تعامل ورودی و خروجی و سیاست پولی بهینه

This paper deals with the implications of factor demand linkages for monetary policy design in a two-sector dynamic general equilibrium model. Part of the output of each sector serves as a production input in both sectors, in accordance with a realistic input–output structure. Strategic complementarities induced by factor demand linkages significantly alter the transmission of shocks and amplify the loss of social welfare under optimal monetary policy, compared to what is observed in standard two-sector models. The distinction between value added and gross output that naturally arises in this context is of key importance to explore the welfare properties of the model economy. A flexible inflation targeting regime is close to optimal only if the central bank balances inflation and value added variability. Otherwise, targeting gross output variability entails a substantial increase in the loss of welfare

This paper deals with the implications of factor demand linkages for monetary policy design. We build a dynamic stochastic general equilibrium (DSGE) model with two sectors that produce services and manufactured goods. The gross output of each sector serves either as a final consumption good, or as an intermediate input in both sectors, according to a realistic input–output structure. Factor demand linkages are empirically relevant1 and their importance in the transmission of both sectoral and aggregate shocks has long been recognized by the literature exploring the sources and channels of propagation of the business cycle. Horvath, 1998 and Horvath, 2000 shows that cross-industry flows of input materials can reinforce the effect of sectoral shocks, generating aggregate fluctuations and co-movement between sectors, as originally hinted by Long and Plosser (1983).2 More recently, Bouakez et al. (2011) and Sudo (2008) have shown that factor demand linkages help at generating positive co-movement between non-durable and durable spending in the face of a monetary innovation, thus overcoming the limits of standard two-sector models that feature heterogeneous degrees of price stickiness across sectors.3 However, none of these papers has taken a normative perspective. The novel contribution of the present study is to explore how monetary policy should be pursued in a model with cross-industry flows of input materials. In our framework the monetary authority cannot attain the Pareto optimal allocation consistent with the full stabilization of output and inflation. Thus, we explore optimal monetary policy under the assumption that the policy maker can credibly commit to a policy rule derived from the minimization of a social welfare function. The loss function balances, along with sectoral inflation variability, a preference to reduce fluctuations in aggregate consumption (or, equivalently, value added). Given the natural distinction between consumption and production in the presence of input materials, it is no longer irrelevant whether the monetary authority targets the output gap or the consumption gap. This result has important implications for both the transmission of exogenous shocks and the selection of policy regimes as alternatives to the optimal policy under commitment. Introducing factor demand linkages into an otherwise standard two-sector model amplifies the loss of social welfare and alters the transmission of shocks to the system, compared to the benchmark economy without input materials. A distinctive feature of the model is that a technology shock to either sector also affects potential output in the other sector, even if preferences over different types of consumption goods are separable. Furthermore, factor demand linkages imply that the relative price of services not only affects the marginal rate of substitution between manufactured goods and services, but also exerts a positive (negative) impact on the real marginal cost in the manufacturing (services) sector. The relative magnitude of the second effect depends on the off-diagonal elements in the input–output matrix. Beyond reconciling conventional two-sector DSGE models with a realistic structure of the economy, this paper detects important differences between the way monetary policy should be pursued and what is otherwise prescribed by the existing literature on multi-sector models without factor demand linkages. We compare the welfare properties of the model under alternative policy regimes and show that a flexible inflation targeting regime delivers a welfare loss close to that attained under the optimal policy. Most importantly, the central bank attains a smaller loss when fluctuations in aggregate or core inflation are balanced with those in real value added, compared to the loss induced by targeting gross output. We also consider the case of asymmetric price stickiness, which implies a natural divergence between core and aggregate inflation. Although such a difference is still relevant within our framework, targeting either core or aggregate inflation makes little difference in terms of welfare loss. By contrast, what matters is the term accounting for real volatility: in this respect, targeting the consumption gap entails substantial benefits compared to targeting the production gap. These results emphasize the distinction between consumption and production that naturally arises in this class of models. The remainder of the paper is laid out as follows: Section 2 introduces the theoretical setting; Section 3 discusses the calibration of the model economy; Section 4 explores the Pareto optimal outcome; Section 5 studies the implementation of the optimal monetary policy under commitment and compares the resulting loss of social welfare with that attainable under a number of alternative policy regimes. Section 6 concludes.

We have integrated inter-sectoral factor demand linkages into a dynamic general equilibrium model with two sectors that produce services and manufactured goods. Part of the output produced in each sector is used as an intermediate input of production in both sectors, according to a realistic input–output structure of the economy. The resulting sectoral interactions have non-negligible implications for the formulation of policies aimed at reducing real and nominal fluctuations. A key role is played by the relative price, which not only acts as an allocative mechanism on the demand side (through its influence on the marginal rate of substitution between different classes of consumption goods), but also on the supply side (through its effect on the sectoral real marginal costs of production). The presence of input materials implies a non-trivial difference between aggregate consumption (or, equivalently, value added) and gross output. Such a distinction proves to be of crucial importance at different stages of the analysis. In fact, the welfare criterion consistent with the second-order approximation to households' utility reveals that the policy maker is faced with the task of stabilizing fluctuations in the sectoral rates of inflation and aggregate value added, rather than gross output. Moreover, strategic complementarities induced by factor demand linkages alter the transmission of shocks to the system, compared to what is commonly observed in otherwise standard two-sector models. These results show how accounting for a realistic feature of multi-sector economies entails non-negligible differences with respect to the policy prescriptions retrievable for frameworks that rule out input–output interactions or consider a vertically integrated production structure. The optimal policy can be closely approximated by a flexible inflation targeting regime. However, it is of crucial importance to target consumption gap variability rather than output gap variability. This strategy allows the central bank to avoid inducing additional loss emanating from inter-sectoral complementarities.