GDP
Source: National statistics
The Economist, January 26th 2010
Trough
Output will not fall indefinitely. It will stop at some minimum level (a trough or depression) because employees retain jobs and spending power where they work in government or in industries supplying food, basic essentials and perhaps export goods. Unemployment and welfare payments, past saving and new borrowing enable other consumers to buy essentials.
Slack demand for investment funds may pull back interest rates making new or replacement investment attractive, at least for the industries providing basic essentials. And with consumer demand steady, investment demand begins to lift the economy again.
In America a recession is technically defined as two consecutive quarters of falling GDP. The snag with this is that if GDP plunges steeply in the first and third quarters of a year, but rises slightly in the second and fourth quarters, then officially an economy has escaped a recession, even though output may end the year sharply lower. Some economists prefer to define a recession as a year-on-year fall in output. Others talk about a “growth recession“ when a country’s GDP growth rate falls below its long-term productive potential. In Japan, for example, annual growth of less than 3% has often been called a recession.
Table 4.6 shows dates for peaks and troughs in selected countries.
Table 4.6 Implied peaks and troughs in GDP Month/year
Causes
There is no general consensus about what causes cycles or even about what is a cause and what is an effect. Major influences include fixed investment and inventory cycles, external shocks and well-meaning or perhaps self-interested government policies. It is not unknown for the government of the day to engineer an economic boom just before an election, thus setting the cycle in motion. Expansionary policies aimed at reducing unemployment followed by contractionary policies to limit the inflationary consequences can cause severe cyclical swings. Moreover, measures to cool a boom may be imposed when automatic contractionary forces are already in motion, thus hastening the downhill slide.
Cyclical patterns in economic indicators
Cyclical patterns can be detected in many economic series. Peaks and troughs do not fall in step. This means that indicators which turn in advance of GDP can be used to predict economic developments. The following timings indicate roughly what might be expected in a standard cycle in an industrial economy.
(Some economists argue that interest rates are a lagging indicator. They are included here as a leading indicator since most people are familiar with the idea of, say, lowering interest rates to boost a flagging economy.)
Leading indicators
Interest rates pass their low point and begin to rise about 18 months ahead of a peak in output.
Business confidence, share prices, housing starts and companies’ financial surpluses peak 8–16 months ahead of output. Consumer credit, car sales and manufacturing orders peak about six months ahead. Retail sales peak 2–3 months in advance.
Coincident indicators
GDP establishes the reference point for the overall cycle, while other indicators which peak within a month or two of GDP are used to confirm that the economy is turning.
Lagging indicators
Manufacturing capacity utilisation peaks about a month after total output. Job vacancies peak about three months later, growth in average earnings after four months and growth in unit labour costs after five months. Productivity and unemployment stop falling and turn upwards six months after the peak in overall activity, and inflation peaks at about that time also. Investment, order backlogs and stocks peak around 12 months after output.
The cyclical indicators
Several countries and OECD statisticians have combined groups of economic indicators into composite cyclical indicators. Generally trends are removed, erratic fluctuations are smoothed, and the series are combined into weighted averages in index form.
The components of the indices vary, reflecting changes in economic habits and analysts’ understanding of the economy. For example, in the early 1990s America downgraded the role of share prices after it was found that an increase in stockmarket values was taken as an indicator of an upturn, which prompted share buying and pushed up the leading indicator still further.
Interpretation
Use composite indicators as a guide to the cycle. Leading indicators turn 6–12 months ahead of GDP; coincident indicators turn with it; lagging indicators turn perhaps six months later.
Watch mainly for changes in direction in leading indicators and use coincident indicators to confirm the change. Most composite indicators are used only to identify turning points. However, American indicators are scaled so that percentage changes are broadly suggestive of the magnitude of fluctuations in the overall level of economic activity.
Composite indicators are frequently published when only a few components are available and are revised in subsequent months as more information comes to hand and as component indicators are themselves revised. The composites should be interpreted with care and supplemented by examination of individual economic series.
Chapter 5
Population, employment and unemployment
Work is the refuge of people who have nothing better to do.
Oscar Wilde
This chapter contains a series of population and employment briefs. Among other things, they highlight the following points.
Trends
The first things to check when assessing longer-term economic trends are employment, productivity and investment. The brief on Productivity (page 52) shows their relationship to growth.
The size and age structure of the population provides an indicator of long-term pressures on the economy. GDP must grow at least as fast as the population if output per head is not to decline, while an increasing number of people of working age may signal enhanced productive potential, or higher unemployment.
Cycles
Unemployment is an excellent indicator of the state of the economic cycle. High unemployment (compared with the average over the past few years) suggests a recessionary gap. Low unemployment at the top of the cycle is broadly indicative of inflationary pressures. Note, however, that unemployment lags behind the cycle by perhaps 6–12 months (see Cyclical indicators, page 55).
Services
The services sector accounts for 65–75% of output in the major industrialised countries, but it is relatively neglected among the commonly followed indicators of output. Employment provides a useful guide to activity in the sector.
Incomes
Employment data provide guides to personal incomes, wages and unit labour costs (see pages 89, 214 and 217). These are the basis for measuring GDP on an incomes basis and they help when assessing inflationary pressures.
Table 5.1 Population
Old story
Which countries have most elderly people?
As the world gets healthier and wealthier, people are living longer. And with fertility rates declining in many rich countries, the number of elderly people as a share of those of working age is projected to rise sharply over the next 40 years, according to the European Commission. The biggest absolute increase in the old-age dependency ratio will be in Japan, which will jump from 35.1% in 2010, already the world’s highest, to 73.8% by 2050. At that point, the number of pensioners in China will be equivalent to 38.8% of its labour force, up from 11.6% in 2010. The European Union, which had 84.6m elderly people last year, will have 148.4m in 2050. And the ratio for the world as a whole will more than double to over 25%.
Old-age dependency ratios
Source: European Commission
The Economist, May 8th 2009
Population
Measures: Total number of people in a country.
Significance: Yardstick for minimum GDP growth.
Presented as: Total number.
Focus on: Age structure and changes.
Yardstick: The Euro area population grew by
0.5% a year between 1997 and 2007.
Average annual growth of 0.2% was forecast for the period 2007–25.
Released: Annually.
Overview
Real GDP must grow as least as fast as the population if living standards are not to fall. This may not be too hard in the industrial countries, where populations are expected to grow by less than 1% a year or even fall in 2007–25. It will be more difficult in sub-Saharan Africa, where populations are forecast to increase by around 2.2% a year over the same period. Developing countries elsewhere are likely to expand a little less rapidly, except for some Arab states and China. The population of China is projected to grow by 0.5% a year in 2007–25.
Migration and the age structure have important effects on output (see Labour or workforce below).
Labour or workforce
Measures: Employees plus self-employed plus the unemployed.
Significance: Indicator of maximum potential output.
Presented as: Total number.
Focus on: Structure and changes.
Yardstick: Average annual growth in the OECD labour force was 0.9% in 1998–2008.
Released: Monthly, at least one month in arrears; see Employment, page 66.
Definitions
The labour force or workforce is the number of people employed and self-employed plus those unemployed but ready and able to work. The grand total is sometimes known as the economically active population. The components of the labour force are notoriously difficult to measure (see also Employment and Unemployment, pages 66 and 68).
The labour force is defined variously to include, for example, people whose age in years is over 15 (Italy, Canada), 16 (America), or in the range 15–74 (Norway, Sweden).
5.1 Labour Force
Source: World Bank
There is a tendency to focus on the civilian labour force (that is, excluding the armed forces). Spain includes professional military personnel but excludes conscripts from its regular figures.
Changes in the labour force
The things to watch for are the three factors which affect the size of the labour force: population, migration and the proportion participating in economic activity.
Population
Birth rates in most industrial countries fell to replacement levels or lower in the 1980s. Meanwhile, earlier population growth boosted to record levels the number of 15–24 year-olds entering the labour force (exceptions include Japan and Switzerland). This implies an older workforce and higher old-age dependency rates (the number of retired people as a percentage of the population of working age) in the future. Output per employee must grow for GDP per head to remain constant. By 2025 20% of the population in industrial economies will be over 65 years of age.
Developing countries have young populations with around 30% under 15 years. This suggests an expanding working-age population with potential problems for housing and job creation.
Table 5.2 Labour force (including armed forces)
Migration
In the industrial countries inflows of foreign workers increased since the late 1980s and a substantial number of illegal immigrants were granted amnesty in America, France, Italy and Spain. The expansion of the EU to include eastern European countries has resulted in an increase in flows of legal migrants to rich European countries, and America’s less welcoming policy towards migrant labour hasn’t prevented large numbers of Mexicans from working illegally in the country.
Inward migration may be a bonus for some economies. For example, the unification of east and west Germany boosted that country’s productive potential. However, large numbers of refugees seeking asylum can have significant adverse effects on income per head.
Wealthier developing countries, especially oil producers, have large proportions of foreigners in their labour forces. Workers frequently make a substantial contribution to the balance of payments in their home countries by remitting savings from their salaries.
Participation
Participation rates (the labour force as a percentage of the total population) generally increased in the 1980s and 1990s with earlier retirement for men, especially in France, Finland and the Netherlands, generally offset by more married women entering the labour force, especially in America, Australia, Britain, New Zealand and Scandinavia.
Women account for a smaller proportion of the workforce in predominantly Muslim countries (30%) and a greater proportion in Africa (around 40%) where they traditionally work on the land.
Employment
Measures: Total employment = employees in employment plus the self-employed.
Significance: Indicator of current output potential.
Presented as: Totals.
Focus on: Structure and changes.
Yardstick: OECD employment grew by 1.0% a year between 1998 and 2008.
Released: Monthly, at least one month in arrears.
Data collection
Measuring employment is tricky. The main sources of data are censuses and surveys of population and employment. Household surveys are generally the most reliable since surveys of employers tend to double count people with more than one job.
Most countries conduct household surveys; some monthly (Australia, Britain, Japan, the United States and Canada), some quarterly (Israel, Italy, New Zealand) and some less frequently still (China, India). Figures for months between main surveys are based on employment surveys or are estimates or interpolation.
Apart from the definitional problems mentioned in the previous brief, other distortions and international inconsistencies arise owing to factors such as the method of counting home workers and domestic servants, part-time staff, people with more than one job and those temporarily ill or laid off. Full employment is usually defined as the workforce less the natural rate of unemployment. (See also Unemployment, page 68.)
Basic analysis
The level of production depends on the number of people employed, hours worked, education, training and the quality of capital equipment. (See also Productivity, page 52.)
Table 5.3 Total employment
Hours worked consists of core hours and overtime. There is a tendency for core hours to decline as economies mature and workers demand more leisure.
Interpretation
Employment and unemployment are highly cyclical. When demand increases, companies first tend to increase overtime. They take on more employees only when higher demand is perceived to be strong and durable. When demand turns down, hours are cut before jobs.
Watch hours worked and overtime for early signals, and employment for confirmation. Survey evidence (see Business conditions, page 115) indicating plans to take on or lay off workers may also provide early warning of changes in employment.
5.2 Employment
Sources: OECD; ILO
Try to identify to what extent an increase in payrolls represents second jobs. These limit employers’ ability to increase output. A high number of second jobs with low unemployment suggests that any increase in consumer demand may be inflationary.
Sectoral trends
Where employment figures are available by sector, they provide a rough-and-ready guide to output trends in parts of the economy. Employment in services is an imperfect but useful indicator for that sector since output data for service industries are hard to find.
Unemployment and vacancies
Measures: Total= people out of work but ready and able to work. Rate = unemployment as a percentage of the labour force.
Significance: Indicator of spare labour capacity (and wasted resources).
Presented as: Total number, percentage.
Focus on: Structure and changes.
Yardstick: The OECD standardised rate of unemployment averaged 6.6% between 1998 and 2008.
Released: Monthly, at least one month in arrears.
Total unemployment
Based on people registered as unemployed (Austria, Switzerland) or claiming benefit (Belgium, Britain) or on survey evidence (Britain again, and many
other countries). Surveys tend to make better indicators because they catch people who would take employment if work was available but who are not registered as unemployed.
Distortions and international inconsistencies arise owing to factors such as students claiming benefits during vacations, the treatment of people temporarily laid off, discouraged workers who do not declare themselves available for work and people who have part-time jobs but who are looking for full-time employment.
The unemployment rate
Usually defined as unemployment as a percentage of the labour force (the employed plus the unemployed). National variations are rife: Germany excludes the self-employed from the labour force; Belgium produces two unemployment rates expressing unemployment as a percentage of both the total and the insured labour force.
By changing the definition, which governments are inclined to do, the unemployment rate can be moved up or, more usually, down by several percentage points.
The International Labour Organisation (ILO) and other international organisations produce standardised unemployment rates which differ from national figures but which provide a consistent basis for cross country comparisons. The 2008 figures in Table 5.4 show some slight differences between standardised and national rates.
Table 5.4 Unemployment Standardised definition, as % of labour force
Total unemployment
Unemployment never drops to zero for various reasons.
Frictional unemployment. There are always people changing jobs and temporarily recorded as unemployed. Their number might be reduced by better information flows (bringing together vacancies and the unemployed) and training.
Structural unemployment. This indicates people whose skills and locations do not match job opportunities, usually because they were trained for industries which are collapsing under competition from modern technology and/or imports. Structural job losses can best be reduced through retraining and improving labour mobility.
Guide to Economic Indicators Page 6
