تمرکز بازار، عدم اطمینان اقتصاد کلان و سیاست های پولی

This paper studies the effect of market structure and macroeconomic uncertainty on the transmission of monetary policy. We motivate our analysis with a simple model which predicts that: (1) investment and production in more concentrated sectors are more affected by demand shocks and (2) high uncertainty makes investment and production more sensitive to demand shocks. The empirical analysis estimates the effect of monetary shocks on sectoral output for different sectors in the US using a structural vector autoregressive (VAR) approach. The results are generally consistent with the theoretical predictions.

In this paper, we study the impact of economic uncertainty on the monetary transmission mechanism with particular emphasis on investment decisions. We develop the intuition through a simple model where firms choose their investment and pricing strategies as a function of the degree of market concentration in their industry. The model shows that different levels of market concentration explain different investment and production reactions to changes in demand. Further, we also show that uncertainty about future demand can have a significant impact on investment because it affects market structure. Our theory differs from the previous literature on investment decisions under uncertainty and their impact on the monetary transmission mechanism for which we can separate two different strands. The first of these focuses on the role of irreversible investment decisions under uncertainty as developed by Dixit and Pindyck (1994).1 The second strand emphasizes the presence of “frictions” in credit markets. Our analysis departs from the literature on irreversible investment since we focus on short-run investment decisions where the number of possible investors is exogenous and investment decisions are sticky in the sense that they cannot be changed once the state of nature is known. Our interest in the short run leads to important implications which explain how market concentration shapes the investment–uncertainty relationship. Our focus on market structure, on the other hand, suggests an alternative to credit restrictions theories. More specifically, in the model we consider an industry where potential investors have access to idiosyncratic opportunities for investment in capacity. These investments can either be made in the present or at some future period, when there will be some uncertainty about demand. The number of potential investors with access to such opportunities is fixed and is a crucial parameter in our model as it captures the degree of concentration in the industry: The smaller this number, the more concentrated the industry will be. Although investment opportunities are individual-specific, the model is entirely symmetric as all potential investors face the same opportunity cost of investment, the same capacity level if they make the investment, the same (zero) marginal cost of production, the same discount factor between periods and the same market demand for the (identical) final product. In each period, investment decisions are made simultaneously by those who have not invested in the previous period. In a given period, those who do invest, proceed simultaneously to select prices. Thus, investment decisions in the present are made by comparing expected profits if the investment is postponed until the future with the expected profits if the investment is made in the present. Our first result is that greater market concentration makes investment decisions more sensitive to changes in present and future demand. The intuition is particularly simple: Greater concentration implies higher profits for each investor in each period. This means that, for example, an expected increase in future demand will increase the opportunity cost of investing today and not in the future more than it would have if smaller profits were at stake. While simple, this result establishes an unambiguous link between market concentration on the one hand, and monetary policy and its impact on demand on the other. Our other results concern the effects of uncertainty on investment. As we will discuss in detail below, the previous literature suggests that the correlation between future uncertainty and present investment is ambiguous, but our model suggests that uncertainty increases the option value to wait to invest in the future in a way that may be mediated by market concentration. In particular, we show that whenever uncertainty is so high that a negative shock to demand will significantly alter the structure of the product market, then the effect on today's investment is stronger than when uncertainty is low and a change in market structure following a demand shock is not likely. In other words, today's investment decisions are more sensitive to changes in uncertainty for future demand, the greater this uncertainty is to begin with. The intuition is that whenever shocks to future demand are so large as to have an impact on market structure, then there is an increase in the value of the option of postponing investment decisions until the uncertainty is realized. We proceed to take the intuitions developed by our model to the data, by considering the effects of monetary shocks on industrial production for different manufacturing sectors in the United States. In order to do this, we follow the traditional macroeconomic literature in assuming that monetary policy drives demand shocks. We consider monthly data from 1972 to 2003 for a group of 21 United States manufacturing sectors classified according to the North American Standard Industrial Classification System (NAICS). The focus on United States sectors will help us in identifying the impact that market concentration in each sector has on the monetary transmission mechanism since this is the least open economy in the OECD.2 We use the vector autoregressive (VAR) approach to estimate the effect of unexpected interest rate changes on sectoral output. Although VAR specifications are relatively atheoretical, they have become a common tool for the sectoral analysis of monetary transmission.3 One advantage of this methodology lies in the fact that all the fundamental variables are endogenously determined in the system. In this context, we identify a structural model so that the effect of unanticipated monetary policy changes can be estimated. This overcomes some of the problems found in the parameter interpretation of reduced form equations where demand shifts fail to be exogenous as they may not be anticipated by economic agents. The literature (and, indeed, the data) suggests that output volatility in the United States has significantly decreased from the second half of the eighties, following the changes introduced by the Monetary Control Act of 1980 and the Garn-St. Germain Act of 1982. Consistent with this view, we estimate industry-level reactions to interest rate shocks in two VAR models: One for the period from the beginning of 1972 to the end of 1982 and one for the period up to 2003. Our results are largely compatible with the predictions developed in the theoretical model. We find that the intensity of the reactions in a given sector depends positively on its degree of market concentration while several proxies related to credit and interest rate channels are not as significant. We also find that macroeconomic uncertainty increases sensitivity to interest rate shocks but this evidence is weaker. The paper is organized as follows. The next section relates our work to the previous literature. A simple model is presented in Section 3. The purpose of the model is not to provide a complete description of the monetary transmission mechanism, but to provide some insights into how market concentration may interact with changes in demand and uncertainty about future prospects. Section 4 presents the data. Section 5 discusses the econometric approach we use for the estimation of sectoral reactions to interest rate shocks while Section 6 presents the results and considers the impact of market concentration. Conclusions are drawn in Section 7.

This paper documents and analyses sectoral reactions of output to interest rate shocks for 21 manufacturing sectors in the US. To motivate the paper, we have presented a simple model which focuses on the interaction of firms’ production and investment decisions when entry is restricted and in the presence of uncertainty about future levels of demand. The model generated two main implications for the empirical analysis: (1) higher levels of market concentration imply stronger short-run reactions from firms to interest rate shocks; and (2) future uncertainty strengthens today's output reactions to interest rate shocks. The empirical analysis uses structural VAR models to estimate output reactions to monetary shocks for three digit NAICS manufacturing industries in the USA for the 1972–2003 period. Our results indicate that, among a number of indicators that account for potential restrictions in the credit market and for demand sensitivity, market concentration is the most important factor in explaining the heterogeneity of output responses. More specifically, we find that, in accordance with our theory, output responses to interest shocks are stronger and faster in highly concentrated sectors. We also find some evidence that reactions are stronger in high uncertainty periods. The results suggest a number of policy implications. For example, in a large, relatively closed economy like the United States, our results suggest that market structure plays an important role in explaining the heterogeneity of the responses to monetary policy, and this more so than the measures of credit market imperfections that have been put forward in the previous literature. The extent to which these results carry over to other, more open economies, remains to be investigated.