Nathaniel A. Throckmorton
Assistant Professor of Economics
Recent Working Papers
[Revised: May 2019][FRB Dallas Working Paper 1808]
This paper shows the recent success of valuation risk (time-preference shocks in Epstein-Zin utility) in resolving asset pricing puzzles rests sensitively on an undesirable asymptote that occurs because the preference specification fails to satisfy a key restriction on the weights in the Epstein-Zin time-aggregator. When we revise the preferences to satisfy the restriction in a simple asset pricing model, the puzzles resurface. However, when estimating a sequence of Bansal-Yaron long-run risk models, we find valuation risk under the revised specification consistently improves the ability of the models to match asset price and cash-flow dynamics.
Journal of Monetary Economics, forthcoming
During the Great Recession, central banks lowered their policy rate to the zero lower bound (ZLB), calling into question linear estimation methods. There are two alternatives: estimate a nonlinear model that accounts for precautionary savings effects of the ZLB or a piecewise linear model that is faster but ignores the precautionary savings effects. This paper compares their accuracy using artificial datasets. The predictions of the nonlinear model are typically more accurate than the piecewise linear model, but the differences are usually small. There are far larger gains in accuracy from estimating a richer, less misspecified piecewise linear model.
Econometrica, July 2018, Volume 86, Pages 1513-1526
Basu and Bundick (2017) show an intertemporal preference volatility shock has meaningful effects on real activity in a New Keynesian model with Epstein and Zin (1991) preferences. We show when the distributional weights on current and future utility in the Epstein-Zin time-aggregator do not sum to 1, there is an asymptote in the responses to such a shock with unit intertemporal elasticity of substitution. In the Basu-Bundick model, the intertemporal elasticity of substitution is set near unity and the preference shock only hits current utility, so the sum of the weights differs from 1. We show when we restrict the weights to sum to 1, the asymptote disappears and preference volatility shocks no longer have large effects. We examine several different calibrations and preferences as potential resolutions with varying degrees of success.
Economic Journal, June 2018, Volume 128, Pages 1730-1757
[Online Appendix, FRB Dallas Working Paper 1405, WSJ: Real Time Economics, Economic Letter, LSE Business Review]
This paper examines the correlation between uncertainty and real GDP growth. We use the volatility of real GDP growth from a VAR, stock market volatility, survey-based forecast dispersion, and the index from Jurado et al. (2015) as proxies for uncertainty. In each case, a stronger negative correlation emerged in 2008. We contend the zero lower bound (ZLB) on the federal funds rate contributed to our finding. To test our theory, we estimate a New Keynesian model with a ZLB constraint to generate a data-driven, forward-looking uncertainty measure. The correlations between that measure and real GDP growth are close to the values in the data.
Economic Inquiry, October 2017, Volume 55, Pages 1593-1624
This paper analyzes forward guidance in a nonlinear model with a zero lower bound (ZLB) on the nominal interest rate. Forward guidance is modeled with news shocks to the monetary policy rule, which capture innovations in expectations from central bank communication about future policy rates. Whereas most studies use quasi-linear models that disregard the expectational effects of hitting the ZLB, we show how the effectiveness of forward guidance nonlinearly depends on the state of the economy, the speed of the recovery, the degree of uncertainty, the policy shock size, and the forward guidance horizon when households account for the ZLB.
Economics Letters, December 2016, Volume 149, Pages 44-48
Structural models used to study monetary policy often include sticky prices. Calvo pricing is more common but Rotemberg pricing has become popular due to its computational advantage. To determine whether the data supports that change, we estimate a nonlinear New Keynesian model with a zero lower bound (ZLB) constraint and each type of sticky prices. The models produce similar parameter estimates and the filtered shocks are nearly identical when the Fed was not constrained, but the Rotemberg model has a higher marginal data density and it endogenously generates more volatility at the ZLB, which helps explain data from 2008-2011.
European Economic Review, August 2015, Volume 78, Pages 76-96
Several proposals to reduce U.S. debt reveal large differences in their targets. We examine how an unknown debt target affects economic activity using a real business cycle model in which Bayesian households learn about a state-dependent debt target in an endogenous tax rule. Recent papers use stochastic volatility shocks to study fiscal uncertainty. In our setup, the fiscal rule is time-varying due to unknown changes in the debt target. Households infer the current debt target from a noisy tax rule and jointly estimate the transition probabilities. Three key findings emerge from our analysis: (1) Limited information about the debt target amplifies the effect of tax shocks through changes in expected tax rates; (2) The welfare losses are an order of magnitude larger when both the debt target state and transition matrix are unknown than when only the debt target state is unknown to households; (3) An unknown debt target likely reduced the stimulative effect of the ARRA and uncertainty about the sunset provision in the Bush tax cuts may have slowed the recovery and led to welfare losses.
Journal of Economic Dynamics and Control, June 2015, Volume 55, Pages 14-38
This article examines monetary policy when it is constrained by the zero lower bound (ZLB) on the nominal interest rate. Our analysis uses a nonlinear New Keynesian model with technology and discount factor shocks. Specifically, we investigate why technology shocks may have unconventional effects at the ZLB, what factors affect the likelihood of hitting the ZLB, and the implications of alternative monetary policy rules. We initially focus on a New Keynesian model without capital (Model 1) and then study that model with capital (Model 2). The advantage of including capital is that it introduces another mechanism for intertemporal substitution that strengthens the expectational effects of the ZLB. Four main findings emerge: (1) In Model 1, the choice of output target in the Taylor rule may reverse the effects of technology shocks when the ZLB binds; (2) When the central bank targets steady-state output in Model 2, a positive technology shock at the ZLB leads to more pronounced unconventional dynamics than in Model 1; (3) The presence of capital changes the qualitative effects of demand shocks and alters the impact of a monetary policy rule that emphasizes output stability; and (4) In Model 1, the constrained linear solution is a decent approximation of the nonlinear solution, but meaningful differences exist between the solutions in Model 2.
The B.E. Journal of Macroeconomics, January 2015, Volume 15, Pages 157-182
When monetary policy faces a zero lower bound (ZLB) constraint on the nominal interest rate, a minimum state variable (MSV) solution may not exist even if the Taylor principle holds when the ZLB does not bind. This paper shows there is a clear tradeoff between the expected frequency and average duration of ZLB events along the boundary of the convergence region—the region of the parameter space where our policy function iteration algorithm converges to an MSV solution. We show this tradeoff with two alternative stochastic processes: one where monetary policy follows a 2-state Markov chain, which exogenously governs whether the ZLB binds, and the other where ZLB events are endogenous due to discount factor or technology shocks. We also show that small changes in the parameters of the stochastic processes cause meaningful differences in the decision rules and where the ZLB binds in the state space.
Computational Economics, December 2014, Volume 44, Pages 445-476
Policy function iteration methods for solving and analyzing dynamic stochastic general equilibrium models are powerful from a theoretical and computational perspective. Despite obvious theoretical appeal, significant startup costs and a reliance on grid-based methods have limited the use of policy function iteration as a solution algorithm. We reduce these costs by providing a user-friendly suite of MATLAB functions that introduce multi-core processing and Fortran via MATLAB's executable function. Within the class of policy function iteration methods, we advocate using time iteration with linear interpolation. We examine a canonical real business cycle model and a new Keynesian model that features regime switching in policy parameters, Epstein-Zin preferences, and monetary policy that occasionally hits the zero-lower bound on the nominal interest rate to highlight the attractiveness of our methodology. We compare our advocated approach to other familiar iteration and approximation methods, highlighting the tradeoffs between accuracy, speed and robustness.
Other Working Papers
[Revised: May 2018][FRB Dallas Working Paper 1705]
This paper develops a new way to quantify the effects of aggregate uncertainty that accounts for exogenous and endogenous sources. First, we use Bayesian methods to estimate a nonlinear New Keynesian model with stochastic volatility and a zero lower bound constraint on the nominal interest rate. We discipline the model by matching data on uncertainty, in addition to common macro time series. Second, we use the Euler equation to decompose output into expected output and the expected variance and skewness of output. We then filter a time series for each term. Our method captures the effects of higher-order moments over horizons beyond 1 quarter by recursively decomposing expected output. Over a 1-quarter horizon, output uncertainty reduced output less than 0.01% every quarter, similar to volatility shocks in our model. Over horizons that remove the influence of expected output, output uncertainty on average reduced output 0.06% and the peak effect was 0.15% during the Great Recession, similar to structural VAR estimates. Other higher-order moments had much smaller effects on output.
[Revised: February 2017][FRB Dallas Working Paper 1606]
This paper examines the importance of using nonlinear methods to capture the zero lower bound (ZLB) on the Fed's policy rate. While it may seem obvious to impose the ZLB, ex-ante it is unclear how large of an effect the ZLB has on parameter estimates, since it was only hit once in recent history. The parameters and marginal likelihoods from a linear and nonlinear model are similar, but the linear model does not fit the data as well and predicts counterfactually large policy shocks when the Fed is constrained. A quasi-linear model performs better than a linear model but it still generates less volatility at the ZLB and is not as conducive to estimation. When we add a banking sector to create an interest rate spread, the ZLB is even more important.
Econometric evidence shows that when higher income inequality and financial liberalization are added to a set of conventional explanatory variables, they predict significantly larger current account deficits in a cross-section of advanced economies. To study this mechanism, we develop a DSGE model where investors' income share increases at the expense of workers, and where workers respond by obtaining loans from domestic and foreign investors. This supports aggregate demand but generates current account deficits, especially if domestic financial markets are simultaneously liberalized. In emerging markets, because domestic workers cannot borrow, investors deploy their surplus funds abroad, leading to current account surpluses.