وقفه انتقال و سیاست پولی بهینه

The credibility problems of monetary policy are enlarged by transmission lags whenever the welfare criterion consists of arguments with differing transmission lags. If, as usually argued, prices react to monetary policy with a longer lag than output, the discretionary bias is substantially increased under a consumer welfare maximizing policy criterion (flexible inflation targeting) in the prototype New Keynesian model. Money growth targeting can significantly reduce the discretionary bias, but is not robust to other specifications of welfare with higher valuation of output stability.

Since Kydland and Prescott (1977) and Barro and Gordon (1983) we have known that an overly ambitious monetary policy which aims to bring output above the natural level is associated with inflation and stabilization biases. If the central bank tries to systematically exploit the short-run trade off between output and inflation, it will lead to higher inflation, and output and inflation being stabilized sub-optimally. Furthermore, due to the lack of commitment to future policies, discretionary policymaking is unable to appropriately influence expectations about the future. At the time policy is implemented, the advantages of the future commitment may already have been realized and the policymaker has incentives to deviate from the pre-announced policy. In the absence of commitment technology, the best thing a policymaker can do is to re-optimize policy in every period. Since people form expectations rationally, this will be anticipated and the only equilibrium is that of the time-consistent optimal discretionary equilibrium which may perform considerably worse than the optimal commitment policy. This paper studies the impact of delayed effects of monetary policy on the economy in the discretionary equilibrium. Delayed effects are commonly referred to as the transmission lags of monetary policy. It is almost universally accepted that monetary policy is subject to rather long transmission lags and that they create various challenges for monetary policy. In this paper we show that if the transmission lags are caused by implementation lags in the private sector, the credibility problems of a welfare-maximizing policymaker that acts under discretion increase. Under the reasonable assumption that pricing decisions of the firms are subject to longer implementation lags than household consumption decisions, the discretionary policy involves no policy-induced stabilization of cost-push shocks in the canonical New Keynesian model.1 The argument is simple: at the horizon the policymaker can affect output gap, inflation (and prices) are already predetermined. The best discretionary policy is then to fully stabilize the output gap. The implementation lags have a severe impact on the discretionary equilibrium in particular if the cost-push shocks are persistent. We argue that when society attaches little weight on output stabilization, adopting a single target for monetary policy, thus having a strict (as opposed to flexible) monetary policy, eliminates the additional credibility problems caused by differing transmission lags. The central bank does not get tempted in deviating from the main nominal target. Our result confirms the results in Söderström (2005) who argues that there is a role for money growth targeting in reducing the discretionary bias. We also show that the relative benefits of money growth targeting over inflation increases when there is an implementation lag in prices. Our results support the Friedman (1960) conjecture that lags in the transmission mechanism could be a reason for adopting money growth targeting, yet this result is not robust to alternative specifications of welfare with higher valuation of output stability. Furthermore, the argument for money growth targeting should be balanced by the potential for instability of money demand. The remainder of the paper is organized as follows. Section 2 present the canonical New Keynesian model and Section 3 derives the optimal discretionary policy strategies under both discretion and commitment. Section 4 discusses most important alternative policy regimes that offer a potential remedy to discretionary bias. Welfare comparisons are then made in Section 5. Section 6 concludes.

It is well established that monetary policy is subject to transmission lags. These lags can be the results of delayed responses of the private sector to economic shocks. We show under the reasonable assumption that when inflation reacts with a longer lag than output gap to changes in monetary policy, the optimal discretionary equilibrium implies no policy-induced stabilization of cost-push shocks. Since inflation is predetermined at the time when monetary policy can influence output gap, the discretionary optimizing policymaker stabilize the output gap perfectly and does not stabilize inflation. This result can be generalized as to having policy only stabilizing the target variable with the shortest transmission lag. For standard specification of welfare we find that implementation lags in prices increase the discretionary stabilization bias severely if cost-push shocks are sufficiently persistent. Money growth targeting reduces the bias substantially, since it features history dependence, similarly to the policy under commitment equilibrium. For the alternative specification of welfare that values output far more, however, the discretionary bias is substantially smaller and money growth targeting does not offer a remedy for reducing the discretionary bias. In fact it does substantially worse. For either specifications, strict targeting of inflation or the price-level does not produce a good outcome. If there are implementation lags, strict price-level targeting seems to be particularly detrimental to the economy.