ESSENTIAL MECHANICS

"Please find me a one-armed economist so we will not always
hear `On the other hand ... "'
Herbert Hoover, US president

When interpreting economic figures it is important to distinguish between the effects of inflation and changes in the real level of economic activity. Indicators measure one of three things:

		volume, such as tons of steel or barrels of oil;

		value, such as the market value of steel or oil produced in one

month or year; or

		price, such as the market price of 1 ton of steel or 1 barrel of oil.

The relationship between these three is simple. Volume times price equals value (see Table 2.1).

There is one possible complication. If the volume of oil or steel produced each year is valued in the prices ruling in, say, 1995, the result is an indicator of output in "1995 price terms". Such a series is in money units, but it is a volume indicator because it provides information about changes in volumes not prices. This is known also as output in constant prices, real prices or real terms.

The value of steel output measured in actual selling prices is known as a current price or nominal price series or a series in nominal terms. Thus:

This chapter looks at some basic methods of interpreting numbers and some of the common associated problems. It also lays the groundwork for analysing any kind of economic data.

Volume, value and price

		values, current prices, nominal prices and nominal terms
	include the effects of inflation; while

		volumes, constant prices, real prices and real terms
	exclude any inflationary influences.

VOLUME, VALUE AND PRICE

Table 2.1 OPEC crude oil production and prices

	volume					times	price	equals	value
					Value of
	Production					Price	production
	m barrels					$/barrel	$bn

1994				9,165	15.65	143.4
1995				9,309	17.19	160.0
1996				9,505	20.52	195.0
1997				9,928	19.11	189.7
1998				10,165	12.64	128.5

Sources: International Energy Agency; Petroleum Economist.

In Table 2.2 column A shows the money value of annual American economic output (gross domestic product or GDP, see page 25), which reflects changes in both output and prices. The next two columns disentangle these factors. Column B shows the volume of output with all goods and services measured in 1992 prices. Column C indicates the path of inflation (but see the comment on current-weighted index numbers below).

The value of output rose in 1990-91 (from $5,744 billion to $5,917 billion), yet in terms of the prices ruling in 1992, real output fell over the same period (from $6,139 billion to $6,079 billion).

Price indicators used to convert between current and constant prices (to deflate) are sometimes called price deflators.

		Current price series divided by constant price series (X 100)
	equals the price deflator.

		Current price series divided by price deflator (X 100) equals
	the constant price series.

		Constant price series times the price deflator (= 100) equals
	current price series.

Any series of numbers can be converted into index numbers, as described below for the constant price series in Table 2.2 column E.

Step 1		A reference base is selected, 1992 in this case.
Step 2		The value in the reference base is divided by 100
(6,244 = 100 = 62.44).
Step 3		All numbers in the original series are divided by the
result of step 2.

For example, the index value for 1990 is 6,138 = 62.44 = 98.3.






2 ESSENTIAL MECHANICS

Table 2.2 US GDP

		Constant				Price	Current	Constant
	Current				1992	deflator	prices	prices
	prices				prices	index	index	index
	$bn			$bn		1992=100	1992=100	1992=100
	A			B		C	D	E

1989			5,438.7	6,060.4	89.7	87.1	97.1
1990			5,743.8	6,138.7	93.6	92.0	98.3
1991			5,916.7	6,079.0	97.3	94.8	97.4
1992			6,244.4	6,244.4	100.0	100.0	100.0
1993			6,558.1	6,389.6	102.6	105.0	102.3
1994			6,947.0	6,610.7	105.1	111.3	103.5
1995			7,265.4	6,761.7	107.5	116.4	102.3
1996			7,661.6	6,994.8	109.5	122.7	103.4
1997			8,110.9	7,269.8	111.6	129.9	103.9
1998			8,510.7	7,551.9	112.7	136.3	103.9

Source: US Commerce Department; IMF; DECD.

Index numbers

Index numbers are values expressed as a percentage of a single base figure. For example, if annual production of a particular chemical rose by 35%, output in the second year was 135% of that in the first year. In index terms, output in the two years was 100 and 135 respectively.
	Index numbers have no units. Chemical production in the second year is referred to as 135, not 135 tonnes or 135%. The advantages are that distracting units are avoided and changes are easier to assess by eye. The arithmetic is straightforward, as shown in Table 2.2.

Composite indices and weighting. Frequently two or more indices are combined to form one composite index. For example, indices of consumer spending on food and on all other items might be combined into one index of total spending.

Base weighting. The most straightforward way of combining indices is to calculate a weighted average using the same weights throughout. This is known as a base-weighted index, or sometimes a Laspeyres index after the German economist who developed the first one. The following is an example of a base-weighted price index for single-person household consumption of wine and cheese each week.

Base data

INDEX NUMBERS

Item							Quantity
		Price					consumed
	1990						1999	1990 	1999

Wine			9.00	10.50
Cheese			5.00	8.00

Where prices are in, say, Deutschemarks and quantities are litres of wine/kilos of cheese:

Weekly expenditure in 1990

_ (1990 quantity of wine X 1990 price of wine) + (1990 quantity of cheese X 1990 price of cheese) _ (5 X 9.00) + (2 X 5.00) = 45.00 + 10.00 = 55.00

Weekly expenditure in 1999, based on 1990 quantities _ (1990 quantity of wine X 1999 price of wine) + (1990 quantity of cheese X 1999 price of cheese) _ (5 X 10.50) + (2 X 8.00) = 52.50 + 16.00 = 68.50

Index number for 1990 = 55.00/55.00 X 100 Index number for 1999 = 68.50/55.00 X 100

= 100.0 = 124.5

Current weighting. The problem with weighted averages is that weights usually need revising from time to time. With the consumer prices index, spending habits change because of variations in relative cost, quality, availability, and so on. One way to proceed is to calculate a new set of current weights at regular intervals, and use these to derive a single long-term index. This is known as a current-weighted index, or occasionally a Paasche index, again after its founder. The following is an example of a current-weighted price index for single-person household consumption of wine and cheese each week.

Base data

Item							Quantity
		Price					consumed
	1990						1999	1990 	1999

9.00		10.50
5.00		8.00
2 ESSENTIAL MECHANICS

Where prices are in, say, Deutschemarks and quantities are litres of wine/kilos of cheese:

Weekly expenditure in 1990, based on 1999 quantities
	_ (1995 quantity of wine X 1990 price of wine)
	+ (1995 quantity of cheese x 1990 price of cheese)
	_ (6 x 9.00) + (3 x 5.00)
	= 54.00 + 15.00 = 69.00

Weekly expenditure in 1999
	_ (1999 quantity of wine x 1995 price of wine)
	+ (1999 quantity of cheese X 1995 price of cheese)

_ (6 x 10.50) + (3 x 8.00)
= 63.00 + 24.00 = 87.00

Index number for 1990 = 69.00/69.00 X 100 = 100.0 Index number for 1999 = 87.00/69.00 x 100 = 126.1

	Neither base weighting nor current weighting is perfect. Base-, weighted indices are simple to calculate but they tend to overstate changes over time. Current-weighted indices are more complex to produce and they understate long-term changes.
	Current-weighted price indices reflect changes in both prices and relative volumes, while base-weighted versions record price changes only. The price deflator in Table 2.2 is actually current-; weighted.
	Mathematically, there is no ideal method for weighting indices; expediency usually rules. Most commonly indices are a combination of base-weighted and current-weighted. A new set of weights might be introduced every five years or so and the new index then spliced or chained to the old index. Table 2.3 shows how two indices are joined.
	It is essential to know the basis for the weighting, as illustrated above.

Chaining index numbers
Step 1 Identify one period when there are figures for both indices; 1992 in Table 2.3.
Step 2 For this period, divide the new figure by old figure;, 83 = 133 = 0.62.
Step 3 Multiply all old figures by the result; each figure in column C = figure in column A X 0.62.
Step 4 Put the rebased data with the new figures to create one E long run of data.

14

INDEX NUMBERS

Table 2.3 Chaining index numbers

		Old index
Old index			New index	rebased
	B		C

A

Chained index D

1992					100	62	62
1993					110	69	69
1994					121	76	76
1995					133	83	83	83
1996					91	91
1997					100	100
1998					110	110

Effects of reweighting/out-of-date weights. To show the effects of reweighting, consider GDP (total output) based on 1990 weights when, say, manufacturing accounted for half of all economic activity. If in 1990 manufacturing grew by 6% while all other activity was static, initial 1990 figures showed total GDP rising by 6 X 0.50 = 3%. By 1995 the results of a major survey were available and GDP from 1988 was reweighted to take account of the fact that the manufacturing sector had shrunk to a mere 10% of total GDP. As a result the revised figure for total growth in 1990 was 6 x 0.10 = 0.6%.
	This is obviously an extreme example, but index numbers can easily become distorted if one item is much less or much more significant than the others. For example, demand tends to grow most rapidly for goods and services which increase least in price, and so on rebasing these items are allocated larger relative weights.

Figure 2.1

Index numbers

Illusory convergence

us GDP INDICES		t75
1989= l OD

us GDP INDICES 199$= i0

'		Constant prices
/~'s

Current Prices/		..
Current paces

es

1989 90 91~ 92 '93		94 95 96 97~9~8
2 ESSENTIAL MECHANICS

	When looking at index numbers it is a good idea to check when they were last rebased and ask whether any component is increasing or decreasing in relative importance. The same approach should be taken to constant price series, such as the GDP data in 1992 dollars in Table 2.2, because these are essentially index numbers with a base value other than 100.

Convergence. Look out also for illusory convergence on the base. Two or more series will always meet at the base period because that is where they both equal 100 (see Figure 2.1). This can be highly misleading. When you encounter indices on a graph, the first thing to do is check where the base is located.

Measuring changes

If an index of stockmarket prices rose from 1,200 to 1,260, you could say either that it rose by 60 points or, alternatively, that it increased by 5%.
	Stating the increase as 60 points (an absolute measure) is simple and straightforward. Yet to interpret the figure it must be judged against another figure, such as the starting level. A rise of 60 points in an index standing at 120 is much more dramatic than an increase of 60 points in an index which started at 12,000.
	The percentage change (a relative measure) is easy to interpret. It indicates the size of a change when the starting level is 100. Percentages therefore provide a consistent yardstick for interpreting changes.

Calculating percentages. This is a matter of simple arithmetic. Basic rules for calculating percentage changes are given below. The various operations in the examples are similar. They are designed to minimise the number of key strokes required when using a calculator. Multiplying or dividing by 100 and adding or subtracting 1 can be done by eye.

1 To find one number as a percentage of another.

	General procedure		Example 1	Example 2
	Yas%ofX		150 as %of120	120 as %of150
Step 1		Divide Y by X	150 = 120 = 1.25	120 = 150 = 0.80
Step 2		Multiply by 100	1.25 X 100 = 125	0.80 X 100 = 80.0
		--150 is 125% of 120		-120 is 80% of 150

16

COMMON TRAPS

2 To find the percentage change between two amounts.

	General procedure			Example 1	Example 2
	Change between X
	and Y as % of X
Step I				Divide Y by X
Step 2				Subtract 1
Step 3				Multiply by 100

change between 120
and 150 as % of 120
	150 = 120 = 1.25
		1.25-1=0.25
	0.25 x 100 = 25.0
		X150 is 25%
		greater than 120

3 To find a given percentage of an amount.

General procedure

change between 150
and 120 as % of 150
	120 = 150 = 0.80
	0.80 - 1 = -0.20
-0.20 x 100 = -20.0
		-1120 is 20%
	smaller than 150

Example 1		Example 2

	X0/0 of Y			125% of 120	80% of 150
Step 1				Divide X by 100	125 = 100 = 1.25	80 = 100 = 0.8
Step 2				Multiply by Y	1.25 x 120 = 150	0.8 x 150 = 120
		X150 is 125% of 120 		-1120 is 80% of 150

4 To find an amount after a given percentage increase or decrease.

	General procedure			Example 1	Example 2
	Y increased by X%			120 increased by 25%	150 reduced by 20%
Step 1				Divide X by 100	25 '-. 100 = 0.25	-20 =.' 100 = -0.2
Step 2				Add 1	0.25 + 1 = 1.25	-0.2 + 1 = 0.8
Step 3				Multiply by Y	1.25 x 120 = 150	0.8 x 150 = 120
		-150 is 25%		-120 is 20%
		greater than 120		less than 150

Basis points. Financiers deal in very small changes in interest or exchange rates. For convenience one unit, say 1% (that is, 1 percentage point) is often divided into 100 basis points.

	1 basis points			=	0.01 percentage point
	10 basis points			=	0.10 percentage point
	25 basis points			=	0.25 percentage point
	100 basis points			= 	1.00 percentage point

Common traps

Units and changes. Do not confuse percentage points with percentage changes. If an interest rate or inflation rate increases from 10% to 13%, it has risen by three units, or 3 percentage points, but the percentage increase is 30% (3 = 10 X 100).

17
2 ESSENTIAL MECHANICS

Up and back. A percentage increase followed by the same percentage decrease results in a figure below the starting level. For example, a 50% rise followed by a 50% cut leaves you 25% worse off.

 $1,000 increased by 50% is $1,500.
 50% of $1,500 is $750.

Starting levels. A 10% pay rise for chief executives earning $500,000 a year puts an extra $50,000 in their annual pay packets. The same percentage increase for cleaners on $10,000 a year gives them a mere $1,000 extra.
	The importance of the base from which changes are calculated is also illustrated in Tables 2.4 and 2.5.
	Using Table 2.4, it could be claimed that in February 1999 the 12-month rate of inflation fell from 10% to 0%. In fact all that happened is that the increase a year earlier fell out of the 12month comparison. In this example, shop prices changed just once during the period January 1997-February 1999, perhaps owing to an increase in the rate of sales tax or VAT.

Table 2.4 When did inflation fall?

	Consumer
	prices
	index			1 month	12 months

change over

1997
January
February
1998
January
February
March
1999
January
February

100 100

100 110 110

0 0

0 10 0

0 0

	Table 2.5 shows that orders in the third quarter of 1998 were down from the previous quarter. However, the figure for the previous quarter was unusually high, and the third quarter figures were better than the first quarter's and any quarter of 1997. When comparing data over several years it is easy to overlook the distortion that can arise from using an unusually high or low starting or ending value.

GROWTH RATES

Table 2.5 Choosing the period for comparison

		change
Orders			Over four	From fourth quarter 1997
;Ebn				quarters	actual	annualised

1997
1st quarter					33.3
2nd quarter					33.8
3rd quarter					34.3
4th quarter					33.6
	Average for year					33.7
1998
1st quarter					34.3	3.0	2.1	8.6
2nd quarter					36.0	6.5	7.1	14.8
3rd quarter					35.2	2.6	4.8	6.4
4th quarter					32.9	-2.1	-2.1	-2.1
	Average for year					34.6	2.7

Growth rates

If consumer spending rises by 1% a month, by how much will it increase over a full year? Not 12%, but 12.7%. Each month expenditure is 1% greater than the month before and each percentage increase is calculated (compounded) from a higher base. Thus 12.7% a year is the same as 1% a month annualised. It is important to distinguish between the following terminology (numerical examples from Table 2.5).

12-month or 4-quarter change. This compares one month or quarter with the same one in the previous year. For example, orders rose 2.6% between the third quarters of 1997 and 1998.
Change this year. This compares the latest figure with the very end of the previous year. For example, when third quarter figures for 1998 were published, commentators might have said that orders had risen by 4.8% over the three quarters to the third quarter of~ 1998.
Annualised change. This is the change which would occur if the movement observed in any period were to continue for exactly 12 months. For example, orders rose 6.4% annualised during the first three quarters of 1998.
Annual change. This compares the total or average for one calendar or fiscal year with the previous one. For example, orders in 1998 were 2.7% higher than in 1997.
Change to end-year. This compares end-year with end-year: e.g., orders fell by 2.1% over the four quarters to end-1998.
2 ESSENTIAL MECHANICS

Table 2.6 Annualised and doubling rates

Observed rate

Doubling		Annualised rate if the observed rate
tune			quarterly	monthly

	0.1	693.5	0.4		1.2
	0.2	346.9	0.8		2.4
	0.3	231.4	1.2		3.7
	0.4	173.6	1.6		4.9
	0.5	139.0	2.0		6.2
	0.6	115.9	2.4		7.4
	0.7	99.4	2,8		8,7
	0.8	87.0	3.2		10.0
	0.9	77.4	3.6		11.4
	1.0	69.7	4.1		12.7
	1.1	63.4	4.5		14.0
	1.2	58.1	4.9		15.4
	1.3	53.7	5.3		16.8
	1.4	49.9	5.7		18.2
	1.5	46.6	6.1		19.6
	1.6	43.7	6.6		21.0
	1.7	41.1	7.0		22.4
	1.8	38.9	7.4		23.9
	1.9	36.8	7.8		25.3
	2.0	35.0	8.2		26.8
	2.5	28.1	10.4		34.5
	3.0	23.4	12.6		42.6
	3.5	20.1	14.8		51.1
	4.0	17.7	17.0		60.1
	4.5	15.7	19.3		69.6
	5.0	14.2	21.6		79.6
	5.5	12.9	23.9		90.1
	6.0	11.9	26.2		101.2
	6.5	11.0	28.6		112.9
	7.0	10.2	31.1		125.2
	7.5	9.6	33.5		138.2
	8.0	9.0	36.0		151.8
	8.5	8.5	38.6		166.2
	9.0	8.0	41.2		181.3
	10.0	7.3	46.4		213.8
	11.0	6.6	51.8		249.8
	12.0	6.1	57.4		289.6
	13.0	5.7	63.0		333.5
	14.0	5.3	68.9		381.8
	15.0	5.0	74.9		435.0
	16.0	4.7	81.1		493.6
	18.0	4.2	93.9		628.8
	20.0	3.8	107.4		791.6

20

THE ARITHMETIC FOR DEALING WITH GROWTH RATES

it

flow to use Table 2.6. Locate in column 1 any observed rate, say a 1 monthly increase in consumer prices. If this rate continues, prices will double after almost 70 months (column 2) and increase by 12.7% in a year (final column). If the 1% change took place over one quarter (three months), the doubling time is 70 quarters (column 2) and the annual rate of increase is 4.1% (column 3).

Table 2.6 shows annualised rates for a selection of simple rates. American commentators tend to focus on annualised rates. This makes it easy to compare monthly or quarterly changes with perhaps more familiar annual rates, but it can be highly misleading. Many economic figures bump around from month to month, and annualised rates exaggerate erratic fluctuations. A mere 0.1
change in a month adds 1.2% to the annualised figure. Columns C and D of Table 2.7 compare simple and annualised changes and show how annualising can emphasise erratic fluctuations.
	Each week the economic indicators pages of The Economist show changes in indicators such as industrial production and retail sales in two different ways: the percentage change over 12 months and the percentage change between the latest three months and the previous three months at an annual rate. The second measure is more susceptible to erratic short-term influences but it is useful for spotting whether economic activity is speeding up or slowing down.

The arithmetic for dealing with growth rates

1 To find the growth rate over several periods when the rate over one period is known.

	General procedure			Example 1	Example 2
	r% per period			0.3% per month	7.5% per annum
	over n periods			for 12 months	for 10 years
Step 1				Divide r by 100	0.3 = 100 = 0.003	7.5 = 100 = 0.075
Step 2				Add 1'	0.003 + 1 = 1.003	0.075 + 1 = 1.075
	Step 3 Raise to power of n			1.00312 = 1.037	1.075 = 2.061
Step 4				Subtract 1	1.037 - 1 = 0.037	2.061 - 1 = 1.061
Step 5				Multiply by 100	0.037 X 100 = 3.7	1.061 X 100 = 106.1
--Growth of 0.3% per -Growth of 7.5% per
	month annualises year equals a 106.1%
		to 3.7% a year increase over 10 years

Note on step 3. Raising a number to the power of n is a shorthand way of saying multiply it by itself n times. For example, 23 - 2 X 2 X 2 = 8. Use the calculator key marked xy (the letters might be slightly different) to perform this operation. If there is no xy key use logarithms (the LOG and lOx or LN and ex keys).

21
2 ESSENTIAL MECHANICS

Replace step 3 with the following.

Step 3a		Take the log	Log 1.003 = 0.0013	Log 1.075 = 0.0314
Step 3b		Multiply by n	0.0013 X 12 = 0.0156	0.0314 X 10 = 0.314
	Step 3c Take the antilog	Antilog 0.0156 = 1.037	Antilog 0.314 = 2.061

The formula for these calculations is [(1 + =/100)n - 1] X 100 or, for PC spreadsheet users, (Exn(LN(1 + r/100)*n) -1)*100.

2 To find the growth rate over one period when the
over several periods is known.

General procedure

Example 1		Example 2

r% over n periods 3.7% over 12 months
Step 1	Divide r by 100	3.7 = 100 = 0.037
Step 2	Add 1	0.037 + 1 = 1.037
	Step 3 Raise to power of 1/n	1.0371112 = 1.003
Step 4	Subtract 1	1.003 - 1 = 0.003
Step s	Multiply by 100	0.003 X 100 = 0.3
		-Growth of 3.7% per
		annum equals
		0.3% per month

106.1% over 10 years
106.1 = 100 = 1.061
	1.061+I=2.061
	2.0611110 = 1.075
	1.075 - 1 = 0.075
	0.075 x 100 = 7.5
-Growth of 106.1%
over 10 years equals
		7.5% per annum

Note on step 3. If your calculator does not have an xl1y key, use logarithms (the LOG and lOX or LN and ex keys). Replace step 3 with the following.

Step 3a Take the log Log 1.037 = 0.0156 Log 2.061 = 0.314 Step 3b Divide by n 0.0156 = 12 = 0.0013 0.314 = 10 = 0.0314 Step 3c Take the antilog Antilog 0.0013 = 1.003 Antilog 0.0314 = 1.07s

The formula for these calculations is [(1 + =/ICI/n - 1] X 100 or, for Pc spreadsheet users, (Exi'(LN(1 + r/100)/n) - 1)*100.

Moving averages

One way to smooth out erratic fluctuations is to look at an average. When reviewing, say, total high street sales in June, you might take an average of figures for May, June and July. A sequence of such averages is called a moving average. Column E of Table 2.7 and the footnote show the calculation of a threemonth moving average for a short run of data.
	The moving average can average any number of periods. A five-year moving average helps to smooth out the economic cycle described on pages sl-s5, although a lot of data would be needed to calculate it. Moreover, the more periods covered by a moving average, the slower it will be to show changes in trend.

22

SEASONALITY

seasonality

Most economic figures show a seasonal pattern that repeats itself every year. For example, prices of seasonal foods rise in the winter, sales of beachwear increase with the onset of summer, and industrial production falls in the months when factories close for annual holidays.

Seasonal adjustment. There is a simple numerical process called seasonal adjustment which adjusts raw data for the observed seasonal pattern (described in detail in The Economist Numbers Guide). Briefly, if sales or output in February are typically 85% of the monthly average, the seasonal adjustment process divides all observations for February by 85%.
	Many published figures are seasonally adjusted to aid interpretation, but it is important to remember that seasonal adjustment is not infallible. For example, in a particularly cold month energy use increases by more than the amount expected by seasonal adjustment, while more building workers than usual are temporarily laid off. The adjusted figures might be erroneously taken to suggest that energy use or unemployment was rising when the underlying situation was very different. Climatic and other influences might be overlooked when viewing the economy from the comfort of seasonally adjusted data.

Coping with seasonality and blips. Table 2.7 indicates some problems of interpreting data which are subject to erratic or seasonal influences.

The figures in column A are an index of retail sales. At first glance it appears that sales in January 1999 were very poor, since there was a 4% decline from the previous month (column B). It seems that this interpretation is confirmed because the 4% fall is worse than the 0.2% decline in the same month a year earlier.
The percentage changes over 12 months (column D) give some encouragement. They indicate that the trend in sales is upward, though growth over the 12 months to January 1999 (6.2%) was slacker than in the previous few months (around 10%).
Column F smooths out short-term erratic influences by comparing sales in the latest three months with sales in the same three months a year earlier. This suggests that the fall in January was not as severe as it appeared at first glance, with the 12-month growth rate remaining at close to 10%.

This final interpretation is the correct one. Indeed, a full run of

23
2 ESSENTIAL MECHANICS

figures would show that sales fell in January only because this was a correction to an exceptionally steep rise in the earlier few months.

Table 2.7 Analysing seasonal and erratic influences

					3-month
Retail				% change over	moving average
sales			1	1 month	12	% change from
index		month annualised months	Index 12 months ago
	A		B		C	D	E	F
1997
Oct			119.2
Nov			120.2	0.8
Dec			121.8	1.3			120.4
1998
Jan			121.6	-0.2			121.2
Oct			130.7	-0.5	-5.8	9.6
Nov			133.4	2.1	27.8	11.0
Dec			134.4	0.7	9.3	10.3	132.8	10.3
1999
Jan			129.2	-3.9	-37.7	6.2	132.3	9.2~

Note: Calculations for January 1999:
Column B: [(129.2 = 134.4) - I] X 100 = -3.9%
Column C: [(129.2 = 134.4)1Z - lj x 100 = -37.7%
Column D: [(129.2 = 121.6) - 1] X 100 = 6.2%
Column E: (133.4 + 134.4 + 129.2) = 3 = 132.3
Column F: [(132.3 = 121.2) - I] x 100 = 9.2%

Commentators are inclined to interpret blips as changes in trend. In general you should examine a run of data, form a view about the trend, and stick to it until there is clear evidence that the trend has changed.

24

MEASURING ECONOMIC
ACTIVITY

"GDP should really stand for grossly deceptive product."
The Economist

Total economic activity may be measured in three different but equivalent ways.
	Perhaps the most obvious approach is to add up the value of all goods and services produced in a given period of time, such as one year. Money values may be imputed for services such as health care which do not change hands for cash. Since the output of one business (for example, steel) can be the input of another (for example, automobiles), double counting is avoided by combining only "value added", which for any one activity is the total value of production less the cost of inputs such as raw materials and components valued elsewhere.
	A second approach is to add up the expenditure which takes place when the output is sold. Since al] spending is received as incomes, a third alternative is to value producers' incomes.
	Thus output = expenditure = incomes.
	The precise definition of economic activity varies. The three main concepts are gross domestic product, gross national product and net national product.

Gross domestic product. GDP is the total of all economic activity in one country, regardless of who owns the productive assets. For example, Britain's GDP includes the profits of a foreign firm located in Britain even if they A remitted to the firm's parent company in another country.

Gross national income or gross national product. cNI, a term which has replaced GNP in national accounts, is the total of incomes earned by residents of a country, regardless of where the assets are located. For example, Britain's GNI includes profits from British-owned businesses located in other countries.

Net national income. The "gross" in GDP and GNI indicates that

25
