Business and Consumer Surveys: accurate and timely indicators complement official statistics

Business and consumer surveys are a proven, and increasingly sophisticated, tool for economic analysis. The results of such surveys reflect economic agents’ judgements about past, current and future economic developments. They provide policymakers, economists and business managers with useful information to assess the current state of the economy and forecast short term developments. The business and consumer surveys produced by ECFIN are used, among others, by the ECB to monitor inflation expectations and other economic variables.

The Joint Harmonised EU Programme of Business and Consumer Surveys (BCS) was set up in 1961, with the first survey, the harmonised business survey in industry, conducted in 1962. Since then the scope of the programme has expanded considerably in terms of both countries and sectors covered.

Currently six harmonised surveys are conducted on a monthly basis: industry, construction, consumers, retail trade, services and financial services. In addition, an investment survey is conducted twice a year and additional questions are posed on a quarterly basis covering industry, consumers, services, construction and financial services. The surveys are unique in that they cover all 27 EU Member States as well as Croatia, the Former Yugoslav Republic of Macedonia and Turkey. More than 125,000 firms and over 40,000 consumers are surveyed every month across the EU.

‘Hard data’ vs. ‘soft data’

The surveys use ‘soft data’ to reflect the sentiment – optimistic, pessimistic or neutral – of managers and consumers. This soft data is surprisingly robust: many of the sentiment indicators closely track the ‘hard data’ (official statistics). The Industrial Confidence Indicator, for example, has closely tracked actual industrial production, as measured by the Industrial Production Index, for nearly 30 years. Similarly, the euro area Economic Sentiment Indicator (ESI) is highly correlated (0.94) with real GDP growth.

The high correlation between sentiment (soft data) and actual results (hard data) is important because it means that the various sentiment indicators can be used for economic surveillance and short-term forecasting. ECFIN’s Markov-switching model, for example, has been used to accurately predict economic turning points, including the recent recession. ‘In the case of the recent crisis, our soft data predicted turning points with a high degree of accuracy,’ states Kristine Vlagsma, Head of ECFIN’s unit for economic studies and business cycle surveys.

Advantages: accuracy, timeliness and frequency

The main advantage of soft data is timeliness. Unlike the official statistics, soft data are typically available within weeks, while hard data are typically available with a minimum delay of 45 days. ‘There’s also an “inertia” in hard data,’ observes Vlagsma. ‘Sentiment may be going up or down while hard data stays the same.’ Expectations adjust much faster, for example, than inventories or order books do.

Another advantage of soft data is that it can be collected with a high frequency – on a monthly basis, and that soft time series – which reflect opinions as expressed at a certain point in time – are not subject to revisions, contrary to what happens with hard data.

Despite these advantages, sentiment indicators do have their limitations: they are less good at forecasting the magnitude of changes. And it is up to economic analysts to determine whether a diverging signal is significant or just a blip.

A solid methodology

Most of the indicators are computed using questions based on a three-option ordinal scale. Participants in surveys answer ‘positive’ (‘increase’, ‘improve’, etc.), ‘neutral’ (‘remain unchanged’, ‘sufficient’, etc.) or ‘negative’ (‘decrease’, ‘not sufficient’, etc.) to questions. The answers are usually then aggregated in the form of a balance statement which is the difference between the percentages of respondents giving positive and negative replies. Institutes at the country level conduct the actual surveys and apply inner weights to reflect the structure of each economy. Weights are calculated based on a sector’s value-added relative to GDP. ECFIN then applies weights at the European level to reflect the size of each country in the EU economy.

The role of national institutes

Administering the surveys and collecting the data is not a trivial exercise. The annual programme budget is EUR 6 million and ECFIN works with 50 institutes throughout Europe, one or more in each Member State. Some of the institutes are government institutions while others are commercial organisations. The institutes work under a three-year framework contract in which ECFIN covers 10-50% of the cost of administering surveys via a system of grants. ECFIN also takes great pains to ensure that the data gathered are accurate. Questions are carefully screened to ensure that they are robust for all cultures and languages. And ECFIN holds an annual workshop with the institutes to discuss both methodological and practical implementation issues, such as changes in questions or sector coverage. Both institutes and the programme as a whole are carefully audited. Institutes have to provide reports about questions they have asked and other information. Moreover, ECFIN checks the procedures and accounts of a random sampling of institutes. An external audit of the programme is also conducted by the EU Court of Auditors.

Additional uses and future developments

The sentiment indicators have proven their usefulness, not only in predicting economic turning points and measuring the economic climate, but also for understanding the behaviour of economic actors. The data have been used, for example, to analyse how consumers’ inflation perceptions have evolved since the cash changeover to the euro: in that case, the discovery of divergences between perceived and actual inflation pointed to the necessity of a better communication policy.

One of the greatest assets of the BCS programme is the availability of historical data. The availability of long time series is particularly useful for researchers who analyse long-term trends. With this in mind, Vlagsma and her team plan to target new applications for the data, as well as in-depth academic studies. In this vein, they regularly participate in academic conferences such as that of the Centre for International Research on Economic Tendency Survey (CIRET).

Containing a wealth of information stretching back as far as four decades, the business and consumer surveys have become a major asset. Yet ECFIN produces these surveys with a small team of just 7 people.

‘We’re using the taxpayer’s money very well – offering a good product at a competitive price,’ says Vlagsma.