RDP2003-01: Business Surveys and Economic Activity 7. Conclusion

Business surveys provide a potentially useful source of information about the economy. They benefit from being released well ahead of comparable official statistics. In some cases they fill in gaps in the official statistics and may provide a different perspective on an issue. Some of the questions posed are forward-looking, and quite often they provide supplementary information on areas of special interest. However, care needs to be taken in using the information in business surveys. They are subject to a number of possible sources of measurement error. Moreover, much of the information is qualitative and therefore needs to be summarised in the form of a net balance statistic, which can be difficult to interpret.

Looking at the performance of the various Australian business survey indicators over the past decade or so, a few observations can be made:

• in terms of information on the current state of the economy, business survey series such as business conditions (including sales/output), employment and selling prices, provide some information about developments in important macroeconomic variables. Timeliness is the obvious advantage, but some of the surveys also appear to be forward-looking;
• the profits component of surveys also performs relatively well in a number of the surveys, in particular for quarterly movements in profits, though they track the company profits measure of corporate profitability better than the national accounts measure of gross operating surplus;
• the results for capital expenditure are mixed. The surveys do not appear to offer much in the way of forecasting quarterly movements in business investment, though the two surveys looking at expectations over the year ahead (ACCI-Westpac and NAB) perform reasonably well; and
• generally speaking, the stocks and export components of the business surveys do not offer much, though there are some isolated exceptions such as the Dun & Bradstreet survey for stocks.

In terms of turning points, the business conditions component of business surveys has probably been a coincident indicator of turning points – and therefore has a timing advantage over official statistics – though they have provided a false signal on occasion. This component has identified periods when the economy is going into, and exiting from, major economic downturns such as the early 1990s recessions, but has been less successful at identifying turning points associated with smaller cyclical variations. In contrast, the employment and selling prices components of business surveys have been useful leading indicators of turning points in comparable official statistics, though to some extent this may reflect the lagging relationship that these indicators have with other important macroeconomic variables, such as growth in output.

Given the variability across the survey components, and even between the different surveys, the recommended approach is to focus on a range of surveys. Common themes can then be extracted and light shed on certain sectors, or aspects of the economy, where timely official data are not available. This can then be compared with other partial indicators of the economy such as growth in credit, share prices and corporate profitability. This approach also ensures that too much weight is not placed on a single result.