SCOTTISH EXECUTIVE

[Previous] [Contents] [Next]

Scottish Economic Statistics 2000

A2 Output, Employment and Income Multipliers, Scotland 1996

Maria Melling and Lynn Graham⁷, Scottish Executive

Introduction

The 'Scottish Input-Output Tables and Multipliers for 1996' were published on 27 October 1999. The purpose of this paper is to provide an overview of the derivation and use of multipliers from the input-output tables and to use illustrative examples to explain the use of economic multipliers as a tool in carrying out various economic analysis of the Scottish economy. The article '1994 Scottish Input-Output Tables' (Scottish Economic Bulletin no 55, 1997) provides a detailed description of the basic Scottish input-output balances.

The Scottish input-output tables provide a complete picture of the flows of products and services in the economy for the year in question. They illustrate the flows between various industries and also between industries and the Final Demand sectors; namely consumers, government, investment, stocks, tourists and exports. These linkages allow estimates to be made of the extent to which Scottish industries contribute both directly and indirectly to the various Final Demand sectors within Scotland.

For example, if there is an increase in Final Demand for a particular commodity, there will be an increase in the output of that commodity, as producers react to meet the increased demand; this is the direct impact. As these producers increase their output, there will also be an increase in demand on their suppliers and so on down the supply chain; this is the indirect impact. As a result of the direct and indirect impacts the level of income throughout the economy will increase, a proportion of this increased income will be re-spent on final goods and services: this is the induced effect. The ability to quantify these multiplier effects is important as it allows economic impact analyses to be carried out on the Scottish economy. Input-output models enable us to derive sets of disaggregated multipliers, recognising that the total impact on output (income or employment) will vary according to the sector which experiences the initial increase in demand. It should be noted that, when discussing the direct, indirect and induced effects calculated within the input-output framework, full displacement and crowding-out have not been assumed. Taking these issues into account is beyond the scope of the analysis in this paper.

Derivation of Multipliers

The main concept of the multiplier is based on the recognition that the various sectors that make up the economy are interdependent. This means that in addition to purchasing primary inputs, such as labour and imports, each sector will also purchase intermediate goods and services produced by other companies within the local economy, in this case Scotland. Manipulation of the Input-Output tables allows estimation of different types of multipliers depending on whether there is an interest in output, employment or income effects. Annex A presents detailed 1996 output, employment and income multipliers for Scotland.

Leontief Inverse

The main building block for calculating multipliers is the Leontief Inverse matrix. This is derived from the symmetric industry-by-industry use matrix and shows how much of each industry's output is required, in terms of direct and indirect requirements, to produce one unit of a given industry's output.

Output Multipliers

Type I

The Type I output multiplier for an industry is expressed as the ratio of direct and indirect output changes to the direct output change due to a unit increase in final demand. Multipliers therefore represent marginal changes and cannot strictly be applied to large changes. The multiplier is derived by summing the entries in the relevant column of the Leontief Inverse matrix.

Type II

It is also possible to calculate a Type II output multiplier. The Type II output multipliers are expressed as the ratio of direct, indirect and induced output changes to the direct output change due to a unit increase in final demand. These multipliers take into account increased consumption and therefore output which will be generated from higher personal incomes. Consumers' spending is included in the calculation of the Type II multipliers by expanding the industry-by-industry matrix to include the income from employment row and the consumers expenditure column.

Although the output multiplier represents total requirements per unit of final output, it is not a particularly useful concept except as an indicator of the degree of structural interdependence between each sector and the rest of the economy. Consequently, when undertaking economic impact studies we are more interested in income and employment generating effects and for this we require income and employment multipliers, which are calculated as follows.

Employment Multipliers

The employment multipliers tend to be the most commonly used of the multipliers made available through extensions to the input-output tables. This is due to their use in economic impact analysis which is often preoccupied with the employment effects of either industrial expansion or company closure.

The employment multiplier which is analogous to the Type I output multiplier is the ratio of direct plus indirect employment change to the direct employment change. Similarly, there is a Type II multiplier which measures the ratio of direct, indirect and induced employment change to the direct employment change.

Income Multipliers

It is also possible to calculate both Type I and Type II income multipliers. These measure the change in income (wages, salaries, profits etc.) which occur throughout the economy as a result of a change in Final Demand.

The Use of Scottish Multipliers: Illustrative Examples

The input-output tables provide a wealth of information on the interactions between industries and Final Demand. Their structure allows us to analyse the effect on the economy of different types of changes in Final Demand, for example -

• the closure of a company
• the opening of a new company
• an increase in consumer spending due to a change in, for example, disposable income
• an increase in exports, due to perhaps a drop in the value of sterling.

The input-output structure provides us with the tools to follow the Final Demand changes through the whole economy and estimate the total effect on the Scottish economy, not simply the known direct effect. The results of such analysis can then be fed into the decision-making process at the appropriate point.

The two most common uses of input-output tables are in Scottish forecasting and in economic impact analysis. The remainder of this article presents examples of these different types of analyses.

Scottish Forecasting - the effect of a change in Final Demand

This type of analysis differs from economic impact analysis in that it is concerned with predicting the effect of across-the-board changes in Final Demand rather than individual sector changes. This type of Scottish forecasting could be used, for example, to analyse an increase in consumer expenditure or an increase in exports.

In the example below, the effect of an illustrative across-the-board increase in consumer expenditure of 1% is considered. Such an increase could arise from a number of changes in the economy, namely a decrease in interest rates or a reduction in the rate of taxation that would mean that individuals will have more disposable income, or a marginal shift in consumer expenditure from imports to domestic products. It is important to stress marginal in this context. If the change was non-marginal, there would be a series of behavioural changes which could in fact alter the structure of the input-output analysis.

Table 1 below shows the amount of consumer expenditure in 1996 within Scotland broken down by industry. It illustrates that consumers spend an estimated £2.2 billion on manufacturing goods, £1.2 million on energy and water supply and £5.8 billion on distribution and catering. Consumers spend more than half of their total domestic expenditure on the distribution & catering and finance and business sectors.

Table 1: Analysis of Consumer Expenditure, 1996

Source: Scottish Executive

To calculate figures for the total effect on the economy, the original consumer expenditure column is multiplied by the Leontief Inverse matrix. For illustration, a 1% increase in consumer expenditure across the board (i.e. in proportion to 1996 expenditure on each industry's output) is considered to produce estimates of the total effect on the Scottish economy of such a change (as illustrated in the column in Table 1). This produces a monetary value for the effect on each industry's output resulting from the increase in consumer's expenditure. It is also interesting to look at the effect, both direct and indirect, on the employment levels within the economy (as shown in the final column of Table 1). This is derived by calculating the employment to unit output ratio and applying this to the increase in output generated by the increase in consumer expenditure.

The analysis in Table 1 illustrates that, for a 1 per cent increase in consumer expenditure across-the-board, there will be an estimated overall increase of 4,300 jobs in the economy as a whole. Over 30 per cent of these jobs will be in the distribution and catering industry.

It is interesting to note that the industry which would experience the largest output effect is not necessarily the industry where the largest number of jobs increase will occur. Estimates of job increases are dependant on employment to output ratios for that particular industry, hence the more labour intensive industries will see a larger proportionate increase in employment. For example, both the energy & water industry and public administration would be expected to experience increases of £22 million, as a result of an increase of 1 per cent in consumer expenditure. However, when this is translated into the related number of jobs, around 130 would be created within energy & water industries whereas around 865 would be created within public administration.

[Previous] [Contents] [Next]