Fiscal Stimulus in Times of High Debt: Reconsidering Multipliers and Twin Deficits

We investigate the impact of fiscal stimuli at different levels of the government debt‐to‐GDP ratio for a sample of 17 European countries from 1970 to 2010. This is implemented in an interacted panel VAR framework in which all coefficient parameters are allowed to change continuously with the debt‐to‐GDP ratio. We find that responses to government spending shocks exhibit strong nonlinear behavior. While the overall cumulative effect of a spending shock on real GDP is positive and significant at moderate debt‐to‐GDP ratios, this effect turns negative as the ratio increases. The total cumulative effect on the trade balance as a share of GDP is negative at first but switches sign at higher levels of debt. Consequently, depending on the degree of public indebtedness, our results accommodate long‐run fiscal multipliers that are greater and smaller than one or even negative as well as twin deficit and twin divergence behavior within one sample and time period. From a policy perspective, these results lend additional support to increased prudence at high public debt ratios because the effectiveness of fiscal stimuli to boost economic activity or resolve external imbalances may not be guaranteed.     

Do Central Banks React to House Prices?

We estimate the Federal Reserve's, the Bank of England's, and the Bank of Japan's responses to house prices. We show that generalized method of moments estimates of a Taylor rule augmented with house prices are biased and dispersed. We then use full‐information methods and estimate the policy rule together with a VAR for the nonpolicy variables. These estimates are also biased. We propose an alternative approach and estimate a dynamic stochastic general equilibrium model embedded with a monetary rule with a direct response to house prices. We find that house prices played a separate role in the reaction functions of these central banks.     

Output Gap Estimation and Monetary Policy in China

Using the Bayesian multivariate Beveridge-Nelson decomposition method, this paper estimates China's output gap based on a multivariate dynamic model featuring distinct interactions among real output, inflation, money, and the exchange rate in China during the period 1980-2010. The authors compare the statistical nature and potential forecasting effects of the resulting multivariate gap measure on monetary policy with those of the output gap measures based on univariate models. The empirical results show that only the measure based on the multivariate system significantly predicts monetary policy, which indicates that the output gap estimated by the multivariate system contains more information than the traditional measures for macroeconomic policy adjustments do.