No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
A PMI-Based Real GDP Tracker for the Euro Area
Abstract
Real-time evidence for the euro area shows that a tracker for real GDP growth using only the Purchasing Managers’ Index (PMI) composite output is of similar accuracy for the final GDP release as the first GDP release. No signs of instability—except during the 2008/09 crisis—in this tracking performance are found. This is surprising given the small size of the underlying PMI panel. From a closer look at what is driving this outstanding track record, seven conclusions emerge: (i) the level of and change in the PMI composite output explain one-third of the GDP revisions; (ii) later available information is more accurate; (iii) services are key; (iv) firm size breakdown is valuable; (v) export status breakdown creates only noise; (vi) aggregated euro area PMI track record is not consistently related to a particular country; (vii) take firm defaults into account during very bad times. These findings imply that PMI surveys are not only valuable for analysts and policymakers as a timely and reliable GDP tracker, but also for statisticians to potentially improve the accuracy of the first preliminary flash estimate of euro area real GDP.
... As one of the important engines of economic growth [3][4][5], the manufacturing sector is indispensable in increasing employment, advancing technology, and boosting exports [1,6]. Researchers typically use the manufacturing purchasing manager's index (PMI) to measure and forecast the trend of manufacturing [7][8][9][10]. The manufacturing PMI originated in the United States. ...
... Masona [30] used the least squares method to analyze the impact of GEPU on enterprise R&D investment and found that the negative impact of GEPU on manufacturing R&D investment is higher than that of the service sector. A few academics have talked about using PMI to track GDP [7], or thought China's manufacturing PMI to be ahead of the US and the EU's [8], or have shown interested in the correlation between manufacturing PMI and macroeconomic indicators such as CPI and PPI [9]. Although the above studies are all related to GEPU or PMI, they have not empirically analyzed the direct impact of GEPU on manufacturing PMI, nor have they demonstrated manufacturing's risk resilience. ...
... The Russia-Ukraine conflict has increased the uncertainty of German economic policy and negatively impacted its financial and economic systems [37][38][39], hence, we find that EU manufacturing suffered most from the Russia-Ukraine conflict. Bondt's [7] research suggests that PMI provides a reliable GDP tracker for policymakers and analysts. Yue [9] believes that when the manufacturing PMI is in the expansion range, the PPI shows an upward trend. ...
Events such as COVID-19 and the Russia–Ukraine conflict have significantly increased the uncertainty and volatility of global economic policies. In the context of economic globalization, the key question we investigate is whether global economic policy uncertainty will have different impacts on the manufacturing of the three major economies in China, the United States, and Europe Union. This study employs the time-varying parameter vector autoregressive (TVP-VAR) model to examine how global economic policy uncertainty (GEPU) affected manufacturing from March 2008 to March 2023. The empirical results show that the effects of GEPU are time varying; its short-term effects on Chinese manufacturing are slightly greater than its medium- and long-term effects, whereas its medium- and long-term effects on manufacturing in the United States (US) and European Union (EU) are significantly greater than its short-term effects. The impact of European debt crisis, the China–US trade war and Russia–Ukraine conflict on EU manufacturing is higher than that of China and the US, and the impact of the COVID-19 pandemic on China’s manufacturing is much smaller than that of the US and the EU; thus, Chinese manufacturing has a greater capacity for risk mitigation than US and EU manufacturing. This study not only provides a new perspective on the study of global economic policy uncertainty; it also provides new empirical evidence on how global economic policy uncertainty affects the manufacturing sector in China, the US and Europe and provides policymakers with guidance for decision making.
... Special thanks goes to IHS Markit for providing us the PMI data for Germany. the ifo Business Climate, and the PMI Composite Output Index) receive considerable media attention each month and are found to be important for tracking economic activity in both the Euro Area and Germany (see, for example, Basselier et al. 2018;de Bondt 2019;Fritsche and Stephan 2002;Lehmann 2020). Recently, the indicators of both survey providers are listed on Bloomberg's "12 Global Economic Indicators to Watch". 1 However, two further and very important survey providers that publish monthly headline indices are mostly neglected in the public debate: the Directorate-General for Economic and Financial Affairs of the European Commission (DF ECFIN) and the Centre for European Economic Research (ZEW). ...
... 2.3 we decided to apply first differences as our baseline specification. However, studies such as Basselier et al. (2018) andde Bondt (2019) show that also level specifications provide sensible results, hence it might be an empirical matter what works best. We therefore compare the forecasting power of the indicators transformed into first differences with their accuracy based on levels. ...
This analysis investigates the predictive power of the headline indices of the four most important German survey providers. We conduct an out-of-sample, real-time forecast experiment for growth of total and private sector gross domestic product and growth of gross value added in both the manufacturing and the service sector. All providers publish valuable leading indicators for both GDP measures, with some advantages for the ifo indicators and the Economic Sentiment Indicator, respectively. For the manufacturing sector, indicators provided by the ifo Institute are clearly superior. For the service sector, all indicators prove to have a similar nowcasting performance, whereas the Economic Sentiment Services of the Centre for European Research is preferable for one quarter-ahead predictions.
... Also the existing academic literature certifies business surveys to be powerful tools for economic forecasting or tracking economic activity. However, most of the studies focus on a rather small number of economic aggregates such as gross domestic product (GDP), (un-)employment or inflation (see, for example, Hansson et al., 2005;Claveria et al., 2007;Angelini et al., 2011;Martinsen et al., 2014;Österholm, 2014;Lehmann and Weyh, 2016;Basselier et al., 2018;de Bondt, G. J., 2019). Furthermore, most of the media attention is gained by the headline indices of a business survey. ...
The ifo Institute is Germany’s largest business survey provider, with the ifo Business Climate Germany as one of the most important leading indicators for gross domestic product. However, the ifo Business Survey is not solely limited to the Business Climate and also delivers a multitude of further indicators to forecast several important economic variables. This paper gives a literature overview over existing studies that deal with the forecasting power of various ifo indicators both for gross domestic product and further economic variables such as exports. Overall, the various indicators from the ifo Business Survey can be seen as leading indicators for a multitude of variables representing the German economy, making them a powerful tool both for an in-depth business cycle diagnosis and for applied forecasting work.
... The study [17] provides an empirical analysis of the industry confidence indicators' impact on real GDP growth during EU and US recessions and growth and concludes that sentiment indicators are statistically significant for economic growth in both good times and bad. The paper summarizes the statement about the absolute correlation of only the Purchasing Managers' Index with GDP data. ...
The paper develops and presents an appropriate model toolkit that allows assessing the relationship between the calculated indices of economic sentiment and confidence for the main types of economic activity. The aim of the study is to experimentally substantiate the relevance of data on the opinions of technological economic agents and assess the value of this information for the statistical description and analysis of macroeconomic trends, including economic cycles and unforeseen and prolonged crises. The main hypothesis about the cyclical sensitivity of composite indices, in particular the index of economic sentiment in relation to the dynamics of the physical volume of GDP, is tested. The authors calculate a composite indicator of aggregate economic sentiment and, based on a consistent analysis of the relationship between the index of physical volume of GDP and the indicator of economic sentiment, identify aggregate empirical patterns and features of the cyclical development of technological enterprises. Accordingly, the turning points of the business cycle are discussed and the leading nature of the proposed index of economic sentiment is affirmed. The importance of the composite indicators in the economic analysis of entrepreneurial behavior in the implementation of technological innovations is shown.
The nature of the calculated Economic Sentiment Index was established, and its predictive capabilities for monthly and annual real GDP growth rates using autoregression and error correction models were investigated. The stages of calculating and setting indices with the application of the DEMETRA+ statistical package were implemented.
The paper investigates the application of endogenous Markov-switching MIDAS models for nowcasting and forecasting during the turbulent economic landscape shaped by the COVID-19 crisis. I extend the standard MS MIDAS model to incorporate time-varying transition probabilities whose dynamics is influenced by various explanatory variables, enabling a more nuanced understanding of GDP dynamics across G-20 economies. The analysis reveals that these models enhance forecasting accuracy compared to standard MIDAS models, particularly in crisis periods, by better capturing the timing and magnitude of economic shifts. I point out the importance of selecting appropriate indicators at varying forecasting horizons, as their effectiveness fluctuates among forecasting horizons. My findings underscore the potential of Markov-switching models, particularly those with endogenous switching, as promising tools for macroeconomic forecasting, advocating for their further development across multiple modeling frameworks.
Plain English Summary
The German Mittelstand is economically more successful compared to SMEs and large enterprises because of its combination of owner-management and firm size. The German Mittelstand is always seen as the “backbone” of the German economy and made responsible for Germany’s ability to cope with large external shocks such as the 2008/2009 global recession. To date, this suggestion is not backed by empirical evidence as common data sources do not incorporate information on the ownership structure of German firms. We enrich the micro panel data from the largest monthly business survey in Germany by two questions on the ownership structure of the interviewed firm and can confirm this suggestion. It turns out that it is not firm size or owner-management alone that leads to a higher crisis resistance but rather the combination of both criteria. Our results have practical and political implications as it is possible with our approach to separate the German Mittelstand from the pool of small- and medium-sized enterprises. The information on the ownership structure in our data can initiate future research activities and shed more light on the relevance of owner-management for the economic success of firms.
Dynamic factor models based on Kalman Filter techniques are frequently used to nowcast GDP. This study deals with the selection of indicators for this practice. We propose a two-tiered mechanism which is shown in a case study to produce more accurate nowcasts than a benchmark stochastic process and a standard model including extreme bounds fragile indicators. Our pseudo-ex-ante forecast nearly measures up to the genuine ex-ante forecast of the European Commission. © 2018, Springer International Publishing AG, part of Springer Nature.
This paper analyses the contribution of survey data, in particular various sentiment indicators, to nowcasts of quarterly euro area GDP. It uses a genuine real-time dataset that is constructed from original press releases in order to transform the actual dataflow into an interpretable flow of news. The latter is defined as the difference between the released values and the prediction of a mixed-frequency dynamic factor model. Our purpose is twofold. First, we aim to quantify the specific value added for nowcasting GDP from a set of heterogeneous data releases including not only sentiment indicators constructed by Eurostat, Markit, the National Bank of Belgium, IFO, ZEW, GfK or Sentix, but also hard data regarding industrial production or retail sales in the aggregate euro area and individually in some of the largest euro area countries. Second, our quantitative analysis is used to draw up an overall ranking of the indicators, on the basis of their average contribution to updates of the nowcast. Among the survey indicators, we find the strongest impact for the Markit Manufacturing PMI and the Business Climate Indicator in the euro area, and the IFO Business Climate and IFO Expectations in Germany. The widely monitored consumer confidence indicators, on the other hand, typically do not lead to significant revisions of the nowcast. In addition, even if euro area industrial production is a relevant predictor, hard data generally contribute less to the nowcasts: they may be more closely correlated with GDP but their relatively late availability implies that they can to a large extent be anticipated by nowcasting on the basis of survey data and, hence, their ‘news’ component is smaller. Finally, we also show that, in line with the previous literature, the NBB’s own business confidence indicator appears to be useful for predicting euro area GDP. The prevalence of survey data remains also under a counterfactual scenario in which hard data are released without any delay. This finding confirms that, in addition to being available in a more timely manner, survey data also contain relevant information that does not seem to be captured by hard data.
The paper compares the most closely watched sentiment indicators with respect to their ability to nowcast quarterly GDP dynamics in the Euro Area and its biggest economies. We analyse cross-correlations and out-of-sample forecast errors generated from equations estimated by rolling regressions in fixed-length window. The results show that models employing PMI Composite perform best in the cases of the Euro Area, Germany, France and Italy, whilst Spanish GDP dynamics is best nowcasted using ESI-based models. PMI-based models generate the most accurate nowcasts at the beginning of the quarter, as well as during periods of high volatility of GDP growth rates
This study shows for the first time that euro area real GDP nowcasts using the PMI survey can be more accurate than Eurostat's flash GDP estimate for extended periods of time in predicting the final GDP release. The nowcast performance of the PMI composite output index, measured as well as linked to GDP in different ways, is examined in detail for the years 2003-2010. PMI-based nowcasts of quarterly real GDP growth have outperformed the flash real GDP estimate in particular over the years 2003-2007, but even in a few cases on average over 2003-2010. Simple PMI-based rules-of-thumb for nowcasting quarter-on-quarter real GDP growth can also be derived at the country level. In contrast to the euro area, the PMI rules do generally not outperform the flash GDP estimate at the country level, with the main exception of the pre-crisis years in Germany due to a comparatively poor performance of the German flash GDP estimate. These findings imply that a close look at the PMI surveys is valuable for a reliable and early assessment of the current state of the euro area economy. are highly appreciated.
This paper uses an extension of the Euro-Sting single-index dynamic factor model to construct short-term forecasts of quarterly GDP growth for the euro area by accounting for financial variables as leading indicators. From a simulated real-time exercise, the model is used to investigate the forecasting accuracy across the different phases of the business cycle. Our extension is also used to evaluate the relative forecasting ability of the two most reliable business cycle surveys for the euro area: the PMI and the ESI. We show that the latter produces more accurate GDP forecasts than the former. Finally, the proposed model is also characterized by its great ability to capture the European business cycle, as well as the probabilities of expansion and/or contraction periods. Copyright © 2014 John Wiley & Sons, Ltd.
We estimate and forecast growth in euro area monthly GDP and its components from a dynamic factor model due to Doz et al. (2005), which handles unbalanced data sets in an efficient way. We extend the model to integrate interpolation and forecasting together with cross-equation accounting identities. A pseudo real-time forecasting exercise indicates that the model outperforms various benchmarks, such as quarterly time series models and bridge equations in forecasting growth in quarterly GDP and its components. JEL Classification: E37, C53.
We propose and evaluate explicit tests of the null hypothesis of no difference in the accuracy of two competing forecasts. In contrast to previously developed tests, a wide variety of accuracy measures can be used (in particular, the loss function need not be quadratic and need not even be symmetric), and forecast errors can be non-Gaussian, nonzero mean, serially correlated, and contemporaneously correlated. Asymptotic and exact finite-sample tests are proposed, evaluated, and illustrated.
This study compares industrial production and gross value added in volume terms in the euro area and euro area countries, because real GDP growth signals from industrial production growth might be misleading and earlier released industrial production growth is not one-to-one translated into industrial value added growth. This is an important issue for analysts and policy makers, because industrial production is a standard element in tools for nowcasting real GDP in real time. It also raises the question about the factors explaining these differences. Differences in terms of (changes in) quarterly growth between production and gross value added include sign reversals and can last for consecutive quarters. Persistent level differences might also exist. The explanatory factors for these differences are the treatment of prices, seasonality and coverage. Data limitations prevent a detailed analysis of the price factor, but the other two factors are more closely evaluated. It turns out that the relative importance of these factors varies over time and thus is difficult to assess ex ante for a specific quarter. A remedy is that statisticians further harmonize national accounts and short-term statistics as well as national practices for seasonal adjustment.
Model specification and selection are recurring themes in econometric analysis. Both topics become considerably more complicated in the case of large-dimensional data sets where the set of specification possibilities can become quite large. In the context of linear regression models, penalised regression has become the de facto benchmark technique used to trade off parsimony and fit when the number of possible covariates is large, often much larger than the number of available observations. However, issues such as the choice of a penalty function and tuning parameters associated with the use of penalised regressions remain contentious. In this paper, we provide an alternative approach that considers the statistical significance of the individual covariates one at a time, whilst taking full account of the multiple testing nature of the inferential problem involved. We refer to the proposed method as One Covariate at a Time Multiple Testing (OCMT) procedure. The OCMT has a number of advantages over the penalised regression methods: It is based on statistical inference and is therefore easier to interpret and relate to the classical statistical analysis, it allows working under more general assumptions, it is computationally simple and considerably faster, and it performs better in small samples for almost all of the five different sets of experiments considered in this paper. Despite its simplicity, the theory behind the proposed approach is quite complicated. We provide extensive theoretical and Monte Carlo results in support of adding the proposed OCMT model selection procedure to the toolbox of applied researchers.
Practices used to address economic forecasting problems have undergone substantial changes over recent years. We review how such changes have influenced the ways in which a range of forecasting questions are being addressed. We also discuss the promises and challenges arising from access to big data. Finally, we review empirical evidence and experience accumulated from the use of forecasting methods to a range of economic and financial variables.
A number of recent studies in the economics literature have focused on the usefulness of factor models in the context of prediction using " big data " (see Bai and Ng (the references cited therein). We add to this literature by analyzing whether " big data " are useful for modelling low frequency macroeconomic variables such as unemployment, in ‡ation and GDP. In particular, we analyze the predictive bene…ts associated with the use of principal component analysis (PCA), independent component analysis (ICA), and sparse principal component analysis (SPCA). We also evaluate machine learning, variable selection and shrinkage methods, including bagging, boosting, ridge regression, least angle regression, the elastic net, and the non-negative garotte. Our approach is to carry out a forecasting " horse-race " using prediction models constructed using a variety of model speci…cation approaches, factor estimation methods, and data windowing methods, in the context of the prediction of 11 macroeconomic variables relevant for monetary policy assessment. In many instances, we …nd that various of our benchmark models, including autoregressive (AR) models, AR models with exogenous variables, and (Bayesian) model averaging, do not dominate speci…cations based on factor-type dimension reduction combined with various machine learning, variable selection, and shrinkage methods (called " combination " models). We …nd that forecast combination methods are mean square forecast error (MSFE) " best " for only 3 of 11 variables when the forecast horizon, h = 1, and for 4 variables when h = 3 or 12. Additionally, non-PCA type factor estimation methods yield MSFE-best predictions for 9 of 11 variables when h = 1, although PCA dominates at longer horizons. Interestingly, we also …nd evidence of the usefulness of combination models for approximately 1/2 of our variables, when h > 1. Most importantly, we present strong new evidence of the usefulness of factor based dimension reduction, when utilizing " big data " for macroeconometric forecasting.
This paper evaluates the impact of new releases of financial, real activity and survey data on nowcasting euro area gross domestic product (GDP). We show that all three data categories positively impact on the accuracy of GDP nowcasts, whereby the effect is largest in the case of real activity data. When treating variables as if they were all published at the same time and without any time lag, financial series lose all their significance, while survey data remain an important ingredient for the nowcasting exercise. The subsequent analysis shows that the sectoral coverage of survey data, which is broader than that of timely available real activity data, as well as their information content stemming from questions focusing on agents' expectations, are the main sources of the ‘genuine’ predictive power of survey data. When the forecast period is restricted to the 2008–09 financial crisis, the main change is an enhanced forecasting role for financial data. Copyright © 2015 John Wiley & Sons, Ltd.
Numerical approximations to the asymptotic distributions of recently proposed tests for structural change are presented. This enables easy yet accurate calculation of asymptotic p values. A GAUSS program is available to perform the computations.
Objectives
The literature typically undervalues the economy-wide importance of confidence, despite a renewed interest since the recent financial crisis in considering also psychological factors such as confidence. This study empirically assesses whether confidence matters for current real GDP growth in the euro area and the United States in addition to a widely applied and reliable predictor, the Purchasing Managers’ Index.Methods
We add confidence indicators to a regression of real GDP growth on the composite PMI output index and check for a different impact of confidence during recessions as opposed to expansions by applying smooth transition regressions.ResultsConfidence matters for economic growth, both in good and bad times. This result is robust across sample periods, models, and proxies for confidence.Conclusions
Confidence is essential for assessing the current stage of the business cycle. Analysts should therefore closely monitor sentiment swings, whereas private and public decisionmakers can boost growth by improving confidence in the economy.
We study the role of the well-known monthly di¤usion indices produced by the Institute for Supply Management (ISM) in nowcasting current quarter US GDP growth. In contrast to the existing literature on ISM surveys, we investigate their marginal impact on these nowcasts when large unbalanced (jagged edge) macroeconomic data sets are used in real time to generate them. We …nd evidence that these ISM indices are helpful in improving the nowcasts when new ISM information becomes available in the beginning of the month, ahead of other monthly indicators. Furthermore, and while the existing literature has focused almost exclusively on manufaturing information, here we establish the increasingly signi…cant role of the recently created non-manufacturing ISM di¤usion indices in such nowcasting contexts.
This paper discusses the contribution of Lahiri and Monokroussos, published in the current issue of this journal, where they investigate the nowcasting power of ISM Business Surveys for real US GDP. The second part of this note includes some empirical considerations on nowcasting quarterly real GDP by using the monthly PMI index for Switzerland. The results indicate that the Swiss PMI is not leading GDP growth; rather, it is coincident, and its nowcasting power is quite good. The signs of the fitted values mostly correspond to the sign of the actual GDP growth, and the important turning points are identified accurately by the model. This also holds true during the recent crisis.
Reliable and timely information about current economic conditions is crucial for policy makers and expectations formation. This paper demonstrates the efficacy of the Survey of Professional Forecasters (SPF) and the Purchasing Manager Indices (PMI) in anticipating US real economic activity. We conduct a fully-fledged real-time out-of-sample forecasting exercise linking these surveys to US GDP and industrial production growth over a long sample period. We find that both indicators convey valuable information for assessing current economic conditions. The SPF clearly outperforms the PMI in forecasting GDP growth, while it performs quite poorly in anticipating industrial production growth. Combining the information included in both surveys further improves the accuracy of both, the PMI and the SPF-based forecast.
We study the role of the well-known monthly di¤usion indices produced by the Institute for Supply Management (ISM) in nowcasting current quarter US GDP growth. In contrast to the existing literature on ISM surveys, we investigate their marginal impact on these nowcasts when large unbalanced (jagged edge) macroeconomic data sets are used in real time to generate them. We …nd evidence that these ISM indices are helpful in improving the nowcasts when new ISM information becomes available in the beginning of the month, ahead of other monthly indicators. Furthermore, and while the existing literature has focused almost exclusively on manufaturing information, here we establish the increasingly signi…cant role of the recently created non-manufacturing ISM di¤usion indices in such nowcasting contexts.
Given two sources of forecasts of the same quantity, it is possible to compare prediction records. In particular, it can be useful to test the hypothesis of equal accuracy in forecast performance. We analyse the behaviour of two possible tests, and of modifications of these tests designed to circumvent shortcomings in the original formulations. As a result of this analysis, a recommendation for one particular testing approach is made for practical applications.
This paper considers tests for parameter instability and structural change with unknown change point. The results apply to a wide class of parametric models that are suitable for estimation by generalized method of moments procedures. The asymptotic distributions of the test statistics considered here are nonstandard because the change point parameter only appears under the alternative hypothesis and not under the null. The tests considered here are shown to have nontrivial asymptotic local power against all alternatives for which the parameters are nonconstant. The tests are found to perform quite well in a Monte Carlo experiment reported elsewhere. Copyright 1993 by The Econometric Society.
Macroeconomic nowcasting and forecasting with big data. Federal Reserve Bank of New York Staff Reports No
- B Bok
- D Caratelli
- D Giannone
- A M Sbordone
- A Tambalotti
Bok, B., Caratelli, D., Giannone, D., Sbordone, A. M. & Tambalotti, A. (2017). Macroeconomic nowcasting and forecasting with big data. Federal Reserve Bank of New York Staff Reports No. 830.
Nowcasting: Trust the Purchasing Managers' Index or wait for the flash GDP estimate
- G J Bondt
- De
Bondt, G. J. de (2012). Nowcasting: Trust the Purchasing Managers' Index or wait for the flash GDP
estimate?. In G. T. Papanikos (Ed.), Economic essays, Athens Institute for Education and Research,
Chapter 8, 83-97; Conference Paper, Presented at EcoMod2012, international conference on economic modeling, July 4-6, Seville.
How the US-Eurozone growth differential gets revised away
- D Vernazza
Vernazza, D. (2018). How the US-Eurozone growth differential gets revised away. UniCredit Macro
Research Economics Thinking No. 67, 27 April.
ESI and other BCS indicators vs PMI-Properties and empirical performance, special topic
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.