تولید مسکن و مدل های بااهمیت قیمت : پیامدها برای سیاست های پولی

I analyze the consequences of including home production in a New Keynesian model with staggered price setting. Home production amplifies responses to technology and monetary policy shocks. Compared to a model without home production, the model generates close to twice the output response to a monetary policy shock. I consider the implications of several nominal interest rate rules and show that a traditional Taylor rule lacks its usual attractive properties. Alternatively, strict inflation targeting implements the constrained efficient allocation.

A substantial amount of individuals’ time endowments are allocated to neither work on the market nor leisure, but to work at home. Home work includes activities such as meal preparation, laundry, and grocery shopping. The amount of hours devoted to home work has remained large over time. Aguiar and Hurst (2007), using evidence from time use surveys spanning five decades, show that adults spend approximately 20 hours per week working at home (Table 1). This compares to an average of 34 hours per week working in the labor market. Goods and services produced at home are, at least in principle, substitutable for market produced goods and services. This suggests that the relative price between home and market consumption influences the allocation of time between market activities and home production. Consequently, to the degree that market prices are sticky, the allocation of labor between the home and market may be distorted. What are the consequences of including home production in an economy with sticky prices? Does home production alter monetary policy tradeoffs? If not, economists studying monetary policy can safely abstract from it. However, if home production changes the consequences of various monetary policies, it becomes an empirical question whether these effects can be safely neglected.This paper investigates the dynamic properties of an economy with nominal rigidities and home production. More specifically, I incorporate household production into a New Keynesian Dynamic Stochastic General Equilibrium (DSGE) model with staggered price setting and analyze the responses of the endogenous variables to technology and money supply shocks. Allowing households to substitute between market and home production provides an amplification mechanism to changes in the money supply. In particular, including home production in a calibrated model generates a market output response to a monetary policy shock that is roughly twice as large as in a model without it. The intuition for the amplified response to a monetary policy shock runs as follows. When the central bank increases the money supply or lowers the nominal interest rate, firms want to raise prices. Because of staggered price setting some firms are stuck at their old prices, which makes the market produced good relatively cheap. Households respond by substituting out of leisure and into consumption. This mechanism is the same in the model with home production as in the model without it. However, in the model with home production, households not only substitute away from leisure, but also away from producing at home. The expansion of market labor supply partially offsets the increase in marginal costs associated with the increase in production. Consequently, output expands by more in the model with home production than without it. Similarly, when there is an increase in market technology, both the model with and without home production exhibit humped shaped output dynamics since price stickiness prevents immediate price reductions to the increase in technology. The output response in the model with home production, however, is bigger because labor is flowing from the home sector to the market sector. Next, I ask what are the consequences of following different nominal interest rate rules in an economy with home production. The attractive properties of a conventionally defined Taylor rule, which include minimizing the output gap and inflation, no longer hold in an economy with home production. Intuitively, if there is a reallocation of labor away from the market sector, due to say a favorable technology shock in the home sector, a central bank following the Taylor rule would cut interest rates, which would stimulate market output and hinder the efficient allocation of resources. In contrast, strict inflation targeting implements the constrained first best allocations. This paper contributes to the work incorporating home production in DSGE models of business cycles, which started with Greenwood and Hercowitz, 1991 and Benhabib et al., 1991 and continues with more recent contributions such as Fisher (2007). Additionally, Arouba et al. (2014), given in the context of a deterministic money search DGE model, analyze how the level of inflation affects home prices. Another strand of related literature studies multiple sector New Keynesian models like Kilian et al. (1995), and more recently, Barsky et al. (2007). The latter paper shows that in an economy where only some prices are sticky, namely those of durable goods, monetary shocks generate large effects. In this paper, the market sector has sticky prices and the home sector, at least implicitly, has flexible prices. Closest to this paper is Ngouana (2012), who includes home production for services in a model with staggered price setting. Ngouana’s model captures the fact that consumption of services is more sensitive to monetary policy shocks than nondurable consumption. This paper is complimentary to Ngouana’s because I consider shocks to technology in addition to monetary policy, derive the optimal monetary policy, and explore the quantitative and qualitative implications of suboptimal policy. In contrast to Ngouana (2012), my model economy has one sector, whereas his has two. In the textbook one sector New Keynesian model, there are well studied welfare results of optimal policy, which makes the results found here easily relatable to optimal policy in the textbook model.1 Moreover, adding another sector means keeping track of hours transitions across sectors, in addition to transitions between the home and market. To preserve the most clarity in the mechanism, I maintain the one sector model. I speculate on how this simplifying assumption may alter the main result in the conclusion.

Households devote a nontrivial amount of time to home production. I show that including home production in an otherwise standard New Keynesian model provides a substantial amplification mechanism for monetary policy shocks and an amplification and propagation mechanism for technology shocks. Despite its intentions, a conventional Taylor rule does not stabilize the economy, because its interpretation of the natural rate of output is incorrect. Strict inflation targeting results in a “divine coincidence” where maximizing welfare is achieved by minimizing inflation variability. This evidence suggests that welfare maximizing central banks should prefer inflation targeting to jointly targeting output and price stability. Throughout the paper I made numerous simplifying assumptions. Most notably, there is only one sector and no capital accumulation. Including capital accumulation in a textbook New Keynesian model does not change the divine coincidence result. Hence, the qualitative results regarding optimal policy will be similar, although the quantitative results might change. Similarly, provided that home and market produced goods are at least somewhat substitutable, the addition of another sector is unlikely to change the qualitative result that responding to the theoretical output gap from the model without home production decreases welfare. However, as Ngai and Pissarides (2008) show, home production is mainly a substitute for market produced services. Hence, to improve upon the quantitative accuracy of the estimates of welfare losses, imbedding a two sector model along the lines of Ngouana (2012) in a medium scale DSGE model could be a valuable extension. Moreover, introducing another sector brings an additional complication for monetary policy, insofar as it must balance the goals of achieving the correct relative prices and limiting price dispersion. These are interesting topics and are left for future research.