Skip to main content

Are the European Commission’s Business and Consumer Survey Results Coincident Indicators for Maltese Economic Activity?


The European Commission’s business and consumer surveys are the most extensive regular surveys of Maltese firms and households. The Economic Sentiment Indicator (ESI) for Malta is closely correlated with real GDP growth, particularly when one focuses on the first vintage of national accounts data. This suggests that the opinions expressed by economic agents are partly driven by news prevailing at the time. The sectoral confidence indicators that underpin the ESI are quite highly correlated, with construction sentiment being the most synchronised with sentiment in other sectors. In general, sectoral expectations on future activity appear to be less strongly correlated to changes in national accounts sectoral value added than survey responses to planned employment changes are to observed changes in sectoral employment. Maltese household economic expectations appear to be mostly reflective of current conditions and could be useful to forecast variables that are issued with some time lag, like real GDP.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7


  1. The questionnaires can be found at

  2. For details on how the ESI is compiled see European Commission (2019), Gelper and Croux (2010) looks at alternative ways how the ESI could be constructed using more sophisticated techniques.

  3. The consumer confidence survey includes a large number of questions to households on their opinion of the current and future state of economic activity. This includes a question on how they evaluate the general economic situation evolved over the last 12 months and how they expect it to change over the next 12 months.

  4. For example, Juriová (2015) examines how foreign sentiment affects small open economies, and finds that foreign economic sentiment can be used for explaining fluctuations of domestic variables of a small open economy. This would, in turn, fit with Grech and Rapa (2019) who argue that growth in Malta has been driven by external demand more than usually thought.

  5. The GDP vintages can be found at For a review of GDP revisions for Malta see Grech (2018a, b).

  6. A Spearman's correlation was run to determine the relationship between the two variables. There was a strong, positive monotonic correlation between the two (R2 = 0.64, n = 64, p < .001).

  7. The latter mainly reflected a large upward revision in the gambling and betting sector. For details, refer to Note 6 in NSO News Release 049/2017: International Economic and Financial Transactions: Q4/2016.

  8. Gayer (2005) finds that the ESI is most useful to forecast GDP growth up to one or two quarters ahead. Moreover, the Commission survey indicators turn out to have a better tracking performance when used in annual differences rather than in levels. In this note the focus is however on levels. This is in line with more recent literature, particularly in factor modelling, such as Giannone et al. (2009) and Bulligan et al. (2012), where qualitative variables are treated as stationary in levels.

  9. Malta and Ireland appear to follow a similar historical path of development. As shown in International Monetary Fund (2003), the synchronisation in business cycles for Malta and Ireland with the EU-3 (France, Germany and Italy) stood at 0.10 for Malta and 0.08 for Ireland for the period 1971–1990, rising to 0.42 and 0.57 respectively for the period 1990–2002.

  10. The relative RMSE for the simple GDP ESI equation stands at 0.6 for the four and three- quarter ahead forecast and at 0.7 for the two-period ahead forecast. These RMSEs were calculated using a 20-quarter moving quarter sample period.

  11. A state space model is then defined, with the common factors and the idiosyncratic components modelled as unobserved states. The equation in the text would be the measurement equation, linking the unobserved states with the data. The transition equations would be \(f_{j,t} = a_{j} f_{j,t - 1} + u_{j,t}\), \(u_{j,t} \, \sim \,N\left( {0,\sigma_{{u_{j} }}^{2} } \right)\) for \(j = 1, \ldots ,r\) and \(e_{i,t} = \rho_{i} e_{i,t - 1} + \varepsilon_{i,t}\), \(\varepsilon_{i,t} \, \sim \,N\left( {0,\sigma_{{\varepsilon_{i} }}^{2} } \right)\). These equations describe the dynamics of the system.

  12. In order to have a more meaningful number of observations, this analysis also includes more recent national account vintages, with the final set of data forecasted being the 2019Q3 GDP growth rate.

  13. An AR(4)-MA(1) process was selected for the seasonally unadjusted Year-on-Year GDP growth rate. This derives from an ARIMA selection method, based on both the AIC and the BIC.

  14. Note that Quarterly National Accounts aggregates are expressed, in this study, as Year-on-Year growth rates.

  15. If one computes correlation for a window of twenty successive quarters (rather than the whole period), the correlation between consumer replies on past economic growth and the last vintage of GDP data starts to break down in 2017. By contrast the correlation with the first vintage of GDP data drops slightly after 2017 but remains strong at nearly 0.64.

  16. The relative RMSE for an equation that forecasts real GDP growth using consumer expectations of growth stands at 0.8 for the four and three- quarter ahead forecast and at 0.9 for the two-period ahead forecast. These RMSEs were calculated using a 20-quarter moving quarter sample period.

  17. That said, at the three-quarter lag, the degree of correlation for Maltese data stands at 0.67, which is slightly higher than that for the EU average.


  • Aruoba, S. B., Diebold, F. X., & Scotti, C. (2009). Real-time measurement of business conditions. Journal of Business and Economic Statistics, 27(4), 417–427.

    Article  Google Scholar 

  • Bok, B., Caratelli, D., Giannone, D., Sbordone, A., & Tambalotti, A. (2017). Macroeconomic nowcasting and forecasting with big data. Federal Reserve Bank of New York Staff Reports, no. 830.

  • Bulligan, G., Marcellino, M., & Venditti, F. (2012). Forecasting economic activity with higher frequency targeted predictors. Temi di DiscussioneBank of Italy, No. 847.

  • De Bondt, G., & Forsells, M. (2017). The recent strength of survey-based indicators: What does it tell us about the depth and breadth of real GDP growth? ECB Economic Bulletin, 8, 44–49.

    Google Scholar 

  • European Commission. (2017). ESI and other BCS indicators vs PMI—Properties and empirical performance, special topic. In European business cycle indicators 2nd quarter 2017, European Economy Technical Paper 017, July, 18–26.

  • European Commission. (2019). The joint harmonised EU programme of business and consumer surveys: User guide. Directorate-General for Economic and Financial Affairs.

  • Gayer, C. (2005). Forecast evaluation of European Commission survey indicators. Journal of Business Cycle Measurement and Analysis, 2(2), 157–183.

    Article  Google Scholar 

  • Gelper, S., & Croux, C. (2010). On the construction of the European Economic sentiment indicator. Oxford Bulletin of Economics and Statistics, 72(1), 47–62.

    Article  Google Scholar 

  • Giannone, D., Reichlin, L., & Simonelli, S. (2009). Nowcasting euro area economic activity in real time: The role of confidence indicators. National Institute Economic Review, 210, 90–97.

    Article  Google Scholar 

  • Grech, A. (2018a). SMEs’ contribution to the Maltese economy and future prospects. Central Bank of Malta, Policy Note.

  • Grech, O. (2018b). An analysis of revisions to Maltese GDP data. Central Bank of Malta Research Bulletin, 1(1), 27–34.

    Google Scholar 

  • Grech, A., & Rapa, N. (2019). A reassessment of external demand’s contribution to Malta’s economic growth. Journal of Forecasting, 35(5), 400–418.

    Google Scholar 

  • International Monetary Fund. (2003). 2003 Article IV consultation-staff report. IMF.

  • Juriová, J. (2015). The role of foreign sentiment in small open economy. International Journal of Economic Sciences, 4(2), 57–68.

    Article  Google Scholar 

  • Kuzmanovic, M., & Sanfey, P. (2012). Can consumer confidence data predict real variables? Evidence from Croatia. European Bank for Reconstruction and Development Working Paper No. 151.

  • Mariano, R. S., & Murasawa, Y. (2003). A new coincident index of business cycles based on monthly and quarterly series. Journal of Applied Econometrics, 18(4), 427–443.

    Article  Google Scholar 

  • Mourougane, A., & Roma, M., (2002). Can confidence indicators be useful to predict short-term real GDP growth? ECB Working Paper No. 133.

  • Posta, V., & Pikhart, Z. (2012). The use of economic sentiment indicator for GDP short-term forecasting: Evidence from EU economies. Statistika, 49(1), 41–55.

    Google Scholar 

  • Soric, P., Lolic, I., & Cizmesija, M. (2016). European economic sentiment indicator: An empirical reappraisal. Quality and Quantity—International Journal of Methodology, 50(5), 2025–2054.

    Article  Google Scholar 

  • Stock, J. H., & Watson, M. W. (1989). “New indexes of coincident and leading economic indicators”, NBER chapters. In NBER macroeconomics annual 1989 (Vol. 4, pp. 351–409). National Bureau of Economic Research, Inc.

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Aaron G. Grech.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Grech, A.G., Ellul, R. Are the European Commission’s Business and Consumer Survey Results Coincident Indicators for Maltese Economic Activity?. J Bus Cycle Res 17, 91–108 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Business sentiment
  • Consumer expectations
  • Macroeconomic forecasts
  • Malta

JEL Classification

  • E20
  • C22
  • E37