The Interest Rate Sensitivity of Investment

Baldi, Guido; Lange, Alexander Article The Interest Rate Sensitivity of Investment Credit and Capital Markets – Kredit und Kapital Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Baldi, Guido; Lange, Alexander (2019) : The Interest Rate Sensitivity of Investment, Credit and Capital Markets – Kredit und Kapital, ISSN 2199-1235, Duncker & Humblot, Berlin, Vol. 52, Iss. 2, pp. 173-190, https://doi.org/10.3790/ccm.52.2.173 This Version is available at: https://hdl.handle.net/10419/293862 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Credit and Capital Markets 2 / 2019 The Interest Rate Sensitivity of Investment Guido Baldi and Alexander Lange* Abstract The interest rate sensitivity of investment has often played an important role in macroeconomic models. However, many vector autoregressive (VAR) models do not include investment to the list of variables. In this paper, we empirically investigate the size and the evolution of the interest rate sensitivity of investment for the United States and the four largest European economies in the last few decades. We use a VAR model with four variables at quarterly frequency: real investment, real gross domestic product (GDP), inflation, and a measure of the short-term interest rate. In our VAR, the structural interest rate shock is identified under the assumption that macroeconomic quantities and inflation react to interest rate innovations with a lag. We test the appropriateness of this specification by comparing our approach with the identification of shocks derived from the changes in volatility approach. For the countries under consideration, we determine a date during either the 1980s or the 1990s where the interest rate sensitivity of investment began to decrease and became less responsive to monetary policy. In addition, we find that the interest rate sensitivity of investment has been higher in the United States than in Europe, particularly in the first subperiod. Die Zinssensitivität der Investitionen Zusammenfassung Die Zinssensitivität der Investitionen spielt oft eine große Rolle in theoretischen makroökonomischen Modellen. In dieser Studie untersuchen wir empirisch die Höhe und die zeitliche Änderung der Zinssensitivität der Investitionen für die Vereinigten Staaten und die vier größten europäischen Volkswirtschaften. Wir verwenden ein VAR-Modell * Guido Baldi, University of Bern, Department of Economics, Schanzeneckstr. 1, CH- 3012 Bern, Switzerland; German Institute for Economic Research (DIW Berlin), D-10108 Berlin, Germany. E-Mail: guido[email protected]m; Alexander Lange, Georg-Au- gust-Universität Göttingen, Faculty of Economic Sciences, Humboldtallee 3, 37073 Göttingen, Germany. E-Mail: [email protected]. We thank the anonymous referees and the editor Hendrik Hakenes for very useful comments and suggestions. In addition, comments by various colleagues at our institutions are gratefully acknowledged. All remaining errors are our own. The views expressed in this paper are those of the authors and not necessarily those of the institutions to which the authors are affiliated. Credit and Capital Markets, Volume 52, Issue 2, pp. 173–190 Scientific Papers OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 174 Guido Baldi and Alexander Lange Credit and Capital Markets 2 / 2019 mit vier Variablen: reale Investitionen, reales Bruttoinlandsprodukt, Inflation und kurzfristige Zinsen. In unserem VAR identifizieren wir den strukturellen Schock unter der Annahme, dass die realen makroökonomischen Variablen verzögert auf einen Zinsschock reagieren. Wir testen die Angemessenheit dieser Spezifikation, indem wir unsere Vorgehensweise mit der Identifikation durch den “changes in volatility approach” vergleichen. Wir finden heraus, dass entweder in den 1980er oder frühen 1990er Jahren ein Strukturbruch stattgefunden und sich die Zinssensitivität der Investitionen verringert hat. Interessanterweise zeigen unsere Resultate zudem, dass die Zinssensitivität der Investitionen in den Vereinigten Staaten höher gewesen ist als in den untersuchten europäischen Ländern– insbesondere bis in die 1980er Jahre. Keywords: Investment, Effects of Interest Rates, Monetary Policy JEL Classification: E22, E43, E52 I. Introduction The interest rate sensitivity of investment has played an important role in many macroeconomic models. In the standard New Keynesian model, the short-term interest rate targeted by the central bank transmits the effects of monetary policy to the economy (for an overview, see e. g. Gali 2010). Our aim in this paper is to empirically investigate the size of the interest rate sensitivity of investment and its evolution over the recent economic history for the United States and the four largest European economies. We use a vector autoregressive (VAR) model with four variables at quarterly frequency: real investment, real gross domestic product (GDP), inflation, and a measure of the short-term interest rate. In many VAR models analyzing the effects of interest rate shocks, investment is not included in the list of variables under investigation. In addition, to the best of our knowledge, potential structural breaks in the interest rate sensitivity of investment have not been analyzed in the literature. In our analyses, we focus on the United States and the four largest European economies and apply a common strategy to identify interest rate shocks for these countries. Using a common model allows us to obtain a general picture of the effects of interest rate shocks. Because we are interested only in the effects of the interest rate shock, our main goal is to identify the interest rate innovations without having to recover the other structural shocks. In the related literature using VAR models, the assumption is often made that macroeconomic quantities such as GDP respond to an interest rate shock with a lag. This is a priori plausible given the various reaction and implementation lags that affect macroeconomic quantities and the price setting process. However, one may also question the appropriateness of this identification strategy.1 Following Lanne/Lütkepohl (2008b), the changes in volatility approach allows us to 1 For alternative approaches, see e. g., Christiano/Eichenbaum/Evans (1999). OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 The Interest Rate Sensitivity of Investment 175 Credit and Capital Markets 2 / 2019 identify all the structural shocks without further restrictions and to test this unrestricted model against our identification. The changes in volatility approach can be used if the volatilities of the shocks differ across sub-periods in our sample. This is a pattern that one may often find in historical time series. Using Chow tests, we find evidence for a structural break point for each country under investigation. These break points occur during either the 1980s or the early 1990s, when the interest rate sensitivity of investment started to decrease. Having verified our identification strategy using the break points determined by the Chow tests, we then split the sample into two sub-periods and find that the sensitivity has decreased since the 1980s and early 1990s. According to our results, expansionary interest rate shocks have had expansionary effects on real investment in the past. In recent decades, interest rate shocks have displayed ambiguous real effects on investment and may be neutral. In addition, our findings imply that the interest rate sensitivity of real investment in the first subperiod was higher in the United States than in Continental Europe. The difference is less pronounced in the second subperiod. Discussions on the relation between interest rates and investment have a long tradition that dates back at least as far as Wicksell ([1898] 1936) and Klein (1947) (for an overview, see e. g., Backhouse/Boianovsky 2016). In theory, investment can be expected to be influenced by a rise in the user cost of capital, of which the interest rate is one component.2 Therefore, a decrease in interest rates is expected to lead to an increase in investment. Despite these clear theoretical predictions, empirical papers report difficulties in determining the sensitivity of investment to the interest rate or, more generally, to the user cost of capital (see, e. g. Guiso/ Kashyap/Panetta/Terlizzese 2002). This is due to a potential simultaneity problem, especially when annual data are used, which may be reflected in a positive correlation between interest rates and actual or expected investment. Intuitively, it occurs because more optimistic expectations typically increase both interest rates and the number of profitable investment opportunities. In addition, a central bank usually seeks to increase its targeted interest rate when it expects investment to rise. If this simultaneity problem is not addressed in empirical analyses, estimations of the interest rate sensitivity may be biased downwards. Previous studies investigating the interest rate sensitivity of investment or, more generally, the sensitivity of investment to the user cost of capital, find mixed results (using methods other than those in our paper). Guiso etal. (2002) and Gilchrist/Zakrajsek (2007) report relatively high long-term elasticities. Ca- 2 The user cost of capital depends not only on the interest rate, but also on the rate of depreciation and the relative price of investment to output (Jorgenson 1963). The size of the effect of the real user cost on investment is influenced by the elasticity of substitution between capital and labor. The higher the elasticity of substitution, the stronger the decline in investment after an increase in the user cost. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 176 Guido Baldi and Alexander Lange Credit and Capital Markets 2 / 2019 ballero (1999) and Schaller (2006) also find a negative effect of the user cost on investment. However, at the same time, several studies report that the interest rate may be less important than are quantity variables in influencing investment decisions (for an early review, see, e. g. Chirinko 1993). In addition, results from a survey among firms do not imply a large interest rate sensitivity of investment (Sharpe/Suarez 2013). Recently, it has been argued that the interest rate sensitivity of the overall economy has declined (see, among others, Willis/Cao 2015). Provided that this argument also applies to investment, capital accumulation in advanced economies may have become less responsive to monetary policy. As a result, central banks lowering their policy rates would have become less successful in stimulating investment than in the past and, at the same time, less successful in curbing investment booms. This raises questions about the real effects of monetary policy shocks. If the interest rate under consideration is the short-term interest rate targeted by the central bank, interest rate shocks may reflect monetary policy shocks to a significant extent. There is a large body of literature investigating monetary policy shocks using VAR models (see, e. g. Bernanke/Blinder 1992; Bernanke/Mihov 1995; Bagliano/Favero 1998 or Uhlig 2005). Several papers, for instance, Peersman (2004) and Boeckx/DosschePeersman (2017), have investigated differences across countries with respect to the effects of monetary policy shocks. In contrast to most of these studies that study the effects of shocks to the interest rate (or monetary aggregates) on inflation and output, our analysis is focused on how investment responds to interest rate shocks. In VAR models, the challenge for researchers is to disentangle policy makers’ responses to nonmonetary developments from monetary innovations. Endogenous changes are reactions of the interest rate to the evolution of macroeconomic variables, whereas exogenous policy captures all other actions and represent fundamental shocks to the economy. Thus, the better we are at disentangling endogenous from exogenous shifts in the interest rate, the more successful we can be at addressing the simultaneity issue between interest rates and investment. An exogenous interest rate shock may be interpreted in a variety of ways (see, e. g. Christiano etal. 1999 or Uhlig 2005). First, one may interpret it as an exogenous shock to the preferences of central bankers as, for example, shifts in the weights given to inflation or the output gap. Second, exogenous shifts may occur because a central bank tries to meet the expectations of financial market participants. Shocks to these expectations will translate to shocks in the interest rate policy of central banks. A third reason for the detection of interest rate shocks in a VAR model is measurement errors in real-time data that are available when a central bank decides on the level of interest rates. Fourth, changes in the interest rate may reflect market forces as well as policy decisions. Central banks usually allow small movements in their targeted short-term interest rate that are driven by market forces. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 The Interest Rate Sensitivity of Investment 177 Credit and Capital Markets 2 / 2019 The rest of this paper is organized as follows. In Section 2, we describe the data and methodology used in this paper. In Section 3, we present the results of our analysis. Finally, Section 4 contains the conclusion. II. Data and Methodology 1. Data We use the following variables in our VAR analyses.3 t I is the log of real gross fixed capital formation. t GDP stands for the log of the real gross domestic product. t P is the log of the price level. For the United States and the United Kingdom, we use the core price index without food and energy. For the other countries, we use the price index with the longest available time series. For Germany and Italy, this is the consumer price index, and, for France, it is the GDP deflator. Finally, t ir is the short-term interest rate. For the United States, this is represented by the federal funds rate. For the United Kingdom, the bank rate is chosen, which is the interest rate set by the Bank of England. We use the threemonth money market rate for the remaining European countries in order to have long time series. For a robustness check, we also use the long-term interest rates for government bonds with a maturity of ten years for all countries. For the United States, data availability allows us to use time series data starting in 1957 Q1. The data for the United Kingdom start in 1965 Q1. For the other European countries, we use data starting in 1970 Q1. These data provide us with fairly long time series. Earlier data were not available for at least one time series in these countries. In the baseline regressions, we use data that end in 2007 Q4 to exclude the period since the financial crisis that could lead to misleading results, as interest rates have remained very low, and quantitative easing measures have supplemented considerably the interest rate channel of monetary policy. However, we also conduct sensitivity analysis using data that end in 2018 Q3. 2. Methodology a) The VAR Model As discussed above, we use a VAR model of order p of the form ( ) 11 , 0, t t ptp t t u y A y A y u with u Σ -- = ++ + ~ 3 Data for the United States comes from the St. Louis Fed and the Bureau of Economic Analysis. For European countries, we use data provided by Eurostat and the national central banks. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 178 Guido Baldi and Alexander Lange Credit and Capital Markets 2 / 2019 where ( ) = 1 ,..., ' t t Kt yy y is a vector of observable variables, the i A are ( ) ´KK coefficient matrices and t u are K -dimensional serially uncorrelated reduced form residuals. The vector of variables in logs is given by ( ) , ,, ' t t ttt y I GDP P ir=. t I stands for real private fixed investment, t GDP is real gross domestic product, t P is the price level, and t ir denotes the short-term interest rate. Using augmented Dickey-Fuller tests and tests on trend stationarity (Dickey/Fuller 1979), we find that all variables are () 1I. Therefore, we express our variables in first differences for the empirical analysis. The structural residuals ε t in a VAR model can be obtained by pre-multiplying the matrix B , which contains the instantaneous effects of the structural shocks on the observed variables. Therefore, the structural shocks are a linear transformation of the reduced form residuals: - == 1 εε tt t t u B or B u We use a Cholesky decomposition of the covariance matrix to identify the structural shocks of our model. This choice is based on approaches followed in previous studies, which assumed that a monetary policy shock has no immediate effect on macroeconomic quantities and inflation (see, e. g. Christiano etal. 1999). However, this identification strategy is not uncontroversial. Since we use quarterly data, while many papers use monthly (although partly interpolated) data, the validity of our identification strategy is not guaranteed. To verify our identification, we use the identification by changes in volatility as described in Rigobon (2003) and Lanne/Lütkepohl (2008b). In this approach, we exploit the fact that we can identify a structural break for each country using Chow tests. This provides us with two distinguishable volatility regimes in the data. Importantly, this identification scheme allows us to test for overidentifying restrictions, i. e., whether our identification of the interest rate shock is supported by the data. The split samples for the pre- and post-break periods can then be used to run a VAR( p ) for each period and compute impulse responses. The lag order p in our model is determined by the Akaike Information Criterion (AIC). b) Using Chow Tests to Detect Structural Breaks As discussed above, after estimating the VAR models for the entire period using a Cholesky decomposition, we use the changes in volatility approach to test for the appropriateness of this identification strategy and to derive separate impulse responses using this approach. This requires determining a structural break point in the parameters of the covariance matrix. To this end, we use two types of Chow tests as described in Doornik/Hendry (1997). The first (“break point”) tries to detect a break point by testing for constant parameters and a changing covariance structure (Hansen 2003). The second (“sample split”) tests only for constant parameters and assumes no changes in the white noise error OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 The Interest Rate Sensitivity of Investment 179 Credit and Capital Markets 2 / 2019 term. These two tests help us investigate whether a structural break is due to structural changes either in the parameters or in the covariance matrix. The first version of the Chow test (break-point “bp”) tests both parameter constancy and the constancy of the white noise variance: ( ) ( ) () ( ) 2 12 1,21122 ˆ ˆˆ ln ln ln , bp TT T T λχ Σ ΣΣ=+ - - » where 1B TT < and £- 2B T TT and ( ) 1 2 '' 1,2 12 11 11 ˆˆˆ ˆˆ TT tt tt t tTT uu uu TT Σ = =- + =+ åå , () ( ) ( ) 11 '1 11 1 ˆˆ 1ˆ T tt t uu T Σ = =å, ( ) ( ) ( ) 2 2 '2 221 1 ˆˆˆ T tt tTT uu T Σ =- + =å. The residuals ˆ t u are obtained by running a VAR over the entire period T , whereas the residuals () 1 ˆt u and ( ) 2 ˆt u are obtained by running a VAR over 1 T and 2 T , respectively. The null hypothesis is no structural break in the parameters and the covariance structure. Since Chow test showed that the 2 χ distribution is a poor approximation, the null hypothesis is rejected too often. To overcome this problem, we use a residual based bootstrap procedure to obtain empirical quantiles. The second Chow statistic tests for a sample split (ss) and is given by: ( ) ( ) () ( ) ( ) 2 1 2 1,2 1 1 2 2 12 1 ln ln ˆ ˆˆ ss TT T T TT λχ Σ ΣΣ æ é ùö ÷ ç êú =+ - + » ÷ ç÷ ç êú èø + ëû The null hypothesis is that there is no structural break. c) Testing the Validity of the Identifying Strategy We attempt to test whether our identification of the interest rate shock is justified by the data. To this end, we use the changes in volatility approach, as in Rigobon (2003) and Lanne/Lütkepohl (2008b). Following this approach, we start by considering the two covariance matrices obtained after imposing the structural break in period B T : [ ] 1 '2 1,..., 1 ,..., , B tt B for t T uu for t T T Σ Σ æ =- ç =ç ç ç= è  The two matrices can be decomposed into 1 'BBΣ= and 2 'BBΣΨ= . As usual, B is the matrix of the instantaneous shock responses and is a diagonal matrix capturing the changes in volatility across the two periods (Lanne/Lütkepohl 2008a). We use the break point found by the Chow test as an exogenous break point. This ensures that the break points are chosen carefully. However, OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 180 Guido Baldi and Alexander Lange Credit and Capital Markets 2 / 2019 even if the break points are fixed incorrectly, the time invariant parameters can still be estimated consistently (Rigobon 2003). The covariance matrices are given by: 1 '' 12 1 11 ; ˆˆ ˆˆ ˆ 11 ˆ TT B tt tt BB t tT B uu uu T TT ΣΣ - == == - -+ åå The estimates ˆ B and ˆ Ψ can be obtained by maximizing the log likelihood: ( ) ( ) ( ) ( ) 11 12 11 ˆˆ ln ln ' ( ') ln ' ( ') . 22 BB T TT BB tr BB B B tr B BΣ Ψ ΣΨ -- - -+ =- + - + Moreover, we use the procedure introduced by Lanne/Lütkepohl (2008b) to improve the estimation precision of this method. The matrices ˆ B and ˆ Ψ obtained from maximizing the log likelihood function are used in an iterative GLS estimation. The GLS method uses ˆ β to update the covariance estimates by ( ) ' ˆ ˆt t tK uyZI β =- Ä . ( ) ( ) ( ) ( ) 1 1 1 '1 ' 1 1 1 11 1 ˆˆˆ ˆ ˆˆ ˆ ˆ ˆˆ ˆ ˆ , ,..., ˆ ( ') ( ') ˆ ( ') ( ') , p TT B tt tt t tT B TT B t tt t t tT B vec A A ZZ BB ZZ B B ZBBy ZBBy βν Ψ Ψ - - -- == - -- == éù =êú ëû éù êú = Ä+ Ä êú êú ëû éù êú ´Ä +Ä êú êú ëû åå åå where '1',..., ' t t tp Zy y -- éù = ëû . This procedure iterates until convergence of the likelihood. To derive standard errors for the estimates, we calculate the square root of the elements of the inverted Fisher information matrix (Hamilton 1994). The initial matrix B is the Choleski decomposition of ˆ u Σ, which is obtained from least squares estimation of the reduced form VAR and the initial Ø matrix is an identity matrix. A necessary condition for identification with changes in volatility is that the elements of the main diagonal of ˆ Ψ are distinct. There seems to be a consensus in the literature for using a type of a Wald test for pairwise comparison Lütkepohl/Netsunajev (2014): ] [ ] [ - = ~ "¹ éù +- ëû 22 1 () 2 ˆˆ ˆˆ , ˆˆ ψψ λχ ψ ψ ψψ ij Wi j ij ij Var Var Cov The null hypothesis is that the elements are not distinct (see, for instance, Herwartz/Ploedt 2016 or Lütkepohl/Netsunajev 2014). If the elements are distinct, the model is identified. Thus, further restrictions that may be relevant OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 The Interest Rate Sensitivity of Investment 187 Credit and Capital Markets 2 / 2019 Figure 2: Impulse Response Functions of the Percentage Change in Investment to an Unexpected 25 Basis Point Cut in the Interest Rates Before and After the Break Point (Data end in 2018 Q3). 68 % Confidence Bands are Based on 2000 Bootstrap Replications −1 0 1 2 3 4 0 1 2 3 4 5 6 7 8 9 10 11 12 Pre 1984 Q4 Post 1984 Q4 USAεi→ −1 0 1 2 3 0 1 2 3 4 5 6 7 8 9 10 11 12 Pre 1991 Q3 Post 1991 Q3 UKεi→ 0 1 2 0123456789101112 Pre 1991 Q4 Post 1991 Q4 Franceεi→ −0.5 0.0 0.5 1.0 0 1 2 3 4 5 6 7 8 9 10 11 12 Pre 1988 Q2 Post 1988 Q2 Germanyεi→ −1 0 1 0 1 2 3 4 5 6 7 8 9 10 11 12 Pre 1996 Q3 Post 1996 Q3 Italyεi→ OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 188 Guido Baldi and Alexander Lange Credit and Capital Markets 2 / 2019 IV. Conclusion In this paper, we empirically investigate the interest rate sensitivity of investment for the United States and the four largest European economies. In particular, we analyze whether this sensitivity has declined in recent decades. We use a vector autoregressive model with four variables: real investment, real gross domestic product, inflation, and a measure of the short-term interest rate. In our model, the structural interest rate shock is identified assuming that macroeconomic quantities and inflation react with a lag to interest rate innovations. The appropriateness of this specification is tested using the changes in volatility approach, exploiting the evidence that the volatilities of the shocks differ across two sub-periods. This approach might be useful for future empirical research in economic history, where changes in volatility may be particularly frequent. Using Chow tests to determine specific break points, we split the sample into pre- and post-break periods. For the countries under consideration, we detect a period during either the 1980s or the 1990s where the interest rate sensitivity of investment started to decrease. According to our findings, expansionary interest rate shocks have had expansionary effects on real investment in the decades before the 1980s, this seems to be less the case more recently. Our results suggest that, in recent decades, interest rate shocks have displayed ambiguous real effects on investment and may be neutral. The decrease in the interest rate sensitivity of investment is particularly pronounced for the United States. The past three decades have been characterized by a decline in the level of interest rates and a subdued evolution of corporate investment. The findings in this paper suggest that these long-term developments are not influenced by monetary policy, but are driven by structural factors. Future research could investigate the robustness of our results using different data and empirical methods. References Backhouse, R. E./Boianovsky, M. (2016): Secular Stagnation: The History of a Macroeconomic Heresy. The European Journal of the History of Economic Thought, Vol. 23(6), pp. 946–970. Bagliano, F. C./Favero, C. A. (1998): Measuring Monetary Policy with VAR Models: An Evaluation. European Economic Review, Vol. 42(6), pp. 1069–1112. Bernanke, B. S./Blinder, A. S. (1992): The Federal Funds Rate and the Channels of Monetary Transmission. American Economic Review, Vol. 82(4), pp. 901–921. Bernanke, B. S./Mihov, I. (1995): Measuring Monetary Policy. NBER Working Papers No. 5145, National Bureau of Economic Research. Boeckx, J./Dossche, M./Peersman, G. (2017): Effectiveness and Transmission of the ECB’s Balance Sheet Policies. International Journal of Central Banking, Vol. 13(1), pp. 297– 333. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 The Interest Rate Sensitivity of Investment 189 Credit and Capital Markets 2 / 2019 Caballero, R. J. (1999): Aggregate Investment. In J. B. Taylor and M. Woodford (editors): Handbook of Macroeconomics, Vol. 1 of Handbook of Macroeconomics, Elsevier, chapter 12, pp. 813–862. Candelon, B./Lütkepohl, H. (2001): On the Reliability of Chow Type Test for Parameter Constancy in Multivariate Dynamic Models. Economics Letters, Vol. 73, pp. 155–160. Chirinko, R. S. (1993): Business Fixed Investment Spending: Modeling Strategies, Empirical Results, and Policy Implications. Journal of Economic Literature, Vol. 31(4), pp. 1875–1911. Christiano, L. J./Eichenbaum, M./Evans, C. L. (1999): Monetary Policy Shocks: What Have we Learned and to What End?. In J. B. Taylor and M. Woodford (editors): Handbook of Macroeconomics, Vol. 1 of Handbook of Macroeconomics, Elsevier, chapter 2, pp. 65–148. Dickey, D. A./Fuller, W. A. (1979): Distribution of the Estimators for Autoregressive Time Series with a Unit Root, Journal of the American Statistical Association. Journal of the American Statistical Association, Vol. 74, pp. 427–431. Doornik, J. A./Hendry, D. F. (1997): Modelling Dynamic Systems using PcFiml 9.0 for Windows. International Thomson Business Press. Gali, J. (2010): Monetary Policy, Inflation, and the Business Cycle: An Introduction to the New Keynesian Framework and Its Applications. Princeton University Press, Second Edition. Gilchrist, S./Zakrajsek, E. (2007): Investment and the Cost of Capital: New Evidence from the Corporate Bond Market. NBER Working Paper No. 13174, National Bureau of Economic Research. Gudmundsson, M. (2017): Global Financial Integration and Central Bank Policies in Small, Open Economies. The Singapore Economic Review, Vol. 62(01), pp. 135–146. Guiso, L./Kashyap, A./Panetta, F./Terlizzese, D. (2002): How Interest Sensitive is Investment? Very (When the Data are Well Measured). mimeo. Hamilton, J. D. (1994): Time Series Analysis. Princeton University Press. Hansen, P. R. (2003): Structural Changes in the Cointegrated Vector Autoregressive Model. Journal of Econometrics, Vol. 114(2), pp. 261–295. Herwartz, H./Ploedt, M. (2016): Simulation Evidence on Theory-based and Statistical Identification under Volatility Breaks. Oxford Bulletin of Economics and Statistics, Vol. 78(1), pp. 94–112. Hoerdahl, P./Sobrun, J./Turner, P. (2016): Low Long-term Interest Rates as a Global Phenomenon. BIS Working Papers No. 574, Bank for International Settlements. Jorgenson, D. (1963): Capital Theory and Investment Behavior. American Economic Review, Vol. 53(2), pp. 247–259. Klein, L. R. (1947): Theories of Effective Demand and Employment. Journal of Political Economy, Vol. 55, pp. 108–108. Lanne, M./Lütkepohl, H. (2008a): A Statistical Comparison of Alternative Identification Schemes for Monetary Policy Shocks. EUI Working Papers 23. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56 190 Guido Baldi and Alexander Lange Credit and Capital Markets 2 / 2019 Lanne, M./Lütkepohl, H. (2008b): Identifying Monetary Policy Shocks via Changes in Volatility. Journal of Money, Credit and Banking Vol. 40(6), pp. 1131–1149. Lütkepohl, H. (2007): New Introduction to Multiple Time Series Analysis, Springer Publishing Company. Lütkepohl, H./Netsunajev, A. (2014): Structural Vector Autoregressions with Smooth Transition in Variances: The Interaction Between U.S. Monetary Policy and the Stock Market. SFB 649 Discussion Paper 31. Peersman, G. (2004): The Transmission of Monetary Policy in the Euro Area: Are the Effects Different Across Countries?’. Oxford Bulletin of Economics and Statistics, Vol. 66(3), pp. 285–308. Rigobon, R. (2003): Identification Through Heteroskedasticity. The Review of Economics and Statistics Vol. 85(4), pp. 777–792. Schaller, H. (2006): Estimating the Long-run User Cost Elasticity. Journal of Monetary Economics Vol. 53(4), pp. 725–736. Sharpe, S. A./Suarez, G. A. (2013): The Insensitivity of Investment to Interest Rates: Evidence from a Survey of CFOs. Finance and Economics Discussion Series 2014–2, Board of Governors of the Federal Reserve System (U.S.). Uhlig, H. (2005): What are the Effects of Monetary Policy on Output? Results from an Agnostic Identification Procedure. Journal of Monetary Economics Vol. 52(2), pp. 381–419. Wicksell, K. ([1898] 1936): Interest and Prices. New York: M. Kelley Publishers, 1965. Willis, J. L./Cao, G. (2015): Has the U.S. Economy Become Less Interest Rate Sensitive?. Kansas City Fed. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.52.2.173 | Generated on 2023-01-16 13:27:56