To demonstrate the relative size of government three measures will be used:
None of these measures is entirely unambiguous though they are often used without being clearly defined. The purpose of this section is to clarify these definitions one by one.
Remembering that:
it seems entirely logical to express government expenditure on goods and services (both consumption and investment) as a proportion of gross domestic product (GDP). The result gives the proportion of total annual resources used to produce government services (which are largely unpriced).
If you consulted a common source of information on international comparisons of the relative size of government, such as the Organization for Economic Co-operation and Development (OECD) or the International Monetary Fund (IMF), you would be able to draw up a table similar to that shown in Table 2.2.
| Country | % |
| USA | 16.0 |
| Japan | 10.1 |
| Germany | 19.7 |
| France | 19.4 |
| UK | 21.5 |
| Italy | 16.9 |
| Canada | 19.8 |
Source: OECD Annual Economic Surveys: Statistical Appendices
Now let us go back to Table 2.1. In the second part of it we can identify government consumption expenditure on goods and services and expenditure on capital formation as items 6 and 7. Item 6 shows that £164.9b. represented purchases of goods and services from firms. In addition, item 7 lists £10.4b. as gross domestic capital formation (investment) by government. The total of these items is £175.3b. But this represents only 52 per cent of the total government payments of £306.2b. Clearly the use of the first measure will understate the relative size of government, using an expenditure indicator!
What happens when we include such missing items can now be shown:
| Government expenditure on goods and services | |
| + transfers + debt interest + subsidies | × 100% |
| Gross domestic product |
Certainly, our definition of government expenditure is more comprehensive, but we have now included items which are not part of the denominator (GDP). Transfers and debt interest payments and subsidies are money payments against which there is no counterflow of goods and services. They represent a redistribution of resources between receivers of these payments and citizens as taxpayers. To include them could produce a situation in which the above ratio could be in excess of 100 per cent. Nevertheless, this ratio is widely used because it is argued that comprehensiveness is more important than consistency.
| Country | % |
| USA | 29.9 |
| Japan | 38.4 |
| Germany | 47.6 |
| France | 51.5 |
| UK | 39.5 |
| Italy | 47.5 |
| Canada | 40.9 |
Source: As for Table 2.2.
Using the same sources of information as in the last example, and defining the sum of the components in the numerator as ‘general government expenditure’, then you would be able to extract figures as shown in Table 2.3.
The third measure – government employment as a percentage of employed population – seems less ambiguous. However, there is one major difficulty in compiling these statistics. It is generally not possible to take account of part-time employment. What evidence is available suggests that part-time employment is more important in the government sector than in the private sector. This suggests that if the government share of total employment could be expressed in full-time equivalents then it would be less than that normally recorded. This will affect international comparisons for there will be variations between countries in the ratio of part-time to full-time employment in government.
International agencies appear to have much more difficulty in compiling international comparisons of government employment than in compiling expenditure comparisons. Using the same sample of countries as before, the latest data estimates are produced in Table 2.4.
| Country | % |
| USA | 13.2 |
| Japan | 5.9 |
| Germany | 15.3 |
| France | 25.1 |
| UK | 14.4 |
| Italy | 15.8 |
| Canada | 18.9 |
Source: As for Table 2.2.
What can we conclude about the choice of a measure of relative size? Only that the choice depends on the use to which the measure is being put. The narrow definition (i) focuses attention on government use of resources available. The wider definition (ii) focuses attention on the spread of government activities which affects all parts of the private sector, including households as well as firms. The employment measure (iii) concentrates on the relative importance of manpower resources used by government. However, the main purpose of explaining these differences at length is to alert you to the fact that if you are given the task of providing a background paper on the influence of government on the economy or of commenting on one which is presented to you by the company economists or statisticians, you have a check-list of questions to pose to yourselves or to those who are providing you with background information. The list will include the following:
What is included in the term ‘government’?
A wide or narrow definition may be used. A wide definition will include all layers of government and sometimes certain comparable expenditures of public enterprises, which do not involve ‘double counting’ in the national income accounts, the most important being public enterprises’ expenditure on gross capital formation. Similarly, care must be taken to find out whether or not a broad definition includes the numbers employed in public enterprises.
What is included in the numerator in the measure of relative size?
As explained in Sections 2.3.2 and 2.3.3, a check needs to be made on which economic transactions of government (see Table 2.1) are included and why.
What is included in the denominator?
The usual denominator used is gross domestic product at factor cost, which is a measure of the annual output of goods and services. Sometimes the denominator is given as the national income or even net national income, which excludes that part of gross capital formation represented by depreciation of the capital stock. The consequential reduction in size of the denominator will increase the relative size of government. If net national income is used then depreciation must be removed from the gross capital formation figure for the government as capital formation is included in the numerator. The difficulties associated with splitting off the narrower definition of the denominator ensure that it is rarely used, unless the investigator has a vested interest in trying to make the relative size of government look as large as possible.
Are the cross-country comparisons consistent?
When a wide variety of different sources is used for compiling comparative figures showing the relative size of government in different countries, care must be taken to see that definitions of government, government expenditure and the GDP are consistent. This problem may be avoided if the donkey work of achieving consistency has been done for one, usually by the international agencies who have a continuous interest in international comparisons of the size of the public sector. However, it cannot be completely avoided if there is a requirement for up-to-date comparative figures in some detail, e.g. figures which break down the expenditure figures more finely according to administrative or functional categories. Usually, consistent, detailed figures will be published with a lag of as much as three years, particularly in the case of government employment figures.
