Canada							8.9	9.1	2.5	2.1	8.0	-0.3
Denmark							11.1	7.4	0.6	0.1	2.8	-2.8
France							...	...	5.8	-3.5	6.1	-2.5
Germany							6.7	3.1	-0.2	1.1	1.8	-1.2
Italy							15.5	15.9	3.2	1.8	7.9	-1.3
Japan							11.1	2.8	-2.9	-1.8	-0.1	-2.0
Netherlands							6.1	4.1	0.0	1.6	1.4	-5.0
Spain							10.2	16.1	2.2	2.4	6.4	-2.1
Sweden							8.2	7.6	3.2	2.9	7.7	-1.3
Switzerland							8.0	-0.1	0.1	-0.1	-0.1	-2.0
UK							10.8	14.0	3.3	3.7	4.2	-0.1
USA							9.8	7.7	2.2	0.9	3.6	-0.1
EU							...	...	...	1.7	4.6	-1.7
DECD							...	...	...	2.7	7.1	-0.3

Source: DECD.

Detail. Producer price numbers are usually available at a high level of detail, covering commodity groups (metals, furniture), stages of processing (crude materials, intermediate goods) and industries (farm machinery, capital goods). Watch especially the prices of finished goods in general and of consumer goods.

Special applications. rPls are frequently used as the basis for price indices for inflation accounting and also for contract price adjustments, often in combination with earnings indices.

SURVEYS OF PRICE EXPECTATIONS

Measures: Manufacturers' perceptions of inflationary pressures.
Significance: Excellent anecdotal warning of potential price changes.
Presented as: Percentage balances (for example, percentage of those
	expecting to raise prices).
Focus on:		Trend in expectations.
Yardstick:		An increase of a few points over a few months is a
	warning of inflationary pressures.
Released:		Monthly; not revised.
13 PRICES AND WAGES

Coverage and interpretation

Surveys of price expectations provide excellent inflation indicators usually straight from the horses' mouths. Various organisations conduct monthly or quarterly surveys (see Business conditions page 103) in which respondents are asked questions such as: "Do you intend to raise your prices within the next four months?" The balance of those answering yes over those saying no is presented as, say, +20% or (50 + 20) = 70. If a net 20% of respondents expect to lower prices, the balance would be -20% or (50 - 20) = 30.
	The absolute balance may not be a good guide, since there may always be an excess of companies planning price rises even in times of low and stable inflation. As a quick guide, see if the latest numbers are above or below figures for recent months; a change in the trend may suggest a potential increase or decrease in cost pressures. Better still, examine a long run of data so that you can put the latest figure in the context of the economic cycle.

WAGES, EARNINGS AND LABOUR COSTS

Measures: Labour costs and influences on consumers' incomes. Significance: Indicator of both cost and demand pressures. Presented as: Usually index form, some figures in cash terms.
Focus on:		Percentage change over 12 months.
Yardstick:

	Compare with growth of output and consumer prices. In
	the OECD, compensation per employee in the business sec
	tor rose by 4. I % a year during the period I 990-98, declin
	ing from 6. I % in 1990 to 3.0% in 1998.
Released:		Mainly monthly, at least one month in arrears; revised. ~

Terminology

Wage rates. Basic pay per period (hour, week, and so on). Manual workers tend to have wages, white collar workers have salaries (are paid monthly).

Earnings. Basic pay plus overtime and bonuses. These may be quoted before or after tax and other deductions. Take-home pay is earnings after deductions.

Wage drift. The tendency for earnings to rise faster than wage rates, for example, due to overtime and bonuses.

WAGES, EARNINGS AND LABOUR COSTS

as social security, free meals and a host of other perks. Nonwage costs are 20-50% of labour costs. (See also page 204.)

The Phillips curve. In the 1950s a New Zealand economist, A,W. Phillips, identified an apparent trade-off between unemployment and the rate of increase in wages. His curve suggested that low annual increases in wages are associated with a high unemployment rate; or, conversely, high wage inflation is associated with a low unemployment rate. (See Unemployment, page 63).

WAGE COSTS
In most industrial countries labour costs have been growing more slowly during the past flue years. Unit labour costs measure the total cost of labour required to produce one unit of output. They are thus driven upward by wage inflation but downward by gains in labour productivity. In nine of the II countries in our chart, the average annual increase in unit labour costs (in local currcncy terms) was lower during the 199o-95 period than it had been during 1985-90. The biggest change took place in Sweden, where unit labour costs rose at an annual rate of 6.6% in the earlier period more than in any of the other countries shown - but have since fallen by o.6% a year. Unit labour costs rose by only 03% a year in America and Canada during 199o-95.
The Economist, August 31st 1996

Manufacturing labour costs
Unit labour costs, annual average % change

SMra:SMawudt banSUWlo.

1se5ao =1ss0-s5

Nknwn. tH1HL

Key figures Cash totals. Many government and private-sector bodies publish money wage rates and earnings in various industries, some-

Labour costs. Sometimes called total compensation, these are wages and salaries

plus pension contributions, payroll taxes suc

201
13 PRICES AND WAGES

Figure 13.4

Compensation per employee in the business sector
h change, 1998

1.8
	1.3 1.3
0.9

-0.1		-0.3

LO										~a	Jcc	~z	as	c1	ao	~a	a~1	ao

OQ. PJt'' ^~0 ~~a~ Laca ,~F			SQ h~a	S1aQ

Source: OECD.

WAGES, EARNINGS AND LABOUR COSTS

Time series. Governments also produce indices showing trends wages and earnings, while various organisations (including those mentioned in Business conditions, page 103) track pay settlements. These are important indicators of both cost pressures and aggregate demand. Personal incomes (see page 81) offer a better guide to potential consumer demand because earnings are usually quoted before tax and other deductions, often cover only part of total employment and may ignore income from self-employment.
	At one extreme, American average weekly earnings exclude salary-earners and cover only production and non-supervisory workers. At the other, British average earnings cover both manual and non-manual employees.

Interpretation

Wages and earnings are closely linked to the economic cycle.

	When aggregate demand begins to recover after a recession, pro
times including the cash value of perks such as company cars.			ducers respond first by increasing overtime and earnings rise faster
The figures are interesting for comparisons between industries,			than wage rates. Only when higher demand seems more estab
sectors and countries. International comparisons, however, are			fished do employers take on more workers, which then puts
complicated by differences in coverage, tax regimes and by fluc-			upward pressure on wage rates. When the cycle turns and output
tuations in exchange rates.	begins to decline, overtime is cut first and earnings rise less rapidly
	or fall. Then staff are laid off. The annual increase in wage rates
Table 13.5 Hourly earnings in manufacturing	tends to decline as unemployment rises, although usually with a lag.
Annual average % change		A change in average earnings can reflect changes in wage
	rates, total hours, the mix of standard hours and overtime, out-
1975-79 1980-84 1985-89 1990-94 1995$ 1998a		put (piecework and profit-sharing), and the relative mix of jobs,
	grades, industry, and so on.
Australia			10.5	9.9	7.0	3.6	2.4	3.2	In general,	if earnings are rising faster than consumer price
Austria			7.3	5.4	4.0	4.8	3.5	2~3	^~ z
Belgium			8.7	6.4	z.s	3.5	2.3	2.3	inflation,	real spending power is growing. It should be noted,
Canada			10.1	9.1	4.0	2.8	1.7	1.8	4,	however, that earnings data can be distorted by industrial dis-
Denmark			1o.s	7.7	6.4	2.2	3.3	4.8 . y~	putes, delays in implementing pay settlements, lump-sum back
France		15.1	13.9		4.2	3.6	2.7	2.0
Germany			7.0	4.9	45.6	5.4	3.4	1.3 rc,:	pay and temporary lay-offs during bad weather.
Italy		20.9	16.8		5.9	6.2	5.3	-0~1
Japan		8.7	4.2		2.9	2.5	0.8	-0.3
Netherlands			6.5	3.4	1.4	2.7	1.3	3.0
Spain		27.5	13.8		8.0	7.5	2.3	1.3
Sweden		9.1	8.7		8.1	4.2	2.7	5.2
Switzerland			3.1	5.0	3.5	5.1	2.4	0.9
UK		13.9	9.2		3.6			6.9
				5.4			2.7
USA		8.5		6.0	2.4	2.7	2.2	3.9.
	EU 	...			5.3	3.5	2.6
	DECD 		...		...	3.6	2.7	2.6

a Compensation per employee in the business sector.
Source: OECD.

202	203
13		PRICES AND WAGES

UNIT LABOUR COSTS

Measures: Significance: Presented as:

	Labour costs per unit of output
	Indicator of cost pressures and competitiveness.
	Index form.
Focus on:		Percentage change over 12 months.
Yardstick:		OECD average unit labour costs in the business sector rose
	by 4.5% a year during the 1980s and by 2.9% during the
	period 1990-95.
Released:		At least one month after the end of the quarter, frequently
	revised.

Significance

Unit labour costs (uLCS) measure the average cost of producing one unit of output; for example, labour costs divided by GDP. This is a key indicator of the cost efficiency of labour. If unit labour costs fall, the same output can be produced for less expenditure on labour.
	Unit labour costs reflect two factors, labour costs and productivity. Britain's rapid increases in unit labour costs in 1989 and 1990 reflected the twin evils of rising wages and falling output, and hence declining productivity.

The cycle

Within the economic cycle, the rate of increase in unit labour costs generally peaks 12-18 months after a peak in activity. When output first begins to fall uLCS rise faster because there is less production for the same spending on labour.

Competitiveness

Relative movements in unit labour costs are important signals of international competitiveness in traded goods. Businesses in a country where unit labour costs are rising faster than in other countries might temporarily absorb the pressures by cutting profit margins or improving efficiency.
	In the longer term deteriorating competitiveness will reduce exports, output and employment and so eventually tame inflation the hard way. Some economists still advocate devaluation to restore price competitiveness, but experience shows that the initial benefit is quickly eroded by faster inflation. A fixed exchange rate can impose a useful discipline on pay bargaining.

204

CONSUMER OR RETAIL PRICES

Table 13.6 Unit labour costs in the business sector
Annual average % change

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

Source: DECD.

1980-84					1985-89	1990-94	1995	1998

8.2 3.9 4.6 5.8 6.0 8.9 2.5 15.2 2.3 1.0 10.4 8.1 4.6 5.6 4.6

6.5
2.5 1.4 3.9 3.9 1.2 1.0 4.8 -0.3
0.5
5.0
7.3
6.5
4.7
3.3

0.2
2.8
3.1
1.4
6.0
1.4
3.3
3.5
1.1
2.2
4.2
1.8
4.3
2.5
2.7

l.s
2.8 0.8 0.8 2.2 2.0 2.0 1.0 0.2 1.0 2.3 0.2 1.8 1.3 2.6 1.9 1.9

1.7
0.8 0.5 0.8 3.9 0.6 -1.3 -0.6
2.2 2.4 2.3 2.4 -0.1
5.6
2.3
1.2
1.7

CONSUMER OR RETAIL PRICES

Measures: 		Price of a basket of goods and services.
Significance: Indicates inflation as experienced by a "typical" household. Presented as: Monthly index numbers.
Focus on:		Percentage changes.
Yardstick:		OECD average consumer prices rose 6.7% a year during the
	1980s and by 5.0% a year during the period I 990-95.
Released:		Monthly, one month in arrears; quarterly in Australia, New
	Zealand and Ireland; rarely revised.

Composition

The consumer price index (CPI) is the indicator most people use to track inflation. The index is familiar and readily available, but not necessarily accurate. The British call it the retail price index (RPI). The European Union's statistical agency, Eurostat, publishes a harmonised index of consumer prices (IItCP) for the Eu-15. Estimates of IIICP for individual member states are also published. The European Central Bank's inflation target is defined in terms of HICp for the 11 Eurozone countries.

Basket contents and weighting. cPls measure the cost of a basket of goods and services purchased by the average household each month. The basket's composition and weighting are usually

205
13 PRICES AND WAGES

based on surveys of household or family expenditure habits.
	Some indices cover "essentials" only. By careful selection a consumer price index can be dominated by subsidised commodities and those subject to official price controls. This is how some developing countries keep down their reported consumer price inflation. Most indices, however, cover a fairly full range of discretionary expenditure. Weights are updated annually in Britain and France, but most countries change their weights only every 5-10 years.

Price data. Taxes on expenditure and subsidies are included in CPIs; it would be difficult to exclude them. Other taxes such as those on incomes are excluded, as are savings, life assurance premiums and capital spending.
	Prices are usually found by observation, perhaps of over 100,000 items each month. Collection points vary from six state capitals in Australia to over 100 urban centres in France and Germany.
	The information is collected by surveys on a particular day so a price change late one month may not be caught in the index until the following month. Indeed, in America, for example, prices of most goods and services other than food and fuel are collected monthly in the five largest geographic areas and every other month in the remaining 80 survey locations. In some other countries major surveys are conducted only every three months.

Housing. The British and Canadian indices exclude the cost of houses and capital repayments on home loans but include mortgage interest payments. As a result, if the government increases

Figure 13.5

Changes in consumer prices
change, 1990-98

40- 39.3 38,2
	37.1 30.3
30		79'7

70

10

0

av						C.~J~		4,J	ca	SP'	aci	ec	cJS	is	~F	ava	i
._S`ye`a		J Va~ S~-6 a e~\a PJ~ ~av0 P~~ Oee

28.2
25.4 74.6
	22.9 22.3 72.0
	t~t~7~'g 76.576.1

CONSUMER OR RETAIL PRICES

Table 13.7 Consumer prices
Annual average % change

1980-84			1985-89 	1990-94

Australia						8.3	5.4	1.1	4.6	0.9
Austria						4.9	1.5	2.8	2.2	0.9
Belgium						5.0	2.8	2.1	1.5	1.0
Canada						6.6	3.5	1.8	2.2	1.0
Denmark						7.9	3.4	1.6	2.1	1.8
France						8.3	2.4	1.9	1.7	0.8
Germany						3.4	0.8	2.9	1.8	0.9
Italy						14.1	4.3	2.3	5.4	1.7
Japan						2.4	0.7	1.4	-0.1	0.6
Netherlands						3.7	0.2	2.1	1.9	2.0
Spain						10.7	5.1	4.2	4.7	1.8
Sweden						7.5	4.2	3.7	2.9	0.4
Switzerland						3.6	1.5	2.8	1.8	0.0
UK						5.9	4.0	2.7	3.4	3.4
USA						4.7	2.9	2.5	2.8	1.6
EU15						9.7	4.4	5.0	3.1	1.6
DECD						9.7	7.1	4.9	5.5 	3.8

Sources: OECD.

interest rates are increased in response to inflationary pressures,

the index rises automatically, which is exactly the opposite to the desired and underlying effect. Most other countries use a more satisfactory rental equivalent for measuring housing costs.

Variations

Many countries produce more than one cm America has two main indices: the CPI-a for urban consumers (about 80% of the population) and the CPI-W for wage earners (32% of the population).
	As well as the "headline" index, ItPI, Britain also publishes an index excluding mortgage-interest payments (RPIx) and an index excluding mortgage-interest payments and indirect taxes (RPIY). The Bank of England's inflation target is defined in terms of

lzplx. If mortgage payments were included in the target, then interest rate increases would have the perverse effect of increasing the target measure of inflation. 1tPIx is sometimes referred to as the "underlying" rate.

Use and abuse

cPls are the most timely and best understood inflation indicators. They are often used in setting wage demands and in determining

207
13 PRICES AND WAGES

index-linked pay, pension or social welfare payments. They are also used to convert wages and prices of consumer goods (including capital items such as houses) into "real" terms.
	Consumer expenditure and GDP deflators are often better guides to inflation, but usually they are not available quickly enough.

Interpretation

Getting at the underlying rate of inflation may not be easy. CPIs are generally not seasonally adjusted and the necessary 12-month comparison is slow to highlight changes in trends. Looking at movements over the latest few months can be misleading because of erratic and distorting factors such as seasonal variations in food prices, annual price-cutting sales promotions, once-off changes in the rate of sales tax and erratic bumps in oil prices.
	It is almost always necessary to make adjustments to highlight the "core" rate of inflation. Many statistical agencies produce helpful sub-indices, usually excluding food and energy prices.

CONSUMER OR PRIVATE EXPENDITURE DEFLATORS

Measures: 		Price changes affecting total consumers' expenditure.
Significance:		Abroad indicator of consumer price inflation which is less
	susceptible to fiddling than consumer price indices.
Presented as: Quarterly and annual index numbers.
Focus on:		Percentage changes.
Yardstick:		OECD average consumer expenditure deflators rose by
	4.6% a year during the period 1990-98, declining from
	6.490 in 1990 to 3.40/6 in 1998.
Released:		Quarterly, at least one month in arrears; frequently revised.

Advantages

Consumer expenditure or private consumption deflators are derived from current and constant price estimates of total consumer expenditure. This reflects actual spending which is arguably better than the consumer prices basket approach of outlays by an average family. It also avoids a selective approach when specifying what goes in the basket.
	In addition, consumer prices data are collected at one point each month, while deflators relate to averages over the period, usually three months at a time. The deflators normally include an imputed figure for house rents, which is less distorting than the interest-rate approach used for the British 1tPI and the Canadian cPL

208

CONSUMER OR PRIVATE EXPENDITURE DEFLATORS

Problems

Deflators are not available as rapidly as consumer price indices and the deflators are revised more often. Movements in implicit price deflators reflect changes in the composition of consumers' expenditure as well as changes in prices (fixed-weight deflators avoid this problem). The personal sector is broader than households; it often also covers unincorporated businesses (such as farms, pension funds and trusts) and private non-profit bodies (such as charities and trade unions).

Table 13.8 Consumer spending deflators Annual average % change

1980-84					1985-89	1990-94	1995	1998

Australia
Austria
Belgium
Canada
Denmark
France
Germany
Italy
Japan
Netherlands
Spain
Sweden
Switzerland
UK
USA
EU15
OECD

Source: DECD.

9.1 5.8
7.1
8.3
11.4
11.0
4.5
16.4
3.9
4.6
13.9
18.8
4.5
9.1
6.7
10.3
9.7

7.3 2.1 2.7 4.0 4.0 3.6 1.2 7.6 1.2 0.9 6.7 6.2 2.3 4.9 3.9 4.3 6.8

3.0
3.4
2.8
2.5
1.8
2.1
3.6
5.6
1.8
2.6
5.9
6.2
3.8
4.7
3.5
4.4
5.1

2.5
2.3
1.5
1.6
1.7
1.6
2.0
5.7
-0.5
1.0
4.6
2.7
13.0
2.6
2.3
3.0
4.7

1.7
0.9
1.0
1.2
1.7
0.3
0.9
2.4
0.4
1.8
2.0
0.5
0.2
2.1
0.8
1.5
1.3

Bottom line

Consumer expenditure deflators should be used with care. They usually provide a valuable alternative to consumer price indices. If nothing else, they are a useful check on the signals from other inflation indices.
13		PRICES AND WAGES

GDP DEFLATORS

Measures:		Overall national price changes.
Significance:
Presented as:
Focus on:
Yardstick:

Released:

Definition

Broadest indicator of inflation. Quarterly and annual index numbers. Percentage changes. OECD average GDP deflators rose by 6.0% a year during the 1980s and by 4.5% a year during the period I 990-95. Quarterly, at least one month in arrears; frequently revised.

Deflators measure the difference between current and constant price GDP and its components. For example, if GDP increases by 2% in real terms and 5% in nominal terms, the implied economywide rate of inflation is 3%.

Table 13.9 GDP deflators
Annual average % change

1980-84				1985-89	1990-94	1995

1998

Australia		7.0	6.0	1.3	2.6	0.7
Austria		4.2	2.2	3.0	2.1	1.2
Belgium		4.5	2.6	2.2	2.2	1.9
Canada		5.5	3.3	1.2	1.7	-0.4
Denmark		6.7	3.4	1.4	1.7	1.4
France		8.0	2.8	1.9	1.6	0.7
Germany		2.8	1.8	2.7	2.2	0.9
Italy		12.4	5.3	5.5	4.8	2.9
Japan		1.8	1.1	0.8	-0.5	0.4
Netherlands			3.0	0.1	1.6	1.9	1.8
Spain		9.7	5.8	4.4	4.8	2.2
Sweden		7.0	5.2	2.9	4.1	0.8
Switzerland			3.9	2.6	2.4	1.4	1.1
UK		5.7	4.3	3.2	2.4	2.5
USA		4.9	2.8	2.2	2.4	1.0
EU15		9.8	4.9	4.3	3.0	1.3
OECD		8.7	5.9	3.7	4.5	1.8

Sources: OECD.

	Deflators can be found at any level of detail from one component of consumer spending or business investment right up to total GDP (see Table 13.1). They are not always published, but they can be readily calculated in index form if current and constant price data are available: for each period divide the current by the constant price data and multiply by 100.
Unless otherwise stated, most deflators are implicit price

GDP DEFLATORS

deflators. These measure changes in the composition of GDP as well as changes in price. Fixed-weight indicators show changes in price only (see page 184).

Use

Deflators are valuable for identifying trends and obtaining advance warning of price changes in many areas.
	The consumer expenditure deflator is an important alternative to the consumer prices index (see page 205). The overall GDP deflator (also known as the index of total home costs per unit of home output) is the best indicator of overall economy-wide inflation.
	Since deflators cover many items and price movements selfcancel to some extent, deflators do not fluctuate as much as narrower indices such as those covering consumer prices or producer prices.
