monetary policy shocks

We use this informationally-robust instrument to identify the shocks in a SVAR-IV (see Stock and Watson, 2012, 2018; Mertens and Ravn, 2013). We start by showing in Section 4 that in a standardly specified monetary VAR such as e.g. the one in Coibion (2012), and contrary to other leading instruments, our identification does not give rise to either output or price puzzles. We also show that the endogenous component in high-frequency monetary surprises – that is due to the information channel –, produces in the VAR responses that are compatible with the effects of aggregate demand shocks that the central bank is likely to respond to. In Section 5 we then study the transmission of monetary policy shocks on a large and heterogenous set of both macroeconomic and financial variables, as well as on private sector forecasts, and medium and long-term interest rates. We find that a monetary contraction is unequivocally and significantly recessionary. The contraction in output is sudden, significant and larger than reported in previous studies. It is accompanied by a contraction in prices, and there is no evid- ence of puzzles. We document evidence compatible with many of the standard channels of monetary transmission, and reflected in a deterioration of prices, domestic demand,

Figure 10 plots the results generated by this model. Estimates from the stochastic volatil- ity model suggest that the variance of exogenous shocks to the interest rate under a fixed inflation target model gradually rises from 1960 through to 1972, remaining large till the early 1980s, and falls sharply from the early 1980s. This is line with the findings presented in Justiniano and Primiceri (2008) who have highlighted the high variance of these shocks during the pre-Volcker period. The time-varying variance of the inflation target follows a similar pattern, with the largest rise coming in the first half of the 1970s, and gradually falling during Volcker’s disinflation. This exercise suggests that shocks to the inflation tar- get were most volatile during the 1973 through to 1978 period and 1980 through to 1981 period. These dates correspond to the findings presented in Justiniano and Primiceri (2008), Boivin (2005) and Romer and Romer (1989), and the most important changes in the infla- tion target seem to largely explain important changes in the interest rate. Since these two shocks are, by construction, orthogonal in the model, this finding justifies that the exogenous monetary policy shock may be misidentified and biased, and shocks to the inflation target may explain a lot of variation attributed to exogenous monetary policy shocks. Therefore, this framework presents a novel contribution to the characterization of changes in monetary policy, which has traditionally been attributed to exogenous and unexplained changes in the reaction function.

We have addressed the robustness of our findings along several dimensions, but additional evidence could corroborate our results. Other shocks, such as technology shocks, should also have a different impact across the calendar year if wage staggering is not uniform. Identification of technology shocks, however, is contentious and would require additional variables in our VAR, thus reducing degrees of freedom at the estimation stage. A more promising avenue, in our view, is the examination of international evidence. Other countries likely exhibit uneven staggering of wage contracts, and the transmission mechanism of monetary shocks should be affected accordingly. In this respect, Japan is particularly relevant, because a large fraction of Japanese firms set wages in the spring season (during the wage setting process known as Shunto, or “spring offensive”). Preliminary findings for Japan indicate that monetary policy shocks that take place in the third quarter (after the Shunto) have a large impact on output, whereas monetary policy shocks occurring in the second quarter (during the Shunto) have virtually no output effect.

However, the paper reveals that in comparison with Hungary, Poland or the Czech Republic, Slovakia remains to be relatively less explored. The structural vector autoregression (VAR) techniques show very good abilities to study economic fluctuations. It is a popular and well suited approach for the empirical analysis of the behaviour of endogenous variables in response to economic innovations. The method ensures a relatively precise definition of an independent shock, characterized from theoretical evidence. Moreover, with impulse response functions, the procedure allows to follow the dynamic development of variables in a given time horizon as a reaction to a shock. In other words, while exploring the transmission mechanism of the monetary policy, the advantage of structural VAR approach is that it allows identifying monetary policy shocks by characterizing their dynamic effects on economic indicators.

Real and nominal effects of monetary policy are significant in both sample periods; however, before 1980, the response of inflation is larger. Eighteen months after a one- standard-deviation shock, the price level is respectively 0.5 percent and 0.1 percent below baseline in the first and second periods; the confidence bands do not overlap (Tab. 4 and Fig. 9); the response of output also decreases, but just above the one-standard deviation band. A different shape of responses to monetary shocks in different periods is also found by Christiano, Eichenbaum and Evans (1997b) for the US, who find smaller response functions in the period 1989-1995 than in 1965-1995; however, they attribute this result to the different size of a standard monetary shock, concluding that in the US the post-1989 period is characterized by smaller shocks but similar responses to a given shock. This is not the case here; the different shape of responses holds also comparing the effects of unit (one percent), rather than one-standard-deviation, interest rate shocks (Tab. 4). It could be argued that the pegging of the exchange rate in the 1980s may have decreased the size and duration of the effects of nominal shocks on prices; the same effect could be expected as a consequence of the orientation of monetary policy to price stability, representing a stronger nominal anchor. 23 In any case a weaker response of prices to monetary policy shocks in the 1970s is not found.

This paper tests empirically the e¤ect of US monetary policy shocks on the composition of US capital ‡ows and the US trade balance, and its channels. It employs a standard structural VAR speci…cation to identify monetary policy shocks, relying on sign restrictions imposed on the impulse response functions of a few macroeconomic variables. The empirical analysis yields two key …nd- ings. First, US monetary policy shocks exert a statistically and economically meaningful e¤ect on US capital ‡ows. An exogenous easing of US monetary policy by 100 basis points (b.p.) induces net capital in‡ows and a worsening of the US trade balance of around 1% of GDP after 8 quarters. The variance decomposition indicates that US monetary policy shocks over the period 1974 to 2007 explain about 20-25% of the variation in both the US trade balance and capital ‡ows at that horizon. As to the channels, it appears that wealth e¤ects play a central role. Equity returns rise on impact by about 6% in response to a 100 b.p. policy easing, while interest rates fall. Both of these responses in turn induce an increase in private consumption for about 8 quarters, and thus a deterioration in the trade balance.

Models with sticky prices predict that monetary policy changes will affect relative prices and relative quantities in the short run because some prices are more flexible than others. In U.S. micro data, the degree of price stickiness differs dramatically across consumption categories. This study exploits that diversity to ask whether popular measures of monetary shocks (for example, innovations in the federal funds rate) have the predicted effects. The study finds that they do not. Short-run responses of relative prices have the wrong sign. And monetary policy shocks seem to have persistent effects on both relative prices and relative quantities, rather than the transitory effects one would expect from differences in price flexibility across goods. The findings reject the joint hypothesis that the sticky-price models typically employed in policy analysis capture the U.S. economy and that commonly used monetary policy shocks represent exogenous shifts.

monetary theory would predict. But what drives this almost text-book like result for the whole period? The cumulated monetary policy shocks plotted in …gure 1 give an answer to this question. They reveal the problem associated with RR’s estimation of monetary policy shocks by using a single monetary policy reaction function spanning the whole time period. Shocks are either persis- tently negative or positive with an obvious turning point around 1979, which is consistent with the compelling anecdotal evidence on the changes in the monetary policy around that time. Keeping in mind that the monetary policy shocks are the residuals from a monetary policy reaction function, the interpretation of this graph is straight forward. Since monetary policy shock is the di¤erence between the intended federal funds rate and the …tted value of the intended federal funds rate, we get consistently negative values of the residuals for the pre-1979 period. By construction, the use of whole sample leads to a lower value of the

This paper analyzes the propagation of monetary policy shocks through the creation of credit in an economy. Models of the monetary transmission mechanism typically feature responses which last for a few quarters con- trary to what the empirical evidence suggests. To propagate the impact of monetary shocks over time, these models introduce adjustment costs by which agents find it optimal to change their decisions slowly. This paper presents another explanation that does not rely on any sort of adjustment costs or stickiness. In our economy, agents own assets and make occupational choices. Banks intermediate between agents demanding and supplying as- sets. Our interpretation is based on the way banks create credit and how the monetary authority affects the process of financial intermediation through its monetary policy. As the central bank lowers the interest rate by buying government bonds in exchange for reserves, high productive entrepreneurs are able to borrow more resources from low productivity agents. We show that this movement of capital among agents sets in motion a response of the economy that resembles an expansionary phase of the cycle.

The conßict between the empirical investigation of monetary policy and the actual setting of policy is troubling in principle. Rudebusch (1998) stresses this conßict, but provides only indirect evidence on the economic importance of the issue. To assess the economic consequences of using revised data, three questions emerge. First, how do the empirical policy shock measures and policy instrument settings differ in equations (2.2) and (2.3)? In general, estimating (2.2) using standard VAR methods will not recover the reaction function and policy shocks in (2.3). Second, how are impulse response functions from policy shocks to other macroeconomic data affected by this conßict? This question is far more difficult to assess than the Þrst one. In most cases, computing impulse response functions from a monetary policy shock requires estimating the VAR equations for the other variables. 5 Although monetary policy may plausibly respond to each new data revision as described in (2.3), this assumption is somewhat more problematic for real GDP . Should we really expect that true output will be affected directly by the government’s announcement that last month’s announced Þgure for real GDP was half a percentage point too high? Further assumptions about the non- policy equations are required: assumptions about the data revision process over time and the information available to economic agents at any point in time. As the discussion in section 4 indicates, the problems posed by data revisions range beyond the monetary policy shock literature. Third, how is the identiÞcation of non-monetary policy shocks affected by data revision issues? Simple examples below suggest that VAR innovations estimated using revised data will include revision errors. Since identiÞcation of exogenous shocks is achieved by factoring particular covariance matrices of VAR innovations, the presence of additional covariation due to revision errors cannot be ignored, in principle.

We then move to other identi…cation schemes not based on a VAR: our third alternative relies instead on …nancial markets information. Kuttner (2001) proposes to gauge a monetary policy shocks by subtracting from the actual change in the federal funds rate its expectation, i.e. compute the di¤erence between Federal funds futures immediately before and after the decision of the FOMC. The idea is that many of the monetary policy decisions (and often the size of the change) are expected and therefore cannot be labelled as “shocks”. The remaining monetary policy ‘surprises’that agents face should hence produce stronger e¤ects. Such series of monetary policy shocks is available since 1989, when the futures market for the Fed funds rate was established at the Chicago Board of Trade. To determine how commodity prices respond to monetary shocks we simply regress the log change in the commodity price index on a constant, its own lagged values, and lagged values of the policy measure. The lagged values of the shock series are included to capture the direct impact of shocks on commodity price change, and the lagged values of commodity price changes are included to control for the normal dynamics of the commodity price index. 11

In the case of Malawi, there are two theoretical studies, one by Bolnick (1991) and another by Phiri (2002), and no empirical analysis on the country’s monetary transmission process. The Bolnick (1991) study was carried out at the time the RBM was converting from direct to indirect tools of monetary management. The study investigated how the changing conditions would weaken or alter links in the country’s monetary transmission process. Unfortunately, Bolnick was unable to predict the emergence of a reasonably competitive financial sector within the foreseeable future and consequently concluded erroneously that after the RBM’s implementation of indirect monetary controls, the transmission of monetary policy shocks in Malawi would be stage-managed by the central bank through informal consultations with commercial bank managers. Phiri’s (2002) study, on the other hand, does not go beyond a theoretical exposition of textbook transmission channels of monetary policy, which it incorrectly relates to Malawi. For instance, he presents the other asset price effects channel operating through an efficient stock market as one of Malawi’s monetary transmission channels. However, with only eight listed companies at the time of the study, it is clear that the Malawi Stock Exchange (MSE) was still in its infancy to act as a conduit of monetary transmission. As at January 2010, the MSE had 15 listed companies and was still classified as an infant industry.

, and the four weighted foreign variables  y ∗ it P it ∗ R ∗ it Q ∗ it . The U.S. monetary policy shock is identified in the United States portion of the GVAR using a Wold ordering, and is then incorporated into the global solution. In this analysis, only the order of the domestic variables matters, and we order these as y it , P it , R it m , R it , and Q it . A major difference in empirical results that we obtain for the Cholesky decomposition-based GVAR as compared to the models considered in the first two steps of our counterfactual analysis is that after a U.S. monetary policy shock, the German short-term interest rate displays no significant response (see Figure 11c). The non-GVAR setting appears to overstate the response of German interest rates to U.S. monetary policy shocks. In addition, the federal funds rate falls back to its original levels within 14 months, a shorter adjustment phase than in the bilateral models. The peak responses of the U.S. Dollar nominal and real effective exchange rates occur in the second month after the U.S. monetary policy shock, but except for the first two months these responses are insignificant, and for all months of very small magnitude.

---- (**) means rejection of the null hypothesis at the 1% and r denotes the rank of the long- run matrix. It identifies the number of cointegating vectors 4.4 Vector Error correction Models The error correction models of each model are presented as two sub-sections: the first part contains estimation of Error Correction terms of the ECMs. These are the tests for short run dynamics of respective variables of the models that show how fast the long run equilibriums are restored following the short run disequilibrium. In the second part, the VAR in first difference of the estimated ECMs, reports the coefficients of regression of the each variable on its own lags and lags of other variables in the system. The VARMs explain the vector composed of the differences of natural logarithms of respective variables of the model under explanation. The motive here is to find out how output and price levels respond to different monetary policy shocks for Ethiopian Economy. Taking the advantage of the tri-variate VARMs allows to investigating causality among variables of the model with the null hypotheses of no causations.

For comparison with the previous measures of monetary policy shocks, I extract the filtered monetary policy shock series from Smets and Wouters (2007). These shocks are available quarterly over the same sample as the R&R shocks. The correlation between the two shock series is surprisingly high (0.40) given the very different methodologies, and the Smets-Wouters measure is 69% more volatile, as measured by the standard deviation of the shocks. Appendix Figures 7 and 8 display a scatter of the two shock series and their cumulative sums respectively. While the latter tell qualitatively similar stories, the Smets-Wouters shocks point to much more contractionary innovations since the early 1980s than the original R&R shocks, with the sum of the shocks rising from approximately 5% in 1984 to over 20% by the end of 1996. Figure 11 reproduces the sensitivity results of section 2 but applied to the Smets- Wouters shocks using quarterly data. The results closely resemble those found using the R&R GARCH shocks. First, the estimated peak effects using the lag structure from R&R (converted to quarterly frequency) are in the small to medium range: a one-hundred basis point innovation lowers industrial production by a maximum of 1.5% and raises the unemployment rate by slightly less than 0.40 percentage points, when estimated over the whole sample, while the price level is predicted to fall by approximately 1.5%. These smaller estimates of monetary policy shocks likely reflect the same features as in the standard VAR: 1) the use of the actual FFR rather than intended policy changes and 2) not controlling for the real-time information available to the Fed. The results are insensitive to any quarterly shock being dropped from the estimation and are also insensitive to dropping the period of reserves targeting. Finally, the results are much less sensitive to the assumed lag length than the original specification of R&R.

such as Qatar, Saudi Arabia, and the U.A.E., former Eastern Bloc countries, middle income countries like Argentina, Indonesia, and Mexico, and low income countries. The sample period for investigation extends from 1968 to 2008, including events such as Great inflation and the end of the Cold War to examine the effects of monetary policy shocks across different countries under different conditions. Description of variables and data sources are provided in Appendix B. 12 The empirical models (1) and (2), or (3) and (4) are estimated jointly with the equations that determine agents’ forecasts of variables that enter the empirical model (see Appendix A for details). As described in the Appendix, the basic model consists of AD and AS equations and the model is closed with the equations for the monetary policy authority (some sort of Taylor rule), the fiscal policy rule and the exchange rate model.

the fact that de jure exchange rate regimes are rather imprecise, in particular for emerging markets, in describing the true actions of central banks with regard to exchange rate policy. Finally, we find that the nature and degree of real and financial integration is a key determinant for the transmission of US monetary policy shocks. Our database to test for financial integration is unique in that it contains holdings of capital stocks vis-à- vis the United States as well as the rest of the world for all elements of the capital account – FDI, portfolio equity investment, portfolio debt investment and loans. Thus this database allows us to conduct a thorough and comprehensive analysis of the role in particular of financial integration. We find that stock markets in countries that hold a large amount of foreign financial assets (relative to domestic GDP) and also that owe a large amount of domestic financial assets to foreigners react two to three times more strongly to monetary policy shocks than less financially integrated countries.

Moreover, the long-run as well as short-run interest rates and money supply are potential measures of monetary policy which are used in the same equation without addressing the issue of multicollinearity. Jang and Ogaki (2004) examine the relationship between monetary policy shocks and Dollar/Yen exchange rate, prices and output level for USA. The empirical analysis is carried out, following the model of Jang (2000), through structural VECM and VAR by employing long-run and short- run restrictions on the model. They use seven macroeconomic variables in their empirical investigation. These variables include domestic and foreign output levels, domestic prices, domestic and foreign interest rates, real exchange rate, whereas monetary variables include FFR and non-borrowed reserves (NBR). They find that an appreciation of exchange rate is the result of a contractionary monetary policy. This confirms the overshooting hypothesis and is also in line with the uncovered interest rate (UIP) theory. Furthermore, they find that output in domestic and foreign country significantly decreases due to the long-run neutrality restrictions with an exception of USA where a decline in output becomes negligible after four years. Finally, a fall in price is observed as a result of tight MP. While, estimating VECM and VAR with short-run restrictions for variables in their levels they fail to accept the UIP condition, they find strong evidence in support of the existence of price puzzle.

In this paper I will focus on the effect of monetary policy shocks on output. 3 This is a particularly interesting question in that it is one that has generated a huge literature 4 and at the same time constitutes a prime example of a central question of macroeconomics on which the profession has still not converged to a unified answer. At this point, the dispute here is much less about how to read the evidence from a VAR as about how to reconcile the results of VAR studies with those of other types of macroeconomic models, notably structural models in the RBC/DSGE 5 tradition. Reduced-form data analyses such as VARs show a fairly robust finding about the effect of money on output: an unanticipated contractionary impulse to money results in a long-lived, hump-shaped response of output. The difficulty with standard micro-founded structural models is that these models can only reproduce a strong and persistent output response to a monetary impulse if prices are sticky for a very long time. 6 The reaction with respect to this apparent discrepancy has generally been to amend structural theoretical models with nominal rigidities thus enabling them to quantitatively

non-decreasing function of the forecast horizon, or projection lag. Empirical Findings. Using our methodology, we study the transmission of mon- etary policy shocks on a large and heterogenous set of both macroeconomic and fin- ancial variables, as well as on private sector expectations, and medium and long-term interest rates. We find that a monetary contraction is unequivocally and significantly recessionary. Output and prices contract and there is no evidence of puzzles. We docu- ment evidence compatible with many of the standard channels of monetary transmission (Mishkin, 1996). We analyse in detail the response of interest rates at short, medium, and very long maturities and find important but very short-lived effects of policy on the yield curve (Romer and Romer, 2000; Ellingsen and Soderstrom, 2001). Also, we find evidence of a powerful credit channel that magnifies the size of the economic contrac- tion through the responses of both credit and financial markets (Bernanke and Gertler, 1995; Gertler and Karadi, 2015; Caldara and Herbst, 2016). Moreover, we document a deterioration of the external position sustained by a significant appreciation of the domestic currency. Finally, the expectational channel is activated: agents revise their macroeconomic forecasts in line with the deteriorating fundamentals. Finally, we doc- ument that BLP responses optimally deviate from the VAR responses as the horizon grows. As a result of this BLP IRFs revert to trend much faster than VAR IRFs do.