PDF C.V.
Education
Ph.D. - University of Oregon (2013)
B.S. in Economics and Mathematics - Westminster College (2008)
Positions
Senior Lecturer - University of Sydney 2018 - present
Economic Research Advisor - Reserve Bank of Australia 2017 - present
Visiting Scholar - Bank of Finland 2019 Jan - Feb
Lecturer - UNSW Sydney 2013 - 2018
Visiting Economist - Reserve Bank of Australia 2016 - 2017
Graduate Teaching Fellow - University of Oregon 2008 - 2013
Publications
Gibbs, C.G., 2017, "Forecast combination, non-linear dynamics, and the macroeconomy," Economic Theory, 63(3), 653-686
DOI 10.1007/s00199-016-0951-x
Mathematica files containing the proofs for Lemma 1 and Lemma 2; Files to replicate key results for the Lucas monetary model and the Fisherian model.
Working paper version available here.
Gibbs, C.G. and Kulish, M., 2017, "Disinflations in a model of imperfectly anchored expectations," European Economic Review, 100(11), 157-174
Online appendix. Replication files.
Cuffe, H.E. and Gibbs C.G., 2017, "The Effect of Payday Lending Restrictions on Liquor Sales," Journal of Banking & Finance, 85(12), 132-145
Media coverage: Washington Post, Economist's View, BusinessThink
SSRN Top Ten download list for ERN: Urban Economics & Public Policy and ERN: Other Microeconomics: Intertemporal Consumer Choice & Savings
Gibbs, C.G., 2018, "Learning to Believe in Secular Stagnation," Economics Letters. 163(2), 50-54
Appendix B explores the E-stability properties of the liquidity trap equilibrium of Schmitt-Grohe and Uribe (2016)
Working Papers
Optimal Monetary Policy at the ZLB with Imperfect Knowledge - with Stefano Eusepi (UT - Austin) and Bruce Preston (University of Melbourne)
Presentation slides from Bank of Finland and SNDE conference.
Abstract: This paper explores optimal monetary policy at zero lower bound (ZLB) when agents have imperfect knowledge of the long run values of key endogenous variables. We study a New Keynesian environment, where agents make long-run forecasts based on current observable endogenous variables. The environment generates low-frequency drift in expectations in response to short-run forecasts errors, which is consistent with survey data on expectations of many macroeconomic variables. We vary the degree of drift to analyze the robustness of the optimal monetary policy solution at the ZLB of Eggertsson and Woodford (2003) when forward guidance is ruled out by assumption. We find that optimal policy often requires the central bank to hold interest rates at zero beyond when the ZLB naturally binds, even when the forward guidance channel is ruled out by assumption. The resulting optimal policy shares some features that are consistent with the rational expectations (RE) commitment policy but appears to favor inflation targeting over price level targeting. Specifically, the degree to which inflation expectations are anchored plays a significant role in the duration of the ZLB. If expectations are not well-anchored, the ZLB may become a trap from which the economy cannot escape. We then construct a hybrid case that allows for RE state dynamics and drifting long-run beliefs. We find that our policy recommendations are largely robust in this environment. Finally, we take a medium-scale version of the hybrid model to the data to explore the quantitative impact of imperfect knowledge when there is a binding ZLB.
A Unified Theory of Learning to Forecast - with George Evans (University of Oregon) and Bruce McGough (University of Oregon)
Preliminary draft available by request. Slides from SNDE 2018 presentation.
Abstract: We propose a model of boundedly rational decision making that nests adaptive and eductive learning as special cases to rationalize observed behavior in Learning-to-Forecast Experiments when laboratory participants have significant knowledge of the market environment. We assume that there exist heterogeneous agents who engage in the well-known behavior of level-k reasoning, where level-0 agents are adaptive learners and level-infinity agents are eductive learners. Agents may increase their depth reasoning based on past forecast errors following a type of evolutionary dynamic. We find that this setup allows us to flexibly model boundely rational forward-looking behavior, allows us to explain bounded but non-convergent dynamics in experimental settings where convergence under adaptive learning should be obtained but is not, and that eductively stability can be the outcome of an adaptive process.
Expectations and the Empirical Fit of DSGE Models - with Eric Gaus (Haverford College) - (2018)
Abstract: This paper studies the improvement in empirical fit of dynamic stochastic general equilibrium (DSGE) models that assume adaptive learning in lieu of rational expectations (RE). The literature finds that estimated DSGE models with adaptive learning generate near universal improvements in fit, while inference on structural parameters is mostly unchanged. Improvements are attributed to the increased persistence generated by backward-looking expectations. We show, however, that improvements often result from altered cross-equation restrictions and not additional persistence assumptions. Nested comparisons of Euler-equation and infinite-horizon adaptive learning both significantly improve upon RE but only the latter's improvements are due to expectation formation. Bounded rationality assumptions offer an intuitive way to improve both in-sample and out-of-sample DSGE model fit. But our results suggest that learning models best-capture persistent deviation in beliefs from fundamentals rather than temporary deviations at business cycle frequencies.
Conditionally Optimal Weights and Forward-Looking Approaches to Combining Forecasts - with Andrey Vasnev (University of Sydney) - (2017)
This paper has undergone a substantial revision. Below is an old draft of the idea.
"Overcoming the Forecast Combination Puzzle: Lessons from the Time-Varying Efficiency of Phillips Curve Forecasts of U.S. Inflation" - (2015) - Old Draft
Previously circulated under the title: Systematic Forecast Errors, Forecast Combination, and the Forecast Combination Puzzle
Abstract: In applied forecasting there is a trade-off between in-sample fit and out-of-sample forecast accuracy. Parsimonious model specifications usually outperform richer model specifications. As a consequence, there is often predictable information in forecast errors that is hard to exploit. However, we show how this predictable information can be exploited in forecast combinations. In this case, optimal combination weights should minimize conditional mean squared error, or a conditional loss function, rather than the unconditional variance as in the commonly used framework of Bates and Granger (1969).
We prove that our conditionally optimal weights lead to better forecast performance. The conditionally optimal weights support other forward-looking approaches to combining forecasts, where forecast weights depend on expected model performance. We show that forward-looking approaches can robustly outperform the benchmark random walk and many commonly used forecast combination strategies including equal weights in real-time out-of-sample forecasting exercises of inflation.
DSGE Reno: Housing and Commodity Investment Booms in a Multisector Small Open-Economy Model - Gabriela Nodari (RBA) and Jonathan Hambur (RBA)
Previous version: DSGE Reno: Adding a Housing Block to a Small Open Economy Model - with Gabriela Nodari (RBA) and Jonathan Hambur (RBA) - (2017)
Abstract: We propose a straightforward approach to adding a housing sector to a large-scale multisector small open-economy DSGE model to explore the effect of global commodity price shocks on domestic housing investment. The model predicts that housing investment booms may follow commodity booms. Commodity booms have a persistent effect on housing services price inflation as they ‘crowd out’ housing investment, which leads to a lower capital stock and higher prices. When the boom ends, the combination of these higher prices and falling interest rates induces a significant housing investment response. The model attributes a sizeable portion of the recent increase in housing investment in Australia directly to falling commodity prices.
Strongly Rational Routes to Randomness - (2016)
Abstract: Many behavioral models of expectation formation in macroeconomics and finance assume agents choose from a menu of predictor rules over time to construct forecasts. The menus often include a costly rational predictor that yields a correct prediction taking into account the heterogeneity of expectations. In this paper, I study the eductive justification of this rational predictor in the sense of Guesnerie (2002). I show that the condition for eductive stability is strictly weaker than in the homogeneous expectations case under commonly held assumptions. This implies that many rational routes to randomness identified by Brock and Hommes (1997) are actually strongly rational and that rational forecasting behavior is not necessarily ruled out when eductive stability in the homogeneous expectations case fails. The latter has implications for when rational behavior may occur in laboratory experiments on expectations.
Work in Progress
"Forecast Combination and Sunspot Equilibria" - with Bruce McGough (University of Oregon)
Conference Presentations
Symposium of the Society for Nonlinear Dynamics and Econometrics - Tokyo, Japan (2018)
WEAI Conference - San Diego, CA (2017)
Symposium of the Society for Nonlinear Dynamics and Econometrics - Paris, France (2017)
Workshop of the Sydney Macro Reading Group - University of Sydney (2016)
Workshop of the Australasian Macroeconomic Society - UNSW Australia (2015)
Expectations in Dynamic Macroeconomic Models Conference - University of Oregon (2015)
Continuing Education in Macroeconomics - University of Tasmania (2014)
Symposium of the Society for Nonlinear Dynamics and Econometrics - NY, NY (2014)
Continuing Education in Macroeconomics - UNSW, Sydney NSW (2013)
Expectations in Dynamic Macroeconomic Models Conference - Federal Reserve Bank of San Francisco (2013)
Canadian Economic Association Meetings - HEC Montreal (2013)
32th Annual International Symposium on Forecasting - Boston, MA (2012)
Served as Referee for
Journal of Economic Behavior and Organization
Macroeconomic Dynamics
European Economic Review
Journal of Economic Dynamics and Control
Economic Inquiry
Studies in Nonlinear Dynamics and Econometrics
Journal of Macroconomics
Journal of Economic Surveys
Economic Record
Education
Ph.D. - University of Oregon (2013)
B.S. in Economics and Mathematics - Westminster College (2008)
Positions
Senior Lecturer - University of Sydney 2018 - present
Economic Research Advisor - Reserve Bank of Australia 2017 - present
Visiting Scholar - Bank of Finland 2019 Jan - Feb
Lecturer - UNSW Sydney 2013 - 2018
Visiting Economist - Reserve Bank of Australia 2016 - 2017
Graduate Teaching Fellow - University of Oregon 2008 - 2013
Publications
Gibbs, C.G., 2017, "Forecast combination, non-linear dynamics, and the macroeconomy," Economic Theory, 63(3), 653-686
DOI 10.1007/s00199-016-0951-x
Mathematica files containing the proofs for Lemma 1 and Lemma 2; Files to replicate key results for the Lucas monetary model and the Fisherian model.
Working paper version available here.
Gibbs, C.G. and Kulish, M., 2017, "Disinflations in a model of imperfectly anchored expectations," European Economic Review, 100(11), 157-174
Online appendix. Replication files.
Cuffe, H.E. and Gibbs C.G., 2017, "The Effect of Payday Lending Restrictions on Liquor Sales," Journal of Banking & Finance, 85(12), 132-145
Media coverage: Washington Post, Economist's View, BusinessThink
SSRN Top Ten download list for ERN: Urban Economics & Public Policy and ERN: Other Microeconomics: Intertemporal Consumer Choice & Savings
Gibbs, C.G., 2018, "Learning to Believe in Secular Stagnation," Economics Letters. 163(2), 50-54
Appendix B explores the E-stability properties of the liquidity trap equilibrium of Schmitt-Grohe and Uribe (2016)
Working Papers
Optimal Monetary Policy at the ZLB with Imperfect Knowledge - with Stefano Eusepi (UT - Austin) and Bruce Preston (University of Melbourne)
Presentation slides from Bank of Finland and SNDE conference.
Abstract: This paper explores optimal monetary policy at zero lower bound (ZLB) when agents have imperfect knowledge of the long run values of key endogenous variables. We study a New Keynesian environment, where agents make long-run forecasts based on current observable endogenous variables. The environment generates low-frequency drift in expectations in response to short-run forecasts errors, which is consistent with survey data on expectations of many macroeconomic variables. We vary the degree of drift to analyze the robustness of the optimal monetary policy solution at the ZLB of Eggertsson and Woodford (2003) when forward guidance is ruled out by assumption. We find that optimal policy often requires the central bank to hold interest rates at zero beyond when the ZLB naturally binds, even when the forward guidance channel is ruled out by assumption. The resulting optimal policy shares some features that are consistent with the rational expectations (RE) commitment policy but appears to favor inflation targeting over price level targeting. Specifically, the degree to which inflation expectations are anchored plays a significant role in the duration of the ZLB. If expectations are not well-anchored, the ZLB may become a trap from which the economy cannot escape. We then construct a hybrid case that allows for RE state dynamics and drifting long-run beliefs. We find that our policy recommendations are largely robust in this environment. Finally, we take a medium-scale version of the hybrid model to the data to explore the quantitative impact of imperfect knowledge when there is a binding ZLB.
A Unified Theory of Learning to Forecast - with George Evans (University of Oregon) and Bruce McGough (University of Oregon)
Preliminary draft available by request. Slides from SNDE 2018 presentation.
Abstract: We propose a model of boundedly rational decision making that nests adaptive and eductive learning as special cases to rationalize observed behavior in Learning-to-Forecast Experiments when laboratory participants have significant knowledge of the market environment. We assume that there exist heterogeneous agents who engage in the well-known behavior of level-k reasoning, where level-0 agents are adaptive learners and level-infinity agents are eductive learners. Agents may increase their depth reasoning based on past forecast errors following a type of evolutionary dynamic. We find that this setup allows us to flexibly model boundely rational forward-looking behavior, allows us to explain bounded but non-convergent dynamics in experimental settings where convergence under adaptive learning should be obtained but is not, and that eductively stability can be the outcome of an adaptive process.
Expectations and the Empirical Fit of DSGE Models - with Eric Gaus (Haverford College) - (2018)
Abstract: This paper studies the improvement in empirical fit of dynamic stochastic general equilibrium (DSGE) models that assume adaptive learning in lieu of rational expectations (RE). The literature finds that estimated DSGE models with adaptive learning generate near universal improvements in fit, while inference on structural parameters is mostly unchanged. Improvements are attributed to the increased persistence generated by backward-looking expectations. We show, however, that improvements often result from altered cross-equation restrictions and not additional persistence assumptions. Nested comparisons of Euler-equation and infinite-horizon adaptive learning both significantly improve upon RE but only the latter's improvements are due to expectation formation. Bounded rationality assumptions offer an intuitive way to improve both in-sample and out-of-sample DSGE model fit. But our results suggest that learning models best-capture persistent deviation in beliefs from fundamentals rather than temporary deviations at business cycle frequencies.
Conditionally Optimal Weights and Forward-Looking Approaches to Combining Forecasts - with Andrey Vasnev (University of Sydney) - (2017)
This paper has undergone a substantial revision. Below is an old draft of the idea.
"Overcoming the Forecast Combination Puzzle: Lessons from the Time-Varying Efficiency of Phillips Curve Forecasts of U.S. Inflation" - (2015) - Old Draft
Previously circulated under the title: Systematic Forecast Errors, Forecast Combination, and the Forecast Combination Puzzle
Abstract: In applied forecasting there is a trade-off between in-sample fit and out-of-sample forecast accuracy. Parsimonious model specifications usually outperform richer model specifications. As a consequence, there is often predictable information in forecast errors that is hard to exploit. However, we show how this predictable information can be exploited in forecast combinations. In this case, optimal combination weights should minimize conditional mean squared error, or a conditional loss function, rather than the unconditional variance as in the commonly used framework of Bates and Granger (1969).
We prove that our conditionally optimal weights lead to better forecast performance. The conditionally optimal weights support other forward-looking approaches to combining forecasts, where forecast weights depend on expected model performance. We show that forward-looking approaches can robustly outperform the benchmark random walk and many commonly used forecast combination strategies including equal weights in real-time out-of-sample forecasting exercises of inflation.
DSGE Reno: Housing and Commodity Investment Booms in a Multisector Small Open-Economy Model - Gabriela Nodari (RBA) and Jonathan Hambur (RBA)
Previous version: DSGE Reno: Adding a Housing Block to a Small Open Economy Model - with Gabriela Nodari (RBA) and Jonathan Hambur (RBA) - (2017)
Abstract: We propose a straightforward approach to adding a housing sector to a large-scale multisector small open-economy DSGE model to explore the effect of global commodity price shocks on domestic housing investment. The model predicts that housing investment booms may follow commodity booms. Commodity booms have a persistent effect on housing services price inflation as they ‘crowd out’ housing investment, which leads to a lower capital stock and higher prices. When the boom ends, the combination of these higher prices and falling interest rates induces a significant housing investment response. The model attributes a sizeable portion of the recent increase in housing investment in Australia directly to falling commodity prices.
Strongly Rational Routes to Randomness - (2016)
Abstract: Many behavioral models of expectation formation in macroeconomics and finance assume agents choose from a menu of predictor rules over time to construct forecasts. The menus often include a costly rational predictor that yields a correct prediction taking into account the heterogeneity of expectations. In this paper, I study the eductive justification of this rational predictor in the sense of Guesnerie (2002). I show that the condition for eductive stability is strictly weaker than in the homogeneous expectations case under commonly held assumptions. This implies that many rational routes to randomness identified by Brock and Hommes (1997) are actually strongly rational and that rational forecasting behavior is not necessarily ruled out when eductive stability in the homogeneous expectations case fails. The latter has implications for when rational behavior may occur in laboratory experiments on expectations.
Work in Progress
"Forecast Combination and Sunspot Equilibria" - with Bruce McGough (University of Oregon)
Conference Presentations
Symposium of the Society for Nonlinear Dynamics and Econometrics - Tokyo, Japan (2018)
WEAI Conference - San Diego, CA (2017)
Symposium of the Society for Nonlinear Dynamics and Econometrics - Paris, France (2017)
Workshop of the Sydney Macro Reading Group - University of Sydney (2016)
Workshop of the Australasian Macroeconomic Society - UNSW Australia (2015)
Expectations in Dynamic Macroeconomic Models Conference - University of Oregon (2015)
Continuing Education in Macroeconomics - University of Tasmania (2014)
Symposium of the Society for Nonlinear Dynamics and Econometrics - NY, NY (2014)
Continuing Education in Macroeconomics - UNSW, Sydney NSW (2013)
Expectations in Dynamic Macroeconomic Models Conference - Federal Reserve Bank of San Francisco (2013)
Canadian Economic Association Meetings - HEC Montreal (2013)
32th Annual International Symposium on Forecasting - Boston, MA (2012)
Served as Referee for
Journal of Economic Behavior and Organization
Macroeconomic Dynamics
European Economic Review
Journal of Economic Dynamics and Control
Economic Inquiry
Studies in Nonlinear Dynamics and Econometrics
Journal of Macroconomics
Journal of Economic Surveys
Economic Record
