Forecasting macroeconomic variables using disaggregate survey data
Introduction
Many central banks conduct surveys which yield regional and sectoral information on the general economic outlook. Following the example of the Federal Reserve’s Beige Book, which was implemented in 1970, and the Bank of England’s Agents survey, which begun in 1997, other central banks such as the Bank of Canada, Norges Bank, Sveriges Riksbank, and the Swiss National Bank have also initiated their own surveys. The information provided by these surveys is typically anecdotal and qualitative, unlike the well-known, quantitative Livingston survey, the Michigan survey, or the Survey of Professional Forecasters (see Thomas, 1999 for supplementary information about these surveys). While it is well-documented that the information obtained from quantitative surveys has strong forecasting power for macroeconomic variables (see for example Ang, Bekaert, & Wei, 2007; Fama & Gibbons, 1984; Mehra, 2002; and Thomas, 1999), there is less evidence of the forecasting power of information obtained from qualitative surveys.
This paper investigates the abilities of the Norges Bank’s regional survey and the Swedish Business Tendency Survey to forecast the gross domestic product (GDP) growth, consumer price inflation, and the unemployment rate for Norway and Sweden. Each survey consists of both backward- and forward-looking qualitative information. Studies such as those of Abberger (2007), Claveria, Pons, and Ramos (2007) and Lui et al., 2011a, Lui et al., 2011b focus on examining specific survey questions in order to predict individual macroeconomic variables. Our approach is different, applying a dynamic factor model to the full database in order to construct regional and sectoral factors. These factors should contain the most relevant information for the regions and sectors from which they are extracted.
Our approach is similar to that of Hansson, Jansson, and Löf (2005), who use a dynamic factor model (based on net balance indices, representing differences between the shares of firms that have specified increases and decreases for a particular economic activity) from the Swedish Business Tendency Survey to forecast the Swedish GDP. Hansson et al. (2005) find that their factor model outperforms popular alternatives such as econometric VAR models in most cases. We extend their analysis in at least four directions. First, we consider the Norges Bank’s regional survey, which is more comprehensive in terms of sectors and regions of the economy. Our choice follows the claims made by Beck, Hubrich, and Marcellino (2009) that highly disaggregated regional and sectoral information is important in explaining aggregate Euro area and US inflation rates. Second, we work at a higher level of disaggregation and construct regional and sectoral factor models from the surveys. Out of ten sectors and seven regions for the Norwegian economy, and three sectors for the Swedish economy, our results identify which ones perform particularly well at forecasting different variables at various horizons. Third, we mitigate the uncertainties in the construction of the factors, the numbers of factors, and the relationships with the variable of interest by investigating two different classes of factor models where the number of factors is fixed a priori (denoted as model A) or estimated via a selection criterion (model B). Finally, we use forecast combinations to address the model uncertainty created by the use of several factors constructed by different datasets (regions or sectors). Each factor model is used to extract information and produce forecasts from a given dataset (regions or sectors) for the particular variable of interest.
We find that factor models based on several regions and sectors systematically beat the nowcasts and one-quarter-ahead forecasts of Norwegian inflation and unemployment rate given by the benchmark model. However, the factor models are most successful in nowcasting and forecasting GDP growth. Forecast combinations of the regional and sectoral models based on past performances are more accurate than the best regional or sectoral model in several cases, and provide more accurate forecasts than the benchmark model in almost all cases. Furthermore, we empirically find that aggregating the survey data either by pooling all of the Norwegian regional and sectoral survey information in a single factor model or by aggregating individual question-based forecasts via model combinations, to account for the heterogeneity in individual survey questions, results in less accurate forecasts than our regional and sector factor models. This finding is qualitatively similar when we use the Swedish Business Tendency Survey.
The paper proceeds as follows: Section 2 outlines the methodological aspects of our dynamic factor model, and Section 3 explains the forecasting models. Section 4 describes Norges Bank’s regional survey data, presents the factors and discusses the forecasting results. Section 5 reports results using the Sweden Business Tendency survey. Finally, Section 6 concludes.
Section snippets
A dynamic factor model
The increasing availability of information on economic activities and their disaggregate components makes factor models a very attractive approach for handling macroeconomic data. Applying a factor model to a large dataset of possibly correlated variables reduces the dimension of the dataset while retaining as much of the variation in the data as possible. This reduced form can be useful for forecasting, since models which are more parsimonious reduce the estimation errors and may yield more
Forecasting
This paper’s ultimate goal is to forecast inflation, GDP growth, and the unemployment rate for Norway and Sweden using the factors derived from the surveys. We produce nowcasts of the current quarter, as well as one-, two-, three-, and four-quarter-ahead forecasts for a total of five horizons. Survey data become available at the end of the second month of the current quarter, and we use this information in nowcasting and forecasting.
We compare two different factor models with an autoregressive
The Norges Bank’s regional survey
In 2002, Norges Bank established regional networks of enterprises, organizations, and local authorities throughout Norway. By conducting interviews with its contacts, Norges Bank obtains information concerning their current economic situation and their plans for the coming months. The survey reflects the production side of the economy both geographically and sectorally by dividing the country into regions: Inland, Mid-Norway, North, North–West, South, South–West and East; and sectors:
The Swedish Business Tendency Survey
The Swedish Business Tendency Survey (SBTS) provides fast and accurate information on developments in the Swedish economy. Each month, the Sweden National Institute of Economic Research asks a large number of businesses for their assessments of the current economic situation. The questions include asking the firms for their views on output, new orders, employment, and prices. The SBTS’s aim is to produce timely information on the economy’s current situation and to provide short-term forecasts
Concluding remarks
This paper proposes a factor model approach to forecasting macroeconomic variables using information from large qualitative surveys, where the questions used to collect information can be very different and may refer to disaggregate information for the variables of interest. We apply our methodology to the Norges Bank’s regional survey and to the Swedish Business Tendency Survey, and find several interesting results. First, regarding the factor estimation based on a dynamic factor model, the
Acknowledgments
We thank the referees, the associate editor and the editor Michael Clements for their very useful comments on an earlier version of our paper. We also thank Knut Are Aastveit, Raffaella Giacomini, James Mitchell, Elizabeth Murry, Christian Kascha, Shaun Vahey, and seminar participants at the 31st Annual International Symposium on Forecasting 2011 and Norges Bank for helpful comments. The views expressed in this paper are our own and do not necessarily reflect those of Norges Bank.
Kjetil Martisen is an economist in the economic department of Norges Bank Monetary Policy wing. Kjetil has finished a master at University College London in 2010.
References (29)
Qualitative business surveys and the assessment of employment—a case study for Germany
International Journal of Forecasting
(2007)- et al.
Do macro variables, asset markets, or surveys forecast inflation better?
Journal of Monetary Economics
(2007) - et al.
Business and consumer expectations and macroeconomic forecasts
International Journal of Forecasting
(2007) Combining forecasts: a review and annotated bibliography
International Journal of Forecasting
(1989)- et al.
A two-step estimator for large approximate dynamic factor models based on Kalman filtering
Journal of Econometrics
(2011) - et al.
A comparison of inflation forecasts
Journal of Monetary Economics
(1984) - et al.
Business survey data: do they help in forecasting GDP growth?
International Journal of Forecasting
(2005) - et al.
Testing the equality of prediction mean squared errors
International Journal of Forecasting
(1997) - et al.
The utility of expectational data: firm-level evidence using matched qualitative-quantitative UK surveys
International Journal of Forecasting
(2011) - et al.
Survey expectations
Evaluating factor forecasts for the UK: the role of asset prices
International Journal of Forecasting
Nowcasting Norwegian GDP: the role of asset prices in a small open economy
Empirical Economics
Combination of forecasts
Operational Research Quarterly
Cited by (40)
Economic forecasting with evolved confidence indicators
2020, Economic ModellingCitation Excerpt :Juhro and Iyke (2020) also found that business and consumer sentiments improved the accuracy of the Indonesian consumption forecasts. There is ample evidence that survey expectations are useful for predicting economic variables (Altug and Çakmakli, 2016; Claveria et al., 2007; Girardi et al., 2015; Klein and Özmucur, 2010; Martinsen et al., 2014). In this sense, the obtained results are consistent with recent research regarding the predictive content of survey expectations.
Unemployment expectations: A socio-demographic analysis of the effect of news
2019, Labour EconomicsCitation Excerpt :Finally, the discussion and conclusions are drawn in Section 5. Economic agents’ capability to accurately assess and predict micro and macroeconomic variables has long been one of the focal points of applied economic research (Abberger, 2007; Billstam et al., 2017; Bruestle and Crain, 2015; Clements, 2019; Graff, 2010; Hanson et al., 2005; Lehmann and Wohlrabe, 2017; Martinsen et al., 2014; Mitchell et al., 2005a,b; Vermeulen, 2014; Wilms et al., 2016). When focused on consumer sentiment, this strand of research has provided an abundance of empirical evidence in favour of the Katonian hypothesis, which refers to the leading properties of consumer confidence, or economic sentiment in general, with regards to various forms of economic activity.
Using survey data to forecast real activity with evolutionary algorithms. A cross-country analysis
2017, Journal of Applied EconomicsBusiness tendency surveys and macroeconomic fluctuations
2017, International Journal of ForecastingThe Forecasting Power of the ifo Business Survey
2023, Journal of Business Cycle ResearchQuantum Computing and Deep Learning Methods for GDP Growth Forecasting
2022, Computational Economics
Kjetil Martisen is an economist in the economic department of Norges Bank Monetary Policy wing. Kjetil has finished a master at University College London in 2010.
Francesco Ravazzolo is a senior researcher in the monetary policy wing of Norges Bank. Ravazzolo’s fields of interests are (Bayesian) econometrics and forecasting. He has published several papers in international journals such as Journal of Econometrics, Journal of Business and Economic Statistics, Journal of Money, Credit and Banking, and Journal of Forecasting.
Fredrik Wulfsberg has his Ph.D. from the University of Oslo in 1997. Wulfsberg’s fields of interests are macroeconomics and labour market economics, and his papers have been published in Journal of Monetary Economics, The B.E. Journal of Macroeconomics (Advances) and Scandinavian Journal of Economics.