I regress real GDP growth rates on the IMF’s growth forecasts and find that IMF forecasts behave similarly to those generated by overfitted models, placing too much weight on observable predictors and underestimating the forces of mean reversion. I identify several such variables that explain forecasts well but are not predictors of actual growth. I show that, at long horizons, IMF forecasts are little better than a forecasting rule that uses no information other than the historical global sample average growth rate (i.e., a constant). Given the large noise component in forecasts, particularly at longer horizons, the paper calls into question the usefulness of judgment-based medium and long-run forecasts for policy analysis, including for debt sustainability assessments, and points to statistical methods to improve forecast accuracy by taking into account the risk of overfitting.

This note explains the value of strategic foresight and provides implementation advice based on the IMF’s experience with scenario planning and policy gaming. Section II provides an overview of strategic foresight and some of its tools. Scenario planning and policy gaming have been the Fund’s main foresight techniques so far, though other tools have been complementary. Accordingly, section III focuses on the scenario planning by illustrating applications before detailing the methods we have been using, while section IV describes policy gaming including the matrix policy gaming approach with which we have experimented so far. Section V summarizes the key points. In so doing, the note extends an invitation to those in the economics and finance fields (e.g., researchers, policymakers) to incorporate strategic foresight in their analysis and decision making.

In this paper I assess the ability of econometric and machine learning techniques to predict fiscal crises out of sample. I show that the econometric approaches used in many policy applications cannot outperform a simple heuristic rule of thumb. Machine learning techniques (elastic net, random forest, gradient boosted trees) deliver significant improvements in accuracy. Performance of machine learning techniques improves further, particularly for developing countries, when I expand the set of potential predictors and make use of algorithmic selection techniques instead of relying on a small set of variables deemed important by the literature. There is considerable agreement across learning algorithms in the set of selected predictors: Results confirm the importance of external sector stock and flow variables found in the literature but also point to demographics and the quality of governance as important predictors of fiscal crises. Fiscal variables appear to have less predictive value, and public debt matters only to the extent that it is owed to external creditors.

Model selection and forecasting in stress tests can be facilitated using machine learning techniques. These techniques have proved robust in other fields for dealing with the curse of dimensionality, a situation often encountered in applied stress testing. Lasso regressions, in particular, are well suited for building forecasting models when the number of potential covariates is large, and the number of observations is small or roughly equal to the number of covariates. This paper presents a conceptual overview of lasso regressions, explains how they fit in applied stress tests, describes its advantages over other model selection methods, and illustrates their application by constructing forecasting models of sectoral probabilities of default in an advanced emerging market economy.

This paper discusses the impact of the rapid adoption of artificial intelligence (AI) and machine learning (ML) in the financial sector. It highlights the benefits these technologies bring in terms of financial deepening and efficiency, while raising concerns about its potential in widening the digital divide between advanced and developing economies. The paper advances the discussion on the impact of this technology by distilling and categorizing the unique risks that it could pose to the integrity and stability of the financial system, policy challenges, and potential regulatory approaches. The evolving nature of this technology and its application in finance means that the full extent of its strengths and weaknesses is yet to be fully understood. Given the risk of unexpected pitfalls, countries will need to strengthen prudential oversight.

Is over-optimism about a country's future growth perspective good for an economy, or does over-optimism also come with costs? In this paper we provide evidence that recessions, fiscal problems, as well as Balance of Payment-difficulties are more likely to arise in countries where past growth expectations have been overly optimistic. To examine this question, we look at the medium-run effects of instances of over-optimism or caution in IMF forecasts. To isolate the causal effect of over-optimism we take an instrumental variables approach, where we exploit variation provided by the allocation of IMF Mission Chiefs across countries. As a necessary first step, we document that IMF Mission Chiefs tend to systematically differ in their individual degrees of forecast-optimism or caution. The mechanism that transforms over-optimism into a later recession seems to run through higher debt accumulation, both public and private. Our findings illustrate the potency of unjustified optimism and underline the importance of basing economic forecasts upon realistic medium-term prospects.

We compared the predictive performance of a series of machine learning and traditional methods for monthly CDS spreads, using firms’ accounting-based, market-based and macroeconomics variables for a time period of 2006 to 2016. We find that ensemble machine learning methods (Bagging, Gradient Boosting and Random Forest) strongly outperform other estimators, and Bagging particularly stands out in terms of accuracy. Traditional credit risk models using OLS techniques have the lowest out-of-sample prediction accuracy. The results suggest that the non-linear machine learning methods, especially the ensemble methods, add considerable value to existent credit risk prediction accuracy and enable CDS shadow pricing for companies missing those securities.