سرمایه گذاری و سیاست های پولی در منطقه یورو

This paper analyses the effects of a change in monetary policy on firms’ investment in Germany, France, Italy and Spain using a data set which provides aggregated balance sheet and profit and loss account data for 17 different industries and three different size classes. The main findings are twofold. First, in each of the four countries a change in the user cost of capital, which in turn is affected by interest rates, has both statistically and economically significant effects on investment. Second, while the average interest rate on debt is generally higher for small firms than for large firms, there is little evidence that the effects of monetary policy on small firms are larger.

A good understanding of the monetary transmission mechanism in the euro area is essential for the efficient implementation of the ECB’s single monetary policy. While there is a large literature which focuses on the macroeconomic effects of a change in policy-controlled interest rates in the various euro area countries (Kieler and Saarenheimo, 1998), much less comparative work has been done based on microeconomic evidence. Nevertheless, such evidence is important for at least two reasons. First, it has proven to be very difficult to find significant interest rate effects on investment using aggregate data (Blanchard, 1986). Using the cross-sectional variation in disaggregated data, it may be easier to empirically identify such interest rate effects. For the euro area there is hardly any microeconomic evidence on the elasticity of investment with respect to the user cost of capital.2 Most of the studies that estimate Euler equations of investment postulate a production function that is homogenous of degree one and therefore impose a unit elasticity without testing it.3 Second, it has been argued that differences in financial systems could lead to asymmetries in the transmission as some countries are more affected by financial accelerator phenomena than others.4 Typically, such transmission channels imply that monetary policy has distributional effects, which can only be tested using disaggregated data. In this paper we analyse industry-specific investment behaviour in the four largest countries of the euro area (i.e. Germany, France, Italy and Spain) using a semi-aggregate data set on firms’ balance sheets assembled by the European Commission. The data comprises 17 industries (both manufacturing and services) for each country. For each industry we have disaggregated balance sheet information for three different firm-size classes bringing the total number to 51 “representative industries”. Starting from a neoclassical model for investment, we first examine the elasticity of investment with respect to the user cost of capital, we then examine the effects of an interest rate change on the user cost of capital. Finally, we also study to what extent the strength of the effect of monetary policy on industry investment is related to the size of the firms within the industry. The contribution of this paper is threefold. First, we use a consistent data set and methodology to estimate investment equations for the four largest countries of the euro area. This increases the comparability of the results across countries. Moreover, these four countries cover around 80% of total euro area GDP. Second, while most studies that estimate the effect of changes in the user cost of capital on investment focus on variations in tax rates, this paper is the first one to use a time and industry varying interest rate on debt to build firm-specific measures of the user cost of capital. We find a significant negative effect of the user cost on investment in all four countries. While the short-term dynamics differs across countries, it is striking that the long-run parameters are quite similar. The long-run elasticities of the capital stock with respect to both sales and the user cost are not significantly different from 1, implying that a simple Cobb–Douglas specification of the production function cannot be rejected. This result is in stark contrast with the large literature on aggregate investment functions (Blanchard, 1986) and with one of the major arguments of proponents of the credit channel that the empirical evidence on a sizable impact of the user cost on investment is very weak (Bernanke and Gertler, 1995). Our results are in line with the conclusion of Hasset and Hubbard (1997) that recent empirical research on the sensitivity of investment to the user cost with microdata has resulted in substantial estimates of the elasticity ranging from −0.5 to −1.5 While these studies refer to the US economy, we show that the long-run elasticity of the capital stock with respect to the user cost also fall in this range for the four largest countries of the euro area. Third, we are one of the few studies to use microlevel data to test whether, as suggested by the credit channel, the external finance premium paid by small firms reacts to changes in monetary policy.6 While there is a relatively large literature which tests the effects of financing constraints on investment in countries of the euro area, only a few studies directly address to what extent liquidity constraints interact with the stance of monetary policy.7In this paper, we use firm-specific interest rates to examine financial accelerator phenomena. Although we find that the interest rate paid by small firms is on average higher than that paid by larger firms, we do not find evidence that it is also more sensitive to changes in market interest rates. Similarly, the investment by small firms does not appear to be more sensitive to changes in the user cost of capital. These two results cast some doubt on whether the financial accelerator has played an important role in the link between interest rate changes and investment in Germany, France, Italy and Spain over the sample period. The rest of the paper is structured as follows. In the next section we briefly describe the model we estimate. We follow the derivation of a neoclassical investment model as in Hall et al. (1999). In Section 3 we describe the data we use. We also explain how we construct our measure of the user cost of capital and describe the other variables used in the regressions. Section 4 contains the main estimation results. First, we report the estimate of the basic neoclassical investment model. Next, we analyse how the firm-specific interest rates are affected by changes in market interest rates. Finally, we test whether the elasticity of investment with respect to the user cost of capital depends on the size of the firms. Finally, we present the main conclusions in Section 5.

In this paper we have analysed the effects of a change in the user cost of capital and of sales on firms’ investment in Germany, France, Italy and Spain using a data set which provides aggregated balance sheet and profit and loss account data for 17 different industries and three different size classes. Our main findings are twofold. First, in each of the four countries a change in the user cost of capital has both statistically and economically significant effects on investment. The long-run effect of changes in the user cost on the capital stock does not reject a Cobb–Douglas production technology. We obtain a somewhat higher and more significant user cost elasticity than what is usually obtained with panels of individual firm-level data where mostly tax changes are used to identify changes in the user cost. In part, our stronger results could also be due to the fact that the data we use are actually for “representative firms” for which some of the measurement problems typical of panels are averaged out. Overall, the effects of the user cost on investment suggest that the interest rate channel of monetary policy is operative in the euro area. Changes in the level of interest rates have an impact on firms’ investment through the user cost of capital. Second, while the average interest paid by small firms is significantly larger than the average interest paid by large firms, there is no evidence that the premium paid by small firms reacts to changes in the interest rate. There is also no evidence that investment of small firms is more sensitive to the user cost of capital than investment of large firms. These findings put some doubt on the possibility that financial accelerator phenomena play an important role in the transmission mechanism of monetary policy for the large countries of the euro area during the sample period. However, a number of drawbacks of the data we used need to be kept in mind. First, the data does not allow to test for distributional effects of monetary policy among the firms within each size class. Second, our measure of the cost of debt is an average interest rate rather than the marginal cost of debt.