5250.0 - Australian Business Expectations, Mar 2003 and Dec 2003 (Final)  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 19/12/2002  Ceased
   Page tools: Print Print Page Print all pages in this productPrint All

INTRODUCTION

1 This publication contains estimates of future economic activity based on the business expectations of senior executives, managers and proprietors of businesses operating in Australia. The estimates have been compiled from data collected by the Australian Bureau of Statistics (ABS) in its quarterly survey of business expectations.

2 This survey commenced with short term expectations for the December quarter 1993 and medium term expectations for the September quarter 1994. This quarter’s publication contains estimates of the expected change between the December quarter 2002 and the March quarter 2003 and the December quarter 2003.

3 The seasonal adjustment of these series has been reviewed. For this issue, the original, seasonally adjusted and trend series are the published indicators for short-term business expectations, with the trend providing an indication of the underlying direction of the series. The volatility of the medium-term seasonally adjusted series is such that the original and trend series will continue to be the published indicators, with the trend providing an indication of the underlying direction of the series. Seasonally adjusted estimates are provided on page 12, with an explanation in paragraphs 28 to 32 of the Explanatory Notes. Trend estimates are provided on pages 12 and 20, with an explanation in paragraph 33 of the Explanatory Notes.

4 The survey is conducted by mail each quarter. This quarter’s survey was collected during October and November 2002.

5 It is based on a stratified random sample of approximately 4,700 businesses selected from the ABS annual Economic Activity Survey (EAS). EAS in turn derives its survey population from the ABS central register of business units.

6 The sample is stratified by industry, sector (private and government business) and size of business (measured by number of employees). Within each stratum businesses are sampled randomly, with each business in a stratum having the same probability of selection.

7 The sample is supplemented by a further sample of businesses which have been added to the ABS business register since the original EAS sample was selected. This ensures that the expectations of relatively new businesses are taken into account in the overall estimates.


SCOPE OF THE SURVEY

8 The statistics in this publication relate to employing businesses in all industries and sectors of the Australian economy except: agriculture, forestry and fishing; general government.

9 Data related to intended sheep matings are collected annually by the ABS and published in catalogue 7111.0 and 7113.0 publications from the Agricultural Commodity Survey.

10 The Australian Bureau of Agricultural and Resource Economics (ABARE) publishes its forecasts for specific commodities and for the Agriculture industry generally, as part of the annual Outlook conference in February each year. ABARE updates these forecasts in its quarterly publication Australian Commodities-Forecasts and Issues.


CLASSIFICATION

11 Each statistical unit selected in the survey is classified to an industry according to the Australian and New Zealand Standard Industrial Classification (ANZSIC).


BUSINESS SIZE

12 Data presented in this publication is classified by three business sizes :

  • small (less than 20 employees, except for manufacturers where it is less than 100);
  • medium (20 to 99 employees, except for manufacturers where it is 100 to 599 employees); and
  • large (100 or more employees, except for manufacturers where it is 600 or more employees).

PROPORTION OF BUSINESSES SELECTED BY SIZE WITHIN AUSTRALIA

Small
Medium
Large
All businesses
%
%
%
%

Manufacturing
61.6
14.4
24.0
100.0
Other industries
51.0
22.2
26.7
100.0
All industries
53.3
20.6
26.2
100.0



STATISTICAL UNIT

13 The statistical unit used in the survey of business expectations is the management unit. The management unit is the highest level accounting unit within a business for which sub-annual accounts are maintained, having regard for industry homogeneity.

14 In nearly all cases the management unit coincides with the legal entity owning the business (i.e. company, partnership, trust, sole proprietor, etc.).

15 In the case of large diversified businesses, however, there may be more than one management unit, each coinciding with a ‘division’ or ‘line of business’.


BUSINESS PERFORMANCE INDICATORS

16 The survey uses a set of well recognised economic trading indicators in measuring future trading activity. These indicators are: Operating income, selling prices, operating expenses, employment, etc. See Glossary for details.

17 The survey asks for full-time equivalent paid persons working. This is not a usual definition of employment as used by the ABS. It would be incorrect to assume a direct comparison with labour force statistics or other ABS employment statistics for instance.


SIMPLE AND WEIGHTED NET BALANCE

18 The simple net balance for a selected indicator is estimated by subtracting the percentage of respondents predicting a ‘fall’ from the percentage of respondents expecting a ‘rise’.

19 The net balance is a qualitative statistic best suited to indicating the sentiment of businesses about future business conditions, and measures the net proportion of businesses predicting a rise or fall in future business conditions.

20 The weighted net balance is estimated by weighting the surveyed direction of change for each unit by its benchmark level response for the equivalent variable in the EAS.

21 Weighting the responses enables larger businesses to have an influence upon the net balance proportional to the level of their expenditure, employment size, etc. Movements in the weighted net balance indicate the net proportion of business activity predicting a rise or fall in future business conditions.


EXPECTED AGGREGATE CHANGE

22 The expected aggregate change measures the forecasted percentage change in the level of a particular indicator. It is estimated by weighting the expected percentage change reported by respondents to the survey by their proportion of aggregate sales, expenditure, employment, etc. in the economy as measured from the benchmark estimate in the EAS.

23 The weighted aggregate estimate of a particular indicator, combined with an estimated level, can be used to quantify its expected future movement.


COMPARISON OF RESULTS

24 The weighted net balance and expected aggregate change are complementary measures which, in combination, give a broad indication of future business conditions. It is possible to obtain estimates in opposite directions for the net balance and weighted aggregate change estimates.

25 The weighted net balance provides a qualitative measure of the proportion of businesses predicting the direction of change in future business conditions. The expected aggregate change, however, provides a quantitative measure predicting the magnitude of change in a selected variable.

26 A comparison of the various expectations measures is provided in the following table:

COMPARISON OF THREE MEASURES OF BUSINESS EXPECTATIONS

Simple net balance
Weighted net balance
Weighted aggregate
%
%
%

Operating income
-0.9
-6.9
-2.4
Wage costs
9.8
14.8
-0.7
Employment
Full time equivalent
-2.7
-5.2
-1.1



27 For the March quarter 2003 a simple net balance of 2.7% of businesses expect a decrease in employment. The decrease in the employment weighted net balance of 5.2% indicates that the businesses expecting a decrease in employment tend to be larger than those expecting an increase .


SEASONAL ADJUSTMENT

28 The quarterly business expectations series in this publication are affected to some extent by seasonal influences and it is useful to recognise and take account of this element of variation.

29 Seasonal adjustment may be carried out by various methods and the results may vary slightly depending on the procedure adopted. Accordingly, seasonally adjusted statistics are in fact only indicative and should not be regarded as in any way definitive. In interpreting seasonally adjusted data it is important to therefore bear in mind the methods by which they have been derived and the limitations to which the methods used are subject.

30 At least once each year the seasonally adjusted series are revised to take account of the latest available data. The most recent reanalysis takes into account short-term expectations collected up to and including the December quarter 2002, and medium-term expectations collected up to and including the September quarter 2003. Data for subsequent periods are seasonally adjusted on the basis of extrapolation of historical patterns. The nature of the seasonal adjustment process is such that the magnitude of some revisions resulting from reanalysis may be quite significant, especially for data for more recent quarters. Care should be exercised when interpreting quarter to quarter movements in the seasonally adjusted series in the publication, particularly for recent quarters.

31 It should be noted that the seasonally adjusted figures necessarily reflect the sampling and other errors to which the original figures are subject.

32 Details of the seasonal adjustment methods used, together with selected measures of volatility for these series, are available upon request.


TREND ESTIMATES

33 The trend estimates are derived by applying a 7–term Henderson moving average to the published and unpublished seasonally adjusted series. The 7–term Henderson average (like all Henderson averages) is symmetric, but as the end of a time series is approached, asymmetric forms of the average are applied. Unlike the weights of the standard 7–term Henderson moving average, the weights employed here have been tailored to suit the particular characteristics of individual series. While the asymmetric weights enable trend estimates for recent quarters to be produced, they can result in revisions to the estimates for the most recent three quarters as additional observations become available. There may also be revisions because of changes in the original data and as a result of the re-estimation of the seasonal factors. For further Information, see A Guide to Interpreting Time Series - Monitoring Trends: an Overview (Cat. no. 1348.0) or contact the Assistant Director, Time Series Analysis on (02) 6252 6345.


RELIABILITY OF ESTIMATES

34 All of the estimates in this publication are subject to:
    • sampling error;
    • non-sampling error; and
    • benchmark bias


SAMPLING ERROR

35 Sampling error is due to the use of a sample rather than a complete enumeration; that is, the estimates differ from the values that would have been obtained if all units were surveyed. A measure of the likely difference is given by the standard error, which indicates the extent to which an estimate might have varied by chance because only a sample of units was included. There are about two chances in three that the difference will be within one standard error, and about nineteen chances in twenty that the difference will be within two standard errors.


STANDARD ERRORS

36 The table below provides standard errors (SE) for some of the main estimates of this publication. As an example of how the standard errors can be interpreted, given that the short term expectation for Operating Income for Australia is -2.4% with a standard error of 0.4, there would be two chances in three that the true value would be within the range -2.0% and -2.8%.

37 The size of the SE may be a misleading indicator of the reliability of some of the estimates for profit. This situation may occur where an estimate may legitimately include positive and negative values reflecting the financial positions of different businesses. In these cases the aggregate estimate can be small relative to the contribution of individual businesses resulting in an SE which is large relative to the estimate.

STANDARD ERRORS OF KEY ESTIMATES, SHORT-TERM EXPECTATION, AUSTRALIA

Business performance indicator
Survey estimate
Standard error

Operating income
-2.4
0.4
Selling prices
-0.5
0.3
Profit
-15.8
2.9
Capital expenditure
1.1
1.0
Inventories
-1.5
0.4
Employment
-1.1
0.2
wage costs
-0.7
0.3



NON-SAMPLING ERROR

38 All other inaccuracies are referred to collectively as non-sampling error. The major areas of concern are: non-response; mis-reporting of data by respondents; and deficiency in the central register of economic units.

39 Every effort is made to reduce the non-sampling error to a minimum by careful design of questionnaires and efficient editing and operating procedures.

40 The expected aggregate change is designed to reflect business expectations for each business performance indicator, as accurately as possible. However, while the estimates should be appropriate measures of business climate, the expectations may not predict actual movements accurately. Businesses may be too optimistic or pessimistic in their predictions at different times.

41 In addition, actual movements would be partly comprised of activity of relatively recently formed businesses, and businesses which are formed during the expectations reference period which are not immediately represented in BES because they would not have been included on the ABS central register of economic units. Allowance is made in other ABS series for coverage deficiencies relating to newly formed businesses but no allowance for this is made in BES. This is important for some variables, where the contribution of new businesses to growth in that variable is relatively substantial (e.g. employment, capital expenditure or stocks).


BENCHMARKS

42 Benchmark (or base level) information is obtained from the ABS annual Economic Activity Survey (EAS). It is used to weight individual business responses by their relative contribution to each business performance indicator. This enables percentage responses from different businesses to be aggregated.

43 In June each year a new sample is selected from units surveyed by EAS in the previous financial year. New benchmarks are introduced for the survey conducted in August each year. The benchmark data become increasingly out of date as they are used in the surveys conducted in November, February and May. In certain cases, the benchmark data may not accurately reflect the current activity of a business. It is currently not possible to measure the extent of any such inaccuracies.

44 Results from the EAS are published in Business Operations and Industry Performance (Cat. no. 8140.0).


SYMBOLS AND OTHER USAGES

n.p. not available for publication but included in totals where applicable, unless otherwise indicated.