Bayesian forecasting with small and medium scale factor-augmented vector autoregressive DSGE models

December 31, 2015

By Stelios Bekiros and Alessia Paccagnini

http://d.repec.org/n?u=RePEc:ucn:oapubs:10197/7322&r=dge

Advanced Bayesian methods are employed in estimating dynamic stochastic general equilibrium (DSGE) models. Although policymakers and practitioners are particularly interested in DSGE models, these are typically too stylized to be taken directly to the data and often yield weak prediction results. Hybrid models can deal with some of the DSGE model misspecifications. Major advances in Bayesian estimation methodology could allow these models to outperform well-known time series models and effectively deal with more complex real-world problems as richer sources of data become available. A comparative evaluation of the out-of-sample predictive performance of many different specifications of estimated DSGE models and various classes of VAR models is performed, using datasets from the US economy. Simple and hybrid DSGE models are implemented, such as DSGE–VAR and Factor Augmented DSGEs and tested against standard, Bayesian and Factor Augmented VARs. Moreover, small scale models including the real gross domestic product, the harmonized consumer price index and the nominal short-term federal funds interest rate, are comparatively assessed against medium scale models featuring additionally sticky nominal prices, wage contracts, habit formation, variable capital utilization and investment adjustment costs. The investigated period spans 1960:Q4–2010:Q4 and forecasts are produced for the out-of-sample testing period 1997:Q1–2010:Q4. This comparative validation can be useful to monetary policy analysis and macro-forecasting with the use of advanced Bayesian methods.

If you need to read into the literature on Bayesian estimation of DSGE VAR models, especially for the purpose of forecasting, this is the paper you need. It includes a nice survey of the methods, models, and the literature, as well as a horse race between the various models. DSGE-FAVAR seems to be the winner, although I wonder how long that will hold. Indeed, the estimations all use linearization, but as shown earlier on this blog,issues with the zero lower bound can seriously mess things up.

Advertisements

Capital Misallocation during the Great Recession

December 23, 2015

By Alessandro Di Nola

http://d.repec.org/n?u=RePEc:pra:mprapa:68289&r=dge

In this paper I evaluate the contribution of financial frictions in explaining the drop in aggregate TFP through misallocation during the Great Recession. I build a quantitative model with heterogeneous establishments; with the help of the model I compute the counterfactual drop in misallocation: by how much would aggregate TFP have decreased if the credit crunch had been absent. I find that a “real recession” would have caused a drop of only 0.16 percent, as opposed to 1.04 percent found in the data; therefore financial frictions account for a significant part of the drop in aggregate TFP. The key mechanism is the following: the increase in the cost of external finance affects negatively the reallocation of productive inputs from low to high productivity firms, by dampening the growth of small-highly productive firms.

There is a literature that studies on strong the effect of misallocation of capital is on total factor productivity and output. This has shown that some economies are far off their potential. This paper shows how such misallocation could have happened over the last business cycle, thereby highlighting a real consequence of the stronger financial frictions.


How Risky Is College Investment?

December 16, 2015

By Lutz Hendricks and Oksana Leukhina

http://d.repec.org/n?u=RePEc:hka:wpaper:2015-014&r=dge

This paper is motivated by the fact that nearly half of U.S. college students drop out without earning a bachelor’s degree. Its objective is to quantify how much uncertainty college entrants face about their graduation outcomes. To do so, we develop a quantitative model of college choice. The innovation is to model in detail how students progress towards a college degree. The model is calibrated using transcript and financial data. We find that more than half of college entrants can predict whether they will graduate with at least 80% probability. As a result, stylized policies that insure students against the financial risks associated with uncertain graduation have little value for the majority of college entrants.

Student debt has been increasing rapidly in recent years in the United States. As this is debt against better future incomes, this is not necessarily a problem. However, if students do not finish their studies and end up with significant debt, their low incomes will not be helping them out of the hole. It is therefore important that they realize that there is such a risk, and that this risk can be very different from student to student. Such a high-stakes risk also means that there could be underinvestment in education, which makes it important to have some sort of insurance mechanism. This paper is about such important questions.


Loss of Skill and Labor Market Fluctuations

December 9, 2015

By Etienne Lalé

http://d.repec.org/n?u=RePEc:bri:uobdis:15/668&r=dge

This paper studies the effects of skill loss on compositional changes in the pool of unemployed, and their impact on aggregate labor market fluctuations. We develop a computationally tractable stochastic version of the Diamond-Mortensen-Pissarides model, wherein workers accumulate skills on the job and lose them during unemployment. Skill loss provides a mechanism for amplifying fluctuations: the loss of skills shifts the average composition of the unemployment pool towards low-surplus workers, which magnifies the response of vacancies to aggregate productivity shocks. The model, however, cannot generate large compositional changes at business cycle frequency: the dynamics of unemployment remains too fast for the pool of searching workers to deteriorate markedly during downturns. Finally, we find that loss of skill plays a quantitatively important role if skills are destroyed immediately upon job loss, and more so during recessions.

This is an interesting paper. However, it is calibrated to the US economy before 2007, that is, an economy with really short unemployment duration. Since then, long term unemployment has become a more prominent feature of the labor market, like in other developed economies. Large compositional changes should be possible in such circumstances and an appropriate experiment should be included in the paper.