For a number of euro area periphery countries, this paper explores the stability of the link between bank lending rates and yields on sovereign bonds. A stable relationship between these interest rates is important for the ECB's attempt to restore monetary policy transmission by conducting unconventional measures that aim at bringing down government bond rates. Using vector autoregressive models with time-varying parameters, we find that bank lending rates adjusted incompletely to changes in government bond rates before the onset of the financial crisis, while their responsiveness has even further weakened thereafter. Thus, our results suggest that periphery bank lending rates have not only decoupled from policy rates after mid-2008, but also from yields on sovereign bonds.Keywords ECB • Unconventional monetary policy • Euro area crisis • Interest rate link • Time-varying parameter vector autoregressive modelsWe would like to thank two anonymous referees for helpful comments and suggestions. We are also grateful to the participants of the

We derive counterfactual national interest rate paths for the 17 Euro Area countries for the period 1999 to 2012 to approximate the interest rates countries would have implemented had they still been able to conduct independent monetary policy. We find that prior to the financial crisis the counterfactual interest rates for Germany trace the realized EONIA rate very closely, while monetary policy has been too loose especially for the southern European countries. This situation was inverted with the onset of the financial crisis. To shed light on the underlying decision rule of the ECB, we rank different mechanisms according to their ability to aggregate the national counterfactual paths to the EONIA rate. We find that mechanisms which focus on countries which fare economically worse than the Euro Area average explain the EONIA path best.

Summary This paper investigates the extent to which national economic conditions matter in the ECB's decision process. We employ various decision models to aggregate counterfactual national interest rates based on Taylor rule estimates and test which of the resulting interest paths fits best to the actual monetary policy in the euro area. As a novel feature, we introduce decision models where countries that fare economically worse than the euro area average obtain a higher weight in the decision process. Our results suggest that these models explain actual ECB policy better than GDP‐based bargaining models that have been highlighted in the previous literature. Thus, the ECB seems to have emphasized the needs of countries that face an economic crisis disproportionately highly even before the advent of the financial crisis.

We study monetary policy at the ZLB in a traceable three-period model, in which pricelevel targeting emerges endogenously in the welfare function. We characterize optimal pricelevel forward guidance under discretion and commitment. Potentially non-monotonic discretionary welfare losses are lowest with perfectly flexible prices. Price-level targeting introduces a new constraint on optimal forward guidance that restricts the credible amount of overshooting. With this constraint, the zero lower bound may be binding even after the shock has abated. We characterize conditions when the commitment to hold nominal rates at zero for an extended period is optimal. Finally, we introduce government spending and show that under persistently low policy rates optimal government spending becomes more front-loaded, while procyclical austerity fares worse than discretionary government spending. JEL classifications: E43, E52, E58,Key words: zero lower bound, forward guidance, price-level target, optimal policy * We wish to thank two anonymous referees for excellent suggestions. We are also grateful for comments from participants at various seminars and conferences. All remaining errors are our own. †

No abstract