Asymmetric Investment Responses to Firm-specific Forecast Errors
Julian Berner, Manuel Buchholz, Lena Tonzer
Abstract
This paper analyses how firm-specific forecast errors derived from survey data of German manufacturing firms over 2007–2011 affect firms’ investment propensity. Understanding how forecast errors affect firm investment behaviour is key to mitigate economic downturns during and after crisis periods in which forecast errors tend to increase. Our findings reveal a negative impact of absolute forecast errors on investment. Strikingly, asymmetries arise depending on the size and direction of the forecast error. The investment propensity declines if the realised situation is worse than expected. However, firms do not adjust investment if the realised situation is better than expected suggesting that the uncertainty component of the forecast error counteracts positive effects of unexpectedly favorable business conditions. Given that the fraction of firms making positive forecast errors is higher after the peak of the recent financial crisis, this mechanism can be one explanation behind staggered economic growth and slow recovery following crises.
Artikel Lesen
How Forecast Accuracy Depends on Conditioning Assumptions
Carola Engelke, Katja Heinisch, Christoph Schult
IWH Discussion Papers,
Nr. 18,
2019
Abstract
This paper examines the extent to which errors in economic forecasts are driven by initial assumptions that prove to be incorrect ex post. Therefore, we construct a new data set comprising an unbalanced panel of annual forecasts from different institutions forecasting German GDP and the underlying assumptions. We explicitly control for different forecast horizons to proxy the information available at the release date. Over 75% of squared errors of the GDP forecast comove with the squared errors in their underlying assumptions. The root mean squared forecast error for GDP in our regression sample of 1.52% could be reduced to 1.13% by setting all assumption errors to zero. This implies that the accuracy of the assumptions is of great importance and that forecasters should reveal the framework of their assumptions in order to obtain useful policy recommendations based on economic forecasts.
Artikel Lesen
Optimizing Policymakers' Loss Functions in Crisis Prediction: Before, Within or After?
Peter Sarlin, Gregor von Schweinitz
Abstract
Early-warning models most commonly optimize signaling thresholds on crisis probabilities. The expost threshold optimization is based upon a loss function accounting for preferences between forecast errors, but comes with two crucial drawbacks: unstable thresholds in recursive estimations and an in-sample overfit at the expense of out-of-sample performance. We propose two alternatives for threshold setting: (i) including preferences in the estimation itself and (ii) setting thresholds ex-ante according to preferences only. Given probabilistic model output, it is intuitive that a decision rule is independent of the data or model specification, as thresholds on probabilities represent a willingness to issue a false alarm vis-à-vis missing a crisis. We provide simulated and real-world evidence that this simplification results in stable thresholds and improves out-of-sample performance. Our solution is not restricted to binary-choice models, but directly transferable to the signaling approach and all probabilistic early-warning models.
Artikel Lesen
Asymmetric Investment Responses to Firm-specific Uncertainty
Julian Berner, Manuel Buchholz, Lena Tonzer
Abstract
This paper analyzes how firm-specific uncertainty affects firms’ propensity to invest. We measure firm-specific uncertainty as firms’ absolute forecast errors derived from survey data of German manufacturing firms over 2007–2011. In line with the literature, our empirical findings reveal a negative impact of firm-specific uncertainty on investment. However, further results show that the investment response is asymmetric, depending on the size and direction of the forecast error. The investment propensity declines significantly if the realized situation is worse than expected. However, firms do not adjust their investment if the realized situation is better than expected, which suggests that the uncertainty effect counteracts the positive effect due to unexpectedly favorable business conditions. This can be one explanation behind the phenomenon of slow recovery in the aftermath of financial crises. Additional results show that the forecast error is highly concurrent with an ex-ante measure of firm-specific uncertainty we obtain from the survey data. Furthermore, the effect of firm-specific uncertainty is enforced for firms that face a tighter financing situation.
Artikel Lesen
Optimizing Policymakers' Loss Functions in Crisis Prediction: Before, Within or After?
Peter Sarlin, Gregor von Schweinitz
Abstract
Early-warning models most commonly optimize signaling thresholds on crisis probabilities. The ex-post threshold optimization is based upon a loss function accounting for preferences between forecast errors, but comes with two crucial drawbacks: unstable thresholds in recursive estimations and an in-sample overfit at the expense of out-of-sample performance. We propose two alternatives for threshold setting: (i) including preferences in the estimation itself and (ii) setting thresholds ex-ante according to preferences only. We provide simulated and real-world evidence that this simplification results in stable thresholds and improves out-of-sample performance. Our solution is not restricted to binary-choice models, but directly transferable to the signaling approach and all probabilistic early-warning models.
Artikel Lesen
The Financial Crisis from a Forecaster's Perspective
Katja Drechsel, Rolf Scheufele
Kredit und Kapital,
Nr. 1,
2012
Abstract
This paper analyses the recession in 2008/2009 in Germany. This recession is very different from previous recessions in particular regarding their causes and magnitude. We show to what extent forecasters and forecasts based on leading indicators fail to detect the timing and the magnitude of the recession. This study shows that large forecast errors for both expert forecasts and forecasts based on leading indicators resulted during this recession which implies that the recession was very difficult to forecast. However, some leading indicators (survey data, risk spreads, stock prices) have indicated an economic downturn and hence, beat univariate time series models. Although the combination of individual forecasts provides an improvement compared to the benchmark model, the combined forecasts are worse than several individual models. A comparison of expert forecasts withthe best forecasts based on leading indicators shows only minor deviations. Overall, the range for an improvement of expert forecasts in the crisis compared to indicator forecasts is small.
Artikel Lesen
Protect and Survive? Did Capital Controls Help Shield Emerging Markets from the Crisis?
Makram El-Shagi
Economics Bulletin,
Nr. 1,
2012
Abstract
Using a new dataset on capital market regulation, we analyze whether capital controls helped protect emerging markets from the real economic consequences of the 2009 financial and economic crisis. The impact of the crisis is measured by the 2009 forecast error of a panel state space model, which analyzes the business cycle dynamics of 63 middle-income countries. We find that neither capital controls in general nor controls that were specifically targeted to derivatives (that played a crucial role during the crisis) helped shield economies. However, banking regulation that limits the exposure of banks to global risks has been highly successful.
Artikel Lesen
Inflation Expectations: Does the Market Beat Professional Forecasts?
Makram El-Shagi
North American Journal of Economics and Finance,
Nr. 3,
2011
Abstract
The present paper compares expected inflation to (econometric) inflation forecasts based on a number of forecasting techniques from the literature using a panel of ten industrialized countries during the period of 1988 to 2007. To capture expected inflation, we develop a recursive filtering algorithm which extracts unexpected inflation from real interest rate data, even in the presence of diverse risks and a potential Mundell-Tobin-effect.
The extracted unexpected inflation is compared to the forecasting errors of ten
econometric forecasts. Beside the standard AR(p) and ARMA(1,1) models, which
are known to perform best on average, we also employ several Phillips curve based approaches, VAR, dynamic factor models and two simple model avering approaches.
Artikel Lesen
The Financial Crisis from a Forecaster’s Perspective
Katja Drechsel, Rolf Scheufele
Abstract
This paper analyses the recession in 2008/2009 in Germany, which is very different from previous recessions, in particular regarding its cause and magnitude. We show to what extent forecasters and forecasts based on leading indicators fail to detect the timing and the magnitude of the recession. This study shows that large forecast errors for both expert forecasts and forecasts based on leading indicators resulted during this recession which implies that the recession was very difficult to forecast. However, some leading indicators (survey data, risk spreads, stock prices) have indicated an economic downturn and hence, beat univariate time series models. Although the combination of individual forecasts provides an improvement compared to the benchmark model, the combined forecasts are worse than several individual models. A comparison of expert forecasts with the best forecasts based on leading indicators shows only minor deviations. Overall, the range for an improvement of expert forecasts during the crisis compared to indicator forecasts is relatively small.
Artikel Lesen
Inflation Expectations: Does the Market Beat Professional Forecasts?
Makram El-Shagi
IWH Discussion Papers,
Nr. 16,
2009
Abstract
The present paper compares expected inflation to (econometric) inflation forecasts
based on a number of forecasting techniques from the literature using a panel of
ten industrialized countries during the period of 1988 to 2007. To capture expected
inflation we develop a recursive filtering algorithm which extracts unexpected inflation from real interest rate data, even in the presence of diverse risks and a potential Mundell-Tobin-effect.
The extracted unexpected inflation is compared to the forecasting errors of ten
econometric forecasts. Beside the standard AR(p) and ARMA(1,1) models, which
are known to perform best on average, we also employ several Phillips curve based approaches, VAR, dynamic factor models and two simple model avering approaches.
Artikel Lesen